Miriam E. Tucker
February 28, 2014
Functional decline and disability increase the risk for developing diabetes, a large observational study of adults aged over age 50 years shows.
The findings, from over 22,000 participants in the Health and Retirement Study (HRS), were published online February 18 in Diabetes Care by Barbara H. Bardenheier, PhD, and colleagues, of the Centers for Disease Control and Prevention’s Division of Diabetes Translation.
“We were surprised, because we have traditionally thought of diabetes as preceding disability, rather than the other way around. Controlling for mediators did not affect estimates very much. This suggests that there might be other intrinsic aspects of functional decline and disability that affect diabetes risk. If so, it is something that is currently poorly understood,” Dr. Bardenheier told Medscape Medical News.
“We think that long-lasting exercise programs — including muscular-strengthening exercises and walking, as well as programs to maintain healthy weight — may help maintain physical function and at the same time reduce diabetes,” she added.
Asked to comment, Andrea Kriska, PhD, and Bonny Rockette-Wagner, PhD, both from the University of Pittsburgh, Pennsylvania, said while this study may be the first to directly link disability and subsequent diabetes, previous findings have laid the groundwork by linking physical activity to diabetes risk.
“Those relationships are pretty well-known… This study didn’t tie in the mechanism, ie, physical activity, which is likely a big piece of this,” said Dr. Kriska. Nonetheless, noted Dr. Rockette-Wagner, “It’s a well-done study, and the information is useful. Although some of these connections had been shown before, they did add some novel information, including attributable risk and time spent in disability, which is useful from a clinical perspective.”
When Disability Precedes Diabetes
The HRS is a population-based longitudinal survey of people aged 51 and older, who were polled every 2 years between 1998 and 2010. The investigators looked at 2 groups: A “main study sample” of 22,876 participants who did not have diabetes at baseline, irrespective of disability status, and another “substudy” sample of 12,242 who had neither disability nor diabetes at baseline.
Disability was assessed via a series of questions about the level of difficulty experienced while performing everyday activities, including walking, climbing a flight of stairs, stooping, and pushing or pulling a large object. Based on the answers, subjects’ disability levels were classified as none, mild, moderate, or severe. Because disability status changed over time for more than half of the subjects, the investigators also factored in the number of years spent in each disability state.
In the main study sample, 15.5% developed diabetes, based on self-reports — the majority of which would have been type 2 diabetes — over an average 8.7-year follow-up and 25.6% died. In the substudy, 60.5% reported some level of disability over the study period compared with 76.7% in the main study.
Cumulative diabetes incidence in the main study was 10.8 per 1000 person-years among those who never had any disability during the study. In contrast, for those with disability, cumulative diabetes incidences were 15.9, 19.0, and 21.7 per 1000 for mild, moderate, and severe disability, respectively.
Mortality — examined as a competing risk — also increased by disability level at baseline in the main study, with rates of 10.8, 19.5, 36.9, and 71.5 per 1000 person-years for those with zero, mild, moderate, and severe disability, respectively.
In the substudy, diabetes incidence for those with zero, mild, moderate, and severe disability was 10.7, 17.3, 17.8, and 20.5 per 1000 person-years, respectively, and mortality rates were 10.8, 18.9, 37.7, and 83.4 per 1000 person-years, respectively.
Controlling for a list of factors such as sex, race, educational level, age, and baseline body mass index yielded a greater risk for diabetes for each level of disability. In the main study, risk ratios for diabetes compared with those with no baseline disability were 1.28 for mild, 1.43 for moderate, and 1.63 for severe disability.
The diabetes risk was even higher in the substudy, with risk ratios of 1.40, 1.52, and 1.81 for mild, moderate, and severe disability, respectively, compared with none.
Calculated population-attributable fractions, or the estimated proportion of diabetes cases that would not have occurred had the disability not been present, were 11.1% in the main study and 6.9% for the substudy, Dr. Bardenheier and colleagues report.
What is Disability?
Drs. Kriska and Dr. Rockette-Wagner told Medscape Medical News that the substudy appears stronger than the main study, but that it still gives an incomplete picture without data on physical activity or on the specific reasons for the disability.
“In the substudy, the assumption is that what’s changed with the disability is the level of activity and/or the level of sedentary behavior. It’s likely that when you become disabled you do less moderate physical activity and become more sedentary, both of which have been independently shown to be related to the development of diabetes,” Dr. Kriska said.
Dr. Rockette-Wagner suggested that the subjects categorized as having “mild” disability may represent a group that’s simply sedentary, whereas the “moderate” or “severe” groups might have specific problems such as injury or crippling illness. “The severe group might be different,” she noted.
Indeed, the study authors acknowledge this. Dr. Bardenheier told Medscape Medical News, “Yes, this is a limitation. Our study did not examine the upstream factors that led to disability in the first place — particularly, we were unable to assess the effect of physical activity. However, we controlled for age, race, and socioeconomic status, so we don’t think these factors explained the association with subsequent diabetes.”
Dr. Kriska said that the study highlights current reality. “We are taxing our muscles less, so they’re atrophying more. A lot of the problems of aging could be prevented by moving those muscles.
“I think the take-home message is yes, functional disability is a detrimental thing and can lead to diabetes. You need to try to improve functional ability. That’s where activity and inactivity play a large role.”
Dr. Bardenheier and coauthors and Drs. Kriska and Rockette-Wagner have reported no relevant financial relationships.
Diabetes Care. Published online February 18, 2014. Abstract