探討健保醫療資訊雲端查詢系統對安眠藥使用的影響

Abstract

背景： 國內外有許多關於醫師開立安眠藥處方、病患使用安眠藥、重複用藥之研究，顯示臺灣安眠藥重複領藥問題的重要性。2013年衛生福利部中央健康保險署實施「健保醫療資訊雲端查詢系統」，是政府解決重複用藥問題之因應政策。因此，本研究藉由探討健保醫療資訊雲端查詢系統與安眠藥重複用藥之相關性，作為檢驗健保醫療資訊雲端查詢系統政策效果。 目的： 目的一為評估健保醫療資訊雲端查詢系統介入後，對安眠藥重複處方天數及處方情形的影響；目的二則是探討健保醫療資訊雲端查詢系統實施後，新北市安眠類管制藥品健保調劑藥品是否流為自費調劑。 方法： 採用單組前後測設計，藉由觀察2013年健保醫療資訊雲端查詢系統實施前，與2016年實施健保醫療資訊雲端查詢統之後安眠藥處方使用的變化，來瞭解政策實施之效果。目的一係利用全人口健保資料取得處方劑量、平均每張處方安眠藥品種類及重覆處方天數；目的一樣本共有兩個族群，分別為全國門、急診曾經使用安眠藥的患者，以及全國門、急診三年均使用安眠藥的個案，進行描述性統計分析比較此兩組樣本，接著利用廣義估計方程式分析2016年實施健保資訊雲端查詢系統之後相較2013年實施健保資訊雲端查詢系統之前，後者處方劑量、平均每張處方安眠藥品種類及重複處方情形之變化。目的二亦區分為全國及新北市安眠藥調劑機構兩組樣本進行比較，健保調劑劑量是來自全人口健保資料庫，而新北市的自費調劑率是從管制藥品管理資訊系統取得的調劑總量及前述健保調劑劑量計算得之，接著剖析2016年實施健保醫療資訊雲端查詢系統之後，相較於2013年實施前自費調劑率之趨勢變化。 結果： 目的一樣本群體層次部分，全國門急診安眠藥使用者使用人數約占全國人口5%，其中有四分之一有重複處方情形，2016年的重複率為17.3%，較2013年減少了4.16%，每人平均重複處方天數在2013年為15.348日，到2016年下降到10.531日；全國門急診安眠類長期使用者計91,079人，其中有四至五成有重複處方情形，2016年的重複率為39.74%，較2013年減少了10.62%，每人平均重複處方天數在2013年為30.04日，到2016年下降到17.14日。個體層次部分，2016年重複處方的機率為2013年的0.681倍(p<0.0001)，2013年至2016年重複處方機率下降幅度較大； 2016年的重複處方天數顯著較2013年低43.01%(p<0.0001)，2013年至2016年下降的幅度較2010年至2013年稍少。 另外在目的二的樣本部分，新北市的自費調劑率在2010年為19.19%、2013年為16.22%，2016年則13.76%，其中2013年較2010年減少2.97 %，2016年較2013減少2.46%，降低的幅度差不多。 結論： 健保醫療資訊雲端查詢系統可能有效減緩處方劑量、安眠藥併用機率的上升趨勢，以及降低重複處方率，而自費調劑率未因政策介入增加，不過仍須注意安眠藥使用者盛行率上升趨勢之原因，建議加強重複用藥的管制方式，並進一步研究與監控，以避免安眠藥囤積，導致成癮性藥品流為非法用途使用。

Background: Many studies have addressed the prescription behaviors of hypnotics, patients’ utilization patterns of hypnotics, and duplicated medication of hypnotics, and pointed out the significance of duplicated medication in Taiwan. To tackle this problem, National Health Insurance Administration implemented a health information exchange system, entitled “PharmaCloud,” in 2013. In this regard, this study aims to examine the association between “PharmaCloud” and duplicated medication of hypnotics before and after the “PharmaCloud” program. Objective: The study aimed to evaluate the influence of the 2013 “PharmaCloud” program on the prescription, overlapping days and out-of-pocket utilizations related to hypnotics. Through the evaluation, this study is able to further examine whether the policy of “PharmaCloud” turns the controlled medication of hypnotics covered by health insurance into out-of-pocket prescriptions. Methods: One group pre- and post-test design is adopted to examine the trend prior to the introduction of “PharmaCloud” and the changes and effects after the implementation of this policy. To do so, this study extracts the information from National Health Insurance database to access the prescription dosage, the average types of hypnotics within each prescription, and overlapping days of hypnotics. The samples of this study for analysis are divided into two groups, respectively those who once used hypnotics and those who used hypnotics throughout 2010, 2013 and 2016, based on the users of hypnotics in nationwide out-patient clinics and emergency service departments. Then descriptive statistics is used to compare these two sets of samples: the latter sample is kept under track and observation for estimating the changes of prescription quantity, the average medication types of each prescription, and the situation of duplicated medication of hypnotics by using Generalized estimating equation (GEE). Secondly, we divide the region into nationwide and New Taipei City hypnotics-prescribing organizations for conducting a comparative study, with the quantity of hypnotics covered by health insurance extracted from National Health Insurance database and the out-of-pocket rate of prescription derived from the total dispensed quantity of hypnotics in New Taipei City from Controlled Drugs Management Information System. Subsequently we observe the trend change of out-of-pocket prescription rates prior to the implementation of “PharmaCloud” in 2013 and after its implementation in 2016. Results: In the samples of set I, the hypnotics users across nationwide out-patient and emergency visits accounts for 5% of total population, with one-quarter of them had duplicated medication. The rate of duplicated medication in 2016 is 17.3%, and a reduction of 4.16% compared with that in 2013. The average overlapping days drops from 15.348 days in 2013 to 10.531 days in 2016. The sub-sample of continuous users of hypnotics during 2010 and 2016 consisted of 91,079 people, with forty to fifty percent of samples having duplicated medication; the duplication rate in 2016 is 39.74%, and 10.62% less than that in 2013, with the average overlapping days per capita decreasing from 30.04 days in 2013 to 17.14 days in 2016. At individual level, the probability of duplicated medication in 2016 is 0.681 times (p<0.0001) that in 2013, with a steeper slope of drop rate from 2013 to 2016; the overlapping days in 2016 is significantly shorter than that in 2013 by 43.01% (p<0.0001), with a milder slope of drop in 2013-2016 compared to that in 2010-2013. In the sample of set II, the out-of-pocket rate of prescription in New Taipei City is 19.19% in 2010, 16.22% in 2013, and 13.76% in 2016, respectively. There is a reduction of 2.97% during 2010-2013 and a reduction of 2.46% during 2013-2016, revealing a similar extent of dropping. Conclusions: PharmaCloud may have effectively alleviated the rising trend of prescription quantity, of polypharmacy with hypnotics, and lower duplicated medication rate. On the other hand, the quantity of out-of-pocket prescription did not increase with the interference of the policy. The causes behind the rising trend of hypnotic usage still deserve further investigation, implying the needs of stricter regulation over duplicated medication to prevent stockpiling of hypnotics and the illegal usage of addictive drugs.