现在的位置: 首页时讯速递, 进展交流>正文
[JAMA Surg发表论文]:全国安全网医院内手术部位感染的倾向性评分匹配比较
2024年04月22日 时讯速递, 进展交流 [JAMA Surg发表论文]:全国安全网医院内手术部位感染的倾向性评分匹配比较已关闭评论

Research Letter 

February 28, 2024

Nationwide Propensity-Matched Comparison of Surgical Site Infections in Safety Net Hospitals

Lisa Ngo, Luis Quintero, Joshua P. Parreco

JAMA Surg. Published online February 28, 2024. doi:10.1001/jamasurg.2023.7863

Since 2014, the Centers for Medicare and Medicaid Services (CMS) has been imposing financial penalties on hospitals with high rates of health care–associated infections and financially rewarding hospitals with low rates.1 These penalties and rewards have been ineffective at reducing infection rates at safety net hospitals, which disproportionately serve economically disadvantaged and medically complex patients, who often are at higher risk for infections.2 This phenomenon has been termed reverse Robin Hood effect for essentially stealing from hospitals that treat poorer and sicker patients while giving to hospitals with wealthier and healthier patients.3 We compared rates of surgical site infections (SSIs) in patients undergoing gastrointestinal surgery in safety net hospitals vs all other hospitals.

Methods

We obtained data from the Nationwide Readmissions Database (NRD), the most accurate tracker of readmissions for including admissions across different hospitals. Widespread use of new International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes for SSI began from 2019 to 2020. These years were queried for all admissions of survivors of gastrointestinal surgery. In accordance with the Common Rule, this cohort study was exempt from ethics review and informed consent because it was not human participant research. We used deidentified patient data in compliance with the NRD data use agreement. We followed the (STROBE) reporting guideline.

The primary outcome was an SSI identified during initial admission or readmission within 90 days. Safety net hospitals were identified as publicly funded, metropolitan teaching, and large by number of beds per US region. Propensity matching was performed 1:1 with risk factors for age, sex, anatomic location of surgery, elective surgery, trauma, low household income, lack of insurance, and 38 Elixhauser comorbidities.

Two-sided P < .05 indicated statistical significance. Data analysis was performed from January to September 2020 using SciPy 1.10.1.

Results

Of 392 368 patients identified, 23 447 (6.0%) were treated in safety net hospitals (mean [SD] age, 54.25 [17.5] years; 13 264 females [50.2%], 10 183 males [49.7%]). The overall SSI rate was 3.0% (11 617), of which 59.2% of infections (6878) were found during readmission. The SSI rate in safety net hospitals was 7.8% (911; P < .01). Table 1 shows matching was successful, and there were no significant differences between the 2 groups. After matching, the overall SSI rate was 3.5% (1626) and SSI risk was increased in safety net hospitals (odds ratio [OR], 1.29; 95% CI, 1.16-1.42; P < .01). Table 2 shows that risk was increased in safety net hospitals for each site (eg, superficial incisional: OR, 1.27; 95% CI, 1.07-1.52; P < .01).

Discussion

To our knowledge, this cohort study was the first to evaluate SSI rates in patients undergoing gastrointestinal surgery since the widespread use of ICD-10 codes, corresponding to different types of SSI and implementation of CMS’s financial incentive programs to reduce infection rates. These programs attempt to provide some risk adjustment for SSIs by comparing hospitals using formulas that calculate the expected number of infections. However, these formulas use fewer than a dozen risk factors, including few comorbidities and 0 socioeconomic factors.4 The present study used 38 comorbidities and included insurance status and household income. Despite controlling for these factors through propensity score matching, we found that patients treated in safety net hospitals were still at an increased risk for SSIs. In 2019, CMS made adjustments to hospital penalties for readmissions, accounting for socioeconomic differences among patient populations treated in safety net hospitals.5 This study demonstrates that similar adjustments for SSIs are needed.

Study limitations include the challenges inherent in using administrative databases, such as coding errors that could alter study results. Additionally, the NRD does not include data on race and ethnicity, preventing analysis of these factors. Nevertheless, the NRD is ideally suited for this study since most SSIs were found during readmission, and this database includes up to 23% of surgical readmissions that are missed by single-institution studies.6

抱歉!评论已关闭.

×
腾讯微博