分析抗焦慮劑/安眠劑之使用的影響因子在重度憂鬱症及廣泛性焦慮症病人和一般大眾的處方形態 研究背景: 苯二氮平(benzodiazepine)和苯二氮平接受器致效劑(benzodiazepine receptor agonist,俗稱Z-drug)這類的抗焦慮劑和鎮靜安眠劑通常被用來在重度憂鬱症(major depressive disorder)及廣泛性焦慮症 (generalized anxiety disorder) 的病人中，緩解病人焦慮以及失眠的情形。使用鎮靜安眠劑所要考量到的是，它可能會有耐藥性(tolerance)、成癮(dependence)、車禍或是易造成老年人跌倒的危險；所以治療指引建議這類僅使用在病人處於急性期的狀態。過去研究指出，在某些的族群中(特別是老年人)，哪些因子會影響這類藥物的使用。然而在重度憂鬱症及廣泛性焦慮症的病人族群中，這樣的研究則是很少。台灣在2005年發行BZD的使用治療指引，但沒有研究指出醫師是否有遵照此指引來開立BZD的處方。 研究目的: 本研究主要的目的是要去觀察抗焦慮劑/安眠劑的使用，在重度憂鬱症及廣泛性焦慮症的病人中；以及利用2004及2006年台灣全民健康保險資料庫來比較抗焦慮劑/安眠劑的使用狀況。 研究方法: 本研究的資料來源為全民健康保險資料庫(2004-2006年)，以及臺北醫學大學‧萬芳醫學中心的資料庫(2004年-2009年1月)。本研究分成兩個部份。第一個部份，研究的對象為有被診斷為重度憂鬱症(major depressive disorder)，且或者是廣泛性焦慮症(generalized anxiety disorder)的病人；病人如果其它疾病像是癲癇(seizure)，或者是中樞系統的疾病，則會被排除，因為這些疾病可能會影響病人使用鎮靜安眠劑的時間。而萬芳醫學中心組的部份，病人如能經由醫生指示完成抗憂鬱劑的治療療程，則納入。本研究使用單因子迴歸分析 (univariate regression model) 來選擇與抗焦慮劑/安眠藥使用有顯著相關之因子，然後使用多因子逐步分析迴歸模式(backward stepwise multivariate regression model)來找出對抗焦慮劑/安眠藥使用更具有影響力的相關因子。在本研究的第二部份，以2004及2006年全民健康保險資料庫為資料來源，每一年資料庫提供一百萬人的資料；將有拿慢性處方籤之病人資料收集，用來比較在治療指引發行前一年及後一年，藥品處方的情況有何不同。本研究使用卡方檢定(chi-square test) 來分析其統計上的差異。 研究結果: 在本研究的第一個部份，在2004 到2006年期間的全民健保資料庫中，有17,266位病人被診斷為重度憂鬱症或廣泛性焦慮症；其中有8,902位病人是重度憂鬱症，有9,334位病人是廣泛性焦慮症，而有1,010位病人同時被診斷有這種疾病。將這些病人符合本研究之納入、排除條件之後，共399位病人被納入本研究的健保資料庫組，而有63位病人被納入至萬芳醫學中心組。在兩組資料庫中，抗憂鬱劑的開立天數及抗憂鬱劑的使用個數是兩個具有顯著相關的因子。年齡及大於四個月的抗焦慮劑/安眠劑的使用在抗憂鬱劑的治療前，則有顯著相關在健保資料庫組。在本研究的第二部份中，在每年一百萬人的全民健康保險資料庫中，在2004年有11,039位病人有拿BZD的慢性處方籤，而在2006年則顯著上升至16,193位病人。而有拿開立BZD超過30天的慢性處方籤，在2004年有6,505位病人，在2006年則顯著上升至8,790位病人。另一方面，有拿開立BZD 30天以下之病人，在2004年有4,534位病人，在2006年則顯著上升至7,401位病人。 研究結論: 本研究的結果顯示，抗憂鬱劑的開立天數及抗憂鬱劑的使用個數對抗焦慮劑/安眠劑的使用有顯著的相關性。而年齡及大於四個月的抗焦慮劑/安眠劑的使用在抗憂鬱劑的治療前，有顯著的相關性在健保資料庫組中。在治療指引發行之後，2006年的BZD處方量比較在指引發行前的2004年，顯著的上升。明顯地，在台灣醫師開立BZD的處方行為上，並沒有因為2005年指引的發行而有所改變。

Analysis of Factors Associated with the Use of Anxiolyitcs/Hypnotics among MDD and GAD Patients and the Prescription Pattern among General Population Background: Benzodiazepines (BZDs) or benzodiazepine receptor agonists, also called Z-drugs (e.g. zolpidem, zopiclone or zaloplon), have anxiolytic and hypnotic functions to relieve symptoms among major depressive disorder (MDD) and generalized anxiety disorder (GAD) patients. The concerns of using anxiolytic and hypnotic agents are tolerance, dependence, accident proneness, and increased risk of falls in elderly. Thus, these agents should be used only in the acute phase, indicated by treatment guidelines. Factors associated with the use of these agents have been studied within whole population, especially in the elderly. However, the factors have not been studied in MDD and GAD patients. Taiwan’s government authority implemented the guideline in 2005. Whether physicians would have followed this guideline to prescribe BZDs, is still unknown. Objective: The goal was to investigate the use of anxiolytic/hypnotic agents in the MDD and GAD patients and to compare the prescription pattern of anxiolytic/hypnotic agents in National Health Insurance Research Database (NHIRD) 2004 and 2006 in Taiwan. Methods: The data of this study were obtained from NHIRD (for 2004 to 2006; 1,000,000 people per year) and TMU-Wan Fang Medical Center database (T-WFMCD) during 2004 to 2009. In the first part of this study, patients who were diagnosed of either MDD or GAD at clinics were selected, and those who had other medical conditions (e.g. seizures or other CNS disorders) that affect the duration of hypnotic-use were excluded. Patients who had finished their antidepressant treatment with doctor’s guide were included in T-WFMCD group. Univariate regression models were performed to select the significant correlated factors with the use of anxiolytic/hypnotic agents. Then backward stepwise multivariate regression models were performed to analyze the factors with the more powerful effect on the use of anxiolytic/hypnotic agents among these significant correlated factors. In the second part of this study, the data were based on the 1,000,000 people in 2004 and 2006 NHIRD. The numbers of patients who received the BZD prescriptions in chronic use were collected for comparison of the prescription pattern in pre- and post-guideline year. Chi-square test was used for testing the statistic difference. Results: In the first part of this study, 17,226 patients were diagnosed of MDD and/or GAD during 2004 to 2006. The sum of 8,902 and 9,334 patients were diagnosed as MDD and GAD, respectively in these three years. A total of 1,010 patients were comorbid with both MDD and GAD. After fulfilling the inclusion and exclusion criteria, 399 patients were included into NHIRD group. 63 patients were included into T-WFMCD group. The antidepressant-prescribing days, and the number of antidepressants used were found to be significantly correlated in both databases. Age and the four month or more of the use of anxiolytic/hypnotic agents before antidepressant therapy were found to be significantly correlated in NHIRD group., Based on the 1,000,000 patient population in the second part of this study, the overall numbers of patients who received the BZD prescriptions in chronic use were found to be increased significantly from 11,039 to 16,193 in 2004 to 2006. And the numbers of patients who received more than 30 days of BZDs were increased significantly from 6,505 to 8,790. While the numbers of patients with less than 30 days of BZDs were found to be increased significantly from 4,534 to 7,401. Conclusion: The results of present study suggested that the antidepressant-prescribing days, and the numbers of antidepressants used are significant associated with the use of anxiolytic/hypnotic agents. And age and the four month or more of the use of anxiolytic/hypnotic agents have significant association in NHIRD group. After the publication of the guideline, the BZDs prescriptions in 2006 have been found to be significantly increased over the pre-guideline year of 2004. Apparently the prescription pattern of BZDs by Taiwanese physicians was not changed by the publication of the 2005 guideline of the management and use of BZDs.