疾病管理對糖尿病患照護品質及中長期醫療資源利用之成效

研究目的：全民健康保險糖尿病醫療給付改善方案(以下稱疾病管理)自2001年實施迄今已進入第10年，有關照護計畫之照護品質及中長期醫療資源利用成效值得分析。本研究目的一、探討糖尿病疾病管理組人口學特質、疾病嚴重程度及就醫院所特性。二、探討糖尿病疾病管理對糖尿病照護品質之影響和預測因子。三、探討糖尿病疾病管理組和非疾病管理組特質及中長期醫療資源利用。四、探討糖尿病疾病管理組和非疾病管理組中長期醫療資源利用影響因子。五、探討糖尿病疾病管理組和非疾病管理組有無急診、住院及其影響因子。六、探討糖尿病疾病管理組和非疾病管理組之存活狀況及其影響因子。研究方法：以2006年於高屏地區健保局特約院所參加糖尿病疾病管理新收案之成人第二型糖尿病患者，且於2006~2008年持續參加者，為疾病管理組研究樣本2,181人；另以2006年1~3月醫療費用檔(篩選ICD-9-CM主診斷碼250.X0、250.X2且有使用糖尿病藥物治療者)且未參加疾病管理者資料，依疾病管理組個案之年齡、性別，一對一匹配產生非疾病管理組研究樣本。分析疾病管理組糖尿病照護品質(臨床指標)、研究對象2003~2009年醫療資源利用、2007~2009年存活狀況及探討各項影響因子。研究結果：一、在臨床照護品質方面：疾病管理組HbA1C收案後1年較收案時減少0.86%，收案後2年較收案時減少0.93%。LDL-C收案後1年較收案時減少11.2 mg/dl，收案後2年較收案時減少15.67 mg/dl。AC收案後1年較收案時減少15.85 mg/dl，收案後2年較收案時減少18.75 mg/dl (P＜0.001)。顯示這些生化指標有因持續照護而持續改善之現象。另照護時間、年齡、性別及罹病時間為七項臨床指標之影響因子。二、在中長期醫療資源利用方面：疾病管理介入後第一、二、三年(2007、2008、2009)，在門診次數、急診次數、急診費用、住院次數、住院天數、住院費用及總醫療費用，非疾病管理組均顯著高於疾病管理組；另門診費用，非疾病管理組均顯著低於疾病管理組。疾病嚴重度、慢性併發症為2006、2007、2008、2009年各年度住院費用、總醫療費用之顯著影響因子。而有無參加疾病管理為2007、2008、2009年各年度住院費用顯著影響因子(非疾病管理組高於疾病管理組依序為9,496元、10,861元、5,177元)。三、照護品質不好指標方面(住院、急診人數)：2003年~2009年各年度非疾病管理組住院人數占率均高於疾病管理組；惟2006年未達統計上差異；另2003年~2009年各年度非疾病管理組有急診之人數占率均高於疾病管理組，惟2004年、2006年、2009年未達統計上差異。疾病嚴重度、慢性併發症疾病嚴重度、慢性併發症(主要併發症為心血管疾病併發、神經病變、腎臟病變併發症及其他併發症)為2007、2008、2009年各年度住院、急診之顯著影響因子。有無參加疾病管理為2007年急診(OR=1.25)及2008年住院(OR=1.33)顯著影響因子。四、2007~2009年各年每年死亡人數，非疾病管理組顯著高於疾病管理組。另Kaplan-Meier 存活分析結果發現，疾病管理組存活狀況優於非疾病管理組。有無參加疾病管理、年齡及疾病嚴重度為存活預測因子。2007、2008、2009年各年度非疾病管理組死亡風險顯著高於疾病管理組 (HR =21.72、6.57、3.44)。結論與建議：疾病管理組持續接受照護，在HbA1C、LDL-C、AC、TG及DBP皆有持續改善，且因獲得適當及完整的門診照護，可減少可避免的住院及急診、、降低死亡風險及減少整體醫療資源之耗用，因此此糖尿病照護計畫值得繼續推動。惟目前仍有近七成以上的糖尿病病患未參與該計畫，提升糖尿病照護率，應是衛生主管機關應重視的課題。

關鍵字

糖尿病；糖尿病醫療給付改善方案；疾病管理；品質；醫療利用

並列摘要

Research Objectives: It has been ten years ever since the implementation of Diabetes Medical Benefit Improvement Project (hereinafter referred to as “disease management”) by National Health Insurance in 2001. The care quality of the related care plan as well as the effect of medium- and long-term medical resource use are worthy of making further analysis. The study has six objectives: 1. Explore the demographic features of disease management group for diabetes, the overall seriousness of the disease, and the characteristics of hospitals where diabetes patients seek medical treatment. 2. Explore the effects of disease management for diabetes on the quality of diabetes care, and the predictive factors. 3. Explore the characteristics of the disease management group and non-disease management group for diabetes as well as the medium- and long-term medical resource use. 4. Explore the impact factors of medium- and long-term medical resource use of the disease management group and non-disease management group for diabetes. 5. Explore whether there is any case of emergency treatment or hospitalization in the disease management group and non-disease management group for diabetes, as well as the impact factors. 6. Explore the survival situation in the disease management group and non-disease management group for diabetes, as well as the impact factors. Research Design: The research samples of the disease management group were 2,181 adult patients of diabetes type 2 having joined the new diabetes disease management cases of the designated hospitals of Kaohsiung and Pingtung Area Health Insurance Bureau in 2006, and these patients continued joining the Project in 2006~2008. Besides, from the file of medical expenses in January ~ March, 2006 (screening the patients of Diagnosis Codes 250.X0 and 250.X2 of ICD-9-CM who took diabetes medication), the patients without joining the disease management were one-to-one paired according to the age and gender of the cases of disease management group, thus producing the research samples of the non-disease management group. The study analyzes the quality of diabetes care of the disease management group (clinical indicator), the medical resource use of the research targets in 2003 ~ 2009 and the survival situation of diabetes patients in 2007 ~ 2009, and also explores different impact factors. Research Results: 1. Regarding the quality of clinical care: When compared with the time that the disease management group accepted the cases, the HbA1C was decreased by 0.86% one year after case acceptance, and decreased by 0.93% two years after case acceptance; the LDL-C was decreased by 11.2 mg/dl one year after case acceptance, and decreased by 15.67 mg/dl two years after case acceptance; and the AC was decreased by 15.85 mg/dl one year after case acceptance, and decreased by18.75 mg/dl (P＜0.001) two years after case acceptance. The results show that all these biochemical indicators improve continuously because of the continuous care. Besides, the care time, age, gender and morbidity period are the impact factors of seven clinical indicators. 2. Regarding medium- and long-term medical resource use: In the first, second and third years after disease management was involved (in 2007, 2008 and 2009), for the number of times of outpatient service, number of times of emergency treatment, expense of emergency treatment, number of times of hospitalization, number of hospitalization days, hospitalization expense and total medical expenses, the non-disease management group has significantly higher figures than the disease management group; and as to the expense of outpatient service, the non-disease management group has significantly lower figures than the disease management group. The seriousness of disease and chronic complications are the significant impact factors of hospitalization expense and total medical expenses in each year from 2006 to 2009. And whether disease management is joined is the significant impact factor of hospitalization expense in each year from 2007 to 2009 (the hospitalization expense of the non-disease management group is higher than that of the disease management group by $9,496, $10,861 and $5,177 in 2007, 2008 and 2009 respectively). 3. Regarding the poor care quality indicator (the number of hospitalized patients and the number of emergency patients): In each year from 2003 to 2009, the percentage of the number of hospitalized patients of the non-disease management group is higher than that of the disease management group, but there is no statistical difference in 2006. Besides, in each year from 2003 to 2009, the percentage of the number of emergency patients of the non-disease management group is higher than that of the disease management group, but there is no statistical difference in 2004, 2006 and 2009. The seriousness of disease, seriousness of chronic complications, and chronic complications (the main complications caused are cardiovascular complications, neuropathy, nephropathy complications and other complications) are the significant impact factors of hospitalization and emergency treatment in 2007, 2008 and 2009. Whether disease management is joined is the significant impact factor of emergency treatment (OR=1.25) in 2007 and that of hospitalization (OR=1.33) in 2008. 4. Regarding the number of deaths in each year from 2007 to 2009, the non-disease management group is significantly higher than that of the disease management group. Besides, as found in the result of Kaplan-Meier survival analysis, the survival situation of the disease management group is better than that of the non-disease management group. Whether disease management is joined, age and the seriousness of disease are the predictive factors of patients’ survival. In each year from 2007 to 2009, the risk of death of the non-disease management group is significantly higher than that of the disease management group (HR =21.72、6.57、3.44). Conclusions and suggestions: After the patients of the disease management group received care continuously, the patients’ HbA1C, LDL-C, AC, TG and DBP were all improved continuously. In addition, because of the appropriate and complete care for outpatients, hospitalization and emergency treatment could be decreased, the risk of death could be reduced, and the consumption of overall medical resources could be decreased. Therefore, such a caring project for diabetes is worthy of continuous promotion. Nevertheless, up to now, there is nearly 70% or more diabetes patients having not joined the Program yet. The enhancement of diabetes care rate should be an issue that the competent authorities of public health have to pay concern for.