The Journal of Nutrition

Nutritional Epidemiology

Healthy Dietary Habits Score as an Indicator of

Diet Quality in New Zealand Adolescents1–3
Jyh Eiin Wong,4,6 Paula M. L. Skidmore,4 Sheila M. Williams,5 and Winsome R. Parnell4*

Departments of 4Human Nutrition and 5Preventive and Social Medicine, University of Otago, Dunedin, New Zealand; and 6School of
Healthcare Sciences, Faculty of Healthy Sciences, Universiti Kebangsaan Malaysia, Kuala Lumpur, Malaysia

Abstract

Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021

Adoption of optimal dietary habits during adolescence is associated with better health outcomes later in life. However, the
associations between a pattern of healthy dietary habits encapsulated in an index and sociodemographic and nutrient
intake have not been examined among adolescents. This study aimed to develop a behavior-based diet index and examine
its validity in relation to sociodemographic factors, nutrient intakes, and biomarkers in a representative sample of New
Zealand (NZ) adolescents aged 15–18 y (n = 694). A 17-item Healthy Dietary Habits Score for Adolescents (HDHS-A) was
developed based on dietary habits information from the 2008/2009 NZ Adult Nutrition Survey. Post hoc trend analyses
were used to identify the associations between HDHS-A score and nutrient intakes estimated by single 24-h diet recalls
and selected nutritional biomarkers. Being female, not of Maori or Pacific ethnicity, and living in the least-deprived
socioeconomic quintile were associated with a higher HDHS-A score (all P < 0.001). HDHS-A tertile was associated
positively with intake of protein, dietary fiber, polyunsaturated fatty acid, and lactose and negatively with sucrose.
Associations in the expected directions were also found with most micronutrients (P < 0.05), urinary sodium (P < 0.001),
whole blood (P < 0.05), serum (P < 0.01), and RBC folate (P < 0.05) concentrations. This suggests that the HDHS-A is a
valid indicator of diet quality among NZ adolescents. J. Nutr. 144: 937–942, 2014.

Introduction
Healthful dietary habits are associated with better nutrient intake
and higher diet quality, which in turn leads to positive health out-
comes (1–3). Because dietary habits are likely to track into adult-
hood (4), adoption of optimal dietary habits during adolescence may
have a protective effect against chronic diseases later in life (5).
Food habits questionnaires have been used to collect qual-
itative information on food behaviors of adolescents, including
food types, food preparation, cooking practices, snacking pat-
terns, and intake frequency of certain food groups, which can
then be compared with dietary guidelines (6–9). Compared with
quantitative reporting of food intake, usual dietary habits are
perceived to be more easily and accurately documented (9).
Because dietary habits are interrelated and tend to cluster (2,10),
there is a potential synergetic effect of multiple dietary habits, so
it is important to examine them as a group rather than studying
them in isolation (10). However, dietary habits information has
seldom been examined comprehensively by means of diet index
scores. In addition, the association of a constellation of desirable
dietary habits, as assessed by a diet index, with sociodemo-
1
Supported by grants from the New Zealand Ministry of Health.
2
Author disclosures: J. E. Wong, P. M. L. Skidmore, S. M. Williams, and
W. R. Parnell, no conflicts of interest.
3
Supplemental Tables 1 and 2 are available from the ‘‘Online Supporting
Material’’ link in the online posting of the article and from the same link in the
online table of contents at http://jn.nutrition.org.
* To whom correspondence should be addressed. E-mail: winsome.parnell@
otago.ac.nz.
graphic, nutrient intake, and biomarker data are less clear
because few studies allow investigation of these relations.
To date, 12 studies of children and adolescents used an index
embedding information on dietary habits. They mostly examine
the association of the index with food or nutrient intakes (11–
16), body composition measures (11–19), and 1 with nutritional
biomarkers (11). They were conducted only in convenience
samples of children and adolescents aged 9–24 y. None of them
examined the relation of a dietary habits index with nutrient
intakes and nutritional biomarker levels in a nationally repre-
sentative sample of adolescents.
The aims of this study are 2-fold: 1) to develop Healthy
Dietary Habits Scores for Adolescents (HDHS-A)7 based on
dietary habits information; and 2) to examine the validity of the
HDHS-A based on its internal reliability and associations with
sociodemographic factors, nutrient intakes, and nutritional
biomarkers in adolescents aged 15–18 y who participated in
the 2008/2009 New Zealand (NZ) Adult Nutrition Survey.
Methods
Study design and sample. The study design included a secondary
analysis of the 2008/2009 NZ Adult Nutrition Survey, a cross-sectional
population-based survey in a representative sample of New Zealanders
7
Abbreviations used: DHQ, dietary habits questionnaire; HDHS-A, Healthy
Dietary Habits Score for Adolescents; NZ, New Zealand; NZDep2006, 2006 New
Zealand Deprivation Index; NZEO, New Zealand European and others.

ã 2014 American Society for Nutrition.

Manuscript received November 14, 2013. Initial review completed January 5, 2014. Revision accepted March 14, 2014. 937
First published online April 17, 2014; doi:10.3945/jn.113.188375.
aged $15 y. The methodology for this national nutrition survey was
granted ethical approval from the NZ Health and Disability Multi-Region
Ethics Committee and was described in detail previously (20). Briefly, a
multistage, stratified, probability-proportional-to-size sampling was used,
and 4721 home interviews were conducted from October 2008 to October
2009. Data from those aged 15–18 y were included in the present study.
Dietary intake. Dietary intake was assessed using a dietary habits
questionnaire (DHQ) and an interview-administered 24-h diet recall. The
DHQ containing 25 questions was used to collect information on key eating
habits associated with diet quality and nutritional status (21,22). This
questionnaire comprised frequency questions on the past 4-wk consump-
tion of breakfast, breads, meats (red meat, chicken, processed meat pro-
ducts, fish, or shellfish), fast foods or takeaways, french fries, sugary
beverages, and confectionery. The other questions focused on daily intake of
servings of fruit and vegetables and dietary practices pertaining to the use of
fat and salt, e.g., removing excess fat from meat and adding salt to food on
the plate before consumption and their usual choices of types of bread, milk,
fat spread, and cooking fat. The participantsÕ responses were directly
entered into computer-assisted personal interview software (Abbey Re-
search Software, Life in New Zealand Health and Activity Research Unit,
University of Otago). The DHQ was cognitively tested for suitability in
assessing usual eating habits (20), and each component was used previously
in other large nationally representative studies in NZ (23,24).
Using a standardized computer-prompted protocol, participants were
each interviewed by a trained interviewer about their food intake in the
previous 24 h, including foods, beverages, and dietary supplements
consumed both at and away from home. Interviews were conducted on
weekdays and weekends using a multiple-pass technique with 4 stages.
First, a ‘‘quick list’’ of foods and beverages consumed on the previous day
was obtained. Detailed information of these foods, such as the brand and
product name, cooking method, recipe used, time consumed, and the place
the food was sourced, was collected in the second stage. In the third stage,
the quantities of food consumed were estimated using food photographs,
shape dimensions, food portion assessment aids, and packaging informa-
tion. Finally, items were reviewed in chronological order to allow additions
or changes to all aspects of the data. Food and beverages from the 24-h diet
recalls were matched to the NZ Food Composition Database, or overseas
food composition data when appropriate for nutrient estimation. Nutrient
intakes were calculated without including dietary supplements (20).
Development of the HDHS-A. The HDHS-A index was developed using
information from the DHQ after consultation with 2 expert nutritionists and
a dietitian. The index items were selected on the basis that they were relevant
to and captured the key nutrients that are important in determining the diet
quality of NZ adolescents, as guided by existing scientific evidence and the
NZ Food and Nutrition Guidelines for Healthy Children and Young People
(21). Four index prototypes were initially created. Using 21 questions from
the DHQ, the final prototype included 17 items grouped into 5 clusters: 1) fat
from meat, poultry, and fish; 2) other fats; 3) fruit, vegetables, and bread; 4)
sugar sources; and 5) meal habits. These clusters were named after main
nutrients or domains of diet reflected by the healthy dietary habits.
A response that aligned with a more positive dietary habit was
assigned a higher score using a 5-point scoring system ranging from 0 to 4.
The total HDHS-A was a summation of scores from the 17 items and
ranged from 0 to 68. A greater total score represents a dietary pattern
reflective of healthier dietary habits. The 17 items derived from the DHQ
and their respective scoring criteria are presented in Supplemental Table 1.
Sociodemographic information. Age was derived from the date of
birth and interview start date. Participants were asked to identify
$1 ethnic groups to which they belonged, and participants were
classified hierarchically into 3 ethnic groups: 1) Maori; 2) Pacific; and 3)
NZ European and others (NZEO) (25).
Socioeconomic status was estimated using an area-based scale of
deprivation, namely the 2006 NZ Deprivation Index (NZDep2006)
(26). This proxy measures deprivation based on 8 dimensions of
deprivation for each neighborhood (i.e., a mesh block containing ;87
people) in NZ, including income and benefit receipt, home ownership,
support for sole-parent families, employment status, qualifications,
938 Wong et al.
household number, communication, and transport (26). NZDep2006
scores were assigned to participants based on their home addresses and
divided into quintiles, in which quintile 1 represents participants with the
lowest amount of deprivation (i.e., from the least-deprived areas) and
quintile 5 represents participants with the most deprivation.
Anthropometric measurements. Heights and weights were measured
in the participantÕs home by trained interviewers using a standardized
protocol (20). Standing height was measured to the closest 0.1 cm using a
portable stadiometer (Seca 214; Seca), and body weight was measured in
minimum clothing to the nearest 0.1 kg using an electronic weighing scale
(model HD-351; Tanita). BMI was calculated by dividing weight in
kilograms by height in meters squared. Weight status was defined using the
International Obesity Taskforce age- and sex-specific BMI cutoffs for
children and adolescents (27,28).
Biochemical assessment. Within 2 wk after the home interviews,
nonfasting blood and spot midstream urine samples were obtained from
Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021
consenting participants at affiliated laboratories in local areas across
NZ. Blood and urine biomarkers of dietary status included in this study
were as follows: 1) whole-blood folate; 2) serum folate; 3) RBC folate; 4)
urinary iodine; 5) urinary sodium; and 6) urinary potassium.
Blood was collected from a forearm vein into 2 4-mL vacutainers
(EDTA1 and EDTA2) containing EDTA and 1 10-mL vacutainer with no
additive. All samples were stored at 4°C until analyzed. Complete blood
count was determined from EDTA1 at local laboratories. After trans-
portation to the Department of Human Nutrition, University of Otago,
whole-blood folate was analyzed from EDTA2 using a microbiologic assay
on 96-well microtiter plates with chloramphenicol-resistant Lactobacillus
casei as the test microorganism (29). Using the same method, serum folate
was analyzed using serum separated from vacutainers without additive.
RBC folate was calculated by subtracting serum folate from whole-blood
folate and correcting for hematocrit (30). RBC concentration of folate was
shown previously to be a good indicator of long-term folate status (31).
Spot urine samples were analyzed at the Canterbury Health
Laboratories for urinary sodium, potassium, and iodine. Urinary sodium
and potassium were analyzed using an ion-selective electrode analyzer
with integrated chip technology (Abbott Architect C8000 biochemical
analyzer), whereas urinary iodine was determined by the Sandell-
Kolthoff colorimetric method (32).
Statistical analysis. Participants aged 15–18 y who completed a single
24-h diet recall and at least 75% of the DHQ (i.e., 19 of 25 questions) were
considered eligible for the analysis (n = 695). Data for eligible participants
with 1–3 missing responses (n = 103), entered either as ‘‘did not answer’’ or
‘‘donÕt know,’’ were estimated from other information that the participant
supplied as long as at least 50% of questions related to the items within a
cluster were completed. Mean values of nonmissing items in a cluster were
estimated and substituted for the missing values (1 participant was
excluded because >50% of the data were missing in 1 of the clusters).
Because the data were part of a nationwide survey, the Stata survey
procedures, which account for the complex structure of the sample and the
sampling weights, were used to analyze the data. These provide the correct
SE for the estimates. A natural logarithm was applied to transform non-
normally distributed data. Continuous variables were presented as means
or geometric means if log-transformed and SEs or 95% CIs. Categorical
data were presented as relative frequencies and percentage. Survey post-
estimation tests (lincom command in Stata) were used to compare
continuous variables (e.g., age, BMI Z-score, HDHS-A score) between the
sexes. All analyses were performed using the statistical software package
Stata (version 11.2; StataCorp), with statistical significance set at P < 0.05.
Evaluation of the HDHS-A. Content validity of the HDHS-A was
established through an expert review to ensure that all 17 items of the
index were important dietary habits contributing to the diet quality of
NZ adolescents. The index was also tested for internal reliability using
correlations and CronbachÕs a coefficient. To examine construct validity,
3 hypotheses were generated with regard to the associations between
HDHS-A scores and sociodemographic factors, 24-h nutrient intakes,
and nutritional biomarkers. It was hypothesized that higher HDHS-A
scores would be found in participants with the following: 1) lower TABLE 1 Association between HDHS-A score and socio-
quintiles of NZDep2006; 2) more desirable nutrient profiles from the 24-h demographic factors1
diet recall; and 3) more favorable blood and urinary biomarker profiles.
Regression analyses were used to examine the associations between Participants HDHS-A score P2
HDHS-A scores and sociodemographic variables (sex, age group,
ethnicity, NZDep2006 quintile, BMI categories).When an association Total 694 43.7 6 0.4
was significant, post hoc linear combinations with BonferroniÕs correc- Sex ,0.001
tions were used to examine the differences among the groups. After Males 325 (47) 42.5 6 0.5a
categorizing total HDHS-A into thirds of low, medium, and high scores, Females 369 (53) 45.0 6 0.5b
ratios of nutrients-to-energy were calculated. A similar nutrient density Age (y) 0.84
approach for energy adjustment was adopted previously (33,34). Post 15 148 (21) 44.2 6 0.7
hoc trend analyses were used to identify the presence of linear trend in
16 214 (31) 43.7 6 0.5
nutrient intakes across the thirds of HDHS-A. To examine the associ-
17 181 (26) 43.8 6 0.7
ation between nutritional biomarkers and HDHS-A score, trend analyses
18 151 (22) 43.2 6 0.9
were conducted using biomarkers as the dependent variables and HDHS-
A thirds as the independent variables in the regression models. Ethnicity ,0.001
NZEO 521 (75) 45.0 6 0.4a
Maori 110 (16) 39.8 6 0.7b

Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021

Results Pacific 63 (9) 41.5 6 1.1b
NZDep2006 quintile ,0.0014
Participant characteristics. A total of 694 participants aged
1 (least deprived) 133 (19) 45.5 6 0.9
15–18 y (325 males, 369 females) who completed the DHQ were
2 155 (22) 44.9 6 0.5
included in the study. The mean 6 SE age was 16.5 6 0.1 y.
3 123 (18) 44.2 6 0.9
Approximately 60% (n = 422, 54% females) of participants who
4 136 (20) 43.1 6 0.8
had complete blood (n = 421) and urine (n = 411) variables were
5 (most deprived) 147 (21) 40.6 6 0.7
included in the biomarker subgroup analysis. There were no
BMI categories3 0.52
significant differences in the sex distribution, age, BMI, and
Normal weight 394 (58) 43.5 6 0.4
HDHS-A scores between those excluded and included in the
Overweight 171 (25) 43.8 6 0.7
biomarker analyses (all P > 0.05; data not shown).
Obese 85 (13) 45.2 6 1.2

Internal reliability of the HDHS-A. The items in the HDHS-A 1

had low intercorrelations (0.10 # r # 0.13). The correlations
between individual items with the total score were highest for item 9
[i.e., intake of potato and kumara (root vegetable) fries], followed by
item 14 (i.e., soft drink or energy drink consumption) (Supplemental
Table 2). The overall CronbachÕs a coefficient of the HDHS-A was
0.69, indicating that the index had good internal reliability (7).
Values are n (%) or means 6 SEs weighted for the survey design. a,bSignificant differences
between the groups when tested using post hoc linear combinations with BonferroniÕs
adjustment (P , 0.017). HDHS-A, Healthy Dietary Habits Score for Adolescents; NZDep2006,
2006 New Zealand Deprivation Index; NZEO, New Zealand European and others.
2
Indicates level of significance in unadjusted regression models for the total sample.
3
As defined by Cole et al. (27,28).
4
Significant trend analysis.

HDHS-A scores vs. sociodemographic variables. The mean
6 SE HDHS-A was 44 6 0.4 (range, 16 to 66). HDHS-A score
was associated with sex (P < 0.001), ethnicity (P < 0.001), and
NZDep2006 quintile (P < 0.001) (Table 1). Females had significantly
higher total HDHS-A scores than males. NZEO had higher HDHS-
A scores than Maori (P < 0.001) and Pacific (P < 0.01) groups.
HDHS-A score vs. nutrient intakes and nutritional biomarkers.
The associations between HDHS-A scores and 24-h nutrient intakes
and nutritional biomarker levels are shown in Table 2. Because
HDHS-A score was negatively associated with energy intake
(r = 20.18, P < 0.001), all nutrients were adjusted for total energy
intake (in megajoules). Overall, higher relative intakes of protein,
dietary fiber, PUFA, and lactose and lower intakes of sucrose were
associated with increasing thirds of HDHS-A. Associations in the
expected directions were also found with most micronutrient
intakes, urinary sodium excretion, and whole-blood, serum, and
RBC folate concentrations (P < 0.05).
Discussion
This study describes the development and validation of a
behavioral-based diet index, namely the HDHS-A, derived
from self-reported dietary habits captured in a DHQ. The
strength of this simple diet index lies in its practicality, because it
is calculated based on dietary habits information that is easier to
obtain than quantitative food and nutrient intake data. The use
of 24-h diet recall and biomarker information allowed an
in-depth and independent evaluation of the HDHS-A. Using a
nationally representative sample and 2 independent methods for
validation of the HDHS-A, the results from this study can
therefore be generalized to NZ adolescents aged 15–18 y.
For the first time, it was demonstrated by means of a diet
index that the dietary habits of NZ adolescents differed by sex,
ethnicity, and socioeconomic status. Compared with males, fe-
males scored higher in the HDHS-A, particularly in the fol-
lowing 3 clusters: 1) fat from meat, poultry, and fish; 2) other
fats; and 3) fruit, vegetables, and bread (data not shown). This
finding is in agreement with previous studies of adolescents that
showed that females compared with males had intakes of fruit
(35) and meat and meat products (36) that were more in line with
dietary guidelines and better overall diet quality as measured by
the Healthy Eating Index (14,37) and Mediterranean Dietary
Score (38). The mean HDHS-A of the Maori and Pacific participants
were 12% and 8%, respectively, lower than their NZEO
counterparts. In addition, there was a trend for poorer dietary
habits with increasing amount of deprivation (P-trend < 0.001).
The associations of HDHS-A score with ethnicity and NZDep2006
were likely explained by the fact that a higher proportion of Maori
and Pacific participants experience a higher amount of deprivation
(P < 0.001; data not shown). In a previous regional study in
Auckland, Maori and Pacific adolescents appeared to have less
satisfactory dietary intakes as marked by higher fat intake and
larger-than-standard portions of most food items but fewer daily
servings of vegetables and cheese compared with NZ Euro-
pean and Asian adolescents (39). Considering dietary habits as a
Healthy dietary habits score for adolescents 939
 TABLE 2 Adjusted nutrient intakes and nutritional biomarker levels by thirds of HDHS-A1

Thirds of HDHS-A
Low Medium High P-trend2

HDHS-A 36 (35, 36) 45 (45, 45) 53 (53, 54) ,0.001

Energy, MJ 10 (9.5, 11.4) 9.6 (8.9, 10.3) 8.4 (7.9, 9.0) ,0.001
Protein, % energy 15 (14, 16) 16 (15, 16) 17 (16, 17) ,0.001
Carbohydrate, % energy 49 (48, 51) 49 (48, 51) 50 (48, 52) 0.95
Fat, % energy 34 (33, 36) 34 (33, 36) 34 (32, 35) 0.60
SFA, % energy 14 (14, 15) 14 (13, 15) 14 (13, 15) 0.18
U:S ratio 1.2 (1.2, 1.3) 1.3 (1.2, 1.4) 1.4 (1.3, 1.5) 0.009
Protein, g/MJ 8.9 (8.4, 9.5) 9.3 (8.6, 10.0) 10.0 (9.6, 10.4) 0.006
Total fat, g/MJ 9.3 (8.8, 9.8) 9.1 (8.8, 9.5) 9.0 (8.6, 9.4) 0.48
Total carbohydrate, g/MJ 29 (28, 30) 29 (28, 30) 30 (28, 31) 0.51
Dietary fiber, g/MJ 1.8 (1.7, 1.9) 1.9 (1.8, 2.1) 2.4 (2.2, 2.5) ,0.001
Cholesterol, g/MJ

Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021

32 (27, 36) 28 (25, 30) 28 (25, 31) 0.12
SFA, g/MJ 3.9 (3.6, 4.1) 3.8 (3.6, 3.9) 3.7 (3.4, 3.9) 0.23
MUFA, g/MJ 3.4 (3.2, 3.5) 3.3 (3.1, 3.5) 3.2 (3.0, 3.3) 0.10
PUFA, g/MJ 1.1 (1.1, 1.2) 1.2 (1.1, 1.2) 1.3 (1.2, 1.4) ,0.001
Fructose, g/MJ 2.5 (2.2, 2.8) 2.4 (2.1, 2.6) 2.4 (2.2, 2.7) 0.57
Glucose, g/MJ 2.4 (2.1, 2.7) 2.2 (1.9, 2.4) 2.2 (2.0, 2.4) 0.18
Lactose, g/MJ 1.2 (1.0, 1.3) 1.5 (1.3, 1.7) 1.7 (1.5, 1.8) ,0.001
Maltose, g/MJ 0.6 (0.5, 0.6) 0.4 (0.4, 0.5) 0.5 (0.5, 0.6) 0.31
Sucrose, g/MJ 7.4 (6.4, 8.5) 7.0 (6.3, 7.8) 6.1 (5.5, 6.7) 0.03
Total sugars, g/MJ 14 (13, 16) 14 (13, 15) 13 (12, 14) 0.17
Calcium, mg/MJ 86 (76, 97) 87 (80, 95) 106 (98, 114) 0.004
Iron, mg/MJ 1.1 (1.1, 1.2) 1.2 (1.1, 1.3) 1.4 (1.2, 1.4) ,0.001
Magnesium, mg/MJ 26 (25, 27) 29 (28, 30) 33 (32, 34) ,0.001
Phosphorus, mg/MJ 137 (127, 147) 143 (135, 152) 159 (154, 165) ,0.001
Potassium, mg/MJ 268 (254, 281) 295 (278, 313) 328 (313, 344) ,0.001
Selenium, mg/MJ 5.2 (4.6, 5.8) 5.6 (5.0, 6.3) 6.3 (5.7, 6.9) 0.015
Zinc, mg/MJ 1.2 (1.1, 1.3) 1.2 (1.1, 1.3) 1.2 (1.2, 1.3) 0.42
b-Carotene equivalents, mg/MJ 158 (138, 178) 236 (195, 277) 287 (251, 324) ,0.001
Retinol, mg/MJ 40 (35, 45) 36 (33, 39) 38 (34, 42) 0.51
Total vitamin A, mg/MJ 66 (60, 73) 76 (68, 83) 86 (80, 92) ,0.001
Thiamin, mg/MJ 0.1 (0.1, 0.2) 0.2 (0.1, 0.2) 0.2 (0.2, 0.2) ,0.001
Riboflavin, mg/MJ 0.2 (0.2, 0.2) 0.2 (0.2, 0.2) 0.2 (0.2, 0.3) 0.002
Niacin, mg/MJ 2.2 (1.9, 2.4) 1.9 (1.8, 2.1) 2.2 (2.0, 2.3) 0.91
Vitamin B-6, mg/MJ 0.3 (0.2, 0.4) 0.2 (0.2, 0.2) 0.3 (0.2, 0.3) 0.35
Vitamin B-12, mg/MJ 0.5 (0.4, 0.5) 0.4 (0.3, 0.4) 0.4 (0.4, 0.5) 0.49
Vitamin C, mg/MJ 10 (8, 11) 10 (9, 12) 13 (11, 14) 0.005
Vitamin E, mg/MJ 1.0 (0.9, 1.1) 1.0 (0.9, 1.0) 1.0 (1.0, 1.1) 0.496
Urinary iodine, mmol/L 0.4 (0.4, 0.5) 0.5 (0.4, 0.6) 0.4 (0.4, 0.5) 0.56
Urinary sodium, mmol/L 142 (131, 154) 112 (100, 125) 101 (90, 112) ,0.001
Urinary potassium, mmol/L 72 (67, 78) 68 (61, 76) 66 (59, 74) 0.29
Whole-blood folate, nmol/L 293 (268, 319) 306 (285, 330) 340 (312, 370) 0.017
Serum folate, nmol/L 19 (17, 21) 22 (20, 24) 25 (22, 28) 0.002
RBC folate, nmol/L 671 (617, 730) 689 (639, 744) 790 (722, 865) 0.012
1
Values are means (95% CIs) weighted for the survey design. n = 694, participants who completed the dietary habits questionnaire; n = 422,
participants who supplied a blood sample; n = 411, participants who supplied a urine sample. For all dietary variables, n = 255 for Low, n = 232
for Medium, and n = 207 for High categories. For all other variables, n differs for each parameter in the Low, Medium, and High categories. For
biomarker variables (urinary iodine, urinary sodium, urinary potassium, whole-blood folate, serum folate, and RBC folate), values are geometric
means (95% CIs). HDHS-A, Healthy Dietary Habits Score for Adolescents; U:S ratio, unsaturated fat (PUFA + MUFA)-to-SFA ratio.
2
Linear trend analysis using survey post-estimation.

whole by using a diet index approach, this study confirms
findings of previous studies that disparities in diet quality of NZ
adolescents are associated with ethnicity and socioeconomic
status. This also provides evidence of the construct validity of
the HDHS-A because it distinguishes differences in diet quality
among ethnic and socioeconomic groups in NZ.
HDHS-A scores were significantly associated with more
favorable intake profiles of dietary fiber, PUFA, lactose, sucrose,
940 Wong et al.
calcium, iron, magnesium, phosphorus, potassium, selenium,
and vitamins A, B-1 (thiamin), B-2 (riboflavin), and C. Of
particular interest was the observed higher lactose but lower
sucrose intakes. This finding is reflective of the index scoring
system that rewards milk consumption but penalizes intakes of
sugary foods, such as fruit juice, soft drinks, and confectionery
(Supplemental Table 1). In addition, the HDHS-A seems to be a
good indicator of types of fat in the diet. Although the total fat
intake was constant within the acceptable macronutrient distri-
bution range of 20–35% of total energy (40), the unsaturated-
to-saturated fat ratio increased significantly across the thirds of
HDHS-A (P < 0.001). This increasing fat ratio reflected a higher
proportion of PUFA intake relative to SFA intake, i.e., a better
fat quality across thirds of HDHS-A.
Four studies examined a diet index in relation to nutritional
biomarkers in children and adolescents (11,41–43). Recently,
Vyncke et al. (42,43) studied the Flemish food-based Diet
Quality Index for Adolescents in relation to a comprehensive
range of blood biomarkers in a large multicenter cohort of
European adolescents [the HELENA study (for Healthy Lifestyle
in Europe by Nutrition in Adolescence)]. The Diet Quality Index
for Adolescents was strongly associated with 2 nutritional
biomarkers of long-term intake of vitamin D and vitamin B-12
and weakly associated with some serum FAs, such as EPA and
DHA (42,43). However, they found no association with plasma
folate (42), an indicator of acute folate status (44). In contrast,
this study found significant associations between HDHS-A
scores and 3 indices of folate status (i.e., serum, whole-blood,
and RBC folate concentrations), suggesting that better short-
and long-term folate status is associated with diet quality. This
seems consistent with studies in adolescents and adults that also
showed that a higher diet quality, using a diet index, was
positively associated with either a higher serum folate (41,45) or
both serum and RBC folate concentration (46).
Given that discretionary use of salt cannot be accurately
measured and the lack of a reliable local food composition
database for dietary sodium analysis (20), urinary sodium was
used as an indicator of sodium intake. Although dietary habits
relating to sodium intake were not represented in the construct
of the HDHS-A, a higher HDHS-A score was significantly
associated with lower urinary sodium. This significant inverse
association between a diet quality index and a biomarker of
sodium has not been reported previously in the literature. Good
habits with respect to sodium intake may also be practiced by
participants with healthful dietary habits. Previous studies
showed that health-promoting behaviors are likely to coexist
(47,48), and our results may reflect such a phenomenon.
There are several limitations to this study. A single 24-h diet
recall provides a snapshot of dietary intake but does not rep-
resent the individualÕs habitual dietary intake, particularly for
adolescents who are also known to have large within- and
between-participant variability in intake (49). Nevertheless, single
24-h diet recalls are appropriate for estimating group means of the
nutrient intake of a population when conducted within a random
sample (50). As with other self-reported dietary methods, 24-diet
recall data are susceptible to measurement errors related to
forgetfulness and poor portion size estimation among adolescents
(49,51,52). As part of a sensitivity analysis in this study, data were
analyzed using a subsample with plausible energy reporting
(n = 416), i.e., those with a ratio of total energy intake-to-basal
metabolic rate of >1.1 (53). After excluding energy under-
reporters, the associations between HDHS-A scores and nutrients
remained significant for most nutrients, except for lactose, selenium,
riboflavin, niacin, and vitamin C. In addition, significant inverse
associations were found for total carbohydrate (P-trend < 0.05),
SFA (P-trend < 0.05), and total sugars (P-trend < 0.001) across
the thirds of HDHS-A. To preserve the representativeness of the
survey data, possible underreporters were included in the current
analysis. Therefore, the results presented represent a conservative
estimate of the validity of the HDHS-A.
The scoring system of the HDHS-A may have implications for
its usefulness. Higher scores were assigned to dietary habits that
conform more closely to the NZ Food and Nutrition Guidelines
for Healthy Children and Young People (21). However, it must be
noted that the scoring of the HDHS-A remains arbitrary for some
items, because not all dietary habits could be ordered on a scale
according to their healthfulness. The scoring may be less sensitive
in discerning subtleties in diet quality when a dietary habit was
defined as neither optimum nor poor. Because of the cross-sectional
nature of this study, reliability or temporal stability of the HDHS-A
was not examined. Future work should examine the predictive
capacity of the HDHS-A in relation to health outcomes in
longitudinal studies. Contrary to most diet indices that use
quantitative measures of nutrients and foods, the current index is
based on 17 key dietary habits arranged in 5 dimensions or clusters.
The HDHS-A measures both desirable and nondesirable dietary
habits that are of nutritional concern in adolescence. The index also
penalizes undesirable dietary habits relevant to youth culture, such as
Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021
frequent intake of processed meat and soft drinks and purchasing
food away from home (54,55). The significant associations between
HDHS-A score and more favorable nutrient and biomarker profiles
demonstrated the face and construct validity of the index.
Overall, this study demonstrated that adoption of more health-
ful dietary habits, as measured by the HDHS-A, was associated
with being female, not of Maori or Pacific ethnicity, and experienc-
ing the least deprivation. The HDHS-A is valid as an indicator of
diet quality because a higher diet score was associated with a
more favorable nutrient and biomarker profiles. Derived from
dietary habit questions, the HDHS-A is simple and practical to
use and hence may be applied to small- and large-scale studies of
adolescents in NZ, including those not collecting in-depth dietary
intake data. Future studies are recommended to establish the
predictive validity of HDHS-A in assessing health outcomes.
Acknowledgments
J.E.W., P.M.L.S., and W.R.P., designed the study; J.E.W. and
S.M.W. analyzed the data; J.E.W. drafted the paper; and W.R.P.
had responsibility for the final content. All authors read and
approved the final manuscript.
Literature Cited
1. Deshmukh-Taskar PR, Nicklas TA, OÕNeil CE, Keast DR, Radcliffe JD,
Cho S. The relationship of breakfast skipping and type of breakfast
consumption with nutrient intake and weight status in children and
adolescents: the National Health and Nutrition Examination Survey
1999–2006. J Am Diet Assoc. 2010;110:869–78.
2. Rasmussen M, Krolner R, Klepp KI, Lytle L, Brug J, Bere E, Due P.
Determinants of fruit and vegetable consumption among children and
adolescents: a review of the literature. Part I: Quantitative studies. Int J
Behav Nutr Phys Act. 2006;3:22.
3. Sjöberg A, Hallberg L, Hoglund D, Hulthen L. Meal pattern, food
choice, nutrient intake and lifestyle factors in The Goteborg Adoles-
cence Study. Eur J Clin Nutr. 2003;57:1569–78.
4. Mikkilä V, Rasanen L, Raitakari OT, Pietinen P, Viikari J. Consistent
dietary patterns identified from childhood to adulthood: the cardiovas-
cular risk in Young Finns Study. Br J Nutr. 2005;93:923–31.
5. Hoelscher DM, Evans A, Parcel GS, Kelder SH. Designing effective nu-
trition interventions for adolescents. J Am Diet Assoc. 2002;102:S52–63.
6. Buzzard IM, Stanton CA, Figueiredo M, Fries EA, Nicholson R, Hogan
CJ, Danish SJ. Development and reproducibility of a brief food frequency
questionnaire for assessing the fat, fiber, and fruit and vegetable intakes of
rural adolescents. J Am Diet Assoc. 2001;101:1438–46.
7. Magarey A, Golley R, Spurrier N, Goodwin E, Ong F. Reliability and
validity of the ChildrenÕs Dietary Questionnaire; a new tool to measure
childrenÕs dietary patterns. Int J Pediatr Obes. 2009;4:257–65.
8. van Assema P, Brug J, Ronda G, Steenhuis I. The relative validity of a short
Dutch questionnaire as a means to categorize adults and adolescents to
total and saturated fat intake. J Hum Nutr Diet. 2001;14:377–90.
Healthy dietary habits score for adolescents 941
9. Yaroch AL, Resnicow K, Khan LK. Validity and reliability of qualitative
dietary fat index questionnaires: a review. J Am Diet Assoc. 2000;100:240–4.
10. Kontogianni MD, Farmaki AE, Vidra N, Sofrona S, Magkanari F,
Yannakoulia M. Associations between lifestyle patterns and body mass
index in a sample of Greek children and adolescents. J Am Diet Assoc.
2010;110:215–21.
11. Chiplonkar SA, Tupe R. Development of a diet quality index with
special reference to micronutrient adequacy for adolescent girls
consuming a lacto-vegetarian diet. J Am Diet Assoc. 2010;110:926–31.
12. Farajian P, Risvas G, Karasouli K, Pounis GD, Kastorini CM,
Panagiotakos DB, Zampelas A. Very high childhood obesity prevalence
and low adherence rates to the Mediterranean diet in Greek children:
the GRECO study. Atherosclerosis. 2011;217:525–30.
13. Feskanich D, Rockett HR, Colditz GA. Modifying the Healthy Eating
Index to assess diet quality in children and adolescents. J Am Diet
Assoc. 2004;104:1375–83.
14. Hurley KM, Oberlander SE, Merry BC, Wrobleski MM, Klassen AC,
Black MM. The healthy eating index and youth healthy eating index are
unique, nonredundant measures of diet quality among low-income,
African American adolescents. J Nutr. 2009;139:359–64.
15. Lazarou C, Panagiotakos DB, Matalas AL. Level of adherence to the
Mediterranean diet among children from Cyprus: the CYKIDS study.
Public Health Nutr. 2009;12:991–1000.
16. Serra-Majem L, Ribas L, Garcia A, Perez-Rodrigo C, Aranceta J.
Nutrient adequacy and Mediterranean Diet in Spanish school children
and adolescents. Eur J Clin Nutr. 2003;57(Suppl 1):S35–9.
17. Kosti RI, Panagiotakos DB, Mariolis A, Zampelas A, Athanasopoulos P,
Tountas Y. The Diet-Lifestyle Index evaluating the quality of eating and
lifestyle behaviours in relation to the prevalence of overweight/obesity
in adolescents. Int J Food Sci Nutr. 2009;60(Suppl 3):34–47.
18. Lazarou C, Panagiotakos DB, Spanoudis G, Matalas AL. E-KINDEX: a
dietary screening tool to assess childrenÕs obesogenic dietary habits.
J Am Coll Nutr. 2011;30:100–12.
19. Turconi G, Guarcello M, Maccarini L, Cignoli F, Setti S, Bazzano R,
Roggi C. Eating habits and behaviors, physical activity, nutritional and
food safety knowledge and beliefs in an adolescent Italian population.
J Am Coll Nutr. 2008;27:31–43.
20. University of Otago, Ministry of Health. Methodology report for the
2008/09 New Zealand Adult Nutrition Survey. Wellington, New
Zealand: Ministry of Health; 2011.
21. Ministry of Health. Food and Nutrition Guidelines for Healthy
Children and Young People (Aged 2–18 Years). A background paper.
Wellington, New Zealand: Ministry of Health; 2012.
22. Ministry of Health. Healthy Eating—Healthy Action Oranga Kai—
Oranga Pumau: a strategic framework. Wellington, New Zealand:
Ministry of Health; 2003.
23. Russell DG, Parnell WR, Wilson NC. NZ Food: NZ People. Key results
of the 1997 National Nutrition Survey. Wellington, New Zealand:
Ministry of Health; 1999.
24. Clinical Trials Research Unit, Synovate. A national survey of children and
young peopleÕs physical activity and dietary behaviours in New Zealand:
2008/09. Key findings. Wellington, New Zealand: Ministry of Health; 2010.
25. Statistics New Zealand. 2006 Census Data—QuickStats about New
ZealandÕs population and dwellings—tables. 2006 [cited 2013 Dec].
Available from: http://www.stats.govt.nz/Census/2006CensusHomePage/
QuickStats/quickstats-about-a-subject/nzs-population-and-dwellings.aspx.
26. Salmond C, Crampton P, King P, Waldegrave C. NZDep: a New Zealand
index of socioeconomic deprivation for individuals. Soc Sci Med. 2006;62:
1474–85.
27. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard
definition for child overweight and obesity worldwide: international
survey. BMJ. 2000;320:1240–3.
28. Cole TJ, Flegal KM, Nicholls D, Jackson AA. Body mass index cut offs
to define thinness in children and adolescents: international survey.
BMJ. 2007;335:194.
29. OÕBroin S, Kelleher B. Microbiological assay on microtitre plates of
folate in serum and red cells. J Clin Pathol. 1992;45:344–7.
30. Senti FR, Pilch SM. Analysis of folate data from the second National
Health and Nutrition Examination Survey (NHANES II). J Nutr.
1985;115:1398–402.
31. Institute of Medicine. Dietary reference intakes for thiamin, riboflavin,
niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and
choline. Washington: National Academies Press; 1998.
942 Wong et al.
32. Pino S, Fang SL, Braverman LE. Ammonium persulfate: a safe alternative
oxidizing reagent for measuring urinary iodine. Clin Chem. 1996;42:239–43.
33. Brown CC, Kipnis V, Freedman LS, Hartman AM, Schatzkin A,
Wacholder S. Energy adjustment methods for nutritional epidemiology:
the effect of categorization. Am J Epidemiol. 1994;139:323–38.
34. Kipnis V. Structure of dietary measurement error: results of the OPEN
Biomarker Study. Am J Epidemiol. 2003;158:14–21.
35. Savige GS, Ball K, Worsley A, Crawford D. Food intake patterns among
Australian adolescents. Asia Pac J Clin Nutr. 2007;16:738–47.
36. Kobe H, Stimec M, Ribic CH, Fidler Mis N. Food intake in Slovenian
adolescents and adherence to the Optimized Mixed Diet: a nationally
representative study. Public Health Nutr. 2012;15:600–8.
37. de Andrade SC, de Azevedo Barros MB, Carandina L, Goldbaum M, Cesar
CL, Fisberg RM. Dietary quality index and associated factors among
adolescents of the state of Sao Paulo, Brazil. J Pediatr. 2010;156:456–60.
38. Martı́nez E, Llull R, Del Mar Bibiloni M, Pons A, Tur JA. Adherence to
the Mediterranean dietary pattern among Balearic Islands adolescents.
Br J Nutr. 2010;103:1657–64.
Downloaded from https://academic.oup.com/jn/article/144/6/937/4616048 by guest on 15 February 2021
39. Sluyter JD, Schaaf D, Metcalf PA, Scragg RK. Dietary intakes of Pacific,
Maori, Asian and European adolescents: the Auckland High School
Heart Survey. Aust N Z J Public Health. 2010;34:32–7.
40. National Health and Medical Research Council. Nutrient reference values
for Australia and New Zealand including recommended dietary intakes.
Canberra, Australia: National Health and Medical Research Council; 2006.
41. Kleiser C, Mensink GB, Scheidt-Nave C, Kurth BM. HuSKY: a healthy
nutrition score based on food intake of children and adolescents in
Germany. Br J Nutr. 2009;102:610–8.
42. Vyncke K, Cruz Fernandez E, Fajo-Pascual M, Cuenca-Garcia M, De
Keyzer W, Gonzalez-Gross M, Moreno LA, Beghin L, Breidenassel C,
Kersting, M, et al. Validation of the Diet Quality Index for Adolescents
by comparison with biomarkers, nutrient and food intakes: the
HELENA study. Br J Nutr. 2013;109:2067–78.
43. Vyncke KE, Huybrechts I, Dallongeville J, Mouratidou T, Van Winckel MA,
Cuenca-Garcia M, Ottevaere C, Gonzalez-Gross M, Moreno LA, Kafatos
AG, et al. Intake and serum profile of fatty acids are weakly correlated with
global dietary quality in European adolescents. Nutrition. 2013;29:411–9.
44. Gibson RS. Reproducibility in dietary assessment. Principles of Nutritional
Assessment. 2nd ed. New York: Oxford University Press; 2005:129–48.
45. Kant AK, Graubard BI. A comparison of three dietary pattern indexes for
predicting biomarkers of diet and disease. J Am Coll Nutr. 2005;24:294–303.
46. Weinstein SJ, Vogt TM, Gerrior SA. Healthy Eating Index scores are
associated with blood nutrient concentrations in the third National Health
and Nutrition Examination Survey. J Am Diet Assoc. 2004;104:576–84.
47. Ottevaere C, Huybrechts I, Benser J, De Bourdeaudhuij I, Cuenca-Garcia
M, Dallongeville J, Zaccaria M, Gottrand F, Kersting M, Rey-López JP,
et al. Clustering patterns of physical activity, sedentary and dietary
behavior among European adolescents: the HELENA study. BMC Public
Health. 2011;11:328.
48. Sanchez A, Norman GJ, Sallis JF, Calfas KJ, Cella J, Patrick K. Patterns
and correlates of physical activity and nutrition behaviors in adoles-
cents. Am J Prev Med. 2007;32:124–30.
49. Livingstone MB, Robson PJ. Measurement of dietary intake in children.
Proc Nutr Soc. 2000;59:279–93.
50. McPherson RS, Hoelscher DM, Alexander M, Scanlon KS, Serdula MK.
Dietary assessment methods among school-aged children: validity and
reliability. Prev Med. 2000;31:S11–33.
51. Collins CE, Watson J, Burrows T. Measuring dietary intake in children
and adolescents in the context of overweight and obesity. Int J Obes
(Lond). 2010;34:1103–15.
52. Livingstone MB, Robson PJ, Wallace JM. Issues in dietary intake assessment
of children and adolescents. Br J Nutr. 2004;92(Suppl 2):S213–22.
53. Goldberg GR, Black AE, Jebb SA, Cole TJ, Murgatroyd PR, Coward
WA, Prentice AM. Critical evaluation of energy intake data using
fundamental principles of energy physiology: 1. Derivation of cut-off
limits to identify under-recording. Eur J Clin Nutr. 1991;45:569–81.
54. Gillis LJ, Bar-Or O. Food away from home, sugar-sweetened drink
consumption and juvenile obesity. J Am Coll Nutr. 2003;22:539–45.
55. Taveras EM, Berkey CS, Rifas-Shiman SL, Ludwig DS, Rockett HR,
Field AE, Colditz GA, Gillman MW. Association of consumption of
fried food away from home with body mass index and diet quality in
older children and adolescents. Pediatrics. 2005;116:e518–24.