長期照護安排，醫療照護連續性與醫療服務利用

Long-term Care Arrangement, Continuity of Care, and Medical Care Utilization

張文瓊(Wen-Chiung Chang)

指導教授 ： 吳淑瓊

國立臺灣大學/公共衛生學院/健康政策與管理研究所/博士(2014年)

https://doi.org/10.6342/NTU.2014.02376

摘要

研究背景及目的：隨人口快速老化，我國面臨逐漸增加的長期照護及醫療照護需要，而長期照護需要者同時有高度的醫療需求。除了社會人口學特質、健康狀況等因素之外，公共政策制度的設計會形塑出整體長期照護安排的樣貌，進而影響醫療服務的利用。長期照護需要者有多元複雜的照護需要，故照護連續性可能對照護結果及後續醫療服務利用有重要影響。我國目前處於即將開始建立正式長期照護體系之際，應就實證資料加以分析進行瞭解長期照護安排與醫療服務利用之關係，以提供我國長期照護體系制度規劃之參考依據。本研究目的包括(1)瞭解中老年長期照護需要者之醫療服務利用情形；(2)探討長期照護安排與醫療服務利用之關係；(3)分析照護連續性與醫療服務利用之關係；(4)瞭解長期照護安排及照護連續性共同對醫療服務利用之影響；以及(5)檢視在各類長期照護安排下，醫療服務利用的影響因素之異同。
研究方法：本研究資料來源為行政院衛生署(今衛生福利部)委託國立台灣大學執行之「全國長期照護需要評估第二年計畫」問卷調查資料，並串連2002-2003年全民健康保險申報資料庫。研究樣本為(1)日常生活活動中有任何1項功能障礙，或(2)工具性日常生活活動中有5項以上因健康相關因素具功能障礙，或(3)SPMSQ量表答錯6題以上者，或(4)因個案本人罹患失智症而無法自答SPMSQ量表者。共13,110名樣本。依變項為醫療服務利用情形，包括一年內是否使用門診、年度門診次數、年度門診總金額、一年內是否使用急診、年度急診次數、年度急診總金額、一年內是否有住院、年度住院次數、年度住院總日數、年度住院總金額、一年內是否有可避免住院、年度可避免住院次數、年度可避免住院總日數、年度可避免住院總金額、以及年度醫療費用總金額。自變項為長期照護安排以及照護連續性。長期照護安排分為(1)家庭照顧者或居家/社區式服務，(2)無照顧支持者，(3)雇用全職在宅看護，(4)住機構等四類。照護連續性分數以Bice & Boxerman提出之計算方法測量之。控制變項包括(1)前傾因素：性別、年齡、教育程度；(2)使能因素：是否免部分負擔、居住地區都市化程度、婚姻狀況。(3)需要因素：疾病複雜度、憂鬱症狀、認知功能障礙、失能程度，未滿足需求。以描述性統計量、χ2 test、ANOVA、t-test、multiple logistic regression、以及generalized linear models (GLM)等方法進行分析。
研究結果：樣本之長期照護安排為家庭照顧者或居家/社區式服務者佔60.7%，無照顧支持者有14.3%，雇用全職在宅看護者百分比為16.0%，住機構者則有8.9%。各項醫療服務利用之年度利用率及年度使用次數平均值分別為門診(94.1%，26.1次/年)、急診(32.6%，0.7次/年)、住院(36.3%，0.8次/年)，及可避免住院(17.5%，0.3次/年)；年度住院總日數及可避免住院總日數分別為12.8日及3.6日。各類長期照護安排在各項醫療服務利用達統計上之顯著差異。控制前傾、使能及需要因素等基本特質後，長期照護安排及照護連續性對各項醫療服務利用之影響達統計上之顯著意義。相較於家庭照顧者或居家/社區式服務組，無照顧支持者之住院服務利用及各項醫療費用較低，雇用全職在宅看護者之各項門診服務利用、急診機會(OR=1.17, p<0.01)、住院總金額(RR=1.11, p<0.05)、可避免住院次數(RR=1.15, p<0.05)、可避免住院總日數(RR=1.31, p<0.01)、可避免住院總金額(RR=1.27, p<0.01)，及醫療費用總金額(RR=1.18, p<0.001)都較高；住機構者則是各項門診服務利用、住院機會(OR=1.40, p<0.001)、次數(RR=1.52, p<0.001)、總日數(RR=1.74, p<0.001)、總金額(RR=1.18, p<0.01)，及可避免住院風險(OR=1.73, p<0.001)、可避免住院次數(RR=1.71, p<0.001)、可避免住院總日數(RR=1.85, p<0.001)，及可避免住院總金額(RR=1.35, p<0.01)，及醫療費用總金額(RR=1.25, p<0.001)都較高，而急診機會(OR=0.73, p<0.001)、急診次數(RR=0.63, p<0.001)、急診總金額(RR=0.67, p<0.001)較低。照護連續性的影響為照護連續性愈高，門診、急診、住院及可避免住院的使用機會和次數愈低。共同納入照護連續性之後，長期照護安排對醫療服務利用之影響仍存在，但其影響程度對雇用看護者而言在住院及可避免住院有較明顯的降低。依長期照護安排分層分析後發現，照護連續性對門診服務利用的影響在雇用全職在宅看護者最為明顯，對急診、住院、可避免住院，及醫療費用總金額的影響則是在住機構者最明顯。疾病複雜度提高醫療服務利用的作用對無照顧支持者的影響較強。失能程度對住社區者而言可能減少門診使用而增加住院服務利用，對住機構者則相反，會增加門診使用而在住院服務利用較低。
結論：長期照護安排確實對醫療服務利用有影響，整體而言雇用全職在宅看護或住機構者可能有較高的醫療服務利用。對雇用全職在宅看護者而言，照護連續性在其與醫療服務利用間之關係有部分的中介效果。各類不同的長期照護安排下，照護連續性及其他影響因素對醫療服務利用產生的作用會有所不同。本研究結果指出長期照護與醫療體系之間的聯結，受到長期照護政策之引導而形成的長期照護安排，會在醫療體系對醫療服務利用量產生顯著的影響。應加強發展多元化之居家/社區式長期照護服務，支持家庭照顧資源，以改善目前我國長期照護需要者過度依賴看護，或當家庭資源無法支持時只能接受機構式照護之選項的限制，才能提升照護品質並且有效率的提供健康照護。

關鍵字

長期照護安排 ； 醫療服務利用 ； 照護連續性 ； 中老年人

並列摘要

Background and Objectives: With the trend of population aging, the society faces the challenge of increasing needs on long-term care and medical care. People with long-term care needs also have high demand on medical care. In addition to the factors of sociodemographic characteristics and health status, long-term care arrangement, which is shaped by policy design, could also exert influences on medical care utilization. Long-term care users usually have multiple and complex health care needs; therefore, the continuity of care could influence their health-care outcomes and subsequent medical care uses. At the present time, Taiwan is developing the formal long-term care system. It is needed to analyze experimental data to clarify the relationship between long-term care arrangement and medical care utilization, and provide this base information for system planning. The aims of this study include: 1) describing the medical care utilization of the middle-aged and older adults with long-term care needs, 2) exploring the relationship between long-term care arrangement and medical care utilization, 3) analyzing the association between continuity of care and medical care utilization, 4) examining the joint effects of long-term care arrangement and continuity of care on medical care utilization, and 5) comparing the effects of realted factors of medical care utilization among different long-term care arrangements.
Methods: The data analyzed in this study was from the 2002 interview data of the Assessment of National Long-Term Care Need in Taiwan (ANLTCNT) and 2002-2003 claims data of the National Health Insurance (NHI). To estimate the long-term care needs in Taiwan, the ANLTCNT, a two-stage nationwide survey, was first launched in 2001. Subjects who met one of the following four criteria were defined as with long-term care needs and entered the second-stage survey: 1) one or more ADLs disability; 2) five or more IADLs disabilities; 3) cognitive impairment as measured with the Short Portable Mental Status Questionnaire (SPMSQ); or 4) unable to response SPMSQ due to dementia. A total of 13,110 individuals were analyzed in this study. Dependent variables were medical care utilization of outpatient visit, emergency department (ED) visit, hospitalization, and potentially avoidable hospitalization (PAH). For each type of medical care, whether used, number of episodes, length of stay (LOS) in one year (for hospitalization and PAH), and expenditures in one year were analyzed, In addition, the total medical expenditure in one year was also included in analyses. Independent variables were long-term care arrangement and continuity of care. Long-term care arrangement was categorized as 1) family caregiver or home- and community based services (HCBS), 2) without caregiver, 3) full-time, in-home care assistant, and 4) long-term care institution. Continuity of care index was computed by applying the method proposed by Bice & Boxerman. Control variables included 1) predisposing factors: sex, age, and educational level, 2) enabling factors: copayment exemption, level of urbanization, and marital status, and 3) need factors: morbidity burden, depressive symptom, cognitive impairment, disability level, and unmet need. Descriptive statistics, χ2 test、ANOVA、t-test、multiple logistic regression, and generalized linear models (GLM) were used in statistical analyses.
Results: Of the study samples, 60.7% were cared by family caregiver or HCBS, 14.3% had no caregiver, 16.0% hired care assistant, and 8.9% resided in institutions. The rates and numbers of using were 94.1% and 26.1 visits/year(yr) for outpatient visit, 32.6% and 0.7 visits/yr for ED visit, 36.3% and 0.8 episodes/yr for hospitalization, and 17.5% and 0.3 episodes/yr for PAH. Comparing with the family caregiver/ HCBS group, the individuals without caregiver had less hospitalization utilization and medical expenditures in all types of medical care. The care assistant group had higher utilization in outpatient and ED visits, hospitalization, and PAH, and total medical expenditure. The institution residents used more in outpatient visit, hospitalization, PAH, and total medical expenditure, but less in ED visit. Higher continuity of care was associated with lower risk and numbers of using outpatient and ED visits, hospitalization, and PAH. After controlling for continuity of care, the effects of long-term care arrangement on medical care utilization were still significant, but the magnitude of influence weakened, in particular for hospitalization and PAH. In the analyses stratified by long-term care arrangements, the effect of continuity of care on outpatient visit utilization was most significant in the care assistant group; however, the effects on ED visit, hospitalization, PAH, and total medical expenditure were more significant in the institutionalized group. The effects of morbidity burden on increased medical utilization were more relevant for those without caregiver. Higher disability was associated with lower outpatient visit and higher hospitalization utilization for the community-dwellers, but for the institution residents, the direction of effects was reverse.
Conclusions: Long-term care arrangement could affect medical care utilization, and this association is partly mediated by continuity of care. Under different long-term care arrangements, the effects of the related factors of medical care utilization might be different. The findings of this study indicate the linkage between long-term care and medical care systems. Long-term care arrangement, which is directed by long-term care policy, could greatly influence the amount of medical care utilization in medical system. To improve the quality and efficiency of health care, policy makers shoud reinforce developing diverse HCBSs to support family caregiving, and modify the trend of over-dependency on foreign care assistants and institutionalization.

並列關鍵字

long-term care arrangement ； medical care utilization ； continuity of care ； middle-aged and older population

參考文獻

Wu SC, Chuang KY, Dai YT, Lee SD, Chang HJ, Day GI, Chan HM. Expanding the role of health service providers and the health care profession in the delivery of long-term care. Taiwan Journal of Public Health. 2005; 24(3): 189-91.

戴桂英：中風出院病人後續醫療照護之研究。臺灣大學衛生政策與管理研究所博士論文，2007。

紀玫如：中老年人多重慢性病集群型態與醫療服務使用之相關研究。臺灣大學衛生政策與管理研究所博士論文，2008。

行政院衛生署。長期照護十年計畫成果。2012；台北：行政院衛生署。

楊惠真、鄭讚源、林四海、方志琳、丁增輝：不同照顧場所之中風失能老人復健照護利用及其長期照護政策意涵。健康管理學刊, 2011; 9: 1-16。