Objective: The aim was to determine whether mini mental state examination (MMSE) scores improved in older participants of a Functioning Improvement Tool (FIT) home-visit program. Methods: Two hundred fifty-two participants aged 65 years or older living at home and receiving preventive services or a community long-term care prevention project according to the Japanese social long-term care insurance system were enrolled and randomly assigned to an intervention group (n = 128) or a control group (n = 124). Intervention group subjects received a 60-min FIT home-visit program for 3 months, which included guidance, assistance, and help in writing and teaching calculation in order to complete the FIT. Control subjects did not receive any home visits. Cognitive function was evaluated by MMSE. Analysis of covariance was used to examine the effects of the FIT adjusting for baseline MMSE scores, age, and sex. Results: Fifty-three subjects were excluded because of withdrawal, hospitalization, death, relocation, or missing data of MMSE; 199 subjects (60 men, 139 women; age 78.6 +/- 7.4 years) were analyzed. The baseline MMSE scores did not differ between the intervention and control groups (24.2 +/- 4.3 vs. 24.1 +/- 4.7, p = 0.90). After the study period, the change in the MMSE score was significantly better in the intervention group than in the control group (0.8 +/- 0.3 vs. -0.1 +/- 0.2, p = 0.04). Stratified analyses showed that the intervention strategy was most effective in subjects with mild cognitive decline, with baseline MMSE scores from 18 to 23 points (1.9 +/- 0.5 vs. -0.1 +/- 2.8, p = 0.04). Conclusions: Our FIT home-visit program improved MMSE scores in older participants with mild cognitive decline. Copyright (C) 2011 John Wiley & Sons, Ltd.