Background
Cognitive impairment and reduced physical performance frequently coexist in older adults and are associated with frailty, functional dependence, falls, hospitalization, and other adverse health outcomes. Although this coexistence is often discussed within the broader concept of cognitive frailty, the present study evaluated cognitive status and physical performance as separate domains.
Aims
To analyze the association between cognitive status and lower-extremity physical performance in older adults attending a day-care center and to examine the robustness of this association across prespecified clinical adjustment sets.
Methods
This single-center cross-sectional study included 106 older adults attending one day-care center. Cognitive status was assessed using the Lobo Mini-Cognitive Examination (MEC-35), and physical performance was evaluated using the Short Physical Performance Battery (SPPB). Functional status, fall risk, chronic disease status, previous fractures, surgeries, and hospitalizations were recorded. Group differences were assessed using non-parametric tests, and directed acyclic graph (DAG)-informed, double-penalized generalized additive models (GAMs) were fitted using prespecified direct- and total-association adjustment sets. Bootstrap internal validation and a parsimonious sensitivity analysis were performed.
Results
Participants were predominantly women (64.2%) and had a mean age of 83.21±8.40 years. Mean SPPB and MEC-35 scores were 5.98±2.44 and 20.77±8.24, respectively. SPPB scores did not differ across MEC-35 categories. A positive association was observed only in the fully adjusted direct-association specification (p=0.038), whereas the total-association specification and the age- and sex-adjusted sensitivity model were null (p=0.221 and p=0.167, respectively). Model-predicted SPPB scores differed by approximately 2 points across the full MEC-35 range, although confidence intervals were wide at the extremes.
Discussion
The observed association was dependent on the selected adjustment set and was not reproduced in the unadjusted or parsimonious analyses. It should therefore be interpreted cautiously as exploratory and hypothesis-generating rather than as evidence of a causal, stable, or clinically established relationship.
Conclusions
Cognitive status and physical performance may be related in clinically vulnerable older adults, but the present findings were model-dependent and require confirmation in larger longitudinal cohorts. Multidomain geriatric assessment remains appropriate for clinical characterization; however, intervention implications cannot be inferred from this cross-sectional study.
