『保健医療科学』 保健医療科学 2025 Vol.74 No5 p.479-487（2025年12月）
＜原著＞
中高年期における主観的健康観の規定要因―日本の社会的背景を踏まえた性別層別分析―
岩瀬裕三子1），細川陸也1.2）
1） 京都大学大学院医学研究科
2）京都府立医科大学大学院 保健看護学研究科
Key determinants of self-rated health among middle-aged and older adults: A sex-stratified analysis in the context of Japanese society
IWASE Yumiko1）， HOSOKAWA Rikuya 1.2）
1） Graduate School of Medicine, Kyoto University
2） Graduate School of Nursing for Health Care Science, Kyoto Prefectural University of Medicine
＜抄録＞
目的：主観的健康観（Self-rated health: SRH）は，罹患率や死亡率を予測する指標であり，健康状態と社会的背景により形成される．日本では「介護は家族が担い，男性が家計を支える」といった性別規範に根差した社会的背景が存在する．これまでもSRHにおける性差を検討した研究は存在するが，日本の社会的背景を考慮し，理論的枠組みに基づいた研究は限られている． Jylhäの理論によると，SRHは疾病や身体の不調などの健康構成要素と性別や文化などの評価枠組みとの相互作用により形成される．本研究では，Jylhäの理論的枠組みに基づき，50歳以上の日本人を対象として，SRHと身体的健康，社会経済的要因，生活習慣との関連を性別に検討した．
方法：本研究では，50歳以上の日本人を対象とした 2012年の「中高年者パネル調査」データ（n = 2,826）を用いて横断的分析を行った．SRHを 3 水準の順序変数とし，性別に層別した順序ロジスティック回帰分析を実施した．独立変数には身体的健康(肥満関連疾患，日常生活における介助の必要性),社会経済的要因(就業状況，世帯収入，学歴，管理職経験)，婚姻状況，5 つの健康行動を含め，Brant検定によりモデルの仮定を確認した．
結果：日常生活における介助の必要性は，男性でオッズ比（Odds Ratio: OR）= 0.17，95%信頼区間（Confidence Interval: CI）:0.11–0.28，女性でOR = 0.12，CI: 0.07–0.19（男女ともp < 0.001）となり，SRHの低下と最も強い関連があった．就業状況は，男女ともSRHの高さと関連し，女性でやや強かった（男性: OR = 1.32, CI: 1.00–1.74,p = 0.048, 女性: OR = 1.45, CI: 1.11–1.90,p = 0.006）．世帯収入は，男性の中所得層においてのみSRHの高さと有意な関連が認められた（OR = 1.38,CI: 1.02–1.85,p = 0.034）．定期的な運動は，男性（OR = 1.48,CI: 1.15–1.90,p = 0.002），女性（OR = 1.38, CI: 1.08–1.77, p = 0.011）ともにSRHの高さと有意に関連していた．一方，規則的な生活習慣は女性でのみ有意な関連がみられた（OR = 1.30, CI: 1.02–1.67, p = 0.037）．Brant検定により比例オッズの仮定は確認され，多重共線性もなかった．
結論：Jylhäの理論的枠組みに基づき，性別層別分析した結果,日常生活における介助の必要性がSRHとの関連で最も強かった．就業は男女ともにSRHと関連し，女性でより強い関連がみられた．世帯収入は，女性では有意な関連がみられず，男性でも高所得との関連は認められなかった一方，中所得層では有意な関連が確認された．これらの結果は，収入の評価に性差がある可能性や，一定以上の所得ではSRHとの関連が弱まる天井効果を示唆する．さらに,健康的な生活習慣との関連も性別により異なっており, SRHが健康状態と社会的背景の双方に基づくというJylhäの理論的枠組みに合致していた．SRHの向上には，性差を考慮した多面的支援が求められる．
キーワード：主観的健康観，性差，健康行動，社会経済的要因，就業状況
Abstract
Objectives: Self-rated health (SRH) is a strong predictor of morbidity and mortality, shaped by both health conditions and social context. In Japan, gender norms assign caregiving to families and financial roles to men. While sex differences in SRH have been studied, few have used a theoretical framework. According to Jylhäʼs framework, SRH is formed through the interaction between health conditions such as disease and physical
symptoms, and evaluative factors including gender, age, and cultural context. This study applied Jylhäʼs framework to examine SRH associations by sex among Japanese adults aged 50 and older.
Methods: We conducted a cross-sectional analysis using data from the 2012 Middle-aged and Older Adults Panel Survey of individuals aged 50 and older (n = 2,826). Sex-stratified ordinal logistic regression was applied with SRH as a three-level outcome. Independent variables included physical health (obesity-related diseases and need for daily assistance), socioeconomic factors (employment status, household income, educational
attainment and managerial experience), marital status, and five health behaviors. The Brant test confirmed model assumptions.
Results The need for daily assistance was most strongly associated with lower SRH in both men (odds ratio [OR] = 0.17, 95% confidence interval [CI]: 0.11–0.28, p < 0.001) and women (OR = 0.12, CI: 0.07–0.19,p < 0.001). Employment status was positively associated with higher SRH in both men (OR = 1.32, CI:1.00–1.74, p = 0.048) and women (OR = 1.45, CI: 1.11–1.90, p = 0.006). Household income was significantly associated with higher SRH only among middle-income men (OR = 1.38, CI: 1.02–1.85, p = 0.034). Regular exercise was positively associated with SRH in both sexes (men: OR = 1.48, CI: 1.15–1.90, p = 0.002;women: OR = 1.38, CI: 1.08–1.77, p = 0.011), while regular lifestyle was significant only in women (OR =1.30, CI: 1.02–1.67, p = 0.037).
Conclusion: Based on Jylhäʼs theoretical framework, sex-stratified analyses revealed that the need for daily assistance showed the strongest association with SRH. Employment was positively associated with SRH in both sexes,with a stronger association observed in women. Household income showed no significant association with SRH in women or in high-income men; however, a significant association was observed in middle-income men. These findings suggest sex differences in income perception and a ceiling effect, where the association between income and SRH may weaken beyond a threshold. In addition, associations with health-related lifestyle behaviors differed by sex, supporting the applicability of Jylhäʼs framework that SRH is shaped by both health conditions and social context. Associations with health-related lifestyle behaviors also differed by sex, highlighting the need for multifaceted support strategies that account for sex differences in improving SRH.
keywords : self-rated health, sex differences, health behaviors, socioeconomic factors, employment statussocioeconomic factors, employment status