Population-based coverage survey results following the mass drug administration of azithromycin for the treatment of trachoma in Amhara, Ethiopia

Abstract

Background: Trachoma is the leading infectious cause of blindness worldwide. In communities where the district level prevalence of trachomatous inflammation-follicular among children ages 1-9 years is ≥5%, WHO recommends annual mass drug administration (MDA) of antibiotics with the aim of at least 80% coverage. Population-based post-MDA coverage surveys are essential to understand the effectiveness of MDA programs, yet published reports from trachoma programs are rare.

Methods: In the Amhara region of Ethiopia, a population-based MDA coverage survey was conducted 3 weeks following the 2016 MDA to estimate the zonal prevalence of self-reported drug coverage in all 10 administrative zones. Survey households were selected using a multi-stage cluster random sampling design and all individuals in selected households were presented with a drug sample and asked about taking the drug during the campaign. Zonal estimates were weighted and confidence intervals were calculated using survey procedures. Self-reported drug coverage was then compared with regional reported administrative coverage.

Results: Region-wide, 24,248 individuals were enumerated, of which, 20,942 (86.4%) individuals were present. The regional self-reported antibiotic coverage was 76.8% (95%Confidence Interval (CI):69.3-82.9%) in the population overall and 77.4% (95%CI = 65.7-85.9%) among children ages 1-9 years old. Zonal coverage ranged from 67.8% to 90.2%. Five out of 10 zones achieved a coverage >80%. In all zones, the reported administrative coverage was greater than 90% and was considerably higher than self-reported MDA coverage. Main reasons reported for MDA campaign non-attendance included being physically unable to get to MDA site (22.5%), traveling (20.6%), and not knowing about the campaign (21.0%). MDA refusal was low (2.8%) in this population.

Conclusions: Although self-reported MDA coverage in Amhara was greater than 80% in some zones, programmatic improvements are warranted throughout Amhara to achieve higher coverage. These results will be used to enhance community mobilization and improve training for MDA distributors and supervisors to improve coverage in future MDAs.

Fig 1. Administrative levels and generalized population of Amhara, Ethiopia, 2016.

Fig 2. Geographic distribution of self-reported drug coverage by zone, Amhara, Ethiopia, 2016.

Map created in ArcGIS 10.4.1 (ESRI, Redlands, CA) using shapefile sourced from the GADM database (gadm.org).

Fig 3. Administrative coverage, coverage estimates by self-report, by self-report with proxy responses, and by self-report of head of household by zone, Amhara, Ethiopia, 2016.

Fig 4. Cluster-level distribution of self-report of “campaign did not come to village,” Amhara, Ethiopia, 2016.