PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Public Health Nutr. Author manuscript; available in PMC 2010 August 1.
Published in final edited form as:
PMCID: PMC2705475
NIHMSID: NIHMS69749

Maternal cereal consumption and adequacy of micronutrient intake in the periconceptional period

Abstract

Objective

To assess the adequacy of periconceptional intake of key micronutrients for perinatal health in relation to regular cereal consumption of pregnant women.

Design, setting, and subjects

Low-income pregnant women (n=596) in Pittsburgh, PA who enrolled in a cohort study <20 weeks gestation. These women reported usual dietary intake in the three months around conception on a food frequency questionnaire. Cereal consumers were women who reported consuming any dry cereal at least three times per week. High risk for nutrient inadequacy was defined as intake less than the Estimated Average Requirement.

Results

About 31% of the women regularly consumed cereal. After adjusting for energy intake, race/ethnicity, marital status, breakfast consumption, and supplement use, cereal eaters had significantly higher intakes of folate, iron, zinc, calcium, fiber, and vitamins A, C, D, and E (all P<0.01) and were approximately 2 to 6 times as likely to have intakes in the highest third of the distribution for folate, iron, zinc, calcium, vitamins A and D, and fiber (all P<0.01) than non-cereal eaters. Cereal consumption was also associated with 65% to 90% reductions in risk of nutrient inadequacies compared with non-consumption (all P<0.01).

Conclusions

Encouraging cereal consumption may be a simple, safe, and inexpensive nutrition intervention that could optimize periconceptional intake for successful placental and fetal development.

Keywords: cereal, diet, periconception, pregnancy

Introduction

The periconceptional period represents a specific stage in a woman’s life in which adequate nutrient intake is especially important for both her and her fetus. Optimal pregnancy outcomes rely on successful implantation and early placental and fetal development. Such physiological processes involve remodeling of the maternal arteries underlying the placenta, extensive cell division and differentiation, and well-regulated responses to increases in inflammation and generation of reactive oxygen species—all of which may be influenced by maternal nutritional status13. Not surprisingly, inadequate periconceptional nutrition is associated with pregnancy complications such as congenital abnormalities4, preterm birth5, fetal growth restriction6, 7, and preeclampsia8. Thus, ensuring that women receive adequate nutrition around the time of conception is essential. Nevertheless, many US women of reproductive age are at high risk of nutrient inadequacies9, 10.

The consumption of ready-to-eat breakfast cereals may help women meet the high nutrient requirements of pregnancy. Among non-pregnant individuals, breakfast eaters are less likely than breakfast skippers to have nutrient inadequacies1113, in part because of the typical intake of ready-to-eat breakfast cereals in the morning meal. Many ready-to-eat cereals are fortified with key nutrients, such as folic acid, iron, zinc, calcium, and vitamins A, C, and D, all of which are linked to successful placental and fetal development. Among US adults, cereal is the top food source of folate, iron, and vitamin B6 and is among the top 10 food sources for many other micronutrients14, trends also observed in pregnant women15. In fact, micronutrient fortification of cereals makes significant contributions to overall daily intakes, with a higher contribution in women than in men16. Indeed, children and non-pregnant adults who consume cereal regularly have higher intakes of most vitamins and minerals11, 12, 1620, and lower prevalence of nutrient inadequacy than those who do not consume cereal11, 13, 16, 19, 20. Despite the importance of maternal micronutrient status around the time of conception, cereal intake has not been explored in relation to nutrient adequacy in pregnant women. As nutrient needs dramatically increase during pregnancy21, and food preferences may change, studying the association between cereal intake and nutrient adequacy specifically during this stage of the woman’s lifecycle is important.

Our objective was to assess the adequacy of periconceptional intake of key micronutrients for perinatal health in relation to regular cereal consumption. We hypothesized that pregnant women who ate cereal regularly in the periconceptional period would have higher absolute nutrient intakes and lower risk of nutrient inadequacies compared with women who did not eat cereal regularly.

Methods

Data came from two pregnancy cohort studies conducted at Magee-Womens Hospital in Pittsburgh, Pennsylvania from 2003 to 2005. The studies were designed to explore the effects of nutrition and other maternal factors on pregnancy outcomes. Both studies enrolled women at <20 weeks gestation (mean 17.6 weeks (SD 3.9)). Eligible women were aged 14 to 50 years, carrying singleton pregnancies, planning to deliver at Magee-Womens Hospital, and without history of diabetes, hypertension, autoimmune disease, or other preexisting medical conditions. After providing informed, written consent, subjects in both studies completed an interviewer-administered questionnaire at enrollment to collect data on sociodemographics, medical history, and health behaviors, including smoking, physical activity, and television watching. At enrollment, women completed a self-administered food frequency questionnaire (FFQ) assessing periconceptional dietary intake, supplement use, and meal patterns (discussed below). The University of Pittsburgh Institutional Review Boards approved these studies. A total of 596 out of 829 (72%) women had complete dietary data and were available for the current analysis.

Dietary Assessment

Periconceptional dietary intake was assessed using a modified Block98 FFQ (approximately 120 food/beverage items)22, 23. The questionnaire asks about usual dietary intake in the one month before and two months after conception. The food list for this questionnaire was developed from the NHANES III dietary recall data. An individual portion size is asked for each food, and pictures are provided to enhance accuracy of quantification. The FFQ has been validated in numerous samples2426, and is currently being validated in our own population. Slight modifications were made to the questionnaire to include a more extensive list of fish and seafood items, to focus on a three-month time period, and to be specific for pregnancy. Because this FFQ is semi-quantitative, it provides a projection of the amount of nutrients consumed. The questionnaire also included an assessment of the type and frequency of dietary supplements used in the periconceptional period.

Questionnaires were sent to Block Dietary Data Systems (Berkeley, CA) for optical scanning and nutrient analysis using software originally developed at the National Cancer Institute. The software used for nutrient analysis produces estimates of usual intake for a wide array of nutrients. It calculates the frequency of daily intake and total daily grams of food consumed for each food item, and provides the gram weight for each serving size. Nutrient values were calculated by multiplying the nutrient content of the food by the gram weight and frequency, and summing across all food items. The nutrient values were updated based on the USDA 1994–1996 Continuing Survey of Food Intake by Individuals for women ages 19–44 years and updated folate values for fortified foods from the USDA 1998 nutrient database27. Folate values were adjusted for increased bioavailability of fortified folate. For foods without added folic acid, micrograms of dietary folate equivalents (DFEs) were equal to the micrograms of naturally occurring food folate. For foods with added folic acid, DFEs were calculated using the following formula: micrograms of naturally occurring food folate + (micrograms of added folic acid × 1.7)28.

The FFQ included three dry cereal line items: high-fiber cereals, highly-fortified cereals, and all other dry cereals. We summed the frequency reported for each cereal line item to obtain an overall frequency of any ready-to-eat cereal in the periconceptional period. We classified women as regular cereal eaters if they reported consuming cereal at least three times per week at any point throughout the day.

Women were defined as being at high risk for nutrient inadequacy if their nutrient intake was less than the Estimated Average Requirement (EAR) for pregnant women. The EAR is one of the Dietary Reference Intakes used to assess adequacy of a population’s intake29. The EAR is the average daily nutrient intake to meet the requirements of half the healthy individuals for the specific life stage and gender group. We elected to assess periconceptional diets relative to the EARs for pregnant women because our definition of the 3 periconceptional months included two postconception months and one preconception month (a time when women should be preparing for pregnancy). To assess the likelihood of inadequacy for iron, we used the cut-point approach rather than the full probability approach30. Too few data were available for the Institute of Medicine to simulate an iron requirement distribution for pregnant and lactating women. Because pregnant women have no menstrual losses, it is assumed that the requirements are normally distributed around the EAR (Suzanne Murphy, personal communication, 2008). An EAR has not yet been established for calcium and vitamin D,31 so we did not calculate the risk of nutrient inadequacy for these micronutrients.

Covariates

Intake of breakfast was determined using a diet pattern questionnaire, which assessed the number, type, and timing of meals and snacks consumed in a typical week around conception32. The interpretation of what constituted a meal or snack was left up to the participants’ discretion. We defined breakfast consumption as a self-reported breakfast meal.

Sociodemographic and behavioral covariate information was acquired through interview-based self-report. Education (<12 years, 12 years, or >12 years), prepregnancy smoking status (smoker, nonsmoker), martial status (married, unmarried), and parity (0, 1 or more) were available. Race/ethnicity was self-reported and subsequently categorized as black or non-black because few women reported races other than black or white. Prepregnancy body mass index (BMI) [weight (kg)/height (m)2] was calculated from maternal self-report of prepregnancy weight and height measurement at initial visit. Women were asked to categorize their usual amount of time spent watching television in the year before the index pregnancy as 0–1, 2–3, or ≥4 hours/day. At enrollment, subjects also self-reported their use of multivitamins or prenatal vitamins in the periconceptional period. Periconceptional supplement users were women who reported using a prenatal vitamin or multivitamin at least once per month in the 3 months around conception.

Statistical Analysis

We compared maternal characteristics by cereal intake using Pearson chi-squared statistics. Multivariable linear regression was used to assess differences in mean nutrient intake by cereal use after adjustment for confounders. Nutrients were log-transformed because nutrient intake was highly skewed. Mean micronutrient and fiber intakes were energy-adjusted by dividing the total amount of the nutrient by the number of calories, and then multiplying by 2000. These nutrient density values were then divided into thirds based on tertiles. We used chi-squared tests to compare the distribution of cereal and non-cereal eaters among tertiles of nutrient intake. We then used multivariable logistic regression to assess the independent association between cereal intake and the odds of being in the upper tertile of nutrient intake compared with the lower two tertiles. Finally, we calculated the proportion of women who consumed less than the EAR for each micronutrient and used logistic regression to determine the odds of micronutrient inadequacy by regular cereal use. We also performed the aforementioned analyses excluding 19 women who reported regular consumption of highly-fortified cereals to ensure that the results were not driven by their intakes.

We fit parsimonious regression models by specifying a full model with potential confounding variables (energy intake, breakfast consumption, maternal age, race/ethnicity, parity, education, marital status, smoking, prepregnancy BMI, supplement use, and prepregnancy television watching). Potential confounders were considered to not be influential and were removed from the model if their inclusion did not satisfy our a priori change-in-estimate criterion (a change in the coefficient of >10%).

Results

About 31% of women regularly consumed cereal. Cereal eaters were more likely than non-cereal eaters to be married, multiparous, and users of dietary supplements (Table 1). Cereal users also tended to be leaner and more likely to consume a breakfast meal compared with non-cereal users, although not statistically significant.

Table 1
Subject characteristics by regular cereal consumption*

Energy intake was significantly greater among cereal eaters than non-cereal eaters (Table 2). After adjusting for energy intake, race/ethnicity, marital status, breakfast consumption, and supplement use, women who regularly consumed cereal had significantly higher mean nutrient intakes of folate, iron, zinc, calcium, vitamin A, vitamin C, vitamin D, vitamin E, and fiber (all P<0.01) (Table 2). Nutrient intakes were 14% to 83% greater among users than non-users.

Table 2
Adjusted mean nutrient intakes by regular cereal consumption*,

Compared with women who did not consume cereal regularly, regular consumers were significantly more likely to have folate, iron, zinc, calcium, vitamin A, vitamin D, and fiber intakes in the highest third of the distribution (Table 3). Notably, regular cereal consumers were about 5 and 6 times as likely to be in the highest tertile of periconceptional folate and iron intake, respectively, than non-consumers after confounder adjustment.

Table 3
Association between regular cereal consumption and odds of nutrient intake in the highest third of the distribution

Compared with cereal users, a significantly higher proportion of non-users of cereal failed to achieve the EAR for folate, iron, zinc, vitamin A, vitamin C, and vitamin E (all P<0.001) (Table 4), and were therefore at high risk of nutrient inadequacy. For instance, about 27% of cereal users and 71% of non-users had folate intakes less than the EAR. After adjusting for energy intake, race/ethnicity, marital status, breakfast intake, and supplement use, regular cereal consumption was associated with significant reductions in risk of nutrient inadequacies compared with non-cereal consumption (all P<0.01). Women who regularly used cereal had approximately 90% reductions in risk of inadequacy in folate and iron, and 65% to 75% reductions in risk of insufficiency in zinc, vitamin A, vitamin C, and vitamin E.

Table 4
Association between regular cereal consumption and risk of nutrient inadequacy

The exclusion of the 19 cereal users who reported regular consumption of highly fortified cereals did not meaningfully influence the findings (data not shown). Further adjustment of age, education, smoking status, parity, prepregnancy BMI, and television watching did not considerably alter any of the results (data not shown).

Discussion

We observed that pregnant women who ate cereal at least three times per week around the time of conception had significantly higher intakes of key micronutrients compared with women who did not eat cereal regularly, even after controlling for confounders such as energy and breakfast intake. Moreover, regular cereal users had significant reductions in risk of periconceptional micronutrient inadequacies, including that of folate and iron, compared with non-users. These results are meaningful as these micronutrients have been highlighted as shortfall nutrients in women of reproductive age10. In addition, the positive effects of eating cereal were not limited to women who regularly consumed highly-fortified cereals.

To our knowledge, this is the first study to explore the association between cereal intake and micronutrient adequacy in pregnant women. Our findings are consistent with numerous studies relating cereal intake to nutritional status among children and non-pregnant adults11, 16, 19. For example, in a nationally representative US sample of individuals aged 9 years and older, nutrient intakes were 14% to 63% higher in cereal consumers than non-consumers as defined by self-report on 24-hour dietary recalls19. Additionally, another study of children and young adults in the US found that cereal eaters were significantly more likely to meet at least two-thirds of the Recommended Dietary Allowance for folate, iron, calcium, zinc, and vitamins A, C, and D11. Similarly, in 717 non-pregnant Irish women aged 18–64 years, Galvin et al.16 reported that cereal consumption was associated with lower prevalence of inadequate intakes of folate, iron, calcium, zinc, and vitamin C according to European Average Requirements. Moreover, the proportion of women with nutrient inadequacies decreased as cereal consumption increased.

We used an FFQ to assess cereal intake, while most previous studies used dietary recalls or records. Nevertheless, the percentage of pregnant women regularly consuming cereal in our study was comparable to that previously reported in women of reproductive age17, 19, 33, 34. Although investigators of past studies tended to focus on cereal intake at breakfast, we included cereal intake throughout the day because a sizable proportion (36%) of regular cereal eaters in our study reported not usually consuming breakfast. This suggests that cereal was often consumed as a snack or part of another meal. This finding is consistent with a study of US children and young adults, which found that around 40% of cereal eaters consumed cereal as either lunch, dinner, or a snack and that those who consumed cereal at any time of day had higher micronutrient intakes11. Taken together, these data highlight that cereal intake may not need to be part of breakfast to have a positive impact on diet quality.

Some health care professionals may wonder about the negative aspects of promoting cereal intake, such as excessive weight gain, high sugar intakes, and/or micronutrient intakes above tolerable levels. Nevertheless, cross-sectional studies have shown that cereal eaters are often leaner17, 34, 35 and also have lower daily intakes of fat and cholesterol1619, 34 than non-cereal eaters. We also found that cereal users tended (P=0.07) to have lower BMI values than non-cereal users. Moreover, ready-to-eat breakfast cereals contribute minimally to overall daily sugar intake13. Cereal consumption should, however, be encouraged in moderation, with attention paid to serving size, and in the context of a healthy diet.

The positive effect we observed of cereal intake on micronutrient intake may not have been due to the cereal itself, but rather foods commonly eaten by cereal users or healthy eating behaviors. For example, the higher intakes of calcium and vitamin D seen among cereal users may be due to use of milk on cereal. We could not separate these effects because 85% of the cereal consumers reported regularly using milk on their cereal. In addition, there is unfortunately no gold standard for assessing usual dietary intake. FFQs are limited by a restricted, culture-specific food list; their representation of a person’s perceived intake rather than actual intake; and their limited ability to estimate absolute energy intake36. However, FFQs have the advantages of capturing usual past intake of micronutrients, having a low respondent burden, and ranking individuals relative to one another (i.e., high vs. low consumers)36. We performed our analyses assessing both tertiles of nutrient intake (to capture a ranking of subjects) and micronutrient intakes less than the EAR (an absolute measure of intake) and came to comparable conclusions. Another limitation was the time lapse between the period of recall (periconceptional period) and completion of the FFQ (mid-pregnancy). A study to validate the FFQ in our population is currently underway. Lastly, our small sample size prohibited us from studying whether cereal intake protected against adverse pregnancy outcomes. Future investigations should explore this intriguing research question.

Our results suggest that regularly consuming cereal in the periconceptional period may help reduce the prevalence of deficiencies in essential micronutrients, including folate and iron. Inquiring about regular intake of cereal could be used as an instant assessment of micronutrient inadequacies in pregnant women. Encouraging cereal consumption as part of a healthy lifestyle around conception may be a simple, safe, and inexpensive nutrition intervention that could optimize maternal periconceptional intake for successful placental and fetal development.

References

1. Cross JC, Werb Z, Fisher SJ. Implantation and the placenta: key pieces of the development puzzle. Science. 1994;266:1508–1518. [PubMed]
2. Goodger AM, Rogers PA. Uterine endothelial cell proliferation before and after embryo implantation in rats. J Reprod Fertil. 1993;99:451–457. [PubMed]
3. Jauniaux E, Watson AL, Hempstock J, Bao YP, Skepper JN, Burton GJ. Onset of maternal arterial blood flow and placental oxidative stress. A possible factor in human early pregnancy failure. Am J Pathol. 2000;157:2111–2122. [PubMed]
4. Czeizel AE, Dudas I. Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation. N Engl J Med. 1992;327:1832–1835. [PubMed]
5. Bloomfield FH, Oliver MH, Hawkins P, Campbell M, Phillips DJ, Gluckman PD, et al. A periconceptional nutritional origin for noninfectious preterm birth. Science. 2003;300:606. [PubMed]
6. MacLaughlin SM, Walker SK, Roberts CT, Kleemann DO, McMillen IC. Periconceptional nutrition and the relationship between maternal body weight changes in the periconceptional period and feto-placental growth in the sheep. J Physiol. 2005;565:111–124. [PubMed]
7. Oliver MH, Hawkins P, Harding JE. Periconceptional undernutrition alters growth trajectory and metabolic and endocrine responses to fasting in late-gestation fetal sheep. Pediatr Res. 2005;57:591–598. [PubMed]
8. Roberts JM, Balk JL, Bodnar LM, Belizan JM, Bergel E, Martinez A. Nutrient involvement in preeclampsia. J Nutr. 2003;133:1684S–1692S. [PubMed]
9. U.S. Department of Health and Human Services. Healthy People 2010: National Health Promotion and Disease Prevention Objectives. Washington, DCed:
10. U.S. Department of Health and Human Services and U.S. Department of Agriculture. The Report of the Dietary Guidelines Advisory Committee on Dietary Guidelines for Americans. 6th ed. Washington, DC: 2005.
11. Nicklas TA, O'Neil CE, Berenson GS. Nutrient contribution of breakfast, secular trends, and the role of ready-to-eat cereals: a review of data from the Bogalusa Heart Study. Am J Clin Nutr. 1998;67:757S–763S. [PubMed]
12. Preziosi P, Galan P, Deheeger M, Yacoub N, Drewnowski A, Hercberg S. Breakfast type, daily nutrient intakes and vitamin and mineral status of French children, adolescents, and adults. J Am Coll Nutr. 1999;18:171–178. [PubMed]
13. Williams P. Breakfast and the diets of Australian adults: an analysis of data from the 1995 National Nutrition Survey. Int J Food Sci Nutr. 2005;56:65–79. [PubMed]
14. Cotton PA, Subar AF, Friday JE, Cook A. Dietary sources of nutrients among US adults, 1994 to 1996. J Am Diet Assoc. 2004;104:921–930. [PubMed]
15. Siega-Riz AM, Bodnar LM, Savitz DA. What are pregnant women eating? Nutrient and food group differences by race. Am J Obstet Gynecol. 2002;186:480–486. [PubMed]
16. Galvin MA, Kiely M, Flynn A. Impact of ready-to-eat breakfast cereal (RTEBC) consumption on adequacy of micronutrient intakes and compliance with dietary recommendations in Irish adults. Public Health Nutr. 2003;6:351–363. [PubMed]
17. Barton BA, Eldridge AL, Thompson D, Affenito SG, Striegel-Moore RH, Franko DL, et al. The relationship of breakfast and cereal consumption to nutrient intake and body mass index: the National Heart, Lung, and Blood Institute Growth and Health Study. J Am Diet Assoc. 2005;105:1383–1389. [PubMed]
18. Gibson S. Micronutrient intakes, micronutrient status and lipid profiles among young people consuming different amounts of breakfast cereals: further analysis of data from the National Diet and Nutrition Survey of Young People aged 4 to 18 years. Public Health Nutr. 2003;6:815–820. [PubMed]
19. Song WO, Chun OK, Kerver J, Cho S, Chung CE, Chung SJ. Ready-to-eat breakfast cereal consumption enhances milk and calcium intake in the US population. J Am Diet Assoc. 2006;106:1783–1789. [PubMed]
20. van den Boom A, Serra-Majem L, Ribas L, Ngo J, Perez-Rodrigo C, Aranceta J, et al. The contribution of ready-to-eat cereals to daily nutrient intake and breakfast quality in a Mediterranean setting. J Am Coll Nutr. 2006;25:135–143. [PubMed]
21. Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes: the essential guide to nutrient requirements. Washington, D.C.: National Academies Pressed; 2006.
22. Block G, Coyle LM, Hartman AM, Scoppa SM. Revision of dietary analysis software for the Health Habits and History Questionnaire. Am J Epidemiol. 1994;139:1190–1196. [PubMed]
23. Block G, Hartman AM, Dresser CM, Carroll MD, Gannon J, Gardner L. A data-based approach to diet questionnaire design and testing. Am J Epidemiol. 1986;124:453–469. [PubMed]
24. Block G, DiSogra C. WIC dietary assessment validation study. Final report. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Serviceed;
25. Block G, Thompson FE, Hartman AM, Larkin FA, Guire KE. Comparison of two dietary questionnaires validated against multiple dietary records collected during a 1-year period. J Am Diet Assoc. 1992;92:686–693. [PubMed]
26. Block G, Woods M, Potosky A, Clifford C. Validation of a self-administered diet history questionnaire using multiple diet records. J Clin Epidemiol. 1990;43:1327–1335. [PubMed]
27. U.S. Department of Agriculture, Agricultural Research Service. USDA Nutrient Database for Standard Reference, Release 12. ed.
28. Institute of Medicine, Food and Nutrition Board. Folate. Dietary Reference Intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. Washington, DC: National Academies Pressed; 2000. [PubMed]
29. Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes (DRIs): Estimated Average Requirements for Groups. Washington, DC: National Academies Pressed; 2002.
30. Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Washington, DC: National Academies Pressed; 2001. [PubMed]
31. Institute of Medicine, Food and Nutrition Board. Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington, D.C.: National Academies Pressed; 1997.
32. Siega-Riz AM, Herrmann TS, Savitz DA, Thorp JM. Frequency of eating during pregnancy and its effect on preterm delivery. Am J Epidemiol. 2001;153:647–652. [PubMed]
33. Siega-Riz AM, Popkin BM, Carson T. Differences in food patterns at breakfast by sociodemographic characteristics among a nationally representative sample of adults in the United States. Prev Med. 2000;30:415–424. [PubMed]
34. Song WO, Chun OK, Obayashi S, Cho S, Chung CE. Is consumption of breakfast associated with body mass index in US adults? J Am Diet Assoc. 2005;105:1373–1382. [PubMed]
35. Cho S, Dietrich M, Brown CJ, Clark CA, Block G. The effect of breakfast type on total daily energy intake and body mass index: results from the Third National Health and Nutrition Examination Survey (NHANES III) J Am Coll Nutr. 2003;22:296–302. [PubMed]
36. Willet WC. Nutritional Epidemiology. 2nd ed. New York: Oxford University Press; 1998.