Risk Factors for Falls Identified in Older Adults With Diabetes

In older adults with diabetes, decreased peroneal compound muscle action potential, higher levels of cystatin-C, and poor contrast sensitivity each increased the risk for falls, according to the results of a study reported in the March issue of Diabetes Care.

"Older adults with type 2 diabetes are more likely to fall, but little is known about risk factors for falls in this population," write Ann V. Schwartz, PhD, from the University of California, San Francisco in San Francisco, and colleagues from the Health, Aging, and Body Composition Study. "We determined whether diabetes-related complications or treatments are associated with risk of falls in older diabetic adults."

At annual visits, 446 participants with diabetes in the Health, Aging, and Body Composition cohort of well-functioning older adults reported falls in the previous year. Mean age at enrollment was 73.6 years, and average duration of follow-up was 4.9 years. Odds ratios (ORs) for more frequent falls were estimated with continuation ratio models.

The proportion of patients who reported falling was 24% in the first year and 22%, 26%, 31%, and 30% in each subsequent year, respectively.

Adjusted models revealed that factors associated with the risk for falls were decreased peroneal nerve response amplitude (OR 1.50; 95% confidence interval [CI], 1.07 - 2.12, worst quartile vs others); higher cystatin-C, which is a marker of reduced renal function (OR, 1.38; 95% CI, 1.11 - 1.71, for 1 SD increase); poorer contrast sensitivity (OR 1.41; 95% CI, 0.97 - 2.04, worst quartile vs others); and low hemoglobin A_1c (HbA_1c) levels in insulin users (OR, 4.36; 95% CI, 1.32 - 14.46; HbA_1c level ≤ 6% vs > 8%). In patients treated with oral hypoglycemic agents but not insulin, low HbA_1c level was not associated with the risk for falls (OR, 1.29; 95% CI, 0.65 - 2.54, HbA_1c level ≤6 vs > 8%).

Although adjustment for physical performance explained some of these associations, it did not explain them all.

"In older diabetic adults, reducing diabetes-related complications may prevent falls," the study authors write. "Achieving lower A1C levels with oral hypoglycemic medications was not associated with more frequent falls, but, among those using insulin, A1C ≤6% increased risk of falls."

Limitations of the study include observational design limiting causal inferences; self-reported falls causing the potential for misclassification; peripheral nerve function and vision measured 3 and 2 years, respectively, after baseline; enrollment limited to well-functioning participants; lack of gold standard measurement of glomerular filtration rate; and participants not queried about hypoglycemia.

"Diabetes-related complications (reduced peripheral nerve function, renal function, and vision) contribute to risk of falls in older adults with diabetes," the study authors conclude.

The National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute on Aging, and the Intramural Research Program of the National Institutes of Health, National Institute on Aging, supported this study. The costs of publication of this article were defrayed in part by the payment of page charges, mandating that it must therefore be hereby marked "advertisement."

Clinical Context

Older patients with type 2 diabetes have an increased risk for falls, likely associated with complications of peripheral neuropathy, reduced vision, and poor renal function, but little is known about the magnitude of risk factors for falls in these patients. Other factors such as weight loss and level of glycemic control may also affect fall risk. Current recommendations for an HbA_1c goal of 7% or lower for older adults are not met for many older patients with diabetes, and a higher goal is recommended for adults who are frail.

This is a longitudinal cohort study within a study of 3075 men and women aged 70 to 79 years from 2 US sites in the Health, Aging, and Body Composition Study, of whom 446 had diabetes. In these patients, peripheral nerve function was measured 3 years after baseline, and the patients were observed for risk factors for falls.

Study Highlights

• Adults studied had type 1 or 2 diabetes by self-report, use of hypoglycemic agents, clinician diagnosis, or high fasting glucose level or 2-hour oral glucose tolerance test at baseline.
• Participants were asked at the annual visit, "In the past 12 months, have you fallen and landed on the floor or ground?" with 1, 2 to 3, 4 to 5, and 6 or more as possible responses.
• Follow-up visits occurred once yearly for 4.9 years.
• Of the 446 participants, 6% died between visits 4 and 6, and 2.5% did not complete visit 6.
• At visit 3, high contrast visual acuity, contrast sensitivity, and depth perception were measured.
• Peripheral nerve function was measured at visit 4.
• Light touch discrimination was measured with esthesiometer probes and lower extremity vibration sensitivity with a Vibratory Sensory Analyzer (Medoc, Minneapolis, Minnesota).
• Motor nerve conduction studies were performed for the peroneal nerve.
• Nerve conduction velocity was determined between the ankle and the popliteal fossa or fibular head.
• Estimated glomerular filtration rate was calculated to determine the presence of renal disease, and cystatin-C was measured.
• Blood pressure was measured at baseline and at visits 2, 4, and 6; prescription and over-the-counter medications were assessed at all visits except visit 4.
• Physical performance tests were obtained at baseline and at visits 4 and 6, and strength was tested at visit 2.
• Participants completed a 6-meter walk, a 6-meter narrow walk, and 5 chair stands; standing balance was assessed with semi-tandem, full tandem, and 1-leg balance tests.
• Mean age at baseline was 73.6 years, 55.4% were men, HbA_1c level was 7.6%, and 14% of patients reported insulin use at baseline.
• Mean body mass index (BMI) was 28.8 kg/m², mean weight was 80.7 kg, mean duration of diabetes was 12.8 years, and of those who used oral agents, 40.7% reported using sulfonylureas.
• Among patients using insulin, 6.1% had an HbA_1c level of 6% or less, 17.6% had an HbA_1c level of 6% to 7%, 34.2% had an HbA_1c level of 7% to 8%, and 42.1% had an HbA_1c level of more than 8%.
• At baseline, 62% had known diabetes and the remaining participants were diagnosed with use of blood glucose measurements.
• In the first year, 24% reported falling and the rate of falls was 22%, 26%, 31%, and 30% in the subsequent years.
• In patients who did not use insulin, an HbA_1c level was not associated with fall risk.
• In those who used insulin, HbA_1c level of 6% or lower was associated with an increased OR for falls of 4.36 vs an HbA_1c level of more than 8%.
• Low peroneal nerve response amplitude at the popliteal fossa was independently predictive of falls.
• Peroneal nerve conduction velocity between the ankle and popliteal fossa or fibular head was not associated with higher risk for falls.
• Seated or standing blood pressure was not associated with falling, but diastolic, and not systolic, change in blood pressure was associated with increased fall risk.
• Weight loss, but not BMI, was associated with increased fall risk.
• Grip strength; knee extensor strength; standing balance time; chair stands; and the 6-meter walk, but not the 6-meter narrow walk, were associated with falls.
• Higher cystatin-C level (OR, 1.38), poor contrast sensitivity (OR, 1.41), and reduced peroneal nerve amplitude (OR, 1.50) were associated with falls.

Pearls for Practice

• An HbA_1c level of 6% or lower vs 8% or higher is associated with falls in older patients with diabetes who use insulin but not in those who use oral hypoglycemic agents.
• Predictors of falls in older patients with diabetes are reduced peripheral nerve function, poor vision, weight loss, and poor renal function.