Daniel M. Keller, PhD

May 03, 2013

Taipei, Taiwan — A study of almost 400 elderly adults in Taiwan has shown no association between mild cognitive impairment (MCI) and the risk for falls after adjustment for confounding factors. However, an expert consultant on elder care issues raises some other possible factors that may contribute to fall risk.

"Before we adjusted for the potential risk factors...[cognitive test] score and attention show a significant relationship for risk, which means that if the elder has a better performance on the cognitive function, their fall risk is lower," said lead author Han Fu, a graduate student in epidemiology and preventive medicine at National Taiwan University in Taipei, Taiwan. "However, after we adjusted for the potential risk factors, the significant association disappeared."

She presented the findings here at the Alzheimer's Disease International (ADI) 28th International Conference.

Leading Cause of Injury

Citing data from the Centers for Disease Control and Prevention, she said accidental falls are the leading cause of fatal and nonfatal injuries among people older than 65 years. A study in Taipei had shown that 1 in 6 seniors had fallen at least once in the prior year. Among the Taiwanese elderly population, 85% to 10% have cognitive impairment.

Previous studies around the world have shown that fall risk correlated with diminished memory scores, global cognitive function, and executive function.

The present cross-sectional study aimed to investigate the association between individual cognitive domains and fall risk among elderly persons in Taiwan. It involved people 65 years of age or older attending an annual wellness visit.

Researchers compared cognitive function in the cohort experiencing 1 or more falls ("fallers") in the past year with cognitive function among a cohort without falls ("nonfallers") using a self-reported questionnaire. Cognitive function was assessed using the MoCA-T, a Taiwanese version of the Montreal Cognitive Assessment, a rapid screening instrument for MCI (total possible score of 30; 26 or greater is considered normal).

The cohort of fallers (n = 86) was slightly older, with a mean age of 74.2 years, than nonfallers (n = 306), whose mean age was 72.3 years (P = .004). Fewer of the fallers (9%) engaged in vigorous physical activities compared with nonfallers (19%; P = .04). The cohorts were similar with regard to sex, demographic characteristics, education, body mass index (BMI), comorbid conditions, alcohol and cigarette use, and depression scores.

However, the cohorts differed in their MoCA-T total scores and in the attention domain of the test. Fallers had a total score of 24.58 ± 5.44 vs 25.71 ± 3.29 (P = .02) for nonfallers. Both of these groups would be considered to have some MCI. Attention scores were 5.35 ± 1.14 vs 5.64 ± 0.69, respectively (P = .005). Their visuospatial/executive, naming, language, abstraction, memory, or orientation scores did not significantly differ.

Unadjusted odds ratios for fall risk were 0.93 (95% confidence interval [CI], 0.88 - 0.99) according to the MoCA-T total score and 0.68 (95% CI, 0.53 - 0.89) for the attention domain. However, after adjustment for age, sex, years of education, physical activity, BMI, hypertension, and depression score, these variables were no longer significantly associated with fall risk.

Fu concluded that the prevalence of at least 1 fall in the previous year was 22% but that cognitive function was not significantly associated with fall risk after adjustment for confounders. In light of several previous studies showing a relationship, she suggested that "the tool we used now is not complete enough" and that future research should use more sensitive cognitive assessments to clarify the risk for accidental falls in the elderly.

Key Factors

But session moderator David Troxel, MPH, a long-term care consultant and writer in the field of dementia in Sacramento, California, commented to Medscape Medical News about other key factors contributing to falls in the elderly, which may have overshadowed the cognitive measures that were formally measured in the study.

"A person with dementia, they don't remember to use their walker. They have visual-spatial issues. So I think the consensus certainly right now is that people with dementia are at a higher risk for falls, but on the other hand, there's just also that general issue of are they getting enough exercise, are they weak, are they eating well," he said. "So I think from her study I guess what I'd say is maybe not blame it all on the dementia. Really take a look at the broader factors."

Some of those factors in the elderly in general would be visual problems leading to missing a step or a grab bar. Adding to the risk for dementia in elderly people may be poor judgment, leading, for example, to chasing a pet that ran out the door, Troxel suggested.

"I think her study is interesting, and maybe the difference [in falls] is modest between dementia and nondementia," he said, given all the other possible contributing factors. Regardless of the cause of falls, they are devastating to older people; the average life expectancy is only 18 months after an elder breaks a hip, Troxel noted.

Ms. Fu has disclosed no relevant financial relationships. Mr. Troxel is a dementia consultant to Atria Senior Living, Home Instead, American Baptist Homes of the West, and Prestige Care, and he previously consulted for Genentech Roche.

Alzheimer's Disease International (ADI) 28th International Conference.  Abstract OC019. Presented April 19, 2013.