Impact of Longitudinal Changes in Dynamic Lower-Limb Function on Care-Needs Certification: A 4-Year Prospective Study of Older Women Attending Community Gathering Places (Kayoi-no-ba)

Background: Japan’s rapidly aging society faces a critical challenge in sustaining the public long-term care insurance (LTCI) system. Community gathering places (Kayoi-no-ba) have been promoted for preventive care, but the predictive value of longitudinal changes in physical performance measures (PPMs) within these settings is not fully understood. Objectives: Our objective was to clarify the association between baseline PPMs and their longitudinal change over four years with subsequent care-needs certification among community-dwelling older women. Methods: We conducted a 4-year prospective study of independent older women (n = 223–226), assessing handgrip strength (HGS), Timed Up and Go (TUG), the 30-second chair stand test (CS-30), and the one-leg standing test (OLST). Analysis 1 examined baseline risk; Analysis 2 tracked functional transition (maintenance vs. decline) among those at “no risk” at baseline. Results: Baseline PPM status was not significantly associated with future care-needs certification after adjusting for age and BMI. A longitudinal decline in TUG and CS-30 was associated with a significantly higher risk of certification (TUG: Hazard Ratio [HR] = 17.34, p < 0.001; CS-30: HR = 7.087, p = 0.002), which remained significant after Benjamini–Hochberg false discovery rate correction for multiple comparisons. However, a sensitivity analysis excluding participants certified before their 2023 re-measurement showed that the CS-30 association was no longer supported, while TUG remained nominally significant but with an extremely wide confidence interval (HR = 18.99, 95% CI 1.155–312.390, p = 0.039) reflecting very few events. A complementary continuous-change analysis corroborated this pattern (TUG: HR = 2.34 per 1-SD worsening, 95% CI 1.64–3.33, p < 0.001); no association was observed for HGS, CS-30, or OLST in this or the primary analyses. Conclusions: A longitudinal decline in TUG performance may be associated with increased care-needs certification risk among older women, but this finding is sensitive to the temporal ordering of measurement and certification and requires confirmation in larger, prospective studies. The CS-30 finding did not withstand sensitivity analysis and should not be treated as a robust predictor.