1Department of Epidemiology and Public Health,University of Maryland School of Medicine,Baltimore,Maryland.
3Veterans Affairs Maryland Healthcare System Geriatrics Research,Education, and Clinical Center,Baltimore,Maryland.
Infect Control Hosp Epidemiol. 2017 Oct;38(10):1167-1171. doi: 10.1017/ice.2017.179.
OBJECTIVE To determine whether patients using the Centers for Medicare and Medicaid Services (CMS) Hospital Compare website (http://medicare.gov/hospitalcompare) can use nationally reported healthcare-associated infection (HAI) data to differentiate hospitals. DESIGN Secondary analysis of publicly available HAI data for calendar year 2013. METHODS We assessed the availability of HAI data for geographically proximate hospitals (ie, hospitals within the same referral region) and then analyzed these data to determine whether they are useful to differentiate hospitals. We assessed data for the 6 HAIs reported by hospitals to the Centers for Disease Control and Prevention (CDC). RESULTS Data were analyzed for 4,561 hospitals representing 88% of registered community and federal government hospitals in the United States. Healthcare-associated infection data are only useful for comparing hospitals if they are available for multiple hospitals within a geographic region. We found that data availability differed by HAI. Clostridium difficile infections (CDI) data were most available, with 82% of geographic regions (ie, hospital referral regions) having >50% of hospitals reporting them. In contrast, 4% of geographic regions had >50% of member hospitals reporting surgical site infections (SSI) for hysterectomies, which had the lowest availability. The ability of HAI data to differentiate hospitals differed by HAI: 72% of hospital referral regions had at least 1 pair of hospitals with statistically different risk-adjusted CDI rates (SIRs), compared to 9% for SSI (hysterectomy). CONCLUSIONS HAI data generally are reported by enough hospitals to meet minimal criteria for useful comparisons in many geographic locations, though this varies by type of HAI. CDI and catheter-associated urinary tract infection (CAUTI) are more likely to differentiate hospitals than the other publicly reported HAIs. Infect Control Hosp Epidemiol 2017;38:1167-1171.
目的 确定使用医疗保险和医疗补助服务中心（CMS）医院比较网站（http://medicare.gov/hospitalcompare）的患者是否可以使用全国报告的医疗保健相关感染（HAI）数据来区分医院。
设计 对 2013 年公开可用的 HAI 数据进行二次分析。
方法 我们评估了地理位置相近的医院（即同一转介区域内的医院）HAI 数据的可用性，然后分析这些数据以确定它们是否有助于区分医院。我们评估了医院向疾病控制和预防中心（CDC）报告的 6 种 HAI 的数据。
结果 分析了代表美国注册社区和联邦政府医院 88%的 4561 家医院的数据。只有在地理区域内有多个医院提供 HAI 数据的情况下，这些数据才对比较医院有用。我们发现数据可用性因 HAI 而异。艰难梭菌感染（CDI）数据的可用性最高，82%的地理区域（即医院转介区域）有超过 50%的医院报告。相比之下，只有 4%的地理区域有超过 50%的成员医院报告了子宫切除术的手术部位感染（SSI），这是最低的。HAI 数据区分医院的能力因 HAI 而异：72%的医院转介区域至少有一对医院的风险调整后 CDI 率（SIR）存在统计学差异，而 SSI（子宫切除术）为 9%。
结论 在许多地理位置，HAI 数据通常由足够数量的医院报告，以满足有用比较的最低标准，但这因 HAI 的类型而异。CDI 和导管相关尿路感染（CAUTI）比其他公开报告的 HAI 更有可能区分医院。感染控制与医院流行病学 2017；38:1167-1171.
