Review: Several risk factors are associated with functional status decline in community-living older persons
ACP J Club. 1999 Nov-Dec;131:76. doi:10.7326/ACPJC-1999-131-3-076
Stuck AE, Walthert JM, Nikolaus T, et al. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999 Feb;48:445-69.
Which individual risk factors are associated with functional status decline in elderly persons living in the community?
Studies published in peer-reviewed journals in English, French, German, Italian, and Spanish were identified by searching MEDLINE, EMBASE/Excerpta Medica, PsycInfo, and SOCA (Sociological Abstracts) from 1985 to 1997. Search terms included aged; disab: (with various endings); impaired, limit: (with various endings); and decline combined with function and study or trial. Bibliographies of relevant papers and book chapters were reviewed, and experts were contacted.
Longitudinal studies were selected if they used functional outcome measures, including difficulties in activities of daily living, disability, or physical limitation. Studies were excluded if they were case-control or cross-sectional studies; assessed only cognitive, emotional, or sensory functioning or only specific subcomponents of functional status; used a single-item measure of functional status; only addressed sociodemographic factors; reported no statistical analyses about the predictive value of risk factors; or included participants primarily selected from nursing homes or hospitals.
Data were extracted on definition of functional outcome, sample size, base-line age and mortality rate, and length of follow-up. Methodologic quality of each study was assessed on the basis of sample derivation, sample characteristics at baseline, collection of outcome data, attrition, and statistical analyses.
78 studies met the inclusion criteria. Risk factors assigned the highest rating for strength of evidence (i.e., ≥ 2 appropriate analyses from separate databases, excluding persons with functional impairment at baseline and reporting a significant risk increase) were depression (11 studies), no alcohol consumption (3 studies), cognitive impairment (15 studies), comorbid conditions (21 studies), reduced observed lower-extremity performance (7 studies), low body mass index (5 studies), high body mass index (5 studies), low physical activity (21 studies), poor self-rated health (13 studies), smoking (17 studies), low frequency of social contact (13 studies), and poor self-reported vision (12 studies).
In elderly persons living in the community, an increased risk for functional status decline is associated with depression, no alcohol consumption, cognitive impairment, comorbid conditions, reduced observed lower-extremity performance, low or high body mass index, low physical activity, poor self-rated health, smoking, low frequency of social contact, and poor self-reported vision.
Sources of funding: Robert Bosch Foundation; Swiss National Science Foundation; Novartis Foundation for Gerontological Research.
For correspondence: Dr. A.E. Stuck, Geriatric Research Unit, Department of Geriatrics and Rehabilitation, Zieglerspital, Morillonstrasse 75, CH-3007 Bern, Switzerland. FAX 41-31-970-7767.
The purpose of the systematic review by Stuck and colleagues was to identify risk factors in community-dwelling elderly persons that predict subsequent functional status decline. The extent of functional decline was not specified, and 25 of 78 studies included persons who were < 65 years of age. From an evidence-based perspective, the review is sound; it is based on an extensive review of the literature with a priori exclusion criteria and rankings of the methodologic strengths of each study. However, the length of follow-up in the studies varied from 12 to 324 months, and some of the studies did not report relative risks with confidence intervals for each risk factor.
The findings make clinical sense. Depression, cognitive impairment, and baseline functional status impairment have been associated with increased mortality, morbidity, and likelihood of nursing home placement among hospitalized elderly persons (1-3). Comorbid conditions were defined as the number of chronic medical conditions. However, the nature and severity of the medical problems also affect function and prognosis. Indeed, the authors found that certain self-reported medical conditions (hypertension, stroke or transient ischemic attack, arthritis, and diabetes mellitus) were predictive of functional decline. Dementia is an important omission. In addition, it does not make intuitive sense that "no alcohol intake" should be a risk factor for functional decline, except that alcohol may be protective against coronary disease.
The findings of this study are useful, from a preventive standpoint, to identify the elderly persons most at risk for functional decline over time. Effective measures can then be implemented as part of the therapeutic plan to correct the modifiable factors.
Suzanne D. Fields, MD
University Medical Center at SUNY Stony BrookStony Brook, New York, USA