The Effect of Middle‐ and Old‐Age Body Mass Index on Short‐Term Mortality in Older People
To determine the effect of body mass index (BMI) at old age and at age 50 on short-term survival among persons age 65 and older.
Cross-sectional, using the 4,791 respondents to the community interview of the 1994 National Long Term Care Survey (NLTCS).
United States of America.
Persons age 65 and older who lived in community settings as of the 1994 NLTCS interview.
Short-term mortality was measured from the date of the 1994 NLTCS through year-end 1995. BMI (kg/m2) (at three points: 1994 NLTCS, 1 year before, age 50) and all other variables, including three other modifiable risk factors known to be related to mortality—cigarette smoking, alcohol consumption, and exercise—were based on self-report.
Both the unadjusted and adjusted nadirs of mortality in relation to BMI at old age were found in older persons with a BMI between 30 and 34.9; this was true for males and females in all age groups. The highest mortality rates were found for older persons with very low BMI (<18.5). In contrast, BMI at age 50 was positively related to mortality, with those in the lowest BMI category (<18.5) at age 50 having the lowest mortality. Persons who were obese at age 50 and who were no longer obese at the 1994 NLTCS had lower mortality than persons with stable weight.
Weight reduction by middle-aged persons who are obese should be reinforced as a public health priority, because there is evidence that long-term weight loss results in better short-term survival. Further study of healthy older survivors to determine why they are not harmed by heavier weight in old age may provide useful insights into successful aging. J Am Geriatr Soc 49:1319–1326, 2001.
Figures in this publication
Available from: John J Sollers
- "The lack of association for either measure of obesity at older ages supports recent arguments that “a little extra” around the middle may be good for older adults, or at least not as harmful as in young and middle age [46, 47]. Our findings lend some support for this or, at least, fail to demonstrate that central adiposity, within the range observed in this population, was associated with poorer ANS function in the oldest participants. "
[Show abstract] [Hide abstract]
ABSTRACT: While frank obesity is associated with reduced HRV, indicative of poorer autonomic nervous system (ANS) function, the association between body mass index (BMI) and HRV is less clear. We hypothesized that effects of adiposity on ANS are mostly mediated by visceral fat and less by subcutaneous fat; therefore, centrally distributed adipose tissue, that is, waist circumference (WC), should be more strongly associated with HRV than overall adiposity (BMI). To examine this hypothesis, we used data collected in a subset of the Baltimore Longitudinal Study of Aging to compare strength of association between HRV and WC to that of HRV and BMI. Time domain HRV variables SDNN (standard deviation of successive differences in normal-to-normal (N-N) intervals) and RMSSD (root mean square of successive differences in N-N intervals) were calculated from 24-hour Holter recordings in 159 participants (29–96 years). Increasing WC was associated with decreasing SDNN and RMSSD in younger but not older participants (
value for WC-by-age interaction = 0.003). BMI was not associated with either SDNN or RMSSD at any age. In conclusion, central adiposity may contribute to sympathetic and parasympathetic ANS declines early in life.
Journal of obesity 05/2012; 2012(5):149516. DOI:10.1155/2012/149516
Available from: Linda K George
- "The underwhelming performance of moderate alcohol consumption was somewhat unexpected given recent evidence on the benefits of drinking (O'Keefe, Bybee, and Lavie 2007), and when significant, the effect was incongruent with our hypothesis. Similarly, the fact that normal BMI had inconsistent effects lends credence to recent findings that elevated BMI in later life is more favorable for health and longevity and thinness is more often a marker of frailty (D. H. Taylor and Ostbye 2001; Stevens et al. 1998). Recent research suggests that it is important to consider levels of activity in relation to BMI-related health disparities (e.g., McAuley et al. 2010), particularly among older adults. "
[Show abstract] [Hide abstract]
ABSTRACT: Countless studies show that socioeconomic status (SES) is strongly related to morbidity and mortality. However, few studies consider the substantial variability in health within socioeconomic strata. In this article, the authors examine the incompatibility between stratification-based theories of health inequality and empirical patterns of exceptional health among the socially disadvantaged. Using panel data from the Health and Retirement Survey (1992-2008), the authors test the mediating and moderating effects of various predictors of exceptional health (no chronic diseases or physical limitations) for middle-aged and older adults with and without a high school education. Results suggest that a combination of demographic characteristics, family and religious factors, socioeconomic resources, health behaviors, psychological makeup, and biological attributes play differing roles in protecting the health of disadvantaged men and women. The findings underscore the complex associations among SES, protective mechanisms, and health and offer new insight into how disadvantaged adults defy their odds of poor health.
Research on Aging 01/2011; 33(2-2):115-144. DOI:10.1177/0164027510391988 · 1.23 Impact Factor
Available from: Patama Vapattanawong
- "This study observes the U-shaped and reverse J-shaped association of BMI with all-cause mortality in older persons as found in many studies [33-36]; however, there is a no significant association in the highest BMI category due to the small number of subjects with BMI ≥ 35.0 kg/m2 (0.3% for men, 1.3% for women). The finding of BMI and mortality is consistent with several studies in the West and Asia [17,20-23,33-35]. "
[Show abstract] [Hide abstract]
ABSTRACT: To assess the association of body mass index with mortality in a population-based setting of older people in Thailand.
Baseline data from the National Health Examination Survey III (NHES III) conducted in 2004 was linked to death records from vital registration for 2004-2007. Complete information regarding body mass index (BMI) (n = 15997) and mortality data were separately analysed by sex. The Cox Proportional Hazard Model was used to test the association between BMI and all-cause mortality controlling for demographic, socioeconomic, and health risk factors.
During a mean follow-up time of 3.8 years (60545.8 person-years), a total of 1575 older persons, (936 men and 639 women) had died. A U-shaped and reverse J-shaped of association between BMI and all-cause mortality were observed in men and women, respectively. However there was no significant increased risk in the higher BMI categories. Compared to those with BMI 18.5-22.9 kg/m2, the adjusted hazard ratios (HR) of all-cause mortality for those with BMI <18.5, 23.0-24.9, 25.0-27.4, 27.5-29.9, 30.0-34.9, and ≥35.0 were 1.34 (95% CI, 1.14-1.58), 0.79 (95% CI, 0.65-0.97), 0.81 (95% CI, 0.65-1.00), 0.67 (95% CI, 0.48-0.94), 0.60 (95% CI, 0.35-1.03), and 1.87 (95% CI, 0.77-4.56), respectively, for men, and were 1.29 (95% CI,1.04-1.60), 0.70 (95% CI, 0.55-0.90), 0.79 (95% CI, 0.62-1.01), 0.57 (95% CI, 0.41-0.81), 0.58 (95% CI, 0.39-0.87), and 0.78 (95% CI, 0.38-1.59), respectively, for women.
The results of this study support the obesity paradox phenomenon in older Thai people, especially in women. Improvement in quality of mortality data and further investigation to confirm such association are needed in this population.
BMC Public Health 10/2010; 10(1):604. DOI:10.1186/1471-2458-10-604 · 2.26 Impact Factor
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.