Feasible Surgical Site Infection Surveillance in Resource-Limited Settings: A Pilot in Sierra Leone

Abstract

Background: Surgical site infections (SSIs) are among the most common healthcare-associated infections (HAIs) in low- and middle-income countries (LMICs). SSI surveillance can be challenging and resource-intensive to implement in LMICs. To support feasible LMIC SSI surveillance, we piloted a multisite SSI surveillance protocol using simplified case definitions and methodology in Sierra Leone. Methods: A standardized evaluation tool was used to assess SSI surveillance knowledge, capacity, and attitudes at 5 proposed facilities. We used simplified case definitions restricted to objective, observable criteria (eg, wound purulence or intentional reopening) without considering the depth of infection. Surveillance was limited to post-cesarean delivery patients to control variability of patient-level infection risk and to decrease data collection requirements. Phone-based patient interviews at 30-days facilitated postdischarge case finding. Surveillance activities utilized existing clinical staff without monetary incentives. The Ministry of Health provided training and support for data management and analysis. Results: Three facilities were selected for initial implementation. At all facilities, administration and surgical staff described most, or all, infections as “preventable” and all considered SSIs an “important problem” at their facility. However, capacity assessments revealed limited staff availability to support surveillance activities, limited experience in systematic data collection, nonstandardized patient records as the basis for data collection, lack of unique and consistent patient identifiers to link patient encounters, and no quality-assured microbiology services. To limit system demands and to maximize usefulness, our surveillance data collection elements were built into a newly developed clinical surgical safety checklist that was designed to support surgeons’ clinical decision making. Following implementation and 2 months of SSI surveillance activities, 77% (392 of 509) of post-cesarean delivery patients had a checklist completed within the surveillance system. Only 145 of 392 patients (37%) under surveillance were contacted for final 30-day phone interview. Combined SSI rate for the initial 2-months of data collection in Sierra Leone was 8% (32 of 392) with 31% (10 of 32) identified through postdischarge case finding. Discussion: The surveillance strategy piloted in Sierra Leone represents a departure from established HAI strategies in the use of simplified case definitions and implementation methods that prioritize current feasibility in a resource-limited setting. However, our pilot implementation results suggest that even these simplified SSI surveillance methods may lack sustainability without additional resources, especially in postdischarge case finding. However, even limited phone-based patient interviews identified a substantial number of infections in this population. Although it was not addressed in this pilot study, feasible laboratory capacity building to support HAI surveillance efforts and promote appropriate treatment should be explored.

Funding: None

Disclosures: None