With increasingly aging populations and increasing prevalence of diabetes, there are fears that the gains in healthy life expectancy that have been made may soon be reversed. Estimations of the impact of these trends on the burden of ill health and disability in old age depend on accurate estimations of the associated risks. Although it is clear that diabetes is associated with an increased risk of physical disability, the magnitude of this association is unclear.
Diabetes mellitus is a metabolic condition characterized by hyperglycemia due to defects in insulin action or secretion or both, resulting in end-organ damages such as ischemic heart disease, chronic kidney disease, strokes, peripheral neuropathy and peripheral vascular disease . Diabetes may lead to disability either via one of the above mentioned pathways or independently .
Physical disability is defined by ‘long term limitations in major activities of daily life’ . There are various measures of this limitation including limitations to activities such as feeding, washing, dressing, transferring in and out of a chair or bed, and getting to and using a toilet, all of which can be categorized as limitations to self-care activities of daily living (ADL). Mobility limitations are also used to measure disability and these can be self-reported limitations to various walking tasks or measured functions such as walking-speed.
Estimates of the association between diabetes and disability vary. A study of a prospective cohort of elderly Mexican-Americans free of disability at baseline (1993–1994) reported that participants with diabetes were twice as likely to report development of any limitation in lower body disability during a seven-year follow-up (hazard ratio (HR) 2.05, 95% confidence interval (CI) 1.58-2.67) compared to those without diabetes . Similarly, a prospective study of elderly women, commenced in 1986, reported a HR of 1.42 (95% CI 1.23-1.65) . In contrast, a cross-sectional study in Taiwan, revealed that for both men and women the risk of disability among those with or without diabetes did not significantly differ for men (odds ratio (OR) 1.1 (95% CI 0.6-2.2)) or women (OR 1.7 (95% CI 0.7-4.2)) . In a similar population cross-sectional survey in Hong Kong, Chau et al. reported an OR of 1.7 (95% CI 1.51-1.8) for disability in those undergoing treatment for diabetes when compared to those without diabetes .
Variations in these risk estimates are likely to reflect differences in the measurement of disability, the populations examined, baseline age of participants in the study, follow-up time, diabetes duration, method of ascertaining diabetes and disability status, as well as confounders adjusted for in multivariable regression models. Measurements of disability in various studies to date have included self-reported limitations to ADLs [4, 7, 8], instrumental ADLs [5, 8] or mobility tasks [4, 5] and measured mobility limitation such as a timed 8-foot walk , lower extremity function, balance , and hand function . Several of these studies have included only women.
Other factors which appear to influence the relationship between diabetes and disability include age, sex, education, smoking, lack of physical activity, ethnicity, obesity and other comorbidities including depression [2, 10, 11].‘Prediabetes’ defined by impaired glucose tolerance (IGT) or impaired fasting glucose (IFG) is an intermediate state where glucose levels do not meet the criteria for a diagnosis of diabetes but are above normal accepted levels . The risk of disability associated with prediabetes is less studied. In a small cohort (n=88), an increased risk of disability was observed for those having diabetes for over 15 years and increased glycosylated hemoglobin (HbA1C) but this was not the case for short-term or newly diagnosed diabetes or those with IGT .
We aim to conduct a systematic review of the association between diabetes, prediabetes and physical disability and quantify this relationship using a meta-analysis. If sufficient data permit, we further aim to examine the moderating effect of sex, age and duration of diabetes on the above mentioned associations.