Abstract
Objectives:
Here we examined the reproducibility and validity of a dietary screener which was translated and adapted to assess diet quality among pregnant Nepalese women.
Methods:
A pilot cohort of singleton pregnant women (N=101; age 25.9±4.1 years) was recruited from a tertiary, periurban hospital in Nepal. An adapted Nepali version of the PrimeScreen questionnaire, a brief 21-item dietary screener that assesses weekly consumption of 12 healthy and 9 unhealthy food groups, was administered twice, and a month apart, in both the 2nd and 3rd trimesters. Up to four inconsecutive 24-hr dietary recalls (24-HDRs) were completed each trimester and utilized as the reference method for validation. For each trimester, data from multiple 24-HDRs were averaged across days, and items were grouped to match the classification and three weekly consumption categories (0-1, 2-3, or 4+ servings/week) of the 21 food groups represented on the PrimeScreen.
Results:
Gwet’s agreement coefficients (AC1) were used to evaluate the reproducibility and validity of the adapted PrimeScreen against the 24-HDRs in both the 2nd and 3rd trimester. AC1 indicated good to excellent (≥0.6) reproducibility for the majority (85%) of food groups across trimesters. Moderate to excellent validity estimates (AC1 ≥ 0.4) were found for all food groups except for fruits and vegetables in the 2nd trimester, and green leafy vegetables and eggs in both the 2nd and 3rd trimesters.
Conclusions:
The modified PrimeScreen questionnaire appears to be a reasonably valid and reliable instrument for assessing the dietary intake of most food groups among pregnant women in Nepal.
Keywords: Dietary assessment, reproducibility, validity, Nepal, Pregnant women
Introduction
The diverse South Asian country of Nepal has made significant improvements in nutrition policies and programs for pregnant women over the past two decades (Cunningham, Headey, Singh, Karmacharya, & Rana, 2017). While the prevalence of underweight (body mass index (BMI) <18.5kg/m2) among women of childbearing age in Nepal declined from 24% in 2006 to 17% in 2016, rates of overweight and obesity among this group increased from 9% to 22% during this time (Ministry of Health, 2017). Concurrent changes in diet due to rapid urbanization and economic growth may partly explain the increasing overnutrition and BMI among women of childbearing age in Nepal (Cunningham et al., 2017; Wei et al., 2019). Patterns and changes in diet are important to consider in this population as maternal diet is associated with several pregnancy outcomes, and long-term cardiometabolic health in both women and children (Kominiarek & Peaceman, 2017). In many low-income countries like Nepal, the impact of diet during pregnancy on maternal and neonatal outcomes is not well examined, primarily due to lack of validated and culturally appropriate population-based tools to assess diet among this population.
To address this gap, we translated and adapted the PrimeScreen questionnaire, a brief dietary screener, for use among Nepalese pregnant women. The PrimeScreen questionnaire, which assesses the consumption frequency of 21 food groups classified as healthy or unhealthy, was selected for several reasons. First, it has been previously validated in other pregnant populations (Gicevic et al., 2018). Second, in addition to providing an overall measure of diet quality, it assesses consumption of major food groups known to be associated with both non-communicable diseases (NCDs) and maternal and fetal outcomes (Fung, Isanaka, Hu, & Willett, 2018; Gicevic et al., 2018; Gicevic, Mou, Bromage, Fung, & Willett, 2021). The PrimeScreen is also inexpensive and quick to complete, with low respondent burden, making it more suitable for the assessment of diet-disease associations within large population-based studies and pregnancy/birth cohort studies (Fung et al., 2018; Gicevic et al., 2018; Gicevic et al., 2021). In the present study, we aimed to examine the reproducibility and relative validity of the modified PrimeScreen questionnaire which was adapted for the Nepalese diet and administered in a prospective birth cohort study conducted in a periurban hospital in Nepal.
Materials and Methods
Study Design
This prospective cohort study enrolled a convenience sample of 101 pregnant women from the Obstetric Outpatient Department at Dhulikhel Hospital in Dhulikhel, Nepal between January and December 2019. The primary aim of the pilot cohort study is to validate dietary and physical activity assessment tools among Nepalese pregnant women, and to evaluate the feasibility of establishing a larger birth cohort study in this setting. Dhulikhel Hospital is a community-based tertiary level university hospital of Kathmandu University, which is located approximately 30 kilometers outside of Kathmandu, has a catchment population of 1.9 million people, and delivers an estimated 3,500 infants each year. A trained research assistant (RA) recruited eligible participants, completed initial screening, and obtained written informed consent. Inclusion criteria were; being 18 years or older, ≤14 weeks gestation at enrollment, and carrying a single fetus. At a baseline visit during the 1st trimester, the RA collected each participant’s sociodemographic and clinical information via structured questionnaires and medical chart review. Women were then followed throughout pregnancy and up to six weeks postpartum. Diet, lifestyle, and pregnancy data were collected by the RA at follow-up antenatal (ANC) visits via structured questionnaires; clinical and obstetric outcomes at delivery were abstracted via medical record review. The research was conducted in accord with prevailing ethical principles and was reviewed and approved by the Rutgers Newark Health Sciences Institutional Review Board (Pro2018001976) and the Ethical Review Board of the Kathmandu University School of Medical Sciences (102/18).
Assessment of Dietary Intake
The PrimeScreen Questionnaire
Usual dietary intake was assessed using a modified version of the PrimeScreen questionnaire, a 21-item dietary screener that assesses consumption of food groups commonly associated with NCDs (Fung et al., 2018; Gicevic et al., 2018; Gicevic et al., 2021). The PrimeScreen has been previously validated for use in the United States among non-pregnant female adults (Gicevic et al., 2021), adults in the primary care setting (L Rifas-Shiman et al., 2001), adults at risk for ischemic heart disease (Fung et al., 2018), and among pregnant women with gestational diabetes or hypertensive disorders (Gicevic et al., 2018). To adapt the PrimeScreen to the Nepalese diet and to make it culturally appropriate, the questionnaire was reviewed by a panel of experts (including researchers, nutritionists and dietitians) who determined relevant local food examples for each of the 21 food groups on the PrimeScreen. The questionnaire was also translated into Nepali and cognitively tested among eight local volunteers to ensure its suitability in capturing usual dietary habits in this population.
The modified PrimeScreen questionnaire was administered by the trained RA twice, about a month apart, in both the 2nd (15-27 weeks gestation) and 3rd (≥28 weeks gestation) trimester. The 21 food groups included on the questionnaire include 12 healthy food groups (e.g. fruits, vegetables, whole grains, etc.), and 9 unhealthy food groups (e.g. red meat, refined grains, sugar sweetened beverages, etc.). Weekly consumption frequency for each food group is ascertained for the past month, which is desirable as dietary habits may change within each trimester due to the psychological and physiological changes of pregnancy (Ancira-Moreno et al., 2020; Crozier, Robinson, Godfrey, Cooper, & Inskip, 2009; Rifas-Shiman et al., 2006) For each food group, the modified PrimeScreen offered three consumption frequency choices including: 0-1 servings per week, 2-3 servings per week, and 4 or more servings per week.
24-Hour Dietary Recall (24-HDR)
The repeated 24-HDR was utilized as the comparison method to validate the PrimeScreen questionnaire. The 24-HDR is commonly utilized in validation research (Allehdan, Tayyem, Agraib, Thekrallah, & Asali, 2019; Boparai, Singh, & Kathuria, 2018; Kuppuswamy, Venugopal, & Subramaniam, 2018; Voortman et al., 2020), as it is known to be an accurate and valid method when multiple days are assessed (Willett & Lenart, 2012). Up to four 24-HDRs were completed for each participant during both the 2nd and 3rd trimester. Using the 5-step multiple-pass approach, each of the 24-HDRs was conducted in person or via phone by the same trained RA, on non-consecutive and randomly selected week and weekend days (Conway, Ingwersen, Vinyard, & Moshfegh, 2003).
For each trimester, data from multiple 24-HDRs were averaged across days, and items were grouped to match the classification and three weekly consumption categories (0-1, 2-3, or 4+ servings/week) of the 21 food groups represented on the PrimeScreen. To achieve this, individual recalls were first analyzed by food item to calculate total servings consumed that day for each food item. The daily servings for each food item were then totaled and categorized by the 21 food groups represented on the PrimeScreen questionnaire. Next, the total daily servings for each food group were averaged across the multiple 24-HDRs obtained for each participant and were multiplied by seven to reflect weekly food group consumption rates. Finally, weekly food group consumption was coded into the three consumption frequency options on the PrimeScreen questionnaire (0-1, 2-3, or 4+ servings/week)
Sub-Group Variables
In order to test the robustness of our findings, we also evaluated whether the modified PrimeScreen questionnaire demonstrated variation in reproducibility and validity for sub-groups of the sample. Based on their previously demonstrated associations with dietary intake (Doyle, Borrmann, Grosser, Razum, & Spallek, 2017), we examined the potential impact of the following variables on the reproducibility and validity estimates: age (stratified by median; <26 vs. ≥26 years), ethnicity (Newar/Brahmin vs. Non-Newar/Brahmin), education level (<10 years vs. ≥10 years), nausea/vomiting by trimester (yes vs. no response), parity at enrollment (nulliparous vs. parous), and pre-pregnancy BMI at enrollment (<25kg/m2 vs. ≥25kg/m2).
Statistical Analysis
Our power calculations showed that a sample size of 84 was required to detect a moderate correlation of 0.3 between two food items with a power of 80%. We recruited an additional 17 women (total n= 101) to account for possible attrition. Demographic and clinical characteristics to describe the sample were presented as means±standard deviation (SD) for continuous parametric variables and n (%) for categorical variables.
Due to low reported frequencies within the sample, we combined the carrots and other vegetables food groups in our analyses. For reproducibility, the degree of agreement between the two PrimeScreen administrations in each trimester was evaluated by calculating the Gwet’s agreement coefficient (AC1) for each food group (Gwet, 2002). Gwet’s AC1 was used because the categorical response data on the PrimeScreen were skewed; AC1 statistic overcomes the noted limitations of the Cohen’s kappa statistic with skewed data and provides a more precise, stable agreement statistic (Gwet, 2002). In addition, percentage agreement was calculated based on the percentage of participants categorized into the same PrimeScreen response category across two visits in each trimester. Gwet’s AC1 was also used to validate the PrimeScreen data for each food group against the average intake data from multiple 24-HDRs in each trimester. In addition, the percentage agreement between the two dietary assessment methods was evaluated by calculating the percentage of participants categorized into the same response category across both methods. The reproducibility and validity (e.g. AC1 analyses) were interpreted based on cut-offs proposed by Cicchetti and Sparrow (Cicchetti & Sparrow, 1981): <0.40 (poor), 0.40-0.59 (fair/moderate), 0.60-0.74 (good), and ≥0.75 (excellent). The data was analyzed using Statistical Package for Social Sciences (SPSS) version 25 (“SPSS Statistics for Windows”). A p-value of <0.05 was considered statistically significant.
Results
The mean age of participants (N=101) was 25.9±4.1 years, and on average participants had completed 11.7±3.2 years of schooling (Table 1). Newars, who are indigenous to the study site Dhulikhel, were the predominant ethnicity represented in the sample (38.6%,), followed by Brahmin who represented 21.8% of the sample. The majority of the sample (55.4%) were normal weight prior to pregnancy, and slightly more than half of the sample were nulliparous (Table 1). Out of 101 participants enrolled at baseline, 88 participants completed the follow up questionnaires, including both administrations of the modified PrimeScreen and at least two 24-HDR, in the 2nd trimester; 81 participants were retained in the 3rd trimester. No significant differences in demographic and clinical characteristics were found between women who did and did not complete the follow-up visits in the 2nd and 3rd trimesters.
Table 1.
Summary statistics of demographic and clinical characteristics of study participants at enrollment (N=101)
| Characteristic | Mean (SD) |
|---|---|
| Age (years) | 25.9 (4.1) |
| BMI, pre-pregnancy (kg/m2) | 24.3 (3.7) |
| Education level (years) | 11.8 (3.2) |
| Gestational age, baseline/1st trimester (weeks), N=101 | 10.6 (2.2) |
| Gestational age, 2nd trimester 1st visit (weeks), n=92 | 17.7 (2.3) |
| Gestational age, 2nd trimester 2nd visit (weeks), n=90 | 22.3 (2.2) |
| Gestational age, 3rd trimester 1st visit average (weeks), n=85 | 29.9 (1.7) |
| Gestational age, 3rd trimester 2nd visit average (weeks), n=83 | 34.5 (1.9) |
| BMI Category | n (%) |
| BMI <25kg/m2 | 60 (59.4) |
| BMI ≥25.0kg/m2 | 41 (40.6) |
| Average Monthly Income level (Nepali Rupees) | |
| 10,000 – 30,000 | 3 (3.0) |
| 30,000 – 50,000 | 58 (57.4) |
| >50,000 | 40 (39.6) |
| Ethnicity | |
| Newar | 39 (38.6) |
| Brahmin | 22 (21.8) |
| Magar/Tamang/Rai/Limbu | 20 (19.8) |
| Chetri/Thakuri/Sanyasi | 16 (15.8) |
| Kami/Damai/Sarki/Gaaine/Baadi | 4 (4.0) |
| Nausea or Vomiting | |
| 1st Trimester (only) | 67 (66.3) |
| 2nd Trimester (only) | 5 (5.0) |
| 1st and 2nd Trimesters (both) | 17 (16.8) |
| 1st and 2nd Trimesters (neither) | 12 (11.9) |
| Number of Prior Births | |
| 0 | 52 (51.5) |
| 1 | 46 (45.5) |
| 2 | 3 (3.0) |
Key: BMI, body mass index; SD, standard deviation
Consumption frequencies of the 21 food groups reported on the modified PrimeScreen are shown in Figures 1a–1d.
Figure 1.
a. Highest consumption (4+ servings) for healthy food groups across 24-HDR and two PrimeScreen questionnaires administered in the 2nd trimester.
Abbreviations: 24-HDR, 24-hour dietary recall
b. Highest consumption (4+ servings) for healthy food groups across 24-HDR and two PrimeScreen questionnaires administered in the 3rd trimester.
Abbreviations: 24-HDR, 24-hour dietary recall
c. Highest consumption (4+ servings) for unhealthy food groups across 24-HDR and two PrimeScreen questionnaires administered in the 2nd trimester.
Abbreviations: 24-HDR, 24-hour dietary recall; SSBs, sugar sweetened beverages
d. Highest consumption (4+ servings) for unhealthy food groups across 24-HDR and two PrimeScreen questionnaires administered in the 3rd trimester.
Abbreviations: 24-HDR, 24-hour dietary recall; SSBs, sugar sweetened beverages
With respect to food group intake, red meat, processed meat, poultry, fish, whole grains, baked goods, sugar sweetened beverages (SSBs), fried foods, and desserts had notably low rates of consumption among the sample, with the majority of women reporting intakes of only 0-1 servings/week (Table 2). Conversely, intake of other fruits, legumes, liquid vegetable oils, potatoes, whole milk dairy, and refined grains was consistently high (4+ servings/week) across both trimesters. Overall, consumption patterns for the food groups on the modified PrimeScreen were similar across the 2nd and 3rd trimester (Table 2). However, reported intake of both cruciferous vegetables and citrus fruits on the modified PrimeScreen decreased from the 2nd to the 3rd trimester.
Table 2.
Comparison of trimester-specific food group intake via 24-HDR and PrimeScreen (PS) methods
| Food Group | Instrument | 2nd trimester n (%) | 3rd trimester n (%) | ||||
|---|---|---|---|---|---|---|---|
| 0-1 servings | 2-3 servings | 4+ servings | 0-1 servings | 2-3 servings | 4+ servings | ||
|
Green Leafy Vegetables
|
24HDR | 9 (9.9) | 52 (57.1) | 30 (33.0) | 14 (16.1) | 47 (54.0) | 26 (29.9) |
| PS 1st Visit | 3 (3.3) | 30 (32.6) | 59 (64.1) | 1 (1.2) | 16 (18.8) | 68 (80.0) | |
| PS 2nd Visit | 0 (0.0) | 37 (41.1) | 53 (58.9) | 1 (1.2) | 12 (14.5) | 70 (84.3) | |
| Cruciferous Vegetables | 24HDR | 53 (58.2) | 35 (38.5) | 3 (3.3) | 73 (83.9) | 14 (16.1) | 0 (0.0) |
| PS 1st Visit | 3 (3.3) | 26 (28.3) | 63 (68.5) | 52 (61.2) | 10 (11.8) | 23 (27.1) | |
| PS 2nd Visit | 9 (10.0) | 7 (7.8) | 74 (82.2) | 81 (97.6) | 2 (2.4) | 0 (0.0) | |
| Other Vegetables | 24HDR | 45 (49.5) | 41 (45.1) | 5 (5.5) | 33 (37.9) | 41 (47.1) | 13 (14.9) |
| PS 1st Visit | 73 (79.3) | 18 (19.6) | 1 (1.1) | 52 (61.2) | 33 (38.8) | 0 (0.0) | |
| PS 2nd Visit | 73 (81.1) | 17 (18.9) | 0 (0.0) | 20 (24.1) | 63 (75.9) | 0 (0.0) | |
| Citrus Fruits | 24HDR | 79 (86.8) | 10 (11.0) | 2 (2.2) | 85 (97.7) | 2 (2.3) | 0 (0.0) |
| PS 1st Visit | 16 (17.4) | 52 (56.5) | 24 (26.1) | 84 (98.8) | 0 (0.0) | 1 (1.2) | |
| PS 2nd Visit | 52 (57.8) | 27 (30.0) | 11 (12.2) | 82 (98.8) | 1 (1.2) | 0 (0.0) | |
| Other Fruits | 24HDR | 11 (12.1) | 33 (36.3) | 47 (51.6) | 10 (11.5) | 16 (18.4) | 61 (70.1) |
| PS 1st Visit | 0 (0.0) | 15 (16.3) | 77 (83.7) | 0 (0.0) | 7 (8.2) | 78 (91.8) | |
| PS 2nd Visit | 0 (0.0) | 13 (14.4) | 77 (85.6) | 0 (0.0) | 6 (7.2) | 77 (92.8) | |
| Legumes | 24HDR | 3 (3.3) | 18 (19.8) | 70 (76.9) | 7 (8.0) | 16 (18.4) | 64 (73.6) |
| PS 1st Visit | 0 (0.0) | 34 (37.0) | 58 (63.0) | 1 (1.2) | 1 (1.2) | 83 (97.6) | |
| PS 2nd Visit | 1 (1.1) | 40 (44.4) | 49 (54.4) | 0 (0.0) | 1 (1.2) | 82 (98.8) | |
| Poultry | 24HDR | 88 (96.7) | 1 (1.1) | 2 (2.2) | 84 (96.6) | 2 (2.3) | 1 (1.1) |
| PS 1st Visit | 85 (92.4) | 6 (6.5) | 1 (1.1) | 81 (95.3) | 4 (4.7) | 0 (0.0) | |
| PS 2nd Visit | 88 (97.8) | 2 (2.2) | 0 (0.0) | 68 (81.9) | 12 (14.5) | 3 (3.6) | |
| Fish | 24HDR | 84 (92.3) | 4 (4.4) | 3 (3.3) | 83 (95.4) | 4 (4.6) | 0 (0.0) |
| PS 1st Visit | 92 (100.0) | 0 (0.0) | 0 (0.0) | 84 (98.8) | 1 (1.2) | 0 (0.0) | |
| PS 2nd Visit | 90 (100.0) | 0 (0.0) | 0 (0.0) | 83 (100.0) | 0 (0.0) | 0 (0.0) | |
| Eggs | 24HDR | 28 (30.8) | 47 (51.6) | 16 (17.6) | 30 (34.5) | 25 (28.7) | 32 (36.8) |
| PS 1st Visit | 10 (10.9) | 29 (31.5) | 53 (57.6) | 6 (7.1) | 25 (29.4) | 54 (63.5) | |
| PS 2nd Visit | 9 (10.0) | 37 (41.1) | 44 (48.9) | 5 (6.0) | 25 (30.1) | 53 (63.9) | |
| Whole Grains | 24HDR | 88 (96.7) | 3 (3.3) | 0 (0.0) | 84 (96.6) | 2 (2.3) | 1 (1.1) |
| PS 1st Visit | 91 (98.9) | 0 (0.0) | 1 (1.1) | 83 (97.6) | 2 (2.4) | 0 (0.0) | |
| PS 2nd Visit | 90 (100.0) | 0 (0.0) | 0 (0.0) | 83 (100.0) | 0 (0.0) | 0 (0.0) | |
| Vegetable Oils | 24HDR | 0 (0.0) | 5 (5.5) | 86 (94.5) | 0 (0.0) | 0 (0.0) | 87 (100.0) |
| PS 1st Visit | 0 (0.0) | 0 (0.0) | 92 (100.0) | 0 (0.0) | 1 (1.2) | 84 (98.8) | |
| PS 2nd Visit | 0 (0.0) | 0 (0.0) | 90 (100.0) | 0 (0.0) | 1 (1.2) | 82 (98.8) | |
| Red Meat | 24HDR | 74 (81.3) | 17 (18.7) | 0 (0.0) | 59 (67.8) | 23 (26.4) | 5 (5.7) |
| PS 1st Visit | 91 (98.9) | 1 (1.1) | 0 (0.0) | 83 (97.6) | 2 (2.4) | 0 (0.0) | |
| PS 2nd Visit | 89 (98.9) | 1 (1.1) | 0 (0.0) | 79 (95.2) | 4 (4.8) | 0 (0.0) | |
| Potatoes | 24HDR | 0 (0.0) | 15 (16.5) | 76 (83.5) | 1 (1.1) | 9 (10.3) | 77 (88.5) |
| PS 1st Visit | 0 (0.0) | 1 (1.1) | 91 (98.9) | 1 (1.2) | 2 (2.4) | 82 (96.5) | |
| PS 2nd Visit | 0 (0.0) | 2 (2.2) | 88 (97.8) | 1 (1.2) | 1 (1.2) | 81 (97.6) | |
| Processed Meat | 24HDR | 89 (97.8) | 2 (2.2) | 0 (0.0) | 86 (98.9) | 1 (1.1) | 0 (0.0) |
| PS 1st Visit | 92 (100.0) | 0 (0.0) | 0 (0.0) | 85 (100.0) | 0 (0.0) | 0 (0.0) | |
| PS 2nd Visit | 90 (100.0) | 0 (0.0) | 0 (0.0) | 82 (98.8) | 1 (1.2) | 0 (0.0) | |
| Whole Milk Dairy | 24HDR | 9 (9.9) | 23 (25.3) | 59 (64.8) | 7 (8.0) | 23 (26.4) | 57 (65.5) |
| PS 1st Visit | 12 (13.0) | 21 (22.8) | 59 (64.1) | 8 (9.4) | 11 (12.9) | 66 (77.6) | |
| PS 2nd Visit | 8 (8.9) | 15 (16.7) | 67 (74.4) | 9 (10.8) | 13 (26.5) | 61 (73.5) | |
| Refined Grains | 24HDR | 0 (0.0) | 0 (0.0) | 91 (100.0) | 0 (0.0) | 0 (0.0) | 87 (100.0) |
| PS 1st Visit | 0 (0.0) | 0 (0.0) | 92 (100.0) | 0 (0.0) | 0 (0.0) | 85 (100.0) | |
| PS 2nd Visit | 0 (0.0) | 1 (1.1) | 89 (98.9) | 0 (0.0) | 0 (0.0) | 83 (100.0) | |
| Baked Goods | 24HDR | 71 (78.0) | 17 (18.7) | 3 (3.3) | 65 (74.7) | 20 (23.0) | 2 (2.3) |
| PS 1st Visit | 82 (89.1) | 10 (10.9) | 0 (0.0) | 76 (89.4) | 9 (10.6) | 0 (0.0) | |
| PS 2nd Visit | 86 (95.6) | 4 (4.4) | 0 (0.0) | 74 (89.2) | 9 (10.8) | 0 (0.0) | |
| SSBs | 24HDR | 50 (54.9) | 36 (39.6) | 5 (5.5) | 57 (65.5) | 24 (27.6) | 6 (6.9) |
| PS 1st Visit | 90 (97.8) | 2 (2.2) | 0 (0.0) | 84 (98.8) | 1 (1.2) | 0 (0.0) | |
| PS 2nd Visit | 86 (95.6) | 4 (4.4) | 0 (0.0) | 78 (94.0) | 5 (6.0) | 0 (0.0) | |
| Fried Foods | 24HDR | 67 (73.6) | 13 (14.3) | 11 (12.1) | 63 (72.4) | 18 (20.7) | 6 (6.9) |
| PS 1st Visit | 79 (85.9) | 13 (14.1) | 0 (0.0) | 77 (90.6) | 8 (9.4) | 0 (0.0) | |
| PS 2nd Visit | 74 (82.2) | 15 (16.7) | 1 (1.1) | 79 (95.2) | 4 (4.8) | 0 (0.0) | |
| Desserts & Ice Cream | 24HDR | 62 (68.1) | 24 (26.4) | 5 (5.5) | 56 (64.4) | 22 (25.3) | 9 (10.3) |
| PS 1st Visit | 90 (97.8) | 2 (2.2) | 0 (0.0) | 85 (100.0) | 0 (0.0) | 0 (0.0) | |
| PS 2nd Visit | 90 (100.0) | 0 (0.0) | 0 (0.0) | 82 (98.8) | 1 (1.2) | 0 (0.0) | |
Key: 24HDR, 24-hr dietary recall; PS, PrimeScreen; SSBs, sugar sweetened beverages
Note: Bold typeface indicates majority intake for each food group by assessment tool and trimester.
Reproducibility of the Modified PrimeScreen in the 2nd and 3rd Trimester
In the 2nd trimester, the modified PrimeScreen demonstrated good to excellent reproducibility (AC1 ≥ 0.6) for the majority of food groups; the reproducibility was fair/moderate for green leafy vegetables, legumes, and whole milk dairy (AC1 = 0.40-0.59), and poor (AC1 < 0.4) for citrus fruits and eggs (Table 3). Percentage agreement across the two visits in the 2nd trimester ranged from a low of 41.1% (citrus fruits) to a high of 100.0% (for fish, vegetable oils, and processed meat). While poor agreement was noted for citrus fruits and eggs in the 2nd trimester, agreement improved for both food groups in the 3rd trimester. In the 3rd trimester, AC1 for reproducibility of the PrimeScreen ranged from 0.38 (poor agreement) to 1.00 (excellent agreement), with values ≥ 0.6, indicating good to excellent reproducibility, for 85% (n=17) of the food groups. Other vegetables was the only food group with poor reproducibility in the 3rd trimester (AC1< 0.4). In fact, we observed reproducibility coefficients ≥0.40 for 92.5% of food groups, and ≥0.60 for 80% of food groups, across the 2nd and 3rd trimester (mean:0.81; range 0.14-1.00). Percentage agreement across both visits in the 3rd trimester ranged from a low of 53.1% (other vegetables) to a high of 100.0% (refined grains). The fish, whole grains, vegetable oils, red meat, potatoes, processed meat, refined grains, SSBs, and dessert/ice cream food groups had consistently high percentage agreement (>90.0%) across both trimesters.
Table 3.
Comparison of validity, reproducibility, and % agreement for each food group by trimester
| Nutrient Variable | Reproducibility Analyses | Validity Analyses | ||||||
|---|---|---|---|---|---|---|---|---|
| 2nd Trimester (n=90) | 3rd Trimester (n=82) | 2nd Trimester (n=88) | 3rd Trimester (n=81) | |||||
| Gwet’s Alpha | Percent Agreement | Gwet’s Alpha | Percent Agreement | Gwet’s Alpha | Percent Agreement | Gwet’s Alpha | Percent Agreement | |
| Green Leafy Vegetables | 0.41 a | 55.6 | 0.65 a | 70.4 | 0.32a | 50.6 | 0.10 | 35.2 |
| Cruciferous Vegetables | 0.60 a | 67.8 | 0.49 a | 58.0 | −0.24 | 16.8 | 0.59 | 66.8 |
| Other Vegetables | 0.65 a | 71.1 | 0.38a | 53.1 | 0.37 | 51.7 | 0.41 a | 57.0 |
| Citrus Fruits | 0.14 | 41.1 | 0.98 a | 97.5 | 0.13 | 36.8 | 0.96 a | 96.4 |
| Other Fruits | 0.74 a | 77.8 | 0.88 a | 88.9 | 0.39a | 53.4 | 0.62 a | 67.9 |
| Legumes | 0.56 a | 66.7 | 0.96 a | 96.3 | 0.41 a | 54.5 | 0.70 a | 73.4 |
| Poultry | 0.92 a | 92.2 | 0.79 a | 81.5 | 0.91 a | 91.6 | 0.85 a | 86.1 |
| Fish | 1.00 a | 100.0 | 0.99 a | 98.8 | 0.92 a | 92.2 | 0.94 a | 94.6 |
| Eggs | 0.39a | 56.7 | 0.56 a | 66.7 | 0.14 | 41.1 | 0.20b | 44.9 |
| Whole Grains | 0.99 a | 98.9 | 0.98 a | 97.5 | 0.97 a | 96.7 | 0.95 a | 95.2 |
| Vegetable Oils | 1.00 a | 100.0 | 0.98 a | 97.5 | 0.94 a | 94.4 | 0.98 a | 98.8 |
| Red Meat | 0.98 a | 97.8 | 0.92 a | 92.6 | 0.79 a | 80.9 | 0.62 a | 67.9 |
| Potatoes | 0.97 a | 96.7 | 0.96 a | 96.3 | 0.81 a | 82.6 | 0.88 a | 89.1 |
| Processed Meat | 1.00 a | 100.0 | 0.99 a | 98.8 | 0.98 a | 97.8 | 0.98 a | 98.2 |
| Whole Milk Dairy | 0.59 a | 68.9 | 0.73 a | 77.8 | 0.56 a | 66.9 | 0.51 a | 62.5 |
| Refined Grains | 0.99 a | 98.9 | 1.00 a | 100.0 | 0.99 a | 99.5 | 1.00 a | 100.0 |
| Baked Goods | 0.86 a | 86.7 | 0.84 a | 85.2 | 0.71 a | 74.7 | 0.72 a | 75.8 |
| SSBs | 0.94 a | 94.4 | 0.95 a | 95.1 | 0.47 a | 57.4 | 0.58 a | 64.3 |
| Fried Foods | 0.72 a | 75.6 | 0.87 a | 87.7 | 0.59 a | 65.8 | 0.63 a | 68.5 |
| Desserts & Ice Cream | 0.98 a | 97.8 | 0.99 a | 98.8 | 0.63 a | 68.0 | 0.58 a | 64.3 |
Key: SSBs, sugar sweetened beverages
p-value <0.001
p-value <0.05
Note: Gwet’s agreement coefficients interpreted based on the cut-offs proposed by Cicchetti and Sparrow:(Cicchetti & Sparrow, 1981) <0.40 (poor agreement), 0.40-0.59 (fair/moderate agreement), 0.60-0.74 (good agreement), and ≥0.75 (excellent agreement).
Bold typeface indicates significant agreement coefficients.
Relative Validity of the Modified PrimeScreen in the 2nd and 3rd Trimester
Compared to multiple 24-HDRs, the modified PrimeScreen showed fair/moderate to excellent validity (AC1 ≥ 0.4) for all food groups except for green leafy vegetables, cruciferous vegetables, other vegetables, citrus fruits, other fruits, and eggs in the 2nd trimester (n=88) (Table 3). However, in the 3rd trimester (n=81) all food groups except green leafy vegetables and eggs had fair/moderate to excellent validity (AC1 ≥ 0.4). We observed validity coefficients ≥0.40 for 80% of food groups, and ≥0.60 for 57.5% of food groups across the 2nd and 3rd trimesters (mean: 0.67, range −0.24-to 1.00). Percentage agreement for PrimeScreen vs 24-HDRs in the 2nd trimester ranged from a low of 16.8% (cruciferous vegetables) to a high of 99.5% (refined grains), while agreement in the 3rd trimester ranged from a low of 35.2% (green leafy vegetables) to a high of 100.0% (refined grains). The fish, whole grains, vegetable oils, processed meat, and refined grains food groups had consistently high percentage agreement (>90.0%) across both the 2nd and 3rd trimesters.
For the sub-analyses completed, none of the selected variables (e.g. age, ethnicity, education level, nausea/vomiting, parity, and pre-pregnancy BMI) changed the reproducibility or validity estimates significantly (data not shown), although we cannot rule out that our low sample size may have limited our ability to discern any potential differences by sub-group.
Discussion
In the present study, we found the modified PrimeScreen questionnaire to be a reasonably reproducible and valid tool for assessing dietary intake among pregnant women in Nepal. To our knowledge, this diet quality screener is the first to be validated for use among pregnant women in Nepal, and one of the first tools validated for use among pregnant women from a low-income country. This questionnaire may have particular utility in large scale, population-based studies where detailed dietary intake assessment is not the primary focus, but instead the emphasis is on overall diet quality and the study of associations between specific food groups and health outcomes.
Due to rapid physiological and metabolic changes, dietary intake during pregnancy may vary dramatically across the three trimesters of pregnancy. However, only a few validation studies to date have reported on the timing of when a dietary questionnaire was administered during pregnancy, (Brunst et al., 2016; Zhang et al., 2015) and only one study assessed dietary data across both the second and third trimesters (Yuan et al., 2016). In the present study, we did not examine reproducibility and validity of the PrimeScreen in the 1st trimester, as the frequency of nausea and vomiting, and food safety concerns, may alter typical intake during early pregnancy (Ancira-Moreno et al., 2020; Crozier et al., 2009; Rifas-Shiman et al., 2006). However, we examined its reproducibility and validity separately across the second and third trimester, which is an approach that has not been previously utilized in the literature.
In comparison to prior literature examining the validity of dietary screeners among pregnant women (Hartman et al., 2016; Oken et al., 2014; Tsoi et al., 2020), our reproducibility and validity estimates were similar or slightly better than those previously reported. However, compared to other food groups on the modified PrimeScreen, the estimates were lower for certain fruits and vegetables, and eggs. Seasonality could impact the availability of certain fruits and vegetables in Nepal (Campbell et al., 2014); yet, even when they are in season, fruits and vegetables are often a small percentage of the overall diet (Campbell et al., 2014). Low consumption and high within-person variability may explain why weaker correlations were observed for certain fruit and vegetable groups (Willett & Lenart, 2012). Medical conditions including hypothyroidism and acid reflux which are common in the late 2nd to 3rd trimester, may have further contributed to higher within-person variability and weaker correlations for citrus fruits and cruciferous vegetables respectively (Ahmed, Im, Hwang, & Han, 2020; Devkota, Khan, Alam, Regmi, & Sapkota, 2016; Khakurel, Karki, & Chalise, 2021; Willett & Lenart, 2012).
Our study has notable strengths and makes several important contributions. This study is the first to examine the reproducibility and validity of a dietary screener among pregnant women in Nepal, and to do so by utilizing multiple 24-HDRs across both the 2nd and 3rd trimesters of pregnancy. Dietary information was collected prospectively using a standardized protocol. The PrimeScreen questionnaire was interviewer-administered by the same trained RA using the 5-step multiple-pass approach, which enhances the accuracy of the data collected (Conway et al., 2003; Yuan et al., 2016). Our selection of the PrimeScreen questionnaire was intentional, as it is a novel dietary assessment tool that is simple, quick to administer, and easy to interpret across population groups, offering an application to future health and disease research with its focus on food and dietary attributes associated with NCDs, and its ability to calculate an overall diet quality score (Fung et al., 2018; Gicevic et al., 2018; Gicevic et al., 2021).The reproducibility and validity of the PrimeScreen questionnaire was established in both the 2nd and 3rd trimesters, demonstrating its applicability for use among pregnant women in Nepal. Given the scarcity of validated dietary screeners in this population, a significant contribution of our study was to translate and adapt the PrimeScreen questionnaire to include culture-specific Nepalese foods for increased usability and acceptability. Nonetheless, as foods included in the diet may vary by geographic region, the PrimeScreen may require additional modification for use in other areas of the country.
Our study does have some limitations. First, the use of a small convenience sample from a single center means we cannot rule out selection bias, and findings from this sample may not be generalizable to all pregnant women in Nepal. Of note, Dhulikhel Hospital is a community-based private hospital, and thus is likely to serve a higher socioeconomic stratum of pregnant women compared to the majority of women in the country who receive their antenatal care at government hospitals. Compared to the general population in Nepal, our periurban sample was more educated, had relatively higher socioeconomic status, and had greater representation of the Newar ethnicity, but was comparable to national estimates in terms of marital status, employment, and pre-pregnancy BMI (Ministry of Health et al., 2017). Similar to other dietary screeners validated for use among pregnant women (Hartman et al., 2016; Tsoi et al., 2020), the PrimeScreen questionnaire was designed to measure food group intake and diet quality and was not intended to be used as a tool to calculate total energy or nutrient intake. Lastly, given that the PrimeScreen utilized in this study only reported participant intake within three weekly consumption categories, the tool did not capture potentially meaningful differences among women who reported consuming 4+servings/week. Therefore, as previously suggested, we recommend using more encompassing consumption categories (e.g. <1 serving/week, 1 serving/week, 2-4 servings/week, nearly daily or daily, and ≥2 servings/day) when using the PrimeScreen in the future (L Rifas-Shiman et al., 2001). To further explore the utility of this low-cost, low burden dietary tool, future studies should focus on validating this tool in the postpartum phase, and among low-income obstetric populations in Nepal and elsewhere.
Conclusion
The modified PrimeScreen questionnaire appears to be a reproducible and valid instrument for assessing the dietary intake of most food groups among pregnant women in Nepal. The validation of the PrimeScreen, a brief and culturally relevant dietary screener, is a critical step towards bridging the research gap on the links between prenatal nutrition and maternal and fetal outcomes in Nepal. In addition, it has the ability to assist in identifying dietary behaviors that can optimize maternal and fetal health outcomes, including long-term NCD risk. This connection is particularly important in a population where rates of overweight and obesity are increasing, and diet is transitioning to a Westernized pattern (Cunningham et al., 2017; Wei et al., 2019). Even though pregnancy is already recognized as a time of nutritional vulnerability (Akbari, Mansourian, & Kelishadi, 2015; Wood-Bradley, Henry, Vrselja, Newman, & Armitage, 2013), our findings support and facilitate continued global interest in validating and utilizing culturally-specific dietary assessment tools to examine the diets of pregnant women in far-reaching parts of the world.
Significance:
In many low-income countries like Nepal, the impact of diet during pregnancy on maternal and neonatal outcomes is not well examined due to a lack of brief, validated, and culturally appropriate dietary assessment tools. To our knowledge, this diet quality screener is the first to be validated for use among pregnant women in Nepal, and one of the first validated for use among pregnant women from a low-income country. This questionnaire may have particular utility in large scale, population-based studies with an emphasis on overall diet quality and the study of associations between specific food groups and health outcomes.
Funding:
This work was funded by Rutgers Global Health Institute and the National Institutes of Health/FIC under Grant # NIH 1R21TW011377-01.
Footnotes
Conflicts of Interest/Competing Interests: The authors declare that they have no conflict of interest.
Ethics Approval: The research was conducted in accord with prevailing ethical principles and was reviewed and approved by the Rutgers Newark Health Sciences Institutional Review Board (Pro2018001976) and the Ethical Review Board of the Kathmandu University School of Medical Sciences (102/18).
Consent to Participate: A trained research assistant obtained written informed consent from all participants included in this manuscript.
Consent for Publication: Not Applicable
Code Availability: Not Applicable
Availability of Data & Material:
Not Applicable
References
- Ahmed M, Im HB, Hwang JH, & Han D (2020). Disclosure of herbal medicine use to health care providers among pregnant women in Nepal: a cross-sectional study. BMC Complementary Medicine and Therapies, 20(1), 339. doi: 10.1186/s12906-020-03142-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Akbari Z, Mansourian M, & Kelishadi R (2015). Relationship of the intake of different food groups by pregnant mothers with the birth weight and gestational age: Need for public and individual educational programs. Journal of education and health promotion, 4, 23–23. doi: 10.4103/2277-9531.154109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Allehdan SS, Tayyem RF, Agraib LM, Thekrallah FM, & Asali FF (2019). Relative Validity and Reproducibility of a Food Frequency Questionnaire to Assess Food Group Intake in Pregnant Jordanian Women. J Acad Nutr Diet, 119(8), 1349–1361. doi: 10.1016/j.jand.2019.02.009 [DOI] [PubMed] [Google Scholar]
- Ancira-Moreno M, O’Neill MS, Rivera-Dommarco JÁ, Batis C, Rodríguez Ramírez S, Sánchez BN, … Vadillo-Ortega F. (2020). Dietary patterns and diet quality during pregnancy and low birthweight: The PRINCESA cohort. Maternal & Child Nutrition, 16(3), e12972. doi: 10.1111/mcn.12972 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Boparai JK, Singh S, & Kathuria P (2018). How to Design and Validate A Questionnaire: A Guide. Curr Clin Pharmacol, 13(4), 210–215. doi: 10.2174/1574884713666180807151328 [DOI] [PubMed] [Google Scholar]
- Brunst KJ, Kannan S, Ni Y-M, Gennings C, Ganguri HB, & Wright RJ (2016). Validation of a Food Frequency Questionnaire for Estimating Micronutrient Intakes in an Urban US Sample of Multi-Ethnic Pregnant Women. Maternal and child health journal, 20(2), 250–260. doi: 10.1007/s10995-015-1824-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Campbell RK, Talegawkar SA, Christian P, LeClerq SC, Khatry SK, Wu LSF, & West KP Jr. (2014). Seasonal dietary intakes and socioeconomic status among women in the Terai of Nepal. Journal of health, population, and nutrition, 32(2), 198–216. [PMC free article] [PubMed] [Google Scholar]
- Cicchetti DV, & Sparrow SA (1981). Developing criteria for establishing interrater reliability of specific items: applications to assessment of adaptive behavior. Am J Ment Defic, 86(2), 127–137. [PubMed] [Google Scholar]
- Conway JM, Ingwersen LA, Vinyard BT, & Moshfegh AJ (2003). Effectiveness of the US Department of Agriculture 5-step multiple-pass method in assessing food intake in obese and nonobese women. Am J Clin Nutr, 77(5), 1171–1178. doi: 10.1093/ajcn/77.5.1171 [DOI] [PubMed] [Google Scholar]
- Crozier SR, Robinson S.n. M. , Godfrey KM, Cooper C, & Inskip HM (2009). Women’s Dietary Patterns Change Little from Before to During Pregnancy. J Nutr, 139(10), 1956–1963. doi: 10.3945/jn.109.109579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cunningham K, Headey D, Singh A, Karmacharya C, & Rana PP (2017). Maternal and Child Nutrition in Nepal: Examining drivers of progress from the mid-1990s to 2010s. Global Food Security, 13, 30–37. doi: 10.1016/j.gfs.2017.02.001 [DOI] [Google Scholar]
- Devkota R, Khan GM, Alam K, Regmi A, & Sapkota B (2016). Medication utilization pattern for management of pregnancy complications: a study in Western Nepal. BMC Pregnancy Childbirth, 16, 272–272. doi: 10.1186/s12884-016-1068-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Doyle IM, Borrmann B, Grosser A, Razum O, & Spallek J (2017). Determinants of dietary patterns and diet quality during pregnancy: a systematic review with narrative synthesis. Public Health Nutr, 20(6), 1009–1028. doi: 10.1017/s1368980016002937 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fung TT, Isanaka S, Hu FB, & Willett WC (2018). International food group-based diet quality and risk of coronary heart disease in men and women. Am J Clin Nutr, 107(1), 120–129. doi: 10.1093/ajcn/nqx015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gicevic S, Gaskins AJ, Fung TT, Rosner B, Tobias DK, Isanaka S, & Willett WC (2018). Evaluating pre-pregnancy dietary diversity vs. dietary quality scores as predictors of gestational diabetes and hypertensive disorders of pregnancy. PLoS One, 13(4), e0195103. doi: 10.1371/journal.pone.0195103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gicevic S, Mou Y, Bromage S, Fung TT, & Willett W (2021). Development of a Diet Quality Screener for Global Use: Evaluation in a Sample of US Women. J Acad Nutr Diet, 121(5), 854–871e856. doi: 10.1016/j.jand.2020.12.024 [DOI] [PubMed] [Google Scholar]
- Gwet K (2002). Inter-Rater Reliability: Dependency on Trait Prevalence and Marginal Homogeneity. Stat Methods Inter-Rater Reliab Assess, 2, 1–9. [Google Scholar]
- Hartman TJ, Elliott AJ, Angal J, Block T, Ferranti EP, Mitchell DC, … Network PR (2016). Relative validation of a short questionnaire to assess the dietary habits of pregnant American Indian women. Food science & nutrition, 5(3), 625–632. doi: 10.1002/fsn3.440 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Khakurel G, Karki C, & Chalise S (2021). Prevalence of Thyroid Disorder in Pregnant Women Visiting a Tertiary Care Teaching Hospital: A Descriptive Cross-sectional Study. JNMA: Journal of the Nepal Medical Association, 59(233), 51–54. doi: 10.31729/jnma.5529 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kominiarek MA, & Peaceman AM (2017). Gestational weight gain. Am J Obstet Gynecol, 217(6), 642–651. doi: 10.1016/j.ajog.2017.05.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kuppuswamy R, Venugopal V, & Subramaniam A (2018). Development and validation of a food frequency questionnaire for pregnant women of Tamil Nadu, India. International Journal of Nutrition, Pharmacology, Neurological Diseases, 8(3), 86–91. doi: 10.4103/ijnpnd.ijnpnd_27_18 [DOI] [Google Scholar]
- Ministry of Health. (2017). Nepal Demographic and Health Survey 2016. http://dhsprogram.com/pubs/pdf/FR336/FR336.pdf
- Oken E, Guthrie LB, Bloomingdale A, Gillman MW, Olsen SF, Amarasiriwardena CJ, … Wright RO (2014). Assessment of dietary fish consumption in pregnancy: comparing one-, four- and thirty-six-item questionnaires. Public Health Nutr, 17(9), 1949–1959. doi: 10.1017/S1368980013001985 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rifas-Shiman SL, Rich-Edwards JW, Willett WC, Kleinman KP, Oken E, & Gillman MW (2006). Changes in dietary intake from the first to the second trimester of pregnancy. Paediatr Perinat Epidemiol, 20(1), 35–42. doi: 10.1111/j.1365-3016.2006.00691.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rifas-Shiman SL, Willett WC, Lobb R, Kotch J, Dart C, & Gillman MW (2001). PrimeScreen, a brief dietary screening tool: reproducibility and comparability with both a longer food frequency questionnaire and biomarkers. Public health nutrition, 4(2), 249–254. 10.1079/phn200061 [DOI] [PubMed] [Google Scholar]
- SPSS Statistics for Windows (Version 25.0) [Computer Software]. IBM. [Google Scholar]
- Tsoi KY, Chan RSM, Li LS, McAuliffe FM, Hanson MA, Tam WH, & Ma RCW (2020). Evaluation of dietary pattern in early pregnancy using the FIGO Nutrition Checklist compared to a food frequency questionnaire. Int J Gynaecol Obstet, 151 Suppl 1(Suppl 1), 37–44. doi: 10.1002/ijgo.13324 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Voortman T, Steegers-Theunissen RPM, Bergen NE, Jaddoe VWV, Looman CWN, Kiefte-de Jong JC, & Schalekamp-Timmermans S (2020). Validation of a Semi-Quantitative Food-Frequency Questionnaire for Dutch Pregnant Women from the General Population Using the Method or Triads. Nutrients, 12(5), 1341–1353. doi: 10.3390/nu12051341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wei J, Bhurtyal A, Dhungana RR, Bhattarai B, Zheng J, Wang L, … Wang Y. (2019). Changes in patterns of the double burden of undernutrition and overnutrition in Nepal over time. Obesity Reviews, 20(9), 1321–1334. doi: 10.1111/obr.12883 [DOI] [PubMed] [Google Scholar]
- Willett W, & Lenart E (2012). Reproducibility and Validity of Food Frequency Questionnaires. In (Vol. 6, pp. 96–141). [Google Scholar]
- Wood-Bradley RJ, Henry SL, Vrselja A, Newman V, & Armitage JA (2013). Maternal dietary intake during pregnancy has longstanding consequences for the health of her offspring. Can J Physiol Pharmacol, 91(6), 412–420. doi: 10.1139/cjpp-2012-0352 [DOI] [PubMed] [Google Scholar]
- Yuan MY, He JR, Chen NN, Lu JH, Shen SY, Xiao WQ, … Qiu X (2016). Validity and Reproducibility of a Dietary Questionnaire for Consumption Frequencies of Foods during Pregnancy in the Born in Guangzhou Cohort Study (BIGCS). Nutrients, 8(8), 454–465. doi: 10.3390/nu8080454 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang H, Qiu X, Zhong C, Zhang K, Xiao M, Yi N, … Yang X (2015). Reproducibility and relative validity of a semi-quantitative food frequency questionnaire for Chinese pregnant women. Nutr J, 14, 56–66. doi: 10.1186/s12937-015-0044-x [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Not Applicable