Poster HSR2020: Health System Considerations for Community-Based Implementation of Automated Respiratory Rate Counters to Identify Childhood Pneumonia in 5 Regions of Ethiopia

Presented at 6th Global Symposium on Health Systems Research, November 8-12, 2020 (Dubai-virtually due to COVID-19) Abstract below

Title: Health system considerations for community-based implementation of automated respiratory counters to identify childhood pneumonia in 5 regions of Ethiopia: A qualitative analysis

Introduction

Pneumonia, the leading infectious cause of death among children under 5, is diagnosed by measuring respiratory rates (RR) in low- and middle-income settings. RR counting has limited accuracy and high variability, especially when performed by community health workers. Automated RR counters could improve outcomes of childhood pneumonia in communities. The Acute Respiratory Infection Diagnostic Aids (ARIDA) intervention trained health workers at community and facility levels and implemented 2 automated RR counters in Ethiopia (ChARM and RAD-G). We aimed to identify the practical ‘health system-level’ lessons from the ARIDA implementation and provide considerations for integrating these devices into health systems.

Methods

We conducted key informant interviews (KIIs) with 57 diverse participants in five regions where ARIDA was implemented. Participants represented communities, facilities, trainers, and district management/decision makers.  Data was analyzed in ATLAS.ti using thematic content analysis. Themes were categorized using the Tanahashi bottleneck analysis.

Results

All participants recommended scaling up the ARIDA intervention nationally. Health workers perceived the devices as: time saving, highly acceptable by parents and children, facilitating accurate diagnosis and referrals and decreasing referrals to higher level facilities. Health workers perceived an increased demand for services, reduced numbers of sick children not seeking care and perceived decreased death rates from pneumonia. Participants recommended increasing the numbers of devices and health workers trained with considerations for device maintenance and power supply. Strengthening supply chains for amoxicillin and gentamicin, improving oxygen gas availability, and strengthening referral networks would maximize perceived benefits. While training improved knowledge, more supportive supervision, integration with current guidelines and more days were recommended.  Participants recommended more guidance, budget, planning and structure related to community engagement. Government should lead efforts to scale-up the use of ARIDA devices with inputs from across the health system and support from implementing partners.

Conclusions

Automated RR counters have the potential to improve decentralized childhood pneumonia diagnosis and management within communities. Health system considerations to ensure sustainability of scaling efforts must be prioritized. Further research is needed on the epidemiological impact, changes to pneumonia case management (including health seeing behavior and rational drug use) and the return on investment of intervention.