| 摘要: | 背景及目的：醫療界有兩種特殊的效應-「七月效應 (July effect)」及「週末效應(weekend effect)」。出現「七月效應」可能是因為新進的醫護工作人員在臨床經驗上不足所導致的醫療不良結果，而「週末效應」可能是因為醫療資源分配不均所造成的。在台灣的全民健康保險制度下是否還會出現「七月效應」或「週末效應」，甚至出現「七月效應」及「週末效應」的合併效果，非常值得深入探討。本研究之主要目的以急性心肌梗塞及中風病人為研究對象，探討「七月效應」或「週末效應」對於急性心肌梗塞或中風病人，在醫療照護品質(是否住院後30日內死亡、是否出院後14日內再入院)及醫療利用(住院天數、醫療費用)等方面的影響及其相關因素。
方法：本研究係以國家衛生研究院發行的全民健康保險資料庫(LHID2000及LHID2005)為資料來源，研究對象為大於18歲且第一次被診斷為急性心肌梗塞或中風的住院病人。研究資料擷取時間為2006年1月1日至2012年9月30日。
結果：無論是急性心肌梗塞或是中風病人，七月入院者和非七月入院者，在住院30日內死亡以及出院後14日內再入院的醫療照護指標，都沒有出現七月效應。而急性心肌梗塞病人，七月入院者的總住院日數以及總醫療費用，均與非七月入院者沒有差異。中風病人七月入院者的總住院日數，是非七月入院者的0.97倍，亦未達到統計上的顯著差異。但在區域醫院、公立醫院治療的中風病人，七月入院者反而比非七月入院者有較低的總住院日數。中風病人在區域醫院、公立醫院治療，的確出現了七月效應，但卻是非七月入院者之總住院日數較高。區域醫院、公立醫院、醫院所在地都市化程度3-7級治療的中風病人，七月入院者反而比非七月入院者有較低的總醫療費用。中風病人在區域醫院、公立醫院、醫院所在地都市化程度3-7級等層級做治療，的確出現了七月效應，但卻是非七月入院者之總醫療費用較高。急性心肌梗塞病人和中風病人都在當次住院總日數出現週末效應(週末入院者的總住院日數為非週末入院者的1.06倍)；無論是在醫學中心治療的中風病人，或是在區域醫院、公立醫院、非公立醫院治療的中風病人，週末入院的住院總日數都顯著比非週末入院者長，「週六或週日但非七月入院者」的住院總日數也比「非七月入院亦非週末入者」長。週末入院雖不影響醫療照護品質，但確實延長了當次住院的總日數，但對於當次住院總醫療費用影響較小。
結論：急性心肌梗塞病人，在檢定住院後30日內死亡、出院後14日內再入院、當次住院總日數、總醫療費用等指標之後，並沒有發現有七月效應，但在當次住院總日數則出現了週末效應。而中風病人在當次住院總醫療費用雖然發現有七月效應的存在，但卻是七月比非七月入院者有較低的醫療費用，中風病人在當次住院總日數也出現了週末效應。
Background and purpose: There are two peculiar phenomena in healthcare: the “July effect,” possibly due to the arrival of new healthcare personnel with insufficient clinical experience, leading to poorer healthcare outcomes; and the “weekend effect,” possibly due to an uneven distribution of healthcare resources. Whether the July effect, the weekend effect, or even a synergistic effect of the two is present under Taiwan’s universal National Health Insurance system warrants an in-depth examination. The main purpose of this study was to examine the July effect and the weekend effect on the care of patients with acute myocardial infarction (AMI) and stroke in terms of healthcare quality (mortality within 30 days of hospitalization and readmission within 14 days of discharge) and healthcare utilization (days of hospitalization and healthcare cost) and to analyze factors associated with these effects.
Methods: The National Health Insurance Research Database (LHID2000 and LHID2005), maintained by the National Health Research Institutes, was used as the data source. Analysis was performed with data for adult patients aged ≥18 years who were admitted with AMI or stroke for the first time and were discharged between January 1, 2010, and September 30, 2012.
Results: Among both the AMI patients and the stroke patients, no July effect was apparent for July admissions versus non-July admissions with respect to healthcare quality indicators, 30-day mortality and readmission within 14 days of discharge. AMI patients admitted in July did not differ from those admitted during non-July months in either total days of hospitalization or total healthcare cost. Stroke patients admitted in July had 0.97 times as many total days of hospitalization as those admitted during non-July months, but the difference was not statistically significant. However, for stroke patients treated at regional hospitals and public hospitals, July admission was actually associated with reduced total days of hospitalization relative to non-July admission, indicating the presence of an opposite kind of July effect. For stroke patients treated at regional hospitals, public hospitals, and hospitals located in urbanization level 3–7 areas, July admission was also associated with a lower total healthcare cost than non-July admission, again indicating an opposite July effect. Among both the AMI patients and the stroke patients, the weekend effect was observed in total days of initial hospitalization, with weekend admissions staying 1.06 times as long as non-weekend admissions. For stroke patients treated at either medical centers, regional hospitals, public hospitals, or non-public hospitals, weekend admission was associated with more total days of hospitalization than non-weekend admission, and weekend (Saturday or Sunday) non-July admissions likewise had longer hospital stays than non-weekend non-July admissions. Thus, although weekend admission did not affect the quality of care, it did lengthen the duration of the initial hospitalization, but had little effect on the total healthcare cost of the initial hospitalization.
Conclusion: For patients admitted with AMI, no July effect was evident in terms of indicators including 30-day mortality, 14-day readmission, days of initial hospitalization, and total healthcare cost of initial hospitalization, whereas the weekend effect was observed in total days of initial hospitalization. For patients admitted with stroke, a July effect was present with respect to the total healthcare cost of initial hospitalization, but with the cost being lower for July admissions than for non-July admissions; the weekend effect was also present in total days of initial hospitalization. |
