Multidimensional dependency subgroups in community-dwelling older adults: A latent class analysis

Area of Science:

• Gerontology
• Public Health
• Sociology

Background:

• Understanding patterns of multidimensional dependency is crucial for effective interventions in aging populations.
• Previous research often employed variable-centered approaches, potentially overlooking individual heterogeneity.

Purpose of the Study:

• To identify distinct subgroups of older adults based on multidimensional dependency patterns using latent class analysis (LCA).
• To explore sociodemographic predictors associated with membership in these identified dependency classes.

Main Methods:

• Longitudinal data from 7,920 older adults in the Mexican Health and Aging Study (MHAS) were analyzed.
• Latent class analysis (LCA) was performed using MPlus to identify dependency profiles.
• Covariates were incorporated to assess their association with class membership.

Main Results:

• Three distinct dependency classes were identified: "Active older adults" (64%), "Relatively independent" (26%), and "Physically impaired" (10%).
• The "Active older adults" group showed high functioning across all indicators.
• "Physically impaired" individuals represented the most vulnerable subgroup.

Conclusions:

• A person-centered approach, identifying typologies of dependency, is more beneficial than a variable-centered approach for older adults.
• Findings support the development of tailored interventions based on identified dependency profiles.
• This study provides a valuable typology of dependency for future research and public health initiatives.