Background: The effect of overweight and obesity on the risk of fatal disease tends to attenuate with age. One plausible explanation is that a proportion of elderly people in the ideal weight category (19 to 25 kg/m ) experienced disease-related weight loss - an effect that will attenuate health risk related to overweight and obesity. One potential approach to controlling for this bias would be to examine fluctuations in weight during the adult lifespan by considering multiple measures of weight or recall of body weight during adulthood. Some of these measures are available in the Adventist Health Study cohorts and provide a rich data set to answer questions about adiposity and disease related weight loss.

Purpose: In Study 1, we have examined whether the relation between Body Mass Index (BMI) and mortality in elderly subjects (>65 years) is confounded by pre-baseline disease related weight loss. In Study 2, we have examined the accuracy of past weight recall in elderly study participants.

Methods: To evaluate whether a decrease in BMI-related mortality is partly attributable to disease-related weight loss, we examined the history of overweight/obesity and obesity-related disease in adults who were in the recommended range (19 to 25 9 9 • kg/m ) or overweight (>25 to 30 kg/m ) at baseline in a cohort study. We conducted an analysis of 7,855 subjects who participated in both Adventist Mortality Study and Adventist Health Study 1. To assess accuracy of recall weight, we calculated mean weight difference (current weight at AHS-1 minus recall weight at AHS-2, 26 years later) and compared recall weight to current weight in a correlation analysis.

Results: Among adults in the ideal weight group (BMI 19 -/m2) at baseline we found that 1) the prevalence of previous overweight/obesity was 20.4% and increased with age, 2) those with overweight/obesity were more likely to have currently diagnosed diabetes (OR = 2.83), coronary heart disease (OR=1.91), and high blood pressure (OR=l .52). In survival analyses, we found that current overweight/obesity was a risk factor for mortality after excluding subjects with previous overweight/obesity. We also found that recall of past body weight and weight reported 26 years prior was strongly correlated (r = 0.88, p-value

Conclusions: Our findings indicate that among adults currently in the recommended range of BMI at baseline in a cohort study, a history of overweight or obesity can be a confounder when using this group to estimate the risk due to current adiposity. Valid measures of weight history can be used to control for some of this confounding.


Epidemiology and Biostatistics


School of Public Health

First Advisor

Pramil N. Singh

Second Advisor

Synnove F. Knutsen

Third Advisor

Gary Fraser

Fourth Advisor

Serena Tonstad

Degree Name

Doctor of Public Health (DrPH)

Degree Level


Year Degree Awarded


Date (Title Page)




Library of Congress/MESH Subject Headings

Obesity -- Prevention; Seventh-day Adventists -- Diseases; Weight control; Obesity -- prevention and control; Body Weight Changes -- in adulthood; Body Mass Index; Cohort Studies.



Page Count

x; 139

Digital Format


Digital Publisher

Loma Linda University Libraries

Usage Rights

This title appears here courtesy of the author, who has granted Loma Linda University a limited, non-exclusive right to make this publication available to the public. The author retains all other copyrights.


Loma Linda University Electronic Theses and Dissertations

Collection Website



Loma Linda University. Del E. Webb Memorial Library. University Archives