| 摘要: | 目的：比較分析思覺失調症患者接受抗精神病藥物的長效針劑型或口服型的醫療照護利用與費用，以及探討其相關因素。
方法：此研究為回溯性配對世代研究(A retrospective matched cohort study design)，資料來源係使用全民健康保險研究資料庫的「2008-2013精神疾病住院病患歸人檔(PSY3)」、「承保資料檔(Registration for beneficiaries)」、「重大傷病證明明細檔(Registry for catastrophic illness patients)」、「醫事人員基本資料檔(Registry for medical personnel )」、「醫事機構基本資料檔(Registry for contracted medical facilities)」及「專科醫師證書主檔(Registry for board-certified specialists)」。研究對象為2008年至2013年間思覺失調症新發個案(ICD-9-CM code = 295.xx)，並使用傾向分數配對法(Propensity score matching; PSM)，依據年齡、性別及共病症指標配對出曝露組(The exposed group;長效針劑型組)及非曝露組(The unexposed group;口服型組)。主要統計方法為兩部模型(Two-part model)及廣義估計方程式(Generalized estimating equations; GEE)以控制群聚效應(The clustering effect)，以能更精準地分析思覺失調症患者接受長效型risperidone針劑治療之醫療照護利用與費用。
結果：雙變項分析結果顯示，醫療機構地點、權屬別、評鑑等級、是否為重大傷病患者及罹病年數在曝露組及非曝露組別間呈現統計上的顯著差異。相較於口服型組患者而言，使用長效針劑型組患者之住院日數顯著減少，但門診、急診及住院次數則顯著增加，導致其總精神疾病醫療費用較高。而廣義估計方程式的分析結果則進一步顯示，相較於醫師年齡大於30歲者而言，醫師年齡小於或等於30歲者，其患者之急診次數、急診費用、住院次數、住院日數及住院費用均較高。此外，女醫師較男醫師而言，其患者之門診次數與費用以及急診次數與費用均較少。在醫療機構特性方面，相較於其它地區之醫療機構而言，位於臺灣北部的醫療機構之患者有較少的住院次數、門診費用及住院費用。相較於其它權屬別之醫療機構而言，公立醫療機構之患者有較多的急診次數與急診費用。而相較於其它評鑑等級之醫療機構，醫學中心之患者有較少之門診次數、急診次數與急診費用，卻有較高的門診費用。最後，在患者特性方面，相較於其它年齡層的患者而言，小於15歲的患者使用較多的門診與住院醫療次數及較長的住院日數。而女性患者則較男性患者而言，其門診次數、急診次數、住院次數、門診費用、急診費用及住院醫療費用均較高。此外，相較於其它投保薪資級距的患者而言，低於兩萬台幣之患者使用較多的住院日數及住院醫療費用。而擁有重大傷病身份之患者則使用較多的門診次數、急診次數、住院次數、住院天數及門診費用、急診費用與住院醫療費用。最後，罹病年數愈長的患者會使用愈多的門診、急診及住院醫療。
結論：本論文發現，相較於口服型組患者而言，使用長效針劑型組患者之住院日數顯著較少，然而由於門診次數及費用較多，導致其總精神科醫療費用較高。本研究的研究發現盼能協助醫藥界及相關主管單位發展出更符合臨床實務與治療效益的思覺失調症患者之治療準則，以能更進一步地提升其照護品質。
Objectives: To describe the current state of prescriptions for long-acting injection of risperidone versus oral medication, compare health service utilization and medical costs between patients with schizophrenia receiving LAI risperidone (the exposed group) and those receiving oral medication (the unexposed group), and evaluate related influencing factors.
Methods: This was a retrospective matched cohort study to evaluate health service utilization and medical costs between patients with schizophrenia who received LAI risperidone and those who took oral risperidone medication. The main data source was the 2008-2013 Psychiatric Inpatient Medical Claim (PIMC) Dataset. Propensity score matching (PSM) was conducted, and propensity scores were calculated using logistic regression analysis of age, gender, and the Charlson comorbidity index (CCI). With respect to data analyses, two-part models were firstly performed. Considering the possibility of clustering effects, generalized estimating equations (GEE) was further conducted.
Results: Analytical results showed that health care institution location, ownership, accreditation level, patient’s catastrophic illness status and years of illness were statistically different between the two study cohorts. In addition, the exposed group had a significant reduction in length-of-stay compared to the unexposed group. However, the exposed group incurred more total psychiatric service costs than the unexposed group thanks to increased utilization of outpatient, emergency, and inpatient services. Generalized estimating equation results showed that characteristics of health care provider and patient had significant impacts on patients'' outpatient, emergency and inpatient service use.
Conclusions: To sum up, the present study found that treatment with LAI risperidone was associated with a reduction of length-of-stay but more total psychiatric treatment costs due to increased utilization of outpatient, emergency, and inpatient services. Findings of the study could provide empirical evidence in helping medical professionals design better clinical guidelines, and thus improve health care quality of patients with schizophrenia. |
