Sustainability of a Surgical Quality Improvement Program at Hospitals in Ethiopia

Surgical infections are associated with significant morbidity and mortality in low- and middle-income countries.¹ Recently, a number of infection prevention and control programs have been implemented in low-income countries with subsequent reduction in surgical site infections.^2,3 Clean Cut is one such adaptive, multimodal quality improvement intervention developed in Ethiopia. High compliance with perioperative infection prevention practices was associated with a 35% relative risk reduction in surgical site infections in the postintervention period compared with baseline.⁴ The persistence of practice changes after quality improvement interventions is rarely assessed,⁵ and whether many surgical infection prevention and control quality improvement programs in low- and middle-income countries provide short-term solutions only or result in long-term cultural and behavioral shifts is unclear.⁶ We aimed to measure the sustainability of improvements made in infection prevention practices after introduction of the Clean Cut program.

This quality improvement study was conducted at 8 hospitals in Ethiopia that completed the Clean Cut program between September 1, 2015, and November 15, 2019. Through surveillance of infection prevention practices, process mapping, training, and action planning, surgical teams improved behaviors in 6 key areas of infection prevention: (1) surgical safety checklist use, (2) skin antisepsis, (3) antibiotic prophylaxis, (4) surgical linen sterility and integrity, (5) instrument sterility, and (6) gauze counting. We conducted a 14-day assessment of perioperative practices through direct observation 6 to 18 months after completion of the program. Seven hospitals were included in the sustainability assessment; the eighth facility had transitioned leadership and did not participate. Data collectors observed all operations in the same operating theaters involved in the initial program during a 14-day period using the same Clean Cut data collection tool. Ethical approval was obtained from the Ethiopian National Institutional Review Board through the Africa Health Research Institute. Informed consent was waived because this study was part of a quality improvement program. Data analysis was performed using Stata, version 15.1 (StataCorp LLC). Compliance was defined as correctly performing 100% of expected behaviors in each of the 6 key areas. A 2-side P = .05 was considered statistically significant.

The study cohort included 3385 patients (2384 [70.4%] female; median age, 27 years [IQR, 20-36 years]), including 738 before implementation of the program, 2178 after implementation, and 469 during the sustainability audit (Table). The case urgency was similar between groups; the sustainability group had slightly fewer comorbidities and included more obstetric and surgical specialty procedures.

Compared with the postimplementation period, the sustainability audit period showed further improvement in compliance with the surgical safety checklist use (1089 [50.0%] vs 274 [58.4%]), skin antisepsis (1254 [57.6%] vs 351 [74.8%%]), antibiotic prophylaxis (1279 [58.7%] vs 308 [65.7%]), and gauze counting (2046 [93.9%] vs 444 [94.7%]). Some attrition in compliance occurred with surgical linen integrity and sterility (1002 [46.0%] vs 183 [39.0%]) and instrument sterility (1192 [54.7%] vs 196 [41.8%]), but performance in these 2 areas remained above baseline (46 [6.2%] and 57 [7.7%], respectively). Multivariate logistic regression confirmed that these findings were not associated with differences in case volume by hospital, age, or procedure type. When performance was measured during more granular periods, all areas had peak performance in the final third of program implementation and a decrease in performance by the sustainability evaluation except antibiotic administration and gauze counting, which continued to improve (Figure).

In this quality improvement study, behavior changes associated with a multimodal surgical quality improvement program were persistent and, in some cases, continued to improve. This study has limitations. Other initiatives focused on surgical quality improvement in Ethiopia may have supported persistent improvement in practices at the 7 hospitals. Furthermore, ensuring linen and instrument sterility requires not only correct behaviors but also the availability of sterility indicators and functioning autoclaves, which rely on biomedical engineering expertise and purchasing decisions at the management level. These elements are not always under the control of surgical teams but are cohesive to the operating room environment.

Despite demonstrable improvement in practices and infectious complications associated with surgical quality improvement programs, the long-term success of most interventions has not been measured, to our knowledge. This study suggests that strategic interventions to improve surgical quality and infection prevention behaviors are worthwhile investments that may provide improved patient safety over time.

Accepted for Publication: July 9, 2021.

Published Online: November 3, 2021. doi:10.1001/jamasurg.2021.5569

Corresponding Author: Nichole Starr, MD, MPH, Department of Surgery, University of California, San Francisco, 505 Parnassus Ave, S-321, San Francisco, CA 94143 (nichole.starr@ucsf.edu).

Author Contributions: Drs Starr and Nofal had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Starr, Gebeyehu, Forrester, Weiser, Mammo.

Acquisition, analysis, or interpretation of data: Starr, Nofal, Derbew, Weiser.

Drafting of the manuscript: Starr, Weiser.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Starr, Nofal, Weiser.

Obtained funding: Nofal, Weiser.

Administrative, technical, or material support: Nofal, Gebeyehu, Forrester, Derbew, Mammo.

Supervision: Starr, Weiser, Mammo.

Conflict of Interest Disclosures: Dr Starr reported receiving grants from the Fogarty International Center during the conduct of the study and being a Lifebox Senior Fellow. Dr Forrester reported being a former fellow at Lifebox Foundation. Dr Derbew reported being a global governance council member at Lifebox Foundation. Dr Weiser reported receiving grants from Lifebox during the conduct of the study and being a consulting medical officer for Lifebox Foundation. Dr Mammo reported being the global clinical lead at Lifebox Foundation. No other disclosures were reported.

Additional Contributions: Senait Bitew, BSN (Lifebox Foundation, Inc, New York, New York); Assefa Tesfaye, MD (St. Peter's Specialized Hospital, Addis Ababa, Ethiopia); Selam Kahsay, BA (Lifebox Foundation, Inc, New York, New York); and Muaad Hussein, MD (Karolinska University, Stockholm, Sweden), provided support for research activities, data cleaning, and program implementation and were not compensated for their help.