The Associations Between Body Mass Index and Health Problems: Evidence from the National Longitudinal Survey Cohort 1979

Background/Purpose:  Much research has been conducted to examine the relationships between body mass index (BMI) and health issues. For example, past research has shown that increasing BMI relates to increased prevalence of Diabetes Mellitus and dyslipidemia. However, research has not sought to parse out the dynamics among BMI, health problems, and other socioeconomic factors.  This study replicates and extends Bays et al. (2007) to assess the role of BMI on health problems, while controlling for social factors.

Methods: The present research utilized data form the National Longitudinal Survey of Youth 1979 (NLSY79). The NLSY79 is a nationally representative survey collecting annual information from a cohort of 12,686 since 1979.  The dependent variable was any of BMI-related health problems reported by respondents prior to 2006, for example high blood sugar, congestive heart failure, high blood pressure, and, myocardial infarctions. Those reporting at least one health condition were coded “1” on the dependent variable, and others were coded “0”. The independent variable was individual’s BMI calculated from self-reported information of height and weight in 2006. BMI was subsequently categorized into four groups with scores of less than 18.5 being underweight, between 18.5 and 24.9 as normal, 25 through 29.9 overweight, and scores greater than 29.9 being obese.  Other control variables include individual demographic and socioeconomic characteristics, such as age, educational attainment, and wage.  Logistic regression was conducted to assess the association between BMI and health problems.

Results:  19.87% of the sample reported having been diagnosed with at least one health problem.  62.24% of the sample was categorized as obese, 21.85% were categorized as overweight, 15.43% were categorized as normal, and only .48% of the sample was underweight. Using the group of individuals with normal BMI as the reference category, the bivariate logistic regression suggests that those underweight and those overweight were more likely to report having health problems, with odds ratios as 1.80 and 1.29, respectively. The odds of reporting health problems for those fell into the obese category were .23 lower than those with normal BMI. However, when control variables were included in multivariate analyses, the odds for those fell into the obese category became 2.64 of those with normal BMI, while the odds for those underweight and those overweight remained relatively constant.

Conclusions and Implications:  The findings demonstrate the significant association between BMI and health problems. Compared to those with normal BMI range, other BMI categories have a high risk to report health problems. In particular, the results suggest a dynamic interaction among socioeconomic background, obesity, and health problems.  Future research should explore the underlying mechanism of this dynamic in depth.  The implications of these findings may reorient the hierarchy of needs for clients in at risk BMI categories.  It may be pertinent to focus on health care for obese clients before other needs.