Dietary patterns predict the development of overweight in women: The Framingham Nutrition Studies

https://doi.org/10.1016/S0002-8223(02)90275-0Get rights and content

Abstract

Objective To investigate relationships between dietary patterns and the development of overweight.

Design Longitudinal analyses during 12 years of follow-up involved the identification of dietary patterns at baseline using cluster analysis applied to a 145-item semiquantitative food frequency questionnaire.

Subjects/setting 737 non-overweight women in the Framingham Offspring/Spouse cohort (mean age, 45 years).

Main Outcome Measure Development of overweight (BMI≥25) at follow-up.

Statistical Analyses Relative risks were calculated using Proc Genmod and multivariate models comprehensively considered potential confounders.

Results Five dietary patterns were identified among the cohort at baseline: Heart Healthy, Light Eating, Wine and Moderate Eating, High Fat, and Empty Calorie. Over 12 years, the crude risk of becoming overweight was 29% overall, ranging from 22% of women in the Wine and Moderate Eating cluster to 41% of women in the Empty Calorie cluster. Compared with women who ate a lower-fat, nutritionally varied Heart Healthy diet, women who ate an Empty Calorie diet that was rich in sweets and fats with fewer servings of nutrient-dense fruits, vegetables, and lean food choices were at higher risk for developing overweight [RR 1.4, 95% CI (0.9, 2.2)] after adjusting for age, smoking status, physical activity, menopausal status, energy intake, intentional dieting, and usual weight pattern. Women who ate an Empty Calorie dietary pattern were also younger and were more likely to smoke.

Conclusions Behavioral interventions for weight management and obesity prevention may be enhanced by creatively targeting differences in eating patterns, dietary quality, and other lifestyle behaviors of distinct subgroups of the population. J Am Diet Assoc. 2002;102:1240-1246.

Section snippets

Methods

This study explored relationships between dietary patterns and the development of overweight [defined as Body Mass Index (BMI)≥25] among female participants in the Framingham Offspring/Spouse study (FOS). FOSExam3 (1984-1988) served as baseline with follow-up assessed over 12 years to Exam 6 (1996-1999). At baseline, participants were grouped into one of five clusters based on food intake patterns identified from individual responses to a food frequency questionnaire. Women who were not

The Framingham Study

The Framingham Study was initiated in 1948 as a longitudinal population-based study of cardiovascular disease. The original Framingham cohort consisted of 5,209 men and women who represented a two-thirds random sample of the residents of the town of Framingham, Mass. In 1971, 5,124 Framingham Study offspring and their spouses were recruited to participate in the Framingham Offspring/Spouse (FOS) study (20).

Members of the FOS cohort are examined every 4 years, on average. At each exam, they

Instruments

The self-administered Framingham food frequency questionnaire (FFQ) is a modified version of the original Willett questionnaire (21). This instrument was validated in the Framingham Offspring cohort (22). It contains 145 individual food items or food groupings with 7 nonoverlapping response categories, ranging from “rarely or never use” to use “4 or more times each day.” Respondents report how often, on average, they consumed a standard portion of each item or grouping during the past year.

Results

Table 2 displays daily servings of food groupings consumed by women in the 5 dietary pattern clusters. Women in the Heart Healthy cluster consumed more servings of vegetables, fruits, low-fat milk, and other low-fat and fiber-rich foods (including whole grains, fish, low-fat cheeses, lean poultry, and legumes) with fewer servings of diet beverages and firm vegetable fats than women in each of the other 4 clusters. Women in the High Fat and Empty Calorie clusters consumed noticeably higher

Discussion

This study used an analytic approach to dietary exposure measurement that considered overall patterns of food intake and multiple aspects of dietary behavior. We believe that our understanding of the complex relationships between diet and the development of overweight was informed by the cluster analytic approach. Discrete exposure categories were differentiated and differences noted among dietary clusters in these subgroup analyses were completely consistent with data published earlier on the

Applications

Understanding overall dietary patterns seems to be important to guide the development of targeted weight-control strategies. We have identified a link between the risk of becoming overweight and an eating pattern that is rich in fats and sweets and lacking adequate servings of fruits, vegetables, fiber-rich whole grains, and lean protein selections. This increased risk is noted in contrast to a dietary pattern consumed by women who choose a diet that is closer to the expert dietary

References (33)

  • D.B. Allison et al.

    Annual deaths attributable to obesity in the United States

    JAMA.

    (1999)
  • J. Stamler

    Epidemic obesity in the United States

    Arch Intern Med.

    (1993)
  • R.J. Kuczmarski et al.

    Increasing prevalence of overweight among US adults. The National Health and Nutrition Examination Surveys, 1960-1991

    JAMA.

    (1994)
  • A. Must et al.

    The disease burden associated with overweight and obesity

    JAMA.

    (1999)
  • J.P. Koplan et al.

    Caloric imbalance and public health policy

    JAMA.

    (1999)
  • S.T. St. Jeor et al.

    Obesity Workshop III. AHA Prevention Conference III. Behavior change and complianceKeys to improving cardiovascular health

    Circulation.

    (1993)
  • Cited by (0)

    View full text