Association Between a Two-Cycle Plan-Do-Study-Act Quality Improvement Intervention and Biomedical Waste Segregation Compliance in a Large Indian Government Tertiary Teaching Hospital: A 51-Month Interrupted Time Series Study.

backgroundDespite guidelines, Indian public sector hospitals continue to face challenges in biomedical waste (BMW) mismanagement, and various gap assessments at the national level show uneven segregation practices. Needle-stick injuries (NSIs), an occupational hazard for healthcare workers (HCWs) in developing countries, have been found to have an incidence of 8.9-10.4 events per 100 beds per year in Indian tertiary care hospitals. As far as we can judge, no study from a large Indian government tertiary setting has evaluated the incidence of NSIs concurrently with Plan-Do-Study-Act (PDSA) cycles for BMW management using interrupted time series (ITS) analysis.

objectiveThis study aimed to evaluate the association between a two-cycle PDSA-driven quality improvement (QI) intervention and (i) BMW segregation compliance and (ii) NSI incidence per 100 occupied bed-years using 51 months of prospective hospital data.

methodsWe conducted a quasi-experimental ITS regression study at a government tertiary teaching hospital (2,327 beds, 95% occupancy, approximately 2,210 occupied beds) in Jharkhand, India. A concurrent control was not feasible because the intervention was implemented hospital-wide as part of preparation for National Accreditation Board for Hospitals and Healthcare Providers (NABH) accreditation. Primary data were drawn from monthly weighbridge logs, periodic waste composition audits, structured audit checklists, and a manual paper-based incident register over 51 consecutive months (January 2022 to March 2026); the intervention began in April 2025. The primary model was segmented ordinary least squares regression with Newey-West standard errors; sensitivity analyses incorporated seasonal harmonic adjustment, lag variation, and alternative autocorrelation specifications.

resultsSegregation compliance showed a significant immediate level increase of 14.3 percentage points (pp) at intervention onset (β₂=14.31; 95% CI: 9.55-19.08; p<0.001) and a positive post-intervention slope change of 2.57 pp per month (β₃=2.567; 95% CI: 1.981-3.152; p<0.001). The pre-intervention trend was flat to slightly negative (β₁=-0.092; p=0.066), and model fit was high (R²=0.939). All four sensitivity analyses were directionally consistent. The descriptive decline in NSI (2.08-0.33 events per month) was not confirmed by the log-linear ITS model (β₂ p=0.397; β₃ p=0.270) and is therefore reported as an associative trend only. Knowledge scores improved across all cadres (p<0.001), with the largest gains among sanitation staff (+37.8 pp) and waste handlers (+36.3 pp).

conclusionsThe PDSA intervention improved BMW segregation compliance, which was significant and sustained over time. Without a simultaneous control group and given institutional confounders, enormous caution in referring to the impacts as causal remains advised. Caution should be exercised in the interpretation of NSI trends as they are not statistically significant and are based on passive reporting. The results indicate that the PDSA-based frameworks are feasible in public hospitals.