地區醫院家庭醫學科之糖尿病照護成效橫斷性結果指標與連續性結果指標之比較研究

The comparison of cross-sectional indicators with sequential indicators to evaluate the intermediate outcomes of type 2 diabetic patients cared for by the family physician in a community hospital

黃信揚(Hsin-Yang Huang)

指導教授 ： 賴美淑

國立臺灣大學/公共衛生學院/預防醫學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.02860

摘要

背景:
糖尿病是國人十大死因之一，有賴不同專業醫療團隊的照護、衞教與定期併發症篩檢。過去研究發現結構式照護會提升糖尿病照護過程與成效，國內少有研究是以地區醫院家庭醫學科的糖尿病結構式照護為標的，且成效皆以橫斷性指標評估，過去也沒有研究用連續性指標評估糖尿病結構式照護。

目的:
利用健保局「糖尿病醫療給付改善方案」在台東某地區醫院實施期間，由家庭醫學專科醫師負責下，遵從糖尿病學會治療指引，以橫斷性指標與連續性指標比較有無接受結構式照護模式的第2型糖尿病患照護過程、成效與病患遵醫囑行為。

方法:
研究期間是民國96年10月1日至97年12月31日，以回溯性世代追蹤方式，以家庭醫學科固定就診的第2型糖尿病患為研究對象，根據病人是否有填寫同意書加入結構式照護，分成結構式照護組與一般照護組。結構式照護過程上配合不同醫療專業團隊介入，導入臨床資訊系統與醫療服務流程重新設計等措施。

結果:
在家庭醫學科固定就醫的第2型糖尿病患，結構式照護組有120位病人，一般照護組有99位。在性別、族別、年齡、教育程度、糖尿病罹病年數與醫師的藥物處方分布並不因病人有無接受結構式照護而差別，但接受結構式照護模式的病人會有較多次慢性病處方的釋出。
結構式照護模式介入顯著提升散瞳眼底檢查比例達90％、足部檢查比例達99.2％、HbA1C檢查≧2次的比例達100％ ，曾接種肺炎鏈球菌疫苗的比例達43.3％，相較於一般照護組有較好的照護過程盡責度。
未調整其他變項下，結構式照護模式介入的橫斷性指標：HbA1C＜7％的比例是32.5％，LDL＜100 mg/dl的比例是53.3％，SBP＜130 mmHg的比例是50％，只有LDL＜100 mg/dl的比例結構式照護組優於接受一般醫療照護組(P=0.002)；連續性指標方面：血糖控制良好或改善的比例是60％，顯著優於接受一般醫療照護的病人（P=0.01），收縮壓控制良好或改善的比例連續性指標是80.0％，總膽固醇控制良好或改善的比例是80.8％，上述兩項指標與一般照護組比較無統計上差異。
調整其他共變項，糖尿病照護成效的橫斷性指標LDL＜100 mg/dl的比例，結構式照護組相較於一般照護組的勝算比是2.58 (95％CI 1.43-4.65）；在SBP＜130 mmHg與HbA1C＜7％的勝算比則兩組無統計上差異；在血糖控制良好或改善的連續性指標結構式照護組相較於一般照護組的勝算比是 3.35 (95％CI 1.61-6.94）；在總膽固醇與收縮壓的連續性指標勝算比兩組 無統計上差異。
結構式照護組的平均MPR是0.91±0.14，一般照護組MPR是0.83 ±0.18 (P=0.001)；較佳遵醫囑性（MPR≧0.8）兩組的比例分別是83.3％與65.7％，兩組在統計上達顯著差異(P=0.003)。調整其他變項，較佳遵醫囑性的比例結構式照護組相較於一般照護組的勝算比是2.27（95％CI 1.15-4.46）。

結論:
結構式照護介入會有較好的糖尿病照護盡責度。介入的效果主要在血糖的控制，特別是讓個人血糖持續維持在理想範圍內或是有血糖下降進步的成效，以糖化血色素＜7％的橫斷性指標是無法呈現其成效，用連續性指標測量才可顯現結構式照護的價值；其次是低密度膽固醇＜100 mg/dl的病人比例較高，但收縮壓的照護成效不論是用何種指標測量皆不明顯，而且接受結構式照護模式的病人糖尿病用藥的遵醫囑行為較佳。

關鍵字

糖尿病 ； 連續性指標 ； 橫斷性指標 ； 結構式照護 ； 遵醫囑行為 ； 盡責度

並列摘要

Background:
Diabetes mellitus (DM) is one of the top ten causes of death in Taiwan. The medical care for DM patients is getting more and more depending on the cooperation of multidisciplinary teams, frequent health educations, and regular follow-up for complications. Few studies have been focused on the effectiveness of DM intervention programs executed by family physicians. In addition, studies in the literature evaluated the effectiveness of DM intervention programs using cross-sectional indicators. None of them evaluated the effectiveness of DM intervention programs using sequential indicators.

Objectives:
Cross-sectional indicators and sequential indicators were used to evaluate the effectiveness of structured care (SC) for type 2 DM patients in a community hospital in Taitung by family physicians as compared to those who received usual care (UC).

Methods:
This is a retrospective cohort study. Patients were recruited between October 1, 2007 and January 31, 2008. After patients’ consents were provided, the patients who were regularly followed up for three months or longer in the outpatient clinics of family medicine before recruitment were assigned to the SC group. Others were assigned to the UC group. All patients were followed up for one year after they were assigned to the SC group or the UC group. Bivariate analysis was performed to explore the association between an independent variable and the dependent variable using Chi-square test (or Fisher’s exact test) for dichotomous independent variables or Student’s t test for continuous independent variables. Multivariate logistic regression analysis was also performed to explore the associations between the independent variables of interest and the dependent variables.

Results:
One hundred and twenty patients were assigned to the SC group and 99 patients were assigned to the UC group.
The accountability (process indicator of diabetic care) in SC group is significantly better compared with UC group: more patients (99.2%) in the SC group had annual foot exam than UC group（24.2%）(p-value ＜0.001); more patients (90%) in the SC group had annual dilated eye exam than UC group（55.6%）(p-value ＜0.001); more patients (43.3%) in the SC group ever received pneumococcal vaccination than UC group（21.2%）(p-value＝0.001)；more patients (100%) in the SC group had biannual check up of HbA1C than UC group（96%）(p-value＝0.04) .
Three cross-sectional indicators were used to compare the two groups before controlling for other confounding variables: more patients (53.3%) in the SC group reach the goal of LDL < 100mg/dL than the UC group (32.3%). (p-value = 0.002); more patients (32.5%) in the SC group reach the goal of HbA1C < 7% than the UC group (21.2%), though the difference is of borderline significance (p-value = 0.062); sixty (50%) participants in the SC group and 40 (40.4%) participants in the UC group reach the goal of SBP < 130 mmHg (p-value = 0.156). As demonstrated by sequential indicators defined as improved or being controlled: 72 (60%) participants in the SC group showed good sequential indicator of HbA1C, while only 42 (42.4%) in the UC group showed good sequential indicator of HbA1C (p-value = 0.010); no significant differences were found between the SC and UC group using sequential indicator of serum total cholesterol (p-value = 1.000) and SBP (p-value = 0.363).
After controlling for other confounding variables using multivariate logistic regression analysis, the participants in the SC group were more likely to reach the goal of LDL < 100mg/dL than the participants in the UC group (O.R. = 2.58, 95% C.I. = 1.43 – 4.65, p-value = 0.002); no significant differences were shown between the two groups as indicated by cross-sectional indicator of SBP (O.R. = 1.61, 95% C.I. = 0.91 – 2.83, p-value = 0.101) and HbA1C (O.R. = 1.57, 95% C.I. = 0.75 – 3.29, p-value = 0.235). Sequential indicator of HbA1C, serum total cholesterol (T-CHO), and SBP were used to evaluate the effectiveness of structured care after controlling for other confounding variables: significantly more participants in the SC group resulted in good sequential indicator of HbA1C than those in the UC group (O.R. = 3.35, 95% C.I. = 1.61 – 6.94, p-value = 0.001); no significant differences between the SC group and the UC group were found using the sequential indicator of T-CHO (O.R. = 1.41, 95% C.I. = 0.72 – 2.74, p-value = 0.318) and sequential indicator of SBP (O.R. = 0.93, 95% C.I. = 0.47 – 1.84, p-value = 0.839).
Higher proportion of participants（83.3%）in the SC group had better adherence, defined as medication possession rate (MPR) greater than 0.8, than those in the UC group（65.7%） (p-value = 0.003). SC group was significantly associated with good MPR as compared to the UC group after controlling for other confounding variables (O.R. = 2.27, 95% C.I. = 1.15 – 4.46, p-value = 0.017).

Conclusion:
The structured care significantly has better accountability of process of diabetic care and improves the intermediate outcome of type 2 DM patients as indicated by sequential indicator of HbA1C, though the effect did not appear if cross-sectional indicator of HbA1C is used to evaluate the effectiveness of the program. Furthermore, MPR demonstrate that the structured care is significantly associated with better adherence.

並列關鍵字

diabetes mellitus ； sequential indicator ； cross-sectional indicator ； structured car ； adherence ； accountability

參考文獻

3. 翁瑞亨;徐瑞祥;謝玉娟. 臺灣地區糖尿病共同照護現況. 臺灣醫學. 91 ;6(4): 569-73.

18. 劉見祥;曲同光;陳玉敏. 糖尿病共同照護與健保給付. 臺灣醫學. 2002;6:4:581-4.

27. 徐慧君;翁慧卿;林育慈. 糖尿病患介入疾病管理在經濟面、臨床面及滿意度成效評估之初探--以南部某區域醫院糖尿病病患為例. 醫務管理期刊. 2004;5 (2):222-42.

29. 李璟佩, 朱正一, 吳篤安, 傅振宗, 李哲全, 王懿萱, et al. Comparison of the Diabetes Regular Care Model and the "Diabetes Share Care Disease Management" Model in Eastern Taiwan. 慈濟醫學. 2007;19:2:66-732.

30. 郝立智;馬瀰嘉;趙海倫. 某地區教學醫院糖尿病門診病患個案管理照護之成效探討. 中華民國內分泌暨糖尿病學會會刊. 2007 20 (4):11-33.