Despite widespread claims that social changes have intensified loneliness, no study has yet examined how loneliness has changed over time in Japan. To address this gap, graduate student Momo Homma from the Department of Psychology at Chuo University, Japan and Professor Kenkichi Takase from the same institution conducted a cross-temporal meta-analysis to investigate how loneliness has changed over time in Japan and identify the factors contributing to this change. The results indicated that loneliness increased in Japan from 1983 to 2023. Furthermore, the increase was particularly pronounced among adolescents and women, and social indicators such as marriage rates were found to covary with loneliness. These findings provide a foundation for future research on loneliness and help identify target populations for interventions aimed at reducing loneliness.
Previous research has linked loneliness to mental and physical health, as well as to health-related behaviors, and has also shown that it is associated with increased mortality risk. The World Health Organization has identified loneliness as "a global public health concern" and has positioned it as a public health priority. In Japan, loneliness is also emerging as a social issue. However, its long-term trends have not been systematically examined. Therefore, in this study, we used a cross-temporal meta-analysis to investigate how loneliness has changed over time in Japan.
In this study, we searched four databases (PubMed, Web of Science, J-STAGE, and CiNii) to identify studies conducted in Japan that used the UCLA Loneliness Scale. In the English-language databases, we used the search terms "UCLA Loneliness Scale, Japan" and "UCLA Loneliness Scale, Japanese." In the Japanese databases, we searched for "UCLA Loneliness Scale (in Japanese)." We also included additional studies identified through the review process. The search yielded 333 records. After removing duplicates and screening abstracts, 251 studies remained. We then applied the exclusion criteria and conducted a full-text review, resulting in a final sample of 81 studies comprising 183 datasets (N = 49,054) included in the meta-analysis. Subsequently, we performed data coding and score adjustments.
The main analysis revealed a significant increase in UCLA Loneliness Scale scores from 1983 to 2023. This finding suggests that loneliness in Japan may have worsened over the past 40 years. Secondary analyses also revealed effects of developmental stage and gender. Regarding developmental stage, loneliness increased among adolescents. In terms of gender, although men consistently showed higher levels of loneliness, an increasing trend was observed only among women. Further analyses also revealed the influence of social indicators. When loneliness scores were compared before and during the COVID-19 pandemic, loneliness score was higher during the pandemic. In addition, the increase in loneliness covaried with social indicators such as the number of single-person households, marriage rates, GDP, and Internet use.
This study is the first to investigate how loneliness has changed over time in Japan. The results show a substantial increase in loneliness and factors contribute to this trend. The findings highlight loneliness as a pressing social issue and provide key evidence supporting the need for further scientific research and policy interventions.
Increases in loneliness were particularly evident among adolescents and women, highlighting the need for targeted support. To address loneliness in adolescence, future work will explore whether dietary interventions can help reduce loneliness. If effective, we will further examine the underlying biological mechanisms, including the gut–brain axis. To support women, particularly mothers, we are developing a generative AI–based support system (an "AI childcare worker") designed to alleviate loneliness. We will evaluate its effectiveness and work toward its real-world implementation.
【Glossary】
1) Loneliness
Loneliness is defined as a subjective negative emotional state resulting from a perceived discrepancy between desired and actual interpersonal relationships. It is distinguished from social isolation, which denotes an objective lack of social connections.
2) Cross-temporal meta-analysis
A method of meta-analysis that uses mean scores of psychological measures to examine changes in those measures over time. Meta-analysis integrates findings from multiple studies to draw overall conclusions and is considered a high level of evidence.
