Examine.com
Research Digest

Issue 3 ◆  January 2015 1

 

Table of Contents
05  Heart benefits of alcohol may not apply
to everyone
  Type 2 diabetes: a preventable disease
17  
 21 Investigating a progression of carb and
saturated fat intakes

 33  Whence the hype?

43  Running on empty: can we chase the fat away?

 51  Fitting into

testing-based
testing-b your
ased genes:
dietary
diet do genetic
ary recommendations
recomme ndations work?

61  Combating obesity through intermitt

intermittent
ent
fasting

70  How does a lifetime of marijuana use affect

the brain?

79  A mouse’s microbiome

microbio me may cause its
brain to leak

87   Ask the Resear

Researcher
cher:: Stuart M. Phillips,
Ph.D., FACN, FACSM
FACSM

92  INTERVIEW: Ramsey Nijem

2

From the Editor
“However, more research is needed ...”
Havee you ever se
Hav seen
en that line in a journal article? O
course you have. It’s a part o almost every article that
we review or ERD. Is more research ever not needed?
A ellow researcher and I would talk about how ubiq- multimodal prevention
uitous this phrase was, and whether it really meant
editor  
anything. He eventually wrote a letter to the editor
o an epidemiology journal,
j ournal, including some analysis
on how ofen the phrase was used in major journals.
Tree years later, I still run across the phrase a dozen
times a day. Tis may never change.
Why is this phrase important? Well, it ties in to one
o the most importan
important,
t, yet least talked about issues
in health research: when are new trials justified, and
what exactly should new trials test? Tere’s a field o
research called “ value o inormation
inormation analysis,” which ing treatments.
analysis,”
places a dollar amount on the public health value o
each unit o new research on a given topic.
Tere are only so many research dollars available. Not Te Human Condition.
every topic can get unding or a large
l arge rando
randomized
mized
trial, and many important topics go unresearched. I’d
like to know whether taking vitamin D in the morn-
ing causes different effects than night-time ingestion.
Will we see research on this topic? Probably not.
Major issues that have already been addressed by
animal studies and observational trials ofen are next
Kamal Patel, Editor-in-Chief 
 
in line or randomized trials, such as the impact o
[INSERT NUTRIENT OR DRUG HERE] on heart dis-
ease biomarkers in [INSERT POPULATION HERE].
reatment is unded more ofen than prevention, and
prevention is unded much less ofen
than interventions investigating
investigating a single method or
pharmaceutical.
Maybe that seems backwards. But it’s not easy to test
the combined impact o getting regular sleep, eating
mostly unrefined oods, getting time outside in the
sun, and carving out time to relax and get some per-
spective. Actually, it’s pretty difficult to test even one
o those interventions. Plus there’s much less money
to be made on prevention, especially when it comes
to ree interventions, than there is to be made by sell-
Tere’s a phrase that reers to the inherent nature o
human existence, including choices and difficulties:
Condition. Sometimes, I tthink
hink there is
a counterpart in Te Research Condition. Health
research is complex and shifing, and somewhat
inherently flawed. Single trials can’t conclusively
answer questions.
questions. Subtle differences in methods and
samples lead to different results. Research doesn’t
really flip flop very ofen — it’s just a much more
iterative and grueling process than the public knows.
And it’s why more research is always needed.
3
 

Contributors
Researchers

Trevor Kashey Alex Leaf  Courtney Silverthorn Margaret Wertheim Zach Bohannan
Ph.D(c) M.S(c) Ph.D. M.S., RD M.S.

Editors

Gregory Lopez Pablo Sanchez Soria Kamal Patel

Pharm.D. Ph.D. M.B.A., M.P.H., Ph.D(c)

Reviewers

Arya Sharma Natalie Muth Stephan Guyenet Sarah Ballantyne Katherine Rizzone
Ph.D., M.D. M.D., M.P.H., RD Ph.D. Ph.D. M.D.

Mark Kern Gillian Mandich

Ph.D., RD Ph.D(c)
4
 

Heart benefits of alcohol

may not apply to everyone
CETP TaqIB genotype modifies the association
between alcohol and coronary heart disease:
The INTERGENE case-control study 

Introduction
With advice coming rom everyone rom physician
physicianss to bartenders, a com-
mon message broadcast during the past couple decades has been that
moderate consumption o alcohol is not just allowable, but beneficial or
heart disease. Indeed, imbibing to the tune o one drink daily or women,
or two drinks daily or men, has been associated with lower risk o cardio-
 vascular disease.
disease.

Proposed mechanisms or the protective effect o alcohol on corona

coronary
ry heart
disease (CHD) include the potential benefits rom the antioxidant
antioxidant effects
o polyphenols in wine, and an increase in high density lipopro
lipoprotein
tein (HDL)
levels. HDL’s most well known unction is to transport cholesterol rom
arteries throughout the body back to the liver, preventing cholesterol rom
being deposited in the arteries, which would cause blockages.

5
 

Lipid-containing
Lipid-containing particles in the blood ofen gain sequences at a given site in the DNA. Both versions
and lose different types o lipids, such as cholester- o the DNA sequence would be considered “normal,”
ol and triglycerides. Te ability o HDL to transer with neither likely to directly cause debilitating dis-
cholesteroll into particles like VLDL is partially reg-
cholestero ease, like a rare mutation might. However, different
ulated by cholesteryl ester transer protein (CEP). polymorphisms may still influence susceptibility to
CEP promotes transer o HDL cholesterol into disease.
VLDL, and in exchange HDL receives triglycerides.
trig lycerides.
CEP is hence thought to reduce HDL cholester
cholesterol,
ol, Tis study looked at how two polymorphisms in
so less CEP in your blood means HDL particles the CEP gene affect the odds o having CHD at
would balloon up with  varying levels o alcohol
alcohol
more cholesterol, and intake. Te two different
more CEP would mean HDL can be alleles (gene variants) o
HDL particles would car- CEP are called B1 and
ry less cholesterol. anti-inlammatory B2. B2 is associated with
decreased CEP mass
Hold on, less HDL cho- or inlammatory, and increased HDL cho-
lesterol … isn’t that a bad
thing? Not necessarily, depending on the lesterol. Given that we
have two copies each o
as HDL is more complex gene, the three different
than just the “good cho- disease state o genotype options in a
lesterol” moniker it has given subject are B1B1,
taken on in public par- the body. B1B2, or B2B2.
lance (and unortunately
physician office parlance as well). HDL also has a A previous study  showed
 showed that men with B2B2 gen-
lesser known, but important role in the immune otype who have an ethanol intake o 50 g (about
system,, perorming a variety o unctions,
system  unctions, such as three drinks) or more per day had about a 60%
binding toxic substances
substances in the blood. HDL can lower risk o heart attacks than men with lower or
be anti-inflammatory
anti-inflammatory or inflammatory , depending no alcohol intake. Tis protective effect o larger
on the disease state o the body. HDL and LDL are amounts o alcohol was not seen in people with the
markers o disease, but they each have physiologi- B1B1 or B1B2 genotypes. On the other hand, in a
cal unctions important to the body, and neither are study in a Mediterran
Mediterranean
ean cohort,
cohort, no interaction
absolute determiners o or protectors against heart between CEP aqIB, alcohol intake, and CHD was
disease. observed.

Back to CEP. Tere is a known polymorphism in
the gene that encodes CEP called CEP aqIB. A
polymorphism is when a particular gene has two
or more relatively common possible nucleotide
Why is that? One reason could be simply different
populations. As seen in Figure 1, different pop-
populations.
ulations can have substantially different CEP
genotype requencies. Rodents such as mice have no

6
 

Figure 1: CETP B2B2 allele frequency in different populations

CEP gene, and also have lower risk o atheroscle-

rosis, though many other actors may be respons
responsible
ible Moderate alcohol intake is ofen encouraged to
this. Complete CEP deficiency is a rare mutation help ward off heart disease. Tis advice is largely
in humans, although it’s much more requent in one based on HDL effects, but these effects may also
area o northern Japan.
Japan. While the requency o this be modified by your genotype.
mutation
mutation is higher in people with heart disease, at
least in that area o Japan, recent studies have shown
that the extremely cholesterol-rich HDL in these Who and what was studied?
people still maintains its antioxidative unction
unction and
 and Population  
ability to move cholesterol
cholesterol out
 out o areas o cholesterol Tis case-control study took place in Sweden as part
buildup.. So the impact o CEP on heart disease is
buildup o the INERGENE research program, which aims
still very much up in the air. to assess the interaction between genetic susceptibil-
ity and chronic disease in southwe
southwest
st Sweden. Cases
Te aim o the current
c urrent study was to re-examin
re-examinee with heart disease were compared against controls
the effect o alcohol intake and its interaction with who didn’t have heart disease, to assess how alcohol

CEP aq1B polymorphism on CHD odds. and genetic variation impacted disease prevalence
prevalence..

Te CHD cases were patients under age 75, admit-

7
 

ted to three regional hospitals or acute corona

coronary
ry alcohol intake or men was considered about one
syndrome and diagnosed with myocardial inarc- drink or more daily, while low intake was about less
tion. O the CHD patient
patientss who agreed to participat
participate,
e, than ½ a drink daily. For women, high alcohol intake
618 patients were included (453 men, 165 women). was classified as about ½ a drink or more daily, while
low alcohol intake was less than ¼ o a drink daily.
O those, 209 men and 86 women had a first-time
myocardial inarction,
inarction, while the remaining 323 Table 1: Tertiles of Ethanol Intake
had an exacerbation o previously diagnosed CHD. Men (g /day) Women (g /day)
Te controls without CHD were randomly selected Low < 6.5 < 3.2
individuals aged 25-74 at the time o sampling, and Medium 6.5-13.1 3.2-6.3
2,921 o them were included. High > 13.1 >6.3

Intervention 
One drink is 14 grams o ethanol, which is the
Te data collected or analysis in this study was
equivalent o about 12 ounces o beer, five ounces o
CEP genotype, as well as sel-reported inorma
inormation
tion
wine, or 1.5 ounces o 40%-alcohol spirits.
about alcohol intake, including requency o intake
o different types o alcohol (low-alcohol beer, medi-
um-strong beer, strong beer, wine, dessert wine, and Tis study examined just over 600 cases o heart
spirits) with eight response categories ranging rom disease and almost 3,000 controls, and classified
never to three or more times a day. how much alcohol they drank into three catego-
ries that differed based on sex.
Alcohol intake inormation reerred to intake over
the previous one-year period or controls and or the
one-year period prior to the most recent coronary What were the findings?
events or study participants. Age and sex-specific Characteristics
Characteristics of Case and Controls 
standard
standard ser
serving
ving sizes or alcoholic b
beverages
everages were For both men and women, there was a smaller per-
used to calculate
calcu late the daily ethanol consumption. centage o alcohol users in the cases compared to the
control groups. For women, 80% o CHD cases and
Daily alcohol intake was divided into three levels 87% o controls reported using alcohol. For men,
(low,, medium, and high), and the odds ratio (OR)
(low 89% o cases, compar
compared
ed to 93% o controls, report-
was calculated or having CHD based on genotype ing drinking alcohol. People with CHD also had
and alcohol intake. Abstainers were classified into a lower average ethanol intake compared to controls.
ourth group, though high/intermediat
hig h/intermediatee intake was Tere were no significant differences in the distri-
compared to the low group, not to the abstainers. bution o CEP genotype (B1B1 versus B1B2 versus
B2B2) between cases and controls.
All models were adjusted or age, body mass index
(BMI), HDL, sex, and smoking habits. Te tertile Te cases were older than controls (around 62 years,
cut-offs are shown in able 1. In this study, high compared to 51) and sicker. Almost 20% o the peo-

8
 

ple with CHD had diabetes, compared to under 5% in

i n the control
Why “odds reduction”
group. In addition to being heavier, people with CHD were more
instead of “risk reduction”?
likely to be smokers.
You may have noticed the word “odds”
popping up a lot in this review. The rea-
Alcohol Intake on CHD 
son stems back to this study not being
In the entire cohort, intermediate drinkers had a 35% lower odds o
a randomized trial. It didn’t actively test
CHD, compared
compared to low drinkers, regardless o genotype. High drink-
ers had a non-significant 10% lower odds compared to low drinkers. interventions on different groups o peo-
ple, and see what develops over time.
Nor did it observe participants and mea-
Tose who abstain rom alcohol are ofen ound in observational
obser vational
sure variables as time progresses, like
studies to have a higher risk o heart disease than
disease than moderate drink-
a prospective observational trial does.
ers. In this study however, both low drinkers and abstainers had
Rather, at one slice in time it estimated
increased odds compared to moderate drinkers, and low drinkers
previous alcohol intake and tested or
did not have lower odds than abstainers. Tis suggests that the ac-
CETP alleles in a group with heart dis-
tors typically attributed to abstainers that may impact heart disease
ease and a group without heart disease.
(differentt social habits, higher previous alcoholism, etc.) may not
(differen
have had a large impact in this population. Since the study was a case-control study,
it can’t use the simpler and more intui-
Genotype on CHD 
tive risk terminology. Randomized trials
Tere were no significant effects o genotype on CHD odds in
happen over time, hence you can be sure
the whole cohort, when researcher
researcherss used B1B1 as a reerence. For
that giving the intervention preceded the
B2B2, the 10% lower CHD odds was not statistically significant.
outcome, and estimate the “risk” o the
outcome based on what intervention
When the same logistic regression model was not adjusted or HDL
was given. That isn
isn’t
’t true o case-c
case-con-
on-
cholesterol,l, the B2B2 genotype was associated with a 29% lower
cholestero
trol studies such as this one, and hence
CHD odds in the whole cohort. Te act tthat
hat adjus
adjustment
tment or HDL
level reduced the effect o B2B2 on CHD odds is not surprising, as you can only measure the “odds” o the
outcome in one group versus anoth-
the CEP gene is known to be
b e involved in the regulation o HDL.
er group. However, when a disease is
rare, happening in around 10% or less
Alcohol Intake and Genotype on CHD 
o the population that’s studied, the
B2B2 homozygotes had a remarkable
remarkable decrease in CHD o
odds
dds when
odds ratio and relative risk
they were intermediate alcohol drinkers (79%) and high drinkers
will be approximate-
(52%) as compared to low drinkers. In B1 carriers (B1B1 or B1B2
ly the same, due the
genotypes), intermediate
intermediate drinkers had a 20% lower odds o CHD,
mathematical
though it was not statistically significant. B1 carriers who were high
ormulas or
drinkers had essentially the same odds as low drinkers.
each con-
verging.

9
 

B2B2 Genotype in Intermediat

Intermediate
e Drinkers 
B2B2 intermediate drinkers had a substantial and significant 59%
reduction in CHD odds compared to non-B2B2 intermediat
intermediatee drinkers.

Prevented Fraction 
Based on the authors’ calculation o prevented raction, this population
would have had around 6% more cases o CHD i the combination o
B2B2 and intermediate/high alcohol consum
consumption
ption had not existed.

While B1B1 and B1B2 genotypes weren’t associated with lower

heart disease risk, B2B2 intermediate drinkers had 79% lower
risk than low drinkers, and B2B2 high drinkers had a 52% low-
er risk. Tese numbers equate to an estimated 6% reduction in
CHD or the overall population
population..

What does the study really tell us?

Based on the results o the current study
study,, intermediate to high alco-
hol intake does not significantly reduce CHD odds in people with
B1B1 or B1B2 genotypes. In B2B2 genotypes, intermediate alcohol
intake was associated with a 79% reduction in CHD odds,
o dds, while
high alcohol intake was associated with a 52% odds reduction.

Tese results also held up to a variety o sensitivity analyses, such as

measuring alcohol intake in our cutoffs rather than three, including
or excluding adjustment or HDL and various other potential con-
ounders, or when analysis was restricted to those age 60 or older or
those who were enrolled at their first cardiac event
event..

One strength o this study was that different cut-offs o alcohol

intake were taken into account, rather than just comparing low and
high intake. Te models were adjusted or age, BMI, HDL, sex, and
smoking habits, to correct or common conounding actors. Te
authorss also tested additional actors, like leisure time physical activ-
author

ity, financial security, education levels, marital status, and diabetes

status,, but these had no effect on the results. IItt could be surmised
status
that intermediate drinkers have more healthy behaviors than the

10
 

high alcohol group, but at least or the actors men-
tioned, this was not the case. Tus, the protective
effect o B2B2 at intermediate and higher alcohol
intakes could not be explained by HDL cholesterol
or other liestyle and socioeconomic variables.
Tat being said, the cases and controls differed
widely on a variety o characteristics associated with
disease, such as age, weight, and diabetic status. It is
possible that there were other important conound-
ers that were not controlled or.
Te study also didn’t discuss potential mechanisms
that may explain the results. Previous research
Norwegians showed that HDL may not be so
in Norwegians showed
important or the protective effect o alcohol on
heart disease. However this Swedish study looked
specifically at CEP, a gene that appears to be only
involved in transer o cholesterol rom HDL to
other lipoproteins.
lipoproteins. Yet it ound that the additional
protectivee effect o CEP in intermediat
protectiv intermediatee and high
drinkers (on top o just the alcohol intake) was not
explained by HDL levels. Tis could be due to a vari-
ety o actors — perhaps a simple measurement o
HDL cholesterol is less important than the number
and type o HDL particles. As was reerenced beore,
HDL can be
b e anti-inflammat
anti-inflammatory
ory or pro-inflammat
pro-inflammatory
ory
depending on physiological context, so simply stick-
ing HDL into a regression may not ully describe the
role o HDL in the relation
relationship
ship between CEP gen-
otype and heart disease odds.
Te study results didn’t change when sensitivi-
ty analysis was perormed with different alcohol
intake cutoffs and different conounders. However
However,,

Earlier studies the cases and controls differed in a variety o

characteristics, and it’s possible that important

didn’t take into potential conounders weren

weren’t
’t controll
controlled
ed or.

account CETP The big picture

Having the B2B2 genotype didn’t have a strong pro-

genotype,
likely showedanda tective effect on its own, and neither did drinking
intermediatee or hig
intermediat highh amounts o alcohol on its own.
But combining these two actors was associated with

less subst
substantial
antial a substantial
substantial reductio
reduction
n in the odds o heart d
disease.
isease.
Te authors ocused mostly on intermediate intakes

but still protective in their discussion, but high intakes also had a sub-
stantial reduction in odds, at 52% (compa
(compared
red to 79%

effect o alcohol in intermediate drinkers). Tis may be because high

intakes come with much higher risks.

intake due to a Earlier studies didn’t take into account CEP gen-

dilutional effect otype, and likely showed a less substantial but still

11

protective effect o alcohol intake due to a dilu-
tional effect — meaning that the substantial odds
reduction in people with B2B2 likely may have been
diluted by the lack o CHD odds
o dds reduction in people
with B1B1 or B1B2 genotypes.
Tese results confirm a previous study , which
showed that men who were B2B2 homozygotes
with an alcohol intake o 50 grams a day or more
had lower myocardial inarction risk, and the risk
reduction was the strongest when the participants
drank 75 grams a day or more. In the current study,
however, the greatest risk reduction was seen at an
alcohol intake o 6.5-13.1 grams a day, significantly
lower daily intake than seen previously.
It is surprisingly easy to derive different conclusions
based on something as simple as cutoff points — the
same data can be sliced into two parts with high
hi gh
 versus low intakes, or several diff
different
erent in
intakes.
takes. And
the reerence group can also differ between studies.
In this study, the reerence group was made up o
low alcohol drinkers, rather than those who totally
abstain, as abstainers can be quite a diverse group
that includes anybody rom ormer alcoholics to
those who don’t drink or religious reasons. Some
large and well known previous studies, such as the
Harvard-run and U.S.-based Nurse’s Health Study  
and Health Proessional Follow-Up Study, suggest a
protectivee effect o the B2 allele. Te reerence group
protectiv
in that analysis, however, was abstainers rather than
those with a low alcohol intake. Women in those
studies were ound to have stronger benefit rom
the B2 allele than did men, which was not ound in
this Swedish study. Because study designs and pop-
ulations differ, it’s hard to directly compare different
CEP studies.
Tis study also had some important methodologi-
cal limitations. Subjects were queried on requency
o alcohol intake, but were not asked about portion
size. Standard portion sizes were used to calculate
daily alcohol intake. Tis could lead to inaccuracies
in daily intake data. In addition, under-reporting
o alcohol intake is common during sel-reporting,
which could skew the intermediate and high tertiles
o intake. Furthermore, CHD cases could also have
reduced their alcohol intake in response to the diag-
nosis or under-report intake i they think they are
supposed to limit intake, but this effect is likely to be
same regardless o CEP genotype. Tis is a weak-
ness o the case-control design, as a prospective study
12

  With comp
comparison
arison groups this small,
this study is just one more step in the
progression
progr ession o studies on the topic, rather
than being the inal word on alcohol and
heart disease.
that collects data beore CHD develops may be less
subject to this kind o under-reporting. It’s also pos-
sible that intermediate alcohol users could also have
generally healthier eating and liestyle habits that
were not captured in the logistic regression model.
Tis is also just one study among several on the
topic, some o which show conflicting results. Tis
paper was done on a geographically limited sample
in Sweden, so the results may not apply to those in
another region, like East Asia or Central America.
Te small sample size also limits the conclusions
that can be made rom this paper. Headlines read-
ing “Heart benefits unlikely rom
 rom alcohol” likely
won’t mention that this study only included 13 cases
who had the B2 allele and were intermediat
intermediatee alco-
hol drinkers. With comparison groups this small,
this study is just one more step in the progression o
studies on the topic, rather than being the final word
on alcohol and heart disease.
It’s important to remember that a variety o actors
could influence the effect o alcohol on heart dis-
ease, other than just genetics, such as age, sex, and
resistance. Observational
insulin resistance. Obser vational studies cannot
attribute causation or lack o causation to HDL or
LDL, no matter how strong the associations appear.
While “HDL = good, CEP = bad” is a simplistic
and inaccurate way o thinking, it is surprisingly
promote
pervasive. CEP may promote heart disease
disease in
 in some
situations, and have no effect
effect in
 in others.
Tus meta-analyses
meta-analyses o
 o CEP’s overall effect on lip-
ids and heart disease risk may inadvertently gloss
over interaction effects rom actors like alcohol
intake levels or other variables that may moderate
CEP’s effects. Te topic o heart disease, alcohol,
and HDL is a great example o how ocusing on a
single article abstract without context, even i that
abstract describes a well-cond
well-conducted
ucted meta-analysis,
can be quite misleading. A meta-analysis is only
as good as the studies it contains, and the more
complex the interactions get and the more hetero-
geneous the study designs are, the higher the risk o
a meta-analysis coming to erroneous conclusions.
A meta-analysis
meta-analysis o seven studies ound
studies ound that alcohol
did not interact with the B2B2 genotype, but it com-
pared current drinkers versus nondrinkers, which
is likely to be too crude o a compariso
comparison
n to uncover
the more complex relationship ound in this study.
13

Tis study confirms some previous evidence while
whil e
conflicting with other evidence, likely due to
dividing alcohol intakes into different levels while
using low drinkers as the reerence group rather
than abstainers. Te study is another part o the
CEP and heart disease puzzle, which is yet to be
ully solved.
Frequently asked questions
Does requency o alcohol consumption
consumption matter?
Would 49 grams o alcohol once weekly (average o
seven grams/day) be just as beneficial or CHD risk
in a B2B2 homozygote as daily alcohol intake o 7
 grams?   domized trial o different alcohol
It’s unclear rom these study results how requency
o alcohol intake affects CHD risk reduction. Since
binge drinking is not advised, the smaller amount
would be more consistent with current health guide-
lines or daily consump
consumption.
tion. Heavy drinking increases
stroke and atrial fibrillation,
the risk o some types o stroke and fibrillation,
which highlights the variety o other cardiovascular
outcomes that are related to alcohol consumption.
Is B2B2 protective or CHD when combined with
intermediate alcohol intake in both men and women?   possible conounders,
It’s unclear at this point whether the B2B2 geno-
type with intermedia
intermediate
te alcohol intake is protectiv
protectivee
against CHD in women. Te study under review
and Nurse’s Health Study may have not had a large
enough number o heart disease cases to detect
these effects. For women who are non-drinkers or
low drinkers, increasing alcohol intake to reduce
CHD risk wouldn’t necessarily be advised, giv-
en other data that suggests a higher risk o other
chronic disease, including breast cancer, linked with
intake..
alcohol intake
Context is also very important: the
additional effect o alcohol on heart dis-
ease won’t be nearly as important or a
young person without many risk actors
as it is or someone who has already
had heart disease. Te combined risks
o alcohol side effects, plus potential
risk o alcoholism, may very well out-
weigh alcohol benefits or heart health
even i one is a B2B2 carrier.
Why are studies on cardiovascular
effects o alcohol and CEP so con-
 flicting?  
It’s not really possible to do a ran-
intakes, and see what the cardio-
 vascular effects are.
are. Witho
Without
ut RC
RCs,
s,
observational studies in different
populations couple with mechanis-
tic and animal studies to orm the
evidence base.
Analyses in observational stud-
ies can use a variety o statistical
methods and control or different
conounders, which could
lead to different conclusions even
using the same data. So, even
though the largest meta-analysis
on CEP to
CEP to date shows that the
B2 allele has a statistically sig-
nificant but weak protective
effect, the result is heavi-
ly dependant on the
methods used
by the studies
it included.
14
 

Additionally, the mechanisms by which CEP may

help prevent or promote heart disease are also not
clear. In other words, this is a research area that is
  [...] there's only
still progressing, and disagreements exist within the
academic community. We will keep our collective
a small portion
eyes out or new studies on this topic.
o the population
Does my CEP allele mean that I have higher risk o
heart disease?  
or whom
Tis is the million
mi llion dollar question, or which there
is only a five cent answer: we don’t know. Although
alcohol intake is
this particular study had compelling result
resultss due
to studying a variety o alcohol intake levels and
protective against
adjustingg or a variety o variables, CEP study
adjustin
results in general are really all over the place.
CHD, and most
For example, one review ound that the effect o
B2B2 differed depending on the populatio
populationn that was all o them
unaware
unawar arethe
e that theyy
looked at.
at. In participants with a high risk o heart
disease it was protective, while in general popula-
tions it promoted
promoted heart disease! Te requency o
have a potentially
B2B2 also differed, being
b eing much less requent among
those with high risk. B2B2 sometimes could predict
protective gene.
whether a lipid-lowering drug would prevent heart
heart disease. Te effect o your genotype may be
disease, and sometimes couldn’t.
modified by your diet, habits, medications taken

Can I take a drug to modiy

modi y my CEP activity and (especially statins) and even other genes. HDL and
LDL by themselves don’t mean that much in isola-
 prevent heart
heart disease?  
tion, and neither does your CEP genotype. Some
Because increased CEP activity decreases HDL
people are able to get a portion o their genomes
levels, this became a research target or new medica-
sequences through services such as 23andme, and
tions in the 2000s. One promising drug, torcetrapib,
torcetrapib,
that may help inorm the effect o alcohol on a par-
reliably raised HDL levels by inhibiting CEP activ-
ticular individual’s heart health. Tat being said, the
ity, as well as lowering LDL. However, the trial was
evidence is nowhere near concrete, and the uncer-
terminated early  due
 due to torcetrapib causing a 25%
tainty about alcohol benefits on heart health is one
increase in cardiovascular deaths alongside a 60%
increase in deaths rom any cause. o the major takeaways on this topic.

So to repeat: we don’t know quite how CEP affects What should I know?
15
 

In short, moderate alcohol consumption may

not protect everyone equally rom heart
he art attacks.
Protective effects likely depend on genetics. Te
results o this study raise the question o whether the
recommendations regarding alcohol intake or the
prevention o CHD are too overarching. Substantial
CHD odds reduction was only seen in people who
were B2B2 homozygotes, with intermediate to high
alcohol intake. For someone giving advice about
how to prevent heart disease (like a physician, or
someone advising an older parent), keep in mind
that the evidence is still quite mixed on this topic.
In the context o public policy, the authors estimat-
ed that 6% o heart disease was prevent
prevented
ed by the
combination o B2B2 and intermediate/high alcohol
intake. Tis is not a huge amount or something that
can have several important detriments like drinking
alcohol does.

It’s important to note that only 19% o the entire

cohort in this study had the B2B2 genotype. While
the requency o this genotype in the general pop-
ulation is unknown, the beneficial effect o alcohol
intake on CHD odds would only apply to the small
segment o the population who are B2B2 homo-
zygotes with intermediate to high alcohol intake.
Perhaps in the uture, genetic testing will help us
determine our behaviors around alcohol. But or
now it seems there’s only a small portion o the pop-
ulation or whom alcohol intake is protective against
CHD, and most all o them are unaware that they
have a potentially protective gene. ◆

We’ll discuss the potentially complex relationship

between alcohol and heart disease in the private
ERD readers’ Facebook group
group.. Join us!

16
 

Type 2 diabetes:
a preventable disease
By Stephan Guyenet, Ph.D.

Three thousand and ive hundred years ago, ancient Egyptian physicians reported
excessive urination in some o their patients—a key diagnostic sign o diabetes.
The mummy o Queen Hatshepsut, a powerul pharaoh who ruled ancient Egypt
during this time period, suggests that she was obese and likely suffered rom type
2 diabetes.
diabetes. Throughout history, other royals have been posthumously diagnosed
with probable
probable type 2 diabetes,
diabetes, including the portly King Henry
Henr y VIII o England. Dia-

betes has been a scourge o the affluent or thousands o years.

Diabetes is defined as a ailure o blood glucose con-
trol, leading to excessively elevated blood glucose.
Tis ailure o blood glucose control results rom
insufficient action o the pancrea
pancreatic
tic hormone insulin,
which normally constrains blood glucose concen-
trations, both in the asting state and afer meals.
During type 1 diabetes (ormerly called juvenile-on
juvenile-on--
set diabetes), the body’s immune system attacks and
destroys insulin-secreting
insulin-secreting beta cells in the pancreas,
leading to a near-total disappearance o circulating
insulin. In type 2 diabetes (ormerly called adult-on-
set diabetes), the body’
body’ss tissues lose their sensitivity
to the insulin signal. Te pancreas compensates by
secreting more insulin, but eventually
eventually the beta cells
are unable to maintain this excessive level o insulin
secretion, insulin levels decline, and blood glucose
levels rise.

17
 

Diabetes is extremely rare in cultures

that maintain
maintain a liestyle similar to our (non-
royal) distant ancestors, yet more than a
third o modern Americans are projected to
develop diabetes at some point in lie.
Tis ailure o blood glucose control, and accompa- bled to
bled to 36 percent. Te rest o the affluent world is
nying metabolic disturbances, leads to the amiliar ollowing closely behind. Excess body at is likely the
signs and symptoms o diabetes: excessive thirst single largest contributor
contributor to the modern epidemic o
and urination, glucose in the urine, excessive hun- diabetes.
ger, weight loss, atigue, slow healing, and eventually,

 vascular disease, kidney ailure,

ailure, as well as nerve and
and Te ollowing graph illustrates the relationship
retinal damage. between body mass index (BMI; a measure o body
atness) and diabetes incidence over a five-year peri-
Te reason type 2 diabetes is no longer called od in
od  in American men:
“adult-onset diabetes” is that it now occurs in chil-
dren as well as adults. Tis trend is part o an Diabetes Risk According to BMI
increase in global diabetes risk  that
 that affects people o
nearly all age groups in all affluent nations. Diabetes
rare in
is extremely rare in cultures that maintain a liestyle
similar to our (non-royal) distant ancestors, yet
more than a third o
third o modern Americans are pr
project-
oject-
ed to develop diabetes at some point in lie. Nearly
all o these cases will be type 2 diabetes. Fortunately
Fortunately,,
the causes o diabetes are well known, so much so
that we know how to prevent the large majority o
cases. Let’s have a look.

Obesity
Over the last century
century,, but particularly the last three  
decades, Americans have bought progres
progressively
sively lon- A BMI between 18.5 and 25 is considered lean, 25
ger belts. In 1971, 15 percent o Americans were to 30 is considered ov
overweight,
erweight, and 30 or great-
obese, yet by 2009, that number had more than dou- er is considered obese. As you can see, the risk o

18

developingg diabetes increases rapidly with increas-
developin
ing BMI, and the relationship is extremely strong.
A man with a BMI greater than 35 (obese) has a
42-old greater risk o developing diabetes than a
man with a BMI below 23 (lean). I we zoom in on
the lower end o the graph, we can see that diabetes
risk increases by 50 percent even beore we leave the
lean BMI range, and more than doubles or people
who are only slightly overweight!
Diabetes Risk According to BMI
experiments show that this is more than
Countless experiments show
 just an association:
asso ciation: eexcess
xcess bo
body
dy at contri
contributes
butes
to the metabolic disturbances that lead to type 2
diabetes. Tis appears particularly
particularly true o the vis-
ceral at that surrounds the organs underneath the
abdominal wall.
Age 
Nearly all liestyle-related disorders are strongly
linked to age, and type 2 diabetes is no exception.
Among the elderly,
elderly, the yearly likelihood o b
being
eing
diagnosed with diabetes is more than 30 times great-
er than
er than among young adults. Part o this excess risk
isn’t linked to age directly, but to the act that most
people gain at, lose muscle, and become more sed-
entary with age.
age.
Physical activity 
Muscle tissue is the single largest user o glucose
in the body, and when its uel needs are high, it
increases its sensitivity to insulin to accelerate glu-
cose uptake. Because o this, physical activity causes
a rapid and proound increase in muscle insulin sen-
sitivity , leading to an increase in whole-body insulin
sensitivity. Tis increase in insulin sensitivity only
lasts a ew days,
days, so regular physical activity is essen-
tial to maintain it.
Not surprisingly, people who are more physically
active have a lower risk o developing diabetes, and
the association is substan
substantial.
tial. People who engage
in regular vigorous exercise
exercise,, or even walk regularly ,
have just over hal the diabetes risk o people who
are the most sedentary.
Genetics 
One o the most effective ways to avoid type 2 diabe-
tes is to choose your parents wisely. All o the most
common orms o diabetes, including type 2 dia-
betes, have a strong genetic component. Like most
liestyle-related disorders, diabetes is not usually
caused by a single gene variant. Rather, it’s caused by
complex interactions between many different gene
 variants
 variants an
and
d the en
environ
vironment
ment in which a per
person
son liv
lives.
es.
groups are
Possibly or genetic reasons, certain racial groups  are
at a higher risk o diabetes than others. For exam-
ple, Asians, including people o Indian descent, are
at a higher risk o developing type 2 diabetes at any
given BMI. In other words, a modestly overweight
Indian or Chinese person may have the same diabe-
tes risk as an obese Caucasian person.
person.
19
 

Te genes that influence type 2 diabetes risk

r isk tend to be
involved in the development and unction
unction o
 o the insu-
lin-secreting pancreas, and to a lesser extent, body atness.
Some o these genes may determine how well beta
b eta cells
are able to cope with the metabolic battering that accom-   diabetes risk
panies obesity and insulin resistance.

Preventing type 2 diabetes 

Some risk actors aren’t modifiable: we simply have to live
incre
increases
ases
percent
perc by
ent eeven
ven50
with them. We can’t change the genetic cards we’ve been
dealt, nor can we roll back the years o our lives that have beore we leave
elapsed. Still, the risk
r isk actors we can control are so pow-
erul that they can eliminate the large majority o type 2 the lean BMI
diabetes risk. Several randomized controlled
controlled trials have
clearly demonstrated
demonstrated this, including the massive Diabetes
Prevention
Prevention Program (DPP) trial. Tis trial reported that
range, and more
a combination o dietary weight loss and regular exercise
reduced the risk o developing diabetes by an astounding than doubles
58 percent
percent over
 over a 2.8-year period in pre-diabetic volun- or people who
teers. Several
Several  similar 
similar trials conducted
trials conducted in other countries
and other racial/ethnic groups reported almost identical are only slightly
results. Tis is one o the greatest triumphs o modern
biomedical science. overweight!
Keep in mind that these trials started with people who
were already nearly diabetic, and who didn’t lose much
weight or adhere particularly closely to the intervention.
Imagine what a lietime o healthy living could do. ◆

Stephan is an obesity researcher

researcher,, neurobiologist, and author.
In addition to his research, he enjoys synthesizing and com-
municating science or a general audience. He has a B.S. in
biochemistry (University o Virginia) and a Ph.D. in neurobiology
(University o Washington). His blog Whole Health Source is a ree
resource or anyone who loves the science o health.

20
 

Investigating
a progression
of carb and
saturated fat
intakes
Effects of step-wise increase
increasess

in dietary carbohydrate on
circulating saturated fatty acids 
acids 
and palmitoleic acid in adults
with metabolic syndrome

Introduction
Saturated at reduction has long been a major target o dietary
guidelines, although recent meta-analyses have
meta-analyses have ailed to show
an association with heart disease. Current recommendations
in the U.S. include limiting saturated at intake to less than
10% o total energy intake. However, a reduction in at intake
leads to an increase in carbohydrate intake. A conse-
typically leads to
overconsumption o
quence o overconsumption  o carbohydrates is increased de
novo lipogenesis (DNL). DNL is a process which involves the
synthesis o atty acids rom non-lipid sources, such as carbo-
hydrates or amino acids. Interestingly, even energy-balanced 
energy-balanced 
diets, and single-meal consumption
single-meal consumption o carbohydrates above
the normal oxidative capacity o the body have been shown to

21
 

increase DNL. Te percentage o ingested carbohy-

drate contributing to DNL is however quite minor
minor in
 in Saturated at is commonly targeted or reduction
people who aren’t insulin resistant and overeeding by dietary guidelines. Tis typically leads to an
on refined carbohydrate. increase in carbohydrate intake, which at high
levels may cause the body to create ats through
Te major end-product o DNL is the saturated at de novo lipogenesis. Tis study investiga
investigated
ted sev-
palmitic acid (denoted 16:0, reerring to 16 carbons eral levels o saturated at and carb intake to see
and zero double bonds), which can be desaturated how they affected plasma saturated ats and pal-
within the body to orm the monounsaturated at mitoleic acid.
palmitoleic acid (16:1). Higher blood levels o pal-
mitoleic acid have been associated with an increased
risk  o
 o metabolic syndrome and greater amount o Who and what was studied?
inflammatory
inflammatory markers. Palmitoleic has mixed eevi-
vi- Te study was an 18-week contro
controlled
lled dietary
dence however,
however, also being associated with some intervention in which the participan
participants
ts were ini-
positive biomarkers such as higher HDL
HDL and
 and greater tially ed a low-carbohyd
low-carbohydrate
rate diet that gradually
insulin sensitivity. Divergent impacts could be due shifed to a high-carbohydra
high-carbohydrate
te diet over six con-
to the effects o different liestyle actors and differ- secutive phases (rom lowest carb to highest
hig hest carb:
ent physiological conditions (such as how much o C1→C2→C3→C4→C5→C6).
DNL is rom adipose tissue versus rom the liver).
Prior to beginning the six eeding intervention
interventions,
s, the
Tis study sought to assess how incremental chang- participants were instructed to ollow a low-carbo-
es in dietary carbohydrate intake and decreases in hydrate “run-in” diet or three weeks that mimicked
saturated at intake affect plasma saturated atty the first low-carbohydrate phase, in order to initiate
acid and palmitoleic acid levels. Te study was con- metabolic adaptations to carbohydrate restriction.
ducted in adults with metabolic syndrome under Baseline and “run-in” nutrient intakes were deter-
hypocaloric condition
conditions.
s. mined with the help o three-day ood logs.

  The percentage o ingested

carbohydrate contributing to DNL is
however quite minor in those who aren’t
insulin resistant and overeeding on
reined carbohydrate.
22

All ood was provided or the subjects during the
18-week intervention. Participants picked up their
meals three to our times per week, and i the sub-
 jects could not travel
travel to pick up their ood, the
researchers arranged or delivery in order to ensure
that every subject received their ood as planned.
Blood testing was done at baseline, afer the run-in
diet, and afer each phase (beore transition to the
next diet) to determine atty acid composition and
other blood markers.
Over the entire 21-week period (intervention and
run-in), the subject’s diets were designed to produce
a 300 kcal deficit per day
day.. Resting energy expendi-
ture (REE) was estimated at baseline with indirect
calorimetry and multiplied by an activity actor to
estimate the total daily energy expenditure (DEE)
o the subjects. Protein was held constant at 1.8
grams per kilogram o ideal bodyweight.
bodyweight. As carbo-
hydrates
hydrates were increased every three weeks over the
six eeding phases, total at was decreased to main-
tain energy intake. Tus, across the entire study,
protein and calorie intake was similar. Saturated at
was also maintained, at 40% o total at intake. In
comparison,
comparison, Americans only derive around 34% o
their calories rom any kind o at, with around 13%
coming rom saturated at.
at.
Indirect calorimetry
Indirect calorimetry measures the production o
carbon dioxide and consumption o oxygen to esti-
mate heat production. This is then entered into an
equation to estimate resting energy expenditure.
Although not without error, indirect calorimetry
remains the gold standard  or measuring energy
expenditure in laboratory settings.
Only very-low and non-caloric products such as
coffee, tea, water, and diet soda were allowed to be
consumed by the participants in addition to the
provided oods. Bee, eggs, and dairy were the pri-
mary protein and at sources, with higher and lower
at versions used depending on the study phase.
Low-glycemic carbohydrates were emphasized
throughout.
Te subjects were 12 overweight and obese men and
our women with metabolic syndrome, between 30
and 66 years old (average 44.9) with BMI ranging
rom 27-50 kg/m2 (averag
(averagee 37.9). Exclusio
Exclusion
n criteria
included having diabetes, liver, kidney, or other met-
abolic or endocrine dysunction. Participants
Participants who
were physically active were asked to maintain their
activity levels while sedentary people
pe ople were asked not
to begin an exercise program.
Tis study investigated the effects o various car-
bohydrate diets on a group o overweight and
obese participants. Study participants initial-
ly ate a low-carbohydrate diet that turned into
a high-carbohydrate diet over 18 weeks, in six
phases.
What were the findings?
Energy intake (EI) across the eeding interventions
averaged about 2,500 kcal per day and protein intake
averaged
averaged about 125g per day (20% EI). As designed,
protein and energy intake remained constant over
the 18-week intervention. As seen in Figure 1, car-
bohydrate intake started at an average o 47 grams
per day (7% EI) and rose to an average o 346 grams
per day (55% EI). otal at intake started at an aver-
age o 209 grams per day (73% EI) and dropped
23

Figure 1: Carb and saturated fat int
intake
ake by study period
to an average o 80 grams per day (28% EI). Te
authorss claim that compliance was high, based on
author
 verbal communication
communication an
and
d inspection o ret
returned
urned
ood containers. Tere were no dropouts.
Both body weight and at mass (measur
(measured
ed by DXA)
showed a significant decline rom baseline to C1
(about seven kilograms aand
nd our kilograms, respec-
tively), and continued to decline throughout the
entire intervention, ultimately resulting in an aver-
age loss o about 10 kilograms o bodyweight and
eight kilograms o body at. Neither weight loss nor
at mass were significantly different between C4 and
C6, suggesting that most o the change occurred in
the first 12 weeks (run-in, C1, C2, & C3).
otal, LDL, and HDL cholesterol values were not
significantlyy altered across any o the eeding phases.
significantl
riglycerides levels dropped about 22% rom base-
line to C1. Tese levels stayed consta
constant
nt through C5
and had a significant return to baseline values at C6.
Compared to baseline, asting glucose & insulin,
Compared
HOMA-IR (measure o insulin resistance), and
systolic and diastolic blood pressure significantly
decreased at C1, but were not significantly altered
throughout the six eeding phases.
Despite saturated at intake starting at 84 grams per
day and decreasing to 32 grams per day, the propor-
tion o total saturated atty acids in blood lipids was
not significantly affected. Palmitic acid (16:0), the
predominant saturated atty acid within blood lipids,
significantlyy increased rom baseline to C1 and sig-
significantl
nificantlyy decreased rom C1 to C2, C3, C4, and C5.
nificantl
C6 was not significantly different rom C1.
Stearic acid (18:0, which is commonly ound in ani-
mal ats and cocoa) was not significantly changed
in cholesterol esters.
esters. But rom baseline to C1, it was
significantly reduced in phospholipids and also
decreased in triglycerides through the intervention
intervention,,
ending with a significant reduction in C6 relative to
C1. Phospholipid concentrations showed an oppo-
24
 

site pattern, increasing throughout the intervention and ending with

a significant increase in C6 relative to C1.

Tere was a significant reduction in total monounsaturated atty acid

concentrations rom baseline to C1 only. Similar to 18:0, as carbohy-
drate increased, plasma oleic acid (18:1) decreased in triglycerides,
but increased in phospholipids. Lipoproteins and
lipid ractions
Palmitoleic
Palmitoleic acid (16:1) was significantly reduced rom baseline to C1
in triglycerides and cholester
cholesterol
ol esters, and trended or an increase in This study looked at how much pal-
phospholipid concentrations. All these markers showed increasing mitoleic acid was contained in three
concentrations with increasing carbohydrate intake and ended the different locations in blood plas-
intervention with significantly greater concentrations o palmitoleic ma: triglycerides, phospholipids,
acid at C6 relative to C1. and cholesterol esters. Lipoproteins
shuttle lipids (such as atty acids
Tere was great individual
i ndividual variation in palmitoleic acid concentr
concentra-
a- and cholesterol) around the body.
tions during each diet phase with notable outliers. For instance, one Phospholipids orm the outer shell
subject had triglyceride concentra
concentrations
tions o palmitoleic acid rise by o lipoproteins, while cholesterol
nearly three-old rom C1 to C4 (2% to about 5.8%) and urther rise esters and triglycerides make up the
rom C4 to C6 (about 5.8% to 7%). However, another subject showed majority o the core.
no changes across the entire intervention, and another showed
reductions as carbohydrate intake increased. So the “phospholipid raction” reers
to the ats that are contained in the
phospholipids, with the same rea-
Study participants lost body weight and at over the 18-week
soning or “triglyceride raction”
intervention,
intervention, with most o the changes occurring in the first 12
and “cholesterol ester raction”.
weeks. Te blood samples researchers analyzed suggest
suggested
ed that
Sometimes these different ractions
carbohydratee intake can influence blood levels o compounds like
carbohydrat
respond the same way to diet, and
palmitoleic, stearic, and palmitic acid.
sometimes they don’t. Hence it’s
importantt to measure all o them.
importan

What does the study really tell us?

Tere are numerous studies showing associations between high-
er proportions o palmitoleic acid in blood and tissue, and adverse
health outcomes such as metabolic syndrome in adults
adults and
 and ado-

lescents, hypertriglyceridemia,
lescents, hypertriglyceridemia, type-2 diabetes
diabetes,, corona
coronary
ry heart
disease,, and prostate cancer
disease cancer.. However, since none o these studies
establish causality,
causality, it is possible that these condition
conditionss lead to high-
 25

er proportions o palmitoleic acid (or example, reductions in weight and at mass were observed,
obser ved,
palmitoleic acid may be the body’s attempt at a pro- making the causative actor difficult to isolate. And
tective response to what is being eaten) rather than there was no weight loss matched control group to
 vice-versa. With
With the mixed associations shown in account or weight loss effects. Between the lower
studies,, it is hard to know or sure what the exact
studies palmitoleic acid concentrations, the weight and at
health effects o palmitoleic acid are. loss, and the reduction in carbohydrate intake, we
cannot say which came first and which led to which.
It is also difficult to quantiy the amoun
amountt o pal-
mitoleic acid needed to increase the risk o these On the other hand, by the end o the intervention,
endpoints, as ew studies have done so. In the when carbohydrate
carbohydrate intake was similar to baseline
Physicians Health Study , one standard deviation intake (346 grams vs. 333 grams)
g rams) plasma palmi-
increase in plasma phosphol
phospholipid
ipid palmitoleic acid toleic acid levels returned to levels similar to those
concentrations was observed at baseline
associated with a despite significantly
significant 17%
higher risk o heart
  With the mixed lower weight and at
mass, strongly sug-
ailure even afer
adjustment or associations shown gesting that it was
carbohydrate intake
BMI, alcohol con-
sumption,
sumption, smoking,
in studies, it is hard to that influenced plas-
ma palmitoleic acid
exercise, and plasma
omega-3 levels.
know or sure what the levels.

In the study under

exact health effects o Te authors also
repeated the entire
review, baseline
palmitoleic
palmit oleic acid are. experiment back-

daily intake o car- wards in five

bohydrate and at
averaged 333 grams and 130 grams, respectively.
During the first phase o the interventio
intervention,
n, carbo-
hydrate intake dropped to an average o 47 grams,
while at intake rose to an average o 209 grams.
It was during this time that the most significant
changes in blood lipid atty acid concentrations
occurred, including a major reductions in palmitole-
ic acid levels. Additionally, this was when significant
improvements in insulin sensitivity, blood pres-
sure, and plasma triglyceride levels were observed.
obser ved.
However, this was also when the most significant
additional subjects
(rom high to low carbohydrate intake) and ound
that plasma palmitoleic acid responded in the exact
opposite pattern as the main study group, which
supports the idea that carbohydrate intake influ-
ences palmitoleic acid concentrations. Even so, the
overall diets were hypocaloric, and we cannot con-
clude how carbohydrate intake would influence
palmitoleic acid levels under eucaloric or hyperca-
loric contexts.
 26

Tis study provides evidence to suggest carbo-

hydrate intake influences
influences palmitoleic acid levels.
Although evidence is mixed, high levels o pal-
pa l-
mitoleic acid in the blood are associated with
metabolic syndrome, type 2 diabetes, coronary
heart disease, and other health problems. In this
study, participants experienced a drop in palmi-
toleic acid levels when they were eating low-carb
er-
meals in the first phase o the study.
entiation,
prolier-
ation, and
The big picture apoptosis) and
With 18 ull weeks’ worth o ood provided or the
has been shown to
participants, this study provided a well-controlled
increase delta-6 desat-
environment
environment in which to study the effects o diet
urase activity (first step
on palmitoleic acid. Yet despite the findings rom
in creating long-chained
this study, the relative risk rom various palmitole-
polyunsaturated atty
ic acid concentrations
concentrations in the blood remains to be
acids such as EPA, DHA, and
determined. In the previously mentioned Physicians’
arachidonic acid rom their short-
Health Study, the highest quartile had an average
chained precursors)
precursors)..
palmitoleic acid level o only 0.50%, whereas in the
current study, even when phospholipid palmitoleic
Te applicability o this study to real-lie situa-
acid concentrations were at their lowest during the
tions is uncertain. Tere were only 16 participants,
low carbohydrate phase, absolute concentrations
with widely varying BMIs, each using a particular
averaged
averaged 0.61%, putting these participants above the
 vast majority
majority o the Physicia
Physicians
ns Health SStudy
tudy subjects. dietary composition or a limited period o time.
Te effect o carbs on blood lipids was cono
conounded
unded
by the weight loss that was designed into the study,
Other blood lipid changes add urther complex-
without a weight loss control group that would help
ity to the implications o this study. For instance,
to isolate the effects o carbs. Also, a variety o differ-
increasing carbohydrate intake led to greater phos-
ent outcomes were measured. So while palmitoleic
pholipid oleic acid concentration, which in contras
contrastt
acid was emphasized in the title and study discus-
to palmitoleic acid, has been shown to attenuate the
attenuate the
sion, other important outcomes had different results.
pro-inflammatory and cytotoxic effects o excessive
saturated atty acid incorporation. Myristic acid,
which showed a reduction with carbohydrate restric- For example, outside o C1, cholesterol and blood
pressure didn’t change regardless o diet. Te sub-
tion, plays a physiologically critical role
role in
 in de novo
 jects in this study already had metabolic syndro
syndrome,
me,
ceramide synthesis (necessary or regulating cell di-
 27

so changes in gram, the Egg

things like blood   A given topic [...] Nutrition Center,
pressure and tri- and the Robert
glycerides may be can be explored in C. And Veronica
more important Atkins Foundation.
than changes in a variety o different Te unding sources
bound plasma atty did not have a say in
acids, since some o
these atty acids are
ways, and the results designing the study
or writing the man-
linked to metabolic can be interpreted by uscript. However,
syndrome (which these organizations
they already have) the study authors in are quite clearly
while blood pres- interested
interested in the
sure may have a different ways
ways as we
well.
ll. research on saturat-
more direct impact ed atty acids, thus
on their health. the variety o stud-
Also, circulating ree atty acids, which are linked to ies unded by them. Te primary investigato
investigators
rs are
metabolic and heart health, were not assessed. also noted low-carb researchers. Tis also doesn’t
mean the study is biased, but it is one thing
t hing to keep
While the total proportion o plasma saturated ats in mind when interpreting the study findings. A
didn’t differ in any o the diet phases, different indi- given topic (here, the effect o carbohydra
carbohydrate
te intake
 vidual plasma atty acids ca
can
n have diff
different
erent effects. on plasma saturated atty acids), can be explored
Palmitic acid, the predominant saturated atty acid in a variety o different ways, and the results can be
which was noted in the paper to be a predictor o interpreted by the study authors in different ways
metabolic syndrome and heart disease, was actually as well. It’s important to look at the broader context

lower in phospholipids (but not the other two lipid o literature and the nitty-gritty study details rather
ractions) rom C2-C5 than it was during the low than just take the author’s word or it.
carb C1 or high-carb C6 periods. Tis finding was
not explained, nor were changes in stearic acid and
Other plasma atty acids, such as palmitic,
oleic acid. So while a variety o atty acids were mea-
myristic, and oleic acid, may be important or
sured and reported, palmitoleic acid was the only
evaluating the health effects o different carbohy-
one ocused on in the discussion. Unortunately it
drate and at intakes. Although measured, these
was also the only ocused on in many news stories
were not a ocus o the study. Nor were more
with inaccurate headlines such
headlines such as “Heart disease and
direct predictors o heart and metabolic health,
diabetes risk linked to carbs, not at, study finds”.
such as blood pressure. Te study was unded by
dairy, bee, and low-carbohydrate organizations.
It must be noted that this study was unded by the
Dairy Research Institute, Te Bee Checkoff pro-
 28

Frequently
Frequently Asked Questions
What else influences plasma palmitoleic acid levels?
lev els?
Te current study lends support to the idea that palmitoleic acid con-
centration in the plasma is more reliant on carbohydrate intake than
at intake. However, the study was conducted under hypocaloric con-
ditions, and previous research has
research has suggested that dietary intake o
palmitoleic acid (which is rich in a ew select oods such as macadamia
nuts) does significantly influence plasma concentrations during weight
maintenance.
maintenance. Alcohol has also been suggested to reduce palmitoleic
acid concentrations, with one study  reporting
 reporting significantl
significantlyy lower le
levels
vels
in people consuming more than 100mL o ethanol consumption per
week (seven regular 12-ounce beers) compa
compared
red to people consuming
less. Tis study also ound palmitoleic acid concentra
concentrations
tions to be inde-
pendent o smoking status.

How do various biomarkers o atty acids in the body differ?

Biomarkers
Biomarkers o atty acid compositio
composition
n differ rom dietary intake, in
that biomarkers reflect
biomarkers reflect both the
t he intake and the utilization o the at-
ty acids. Because not everyone
ever yone is similar in how we absorb, transport,
and metabolize nutrients, biomarkers
biomarkers allow us to look beyond simple
dietary intake and ocus on the physiological consequences o consum-
ing certain substances. Moreover, biomarkers can provide a long-term
picture o dietary intake.

Due to the essential nature o atty acids in cell structure, assessment

can involve numerous
numerous body tissues iin n addition to blood and urine (e.g.
hair, nails, skin, breath, saliva, eces). However, measuring blood plas-
ma is the most common method. Serum triglycerides reflect dietary
intakes over the past hours to days, whereas cholesterol esters and
phospholipids
phospholipids reflect daily intakes. Only body at stores
stores (adipose
 (adipose tissue)
tend to reflect long-term dietary at consumption (e.g. years), and even
this measure can be inaccurate in people who have experienced cycles
c ycles
o at loss and gain.

How strongly
strongly is p
palmit
almitoleic
oleic acid associated with heart disease, when
compared to other biomarkers?
Although statistically significant, the strength o the relation
relationships
ships

Only body at stores tend to relect
long-term dietary at consumption
(e.g. years), and even this measure
can be inaccurate
inaccurate in people who have
experienced cycles o at loss and gain.
between palmitoleic acid and health parameter
parameterss is
low to moderate. For instance, in one study o over
3200 Chinese adults
adults,, palmitoleic acid concentrations
could only explain about 37% o the variance in
triglyceride levels and 14% o the variance in HDL-
cholesteroll levels.
cholestero
It should also be kept in mind that atty acid levels
in any biomarker represent a proportion and not an
absolute measure. Tus, greater integration o cer-
tain atty acids into the biomarker can reduce the
percentage o other atty acids without their absolute
amount changing. All o the aorementioned studies
demonstrating associations between atty acids and
health outcomes were based on percentages, mak-
ing it difficult to draw conclusions as these are not
quantifiable values. One person could have double
the amount o palmitoleic acid in serum as another
person and still have similar percentages i they also
have double the amount o blood lipid.
Tere is also evidence o seasonal variation
variationss in at-
ty acid profiles
profiles.. One early study  showed
 showed greater
proportions o saturated atty acids in the adipose
tissue o the legs and arms during summer com-
pared to winter. Tis difference was attributable to
a reduction o palmitoleic and oleic acid levels, with
29
a simultaneous increase in palmitic, myristic, and
stearic acid levels. Although these changes were in
adipose tissue and not serum biomark
biomarkers,
ers, it rais-
es the question o whether the current study could
have been influenced by seasonal
s easonal changes as its six
month duration, by necessity, spanned more than
one season. Since subtle changes in plasma atty acid
levels were tracked over increments o time, it would
be difficult to differentiate what changes were at least
partly a result o the season.
What dietary sources have a lot o palmitoleic acid
in them?  
According to the USDA nutrient database,
database, roasted
chicken skin rom the leg and thigh
thig h contains the
greatest amount
amount o palmitoleic acid with 2.8 grams
per 100 grams o ood. Bee at ollows with about
1.9 grams, then turkey skin with 1.34-1.5 grams, and
finally butter at 0.96 grams. Poultry skins contain
the most palmitoleic acid on average, ollow
ollowed
ed by
bee at and butter. Macadamia oil is a rich source,
containingg 19% palmitoleic acid.
containin
Keep in mind that palmitoleic acid is different than
trans-palmitoleic
trans-palmitoleic acid. Te latter comes rom very
limited sources, mostly red meat and dairy rom
grass-ed cows, and is not synthesized by the body.
 30

rans-
palmitoleic
acid in plasma lipids and adi-
pose tissue has been repeatedly associated
associated with revolve
better metabolic outcomes, as shown in this paper
paper   around insu-
by ERD reviewer Stephan Guyenet, Ph.D. lin-mediated glucose disposal into
both muscle and at tissue. Tis raises an interesting
 Are there
there benefits to palmito
palmitoleic
leic acid rom diet? IIn
n contradiction, with the studies demonstrating asso-
 plasma? Elsewhere?   ciations between palmitoleic atty acid levels in the
A very recent study pub
published
lished in December o 2014 blood and some adverse health outcomes such as
ound that eeding mice 300 milligrams o pure diabetes. Like certain cholesterol markers, palmitole-
palmitoleic acid per kilogram o bodyweight daily, ic acid may be more o an indicator that something
in addition to their normal diets or ten days signifi- might be physiologically wrong rather than a cause.
cantly increased glucose uptake in
uptake in at tissue through DNL is one possible cause o increased palmitoleic
increased expression o glucose-uptak
glucose-uptakee transpo
transporter
rter acid levels, and very high levels may be a marker that
4 (GLU4; necessary or insulin-stimulated
insulin-stimulated glucose something is increasing DNL to dangerous amounts
uptake into tissues). Tis was despite no changes in (such as prolonged overeating o carbohydrate, or
plasma atty acid levels. worsening glucose tolerance rom uncontrolled
diabetes, both o which can disrupt carbohydrate
Earlier studies have also ound palmitoleic acid to metabolism). Suggesting that palmitoleic acid is
enhance glucose uptake and insulin sensitivity   o
o 100% detrimental does not seem accurate given the
skeletal muscle, and reduce liver at buildup. Te complexity o evidence on the topic.
authorss o this study suggest that palmitoleic acid
author
may act as a major signaling lipid produced rom
at tissue or communication with distant organs. In
What should I know?
Tis study suggests that the presence o certain at-
obese sheep, inusion o palmitoleic acid twice dai-
ty acids in blood lipids appears to depend more
ly or 28 days preserved insulin sensitivity beore
on carbohydrate than at intake under hypocaloric
beginning an obesogenic diet, possibly through a
at..
reduction o intramuscular at conditions in overweight and obese people with
conditions
metabolic syndrome. Tere were minor – but uni-
orm – changes in a ew select atty acids, such as
It appears that the benefits o palmitoleic acid

myristic acid, oleic acid, and palmitoleic acid, but no
significant changes in total saturated and monoun-
saturated atty acid concentrations.
Tere was also inter-individual variance in the pal-
mitoleic concentration response to carbohydrate
intake, which is important given the small sample
size. While most subjects showed a positive asso-
ciation, others stayed relatively unchanged and
some showed an inverse association. Moreover,
there was greater variance as carbohydrate intake
increased. Te absolute palmitoleic concentration
 varied between about
about 2-4% in plasma triglycerides
when carbohydrate intake was lowest during C1, but
 varied between about
about 2-7% during the high-carbo-
hydrate C6 phase.
Still, the implications o changes in plasma palmitole-
ic acid levels have yet to be determined. Many studies
demonstrate associations between adverse health
outcomes and increased palmitoleic acid levels, but
reverse causality cannot be ruled out, nor differing
impacts o palmitoleic acid in different contexts. We
also do not know what influence many other dietary,
liestyle, and environmental actors have.
Rather than having obvious health implications or
differing carb levels, this study serves as additional
evidence or those eating low-carb higher saturated
at diets (and losing weight) who are apprehen-
sive about impacts on their plasma atty acids. As
is the case with cholester
cholesterol,
ol, what you eat does not
translate directly to what is floating around in your
blood. However, the lack o correlation between
dietary saturated at and plasma saturated at was
already shown by a previous paper rom
paper rom the same
research group (albeit
(albeit only the tri
triglyceride
glyceride raction
was studied).
31
It’s also important to know what this study does not
show: it doesn’t show that DNL happens at major or
dangerous rates when eating moderate carb levels,
it doesn’t show that increasing levels o carb intake
increased overall plasma saturated at, and it doesn’t
prove that low-carb diets are superior to moder-
ate carb diets or heart or metabolic health. While
weight loss decreased as carbs were added, that may
 very well be due to increased water
water weight or cha
chang-
ng-
es in compliance.
Te authors conclude that the increased propor-
tions o palmitoleic acid concentrations may signal
impaired carbohydrate
carbohydrate metabolism, yet in vitro and
animal studies have
studies have suggested that palmitoleic acid
is insulin-sensitizing. It seems prudent not to draw
health-based conclusions rom this study. Rather,
the conclusion appears to be that consumption o
carbohydrates can have an impact on plasma atty
acid proportions in overweight and obese individu-
als under hypocaloric conditions. What
Whatever
ever health
implications
implications this may lead to will require urther
testing to illuminat e. ◆
il luminate.
Te health implications o this study are unclear.
Te lack o impact o dietary saturated
saturated at on plas-
ma saturated atty acids was already shown in
previous studies. Tis study did show an effect o
carbohydratee on palmitoleic acid llevels,
carbohydrat evels, but the
relative importance o that is unknown. 
Low carb diets are nothing i not controversial. For
some evidence-based discussion on their potential
health effects, check out the ERD private Facebook
group..
group
 32

Whence the hype?

The association between ex aggerat
aggeration
ion in health
related science news and academic press
releases: retrospective observational study 
immense pressure to write more, ast, which
Introduction
encourages reliance on press releases and summa-
When it comes to health news, even though we
ries rom news agencies, universities, and other
know not to “believe the hype,” hype still happens
public relations outlets.
outlets. Tis is why it’s quite pos-
and it has an impact. Not only is the public’s
public’s use o
sible that journalists are reporting the inormatio
inormation
n
health care services influenced by the media,
media, but
they are receiving (airly) accurately, and instead
even proessionals aren’t immune. Press coverage o
medical research findings is associated with those it is the inormation sources they rely upon which
lead to media hype.
findings being cited more by other scientists
scientists..

Indeed, a previous study  o

 o press releases rom med-
Even doctors in the ER test more or certain inec-
ical centers ound that many provided exaggerated
tions that have been getting heavy press coverage.
coverage .
claims, while ew provided caveats and precautions
Since the press is so influential, it’s important that
about their claims. Similar results were ound in
the media reports medical findings accurately. But
genetics research,
cancer genetics research, where press releases ofen
it doesn’t seem like that’s happening: past research
exaggerated causal claims which were then repeat-
has shown media coverage o medical
medical and
 and nutri-
ed by the media. But the origin o the hype may go
tional research
tional research is ofen distorted. back even urther than press releases. One study  
ound that exaggerated claims ofen could be traced
But all o this doesn’t imply that the blame lies
back to the abstract o the original journal article.
with the science journalists.
j ournalists. Tey are ofen under
 33

Te purpose o the study under review was to With all o this inormation in hand, the researchers
expand upon the research above and trace the rigorously defined how hype was created at each o
source o the hype in health science news. the three stages: original journal article, press release,
and news report.

Hype is ubiquitous in health news reporting.

Tis was accomplished by creating a detailed coding
But this hype may come rom places other than
 journalists exaggeratin
exaggerating
g finding
findings.
s. Health nnews
ews system, which reviewers could use to grade each
source or the kinds o claims they were making in
impacts not only the general public, but also phy-
order to compare the hype level and notice any di-
sicians and other researchers.
erences between the research, press releases, and
news reports. o do this rigorously, the researchers
ocused on three specific areas:
Who and what was studied?
Te researchers began by searching or public-
• Advice-givin
Advice-givingg (e.g. ““Eating
Eating choco
chocolate
late ma
mayy be
ly-accessible press releases rom Russell Group
beneficial or…” or “Doctors should advise
universities (the top 20 research universities in the
patients to…”). Tis was coded at our lev-
UK) that covered research related to human health els depending on how implicit or explicit the
and were based on peer-reviewed research published
advice was.
in scientific journals in 2011. For each press release
based on published scientific research, associated
• Causal statements rom correlational research
print and online news stories were then located.
(e.g. “drinking wine might increase cancer
Broadcast news wasn’t examined in this study.
risk...” rom a study that only observed cor-
relationss between these two things). Tis was
relation
coded at seven different levels, on a continu-
Even doctors um rom statements that explicitly mentioned

in the ER test correlation,

correlation, to those that were ambiguous (e.g.
“wine linked to cancer risk”
risk”)) to those that were
explicitly implying causality (e.g. “drinking
more or certain wine increases cancer risk”).

inections that • Conclusio

Conclusions
ns p
phrased
hrased in human terms when
research was done on animals, cells, or simula-
have been tions (e.g. “a pregnant woman
woman’’s stre
stress
ss le
levels…
vels…””
concerning studies that were only done in
getting heavy rats). Tese were also coded
co ded at different levels
depending on how implicitly or explicitly the
press coverage. conclusions were stated.
 34

When coding or advice, the entire journal article, press release, or news
story was examined. Tere was a total o 213 press releases (116 o these
had news reports related to them), and 360 total news stories included.

Furthermore, or press releases and news stories, only the title and first
two sentences were coded, since news writing is ormulaic and ofen ol-

lows an “inverted pyramid” structure, where the main claims are stated
first. A sample o 182 press releases, 95 with news, and 261 news stories
were used here. Only the abstract and discussion were coded or the
original journal articles. Finally, when examining human conclusions
rom non-human studies, the main statements o 105 press releases (48
with news) and 115 news articles were coded, while only the abstract
and discussion sections o journal articles were coded.

wo other areas were also examined to get a measure o how well-jus-
tified the claims made in press releases and news articles actually were.
Tis was done by noting which press releases and news articles had
explicit caveats to their causal claims, advice, and inerence to humans
(e.g. “Te scientists
s cientists who carried out the study emphasized that they
could not say or certain...”) and explicit justification or any o these
three types o claims (e.g. “even afer taking into account the effect o
extra body weight on blood press
pressure,
ure, there was still a significant link
with sweetened drinks”). In addition to these two areas, some other
acts about the studies being reported were collected as well, such as
duration, sample size, and sources o quotes.

Te researchers explicitly
explicitly took the peer-reviewed journal article as the
baseline or the claims being made in press releases and
and news stories
concerning the research. Te original journal articles themselves were
not act-checked or examined to see
se e i they were over-hypin
over-hypingg anything.
Which is not surprising, given that the authors o this study aren’t like-
ly to be experts in dozens o biomedical and health research areas. So
hype was measured by whether press releases and news articles were
exaggerated
exaggerated compared to the original journal article.

I the original journal article itsel contained hype, this study would not
be able to detect it. But, i hype does exist in the original peer-reviewed
research (and the authors o this study think it’s likely), then any hype
 35

But the origin o the hype may go

back even urther
urther than press releases.

One study
claims ofenound
couldthat
be exaggerated
traced back to the
abstract o the original journal article.
ound in this study is likely an underestimate o 33% o press releases contained more strongly-word-
overall hype, since hype originating in the peer-re- ed claims o causation than the associated journal
 viewed scientific literature
literature is n
not
ot being taken int
into
o article warranted, and 36% o press releases inflated
account. Te researchers were also quite careul to relevance to humans rom non-human studies. So,
make sure that their coding scheme was reprod
reproduc-
uc- it seems that press releases tend to add quite a bit o
ible. Tey did this by double-coding 27% o press hype in all three areas studied.
releases and journal articles, and 21% o news sto-
ries. Tey ound that there was a 91% concordan
concordance
ce It was also ound that 36% o news reports contained
rate in coding. Te researcher
researcherss then ran simulation
simulationss more direct or explicit advice than the corre-
to make sure that a 10% discrepancy in coding sponding journal article. However, this does not
wouldn’t affect their main conclusions, and it didn’t. necessarily imply that the journalists were the ones
inflating the advice. Te odds o exaggerated advice
in news was 6.5 times higher when the press relea
release
se
Researchers examined press releases rom the top
contained exaggerated advice than when it didn’t.
20 research universities in the UK to determine
the origin o hype, or exaggeration, in media
A similar pattern held or the two other areas o
reports on new scientific findings. Exaggeratio
Exaggeration
n
hype examined. While 39% o news articles were
was determined by the presence o advice unsup-
more strongly deterministic than what was warrant-
ported by scientific evidence and inappropriate
ed by the associated journal article, the odds that the
extrapolation o evidence.
news had distorted causal statements was 19.7 times
higher i the press release also contain
contained
ed distortions.
Similarly, 47% o news articles reporting on non-hu-
What were the findings?
Te researchers ound that 40% o press releases con- man studies contained exaggerations,
exaggerations, with the odds
o these exaggerations being 56% higher i
i  the press
tain more direct or explicit advice than the journal
releases contained
contained similar distortions. As seen in
articles upon which they were based did. Similarly
Similarly,,
 36

Figure 1: How press release hype Figure 1, hype occurs in both press releases and news
correlates
correlates with news hype articles, and it’s much more likely to be present in
news articles i the press releases also contain hype.

So there is hype… but why the hype? Te authors

hypothesized that one possible motivation or exag-

gerating claims in press releases could be to increase

the chance that the press release will be picked up and
reported by the news. But when the researchers looked
at the data, they ound that there was no statistically
significant association between the percentage o press
releases that had at least one news story published on
their topic and whether or not the press release was
hyped in any o the three ways this study examined.
Also, the average number o news stories per press
release did not vary between any o the three types o
hype. So, whatever the motivation, hype in press releas-
es is not actually correlated with more press coverage.

Finally, the researchers ound that caveats about and jus-

tifications or the claims being made were quite rare in
both press releases and news stories, with at most only
17% o these claims having some sort o caveat or jus-
tification (depending on the type o claim and source).
Tere was no association observed between caveats and

 justification
 justification in p
pres
resss relea
releases
ses and
and n
news
ews u
uptak
ptake,
e, ho
howeve
weverr.
But there was a strong association between press releas-
es having caveats and justifications about their claims
and news sources having them as well.

Te results o this study show that about 40% o

press releases generated by the scientists contain
the seeds o hype: exaggeration. Moreover, news
reports based on hyped-up press releases tended

to contain more hype and exaggeration than news

reports based on press releases with cautionary
statements.
 37

What does the study Why correlation doesn’t

really tell us? necessarily equal causation
Te study tells us that both biomedical press
releases and news reports contain exagger- Correlation just means that when you see one thing occur a lot,
ations that go beyond the peer-reviewed another thing occurs a lot along with it. For instanc
instance,
e, in this study,
 journal articles upon which they are based. there is a strong correlation between news reports that have hype
Specifically, this study looked at exaggera- and press releases that have hype. Assuming the observed
obser ved correla-
tions o three kinds. Tey ound that 33% tion is actually true, then there are generally three explanations or
to 40% o press releases and 36% to 47%
47 % o why it could occur:
news reports contain stronger inerences
than were warranted by peer-reviewed • A caus
causes
es B: For example, it could be that press releases ccon-
on-
 journal articles, depending
depending on the type o taining exaggerations are indeed picked up by the media
exaggeration. and repeated.

Tis study also tells us that news reports • B causes A: This would be that exaggerated news stories

were much more likely to contain these about a piece o research lead to exaggerated press releases.
types o exaggerations i the associated press Assuming that press releases are written beore the news
releases also had them. Te odds o that stories, though, this possibility is unlikely here, since causes
news reports would contain exaggerations don’t work backward in time.
was 6.5 to 56 times higher i the associated
press release also had such exaggerations
exaggerations.. • Some thir
third
d ac
actor
tor ccauses
auses bot
both
h A an
and
d B: p
perhaps
erhaps journ
journalists
alists
are ignoring the press releases and working directly rom
While the populatio
population
n under scrutiny was the journal articles and interviews with the researchers. And
press releases rom Russell Group univer- perhaps the press releases are doing the same. Thus, the
sities in the UK, the authors explicitly state source o the hype in this case would be the the original
that they have no reason to suspect that researchers.
this group o universities differs rom other
sources o press releases in any significant There is no way to differentiate between these three possibili-
way, although this claim was not supported ties rom a correlation alone. However, one can narrow down the
or argued or in the paper. I the authors are possibilities through independent reasoning, as we did by using
correct, these results should be
b e generalizable temporal reasoning above. However, i possible, the best way to
to press reports outside o Russell Group establish causation is not through observational studies like this
universities and the news based on those one but through careully contr
controlled
olled experiments where research-
reports. ers actively intervene by changing only one variable and then

seeing what happens when compared to a control group. This is

Overall, these results are at least consistent part o the reason why randomized double-blinded, placebo con-
with the hypothesis that a lot o the hype trolled trials are the gold standard in the biomedical scienc
sciences.
es.
 38

ound in medical reporting originates not with the
 journalists reporting
reporting the news, but with the pr
press
ess
releases written by universities. But beore jumping
to conclusions about causality ourselves, an import-
ant caveat
caveat must be mentioned, one with which the
authors o this study were also well-aware: this
ound in news stories were also ound in the press
release, which also points to the media’s reliance on
press releases. Finally, study details such as sample
size and study duration were very rarely reported in
the news i the press release did not include similar
details, but was usually reported in the majority o

study was observational in news articles only i the

nature, which means that associated press release

72%
although it can provide had similar details.
inormation on correla-
tions, causality cannot be
directly inerred.
  So, while causation cannot
be definitivel
definitivelyy established

However, there are several

o quotes ound in this kind o obser-
 vational study
study,, there is
lines o reasoning to sug-
gest that press releases are
in news stories additional evidence that at
least points in the direc-
indeed a major reerence
or news articles. First, were also ound in tion that exaggeration
in press releases leads to
other retrospective and
other retrospective and
prospective
prospective studies
 studies have
the press rele
release,
ase, exaggeration in associated
news articles.
ound that press releases
influence news. Second,
which also points Finally, the authors ound
the researchers o this
study took a look at the
to the media’s no statistically signifi-
cant correlation between
dates, quotes, and areas o
reliance on press whether press releases

ocus in press releases and had any o the three types

news reports, and ound
that these three areas seem
releases. o hype examined and
whether and how much
to point to reliance on they were picked up by the
press releases by the media. Specifically
Specifically,, news stories media. Tey also ound that press releases including
were only selected i they were published within 30 caveats and justifications didn’t seem to affect news
days o the press release date. coverage. So, cautious, careully-crafed press releas-
es do not seem to be correlated with lower press
Furthermore, the authors ound that 87% o news coverage, and over-hyped press releases don’t seem
articles selected were released within one day o the to get more press, either.
publication
publication o the associated press release, leaving
 very little time or the journalists
journalists to do any ad
addi-
di-
tional independent research. Also, 72% o quotes
 39

and scientists themselves can take responsib

responsibility
ility
Tough the study was observational in nature or more accurate biomedical reporting by crafing
and did not attempt to determine i the original more careul press releases.
 journal article contained
contained h
hype,
ype, it pro
provides
vides evi-
dence to suggest press releases can significantly Journalists could in theory take more time to
influence the way news is presented to the public. independentlyy check acts and read the per
independentl pertinent
tinent
Te research also suggests that hyped-up press
releases get the same amount o coverage as press background literature, but the current journalis-
tic culture has
culture has put a lot o pressure on journalists
releases with cautionary statements, due to the to produce more material in less time than ever,
news media’s reliance on press releases. and so journalists may be orced to rely on easier
and quicker sources o inormation, such as press
releases and inormation rom news agencies. And
The big picture an entire journalistic culture can be very hard to
Media is ofen blamed or
blamed or hyping medical findings, change, particularly when it’s encouraged by a
but this study adds to a growing body o research changing industry.
which suggests that the ault does not lie solely with
 journalists. Many
Many o the exagg
exaggerations
erations ound
ound in the It may seem that there are a lot o troubling find-
news were also ound in the press releases on the ings in this study. But because the authors ound no
same topic which preceded the news reports. Since incentive to hype up press releases (since more hype
press releases are ofen crafed in collaboration with doesn’t lead to more press), they end with a hopeul
scientists,, both non-scien
scientists non-scientist
tist writers at universities message: a relatively
relatively small handul o people in uni-

  the current journalistic culture has

put a lot o pressure on journalists to
producee more material in less time than
produc
ever, and so journalists may be orced
to rely on easier and quicker sources o
inormation, such as press releases and
inormation rom news agencies.
 40

 versities can help crea

create
te better health inorma
inormation
tion was ound, that doesn’t mean that everything is
or everyone by crafing more accurate press releases exaggerated — that, itsel, would be an exaggera-
at little cost to themselves.
t hemselves. tion! In this study, a large minority o news reports
had exaggerations
exaggerations o some sort, but it was still the

Frequently
Frequently asked questions
q uestions minority.

Do scientists hype up their results in peer-reviewed

 journal articles?   One o the exaggerations these authors looked at
was extrapolating results rom non-human studies
Tis study didn’t examine hype in the original jour-
to humans. Why is this a bad thing?  
nal articles, instead using the peer-reviewed articles
Because less than 10% o animal findings can be
as a baseline. However, the authors o this study were
used clinically in human
humanss. Tere are lots o reasons
clear that they thought it was quite possible that hype
or this, rom
 rom physiological differences and di-
occurs in peer-reviewed literature, too. However,
erences between how an induced disease model
since assessing spin and hype in the scientific liter-
behaves in an animal when compared to natural-
ature takes some expertise in specialized fields, this
ly-occurring human diseases, to systematic biases
is a much harder question to assess. Although one
and methodological flaws in animal studies, but
study did find that spin o a certain sort could be
traced back to journal article abstracts, which is a overall, there are many unknowns.

good reason to read more than just the abstracts!

Animal experiments are very importan
importantt to point
out promising leads or scientists to test down the
 Just how
how pervasive is “the h
hype”? 
ype”?  
road clinically, and also help our understanding o
It’s important to emphasize that just because hype
basic biomedical science. However, it’s pretty poor
reasoning to think that i something worked once or
twice in a petri dish or a rat, it’d definitely work in a
human. A therapy may even be wildly successul

in rats, but cause terrible headaches and sui-

cidal ideation. Rats don’t really report those
side effects as effectively as humans do.

So, i the press gave me accurate inor-

mation, I’d be able to
make accurate deci-
sions, right?  
Not necessarily.
Te idea that

  It’
It’ss impo
important
rtant to
emphasize that just
because hype was
ound, that doesn’t
mean that everything
is exaggerated —
that, itsel
itsel,, would be
an exaggeration!
all people need is good inormation to draw good
conclusions is called the “deficit model” o the public
understanding o science. I people don’t have a defi-
cit o knowledge, they’d make good choices and have
a greater respect or science. Te cold hard acts are
all that’s needed. But this model suffers rom some
serious flaws.
flaws.
Knowing the acts doesn’t mean you’ll act on
them. Plus good reasoning skills, understanding o
extra-scientific culture and methods, and much more
is ofen needed. One o the motivations behind the
Examine Research Digest is to give you, the reader,
at least one more piece o the puzzle o supplement
science. We’re not just trying to spooneed you
“acts,” but also hope to help you learn how to reason
through research a little better, a little bit at a time.
Also, science is an iterative process
process,, and the popu-
lar press is not a great tool or reflecting that. I you
only read about studies in the media, science seems
41
to contradict itsel constantly, but that’s largely
because individual studies are conducted differ-
ently and some may have had errors. Te overall
weight o the evidence is much less affected
by these individual studies, and that ofen isn’t
reflected in the evening news.
What should I know?
Hype in news coverage o biomedical research
is correlated with hype ound in press releas-
es rom universities. Tis strongly implies (but,
since this was an observational study
study,, does not
definitely establish) that hype mostly starts not
with journalists, but with the university press
releases that summarize the biomedical research.
Journalistss simply report on the hype that
Journalist
already exists in the press releases. Furthermore
Furthermore,,
hyped up press releases don’t seem to draw more
news coverage, so there’s little real incentive or uni-
 versities to hype
hype up biomedical rresearch.
esearch.
Keep in mind that every link in the research and
reporting chain can have an incentive to exaggerate.
While researchers do have a degree o accountability
due to peer review, the system is imperect. Funders
can also indirectly influence
i nfluence research by selectively
unding certain studies, which researchers are well
aware o when attempting to attain grants. And this
study shows that exaggeration or inaccuracies can
be amplified urther at the reporting level. So to tru-
ly understand a research topic ofen requires not just
knowledge o the specific topic at hand, but a deep
and broad knowledge o how research works. ◆
o discuss recent examples o exaggeration in the
media and press releases, join us at the private ERD
readers’ Facebook group
group..
 42

Running on
empty: can we
chase the
away? fat
Body composition changes
associated with fasted versus

non-fasted aerobic exercise

Introduction
Te idea o asted cardio to accelerate at loss has been, or the most
part, based on a key assump
assumption:
tion: with no ood in our system, our at
stores are the go-to energy source, assuming low- to moderate-intensi-
ty cardio training.
training. Te use o at is acilitated by the low levels o liver
glycogen and insulin, and short-term studies suggest that asted cardio
does increase at oxidation over 24 hours. It stands to reason that i
done on a sustained basis, then there might be a greater amount o at
loss compared to i the training was done afer eating breakast or in
the afernoon. But is this assumption correct?
 43

exercise was at the appropriate intensity. Te hypo-

Who and what was studied?
caloric diet consisted o customized dietary plans
Tis is the first study to investigate the chronic
that induced a 500 kcal daily deficit. Food was not
effects o asted and ed cardio training on body
provided by the investigators.
composition
composition during a diet, which is likely a com-
mon situation in dieters. Previous research was
Te Mifflin-St. Jeor Equation multiplied by the
done on isocaloric or hypercaloric diets. wenty
university-going emales (average age o 22.4 years) ‘moderately active’ activity actor was used to esti-
mate total daily energy expenditur
expendituree (DEE), and
were recruited to participate in one hour o tread-
500 kcal was cut rom this value. Protein was set
mill running, three days per week, while ollowing a
at 1.8 grams per kilogram o bodyweight and at
hypocaloric diet or our weeks.
Figure 1: Outline of cardio at 25-30% total kcal.
training protocol and diet Adherence to the
All the women reported
diet was monitored
habitual aerobic exercise
through the participan
participants
ts
several days per week
sel-reporting, using
(some were off-season
MyFitnessPal.
collegiate track and
field athletes), but none
Te women were pair-
were involved in any
matched based on initial
resistance training pro-
body weight and divid-
grams. Te exclusion
ed into two groups:
criteria included injuries
FED and FASED.
and medical complica
complica--
Te number o ath-
tions in an attempt to
letes and non-athletes
ensure the women were
were evenly distributed
otherwise
other wise “healthy.”
between the groups.
All the women com-
As seen in Figure 1,
pleted the exact same
the treadmill run-
our-week diet and exer-
ning consisted o a five
cise program, with the
minute warm-up and
only difference being
cool-down at 50% o
the timing o a meal
the age-determined
replacementt shake (250
replacemen
maximal heart rate
kcal; 20 grams protein;
(MHR), separated by a
0.5 grams at; 40 grams
50 minute bout at 70% carb). Te FASED
MHR. Heart rate moni-
group consumed it
tors were used to ensure
immediately afer the
 44

Mifflin St. Jeor Equation

 M
 Meen: TEE = (10W + 6.25H - 5A - 5)*AF Women: TEE = (10W + 6.25H - 5A - 161)*AF
W = weight (kg) H = height (cm) A = age (years) AF = activity actor

Sedentary: little or no exercise x 1.2

Lightly active: light exercise/sports 1-3 days/week x1.375
Moderately active: moderate exercise/sports 3-5 days/week x1.55
Ver y active: hard exercise/sports 6-7 days/week x1.725
Extr
Extra
a act
ctiv
ive:
e: ve
very
ry hard
ard exer
exerccise/
ise/sspo
port
rtss and phys
hysical job
job x1.9
1.9

training session, while the FED group

g roup cons
consumed
umed it with trends or reductions in body
bo dy at percentage
immediately beore. Te protein was rom whey and and waist circumerence,
circumerence, all while preserving their
the carbohydrates rom maltodextrin. lean body mass. But between the groups there were
no significant differences.
Body composition was assessed with the BodPod,
which has been shown to be
b e reliable when used by Specifically, the average weight lost in the FASED
young athletic women.
women. At baseline, the only signifi- and FED groups was 1.6 vs. 1.0 kilograms, respec-
cant difference between the groups was in age, with tively, while the average at lost was 1.1 vs 0.7
the FED group being, on average, almost three years kilograms, respectively.
respectively. On the surace this may
younger (21 vs 23.8). Te averag
averagee baseline BMI was suggest a small advantage or the FASED group,
23.3 kilograms/m2, and the average baseline body with the relativ
relatively
ely small sample size or short study
at percentage was 26.3% in the FASED and 24.8% duration limiting the statistical power to detect sig-
in the FED group. nificant differences. However, these differences were
not trending towards significance.

Tis study investigated the effects o asted car-

diovascular exercise on young women ollowing a In act, the p-values averaged 0.8-0.9 or the various
body composition measurements (these p-values
hypocaloric diet. Study participant
participantss reported their
were ound through correspondence with a study
ood intake through MyFitnessPal. Researchers
author),, indicating that the difference between the
author)
conducted the one-hour treadmill run test three
groups had at least an 80% probability o being due
times a week, or the duration o the study.
to chance. Finally, the FASED group started with
a slightly greater body at percentage and at mass,
providing greater opportunity or at loss rom the
What were the results? beginning. Still, it’s possible that a larger sample size
Afer the our-week intervention, there was no
or longer duration may have changed the results.
significant difference
difference between the groups in any
measure o body composition. Both groups had
Te study also suggests that there was a dietary dis-
significant reductions
reductions in weight, BMI, and at mass,
connect in the young women during the study. Te
 45

How does the BodPod work?

The BodPod is actually the name o one o two commercially available models or body
composition
composition testing (the other being the PeaPod that is used with inants). The BodPod
works through a method called Air Displacement Plethysmography. The volume o
the body is measured indirectly by determining the volume o air that is displaced
within an enclosed chamber (the BodPod).

Once volume is known, density o the body can be calculated using this value
with the person’s weight. The density is then entered into one o several popula-
tion-speciic conversion
conversion ormulas to estimate the percent body at. The way that it
works is similar as the body-at estimating “dunk tank” (ormerly the gold standard
beore DXA), except it uses air instead o water.

DEE was around 2150 kcal, which would make the mate weight loss or different situations, whether

dietary plans based on 1650 kcal a day. At this 500 rom research or real lie.
kcal per day deficit, average at loss should happen
at around one pound per week, i we assume one
Both the ed and asted groups had lost weight by
pound o at is 3500 kcal (a rule o thumb that is not
the end o the study, with the asted group having
always accurate). However, average at loss was only
lost slightly more. However, these results were not
40-60% this amount and total weight loss was only
significantly differe
different.
nt.
50-90%, which suggests that either the women may
have consumed more kcal than they were told, or
that the DEE overshot actual requirements.
What does the study really
Either way, the women reported consuming an aver-
tell us?
age o around 1240-1280 kcal/day, which is around
Tis study clearly shows that a caloric deficit cou-
400 kcal less than they were told to eat. Similarly,
pled with moderate-in
moderate-intensity
tensity aerobic exercise results
they only consumed about 1.2 grams per kilogram
kilog ram
in weight loss. It is novel in that it also shows that
o bodyweight in protein per day, compared to
there is no difference in weight or at loss when the
the planned 1.8 grams per kilogram.
kilog ram. Whether the
cardio is perormed in a ed or asting state, at least
women were under-reporting, under-eating, or a
over the our weeks tested. Tis may seem count-
combination o both remains unknown. However,
er-intuitive
er-intuitive based on simple biological reasoning,
under-reporting
under-reporting is somewhat likely given its preva-
but makes more sense when we take into account
lence in previous weight loss studies. You can plug how adaptive and complex the human body really is.
weight, calorie intakes, and calorie expenditure into
simulator  to esti-
Dr. Kevin Hall’s NIH body weight simulator to
For instance,
instance, it has been previously demonstrated
demonstrated  
 46

that consumption o a light Mediterranean breakast
beore 36 minutes o moderate-int
moderate-intensity
ensity treadmill
running results in a significantly greater utilization
o at 12 and 24 hours afer the training session
when compared to the same exercise session done
asted. It may be prudent to view
vie w body composition
We cannot necessarily extrapolate these results to
other populations. Light-intensity asted cardio is a
common tactic or
tactic or physique sports, such as body-
building when dieting or extreme leanness, and
it would be a long-shot to generalize these results

goals and at loss over the course o days rather than
hours, since the body uses readily available uel but
then stores lefover uel over time.
  [...] evidence
suggests that
It is also worth considering the effect o the pre-car-
dio meal. Te meal replacemen
replacementt shake used in the premenopausal
current study contained 40 grams o carbohydrate
in combination with 20 grams o protein. It has women derive a
been shown that carbohydrate ingestion beore and
beore and
during moderate-intensity
moderate-intensity cardio exercise reduces greater proportion
the expression o genes involved in at metabolism.
However, it has also been shown that although car- o energy rom at
bohydrate ingestion suppressed at breakdown, the
rate o at breakdown can still be in excess
excess o
 o the during exercise
amountt o at needed or energy production, and
amoun
thus the carbohydrates may not limit at oxidation. when compared
The big picture to men, but that
Is it air to say that this study has put a nail in the
coffin or asted cardio? W
Well,
ell, not really
really.. For young
women at a healthy weight who are truly eating males have a
greater basal at
under maintenance, it is very applicable. But the
duration was airly short at our weeks and the sam-
ple size small. Even with the highly insignificant
oxidation rate.
p-values, we cannot entirely rule out the possibili-
to those individuals without similar longer-term
ty that subtle changes between the asting and ed
studies perormed to compare asted and ed exer-
groups would have taken more time or more people
evidence to
cise conditions. Tat said, there is some evidence to
to become apparent. Another conounding variable
suggest that at oxidation during exercise is inde-
is the uncontrolled dietary intakes: even though ood pendent o body at percent and relies more on
logs were collected daily, inaccurate measurements
cardiorespiratory fitness.
and misreporting could have influenced outcomes.

As or gender, evidence suggests that
suggests that premenopausal
women derive a greater proportion o energy rom
at during exercise when compared to men, but
that males have a greater basal at oxidation rate.
Tis may be due to
due to differences in sex hormones and
sympathetic nervous system responsi
responsiveness,
veness, but we
would need another study like this conducted in
men to say or sure i asted cardio would be superi-
or to ed cardio.
Finally, older men demonstrate a higher basal
respiratory quotient relative
quotient relative to younger men, sug-
gesting less basal at oxidation, but they also show
less change in response to ood intake, suggesting
Figure 2: Other research on fasted vs fed cardio
47
that over the course o the day at oxidation may be
somewhat similar between the age groups. How age
ultimately influences outcomes would require yet
another study.
Tese results are part o a growing picture o how
asted cardio impacts weight loss. Studies have
shown that different populations, diets, and types o
cardio can impact results. As seen in Figure 2, pre-
 vious research
research has been done on yo
young
ung men eating
hypercaloric and isocaloric diets, and overweight/
obese women eating their normal diet, with varying
results depending on the study.
 48

long-term study compares

compares the modalities (HII
Tough this study provides some evidence
evi dence to sug-  vs moderate steady
steady state) in d
diverse
iverse popula
populations.
tions.
gest asted cardiovascular exercise is not more Research shows that while the ofen cited post-ex-
effective or improving the rate o weight loss than ercise calorie burn rom HII isn’t
isn’t that large
l arge,, HII
ed cardio or young women, the limited nature o may still have hormonal and
hormonal  and appetite benefits
benefits that
 that
the study means more research is needed beore impact at loss.
these results can be applied to other populations.
 Are there
there other rreasons
easons to perorm asted car
cardio? 
dio?  
Circumstances
Circumstances will mediate the answer to this ques-
Frequently Asked Questions tion. For instance, asted cardio exercise has been
What about High-Intensity Interval raining (HII) shown to attenuate weight gain, enhance glucose
 or at loss? 
loss?   tolerance and insulin sensitivity, and increase gene
HII is an entirely differen
differentt beast than
t han moderate expression o enzymes involved in at oxidation
intensity cardio exercise. It requires more careul in healthy males ed a at-rich hypercaloric diet,
diet,
programming in the routine to ensure adequate whereas the same exercise protocol perormed afer
recovery and isn’t typically done on a daily basis. In consuming breakast showed weight gain with no

Aside rom actual health beneits,

some people don’t enjoy exercise with
ood in their stomach, and others have
more energy in the morning when ast
asted
ed
training is commonly perormed.
terms o actual at loss, the breakdown and utiliza- detectable improvements
improvements in glucose metabolism.
tion o at or energy is blunted at higher intensities
intensities   Aside rom actual health benefits, some people don’t
in avor o glucose, as higher intensities rely more enjoy exercise with ood in their stomach, and oth-
heavily on the anaero
anaerobic
bic energy system. ers have more energy in the morning when asted
training is commonly perormed.
Tus, between the reduced requency
requency,, shorter dura-
tion, and greater reliance on glucose or energy,
HII may not be superior to steady-state card
cardio
io
What should I know?
Tere are two main takeaways rom this study. Tis
or at loss. But we won’t know or certain until a
 49

is the first study to address the effects o asted or

ed cardio under hypocaloric conditions, and it was
shown that there were no significant differences

between asted and ed cardio in any body com-

position measurements. Te greatest limitation is
likely the study population o young healthy women,
which makes generalizing the results to men and
people o different fitness levels difficult. o make
the results o such trials even more certain or any
given population, longer trial lengths, larger sam-
ple sizes, and finding a way to standardize the diet
more would help.
help. All o these actors can make trials
much more expensive, however.

Second, this study lends urther support to the idea

that we should remain skeptical o drawing long-
term conclusions off short-term interventions. For
example: last February it was shown that measures
correlate with
o muscle protein synthesis did not correlate with
actual muscle growth ollowing a resistanc
resistancee training
routine in untrained males. Again, it would be diffi-
cult to generalize these results to experienced lifers,

but when taken in combina

combination
tion with the current
study it seems prudent to be critical o claims based
only on acute responses. Not every study applies
well to real-lie health and fitness situations. ◆

Have you done asted cardio and lost a bunch o

weight? Burned out? Somewhere in between? Let us
know your n=1 experience, and what you think o
orum..
this study in the Facebook ERD private orum
 50

Fitting into your genes:

genes:
do genetic
dietary testing-based
recommendations
work?
Disclosure of genetic information and change
in dietary intake: A randomized controlled trial
Because it’s such a new field, most o the research
Introduction on the role o genetic testing or health managemen
managementt
Science fiction is ull o stories o genetic testing
has been ocused on diseases with known genetic
and its potential to revolutio
revolutionize
nize medicine and
risk actors, such as BRCA muta
mutations,
tions, which greatly
human perormance. However, it’s not clear i the
increase breast cancer risk. As research progresses,
uturist hopes match scientific reality. Now that
more and more genes and gene variants are being
consumer genetic testing is both cheap and accessi-
identified as risk actors or disease. However, as
ble, researchers have begun to study whether or not
consumer genetic tests become more common,
these services can actually help assess and mana
manage
ge they’ve been used or a variety o lesser known expo-
health risks.
sure-disease associations based on more common
gene variations.
 51

Genetic testing will likely become more prevalent as

it becomes cheaper, and consumers without much   Genes can affect
knowledge o genetics or disease will have access to
inormation that they may not know how to handle. [...] everything rom
Genes can affect a variety o nutrition-related areas
— everything rom
 rom how we metabolize different uel how we metabolize
sources to how we absorb different nutrients.

But does it actually help people to have access to this

different uel
inormation?
inormation? Do people who receive advice based sources
sourc es to how we
on genetic tests change their habits? Te researchers
in this study assessed whether or not genetic test- absorb different
ing and subsequent dietary recommenda
recommendations
tions had
an actual effect on diet, not just in the first days or nutrients.
weeks afer being tested, but up to a year aferward.
consumer genetic testing, which meant they were
mostly young, emale, Caucasian or Asian, and
Te most established associations in genetics are
had at least an undergraduate degree. Tis is obvi-
or mutations that increase susceptibility to major
ously not a typical cohort or representative o the
diseases, such as BRCA or breast cancer. With
Canadian population as a whole.
the advent o direct-to-consumer genetic testing,
a variety o lesser known genes have been tested,
Because this study specifically assessed intake o
some o which can
c an impact nutrients.
our specific substances (caffeine, vitamin C, added
sugar,, and sodium), the inclusion criteria included
sugar
people who consumed at least 100 mg o caffeine a
Who and what was studied?
Tis study was a ollow-up to a previous study   day, at least 10% o calories rom added sugars, at
least 1500 mg o sodium per day, and no vitamin
assessing whether or not people thought genetic
C-containing supplements. Tese measures were
testing and nutrition advice based on that testing
assessed using a ood requency questionnaire that
was useul. Because
Be cause the participants o that previous
was emailed to all o the participan
participants
ts at the start o
study thought that personalized nutritional advice
the study.
based on genetics was better and more understand-
able than general nutrition advice, the researchers
Because o these requirements, only 157 out o 1639
perormed this study to assess whether or not the
participants
participants in the cohort were eligible or this tri-
participants
participants actually used the advice they were given.
al. Eligible participants were then randomized to
receive monthly dietary inormation that was either
Both studies used the same large cohort o Canadian
based on their genetic risk
r isk actors (the intervention
participants, who represented the “typical” users o
group) or general recommendations (the controls).
 52

Dietary recommendations or the intervention group min C deficiency when consuming lower
lower than
were based on whether or not the participants had recommended amounts. Tese genes code or
known variations o our genes, as seen in Figure 1: glutathionee S-transerases, which detoxiy envi-
glutathion
ronmental chemicals. Glutathione and vitamin
• CYP1A2: incr
increased
eased risk o hea
heart
rt at
attack
tack aand
nd C can protect each other rom oxidation, and
high blood pressur
pressuree when consuming >= 200 serum vitamin C levels differ depending on

mg o caffeine. Tis gene encodes proteins in GS genotypes.

the cytochrome p450 amily,
amily, which includes
enzymes that metabolize nutrients and drugs.
One variant makes you a “slow” caffeine metab-
olizer (and hence more stimulated by caffeine)
and another makes you a “ast” metabolizer.
Many different medications can impact this
enzyme, and potentially urther slow down the
breakdown o caffeine.
• GSM1 and GS1: increa
increased
sed risk o vi
vita-
ta-
• AS1R2: increase
increasedd risk o consuming excess
sugars. Tis gene codes or a taste receptor sub-
unit that can influence your sweet tooth.
• ACE: in
increased
creased risk o high blood p
pressure
ressure
when consuming excess sodium. Some people
are more sensitive to salt than others, when it
comes to blood pressure. Te ACE gene plays a
major role in determining salt sensitivity.

Figure 1: Frequency of risk alleles in intervention

inter vention group
 53

A reduction o sodium to 400 mg/

day has been estimated to prevent up to
28,000 deaths and be more effective than
using common medications to manage
high blood pressure
o assess the effects o the emailed dietary recom-
What were the findings?
mendations, the participants were sent additional
For most measures, the monthly recommendations
ood requency questionna
questionnaires
ires at three and 12
based on participan
participantt genotypes did not significantly
months afer the initial enrollment in the study.
affect dietary intakes at three or 12 months, com-
Food requency questionnaires
questionnaires are notoriously inac-
pared to the control group. Te only exception to
curate, but they are regarded as the most cost- and this was sodium intake, and in that case, the differ-
time-effective way o generally assessing the dietary
ence was only seen at 12 months.
habits o a population.

However, despite the difference between the inter-

Te subjects were also sent monthly email reminders
 vention and
and contr
control
ol groups or
or sodium intake aatt 12
o their dietary recommenda
recommendations,
tions, which are like-
months, the intervention group with the ACE gene
ly not the most effective way to modiy a person’s
 variation still
still ailed to meet the recommenda
recommendations
tions
behavior unless that person gets very ew emails and
they were provided, with only 34% meeting the
t he low-
has lots o ree time. However, it does mirror the
er recommended intake at 12 months versus 24% in
real-lie situation o ordering a genetic test rom a
testing service, and getting email as the main orm o the control group.
communication, rather than the more hands-on per-
However, the roughly 300 mg/day reduction in
sonal communication typical o many clinical trials.
sodium by the intervention group is still likely clin-
ically relevant, as recent evidence by the Institute o
Tis study used email to remind study partic- Medicine has pointed to sharp reductions in sodi-
ipants o their nutritional recommendations, um being less beneficial than
beneficial than previously thought. A
which were based off o genetic testing. Email was reduction o sodium to 400 mg/day has been
b een esti-
also used to track what the study participants mated to preven
preventt up to 28,000 deaths and
deaths and be more
ingested during the study, specifically caffeine, effective than using common medications to man-
 vitamin C, added sugar
sugar,, and sodium. age high blood pressur
pressure,
e, which shows that smaller
changes than sometimes deemed optimal can have
major impacts on a population-wide level.
 54

Tese findings are likely related to the act that most
o the participants had daily consumption values
that were within recommendations at baseline (91%
or caffeine, 86% or vitamin C, 76% or added sug-
ars, and 61% or sodium). Te act that sodium
intake significantly changed may also be related to
Tis high variance results in what statisticians reer
to as “noisy” data: measurements in which the varied
initial values make it difficult to make strong statis-
tical conclusions. For instance, i one person drinks
no coffee and another person regularly drinks our
cups a day, that’s a relatively large spread. I that first

the act that 80% o participants

participants with high-risk ACE person starts drinking one cup a day because they
 variants consumed beyon
 variants beyond d the recommended sodi- took a stressul job, and the second cuts down to
um level at baseline. For comparison, only 38% o two cups a day because o the advice they received,
participants
participants with the CYP1A2 variation consumed those are both still within the initial range o zero to
more than the recommended amounts o caffeine our cups per day, so it makes it hard to determine i
at baseline. Blood pressure may also have seemed those changes in consumption are normal or i they
a more critical health issue or some o the
t he partic- were caused by the dietary advice.
ipants than something like vitamin C deficiency,
since or better or or worse the ormer is typically Te large amounts o variability at baseline make it
associated with heart disease, while the
t he latter brings hard to generate meaningul conclusions later.
to mind scurvy
scur vy,, pirates, and colds.

Te majority o study participants did not sig-

Furthermore, as one might expect rom a study con-
nificantly change their dietary habits over the
ducted via emailed ood requency questionna
questionnaires
ires
course o the study. Sodium intake was affected
to a population with highly varied
most, though only 34% o the group met their
baseline data, the estimated intakes
recommended sodium intake by the end o the
 varied greatly among
among partici-
12-month study. Te mean reduction o the group,
pants, which ofen resulted in
while not as large as intended
intended,, would still llikely
ikely be
standard errors larger than
enough to make an important impact on a popu-
the mean values them- lation-wide level.
selves (or example, at 12
months, the change in
the control group’s
caffeine consump-
tion was -0.3 +/- 17.8
mg/day).
 55

On the other hand, one-on-one consultations with

What does the study really genetic counselors or other experts could have a
tell us? greater effect than monthly email reminders, but
Tis study intended to clariy the real-lie impact the difference between these two methods has not
o genetic testing based dietary recommendations, been compared.
and ended up showing that it had little to no effect
when it comes to b
behavioral
ehavioral modification, compa
compared
red
to traditional advice on its own. Like many studies, Based on the findings o this trial, the answer to the
question, “Does nutrition-relat
nutrition-related
ed genetic testing
this is one in which the applicability o the results affect liestyle behaviors?” is “Maybe or some select
is difficult to interpre
interpret,
t, nutrients, but likely not in
and it could be that the a consistent and reliable
significant change noted
in sodium intake at 12
  reviews [...] have ashion.”

months is not applicable

to most other nutrient
nutrientss
(i sodium and blood
pressure is deemed a
pressure
more important issue to
act on than other nutri-
ents), or the results may
be due to the vagaries
ound inorming The Big Picture
Te findings o this study
people
genetic o their
risk actors
are generally in line with
other studies
studies and
 and reviews
that have ound inorm-
does little to ing people o their genetic
risk actors does little to

o statistical variation
(the control group used
actually change actually change behavior.
It should also be noted

or comparison had

increased sodium intake
behavior. that this is still a very new
field, and there is a lack
o data on how genetic
at 12 months, making inormation can best be used or behavioral modi-
inormation
the difference between the groups larger). Tis is
fication. It’s not necessarily clear how to best deliver
especially true because there was no effect on sodi-
genetic inormation to people, and this is a major
um intake at three months, and the decrease in
conounder or any study at the moment. It could be
sodium intake at six months still didn’t lead to the
that genetic inormation can be an effective agent o
participants
participants meeting the recommended intakes.
behavioral change, but we simply don’t know how to
effectively deliver it or pair it with existing interven-
However, it’s also just generally difficult to extrapo-
tional strategies. We don’t even know who will be
late the findings o trials like this, which use samples
responsible or providing personalized recommen-
that are not representative o the general public. I
someone is highly motivated
motivated and dedicated to min- dations — will physicians work with genetic testing
companies? Will consumers mostly be interpreting
imizing their liestyle-associated risk actors, the
results on their own?
results o a genetic test may be more useul.

Tese difficulties apply to many biometric-based
intervention trials (meaning those that use a phys-
iological measurement, such as blood pressure or
blood glucose), which are actually scientifically ar
more complicated than they seem. It seems easy to
ask whether or not biometric test results affect behav-
ior in a meaningul way, but there are many other
aspects to that question that may introduce scientific
uncertainty into those types o studies. For example,
  The results might have been ver
veryy
different had the surveyed population
been more (or less) conc
concerned
optimizing their health. erned with
researchers have to assess the reliability o their tests,
assess the effectiveness o how they report the results
o those tests to participants, decide how to measure
that effectiveness, and then determine whether the
effects they see are real or somehow related to the
nature o the study cohort or statistical anomalies.
Each o these points could be enough to write a the-
sis, so the field needs to grow substantially beore
anything can be said with much scientific certainty.
In this study, these uncertainties were urther com-
pounded by the way the researchers interacted with
the study participants, which was almost entirely
through email. Most peoples’ inboxes are constant-
ly spammed by a variety o newsletters and other
inormation, and it’s very easy or a monthly email
to become a monthly auto-delete or spam older
denizen. Tere usually has to be some sort o major
56
motivation or a person to pay attention to advice,
and then there has to be even more motivation to
ollow that advice in the ace o lie’s daily stress.
Because this study didn’t specifically ask the partici-
pants i they wanted to change their habits based on
their genetic test results, that actor wasn
wasn’t
’t controlled
or at all. Te results might have been very different
had the surveyed population been more (or less)
concerned with optimizing their health. Furthermore,
a similar study on highly motivated populations, like
athletes or people recently hospitalized or health
issues, might have very different results.
Despite these issues, consumer genetic testing is still
a promising field because it offers a way to actually
act on all o the genetic inormatio
inormation
n that has accu-
mulated over the years. Without a cost-effective
way o sequencing individuals, all o the genetic
 variationss that ha
 variation have
ve previous
previously
ly been associated with
disease are relatively useless. For example, even i
we know that CYP1A2 variation is related to ca-
eine-associated
eine-associated hypertension, it does little good
unless we have a cost-effective way o testing what
 variantt a person has. Cons
 varian Consumer
umer genetic testing ma
mayy
provide this “outlet” to make gene association stud-
ies more useul by inorming large populations o
their genetic variant
variants.
s.
 57

However, it’s not clear how actionable this disclosure

is. And i it is actionable, it’s not clear i people actu-
ally care enough to change. Te impact o research
on genetics (or epigenetics or microbiomics or any
other “-ic”) is difficult topic to assess, and despite
the modern advances in sequencing and genetics,
human behavior may be the limiting step in apply-
ing findings. Te more biometric data we’re able to
find out about any given person, the more that an
age-old question applies: “How would you live i you
knew how you were going to die?”
Frequently
Frequently Asked Questions
Is genetic testing useul or general liestyle recom-
mendations?
Genetic testing may help guide liestyle choices,
but many o the tested genes (such as the ones in
this study) only show effects with intakes beyond
recommendations. So ii you adhere to general
recommendations.
recommendations,
recommendations, it may be less useul. It seems
obvious that adjusting your liestyle to address cer-
tain genetic risk actors would help reduce risk, but
that has yet to be definitively proven.

Te study doesn’t address possible negative aspects

o genetic testing. Nutrient-related tests may be less
susceptible to major negative aspects, but it’s quite
  How
possible that consumers could misinterpret a test,
and ocus on a result when the true
t rue source o their
health issues lies elsewhere (in other words, a red would you
live i you
herring). It’s even possible that someone might pin
their hopes on a nutrition-related intervention, and
stop taking a medication when they haven’t cleared
knew how
it with their doctor. Tis is a case o “knowing just
enough to hurt yoursel.” Just because you know
you were
what the MHFR gene does (a gene that regulates
homocysteine, involves B vitamins, and is a topic o
going to
much contention) doesn’t mean that it’s the source
o all your health problems. die?
Who most benefits rom genetic testing?  
Tis study was limited by the population it investi- People with amily histories o diseases may
diseases may find
gated. Even the best advice is ignored i there is no benefit rom genetic testing, but it’s also a dou-
internal motivation or change. Additional stud- ble-edged sword. Tere aren’t always preventative
ies on multiple populations such as those that are strategies available
available or all o the diseases with high-
interested in optimizing their health, or one that risk mutations,
mutations, so it may just uel a sense o atalism.
is more reflective o the general populatio
p opulation,
n, can Similarly, researchers have yet to develop reliable

shed light on the best way to deliver the results o
genetic testing and how to best structure liestyle
changes based on those results.
risk assessment models based
models based on genetic screening.
Genetic counselors are specifically trained to help
people interpret and address the results o genetic
testing and amilial risk actors.

People with amily histories o
diseases may ind beneit rom
rom gene
genetic
tic
testing, but it’s also a double-edged
sword. There aren’t always preventative
strategies available or all o the diseases
with high-risk mutations, so it may just
uel a sense o atalism.
Although we know that genetics has a proound
impact on chronic disease risk (especially rom twin
studies), we don’t know much about which specific
genes are involved. Not to mention that genes can
have complex interactions with other genes, diet, and
environment. In the case o most chronic diseases,
we don’t have the ability to look or specific polymor-
phisms and give meaningul advice on that basis.
I a client brings me a genetic test and wants to train
or eat a certain way, what should I tell him or her?
Tis is a balancing act between your proessional
opinion and your client’s opinions. And different
states vary with regards to what credentials are
needed to give different types o advice, so make
sure to look into what you are and aren’t allowed to
do. Te personalization offered by a plan that caters
to a client’s test results may enhance adherence, and
it’s unlikely to be harmul i it encourages intake o
healthy oods, but you should always thoroughly
research a given topic beore offering advice to any
clients. Most trainers lack the background in genet-
58
ics to ully understand a test result, and it’s easy to
 jump to conclusion
conclusionss that aren
aren’t
’t truly evidence-based.
Te real-lie implications o different genetic tests is
still uncertain, which is part o the reason popular
testing company 23andme was reprimanded in 2013.
Te FDA orced 23andme to stop marketing their
direct-to-consumer genetic testing service, as the
health reports provided by the company were seen
as being too close to disease diagnosis, and 23andme
was preparing to market the tests quite heavily to the
public. Tus 23andme now mostly provides raw data
without as much interpretation as they did previously.
Tis crackdown illustrated the many uncertainties
associated with genetic testing. Someone without
much knowledge o genetic epidemiology (which
is … most everybody) might
mig ht have a har
hard
d time
interpreting test results. It may not be optimal or
consumers to mostly receive raw inormation rath-
er than health reports, but it’s also important not to
“lead on” consumers with test result interpretation
 59

that may not be accurate. So it’s always a good idea
to get an expert opinion, such as rom a genetic
counselor, or important health issues that may be
impacted by genetic tests.
How are the findings o this study comparable to
tle to no change in intake or the other items studied
(caffeine, vitamin C, and added sugar intake).
While genetic testing results may enhance adherence
to diet and supplementation plans, this study only
provides some evidence that it might be possible to a

other biometric testing services such as microbio

microbiome
me
analysis?
Personalized
Personalized biometrics is a rapidly growing field,
but it’s not necessarily clear how all that extra data
can most effectively influence behavior or risk ac-
tors. Inormation
Inormation can change everything, and even
save lives. For some people, collecting tons o data
and tracking everything you do distracts rom bigger
issues that impact health. People somehow managed
to stay healthy long beore “Quantified
“Quantified Sel ” became
a buzzword.
Genetic test results rom this study are quite di-
erent than microbiome analysis. Metrics like
microbiome composition can change relatively rap-
idly in response to behavioral changes. For example,
dietary changes, or even moving to a new home, can
change gut microbiome compositions. But since not
much is known about optimal microbiome com-
small degree. It’s also unclear i this is actually spe-
cific to genetic inormation, or applies to any type o
personalization.
Furthermore, effects may depend on the specific
compound studied. For example, based on the inor-
mation ound in this study, it may be much easier
or most people to reduce sodium intake than it is
or people to reduce caffeine consumption. Future
research testing other personalized recommenda-
tions or other dietary componen
components,
ts, perhaps using
different controls,
controls, will help in developing
developing this very
new research area. ◆
Have you made changes based on genetic tests or
microbiome testing? Discuss genes and nutrition
orum on
over at the ERD private orum on Facebook.

position, microbiome analysis may serve a more

inormational role at this point, rather that to spur
direct and specific action (outside o a generally pro-
gut health liestyle). Tat being said, much o this
is speculation, as it’s a young field with a constantly
evolving research base.

What should I know?

Tis study tested how dietary recommenda
recommendations
tions based

on genetic testing results affected dietary intakes. Te

effects were relatively minor and were only seen in
one o our measures, sodium intake. Tere was lit-
 60

Combating obesity
through intermittent
fasting
Time-Restricted Feeding Is a
Preventative and Therapeutic
Intervention against Diverse
Nutritional Challenges

Introduction
Short-term asting due to religious belies has been practiced or thousands o years.
More recently, intermittent asting (IF) has been becoming more popular. Tere are
different kinds o IF, including randomly skipping a meal/meals, alternate day ast-
ing, and using time-restricted eeding (RF) windows.
 61

A RF protocol has participants consuming all o

their daily energy intake within a set window o
Who and what was studied?
Tis is a very
ver y thorough animal study that looked at
time (our hours, six hours, etc), inducing a 12-22
the effectiveness o RF against a variety o nutri-
hour daily asted window. While human trials are
tional challenges. Te researchers studied high-at,
limited, an increasing number o animal studies
high-ructose, and high-sucrose diets consumed
are showing that RF appears to be beneficial or
within nine hour, 12 hour, and 15 hour eeding
improving many chronic disease risk actors, even windows. In rodent studies the term “high-at diet”
while consuming a diet that should otherwise make
doesn’t just mean a diet high in at. No avocados,
the animal obese and diabetic.
diabetic.
no cheese, no macadamia nuts. It means a purified
high-at
high- at diet based on refined ingredients.
ingredie nts. It’s calo-
Prior to this new study, however, it was not known
rie-dense and not very healthy.
i the benefits o RF extended beyond prot
protection
ection
against high-at diets, or i RF could be protectiv
protectivee
Te study also had groups alternating between five
against excessive
excessive sugar or ructose intake. Questions
days o RF (simulating weekdays) and two days
also remained about RF’s effect on pre-existing
o ree-access to ood (weekends). In addition, they
obesity, as well as its lasting effects.
looked at both the immediate effects, as well as the
legacy effects, when the RF routines were chang
changeded
ime-restricted
ime-restricted eeding (RF) has been a part to allow long periods o unrestricted ood access.
o religious practices or thousands o years. While many o our human readers may ollow a
Recently it has captured the attention o biomedi- high-at diet with no ill-effects, it should be noted
cal research due to promising
promising research or disease that unrestricted (ad libitum) access to a high-at
prevention, mostly in animal studies. diet in mice causes obesity, insulin resistan
resistance,
ce, as well
as associated problems like dyslipidemia, hepatic
steatosis (atty liver), and elevated cholesterol.

  [...] it was not known i the beneits

o TRF extended beyond
beyond protection
against high-at
high-at diets, or i TRF could be
protective against excessive sugar or
ructose intake.

Intervention  (bolded below) all ell into one o these six groups
A total o 392 12-week-old
12 -week-old male mice were subject-
ed to a variety o eeding regimens and divided into
six main cohorts, all maintained on a 12-hour:12-
hour light:dark cycle, and ed during the dark-phase
when time-restricted. Feeding during the dark-
phase is optimal or mice, who are nocturnal. Tis
is opposite to humans, who (should) consume the
majority o their daily energy intake during the light
phase.
As seen in Figure 1, there were A LO o variables
manipulated in this trial, producing many differ-
ent diets. Reer to the figure or list to match up the
alphabet soup o different interventions to speci-
ic diet descriptions
descriptions.. Te different individual diets
Figure 1: The many, many different diets variables tested in the study diets
62
(bolded and italicized):
1. High-fructose: ed a high-ructose diet or 11
weeks either ad lib (FrA) or RF (Fr).
2. High-fa
High-fatt high-sucrose: ed a high-at high-su-
crose diet or 12 weeks either ad lib (FSA) or
RF (FS).
3. High-fat TRF and 5T2A: ed a high-at diet
or 12 weeks. Wi
With
th respect to eeding windows,
there were our different groups: either ad libi-
tum (FA), in a nine hour RF window (9hF),
12 hour RF window ( 12hF), or alternated
between five days o nine hours RF (week-
 63

days) and two days o ad lib (weekends) or 12
weeks (52A).
4. High-fat and normal chow : ed a high-at
diet ad lib (FA) or in a 15 hour RF window
(15hF), or a normal chow diet either ad lib
as mice given unlimited access, but gained hal as
much weight. Interestingly, weight gain was similar
when mice were given a high-ructose or normal
diet either ad lib or RF — suggesting that ructose
isn’t especially attening in rodents.

(NA) or in a 15 hour RF window (15hN) or

nine weeks.
The results
5. Short-term crossover (13:12): ed a high-at
diet or 25 weeks with the eeding regimen
o this study
switched or some mice (to or rom RF) mid-
way through the experiment (the FAA, F,
provides some
FA
FA, and FA groups).
additional
6. Long-term crossover (26:12): ed a high-at
diet or a normal chow diet or 38 weeks with
evidence that
the eeding regimen switched or some mice
afer 26 weeks and then maintained another 12
a calorie is
weeks (FAA, F, F
FA
A, FA eeding groups
on a high at diet and NAA, N, N
NA
A, NA
not always a
eeding groups on normal chow).
calorie, at least
What were the findings? in mice.
Te study authors provided a succinct summary o
its results:
When comparing
comparing a high at diet using nine, 12, and
“TRF protects against excessive body weight gain with-
15 hour RF, ood consumption was equivalent, but
out affecting caloric intake irrespective of diet, time
longer eeding times resulted in greater
g reater increases in
schedule, or initial body weight.
weight.” 
” 
body weight. Te 9 hour group had a 26% weight
gain, while the 15 hour group gained 43%, and
What exactly does that mean?
the group with unlimited access gained 65%. It is
important to remember that all our groups were
Bodyweight: 
consuming the same number o calories per day.
Te results o this study provide some additional
Te authors didn’t give much detail about how calo-
evidence that a calorie is not always a calorie, at least rie intake was measured, as ar as specific methods
in mice. Mice ed a high-at, high-sugar diet within
used. Rodent calorie intake can be difficult to mea-
a nine hour window consumed equivalen
equivalentt calories
sure, depending on experimental condition
conditions,
s, and

  TRF appear
appearss to be ver
veryy effective in
protecting against
against weight gain during
a range o challenges, including high-
at and high-sucrose diets, as well as
promoting weight loss and stabiliz
stabilization
ation
in preexisting diet-induced obesity.
measurementt technique is important or a trial such
measuremen
as this.
o urther test the effects o RF, the researchers
set up three crossover experiments
experiments.. When mice
were alternated between five days o nine hour RF
(weekdays) and two days o ad lib (weekends) or 12
weeks, they only had a 29% gain in body weight, as
opposed to a 61% weight gain or the FA (ad lib ed)
mice. As with the previous cohorts, ood consump
consump--
tion was the same between groups.
Another portion o the study was to determine the
longer term and lasting effects o RF. Mice were
ed a high-at diet or 25 weeks, with the eeding
regimen switched or some mice midway through
the experiment. Te mice who were started on RF
displayed rapid weight gain upon switching to ad
lib eeding, and in the end weighed as much as mice
who were always consuming
consuming ad lib (111% body-
weight gain). In contrast, the group who stayed on
RF or the entire 25 weeks only had a 51% increase
in body weight.
In the longer-term crossover study, mice were kept
64
on RF or 26 weeks and then switched to unrestrict-
ed access. As expected, mice gained weight upon
switching to ad lib eeding though their weights
stabilized at a much lower increase in body weight
(106%) than mice never on RF (157% increase in
body weight). Again, it needs to be noted that equiv-
alent calories were consumed among all groups.
o determine i RF could have benefits or mice
with pre-existing obesity, both the short and long-
term crossover studies included a group which
were switched rom ad lib to RF eeding. During
Dur ing
the 25-week study, these mice had a small drop in
body weight and maintained this weight, which
was not different rom the mice that were always
on RF. Switching mice rom ad lib high-at diet to
RF led to a 5% loss in body weight rom the time
they changed, which is impressive compared to a
25% weight gain in mice who were always allowed
ad lib access to ood. In the longer (38-week) study,
switching mice rom ad lib high-at diet to RF led
to a 12% loss in body weight rom the time they
changed,, compared to an 11% weight gain in mice
changed
who were always allowed ad lib access to ood.

RF appears to be very effective in prot
protecting
ecting
against weight gain during a range o challenges,
including high-at and high-sucrose diets, as well as
promoting weight loss and stabilization in preexist-
ing diet-induced obesity.
Body fat and inflammation:
While each experimental group had comparable
lean mass, it was the differences in at mass that
made up the differences seen in total body weight.
Compared to ad lib ed, mice on RF had reduc-
tions in body at o 62% (high-at, high-sucrose)
and 26% (high-ructose). Increasing the length o
the RF windows (rom nine hour to 12 hour to
15 hour) led to an increase in percenta
percentage
ge o body
at, but even the 15 hour window was protective
compared to ad lib consumption. Mice in the
5:2 group were also protected rom excessive at
accumulation
accumulation (48% less body at than ad lib eed-
ing). Mice on a normal diet that were ed ad lib
but transerred to RF has 55% less at than mice
maintained on ad lib diets.
Reduced inflammation was also seen in mice on
RF (by looking at mRNA levels o pro-inflamma-
tory cytokines
c ytokines NF-a, IL-1b, and pro-inflammato
pro-inflammatory
ry
chemokine Ccl8/Mcp2).
Blood glucose regulation:
When mice were ed a ‘normal’ diet, RF did not
offer any extra advantage over ad lib when looking
at asting glucose levels. However, on a high-
high-at
at or
high-sugar diet, RF reduced asting glucose lev-
els compared with ALF. Fasting insulin levels were
reduced in all RF groups ed a high-at diet. In
the crossover studies, insulin levels were nearly 5
times lower in mice maintained on RF compared
with ad lib, while groups who had some exposure
65
to ad lib and RF had asting insulin in between
those two groups.
A glucose tolerance test was also perormed and
all o the mice, except the mice eating normal ood,
showed improved glucose tolerance compared to
their ad lib counterparts. Te crossover studies also
revealed that RF can reverse prior glucose intoler-
ance as a result o diet-induced obesity.
Lipids:
Tis study suggests RF is protective or a lot o
things, and blood lipids were no exception. Liver
triglyceride levels were reduced in all mice on a RF
high-at, high-sugar diet compared with their ad-lib
ed counterparts. In addition, switching mice to RF
prevented
prevented urther hepatic triglyceride accumulation
in ad lib ed mice, suggesting RF as a possible clin-
ical tool against atty liver disease. Likewise, serum
triglycerides were also normalized when mice were
switched rom ad lib high-at to RF. On a lower at
diet however,
however, ser
serum
um triglycerides were unchanged
between RF and ad lib.
Cholesteroll levels, both absolut
Cholestero absolutee levels as well as the
daily rhythmic variation, can also improve on RF.
Mice ed either high-at or high-sugar
hig h-sugar diets on RF
had significantly lower serum cholesterol levels than
those on ad lib eeding. It should be noted that there
are substantial differences in cholesterol metab-
olism between humans and mice. For example,
rodents have very low LDL compared to humans
due to more rapid clearance by LDL receptors. Most
serum cholesterol is carried by HDL, and they are
extremely resistant to atherosclerosis because o that.
In experiments knocking out their LDL receptors,
their lipids become more human-like and then they
become more likely to develop atheroscler
atherosclerosis.
osis. Tis

would suggest that changes in serum cholestero
cholesteroll in
mice may be caused by different mechanisms than
could occur in humans
humans..
Additional benefits: Mice on RF showed better
coordination skills and improvements in physical
endurance tests (nearly double the endurance per-
ormance o the ad lib group), which were not the
result o greater muscle
muscle strength, fiber type or gly-
cogen storage, but likely rom improved metabolic
responses to mobilizing energy stores. Enzymes that
regulate glycogen synthesis and gluconeogenesis
(creation o glucose rom non-carbohydrate sourc-
es) were affected by RF, as was the anabolic insulin/
Akt and catabolic AMPK pathways, and a variety o
cycling amino acid metabolites resulting in more
avorable daily patterns.
Mice that were subjected to a restricted eeding
window gained less at and had better blood lipid
profiles than mice that were allowed to as much as
they wanted, even though all the mice consumed
the same amount o calories.
What does the study really
tell us?
Although we are not mice, these models can be
extremely valuable or understanding the mecha-
nisms behind metabolic health and disease states.
Tis study offers a great deal more inormation than
previous RF studies, because they used not only a
high-at diet, but also high-sucrose, high-ructose,
 various RF windows
windows (nine, 12, an
and
d 15 hour), five
day RF, two day ad lib eeding, and longer term (25
and 38 week) crossover studies to determine lasting
effects o RF.
66
Tis study confirms that, in animals, RF can be
an effective treatment or a variety o disease states
such as obesity, diabetes, high cholesterol, atty liver,
and circadian dysunction,
dysunction, in the absence o a cal-
orie deficit. While the ‘bad’
‘ bad’ diets showed the more
dramatic effects o RF, mice ed normal-chow still
showed better body composition
composition..
Although we are
not mice, these
models can be
extremely valuable
or understanding
the mechanisms
behind metabolic
health and disease
states.
Keep in mind that these mice were always ed during
the dark phase. It has been previously shown in
rodents that the ood timing relative to the light:dark
cycle is very important (even in RF). Some mice
have experienced an 18-19% increases in body weight
when eating the same number o calories during the
“wrong” (light) phase, compared to the dark phase
(normal eating times). o extrapolate this to huma
humans
ns
we need to think o the opposite, and pick our nine to
12 hour windows during the daytime.
 67

light:dark cycle would be basically impossible.

The big picture However, while rodent studies can control more
RF could promote wider adherence than conven-
actors between groups, randomization in humans
tional dieting methods, because the emphasis is on
could help to minimize variation between groups.
the timing o ood intake and not on calorie count-
Costs o testing so many interventions over a long
ing. Tere will certainly be
b e uture studies which
study may be prohibitive. In addition, all o the
can investigate
investigate the mechanisms o action, as well
as a large-scale randomized control trial (RC) in animals were sacrificed in order to be ully stud-
ied, which doesn’t go over very well with human
humans.
participants
participants or the institu
institutional
tional review boards who
approve the study.
Until uture studies are done, we can only guess at
how much o these results will translate to humans.
Tere are a ew more aspects o the study that would
Tere are a ew existing human studies which use
differ in humans. Humans normally consume ood
RF, but nothing on the scale that is needed. Tere
in a time-restricted manner. I you have breakast at
is a great deal o research on Ramadan, which ea-
7:00 a.m. and dinner at 6:00 p.m., that is an 11 or 12
tures a month-long RF window. However, these
hour eeding window. Most people don’t eat in the
meals aren’t aligned to circadian rhythms, occur- middle o the night, but when mice are ed high at
ring at night, instead o during the day. Tere are a
diets, they ofen do eat in the middle o the night
nig ht
ew other recent studies which show reduced daily
(much more ofen than when they’re ed healthy
energy intake, and either improved or no-change
chow). Also, in this study, the major effects o RF
in insulin action.
only maniested when animals were ed unhealthy,
purified diets. RF didn’t have as much o an effect
rying a RF window or yoursel could offer ben-
when mice were eating a healthy diet. Humans typ-
efits with very little downside. However, i you are
ically aren’t ed purified oils and refined oods in
prone to hypoglycemia, consult your doctor
do ctor beore
high amounts over the course o months in studies.
trying this.
t his. Please se
seee the FAQ or additional pre-
cautions.
cautions. ime will tell i a RC can show similar
Lastly, in humans, as ar as we currently know, a
results to what this study has shown, but it is indeed
calorie really is a calorie (as long as it’s absorbed
 very promising rom
rom a number o an
angles
gles and or a
into circulation).
circulation). Tere is no other ood pr
property
operty
diverse populatio
population.
n.
or diet characteristic known to substan
substantially
tially impact
adiposity in humans. Mice are not like that — many
FAQs studies have shown calorie-independent effects o
Tis is great! Could a similar study be done in diet characteristics
characteristi cs on adiposity. Mice are able to
humans? modiy energy expenditur
expendituree more readily and to a
Not the whole study with all the measuremen
measurements,
ts, larger extent than humans.
but some parts could be done and have been done.
o have complete control o the amount and type Does it matter when my window is?  
o ood eaten or 38 weeks while
whi le controlling the It’s hard to say with certainty, but rom the existing
 68

literature
literature the b
best
est answer would be to keep most those who want to bulk up or are still growing (high
o your ood intake in the light phase, since animal school and college athletes),
athletes), or people in very hig
highh
data seems to suggest that eating in the “wrong”  volume and
and high intensity tra
training
ining phases, suc
such
h as
(sleep) phase leads to greater weight gain. Tis cyclists or triathletes. People with advanced liver
would change with the seasons, during the summer disease should speak to their doctor beore practic-
the times can be more flexible, but during winter it ing RF. While RF may be protective against atty

may be best to keep the window earlier in the day. liver, a bedtime snack is typically recommended or
liver,
Remember, our body’s clock is the light:dark cycle, people with advanced liver disease.
and not the time displayed on your watch.

What I should know?

Do I have to skip breakast?
Tis animal study suggests that keeping ood intake
growing  body  o
No! A growing  o research suggests a high-pro-
within a nine to 12 hour daily eeding window can
tein breakast may have avorable effects on appetite
be beneficial in a number o different ways. Tese
control. Additionally, glucose tolerance is better in
results become more apparent when consuming a
the morning,
morning, compared with later in the day due to
poor diet that would otherwise lead to obesity and
circadian variation. Tis impaired evening glucose
tolerance is likely due to decreases in both insulin metabolic dysunction, but benefits also extend to
animals eating an otherwise ‘norma
‘normal’l’ diet.
secretion and insulin sensitivity . When consider-
ing the circadian variation in glucose
g lucose tolerance, a
Te natural question that arises is: Should I try time
roughly 9:00 a.m. - 6:00 p.m. window may work well,
restricted eeding? We don’t know how well the ben-
although a variety o individual actors play into
efits shown in this study applies to humans, given
exact timing.
the physiological and environmental differences
rom rodents, but restricting ood to moderate
Can my window change rom day to day?
daily eeding windows is unlikely to do harm or
Tis study showed benefits rom five days o RF,
most people. Access to ood at all times o the day,
ollowed by two days o ad lib, suggesting that there during all times o the year is not necessary or most
is some flexibility or the eating p
phase,
hase, and you
humans, and trying a different eating pattern may
do not necessarily need to ollow a rigid daily time
produce quite beneficial results without having to
window. Keeping most o your ood intake to the
micromanage different parts o your diet.◆
light phase, but moving it up or back by a ew hours
depending on the day could probably still be okay.
o discuss all the different possible types o intermit-
Is there anyone who should NO try this? tent asting protocols, and their impacts on humans,
Yes, extended asting windows can be a stress on the check out our private Facebook group
group or
 or ERD
body. Ofen a ‘good’ stress, but someone who is deal- readers.
ing with a lot o other stressors in their lie should
approach this diet conservatively. Also, athletes
should probably not get too ambitio
ambitious,
us, particularly
 69

How does a lifetime

of marijuana use
affect the brain?
Long-term effects of
marijuana on the brain

Introduction
Marijuana use is popular due to the psychoactive effects o gamma-9-tetrahydro-
Marijuana gamma-9-tetrahydro-
cannabinol (HC). It’s known that marijuana has a multitude o effects on the
brain, as seen in Figure 1, but understan
understanding
ding the exact effects can b
bee a complicat-
ed scientific process.

Within
Within the brain, there are two major types o cells: neurons and glial cells,
pictured in Figure 2. Neurons are the cells that respond to and carry electri-
cal signals, while glial cells provid
providee support and protection to the neuron
neurons.
s.
Networks o cells orm either gray matter or white matter tissue. Gray matter and

white matter both are made up o neurons (and glia), but the gray matter is the
cell bodies that contain the nucleus and most o the cellular machinery
machinery,, while
the white matter are the thin “telephone lines” between neurons, wrapped in a
myelin sheath (which is a structure that is part o a specialized class o glial cells).
 70

FIGURE 1: Brain areas affected

affected by marijuana
Gray matter is involved
involved in everything
the brain does — such as processing
and cognition activities, including
i ncluding
decision making and sel-control. It’s
grey due to the lack o myelin, the
insulating sheath around the outside

o some brain cells. White matter

physically connects and coordinates
communication between differ-
ent regions o the brain by carrying
electrical impulses rom neuron to
neuron. Te myelin is white in color,
which distinguishes it visually rom
the gray matter.

Tis particular study investigat

investigated
ed
several specific regions o the brain,
in addition to types o brain tissue.
Te orceps major
major and
 and orceps minor 
minor 
are two regions o white matter in
the brain. Te orceps major con-
Figure 2: Brain terminology
nects the occipital lobes within the
cerebral cortex and the orceps minor
connects the rontal lobes o the
cerebral cortex. Within the rontal
lobes, there is a region o the brain
called the orbitor
orbitorontal
ontal network. Tis
network is made up or our lobes:
the lef and right orbitoron
orbitorontal
tal cortex
(OFC) and the lef and right tem-
poral lobes. Te primary unction 
unction 
o this region o the brain is deci-
sion-making, specifically the analysis
o the possible rewards o a decision.
Tis region o the brain displays high
levels o activation during addic-
tion-seeking behaviors like heavy
drug use.
use. Specifically or marijuana
 71

use, the OFC also has a high concentration o canna-

Who and what was studied?
binoid 1 (CB1) receptors, the receptor that binds HC.
Tis was an observational (non-interventional)
study that compared 48 regular cannabis users
Previous studies looking at the effects o marijua-
with 62 non-users o similar age and sex. A regu-
na have conflicting results. Some studies showed
lar user was defined as someone who sel-reported
increases
increases in
 in tissue volumes in certain regions o the
using marijuana at least our times a week and took
decreases in
brain, others have showed decreases  in the same a drug test (via a urine sample) that was positive
areas o the brain, and still others have shown no
or HC at enrollment. A non-user was defined as
effects.. Tis
effects
someone who
could be due
to differ-   Rese
Research
arch on marijuana sel-reported no
marijuana use
ences in the
study popu- impacts on brain unction and had a nega-
tive drug test at
lations, either
in regard to has had conlicting results. enrollment.

the partici-
pant characteristics,
characteristics, or iin
n the level o marijuana use. All study partic-
ipants took an IQ test at the beginning o the study.
Other potential conounding variables include only
Marijuana users were also assessed or behavioral
investigating a particular age range or duration o
issues related to possible marijuana dependency
marijuana use, the enrollment o subjects who used
through the Marijuana Problem Survey (MPS). Te
other substances along with marijuana, or designing
MPS asks participants to identiy and rate problem
the study to only look at a single region o the brain.
areas such as missing work, conflicts with amily
and significant others, or legal issues as a result o
Tis study attempted to overcome those limitations
their marijuana use.
by looking at a broad range o participant ages,
evaluating a subset o participants who exclusively Once enrolled, study participants underwent three
used marijuana, and using several different types
different MRI scans to assess different structural and
o MRI scans to evaluate a number o actors in the
unctional aspects o the brain:
brain globally.

• a high resolution visual MRI scan to quantiy the

Marijuana affects a variety o brain regions, amount o gray matter in the participant’s brain
including a region called the orbitorontal cortex
(OFC) involved in decision-making. Research on • a unctio
unctional
nal MR
MRII (MRI) scan to det
determine
ermine
marijuana impacts on brain unction has had con- unctional connectivity,
connectivity, or how much blood

flicting results. flow occurred in differen

differentt brain regions

• a diffusion tensor imaging (D

(DI)
I) scan to deter
deter--
 72

mine structural connectivity, or how much white matter

exists between different regions o the brain, and how orga -
nized the white matter is

Because marijuana is ofen used along with other substances, the

researchers separated out a subset o the cannabis users into “exclusively

cannabis users,” who had no sel-reported use o alcohol or tobacco. Tis

allowed the researchers to determine i any o the structural and unc-
tional changes seen in the MRI scans were due to cannabis use alone.

Tree different MRI scans were used to assess gray matter, white
matter, and connectivity between brain regions. Users o marijuana
only, rather than users o marijuana and other substances, were also
tested separately.

What were the findings?

Te researchers noted a statistically significant difference in the
IQ scores o the cannabis users, compared to the non-users.
Mean IQ scores were approximately five points lower among
cannabis
cannabis users, even though the educational levels between the
two groups were similar. However, urther statistical analysis
did not indicate a direct causal link rom marijuana use, to
neural abnormalities that may arise rom its use, to lowered
IQ. A number o alternative actors, such as genetics and envi-
ronment, could be involved in this causal pathway, or even
explain the difference between the groups themselves. Tough
untested, it’s also possible that a lower IQ could increase the
likelihood that someone will become a regular marijuan
marijuanaa user,
rather than lowered IQ being an effect o heavy marijuan
marijuanaa use.

Te average IQ score in the cannabis users group was

approximately five points lower than in the non-users group,

though statistical analysis could not confirm i this was a

cause or an effect o marijuana use, or due to other actors.
 73

Te subgroup o participants who were exclusively

cannabis users had similar MRI results as the can-
nabis users group as a whole. Tis indicated that
[...] the
any changes in brain structure and unction were
correlated with cannabis use and not the use o
tissue that was
other substan
substances.
ces.
there showed
Te study compared the high-resolution MRI scans
o the cannabis users with the non-users. A signifi-
increased
cantly lower volume o gray matter was seen in the
right middle orbitorontal and lef superior orbito-
unctional
rontal regions o the brains o cannabis users. Tis
structural difference, however, cannot be deter-
connectivity
mined to be a result o cannabis use, since this is a
correlational
correlational finding, thus suggesting it is also pos-
at rest when
sible that subjects with lower volume o gray matter
are more likely to become chronic users.
compared to
the non-users
When the researchers looked at MRI scans to
detect brain unction in the gray matter, they
ound that the cannabis users group had more
brains.
unctional connectivity in the our nodes o the
OFC regions o the brain, as measured by blood structure and level o organization o the axons in
flow in the gray matter. Even though there was less the brain tissue.
gray matter in the cannabis users’ brains, the tissue
that was there showed increased unctional con- One indicator in the DI scan is ractional
 ractional anisot-
nectivity at rest when compared to the non-users ropy (FA). FA reflects the density and myelination
brains. Te researchers believe this to be
b e a compen- o the axons, and is measured on a scale rom
 rom 0 to 1.
satory mechanism to maintain brain unction even
e ven A higher FA means that water diffusion is restricted
as brain volume decreased. to a single direction, implying that the local water
is inside long thin fibers (axons) as opposed to lit-
Next, the researchers looked at the structural con- tle lumps (cell bodies). Tis is indicative o a more
nectivity o the same regions o the brain, measuring fibrous and organized region. A lower FA means that
white matter in the orceps minor region that inter- water diffusion is less restricted and indicates
i ndicates a less
connects the different areas o the OFC. Te DI organized and axonally-dense region o the brain.
scan uses
scan uses magnetic resonance to measure the diffu-
sion, or passive movement o water through regions Another indicator
indicator in the DI scan is radial diffusiv-
o the brain, providing inormation about the micro- ity (RD). Tis is a measurement o diffusion along
 74

two axes, which is decreased in more mature white matter brain tissue
and increased when cells in white matter become demyelina
demyelinated.
ted.

Te DI scan showed that cannabis users as a group had higher FA

measurements in the orceps minor, but not in the orceps major
region. Tis effect was localized to the orceps minor region, as no sta-

tistically significant effects were seen in the orceps major region within
the occipital lobes. In summary
summary,, this region looked more organized and
more myelinated in cannabis users. A more organized neuronal net-
work with more myelination can result in more efficient transmission
o electrical signals in the
t he existing brain tissue. Since this increased
organization was seen in the orceps minor, which connects the rontal
lobes, this could possibly translate to compensatory improvements in
short-term memory, attention, and motivation.

Te MRI scans were correlated with the intake data, and some interesting
patterns emerged. While the cannabis-using group as a whole had high-
er FA and lower RD indicators in the DI scans, when the researchers
looked at how long each individual participant had been a cannabis user,
they ound that there were highly significant correlations between the
DI scan indicators and lietime duration o cannabis use. Tis makes
the causal, rather than just correlational explana
explanation
tion a bit more likely.

Tere was an initial improvement in these scores (increased FA and

decreased RD) over the first several years o cannabis use, ollowed by
an overall decline in these indicato
indicators
rs as usage be
became
came more long term.
Te participants who had been using cannabis the longest had indica-
tions that their white matter was less organized and more demyelinated
than participants
participants who had only been using cannabis or a ew years.

Additionally, the researchers ound that the unctional connectivity

measured in the MRI scans showed strong correlations with a partic-
ipant’s score on the MPS. Te less unctional connectivity seen on the
MRI scan, particularly in the lef temporal cortex region o the brain,
the more likely a participant was to have behavioral and social problems
related to their cannabis use, as indicated by higher MPS scores. Within
the exclusive cannabis use group, there was also a statistically signifi-
 cant correlation between gray matter volume in the OFC and scores on
There used to be misconceptions that
the brain was a relatively unplastic organ
afer adulthood, but more and more
research
research is inding that both positive and
negative changes can take place [...]
the MPS: as the amount o gray matter in the brain
decreased, the MPS scores increased.
Te MRI also showed that activity in the OFC cor-
related with the age that the
t he participant began using
marijuana - the earlier the participant had become a
regular user, the greater unctional connectivity was
measured.
Participants
Participants who had only been using cannabis
or several years showed higher indicators o axo-
nal organization and brain tissue maturation, but
these measurements
measurements declined as cannabis usage
became more long-term.
The big picture
Te biggest challenge when interpr
i nterpreting
eting this study is
attempting to determine cause and effect out o all
the correlations in the data. Are people who have
higher IQs or more gray matter less likely to become
a chronic marijuana user, or is the marijuana use
causing that physical change? At least one study  has
 has
suggests that children who had smaller OFC vol-
umes were more likely to become marijuana users in
their teens.
75
Since this study (and many others) only looks at a
single point in time using MRI scans, it’s not pos-
sible to determine which variable is the cause and
which is the effect. Te researchers who conducted
this study noted that longitudinal studies would be
needed to ully
 ully understand this. However
However,, since pe
peo-
o-
ple who had used marijuana o
orr a longer period o
time had strong
stronger
er associations with brain structure
and unction, that does boost the likelihood o the
causal explanation. Some mechanistic plausibility
also exists, as both animal and human studies have
ound potentially neurotoxic effects
effects o
 o marijuana.
Whether the reductions in gray matter are a cause
or effect o cannabis use, the brain is a complicated
organ, and appears to attempt to compensate or
reduced tissue volume by increasing the unctional
connectivity o the present tissue. Tis may be why
the new marijuan
marijuanaa users had more organized white
matter and higher resting activity on the MRI. Over
time, however,
however, these indicato
indicators
rs declined as addi-
tional structural changes took place in response to
cannabis
cannabis use.
Tis is really the most interesting, and perhaps
slightly unexpected, part o the study
study.. Tere used to
 76

be misconception
misconceptionss that
the brain was a relative-
ly unplastic organ afer
adulthood, but more and
more research is finding
that both positive and
levels o HC might have
greater long term effects
  [...] the balance on this region o the brain.
o positive and Other constituents o
marijuana have effects on

negative changes can

take place due to a num- negative impacts the brain as well (although
they are much less psy-
ber o different external rom marijuana is choactive or not at all),
effects. Te initial effects and strains vary in the
o marijuana on the brain hard to evaluate. ratio o HC to these oth-
seem to be the brain’s way er constituents.
o attempting to maintain regular unction in the
ace o tissue loss. Te earlier a subject began using How significant are the differences in IQ seen
marijuana,, the more prono
marijuana pronounced
unced these initial com- between the two groups?
pensatory
pensatory effects were, since the brain still is more Te five-point difference between the marijuana users
neuroplastic
neuroplastic (building and wiring connections) and non-users is within one standard deviation. Both
through adolescence and into the early 20s. Starting groups actually scored ‘above average’ (106 and 111,
to use marijuana later in lie would not be as efficient respectively). Tis is comparable to the difference
at taking advantage o this increased neuroplastic seen between adults with some college education but
stage. no degree, and adults with a college degree. Keep in
mind though, the actual education levels were basi-
Te correlation between gray matter volume and cally the same between the two groups.
scores on the MPS is unsurprising, given that the
unction o gray matter is decision making and Do randomized trials on marijuana show
show impacts
sel-control. A person with a lowered capacity to
sel-control. on cognition?
make decisions and exercise sel-control is more likely Tere have been a variety o cognition-related ran-
to have issues with social and psychological activities. domized trials done on chronic marijuan
marijuanaa users,
with results typically showing some
s ome impairment.

Frequently
Frequently asked questions
q uestions For example, one trial ound that marijuana acutely
decreased blood flow in attention-related areas o
Could different strains o marijuana have different
the brain.
brain. Prospective observational
obser vational studies have
effects on the brain?
also ound potential brain-related harm rom mar-
It’s definitely possible. It’s not known i HC speci-
ijuana use. One ound that persistent marijuana
ically is the cause o any structural and unctional
use over the course o years was associated with
changes in the brain, but the OFC is a region o the increased cognitive problem
problemss and general decline in
brain that has a high level o cannab
cannabinoid
inoid 1 recep-
neuropsychological
neuropsychological unctioning.
tors, which bind HC. Strains that have higher
 77

On the flip side, there has been an increasing amount o research on

potential pain-related and other benefits o marijuana use, which involves
other parts o the nervous system and brain. Systematic reviews o ran-
domized trials have ound benefit or 
or neuropathic
neuropathic pain
pain,, and potential
or helping with other kinds o pain such as that rom fibromyalgia and

rheumatoid arthritis. Given that ew i any randomized trials test chronic
rheumatoid
effects over a period o years, the balance o positive and negative impacts
rom marijuana is hard to evaluate.

What should I know?

Several different types o MRI scans ound differences between the brains
o long-term marijuana users and non-users. Regular marijuana users had
lower volumes o gray matter, but also had indicators o increased connec-
tivity and unctional connectivity in several
s everal regions o the brain. People

who had only used marijuan

marijuanaa or several years had more connectivity in
their brain’s white matter tracts, but these actors declined as use became
more long term. Marijuana users also had slightly lower IQs, but it was
not clear i this was due to marijuana use or other actors.

Examine.com has compiled a wealth o scientific knowledge

knowledge on the effects
page. ◆
o cannabis “supplementation” on their Marijuana page.

o discuss the impact o marijuana on brain unction, but without the

generalities and annoyances that ofen come with debating such a topic on
the web, visit the ERD private orum 
orum on Facebook.
 78

A mouse’s
microbiome
may cause its
brain to leak
influences 
The gut microbiota influences 
blood-brain
blood-brain barrier
permeability in mice

Introduction
Your gut has much more to do with your brain than just the influ-
ence it has when you’re passing by the donut shop. Since our guts
take in all the uel we need rom the outside world, and our brain
brainss
are necessary or navigating the outside world, the two need some
way to communicate with each other. Tis method o communica-
tion between the gut and brain is called the “gut-brain
“gut-brain axis.
axis.”

In the 1970s,
1970s, the molecular mechanism through which the gut and
brain communicated
communicated was beginning to be understood. Several pro-
teins and peptides (which are made out o the same building blocks

Figure 1: What is the blood-brain barrier?
as protein, but are smaller) were discovered, that
were both produced by and affected the gut and
brain. But a problem arose: communication between
the brain and gut largely involved big molecules like
li ke
proteins and peptides. How could such large mole-
cules get across the blood-brain barrier (BBB)?
In order to answer this question, we first need to
understand what the BBB is and what its unction
is. Te BBB exists to make sure that compounds in
the blood don’t necessarily enter the brain, and that
your brain keeps whatever nutrients it needs. You can
79
check out the details in Figure 1. Tere, you can see
that the BBB keeps out large molecules, while being a
little more loose about certain types o smaller mole-
cules, while also selectively letting other molecules in.
Te BBB is mainly made up o endothelial cells (the
kind o cells that line the inside o blood vessels) that
are tightly knit together by “tight junctions,” which
are composed o several types o proteins. Te pur-
pose o tight junctions is to make sure substances
don’t accidently slip in between cells. wo o the pro-
teins which make up tight junctions are claudin and
 80

occludin, which are discussed later in the review. help us) may play a role in this training. Te gut
microbiome is known to contribute to other areas o
Since the BBB is made up o cells that are tightly-wo- mammalian development, such as gut development 
development 
 ven together
together,, it is vvery
ery hard o
orr larg
larger
er molecule
moleculess to (including aspects o how it unctions as a barrier)
pass rom the bloodstream into the brain. I they do and even other aspects o brain development
development.. So it’s
get through, they do so either selectively through not a ar leap to suspect that the gut microbiome

transporters or because the BBB is damaged, leaky, or may influence the BBB as well. Te authors o this
otherwise compromised in some way. paper set out to test exactly this hypothesis in mice.

Why does the BBB exist? Well, the brain is a pret-

Gut bacteria have been observed to influence brain
ty important organ, and so it’s wise to be selective
development, as well as influencing gut integrity.
about what gets into and out o the brain. For
Tis inspired researchers to examine whether or
instance, i you get an inection, the BBB will hope-
not gut bacteria can also influence the integrity o
ully stop the inection rom reaching the brain
brain.. But
the blood-brain barrier (BBB), which selectively
the BBB also plays a major role in the developing
allows molecules in and out o the brain.
brain as well. One way it does so is by helping to
regulate the environment o the growing brain
brain to
 to
create an optimum environment or development. It
also helps protect the growing brain rom toxic out-
Who and what was studied?
wo types o mice were looked at in this study:
side influences, such as bacteria colonizing the gut
pathogen-ree and germ-ree. Pathogen-ree mice
o newborns, during the so-called critical period o
had normal gut microbiota and lacked any kind o
brain development.
development. In summary, i your brain is a
bacteria that normally causes disease in mice. Germ-
club, the BBB is like the bouncer: it does its best to
ree mice, however, had no bacteria in the gut at all.
stick to the list o approved molecules, and doesn’t
hesitate to bar entry to the less desirable clubgoers.
Several questions were addressed in this study:
Like bouncers, the BBB also has to be trained. Te
• Does the gu
gutt microb
microbiota
iota o
o the m
mother
other aff
affect
ect
authors o this study hypothesized that gut micro-
the BBB o the developing mouse etus? Tis
biota (the bacteria that normally live in our gut and
was addressed by comparing how well an anti-
usually don’t cause any problems, and in act can

  [...] the brain is a pretty important organ,

and so it’
it’s
s wise to be selective
gets into and out o the brain. about what
 81

body could cross the BBB o etuses in mothers normal gut flora). Tis difference in permeabili-
who were either pathogen- or germ-ree. ty was observed toward the later part o gestation.
Te increase in permeability was associated with
• Do gut micr
microbiota
obiota aaffect
ffect the permea
permeability
bility o decreased expression o occludin, one o the main
the BBB in adult mice? o address this ques- proteinss which make up tight junctions.
protein
tion, both germ- and pathogen-r
pathogen-ree
ee mice were
compared and the permeability o the BBB How could microbes living in the mother’s gut affect
was tested. the permeability o the etus’ BBB? Te authors did
not determine a specific mechanism, but speculate
• What do the tight junctions o these mice look that lower BBB permeability would be beneficial or
like? Te composition and appearance o tight etuses whose mothers have normal gut flora. Te
 junctionss wer
 junction weree examin
examined
ed to see i tthere
here we
were
re di- reason is that maternal gut microbes may require
erences between germ- and pathogen-ree mice. higher nutritional demands in late pregnancy, which
would require tighter BBB permeability so that these
• Can cha
changing
nging the microbiota change the per- metabolic demands don’t impose a cost on the grow-
meability o the BBB? Germ-ree mice were ing brain o the etus.
colonized by normal gut flora to become patho-
gen-ree mice. Te permeability o the BBB as But what about in adult mice? Do the gut micro-
well as tight junctions were examined beore biota affect the permeability o the BBB in them?
and afer colonization to see i changing the gut Again, the answer is yes. Te permeability o the
microbiome could actually change the BBB. BBB in adult mice was greater in germ-ree mice
than in pathogen-ree mice in three separat
s eparatee tests
o BBB permeability. Te researchers made sure that
Tis study compared the permeability o the BBB
this difference was not due to increased blood ves-
in germ-ree mice (mice who had no gut bacteria
sel penetration o the brain, since a higher density
at all) to pathogen
p athogen-ree
-ree mice (who had normal gut
bacteria, but none that normally cause disease in o blood vessels would mean there are is a higher
chance that something would penetrate randomly,
mice).
 just like how buying
buying several lott
lottery
ery tickets increases
your chance o winning.

What were the findings? Te differences in permeability could be account-

Te researchers ound that the gut microbiota o the
ed or by the differences in tight junctions ound in
mother mouse can affect the BBB permeability o
adult germ-ree versus just pathogen-ree mice. Te
the developing mouse etus. Specifically, BBB per-
tight junctions o germ-ree mice had lower lev-
meability o etuses with germ-ree mothers (who
els o two major tight junction proteins, occludin
had no significant levels o bacteria in their gut) was and claudin-5, as compared to pathogen-ree mice.
greater than the BBB permeability o mice whose
Also, the more leaky tight junctions o germ-ree
mothers were merely pathogen-ree (and so had
 mice looked more diffuse and disorganized under

82

the microscope when compared to those o patho-
gen-ree mice.
Finally,, would taking germ-ree mice with no gut
Finally
flora and colonizing them with normal flora change
their leaky BBB? It turns out that, again, the answer
is yes. Colonizing the gut o germ-ree mice led to a
less permeable BBB, along with increased expres
expression
sion
o occludin and claudin-5.
which produces acetate and propionate), and also by
 just eeding an SCF
SCFA
A (butyrat
(butyrate)
e) to germ-ree mice,
and then measuring the effects on the BBB. Tey
ound decreased BBB permeability in mice colo-
nized by either kind o bacterium, as well as in mice
who were ed butyrate. In act, the BBB became just
as impermeable in these mice as in mice who were
pathogen-ree.

Te researchers ound
ound that the BBB was lless
ess leaky
But how do the gut bacteria “talk” to the BBB and
when mice had normal gut flora. Completely
affect its permeability? One possible mechanism is
germ-ree mice had a more permeable BBB. Also,
 via short-chain atty acids (S
(SCF
CFAs)
As) which are syn
syn--
germ-ree mice subsequently colonized with
thesized specifically by bacteria. SCFAs are known
SCFAs
normal flora experienced decreased BBB permea-
to affect the permeability o the gut
gut,, so perhaps they
bility. Interestingly, the BBB o mouse etuses was
affect the permeability o the BBB too. Te research-
ers tested this hypothesis in two ways: by colonizing also leakier i their mother was germ-ree, than
i she was just pathogen-ree. Te permeab
permeability
ility o
germ-ree mice with single strains o bacteria that
the BBB may be in part
par t affected by short-chain
produce SCFAs (Clostridium
(Clostridium tyrobutyricum,
tyrobutyricum, which
atty acids produced by normal gut bacteria,
produces butyrate, and Bacteroides thetaiotaomicron,
thetaiotaomicron,
which travel through the bloodstream and ulti-
mately help make the BBB less permeable.

  The permeability
What does the study really

o thewas
mice BBBgreater
in adult tell us?
Tis study tells us that having healthy gut microbes
in mice makes their BBB less permeable. Specifically,
Specifically,

in germ-ree mice having normal gut flora in mother mice helps their
etal mice develop
de velop a more impermeable BBB. Also,

than in pathogen- adult mice with normal flora have a less permeable
BBB, which is associated with lower expression o

ree mice in three certain tight junction proteins and more disorga-
nized and diffuse-looking tight junctions.

separate tests o Finally,, these effects o normal gut microbes on the

Finally

BBB permeability
permeability.. BBB seem to be causal, since colonizing germ-ree
 mice with normal flora seems to decrease BBB per-

83

The science o our gut microbiot

microbiota
a is still
young, and we can say little or certain, but
it’s starting to look like the microbes in our
gut could play many roles in maint
maintaining
aining a
healthy body.
meability. Te mechanism or these changes are not
known, but may have something to do with SCFA
The big picture
Te science o our gut microbiota
mi crobiota is still young, and
production o normal gut flora, as both monoco-
we can say little or certain, but it’s starting to look
lonizing germ-ree mice with SCFA-producing
like the microbes in our gut could play many roles
bacterial strains and just eeding germ-ree mice an
in maintaining a healthy body. Te gut microbiota
SCFA decreases BBB permeability.
may affect cognition
cognition,, and could possibly play a role
in obesity . Te impact o probiotics on athletic per-
Let’s also briefly mention what this study does not
ormance has also been examined
examined.. Fecal transplants
tell us. It does not tell us anything about humans
to re-establish healthy gut flora are also starting to
taking probiotics. It does not say much about the
be tested as treatments or disease, as a recent study
consequences o BBB permeability and whether it’s
on ulcerative colitis
colitis in
 in children demonstrat
demonstrates.
es.
“good” or “bad.” It also doesn’t spell out the mech-
anism through which gut microbiota influence the
Tis study adds one more piece to this puzzle by
BBB’s permeability, although it does contribute evi- showing that gut microbiota play a role in creating
dence or a possible answer. Any inerences beyond
a less leaky blood-brain barrier in mice, and per-
the main point o this paper (that healthy gut
haps more surprisingly, that the gut microbiome o
microbes in mice makes their BBB less permeable
a mother can, at least in mice, influence the BBB o
compared to having no gut flora at all) would be
the etus. Gut microbiota influence goes ar!
unreliable guesswork.

However, the results o animal studies don’t nec-

Tis study tells us that germ-ree mice have more essarily hold or humans. It is worth noting that
permeable BBBs than pathogen
pathogen-ree
-ree mice. Any independentt research has been done on the micro-
independen

additional extrapolation is speculation.

specul ation. biome o prenatal and neonatal humans. For
instance, the gut microbiome is markedly differ-
ent in
ent  in babies that were born via natural methods,
 as compared to babies born by c-section. Te gut

84

microbiome o c-section inants was much less diverse, or up to

six months afer birth.
  [...] gut
Tis is the typical time period where the human diet begins to vary,
naturally contributing to microbiome diversity. Te immune sys- microbiota
tem goes through a lot o development during inancy
inancy,, and research
shows that microbe changes via birth method may influence the
shows that
immune system in the long term. Tus, there is emerging evidence play a role in
that the gut microbiome o inants can vary, and that this may have creating a less
long-term effects.
leaky blood-
Tis work also demonstrates the importance o animal research.
Animal studies can be quite important because work that is brain barrier
impractical, cost-prohibitive, or unethical on humans may not be
deemed so or animals. For instance, this study could not have in mice, and
been done on humans. Te work that would go into maintaining
a sterile or monocolonized human GI tract is neither practical or
ethical.
perhaps more
surprisingly,
Te science o how gut microbes affect health is still young.
Tis study adds one more piece to the puzzle by suggesting the
that the gut
possibility that gut microbes can affect distant organs, like the
brain, in unintuitive ways.
microbiome
o a mother
Frequently
Frequently asked questions
q uestions can, at least
What about the brain itsel — does a sterile environment or an
extended period o time impair neurogenesis?
Tis study did not reveal any differences in neurons them-
in mice,
selves between germ- and pathogen-ree mice, and the long
term effects o gut sterility on the brain itsel were not studied.
inluence the
However, there is emerging research that shows the gut may
development , so an impact on neu-
influence local neuronal development,
BBB o the
rogenesis is at least theoretically possible. Also, one study  in
 in
mice ound that impaired neurogenesis due to stress could be
improved by administering a probiotic. etus.
 85

What else influences BBB permeability?
While a lack o gut microbes is related to increased
permeability, there are many other actors
actors that
 that can
dictate BBB permeability. o list just a couple o
examples,, increased ammonia content
examples ammonia content in the blood
during liver ailure and high levels o c-reactive pro-
tein are associated with BBB permeability
tein are permeability.. Also, some
rodent data shows downregulation
downregulation in ttight
ight junction
protein expression in mice that consume ethanol.
ethanol.
What is a good way to keep my gut microbiome
diverse?
Check out ERD # 2’s article “O Mice and Guts” or
more inormation on microbiome diversity.
blood-brain barrier (BBB) than mice with complete-
ly sterile guts, contain
containing
ing no bacteria o any kind.
Interestingly, the gut microbiome o pregnant mice
affected the BBB o their unborn etuses. Mice with
healthy gut flora had etuses whose BBB was less
permeable than those who were germ-ree.
Finally,, colonizing a germ-ree mouse with a healthy
Finally
microbiome induces changes
changes in their BBB, making
it less permeable. One mechanism by which gut
microbiota may influence the BBB is through
t hrough pro-
ducing o short-chain atty acids, which enter the
bloodstream and eventually impact the BBB, making
it less permeable. ◆

What’s your gut instinct about all this? Let us

What
Tis should
journal I know?
article was published in Science know your thoughts on the ERD private orumorum on
 on
ranslational Medicine,
Medicine, a well-regar
well-regarded
ded interdisci- Facebook.
plinary journal whose purpose is to help connect
basic science research with eventual clinical app
appli-
li-
cations. Te study results may not be directly
applicable in terms o directing human interven-
tions, but it connects two massively important areas
— the microbiome and the gut-brain axis.

Te blood-brain barrier can both be be damaged

by disease and a cause o disease, and getting cer-
tain potentially important medications through the
blood-brain barrier is an active area o research.
From this study we’ve learned that the microbiome
may impact the blood-brain barrier, and human
studies on the topic are likely to ollow. In short,
this is a research area that could pay dividends or
human health in the near uture.

Specifically,, this article showed that mice with

Specifically
normal, healthy gut microbiomes had a less leaky
 86

s e esearc e
 

Stuart M. Phillips,
Ph.D., FACN, FACSM

Stuart
Stuart graduated
 graduated with a Ph.D. rom the University o Waterloo in Human Physi-
ology in 1995. He joined McMaster in 1999 and is now a Proessor in the Depart-
ment o Kinesiology and
Kinesiology and an Adjunct Proessor in the School o Medicine at Mc-
Master University.

Stuart is a ellow o the American College o Sports Medicine (ACSM) and the
American College o Nutrition (ACN). His research is ocused on the impact o
nutrition and exercise on human skeletal muscle protein turnover. As well he
is keenly interested in diet and exercise-induced changes in body composition

and the inluence o all o the aorementioned in aging persons. An enthusias-

tic and energetic group o graduate students and research ellows are the true
heart o Dr. Phillips’ more than 180 publications, 120 public scientiic presenta-
tions, and continuing enthusiasm or science and research.

You just published a review of protein for weight
loss. Te somewhat new and mildly controversial
“protein leverage” hypothesis is mentioned. What’s
 your take on
on that?
Te Simpson and Raubenheimer leverage hypothe-
applicability in humans – see this study  or
 or a good
review and meta-regression
meta-regression o sorts. In short, I think
the whole protein-seeking behaviour espoused by
these two researchers and their teams is viable, but it
seems that most people’s ‘natural setpoint’ or pro-
tein intake is around 15-17% o total energy intake
 sis is an interesting one, and one that may have some
intakes and it does appear that leverages energy
rom other macronutrients – at and carbs – and can
controll energy intake.
contro
Tat was a pretty complex topic or an opening ques-
tion. aking a step back — what originally brought
 you into protein
protein re
research? 
search?   and I think it’s been a good career ‘decision’ (I’ll pre-
Ha ha, yup deep end question to begin! I did an
undergraduate
undergraduate degree at McMas
McMaster
ter in Biochemistry
and had an epiphany o sorts when I took a nutri-
tional biochemistry course in my ourth year
year.. It
changed the way I thought about things! I signed
up or a master’s and studied protein and endur-
paper -
ance athletes - my first paper  - rom then on I was
When you ind something you’re
passionatee about and enjoy what you’re
passionat
doing, it’s rarer to actually call it ‘work’.
hooked! I loved learning, I loved school, I loved
nutrition,, and I liked (yes only liked) research … it
nutrition
wasn’t until the last stages o my Ph.D. that I truly
grew to love research. It just evolved rom there.
My passion or protein was honed while I was a
postdoc in Bob Wole’s lab rom 1995 to 1998. I
learned so much there and had a ton o un at the
same time! When you find something you’re pas-
sionate about and enjoy what you’re doing, it’s rarer
to actually call it ‘work’. So I ‘work’ at what I love
doing and so ar it’s been a lot o un!
I you could go back in time, is there someth
something
ing else
… you have to consciously move toward higher
87
I don’t think I’d have done another macronutrient
– at or carbohydrate – because that has so much to
do with obesity and diabetes, which, or whatever
reason, didn’t interest me. With protein, there’s less
people who study it and we’re a smaller group. I’m
happy to have chosen protein, it still intrigues me
tend I chose to study protein … I think it chose me).
But change anything? No, not likely … I am ocused
more now on some paradigm-cha
paradigm-changing
nging work, which
I always think is important. I’ve still got a ew more
years lef and my ocus will likely gradually
g radually chang
changee
in the next ew years.
 Although
 Although you’
you’ve
ve been extremel
extremelyy successul in
research, the struggle o climbing the ivory tower is
tough. What’s your view o the “publish or perish”
environment o academia?  
Tere are many types o ‘ivory towers’, just as there
are many types o occupations that people with a
Ph.D. pursue. In act, most people
pe ople (i.e., more than
80%) with a Ph.D. don’t go into academia. Within
academia there are ivory towers that value teaching
more than research, but at McMaster the climb is
based on research first, and teaching second. So it
does orce a publish or don’t do well atmosphere,
though you can still publish and perish in my view!
With that pressure it does create a stress, since so
  you might
might have ocused on? much o your work is evaluated by nameless and

88

aceless people who figuratively ‘hold the cards’ and commodities, which some automatically
automatically assume
can turn you one way or another. It has also lead means you’re an industry shill and have no morals
to olks doing some pretty weird things, and even and are bound to speak the industry/commodity
twisting or making up data, which really casts a ‘party line’
li ne’. It’
It’ss hard to live wit
within
hin tthat
hat shadow, but
shadow on everyone in science. we’ve managed to blend basic science with, I think,

But, like most jobs, perseverance and patience

tend to win out. Having good mentorship meant I
learned early how to write and craf decent grants, [...] we’ve
which really helped early in my career. I think
(hope) I’ve passed
passed some o that on to others who managed to
have trained with me! Te trainees rom my lab are
now installed in a ew institutions around the world blend basic
and I hope they taken the good (and not so much o
the bad, ha ha) with them. science with,
Has the publish or perish atmosphere had impacts
I think, good
on your lie and how you do research?  
Sure, early in my career I was very absorbed in my science rom
work and was not, at times, the best partner to my
wie and even perhaps the best ather to my kids industrr y money
indust
(I’vee been tr
(I’v trying
ying to make amends beore my oldest
boy – now 15 – begins to see me as nobody other too. I don’t eel
than the jerk who holds the car keys!!). It pays to
have an understanding and supportive spouse. I do, like I’ve sold my
however, think it’s true that those that tend to rise in
academia have to spend a disproportionate amount soul and I sleep
o time in their work enviro
environment
nment to succeed. I
doubt whether that’s untrue in other proessions, well at nights.
however, and like most things that are a passion it
never really eels like work. I do have the greatest
wie on the planet, however, who I might mention good science rom industry money too. I don’t eel
is also an academic in exercise physiology, but defi- like I’ve sold my soul and I sleep well at night. I’ve
nitely keeps our lives and the household show on the been asked by some people why I just don’t do more

road! basic work and get more government money.

From a research perspective, the pressure to publish Honestly, it’s not like I haven’t tried, but unding is
 has also meant doing research with industry and
[...] so much o your work is eevaluat
valuated
ed
by nameless and ac aceless
eless people who
iguratively ‘hold the cards’ and can turn
you one way or another. It has also lead
to olks doing some pretty weird things,
and even twisting or making up data
while we’ve done well, relatively speaking (I’d rate
my grant success rate at just above 20%), the sums
o money and budget restrictions are a big hurdle
or Canadian researchers to be ‘competitive’ inter-
nationally. Still, blessings counted, fingers crossed,
we’ve done and continue to do better than average.
O course all o this is due to the studen
students,
ts, who are
the true lieblood o our success as a group.
Some lifers and athletes go well beyond recom-
mended protein intakes, and approach two grams
 per pound o bodyweight
bodyweight a day
day.. W
Would
ould you expect
nitrogenous waste rom this approach to have side
effects over time?   evidence or their effectiveness or building muscle,
I think the point everybody has orgotten, or
perhaps were never taught, is that nitrogen (the
essential nuclide o amino acids) is metabolically
toxic in mammals. In act, most species have evolved
a mechanism to get rid o nitrogen – ammonia in
fish, uric acid in birds – because there’s no place to
store ‘extra’ amino acids. Amino acids are used or
protein-requiring processes or they are not. You
can’t ‘store’ them, you can’t magically make them
tight, very tight, in Canada and everywhere. And
89
inated (the nitrogen is taken off and transerred to
another compound) and urea gets made.
So when people recommend two grams o protein
per pound (i.e., 4.4 grams per kilogram) they lack
a basic understanding o how and or why higher
protein might even possibly be used by the body
at that kind o level! Now, people can twist studies
and show whole body protein turnover measures
that ‘support’ this estimate, but that’s not muscle!
It’s time or a serious reality check or anybody who
spouts those numbers or recommends huge doses o
supplements like BCAAs (which have next to zero
but that’s another story …).
Back to the question, does this
t his kind o intake cause
‘harm’?
‘harm’? Te biggest bugaboos or the h
higher
igher pro-
tein diet are clearly bone loss and kidney disease.
Te ‘teaching’
‘teaching’ on both, old school teaching, is that
higher protein lowers blood pH, which causes bone
resorption. Calcium is ‘leached’ rom your bones.
b ones.
Tis results in your bones getting brittle as you
 into something to be used later, so they are deam
  I loved learning, I loved school, I loved
nutrition, and I liked (yes only liked)
research…it wasn’t until the last stages o
my Ph.D. that I truly grew to love rese
research.
arch.
theory is known as the acid ash hypothesis – acid
rom protein in blood and ‘ash’, or calcium rom
your bones. Te take home on this theory can be
neatly summarized in a series o nice meta-anal-
yses – thus an evidence-based answer
answer:: “Evidence
suggests a linear association between changes in cal-
cium excretion in response to experimental changes
in net acid excretion. However, this finding is not
evidence that the source o the excreted calcium is
bone or that this calciuria contributes to the devel-
opment o osteoporosis.”
Furthermore
Furthermore,, “Tere is no evidence rom superior
quality balance studies that increasing the diet acid
load promotes skeletal bone mineral loss or osteo-
porosis … Promotion o the ‘alkaline diet’ to prevent
calcium loss is not justified.” And finally , “All o the
findings rom this meta-analysis were contrary to
the acid ash hypothesis … Tis meta-analysis did
not find evidence that phosphate intake contrib-
utes to demineralization o bone or to bone calcium
excretion in the urine.
Dietary advice that dairy products, meats, and
grains are detrimental to bone health due to “acidic”
phosphate content needs reassessment. Tere is no
evidence that higher phosphate
phosphate intakes are detri-
mental to bone health.” I think those analyses ‘close
progress toward osteopenia and osteoporosis. Tis
90
the book’ on some o the claptrap that some olks
spout about higher protein and bone. Renal disease
is a little more granular and harder to nail down,
but I think I’ll go with the quotes rom the WHO
report on Protein and Amino Acid Requirements in
Human Nutrition that states
states,, “…the suggestion that
the decline o glomerular filtration rate that occurs
with advancing age in healthy subjects can be atten-
uated by reducing the protein in the diet appears to
have no oundation.
oundation.””
In addition, in the most recent revision o the DRI
by the Institute o Medicine that section on protein
states that
requirements also states that there is no relationship
between increasing protein intakes and decline in
renal unction in people with normal renal unction.
Now, i you have a diseased kidney, then it’s perhaps
not a good idea to be eating lots o protein, there
is pretty clear evidence that a low(er) protein diet
(exact level not known) does extend liespan
liespan.. My
take: it’s hard to find evidence that intakes higher
than 1.6-1.8 grams o protein per kilogram o body-
bo dy-
weight are able to substantially augment gains in
muscle mass, as reviewed here
here,, here,
here, here
here,, and here
here..
We need to get you back on here sometime Stu.
 Always somethin
 Always somethingg new to learn. Ta
Tanks
nks so much or
or
taking time to talk with us!  ◆

INTERVIEW: 
Ramsey Nijem
First, I would like to thank
t hank the Examine.com team
or the interview. I am honored and will do my best
to not bore the highly
high ly educated audience
audience..
Second, I would like to thank the
t he other members
o our sport science staff who seamlessly combin
combinee
over 60 years o experience in the NBA with an evi-
dence-based approach.
Lastly, I am obliged to say that the responses
below are on my behal and do not represent the
Sacramento Kings or the NBA.
What do you do or the Sacramento Kings? How did
 you get into strength and co
conditionin
nditioning,
g, and when
did sports nutrition come into the picture?   collected regularly. We don’t claim to be the first to
I am the assistant strength and conditioning coach
or the Sacramento Kings. We do not have a “head”
strength and conditioning
conditioning coach. Our Director o
91
sulted or this interview as he oversees everything,
including nutrition
nutrition or our team, and has seen it all
in his over 20 year NBA career. As an aside, Chip
is one o only a handul o people that can claim
to have worked with both Michael Jordan and
Kobe Bryant or all o their championship rings (11
combined).
My role is split, and shared with Chip, between
traditional strength and conditioning responsibil-
ities and sport science. We continually collect data
and use the objective numbers to influence our
treatmentt and training decisions. Everything rom
treatmen
movement screening and joint range o motions to
power characteristics and on-court player loads are
do this, nor do we pretend to have all the answers,
rather we pride ourselves on our interdisciplinary
approach,
app roach, as our ultimate goal as a sport science
 Sport Science, Robert Chip Schaeer, was con- staff is to keep our guys healthy.

92

My desire to become an NBA strength and condi- in their nutrition and are able to stay healthy and
tioning coach came the day I realized I wasn’t going perorm well, it’s my opinion that they are not opti-
to make it as a player. I figured i I couldn’t play in mizing their potential to train, recover
recover,, and perorm
the NBA, then I’d
I’d do everyt
everything
hing I could to train the every night. I have observed that it ofen takes
guys that do. So I went on to earn a Master’s degree something to trigger a player to proactively change
in sport perormance and I am now working toward his eating. Whether it’s a slump in perormance, a
my Doctorate degree in human and sport peror- cold, an injury, the grind o the season, or relative
mance. Sports nutrition is obviously relevant when old age, most these guys need an experience to wake
studying and applying
applying sports science to maximize them up a bit. Perhaps that’s just human nature.
an athlete’s
athlete’s potenti
potential.
al.
Te NBA season is looooooooon
looooooooong.
g. How do players
When considering the demands placed on a cope with the grind o training, competition, stress,
high-level athlete’s and injury?
body, one is remiss Looong indeed.
i nutrition is not
considered every
[...] i I couldn’t play Tis is the essence o
what we are trying
bit as important as
training and recov-
in the NBA, then I’d do to figure out with
all o our data. How
ery. Indeed, what
an athlete puts
everything I could to are guys adapting
to the stress that the
into their body
will influence their
train the guys that do. NBA season brings
and how can we
ability to perorm help them combat
over the course o the season.
se ason. In a world where even the stress that undoubtedly wears them down? Te
the slightest advantage counts, nutrition and sup- short answer is they keep up with treatmen
treatment,
t, train-
plementation offer an opportunity to train harder, ing, and get as much rest as they can. But how we
recover quicker, and ultimately perorm better than go about managing loads on the court and in the
the competition. weight room is the
t he complexity that brings us sport
science nerds to the drawing board. As a sport sci-
Do you find any correlation between what a given ence staff we are able to watch the loads accumulate
 player eats and how we
well
ll he perorms?   over time and see how their body is reacting. We
Guys who eat a nutrient dense diet, especially diets can use that data, in an ideal world, to structure
high in vegetable consumption,
consumption, seem to resist the training and treatment to allow or recovery, yet
 viral inections that
that invariably
invariably run through teams provide enough o a stimulus to keep them strong,

each year. Tese types o inections can affect a explosive, and injury
injury ree. I say ‘ideal world’ because
player or weeks, so avoiding them can have a tre- the most influential stressor to the NBA player is the
mendous impact. Although I am not naïve to the  volume o games,
games, and that is unchangea
unchangeable.
ble. An NBA
 act that some players may not take much stock
[...] i I could orce these guys to take
my advice, then they would take the
principles we have set in place and ully
commit to them year-round. That’s not to
say they don’t do a good job already
already,, but
there is always room or improvement.
average month a team will play 15 games. It’s a war
o attrition.
Te Western Conerence has eight teams that could
 potentially
 potentially compete or the ch
championsh
ampionship.
ip. C
Crazy!
razy!
Oh yeah, this is supposed to be about nutrition ...
what are some interesting things about how other
teams eat and train? Does it differ much rom team
to team?
Te nature o the game day schedule doesn’t allow
or time to sit down and talk shop at a philosophical
level, and most interactions are mid-court banter
during pre-game warm-up. Tus it is hard to com-
ment on the degree o difference between teams,
although I’m sure training and nutrition practic-
es can vary greatly. I’ve seen videos o ellow NBA
strength coaches having players hex-bar rack pull
over 400 pounds rom a mid-shin height (quite
impressive when considering limb lengths and
the amount o work being done), while I’ve heard
stories o other coaches preerring to have player
playerss
perorm banded glute and core work all session long.
team can have our games in five nights, and in an
93
 vide any ried oods or sodas and
and some may p
prohibit
rohibit
“junk ood” on the plane, while other teams have a
more hands-off approach and allow the players to
make their own decisions.
I’m not here to say what the best training and nutri-
tion practices are, but I’m confident enough to
know a ew things to be true. A training approach
that emphasizes the SAID and progressive overload
principles while appreciating
appreciating the value o injury
risk reduction and movement quality is going to
provide a great return on investment. In a similar
even-keeled ashion, a nutritional approach that
emphasizes lean protein options, ruits and veg-
gies, and complex carbohydrates will provide a
great nutrient return on caloric investment. Tese
approaches don’t sell DVDs and t-shirts, but they
produce avorable results and are consistent with the
evidence-base.
It’s hard to tell until you stand next to them, but
 NBA players can
can be quite m
massive.
assive. H
How
ow much does
 On the nutrition side, I know some teams don t pro- a big orward or center, like DeMarcus Cousins, eat?

94

I’d imagine they need quite a bit o uel to run up I you could orce players to take your advice on
and down the court, plus practice and gym work.  strength, conditioning, and nutrition, what are some
I couldn’t tell you exactly how much our guys are importantt gems that they should keep in mind?  
importan
eating, but suffice to say it is a lot. A typical NBA I I could orce these guys to take my advice on
big is likely burning between 4,000 and 5,000 calo- strength, conditioning,
conditioning, and nutrition, it would be to
ries (numbers approximated using Harris Benedict  view these things as ways to optimize
optimize their poten
potential.
tial.
ormula or DeMarcus Cousins). oss in an over- When a guy is making millions playing basketball,
time or two or a big minute guy and you can it is hard to get them to see the value in some o
imagine that number can climb pretty high. We these things. Tey figure, “Hey
“Hey,, I made it here doing
take weights and skinolds regularly (two to three what I’m
I’m doing, so why do I need to change?” Tis
times per month, depending on our schedule) and attitude
attitude is short-sighted, in my opinion. I they can
most guys are able begin to appreciate
to remain calorically the value o training
neutral (neither sur-
plus or deficit) over
An NBA team can (strength and con-
ditioning), recovery,
time on their own, and how nutrition
with little weight
have our games in is involved in all o
changes between
measurements.
ive nights, and in an it, they may see the
potential to play at a

Tis liestyle offers no

average month a team higher level or lon-
ger. Undoubtedly
shortage o calories
or the guys to uel
will play 15 games. It’s a stronger, bet-
ter-conditioned,
up on. In addition
to catered meals at
a war o attrition. better-recovered
athlete is a better-per-
the practice acility
acility,, orming athlete. Not
hotels, plane rides, and arenas, we carry around a to mention, all o these things can prolong careers
nutrition trunk stocked with snack options or the (which o course means more money or them). I
guys. Not eating can be more o an issue with our I could change their perspective on this stuff, my
schedule. When you get into a city at 2:00 a.m. and  job would be much easier.
easier. ime ofen cha
changes
nges their
have shoot-around
shoot-around at 10:00 a.m., guys will sacrifice perspective, as our veterans are willing to dedicate
breakast or an extra 30 minutes o sleep. Although more time to these things.
th ings. But i young guys bought
they are encouraged to make the right choices, like in sooner, the results would speak or themselves.
calling or room service beore going to practice

on an empty stomach, they do not always listen. In When it comes to eating, I would orce players to
these instances, we’ll provide them with something take a more conscious approach to their nutrition
rom the trunk. habits. Tings like increasing lean protein consump-
 tion and limiting sugar consumption (should not

95

be read as an anti-sugar suggestion) to make room or

more nutrient-dense options like vegetables, whole
grains, and healthy ats. As ar as training goes, I
would orce players to dedicate off-season time to
the weight room. Te NBA schedule makes it nearly
impossible to make any meaningul training adapta-
tions during season, which makes the off-season an
important time or improving strength and power,
adding muscle, and improve movement quality and   [...] the
groove movement patterns to reduce injury risk.
dedication
o summarize, i I could orce these guys to take my
advice, then they would take the principles we have set that they have
in place and ully commit to them year-round. Tat’s
not to say they don’t do a good job already, but there is is a habit
always room or improvement.

 Are supplements
supplements co
common
mmon in th
thee NBA? I so
so,, which
that could be
ones?   adapted by
Supplements are absolutely common in the NBA.
Although, similar to nutrition practices, supplemen- anyone in any
tation varies rom team to team. Te most common
supplements are whey protein, creatine, a multi- line o work
 vitamin, fish oil, and
and vitamin D
D.. Other common
supplements
supplements are high glycemic energy chews or gels and would be
and caffeine. For guys with sleep issues, we provide
ZMA, and or chronic pain problems we provide glu- beneicial.
cosamine chondroitin.
chondroitin. Although we acknowledge
some supplements are largely anecdotally supported
(e.g. ZMA), while others may be limited to specific
conditionss (e.g. glucosamine and osteoarthritis), the
condition
 value o anecdotal
anecdotal evidence and p
placebo
lacebo effect cannot
be discounted at this level. I a glucosamine chondro
chondroi-
i-
tin supplement
supplement rids a guy o knee pain even one game

sooner than i he had not taken it, then its use is justi-
fied in my opinion. A potential or benefit with little to
no risk is a win or us.
 96

  In a world where even the slightest

advantage counts, nutrition and
supplementation offer an opportunity
supplementation
to train harder, recover quicker, and
ultimately perorm better than the
competition.

Being surrounded by world-class athletes, have you
noticed any particular habits that readers might
want to know about? Even though Charles Barkley
says he’s not a role model, I know that many people
look up to proessional athletes.  my homework.
Te amount o time these guys put in should be
applauded. Most guys are at practice early or treat-
ment, weight room work, and shooting, yet stay afer
or the same things. Surely they get compensa
compensated
ted
well or what they do, but the dedication that they
have is a habit that could be adapted by anyone in
any line o work and would be beneficial. With that
said, they also have habits that should be avoided.
Not prioritizing their nutrition, strength, and condi-
tioning can have a huge impact on their health and
longevity. Te offseason is not only a time or them
to refine their game, but an opportunity to become
a better, less injury prone athlete. Most guys are only
ocused on playing and not the off court work. Tis
is unortunate and can backfire, as the NBA is only
becoming more and more athletic.
Tank you to the Examine.com team or their
high quality work. Your website and products are
influential in my nutrition and supplementation
approaches and I can’t thank you enough or doing
Tanks so much or taking some time out or us
Ramsey! Tis is really, really cool inside inormation
to learn. Sacramento is an intriguing team in an
extremely tough conerence, so it’s good they have a
smart nutrition and conditioning team supporting
them. We look orward to watching the rest o the
season, and best o luck to you.
Ramsey Nijem is the Assistant
Strength and Conditioning
Coach or the Sacramento
Kings. He has an M.S. in
Kinesiology and is currently in
a doctorate program.
 97

Until Next Issue...

The next issue of ERD will come out the

first week of February
Februar y. In the meantime:

Comments? Questions? Join the

Concerns? Discussion Online
Send your thoughts to Join the conversation on
erdeditor@examine.com Issue 3 in our exclusive
ERD Private Forum 
on Facebook

Spread the Word


Credits
Copy Editor: Dmitri Barvinok
©iStock.com/wetcake
©iStock.com/ponkrit
©iStock.com/macrovector
©iStock.com/Matt_Brown
©iStock.com/miflippo
©iStock.com/razihusin
©iStock.com/Graffizone
©iStock.com/tacojim
©iStock.com/Johny87
©iStock.com/Bouganville
©iStock.com/MmeEmil
©iStock.com/CaroTroconis
©iStock.com/pjohnson1
©iStock.com/erierika
©iStock.com/MariuszBlach
©iStock.com/Acerebel
©iStock.com/hocus-focus
98
©iStock.com/SharonFoelz
©iStock.com/Jag_cz
©iStock.com/aluxum
©iStock.com/Trifonov_Evgeniy
©iStock.com/4kodiak
©iStock.com/YinYang
©iStock.com/amphotora
©iStock.com/julichka
©iStock.com/loooby
©iStock.com/snokid
©iStock.com/RichVintage
©iStock.com/lzf
©iStock.com/zmeel
©iStock.com/EdnaM
©iStock.com/f4f
©iStock.com/bopav
©iStock.com/sshepard