Construct Validity of the Life-Space Assessment in Institutionalized Settings Among Hospitalized Patients: A...

Feasible assessments of hospitalized patients' mobility are essential to support the promotion of mobility. We investigated the construct validity of the Life-Space Assessment in Institutionalized Settings (LSA-IS) and direct observation toward the goal of including patients who are often excluded from device-based assessments.

Cross-sectional study.

Patients admitted to a convalescent rehabilitation hospital (n = 119; median age, 83.0 years; 63.9% female).

The LSA-IS was measured by an assigned physiotherapist. Direct observation was performed on a single day at 10-min intervals between 8:00 and 17:00. Physical activity and location observations were recorded. Locations (own room, within the ward, and outside the ward) and physical activity (lying, sitting, and moving) were categorized into 3 groups for analysis. Construct validity between the LSA-IS results and the direct observations was assessed using Spearman's rank correlation. Hypotheses regarding the expected direction and magnitude of correlations were specified a priori to evaluate construct validity.

The median LSA-IS score was 14.0 points (interquartile range, 9.5). The LSA-IS was significantly positively correlated with the time spent moving (ρ = 0.676), significantly negatively correlated with the time spent lying down (ρ = -0.609), significantly positively correlated with the time spent sitting (ρ = 0.416) and the time spent within and outside the ward (ρ = 0.348 and ρ = 0.427, respectively), and significantly negatively correlated with time spent in the patient's own room (ρ = -0.487). The LSA-IS was more strongly correlated with moving vs lying down or sitting. These results supported our hypotheses and indicate sufficient construct validity. The correlation between the LSA-IS and the time spent moving remained strong after stratifications by age, diagnostic category, and cognitive status.

The LSA-IS demonstrated moderate-to-high correlations with hospitalized patients' physical activity and location, and it may provide a clinically applicable method for evaluating mobility in hospitalized patients.