- 學位論文
睡眠障礙和安眠鎮靜劑使用對糖尿病患者整體成效之評估
Impact of Nonapnea Sleep Disorders and Use of Hypnotics and Sedatives on Diabetics
摘要
背景與動機 睡眠障礙是世界上最常見的健康問題之一，據估計失眠盛行率在一般人口約占10-20%，在臨床醫療中更高達50%患者會有睡眠相關問題。此外，安眠鎮靜用藥的處方比例有逐漸增加的趨勢。過去幾年間已有相關研究提出睡眠障礙可能升高糖尿病及相關慢性疾病風險，其中以睡眠呼吸中止症與糖尿病之間的相關性最多。但目前仍缺乏糖尿病患者其他睡眠障礙及安眠鎮靜藥物使用與慢性病控制及醫療資源耗用的相關研究。 目的 本研究嘗試分析糖尿病患者有無合併睡眠障礙(指排除呼吸中止症後之睡眠障礙，以下皆同)及使用安眠鎮靜劑不同日數，在糖尿病及慢性疾病相關臨床指標、存活、以及醫療資源利用上的差異，作為整合醫療照護、醫療提升醫療品質、減少醫療浪費，以及國際間比較的依據。 研究方法 本研究採用次級資料庫分析，以回溯性方式進行研究。研究樣本為2007-2008年間參加糖尿病論質計酬方案的第二型糖尿病病患，往前追蹤一年中，排除睡眠呼吸中止症者，再依照有無睡眠障礙診斷，及服用安眠鎮靜劑的情形分組，並往後追蹤至2012年。描述其人口學特質、疾病特質、醫療提供者特質並探討睡眠障礙及安眠鎮靜劑使用對於糖尿病及慢性病臨床生化數值、存活情形及醫療資源利用之影響。 研究結果 本研究樣本之樣本數共66,992人，收案前一年內有17.40%有睡眠障礙。女性占62.1%，其中合併焦慮症者占27.6%，合併憂鬱症者占12.5%。控制所有變項以及基期的臨床數值後發現，HbA1C的改善率在有睡眠障礙者較無睡眠障礙者低0.04%；有睡眠障礙者HbA1C>9控制不佳的機會較無睡眠障礙者高8%。 以安眠鎮靜劑使用情形區分， 38.60%有服用安眠鎮靜劑；在有開立安眠鎮靜劑的患者當中，開立≧60日者佔47.01%，其中以女性較多占61.1%，合併焦慮診斷者佔34.0%，合併憂鬱診斷者佔15.9%。控制所有變項以及基期的臨床數值後，HbA1C的改善率在開立≧60日較無開立安眠鎮靜劑組低0.04%，開立<60日組經校正後則無顯著差異；安眠鎮靜開立≧60日組之HbA1C>9控制不佳的機會較無開立組高17%，開立<60日組未達統計上的顯著差異。 分析存活情形發現，有睡眠障礙比無睡眠障礙存活時間較短，有睡眠障礙者比無睡眠障礙者死亡風險高3%，但未達顯著差異。開立<60日者較無開藥者的死亡風險高4%，但未達顯著差異，開立≧60日者較無開藥者的死亡風險高20%。 醫療資源利用在控制所有變項以及基期的臨床數值後，收案四年後的年平均總醫療費用，有睡眠障礙比無睡眠障礙者多4,552元，安眠鎮靜劑開立<60日者比無開立藥品者多5,028元，開立≧60日者比無開立藥品者更多了17,147元。 結論和建議 在論質計酬計畫的糖尿病患者中，合併睡眠障礙者之糖尿病控制改善情形較差，且增加醫療資源利用。開立安眠鎮靜劑60天以上的糖尿病患者在糖尿病控制的改善情形較差，存活情形較差，且增加醫療資源利用。在疾病管理上須考慮睡眠障礙及安眠鎮靜劑長期使用之影響，發展整合醫療照護模式，以提升醫療品質並減少醫療浪費。
並列摘要
Background and Motivation: Sleep disorders are common in clinical conditions. The prevalence of insomnia is approximately 10-20%, with approximately 50% in clinical practice. Besides, the prescription trends of sedative-hypnotics were increased year by year. Over the past decades, there has been a growing recognition that disorders of sleep could increase the propensity for type 2 diabetes and associated conditions. Among sleep disorders, associations between sleep apnea and DM were supported by both epidemiologic surveys and laboratory studies. However, previous studies concerning the association between nonapnea sleep disorders, use of sedative-hypnotics and DM have been relatively scarce. Objective: Our aim is to clarify the impact of nonapnea sleep disorders (NSD) and use of sedative-hypnotics on Diabetics, including HbA1C and associated laboratory data of medical conditions, survival, and medical utilization. Methods: The data used in our study were obtained from two nationwide population-based database. One was a population P4P registry database, and the other source was the NHI administrative claims databases. We first identified newly enrolled type 2 diabetes patients in the Diabetes P4P program between 2007 and 2008. Patients with obstructive sleep apnea diagnosis (ICD-9-CM code with 780.51, 780.53, 780.57) was excluded. Finally, total 66,992 patients was included. We identify the comorbidity of NSD diagnosis (ICD-9-CM codes, 307.4 and 780.5) and the prescription of sedative-hypnotics (ATC code: N05BA, N05CD, N05CF) during 1 year prior enrolled date of these patients. Each patient was followed up to 14 times till 2012, and to analyze the impact of NSD and use of sedative-hypnotics on Diabetics, including diabetes control, survival, and medical utilization Result: Total 66,992 patients was included, 17.40% of the cases had NSD diagnoses during 1 year prior enrolled date. Comorbidity with anxiety in NSD was account for 27%, and comorbidity with depression was account for 12.5%. The HbA1C improvement rate was lower by 0.04%, and greater risk of poor diabetes control (HbA1C>9) was noted in NSD group. (adjusted OR=1.08, 95% CI, 1.01-1.16, P=.03 ). 38.60% of included diabetics had at least one sedative-hypnotic prescription during 1 year prior enrolled date. 47% of them were prescript for over 60 days. Among patients with sedative-hypnotic prescriptions, female were predominant by 61.1%, 34% comorbid with anxiety and 15.9% with depression. HbA1C improvement rate was lower by 0.04%, and greater risk of poor diabetes control (HbA1C>9) was noted in patients with over 60 days of sedative-hypnotic prescription (adjusted OR=1.17, 95%CI, 1.08-1.25, P=<.001 ). In survival analysis, increased mortality risk was found in NSD group (adjusted HR=1.03, 95%CI, 0.96-1.12, P=.388), patients with under 60 days of sedative-hypnotic prescription (adjusted HR=1.04, 95%CI, 0.96-1.11, P=.345), and patients with over 60 days of sedative-hypnotic prescription (adjusted HR=1.20, 95%CI, 1.12-1.3, P=<.001) . Analyzing medical utilization, NSD group cost additional 4,552 TWD in average annual medical cost after 4 years of follow-up. Patients with under 60 days of sedative-hypnotic prescription cost additional 5,028 TWD and patients with over 60 days of sedative-hypnotic prescription cost additional 17,147 TWD than patients without any sedative-hypnotic prescription. Conclusion: Poorer diabetes control and more medical utilization was noted NSD group of type 2 diabetes. Beside, poorer diabetes control, poor survival and more medical utilization was also noted in diabetics with over 60 days of sedative-hypnotic prescription. To disease management point of view, the impact of NSD and use of sedative-hypnotics should be considered and develop integrated health care models in order to improve the quality of care and reduce medical cost of diabetics.