Identifying and Mitigating Disparities in Central Line-Associated Bloodstream Infections in Minoritized Racial, Ethnic, and Language Groups

Abstract

Importance: Although inequitable care due to racism and bias is well documented in health care, the impact on health care-associated infections is less understood.

Objective: To determine whether disparities in first central catheter-associated bloodstream infection (CLABSI) rates existed for pediatric patients of minoritized racial, ethnic, and language groups and to evaluate the outcomes associated with quality improvement initiatives for addressing these disparities.

Design, setting, and participants: This cohort study retrospectively examined outcomes of 8269 hospitalized patients with central catheters from October 1, 2012, to September 30, 2019, at a freestanding quaternary care children's hospital. Subsequent quality improvement interventions and follow-up were studied, excluding catheter days occurring after the outcome and episodes with catheters of indeterminate age through September 2022.

Exposures: Patient self-reported (or parent/guardian-reported) race, ethnicity, and language for care as collected for hospital demographic purposes.

Main outcomes and measures: Central catheter-associated bloodstream infection events identified by infection prevention surveillance according to National Healthcare Safety Network criteria were reported as events per 1000 central catheter days. Cox proportional hazards regression was used to analyze patient and central catheter characteristics, and interrupted time series was used to analyze quality improvement outcomes.

Results: Unadjusted infection rates were higher for Black patients (2.8 per 1000 central catheter days) and patients who spoke a language other than English (LOE; 2.1 per 1000 central catheter days) compared with the overall population (1.5 per 1000 central catheter days). Proportional hazard regression included 225 674 catheter days with 316 infections and represented 8269 patients. A total of 282 patients (3.4%) experienced a CLABSI (mean [IQR] age, 1.34 [0.07-8.83] years; female, 122 [43.3%]; male, 160 [56.7%]; English-speaking, 236 [83.7%]; LOE, 46 [16.3%]; American Indian or Alaska Native, 3 [1.1%]; Asian, 14 [5.0%]; Black, 26 [9.2%]; Hispanic, 61 [21.6%]; Native Hawaiian or Other Pacific Islander, 4 [1.4%]; White, 139 [49.3%]; ≥2 races, 14 [5.0%]; unknown race and ethnicity or refused to answer, 15 [5.3%]). In the adjusted model, a higher hazard ratio (HR) was observed for Black patients (adjusted HR, 1.8; 95% CI, 1.2-2.6; P = .002) and patients who spoke an LOE (adjusted HR, 1.6; 95% CI, 1.1-2.3; P = .01). Following quality improvement interventions, infection rates in both subgroups showed statistically significant level changes (Black patients: -1.77; 95% CI, -3.39 to -0.15; patients speaking an LOE: -1.25; 95% CI, -2.23 to -0.27).

Conclusions and relevance: The study's findings show disparities in CLABSI rates for Black patients and patients who speak an LOE that persisted after adjusting for known risk factors, suggesting that systemic racism and bias may play a role in inequitable hospital care for hospital-acquired infections. Stratifying outcomes to assess for disparities prior to quality improvement efforts may inform targeted interventions to improve equity.

Figure 1.. Run Chart of Central Line–Associated Bloodstream Infection (CLABSI) Rates, 2012-2022, Stratified by Subgroups Experiencing the Highest Rates

LOE indicates language other than English.

Figure 2.. Statistical Process Control Charts of Central Line–Associated Bloodstream Infection (CLABSI) Rates in Subgroups Experiencing the Greatest Inequity, 2012-2022

The solid line represents the center line (CL); the dashed line represents the upper control limit (UCL), and the dot-and-dash lines represent the sigma lines (±1 sigma, +2 sigma). LOE indicates language other than English.

Figure 3.. Interrupted Time Series of Quality Improvement Interventions on Central Line–Associated Bloodstream Infection (CLABSI) Rates in Subgroups Experiencing the Greatest Inequity, 2012-2022

Vertical lines denote the start of institution-wide CLABSI equity interventions (interruption time). For Black patients, the level of change was −1.77 (95% CI, −3.39 to −0.15; P = .03), and the slope change was −0.22 (95% CI, −0.52 to 0.09; P = .16). For patients speaking a language other than English (LOE), the level of change was −1.25 (95% CI, −2.23 to −0.27; P = .01), and the slope change was 0.39 (95% CI, 0.20 to 0.58; P < .001).