影響糖尿病患者之健康相關生活品質(SF-36)因素探討：個人社會經濟地位與鄰里脈絡效應

Abstract

目的：瞭解臺灣地區20-64歲糖尿病患者的個人層次社會經濟地位、縣市層次地區變項和跨層次交互作用對其健康相關生活品質 (SF-36)的影響。 方法：採橫斷式研究設計、次級資料分析。個人層次自變項如人口學特性、健康行為、社會經濟地位變項，與依變項健康相關生活品質(SF-36)取自於國民健康局2005年國民健康訪問暨藥物濫用調查(2005年NHIS)，對象為糖尿病患者。人口學特性包含性別、年齡、婚姻狀況；健康行為包含有無吸菸、飲酒、吃檳榔和運動；個人社會經濟地位包含工作狀況、教育程度、家戶月收入。縣市層次的地區變項有三個，其中之一為地區劣勢因素分數，資料來源為行政院主計處網站的各縣市重要指標統計，利用主成份分析將(1)失業率、(2) 十五歲以上民間人口未受高等教育比率、(3) 低收入戶人口數占該/市人口比率，得到因素分數；另兩個指標為ICE (index of concentration at the extremes)所得指標與ICE教育指標，資料來源為2005 NHIS全數樣本。統計分析採用多層次方法中的階層線性模式分析，包含隨機係數模型、截距預測模型。 結果：刪除遺漏值後，納入本研究之糖尿病患者樣本共有539人，於23個縣市中。初步研究顯示，糖尿病患者SF-36的幾個構面中，分別有1.79% ~ 7.15%的差異來自於縣市層次 (ICC)。經由階層線性模式分析，發現個人社會經濟地位如失業與較低家戶月收入幾乎和SF-36各個構面有相關。而在控制個人層次後，縣市層次的部份，有達統計上顯著水準的地區變項與生理的健康相關生活品質呈現負相關，顯示居住在越劣勢的縣市其生理的健康相關生活品質會越差；但心理的健康相關生活品質則相反，即居住在越劣勢的縣市其心理的健康相關生活品質較好。 結論：不同的個人社會經濟地位變項與縣市層次的地區變項對糖尿病患者健康相關生活品質 (SF-36)有直接影響。

Objective: The purpose of this study is to assess the effects of individual socioeconomic status, county-level variables, and cross-level effects in a sample of diabetes patients aged 20-64 in Taiwan. Methods: Data on individual-level characteristics, health behavior variables, socioeconomic status and health related quality of life (SF-36) were obtained from National Health Interview Survey in 2005 (2005 NHIS). Individual characteristics were included sex, age, and marital status; health behavior variables were including smoking, drinking, betel nut and sports; and individual socioeconomic status was measured by employment, education attainment, and household income per month. One of county-level variables was derived from the Important Indicators of County on the website of Directorate-General of Budget, Accounting and Statistics, Executive Yuan, and was calculated for each county using principal component of the proportion of residents by the three indicators: (1) unemployed, (2) education less than high school, and (3) living below poverty line; the other two county-level variables, index of concentration at the extremes of income and education, were also calculated from 2005 NHIS. Multilevel models including random coefficient model and intercept-as-outcome, were used in the analyses. Results: We excluded missing data on individual information and the remaining study sample included 539 DM patients nested within 23 counties. Individual socioeconomic status such as unemployment and low household income status were related to most of domains of SF36 measures. County-level variables were negatively related to physical-related health but positively related to mental-related health. Conclusion: Different individual-level and county-level variables were related to physical-related health and mental-related health.