Beyond Continuity: A Binary-Outcome Cross-Lagged Panel Model of Ikigai and Social Interaction Among...

To examine the longitudinal and potentially bidirectional association between Ikigai and social interaction among community-dwelling older adults, and to evaluate the feasibility of a binary-outcome cross-lagged panel model (CLPM).

This study analyzed data from the Community Empowerment and Care (CEC) cohort in T-village in central Japan. A total of 504 participants aged ≥ 65 years who took part in both the 2017 and 2020 surveys were analyzed. Ikigai was assessed by a self-reported question. Social interaction was measured with the Index of Social Interaction (ISI). A two-wave CLPM with binary outcomes was modeled using the weighted least squares mean and variance adjusted (WLSMV) estimator in Mplus 8.7, and baseline-adjusted logistic regression analyses were additionally conducted using SPSS 30.0. Covariates assessed at baseline included age, gender, subjective economic status, depressive symptoms, subjective cognitive complaints, negative life events (NLE), chronic diseases, smoking behavior, drinking behavior, social interaction, and Ikigai.

In the CLPM, richer social interaction in 2017 significantly predicted Ikigai in 2020 (β = 0.165, p = 0.006), whereas the reverse pathway from baseline Ikigai to subsequent social interaction did not reach conventional statistical significance (β = 0.102, p = 0.060). Autoregressive effects for both Ikigai and social interaction were substantial, and overall model fit indices indicated an acceptable fit.

Over 3 years, richer social interaction prospectively supported Ikigai, while Ikigai alone did not translate into greater subsequent interaction once temporal dependencies were modeled. Binary-outcome CLPM proved feasible and informative for examining categorically measured psychosocial constructs in aging research. Interventions that expand opportunities for meaningful social interaction may help sustain Ikigai in later life.