Chaudhuri Dipayan, Montgomery Alison, Gulenchyn Karen, Mitchell Morgan, Joseph Philip
Michael G. DeGroote School of Medicine (D.C.), Departments of Medicine (A.M., K.G., P.J.), Radiology (A.M., K.G., P.J.), Faculty of Science (M.M.), and Population Health Research Institute (P.J.), McMaster University, Hamilton, Ontario, Canada.
Circ Cardiovasc Qual Outcomes. 2016 Jan;9(1):7-13. doi: 10.1161/CIRCOUTCOMES.115.001836. Epub 2016 Jan 5.
Between 5% and 25% of cardiac imaging tests are performed for inappropriate indications. Studies have examined the impact of appropriate use criteria-based quality improvement initiatives on inappropriate testing, but they have not been systematically evaluated.
We performed a systematic review of studies evaluating quality improvement initiatives aimed at reducing inappropriate cardiac imaging. The primary outcome was the proportion of inappropriate tests based on appropriate use criteria. Studies were analyzed using a random effects meta-analysis model, and heterogeneity was examined using subgroup analyses. We identified 6 observational studies and 1 randomized control trial. Most interventions (n=6) had a formal education component, and 5 included a mechanism for physician audit and feedback. Although these interventions were associated with lower odds of inappropriate testing (odds ratio, 0.44 [95% confidence interval, 0.32-0.61]; P<0.001), significant heterogeneity was observed (I(2)=70%), which was best explained by the utilization of physician audit and feedback. Interventions that employed physician audit and feedback were associated with significantly lower odds of inappropriate testing (odds ratio, 0.36 [95% confidence interval, 0.31-0.41]; P<0.001; I(2)=0%), whereas those that did not had no effect (odds ratio, 0.89 [95% confidence interval, 0.61-1.29]; P=0.51; I(2)=0%; P value for difference <0.001). All studies had potential sources of bias that could have affected the observed estimates.
Interventions using physician audit and feedback are associated with lower odds of inappropriate cardiac testing. Further research is needed to evaluate a greater diversity of intervention types, with improved study designs.
5%至25%的心脏成像检查是在不恰当的指征下进行的。已有研究探讨了基于合理使用标准的质量改进措施对不恰当检查的影响，但尚未进行系统评估。
我们对评估旨在减少不恰当心脏成像的质量改进措施的研究进行了系统综述。主要结局是基于合理使用标准的不恰当检查比例。使用随机效应荟萃分析模型对研究进行分析，并通过亚组分析检查异质性。我们确定了6项观察性研究和1项随机对照试验。大多数干预措施（n = 6）包含正式教育内容，5项包括医师审核和反馈机制。尽管这些干预措施与不恰当检查的较低几率相关（优势比，0.44 [95%置信区间，0.32 - 0.61]；P < 0.001），但观察到显著的异质性（I² = 70%），这可以通过医师审核和反馈的使用得到最佳解释。采用医师审核和反馈的干预措施与不恰当检查的显著较低几率相关（优势比，0.36 [95%置信区间，0.31 - 0.41]；P < 0.001；I² = 0%），而未采用的则没有效果（优势比，0.89 [95%置信区间，0.61 - 1.29]；P = 0.51；I² = 0%；差异P值 < 0.001）。所有研究都存在可能影响观察估计值的潜在偏倚来源。
采用医师审核和反馈的干预措施与不恰当心脏检查的较低几率相关。需要进一步研究以评估更多样化的干预类型，并改进研究设计。
