西部农村急性心肌梗死患者住院时长及其影响因素分析

目的 了解我国西部农村地区急性心肌梗死（AMI）患者平均住院时长情况和变化趋势，探索与平均住院时长有关的医院水平影响因素。方法 通过随机抽样收集我国西部农村地区9省份30家医院2001、2006、2011年的AMI住院患者病历资料，将纳入研究的医院按照住院时间的中位数划分为3类。采用针对整群数据的卡方检验或方差分析对3类医院患者特征差异进行检验，采用Mann-Kendall法检验住院时长在3个研究年份的变化趋势。采用多因素广义线性模型检验医院特征对住院时长的影响。结果 本研究最终共纳入814份病历。3类医院所入选的AMI患者的入院时间、肾小球滤过率存在显著差异(P＜0.05)。2001、2006及2011年的平均住院时长分别为（14.6±16.6）、（11.6±8.7）和（12.2±7.4）d，各年份间无显著的变化趋势（P=0.50）。调整患者水平特征后的无并发症病例分析结果显示，3个年份的平均住院时长分别为（11.1±8.3）、（10.9±6.9）和（11.6±5.7）d，各年份间仍无显著的变化趋势（P=0.68）。非附属非教学医院的患者平均住院时长较长，增加约6 d（95%CI：1.58～10.19），未开展冠脉造影医院的患者平均住院时长增加约3.7d（95%CI：0.46～6.99）。结论 我国西部农村地区医院AMI住院患者的平均住院时长仍不理想。2001~2011年平均住院时长无显著改善，进一步提高临床诊疗能力，缩短住院时长是改善我国西部农村地区AMI患者诊疗的关键。

Length of stay for acute myocardial infarction patients in western rural area of China

Objective To describe the lengfh of stay and its trend of AMI patients from western rural area of China and identify potential hospital level factors associated with LOS. Methods The study selected patients admitted to 30 hospitals for AMI in 3 years (2001, 2006, and 2011) in 9 provinces from western rural area of China by random sampling. Patient information was obtained from medical record via central abstraction. Hospitals were divided into 3 groups according to the median of LOS. Accounting for differences in case mix and year, patient characteristics were compared across hospitals using the Chi square test or ANOVA for clustered data. Mann Kendall trend test was used for testing time trend of LOS in 3 years. Multivariate generalized linear model was used for testing the association between hospital level characteristics and LOS. Results 814 cases were included in the analyses. Significant differences existed in admission time and eGFR concentrations among 3 hospital groups. The average lengty of stay was 14.6±16.6 days in 2001, 11.6±8.7 days in 2006, and 12.2±7.4 days in 2011, respectively, but without time trend (P=0.50). In non-complication subgroup adjusted individual level factors, the average LOS were 11.1±8.3, 10.9±6.9 and 11.6±5.7 days, respectively, without time trends (P=0.68). Neither affiliated nor teaching hospitals had on average 6 (95%CI: 1.58 -10.19) days longer LOS than affiliated or teaching hospitals. And hospitals without conducting coronary angiography had 3.7 (95%CI: 0.46-6.99) days longer LOS than those conducting coronary angiography. Conclusion The average LOS for AMI in western rural area of China was not satisfactory. The average length of stay for AMI was not significantly improved in western rural China from 2001 to 2011. Further improvement of AMI care in western rural area of China is critical to raise the level of diagnosis and treatment of AMI.