目標：探討在全民健保重大傷病制度下，重大傷病患者的家戶所得與醫療利用及存活狀況之相關。方法：以2003-2006年「台灣地區家庭收支調查檔」串聯全民健保「重大傷病證明檔」，找出僅持有一張重大傷病卡之末期腎病定期血液透析病人或癌症病人為本研究之研究對象，另串聯2002-2007年全民健保申報檔以分別取得社經人口學變項與醫療利用資料。另再串聯2003-2009年「全國死因檔」取得研究對象觀察四年之存活狀況。運用負二項分配、複迴歸及Cox proportional hazards model分析社經變項對醫療利用、存活狀況之相關。結果：醫療利用方面，末期腎病病人最高所得者之住院日數較最低者多8.897日(p<0.05)，但門、住診次數與透析利用則與所得高低無關。癌症病人其最高所得者相對於所得最低者，門診利用IRR=1.18 (p<0.05)，住院利用IRR=2.11 (p<0.001)，以及住院日數高11.36天(p<0.001)。存活分析結果發現，末期腎病病人僅男性(HR=1.82, p<0.05)及共病數與死亡率顯著相關。癌症病人中男性(HR=1.66, P<0.05)及最高所得者死亡風險較高(HR=1.60, P<0.01)，教育程度最高者死亡風險較低(HR=0.52, p<0.05)。結論：家戶所得與醫療利用之相關性因疾病而異，癌症病人之家戶所得與醫療利用呈顯著正相關，另外所得最高組的癌症病人死亡率較高。但末期腎病病人之家戶所得與醫療利用及存活則無顯著相關。建議未來可針對不同重大傷病類別進行更深入之研究。(台灣衛誌 2013；32(4)：331-345)

Objectives: To investigate the association of household income and healthcare utilization as well as survival status of patients with catastrophic illnesses under the NHI system in Taiwan. Methods: Data from the “Survey of Family Income and Expenditure” (2003 to 2006) and the “Registry of patients with catastrophic illness” were first linked to identify subjects. Patients with cancer or end stage renal disease (ESRD) holding only one NHI catastrophic illness card were included in this study. Information related to healthcare utilization was obtained from NHI claims data (2002 to 2007) and survival data was obtained from the “National Registry of Deaths” (2003 to 2009). Negative binominal regression, multiple regression, and the Cox proportional hazard model were used to analyze the relationships among healthcare utilization, survival, and socioeconomic variables. Results: Regarding the healthcare utilization, the average length of stay (ALOS) of ESRD patients in the highest income bracket was 8.987 days longer than that of patients in the lowest bracket (p<0.05). Hemodialysis usage, the number of outpatient visits and hospitalizations presented no correlation with household income. Among cancer patients in the highest income bracket, the IRRs of outpatient visits and hospitalization were 1.18 (p<0.05) and 2.11 (p<0.001), respectively. The ALOS of those in the highest income bracket was 11.36 days longer than that of patients in the lowest income bracket (p<0.001). With respect to survival status, male ESRD patients had a higher mortality than females (HR=1.82, p<0.05). Among cancer patients, being males (HR=1.66, p<0.05) and in the highest income bracket had higher mortality (HR=1.6, p<0.05), those with the highest education level had lower mortality (HR=0.52, p<0.05). Conclusions: The relationship between household income and healthcare utilization varied according to type of diseases. Income level was positively associated with healthcare utilization among cancer patients while patients in the highest income bracket had higher mortality. But the association was not significant among ESRD patients. Future researchers can conduct further analyses on other catastrophic illnesses. (Taiwan J Public Health. 2013;32(4):331-345)