Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


osu1170873381.pdf (644.19 KB)

ETD Abstract Container

Abstract Header

Estimates of obesity-attributable mortality in the United States

Lu, Lingyi
http://rave.ohiolink.edu/etdc/view?acc_num=osu1170873381

Abstract Details

2005, Master of Science, Ohio State University, Public Health.
The increasing prevalence of obesity over the past decades has generated considerable concern about health problems and subsequent economic burden. Approximately 64% of U.S. adults are estimated to be overweight or obese. Over 30% are obese and 5% are extremely obese among them. Obesity has been associated with a variety of chronic diseases and with increased risk of all-cause mortality. The number of annual deaths in the U.S. attributable to obesity estimated from epidemiological cohorts ranges widely from 26,917 to 385,000. The objective of the current study is to estimate a more accurate number of excess deaths attributable to obesity. A sub-sample of 6,913 subjects from the First National Health and Nutrition Examination Survey (NHANES I, 1971-1975) data set and corresponding mortality information from the National Health and Nutrition examination Survey I Epidemiologic Follow-up Study (NHEFS,1992) was employed for building a Cox's proportional hazards model. The number (2,374,029) of total deaths age from 25 and older in the U.S. came from the National Center for Health Statistics 2003 National Vital Statistics Report. Body weight was categorized based upon body mass index (BMI) which is defined as Weight (kg) / Height² (m²). "Underweight" was defined as BMI < 18.5; "Normal weight" was defined as 18.5 ≤ BMI <25; "Overweight" was defined as 25 ≤ BMI <30; "Obesity" was defined as 30 ≤ BMI <35; and "severe obesity" was defined as BMI ≥ 35. A Cox's proportional hazard model adjusted for gender, race, education and smoking status was applied to estimate the relative risk of dying associated with obesity and severe obesity. Both time-on-study and left truncation to adjust delayed entry were used to construct Cox's proportional hazard model. The Population Attributable Risk (PAR) was calculated based on Bruzzi's [1] formula which takes into account multiple exposure levels and adjusted relative risk. The extra deaths attributable to obesity were calculated by multiplying the total number of deaths by the PAR. Our results showed that the estimated extra deaths attributable to obesity and severe obesity were 280,708 per year based on the time-on-study model and 260,041 per year for the left truncation model. We also found that the relative risk of death for underweight was as high as severe obesity, whereas the relative risk of death for overweight was less than but very close to that of normal weight. Even though the same NHANES I data set was employed in the two previous studies, our estimated numbers of extra deaths attributable to obesity were higher than the estimated number 184,670 reported by Allison [2], but lower than the number 298,808 estimated by Flegal [3]. The difference in sample size, reference group, target population, model in control of later entry, PAR formula may be the underlying reasons accounting for the different estimated numbers.
Melvin Moeschberger (Advisor)
59 p.

Recommended Citations

Citations

  • Lu, L. (2005). Estimates of obesity-attributable mortality in the United States [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1170873381

    APA Style (7th edition)

  • Lu, Lingyi. Estimates of obesity-attributable mortality in the United States. 2005. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1170873381.

    MLA Style (8th edition)

  • Lu, Lingyi. "Estimates of obesity-attributable mortality in the United States." Master's thesis, Ohio State University, 2005. http://rave.ohiolink.edu/etdc/view?acc_num=osu1170873381

    Chicago Manual of Style (17th edition)

osu1170873381
578
© 2005, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.