由多重死因診斷檢視糖尿病世代死亡記載有糖尿病之相關分析研究

Abstract

背景：死因資訊涉及病患家屬私人的權利義務，對死者的親屬而言，死亡證明書是重要的證明文件。經由糖尿病世代死亡多重死因有無提及糖尿病做為探討死亡證明書多重死因填寫品質，有助提升原死因判斷之正確性。 目的：探討糖尿病世代病人死亡之特異性、地理特性、院所屬性、醫療提供、就醫住院與否、就醫診斷個數等，對於多重死因糖尿病提及率之影響。 方法：研究資料源自DOH94-NH-1001衛生署計畫取得糖尿病世代次級資料，以該資料之2002-2005年糖尿病世代病人作為本研究對象，使用單變量邏輯斯迴歸模型、多變量逐步邏輯斯迴歸模型，分析研究對象影響多重死因糖尿病提及率之相關因素。 結果：本研究以2002-2005年糖尿病世代病人(n=20,443人)，其中多重死因有提及糖尿病之比率為39%(n=8,072人)。再按糖尿病世代病人死亡證明書開立院所分組，以醫院為研究對象多重死因有提及糖尿病之比率為36%；以診所為研究對象多重死因有提及糖尿病之比率為47%。對於糖尿病世代病人死亡多重死因糖尿病提及率，死亡證明書開立診所相較於醫院、公立院所相較於私立院所，多重死因較會提及糖尿病。糖尿病世代病人生前一年因糖尿病及心血管疾病、腦血管疾病、腎臟疾病三大合併症就醫次數或費用愈多，多重死因糖尿病提及率愈高。生前一年就醫次數或費用最多院所，與死亡證明書開立院所相同者相較於不同者，多重死因糖尿病提及率較高。 結論：就影響糖尿病世代多重死因糖尿病提及率之相關因素，針對死亡證明書填寫內容的準確性、明確性與對政策之有用性進行檢討評估，發展出例行的標準作業流程來改進死因統計品質。

Background: Death certification is important informaion that was concerned with private right and duty of decedents’ kinsfolk. To assess the quality of the information supplied on the certificate can follow diabetes cohort to death mentioning diabetes on the certificate, and it also will improve accuracy of coding underlying cause of death. Purpose: This study discussed the relationships of multiple causes of death mentioning diabetes as factors associated with diabetes cohort (specific, geographic, hospitals or clinics, medical utilization, inpatient or outpatient, and a number of medical certificates). Methods: Our sample was diabetes cohort obtained from secondary data (Depatment of heath: DOH94–NH-1001 in 2002-2005). The objectives of this study were to analyze rate of mentioning diabetes on death certificate using logistic model and stepwise multivariable logistic model. Result: Our diabetes cohort sample (n=20,443) was recorded as diabetes on 39% of death certificate (n=8,072). Stratifying by hospitals and clinics, sample mentioning diabetes were 36% in hospitals and 47% in clinics. We found out some factors related to death rate mentioning diabetes. Death certificate from clinics and public institution, more medical utilization (cost or times) at one year age (attributed to cardiovascular, cerebrovascular disease and nephropathy), institutions using most medical utilization, equal institution from between death certification and medical certification were assiociated with death rate mentioning diabetes. Conclusions: To examine the relationship of factors can potentially provide useful information about accuracy of cause of death coding, policy and assessment of health public and developing standard operating procedure to improve recording quality on death certificates.