Prognostic Factors Associated with Functional Decline Among Long-Term Care Residents: A Systematic Review

Functional decline is common among LTC patients and is linked to poor outcomes, such as higher levels of dependency, hospitalizations, and death. Early phase after admission into facilities can be considered a crucial time frame when residents might have a rapid deterioration. Nevertheless, there is limited knowledge regarding predictors of early functional decline among LTC patients. Main objective of this research is to systematically identify and synthesize prognostic factors associated with early functional decline among long-term care residents. A systematic review was carried out based on PRISMA guidelines. The search strategy involved the use of electronic databases such as PubMed, Embase, Scopus, Web of Science, and Cochrane Library. Studies involving predictors of functional impairment among nursing home residents were included in the review. Extraction and assessment of study data were done using standardized instruments. Heterogeneity across studies led to a narrative synthesis. A total of seven studies with more than 400,000 LTC facility residents were considered. Functional decline was mainly measured using ADL-based measures. Cognitive impairment and preexisting functional state proved to be the most robust and significant predictors of functional decline. Other risk factors like urinary incontinence, co-morbidities, and impaired balance were found to increase the risk. Depression and medication usage showed inconclusive evidence. Geriatrician presence at facilities and disruption at the systems level, like the coronavirus pandemic, decreased and increased functional decline, respectively. The occurrence of functional decline in LTC facilities is complex and results from the interaction between vulnerability and the environment. Early detection of at-risk patients through a thorough assessment can help implement appropriate strategies. There is a need for more studies to validate outcome criteria and modifiable predictors.