See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/15424171

Dietary cravings and aversions during pregnancy

Article  in  Indian Journal of Public Health · July 1994

Source: PubMed

CITATIONS READS
12 312

3 authors, including:

Kumudu Wijewardena
University of Sri Jayewardenepura
80 PUBLICATIONS   876 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ADVANCE study (Addressing Domestic Violence in Antenatal Care Environments) View project

All content following this page was uploaded by Kumudu Wijewardena on 15 June 2015.

The user has requested enhancement of the downloaded file.

Dietary cravings and aversions during
pregnancy1’ 2

Ernest B. Hook, M.D.

ABSTRACT Interviews of 250 women concerning dietary changes during pregnancy were
undertaken immediately after delivery. With regard to beverages, of women who regularly drank
coffee or alcoholic items prior to conception, almost 30% reported a significant drop in ingestion
during pregnancy. For coffee this change was attributed primarily to “endogenous” factors e.g.,
provocation of nausea or a loss of taste for the beverage. Concern regarding maternal or infant
health was the most frequent reason for decrease in alcoholic beverages, although endogenous
factors were also mentioned. Soda beverages were also ingested less frequently, predominantly
because of dieting. The main increase was in milk consumption, primarily attributed to concern for

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

the infant, but many mothers cited only a craving for the beverage. Other foods for which specific
cravings were frequently cited were ice cream, sweets, and candy (particularly chocolate), fruits,
and fish. Foods for which aversions outnumbered cravings were meats, poultry, and sauces flavored
with oregano. Possible explanations for these changes associated with pregnancy are
discussed. Am. J. Clin. Nutr. 31: 1355-1362, 1978.

The development of aversions to or cray- opment of clinical symptoms in response to

ings for ingested items during pregnancy is a the beverages, such as nausea and vomiting
well-known phenomenon. Pica for clay or of pregnancy, or simply to changes in a desire
starch for example, has been extensively in- (or “taste”) for the beverage in question (3).
vestigated in many populations particularly It was then decided to undertake systematic
in the southern United States, although the study ofall food (and nonfood) items ingested
cause of these (or other cravings) during preg- during the entire pregnancy in another sam-
nancy are still not well understood. (1). ple of women.
Despite the fact that there is an extensive
awareness of the changes in desires for nutri- Materials and methods
tional items during pregnancy, there has
From March 6, 1975 to May 20, 1975 250 women who
been, to my knowledge, no previous quanti-
had delivered a liveborn infant were interviewed by a
tative study of the development of food aver- single observer at two local hospitals. These mothers
sions, and only one small such study of crav- were selected from a total of 847 women who had
ings for foods during pregnancy. delivered a liveborn infant at the two hospitals during
this period. Because of a wish to avoid provoking addi-
The present study arose out of an investi-
tional difficulty for those whose pregnancy had just
gation that primarily focused upon the subtle ended with a catastrophic outcome, interviews were lim-
teratogenic effects of cigarette smoking (2). ited to mothers of ostensibly normal infants. That is,
After discovery of significant changes in this mothers of stillbirths or of infants placed in the intensive
habit during pregnancy, primarily a decrease, care unit were excluded. There was no other specific
selection procedure except that the mothers were those
an investigation of beverage ingestion was
available for interview on a weekday but before they left
also undertaken (3), since coffee and alcohol the hospital. Comparisons between our sample and all
consumption are known to be associated with
smoking. Many women reported significant
aversions to coffee and alcoholic beverages From the Epidemiology and Human Ecology Sec-

early in pregnancy, and a craving for milk. tion, Birth Defects Institute, Division of Laboratories
These were, many reported, unrelated to phy- and Research, New York State Department of Health
and Department of Pediatrics, Albany Medical College,
sician’s advice or maternal concern about the Albany, New York.
infant or course of pregnancy. Rather, in 2 Supported in part by a grant from the National
many cases they were related to the devel- Institute of Child Health and Human Development.

The American Journal of Clinical Nutrition 31: AUGUST 1978, pp. 1355-1362. Printed in U.S.A. 1355
1356 HOOK

mothers giving birth to hiveborns at these two hospitals TABLE I

during study in the time frame appear in Table 1. Comparison of selected characteristics of the study
Mothers were first asked (in random starting order) population with mothers of all hivebirths at the two
concerning changes in the first half of their pregnancy hospitals in the study
for any of the seven specific beverages listed in Tables 2
Study Mothers of all
and 3. They were also queried specifically concerning
population Iivebirths
the reasons for any changes. They were then queried in
an unstructured manner (i.e., without mentioning spe- No. 250 847
cific items) concerning cravings or aversions for other %
foods (or nonfood items) that may have developed at
Race
any time in their pregnancy.
White 87.2 88.3
Black 11.6 11.3
Results Other 1.2 b

Not stated or unknown 0.4

The data in Table 1 indicate that there Total 100.0 100.0
were no significant differences between the
Maternal age in years
mothers sampled and the entire group deliv-
<15 0.8 0.7
ering liveborn infants at the two hospitals 15-19 13.6 15.5

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

with regard to race, maternal age, or reported 20-24 28.0 30.0
years of schooling (a rough index of socioec- 25-29 36.0 35.7
30-34 16.4 13.4
onomic status). The mothers queried were
35-39 4.0 4.3
not, of course, a representative sample in that 40-44 0.8 0.5
mothers of infants with catastrophic out- 45+ 0 0
comes were excluded, but the latter com- Not stated 0.4 0.1
prised only about 2% of the total of all live- Total 100.0 100.2’

births at the time in question. Thus there is

Mother’s education (years of
no reason to believe that inclusion of the schooling completedY
latter category in the sample would have 8 1.6 0.9
markedly changed the distribution of total 8-10 10.0 12.4
11 9.2 8.2
preferences expressed. (It is possible of course
12 40.0 44.4
that the small group of mothers in the latter 12 38.4 34.0
category would have expressed a significantly Not stated 0.8 d

different set of preferences. This possibility is TotaY 100.0 999C

under investigation in a separate ongoing a Six of these infants died in the first 4 days after
prospective investigation of mothers in early delivery. b Vital records summaries distinguished only
pregnancy.) “white”, “nonwhite” and “not stated”. C Totals are
occasionally not 100.0% because of rounding
The data on changes in consumption in the
error. d The data on the comparison total population
seven beverages appear in Table 2. In Tables were gathered from vital record summaries. The educa-
3 and 4 are listed the reasons for the decreases tional level of the mother was not recorded on 24.8% of
and increases that were reported. such certificates. The figures in this column give the
In Tables 5 and 6 appear data on the distribution in those mothers for which data were avail-
able.
frequency of cravings and aversions for other
food items during pregnancy.
Because of the great range in replies con-
cerning specific food items, tabulating these women who regularly drank coffee, soda,
systematically proved difficult. The results of beer, wine, or other alcoholic spirits dimin-
the specific food cravings and aversions, as ished consumption in the first half of preg-
far as these could be distinguished, appear in nancy.
TableS. The first footnote of Table 5 explains Decrease in coffee ingestion was primarily
the scoring system used in this analysis. In attributed to the endogenous factors sum-
Table 6 data on some grouped categories are marized in Table 3, and indeed about half of
presented. all women diminishing coffee consumption
specifically cited a response to or provocation
Discussion of nausea as the explanation. On the other
hand, for soda and alcoholic beverages, con-
Concerning the seven beverages queried, a cern for the infant or the mother’s personal
significant fraction, at least a quarter, of the health (e.g., weight control) were cited more
 DIETARY CRAVINGS AND AVERSIONS DURING PREGNANCY 1357

TABLE 2
Changes in ingestion of seven beverages

Coffee Tea Milk Soda Beer Wine Other spirits

No. (and percent of 250 in- 170 204 228 212 83 147 153
terviewed) reporting 68.0% 81.6% 91.2% 84.8% 33.2% 58.8% 61.2%
drinking the beverage
regularly at start of preg-
nancy

No. (and percent) of those 49 19 10 55 31 37 44

who drank beverage who 28.8% 9.3% 4.4% 25.9% 37.3% 25.2% 28.8%
decreased in first half of
pregnancy

No. (and percent) of those 9 17 124 12 0 1 1

250 interviewed who in- 3.6% 6.8% 49.6% 4.8% 0 0.4% 0.4%
creased in first half of
pregnancy

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

TABLE 3
Distribution of reasons cited for decrease in beverage consumption
in the first half of pregnancy

. , . . Other alco-
Reason cited for decrease Coffee Tea Milk Soda Beer Wine . .
holic spirits

Endogenous factors”
Response to or provocation 23 (4) 8 ( l) 2 (l) 5 8 10 9 (3)
of nausea
Lossofurgeortaste 11 (5) 3(2) 3(l) 8 12(l) l2(l) 14(3)
Provocation of “indigestion” 6 (1)” 2 0 6 (1)” 0 4 0
Provocation of other adverse 1 (1) I 3 1 0 1 2
reaction

Concern for personal health

and/or infant
Concernforinfant 1(1)” 0 0 2(3) 4(l)#{176} 6(1)” 8(2)”
Concern for personal health 1 0 0 5 (1)’ 1 0 (1)” 3 (1)”
Doctor’s advice 0 1 0 7 (1)” 0 0 0
Dieting 0 0 1 9(5)” 0(1)’’ 0 0

Other
Change in social habits 0 0 0 0 2 (1)” 1 3
Change in season 0 1 0 1 2 0 0
Because of increase in milk 0 1 (1)” 0 4 0 0 0
ingestion
Other 0 0 0 0(1)” 0 1 (1)” 0(1)#{176}
No reason offered 0 0 0 1 0 0 0

Mothers citing one reason only 43 17 9 49 29 35 39

Mothers citing two of above 6 2 1 6 2 2 5

reasons”

Total mothers citing decrease in 49 19 10 55 31 37 44

first half of pregnancy

a The reasons cited by those mothers who gave two reasons for decreasing appear in parentheses. Thus 23 mothers
cited nausea as the sole reason for increasing coffee ingestion, four cited nausea and some other reason as well. b In

the earlier study we distinguished a category other than those listed in this table among the endogenous reasons for
decreased exposure: “greater susceptibility to alcoholic beverages”. (Here those who gave such a reply initially were
classified further by the specific symptom to which they indicated they were susceptible.)
1358 HOOK

TABLE 4
Distribution of reasons cited for increase in beverage consumption
in the firsthalf of pregnancy

, . . Other alco-
Reasons cited for tncrease Coffee Tea Milk Soda Beer Wine .
holic spirits

Endogenous factors
Greater desire for item” 1 0 15 (1)” 2 0 0 0
Better tasteb 0 0(1)” 17(6)” 2(1)” 0 0 0
Increased thirst 0 0 (1)” 5 (4)” 3 0 0 0
Relief of heartburn cramps or 3 0 3 1 0 0 0
other symptoms
Response to nausea (relief) 0 2 1 3 (1)” 0 0 0

Concern for personal health

and/or infant
Concern for infant 0 0 53 (11)” 0 0 1 0
Doctor’s advice 0 0 14 0 0 0 0

Other

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

Change in social or personal 3 1 0(1) 0 0 0 1
habits
Substitute for another bever- 1 9 (1)” 2 (1)” 0 0 0 0
age
Other 1 2 (1)” 1(2)” 0 0 0 0
No reason offered 0 1 0 0 0 0 0

Motherscitingonereasononly 9 15 111 11 0 1 1
Mothers citing two of above 0 2 13 1 0 0 0
reasons”

Total mothers citing increase in 9 17 124 12 0 1 1

firsthalf of pregnancy

“The reasons cited by mothers who gave two reasons appear in parentheses. Thus, 53 mothers cited “concern” as
the sole reason for increasing milk consumption; another 11 cited this reason and some other reason as well. Mothers
reporting a greater desire or urge for an item who specifically mentioned that it tasted better were coded to the
“better taste” rubric.

often as reasons for decreased ingestion; yet coffee they differ somewhat from results of
here too endogenous factors were frequently Little et a!. (4) who, using a different ap-
cited, e.g., about 25% ofwomen who regularly proach, found an even greater number of
drank any alcoholic beverage before the start women reporting changes. These authors re-
of pregnancy who decreased consumption ported that two thirds of their sample drank
specifically cited nausea in explaining this less alcohol during the first 4 months of preg-
change. nancy than before pregnancy. These authors
Consistency of response is frequently used interviewed mothers during pregnancy, while
as one measure of validity in a survey. As the the results reported here were from interviews
women in this study were interviewed only just after delivery, so that it is possible the
once, this factor could not be measured di- discrepancies are due to memory loss of those
rectly. An even more powerful test of con- women interviewed in this study, although
sistency however, is provided by comparison other explanations are, of course, also possi-
of the results here with those from a study of ble. In any event, both studies are consistent
another population. In this regard, the results in reporting a predominant decrease. Fur-
on beverage consumption (Tables 2 to 4) thermore, in the study of Little et al., 6 1.4%
replicate almost exactly those of an earlier (70 out of 1 14) of the women who drank
investigation which used the same method- coffee before pregnancy reported a decrease
ology in study of a different group of preg- in consumption, compared to the 28.8% re-
nant women (3). With regard to alcohol and porting such a decrease here. This difference
 DIETARY CRAVINGS AND AVERSIONS DURING PREGNANCY 1359

TABLE 5
Distributions of cravings and aversions for specific food items
reported during pregnancy by 250 women”

No. and percent of total reporting

Item no. Item
Craving Aversion

I Chocolate ice cream 14 (5.6%) 0(0%)

2 Other ice cream, or ice cream in general 46 (18.4%) 1 (0.4%)
3 Other dairy items (except milk) or dairy in gen- 14 (5.6%) 0(0%)
eral
4 Chocolate (other than in ice cream) 39 (15.6%) 1 (0.4%)
S Other sweets, candy, or candy in general 29 ( 1 1.6%) 3 (1.2%)
6 Sweet starches (pies, cakes, cookies, etc.)” 24 (9.6%) 2 (0.8%)
7 Starches (bread, rice, etc.) 6 (2.4%) 3 (1.2%)
8 Salty “snack” foods, e.g., potato chips 6 (2.4%) 2 (0.8%)
9 Peanut butter 4 (1.6%) 3 (1.2%)
10 Pickles 10 (4.0%) 2 (0.8%)
11 Other vegetables, or vegetables in generalC 15 (6.0%) 23 (9.2%)
12 Citrus fruits and fruitjuices 30 (12.0%) 5 (2.0%)

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

13 Other fruits (and fruit juices), or fruits in gen- 30 (12.0%) 1 (0.4%)
eral”
14 Fishe 19 (7.6%) 12 (4.8%)
15 Meati 14 (5.6%) 42 (16.8%)
16 Poultry 2 (0.8%) 2 1 (8.4%)
17 Eggs 3(1.2%) 3(1.2%)
18 “Italian food” (sauces)” 12 (4.8%) 24 (9.6%)
19 “Submarine” sandwiches” 6 (2.4%) 0(0%)
20 “Spicy food” in general 0 (0%) 7 (3.2%)
21 “Chinese food” 3 (1.2%) 1 (0.4%)
22 Fried food (in general)’ 0 (0%) 3 (7.2%)
23 Beverages other than those listed in Table 1 or 4 (1.6%) 1 (0.4%)
2’
24 “Nonnutritive” items” 4 ( 1.6%) 1 (0.4%)
25 Salad dressings, mayonnaise 0 (0%) 4 (1.6%)
26 Other 1 (0.4%) 2 (0.8%)

a Any classification of the multiple responses is somewhat arbitrary. In this tabulation foods were coded to one
item, unless the interviewee singled out specifically aspects of the food that applied to two items. For example, if a
mother mentioned a craving for chocolate ice cream she was coded to item 1 only; ifshe said “all chocolate, especially
chocolate ice cream” she was coded to items 1 and 4. If she said “all candy especially chocolate items” she was coded
to items 4 and 5, etc. (An alternative approach is presented in Table 6 in which some items are pooled.) No individual
was coded more than once to the same item. “Submarine sandwiches” (which are sandwiches with French or Italian
bread, lettuce, tomato, cheese, meat, or tuna fish filling of some sort, salad dressing of some sort and usually onions),
were singled out in the analysis only because of the unexpectedly high specific mention of this item. Other
“sandwiches”, of which five were mentioned, were coded to their specific contents, not to starch (bread) item 7 as
well, unless some bread was specflcally mentioned in the description of the sandwich, in which case both the filling
and bread would have been coded. Exceptions to these general rules were made only for “banana splits” see below,
and “Italian food.” b The components of “sweet starches” were not individually coded. Thus raisin bread, was not
coded to fruit as well. Had a mother said “anything with raisins, especially raisin bread” she would have been coded
to both items, however. C Vegetables mentioned more than once for cravings were: tomatoes (4) olives (4), onions
(3), lettuce (2), salad greens (2); among aversions were: onions (5), potatoes (3), tomatoes (or sauce) (2), corn (2),
salads (2), all vegetables (2). d Banana splits (mentioned four times) were coded to fruits, item 13, ice cream, item
2, and chocolate item 4. e Seven cravings and three aversions were specifically for “fried” fish. One craving
and 10 aversions were specifically to meats described as being “fried”. g One craving and five aversions specifically
were for fried chicken or other fried poultry. h The biggest difficulty with classification for these purposes was
with Italian foods. Further inquiry revealed that it was the sauce to which there was a reaction in all cases; not the
item cooked with the sauce. The specific items mentioned which were coded to item 18 were cravingr spaghetti (2),
pizza (9), egg plant parmigian (2), lasagna (I), all italian food (2); aversionr spaghetti (9), spaghetti sauce specifically
(5), pizza (9), lasagna (4), Italian food in general (5). (Some mentioned more than one item.) Mention of these items
resulted in coding only to item 18, and to no other item in the table. ‘ “Fried foods” in general were not coded if
the mother mentioned only a specific fried item. Only specification of all fried foods resulted in coding of item 22.
An alternative analysis of any mention of fried foods is presented in the next table. (See also e, f. g above on meat,
chicken, and fish.) ‘ Cocoa use (mentioned twice) was coded to item 4 only. /t “Nonnutritive” items were: ice
(3), “chalk from a river clay bank” (I).
1360 HOOK

TABLE 6 ported to coffee, but, with regard to alcohol,

Analysis of Table 5 data by some grouped categories” such endogenous factors were not cited as
Number (and %) of total reporting re- frequently as were health related concerns.
Category (and item numbers sponse to at least one item in category
Little et al. did not analyze tea, milk, or cola
from Table 5)
Cravings Aversions ingestion, nor report data on increases in
Any dairy product 87 (34.8%) 10(3.6%) consumption in any beverages.
(milk” and/or any of With regard to milk, the striking increase
items 1-3) in consumption in this study does not appear
Any dairy product 64(25.6%) 1(0.4%)
to be attributable exclusively to the fact that
other than milk (any
the women had been told, or otherwise be-
of items 1-3)
Any ice cream (any of 52 (20.8%) 1(0.4%) lieved that such an increase would be bene-
items 1, 2) ficial for their health or that of their infant.
Any “sweet” item or 118 (47.2%) 10 (4.0%) Out of the 124 women who increased milk
fruit (any of items 1,
consumption 41, (33. 1%) cited only the “en-
2, 4-6, 12, 13)
Any “sweet” (any of 90(36.0%) 6 (2.4%)
dogenous” factors listed in Table 4 as an
items 1, 2, 4-6) explanation, and did not mention concern for

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

Any chocolate (any of 46 (18.4%) 1(0.4%) their health or that of their infant. Yet the
items 1, 4) latter factors are more socially acceptable
Any fruit (any of items 47(18.8%) 6(2.4%)
reasons which one would expect, would be
12, 13)
Any fish, meat, or poul- 30(12.0%) 64(25.6%) highly likely to be cited to the interviewer if
try (any of items they contributed to the change. Thus if any-
14-16) thing the increase in milk consumption attrib-
Any meat or poultry 16 (6.4%) 57 (22.8%)
uted to endogenous factors exclusively is
(any of items 15, 16)
Any mention of fried 9 (3.6%) 19 (7.6%) likely to be a minimum estimate of this com-
food ponent of the change.
a In this table an individual reporting a craving or
Concerning other ingested items in this
aversion to any item in the category listed is scored only study, the analysis of which is presented in
once in that category. For example, the 87 expressing Tables 5 and 6, there are several striking
cravings for any dairy product include 18 who craved findings. The most pronounced cravings ap-
milk and other dairy products, 23 who craved milk only,
peared to be for dairy items and sweets, in-
and 56 who craved other dairy products (mostly ice
cream) but not milk. b Milk response was based on cluding chocolate and fruits. A possible cx-
queries analyzed in Tables 2 to 4, i.e., consumption in planation for cravings for these factors may
first half of pregnancy; “other” dairy products at any be related to the increased need for calcium
time in pregnancy. Those scored as expressing milk
and calories during pregnancy, and a prefer-
“craving” were those attributing the increase to greater
desire, taste, or “thirst”, who did not also mention health
ential desire for foods relatively rich in sugars
related reasons. (There were 4 1 such women of whom 37 and calcium. Yet there is, of course, an in-
were white and four black.) Changes that were scored as creased need for protein as well but no reports
due to “aversions” to milk were those attributable to of predominant cravings for eggs, meats, or
endogenous factor in Table 3. There were nine women
poultry. Indeed an unexpected fmding was
total expressing such of which seven were white and two
black. that the main expressed food aversions were
for the latter two items. No associated culi-
nary factor (such as deep fat frying, sauces,
may also relate to methodological differences etc.) accounted for this. Another item to
in the studies, although, again other expla- which aversions were prominent was “Italian
nations are possible. Little et al. also analyzed foods”. On further inquiry, it was found the
the reported reasons for decrease. While it is reaction was apparently specifically to
hard to compare the analyses precisely, in “sauces” used in cooking. Some women men-
their study “physical reasons” (i.e., loss of tioned in particular those flavored with the
taste for the beverage or stimulation of ad- spice oregano.
verse physiological reaction) were the main Cravings during pregnancy that are fre-
explanations cited by those diminishing al- quently mentioned in the anecdotal litera-
cohol and/or coffee. In this study, similar tures are for pickles and olives. In this study
endogenous factors were predominantly re- cravings for these items were reported more
 DIETARY CRAVINGS AND AVERSIONS DURING PREGNANCY 1361

often than aversions (10 to 2, and 4 to 0, no significant differences in response of white

respectively) but the reasons for this remain and blacks in the study. Thus, for example,
obscure. 93. 1% of 29 blacks reported drinking milk
The only study of which I am aware which regularly at the start of pregnancy and 44.9%
provides any quantitative data for compan- then increased ingestion while among whites
son is an investigation of 300 Negro women the figures were 91.0 and 53.7%, respectively.
who had given birth to infants in 1949 to While the reasons for the discrepancies be-
195 1 in Tuskeegee, Alabama, 150 of whom tween this study and that in Tuskeegee are
had returned questionnaires (5). The study still speculative, it seems likely cultural and
was apparently done some time after delivery local geographic factors play a significant role
although many mothers returning the ques- with regard to cravings.
tionnaire were pregnant again. Of those re- Taggert (6) describing a study from Scot-
turning questionnaires, 55 reported food land presented no quantitative results con-
cravings of some type. The distribution of cerning specific items, but noted about two-
these were markedly different from those re- thirds of women expressed some food craving
ported here. The items for which the greatest during pregnancy, all of which “were for

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

cravings reported were: clay (24), fish (18), foods readily available, the most common
chicken ( 1 1 ), corn starch ( 12), and apples being fruit”. Of interest, sweets and chocolate
(1 1). (In addition cravings for 12 other fruits were mentioned only “occasionally”. The
were reported.) Cravings for candy and milk only comment concerning milk is that “sev-
were reported only twice for these items, in eral” women developed a liking or even a
contrast to the trends in this study. craving for it who did not enjoy it before
Differences between the Albany and Tus- pregnancy. This population seems more sim-
keegee populations may result of course from ilar to the Tuskeegee than the Albany popu-
cultural, geographic, or biological factors. lation in the relative lack of cravings for
Lactase deficiency resulting in milk intoler- sweets (and possibly even milk), compared to
ance in blacks is well known and may account fruits, again suggesting that cultural and geo-
for some of these differences (but see below). graphic factors, such as previous experience
On the other hand, individuals are likely to of and general availability of foods, as well
report a craving only for those items to which as perhaps even social traditions, are likely to
there has previously been exposure, and this, influence expression of nutritional cravings
as well as the patterns of response to endog- in pregnant women. Taggert also reported
enous factors, are likely to be highly influ- dietary aversions, particularly for coffee and
enced by local geographic as well as cultural tea due to development of “peculiar,” “dry,”
factors. Apples, watermelons, grapes, and or “insipid” taste for these beverages.
other fruits may have been frequently eaten Another pertinent report is that of Beal (7)
before pregnancy and it is possible that fruit who studied detailed records of “nutritional
consumption provided the primary source of intake” in women throughout pregnancy.
ingested sugars, in contrast to the population (The data were analyzed by nutrient compo-
reported here. Another difference may relate sition rather than by specific food types.)
to the fact that clays, at least in the past, were Changes resulting from endogenous cravings
occasionally ingested by many black women or aversions for particular foods were not
in the rural south even when not pregnant (1, distinguished from responses to medical ad-
5). Clay has a high calcium content (I) and vice concerning weight control and other fac-
this substance may have provided a signifi- tors. The association of nausea of pregnancy
cant source of calcium during pregnancy in with total calorie intake was specifically stud-
those responding in the Tuskeegee question- ied, however. The relationship was complex
naire, although there are no studies to my but in general the “average magnitude of
knowledge that prove that the calcium in caloric changes in either direction was similar
ingested clay is metabolically available. for women with various degrees of nausea as
The results of the Tuskeegee study for those without”. This was a group of
prompted a reanalysis of the Albany data by women of high educational background and
race of respondent. There were, essentially, may not be representative of the population
 1362 HOOK

at large. sions to alcoholic beverages. Studies of inter-

A frequently discussed hypothesis concern- est ifethically possible, would be a physiolog-
ing cravings in pregnancy is that these reflect ical investigation of the changes in taste
physiological needs of the mother and/or at threshold and of the changes in metabolism,
least indirectly, the fetus. Recently, it was particularly of alcohol, that are associated
suggested that analogously, aversions to foods with pregnancy.
and other items during pregnancy might be
the consequence of homeostatic factors which The interviews were conducted by Christine Westen-
have evolved as general feto-protective mech- dorf. The author thanks Grace Conway for reviewing
the data and assisting with tabulations.
anisms, and that in particular, nausea and
vomiting of pregnancy is one such mecha-
nism (3). Certainly, tobacco smoke and al- References
cohol, two agents both known to have embry-
1. HALSTED, J. A. Geophagia in man: its nature and
otoxic effects, have been found to stimulate
nutritional effects. Am. J. Clin. Nut. 21: 1384, 1968.
nausea and vomiting of pregnancy as a con- 2. HooK, E. B. Some general considerations concerning
sequence of which many women report di- monitoring. In: Birth Defects, Monitoring and En-

Downloaded from ajcn.nutrition.org by guest on February 25, 2013

minishing exposure to these substances. Thus vironment-The Problem of Surveillance, edited by
E. B. Hook, et al. New York: Academic Press, 1971,
the operational effect of the symptom com-
pp. 177-192.
plex, at least with regard to these agents, 3. Hoo#{231}E. B. Changes in tobacco smoking and inges-
seems likely to be feto-protective even if the tion of alcohol and caffeinated beverages during
general biological hypothesis is incorrect. It early pregnancy-are these consequences in part of
would appear of interest to search for embry- feto-protective mechanisms diminishing maternal
exposure to embryotoxins? In: Birth Defects-Risks
otoxic effects of other agents for which aver-
and Consequences, edited by S. Kelly, et al. New
sions frequently develop in pregnancy, al- York: Academic Press, 1976, pp. 173-183.
though I emphasize it should not be inferred 4. LITTLE, R. E., F. A. SCHULTZ AND W. MANDELL.
that any foods for which aversions develop Drinking during pregnancy. J. Stud. Abcoh. 37: 375,
1976.
during pregnancy are embryotoxic until di-
5. EDWARDS, C. H., H. MCSWAIN AND S. HAIRE. Odd
rect evidence is available. dietary practices of women. J. Am. Dietet. Assoc. 30:
A possible factor mediating the develop- 976, 1954.
ment of food aversions and cravings in preg- 6. TAGGERT, N. Food habits in pregnancy. Proc. Nutr.
nancy is change in olfactory and taste sensi- Soc. 20: 35, 1961.
7. BEAL, V. A. Nutritional studies during pregnancy.
tivity. Alternatively, metabolic changes ac-
1. Changes in intakes of calories, carbohydrates, fat,
companying the gravid state might be respon- protein, and calcium. J. Am. Dietet. Assoc. 58: 312,
sible for some changes, particularly in aver- 1971.

View publication stats