類風濕性關節炎患者照護連續性與醫療利用之相關性探討

目的：本研究旨在瞭解類風濕性關節炎患者照護連續性與醫療利用之情形，以及此類病患的照護連續性與其醫療利用兩者之間是否具有關連性。方法：本研究使用健保資料庫2010年的承保抽樣歸人檔(LHID2010)，選取2010年首次診斷為類風濕性關節炎的患者(ICD-9-CM code: 714.0)，且在研究期間(2010年-2012年)與類風濕性關節炎相關的看診次數三次(含)以上者。考量門診資料的變異性不宜太大，因此門診資料排除中醫與牙醫，以及其它特殊案件。由於本研究為縱貫性研究(Longitudinal study)，考慮到同一位醫師或醫院之病患有群聚效應(Clustering effect)的可能性，因此使用廣義估計方程式(Generalized estimating equation, GEE)來控制該群聚效應，以能更精準地分析照護連續性與醫療利用之關連性。結果：在606名研究對象中，女性佔439人(72.44%)，男性則佔167人(27.56%)。在2010年時，個案的平均年齡為56.66歲(標準差為14.426)，照護連續性平均值為0.26(標準差為0.199)，共病指標、急診次數、住院次數、住院日數、及醫療費用的平均值則分別為0.61(標準差為1.080)、0.69 (標準差為3.102)、0.37(標準差為1.027)、3.33(標準差為15.458)、及60,019.99(標準差為99,567.100)。廣義估計方程式的結果顯示，在控制其它變項後(亦即病患特性、醫療院所特性及醫師特性)，照護連續性越高的個案其急診次數、住院次數及醫療費用均較少，呈現負相關的關係。在病患特性方面，女性患者較男性患者而言，其急診次數、住院次數及醫療費用均較少。相較於無重大傷病卡的個案而言，有重大傷病卡個案之急診次數、住院次數、住院日數及醫療費用均較多。此外，合併症愈多者其急診次數、住院次數、住院日數及醫療費用也較多。在主要照護醫療院所特性方面，主要照護醫療院所之層級為區域醫院的就診患者較基層醫療院所的就診患者而言，其急診次數、住院次數、住院日數及醫療費用均較少。結論：本研究發現，類風濕性關節患者的照護連續性與醫療利用具有關連性。此外，病患特性中的性別、有無重大傷病卡、共病指標與醫療利用均具有關連性。然而，主要照護醫療院所特性、主要照護醫師特性與醫療利用則無顯著的關連性。根據研究結果，本研究建議衛生主管機關應提供適度之支付誘因，以鼓勵醫師能針對類風濕性關節炎患者提供整合性照護，使病患能得到較佳之治療成效。

關鍵字

類風濕性關節炎；醫療利用；照護連續性；縱貫性研究；群聚效應；廣義估計方程式

並列摘要

Objectives: To examine the current status of continuity of care (COC) and healthcare utilization for rheumatoid arthritis patients, and evaluate the associations between COC and healthcare utilization for those patients. Methods: The study made use of the LHID2010 dataset. Subjects were patients with rheumatoid arthritis (ICD9-CM code 714.0) who made three or more visits to physicians in 2010-2012. Considering the variability of outpatient data not being too large, therefore, the retrieved data excluded Chinese medicine usage and dentist visits, as well as other special cases. This research was a longitudinal study. Hence, considering the possibility of the clustering effect, associations between COC and healthcare utilization were analyzed by using the generalized estimating equation (GEE). Results: In 606 subjects of the study, 439 were females (72.44%), while 167 were males (27.56%). In 2010, the mean age of subjects was 56.66 years (SD = 14.426), while the continuity of care averaged 0.26 (SD = 0.199). In addition, the means and standard deviations of comorbidity index, emergency department visits, hospitalizations, hospital days, medical costs were 0.61 (SD = 1.080), 35.38 (SD = 23.135), 0.69 (SD = 3.102), 0.7 (SD = 1.027), 3.33 (SD = 15.458), and 60,019.99 (SD = 99,567.100), respectively. Generalized estimating equation results showed that after controlling for other variables (i.e., characteristics of patients, medical care organizations, and physicians), subjects had higher COC scores would incur less emergency room visits, hospitalizations and medical expenses, compared to their counterparts. In terms of characteristics of patients, results showed that female patients had fewer whose emergency visits hospitalization and medical expenses than male patients. Moreover, patients with catastrophic illness cards utilized more emergency department visits, hospitalizations, length of stays, and medical expenses than their counterparts. Additionally, patients who had higher comorbidity index would incur more emergency department visits, hospitalizations, length of stay, and medical expenses than their counterparts. In terms of characteristics of medical care organizations, results showed that healthcare utilizations (with respect to emergency department visits, length of stays, and medical expenses) at regional hospitals were higher, compared to other types of hospitals. Conclusions: The study had demonstrated that COC and healthcare utilization were significantly associated for patients with rheumatoid arthritis. In addition, sex, if having a catastrophic illness card, and Charlson comorbidity index were related to healthcare utilizations. However, characteristics of medical care organizations and physician were not significantly associated with healthcare utilizations. Based on the findings, this study suggested that the government should provide appropriate payment incentives to encourage physicians to deliver integrated care for patients with rheumatoid arthritis as to produce better treatment efficacy.

並列關鍵字

Rheumatoid arthritis ； Healthcare utilization ； Continuity of care ； Longitudinal study ； Clustering effect ； Generalized estimating equation

參考文獻

李丞華、周穎政、陳龍生、張鴻仁(2004)。全民健保中醫門診利用率及其影響因素。臺灣公共衛生雜誌，23(2)，100-107。

Charlson, M. E., Pompei, P., Ales, K. L., & MacKenzie, C. R. (1987). A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of Chronic Diseases, 40(5), 373-383.

中文文獻