[Development of a self-efficacy scale for going out among community-dwelling elderly]

Prevention and support for Tojikomori has been a focus of recent regional preventive efforts in reducing the need for nursing care in the elderly, requiring indicators for evaluating effects including psychological influences on the elderly. Behavior change theory suggests potential benefits of self-efficacy in outing to reduce Tojikomori elderly. However, evaluation scales for such psychological effects have hitherto been lacking. The purpose of this study was to develop a self-efficacy scale regarding going out among community-dwelling elderly (hereinafter referred to as the SEGE) and to assess its reliability and validity.

We collected survey items from 18 community-dwelling older people in A Ward, Tokyo. Based on these items, we developed a 13-item prototype scale through a preliminary survey among 258 community-dwelling elderly in O City, located in Japan. The main survey was conducted through the mail by randomly selecting 8,000 community-dwelling elderly in A Ward. The survey included the prototype scale' items and basic attributes, such as age and gender, and items from evaluation scales to be used to examine the validity of the prototype scale.

A total of 2,627 elderly people (1,145 men and 1,482 women, average age 73.8 +/- 6.6 years) were analyzed. Of these people, 86.1% left home at least once a week. Principal component analysis revealed that the prototype scale created through the preliminary survey had a one-factor structure. Through a stepwise variable selection procedure in exploratory factor analysis, a six-item scale was developed. The alpha coefficient of internal consistency was 0.96 for these six items, confirming high reliability. Lower outing frequencies tended to be associated with lower scores of the SEGE, which correlated significantly with self-efficacy of ADL, self-rated health, and health-related quality of life (QOL), confirming criterion-related and construct validity. In addition, a confirmatory factor analysis showed that SEGE and self-efficacy of ADL, although highly correlated with each other, measured different concepts.

A six-item and one-factor SEGE was developed with high reliability and validity confirmed. With this new indicator, we can measure the psychological effects of prevention and support approaches for Tojikomori. This scale is now expected to widely used in Japan.