低效益醫療服務使用之趨勢及其與市場競爭與總額預算之相關性

Abstract

臺灣醫療支出快速成長，時有醫療浪費之疑慮，而目前主要管控措施係以總額支付制度與高耗用服務控管為主要標的，較少整體考量各項醫療服務的使用是否具合理性，因此本研究試圖透過臺灣低效益醫療服務利用之長期趨勢，以及探討總額支付制度及醫療市場競爭對低效益醫療服務利用之影響。 本研究分為兩部分，第一部分為探討低效益醫療服務利用之長期趨勢，採用2000至2015年之全民健康保險行政申報資料進行分析，以醫院為研究對象，分析項目包含不符適應症之下背痛檢查、低風險手術病人接受心血管檢查、頭痛病人開立鴉片類藥物等13項低效益醫療服務指標，並以轉折點迴歸分析進行長期趨勢分析；第二部分則就總額支付制度及醫療市場競爭對低效益醫療服務利用之影響進行探討，此部分探討重點為高單價之低效益醫療服務，爰從13項醫療服務中篩選出之4項低效益電腦斷層及核磁共振服務項目(下背痛影像檢查、影像檢查診斷非複雜性的頭痛、影像檢查診斷暈眩、影像檢查診斷急性非複雜性鼻竇炎)，並採用1998至2015年之全民健康保險行政申報資料，以時間序列及多階層分析，探討總額支付制度及醫療市場競爭對低效益電腦斷層及核磁共振之相關性。 研究結果顯示「以BZD治療失眠、焦慮或譫妄之老年病人」無論執行件數、執行率、平均每院所執行件數及平均每院所執行率均高，且符合診斷對象明顯有重複用藥情形，應是高齡化社會下須關注的指標；其次則為「低風險病人接受心血管篩檢」，其執行件數高且執行件數、執行率、平均每院所執行件數及平均每院所執行率均呈現持續線性成長；「以影像檢查診斷卵巢囊腫」、「對甲狀腺低下病人執行T3檢驗」其符合診斷個案均有多於1次以上之利用，重複利用情形值得注意。13項指標中「以腦電圖診斷頭痛」在觀察期間無明顯增減，另「以抗生素治療病毒感染引起之上呼吸道感染」及「膝關節內視鏡手術治療膝關節炎」呈下降趨勢，其餘指標均為成長趨勢。與已知國內、外文獻之執行率進行比較，除「以鴉片類藥物治療偏頭痛」、「以抗生素治療病毒感染引起之上呼吸道感染」及「膝關節內視鏡手術治療膝關節炎」明顯較低外，其餘指標多高於其他國家研究結果。 平均每院所執行件數或平均每院所執行率則因受到申報特定診斷院所數之影響，爰計算結果較整體執行件數及執行率低，但其變化趨勢仍與整體執行件數及執行率相近。以平均每院所執行率而言，除「以BZD治療失眠、焦慮或譫妄之老年病人」、「低風險病人接受心血管篩檢」、「以影像檢查診斷卵巢囊腫」及「對甲狀腺低下病人執行T3檢驗」高於30%外，其餘9項指標有3項介於15~20%，有6項均低於5%以下。 層級別間可觀察到地區醫院及基層診所的低效益醫療服務利用多為低基數高成長，醫學中心及區域醫院則為高基數低成長。不同手術及住院服務醫療費用占率也有類似層級別的趨勢存在，手術及住院服務醫療費用占率低者也是低基數高成長，手術及住院服務醫療費用占率高者則為高基數低成長，且比層級別之結果更具的一致性。 總額制度實施對低效益醫療服務利用有抑制的效果，但可能並非直接影響，應係間接透過總量下降影響，而市場競爭程度則與低效益醫療服務之利用有些微影響，但其影響應也與電腦斷層及核磁共振之總利用有關。 本研究透過瞭解臺灣低效益醫療服務利用分布狀況，有助於政府單位適度針對不同層級院所制定不同因應策略，建議針對區域級以上醫院採用自動警示系統，並針對地區醫院及基層診所強化品質監控及管理，但總額制度及市場競爭程度對低效益醫療服務之利用之抑制效果可能與總服務利用量有關，因此單純希望透過醫界自律或是熟能生巧來抑制可能效果不彰，應以更明確之財務誘因配合警示系統協助院所減少低效益醫療服務，並投入更多資源在高價值之醫療服務。

Medical expenditure in Taiwan has increased faster than ever, and is usually criticized for the overuse of unnecessary care. Taiwan government have managed the expenditure by global budget system and also by monitoring high utilizers, high utilization of expensive medical services and medications. However, we have rarely focused on clinical appropriateness of medical services, therefore, this research tried to find out the trend of low-value procedures in Taiwan and explored its association with global budget system and market competition. This study was divided into two parts. First, we examined the trend of selected thirteen low-value procedures indicators, such as “Order low back pain imaging for patient with no red-flag”, “Order cardiac tests on low-risk, asymptomatic patients”, “Prescribing opioids for migraine” and so on. In this part, we included hospitals which had existed during 2000-2015, and employed joint point regression to assess changes of each indicators. Secondly, we selected four indicators out of thirteen low-value indicators as dependent variable, in order to know the impact of global system and market competition on low-value computed tomography (CT) and magnetic resonance imaging (MRI) utilization. We included hospitals which had existed during 1998-2015, and employed a time-series and multilevel model to our 15 years cross-sectional dataset to study the dynamic association among low-value imaging, market competition and global budget. We found “Benzodiazepine use in the elderly diagnosed insomnia, anxiety, and delirium” was prescribed most among thirteen indicators, since it was a usual repeated medication, which should be noticed in aging society. “Order cardiac tests on low-risk, asymptomatic patients” was the second most prescribing procedure, which showed a stable linear growth trend for 15 years. “Imaging for ovarian cyst” and “T3 test for hypothyroidism” were other usual repeated procedures in thirteen indicators, patients who were diagnosed ovarian cyst or hypothyroidism accepted these examinations more than once. “EEG for headache” remained steady, while “Prescribing antibiotics for non-viral acute upper respiratory tract infections” and “Arthroscopic surgery for knee osteoarthritis”showed a downward trend, and the remaining six indicators kept growing gradually. When compared our result with existing literature, except for “Prescribing opioids for migraine”, “P Prescribing antibiotics for non-viral acute upper respiratory tract infections” and “Arthroscopic surgery for knee osteoarthritis” were consistently lower than most references, and the other ten indicators were higher. The volume per medical institution or prescribing rate per medical institution were diluted due to large pool of medical institutions which fulfill criteria of specific diagnosis, although the result was lower than overall volume and overall prescribing rate, the trend was similar. The prescribing rate per medical institution of “Benzodiazepine use in the elderly diagnosed insomnia, anxiety, and delirium”, “Order cardiac tests on low-risk, asymptomatic patients”, “Imaging for ovarian cyst” and “T3 test for hypothyroidism” remained high, and beyond 30%. The prescribing rate per medical institution of remaining 9 indicators, three of them were between 15%~20%, the others were lower than 5%. The trend of indicators among district hospitals and primary care clinics increased significantly with a low original base, while medical centers and regional hospitals showed a stable growth trend with a high original base. The differences among hospitals classified to high or low proportion of operation and hospitalization expenditure to total medical expenditure showed a similar trend as which we found in hospitals with different accreditation level. The global budget system had some suppression effect on low-value CT and MRI imaging; however, the effect might result from total medical utilization decrease which were not a specific effect on low-value CT and MRI imaging. The market competition had mild effect in reducing low value imaging, and the effect might result from association with the total utilization of CT and MRI. This research presents distribution of low-value procedures utilization in Taiwan, government should formulate different policies toward different hospitals. For medical centers and regional hospitals, we should emphasize more on auto-alert system; and for district hospitals and primary care clinics, we should focus more on quality monitor and improvement. Since global budget system and market competition only has mild suppression effect on low value imaging, it is hard to eliminate the utilization through intervention in macro-level. Therefore, we should propose specific financial incentive programs accompanied by auto-alert system which is helpful in reducing utilization of low-value procedures, and also in increasing utilization of high-value procedures.