Abstract
Purpose: :
Within the Partnership for Rapid Elimination of Trachoma (PRET), we examined guardian and program risk factors for households with children who do not participate in two rounds of mass treatment compared to households where all children participated at both rounds.
Methods: :
A case control study was conducted in 28 Tanzanian villages within the Kongwa district from July - October 2009. Cases were households with at least one child, less than 10 years, who did not participate in two rounds of treatment. Controls consisted of households where all children, under age 10, participated in both treatment rounds. Household and community risk factors were assessed through a detailed risk factor questionnaire and information from the PRET census, community treatment assistant (CTA) survey, and mass treatment log book. All 152 case households and a random sample of 460 control households in PRET were contacted to participate in the risk factor survey. Population-averaged logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for risk factors and control for clustering. All ORs were adjusted for the following confounders: family size and village size at baseline census.
Results: :
In total, 140 case households and 456 control households completed the questionnaire. Compared to controls, guardians in case households were younger (OR 1.02 (95% CI 1.00-1.04)), held fewer social network indicators (OR 1.35 (95% CI 1.00-1.83), had good health status (OR 2.83 (95% CI 1.37-5.85)), lived in a village with fewer CTAs ((OR 1.69 (95% CI 1.36-2.08)), and where the distribution time was two (versus five) days (OR 3.64 (95% CI 2.30-5.71)).Furthermore, case households were more likely than controls to describe a family health burden (OR 1.60 (95% CI 1.11-2.31)). These associations persisted after controlling for confounders. There was no association between guardian’s education, perceived trachoma risk, perceived CTA performance, or traditional medicine beliefs and being a case household, following adjustments for confounders.
Conclusions: :
Both guardian and program characteristics are important factors in child non-participation over multiple treatment rounds.
Keywords: clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology • trachoma • clinical (human) or epidemiologic studies: risk factor assessment