Scholarlink

Background <p>The self-medication hypothesis proposes that psychological distress may contribute to increased alcohol use, yet population-level evidence in Japan remains limited. To examine population-level associations between psychological distress (K6) and drinking status among working-age adults in Japan, with comprehensive adjustment for sociodemographic and behavioral covariates.</p> Methods <p>We analyzed anonymized official statistical microdata from the 2019 Comprehensive Survey of Living Conditions (Japan). Respondents aged 20–64 years with complete data were included (<i>n</i> = 3,935). Psychological distress was measured using K6 and categorized as low (0–4), moderate (5–12), and high (≥ 13). Drinking status was derived from frequency and amount per drinking day: heavy drinking was defined as estimated ethanol intake ≥ 44&#xa0;g/day (≥ 2 gou/day); non-drinkers included “rarely,” “quit,” or “do not drink”; remaining participants were moderate drinkers (reference). Multinomial logistic regression estimated adjusted odds ratios (aORs) for heavy vs. moderate and non-drinkers vs. moderate, adjusting for age, gender, household composition, education, employment, income quartile, smoking, and psychiatric outpatient care.</p> Results <p>Of 3,935 participants, 20.6% were heavy drinkers (<i>n</i> = 810), 38.8% moderate drinkers (<i>n</i> = 1,527), and 40.6% non-drinkers (<i>n</i> = 1,598). High distress accounted for 4.5% (<i>n</i> = 177) and moderate distress for 25.3% (<i>n</i> = 995). Psychological distress was not associated with drinking status in cross-tabulations (<i>p</i> = 0.578). In adjusted models, distress was not significantly associated with heavy drinking or non-drinking. Moderate distress showed a borderline tendency toward heavy drinkers (aOR 1.21, <i>p</i> = 0.073), while high distress was null (aOR 1.02, <i>p</i> = 0.927). Covariates showed robust associations, including lower odds of heavy drinkers and higher odds of non-drinkers among women, higher odds of both heavy drinkers and non-drinkers among adults aged 20–39 (vs. 60–64), and strong positive association between smokers and heavy drinkers.</p> Conclusions <p>Psychological distress was not an independent predictor of drinking status after adjustment, suggesting limited support for a simple population-level self-medication pathway in Japan. Longitudinal studies and subgroup analyses are needed.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Psychological distress and drinking patterns among working-age adults in Japan: a population-based analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月10日

查看全文
  • Shimpei Hanaoka,
  • Shigeru Fujita,
  • Takefumi Kitazawa,
  • Kanako Seto,
  • Kunichika Matsumoto,
  • Tomonori Hasegawa

摘要

Background

The self-medication hypothesis proposes that psychological distress may contribute to increased alcohol use, yet population-level evidence in Japan remains limited. To examine population-level associations between psychological distress (K6) and drinking status among working-age adults in Japan, with comprehensive adjustment for sociodemographic and behavioral covariates.

Methods

We analyzed anonymized official statistical microdata from the 2019 Comprehensive Survey of Living Conditions (Japan). Respondents aged 20–64 years with complete data were included (n = 3,935). Psychological distress was measured using K6 and categorized as low (0–4), moderate (5–12), and high (≥ 13). Drinking status was derived from frequency and amount per drinking day: heavy drinking was defined as estimated ethanol intake ≥ 44 g/day (≥ 2 gou/day); non-drinkers included “rarely,” “quit,” or “do not drink”; remaining participants were moderate drinkers (reference). Multinomial logistic regression estimated adjusted odds ratios (aORs) for heavy vs. moderate and non-drinkers vs. moderate, adjusting for age, gender, household composition, education, employment, income quartile, smoking, and psychiatric outpatient care.

Results

Of 3,935 participants, 20.6% were heavy drinkers (n = 810), 38.8% moderate drinkers (n = 1,527), and 40.6% non-drinkers (n = 1,598). High distress accounted for 4.5% (n = 177) and moderate distress for 25.3% (n = 995). Psychological distress was not associated with drinking status in cross-tabulations (p = 0.578). In adjusted models, distress was not significantly associated with heavy drinking or non-drinking. Moderate distress showed a borderline tendency toward heavy drinkers (aOR 1.21, p = 0.073), while high distress was null (aOR 1.02, p = 0.927). Covariates showed robust associations, including lower odds of heavy drinkers and higher odds of non-drinkers among women, higher odds of both heavy drinkers and non-drinkers among adults aged 20–39 (vs. 60–64), and strong positive association between smokers and heavy drinkers.

Conclusions

Psychological distress was not an independent predictor of drinking status after adjustment, suggesting limited support for a simple population-level self-medication pathway in Japan. Longitudinal studies and subgroup analyses are needed.