Targeted Mass Azithromycin Distribution for Trachoma: A Community-Randomized Trial (TANA II)

Abstract

Background: Current guidelines recommend annual community-wide mass administration of azithromycin for trachoma. Targeting treatments to those most likely to be infected could reduce the amount of unnecessary antibiotics distributed.

Methods: In a cluster-randomized trial conducted from 1 November 2010 through 8 November 2013, 48 Ethiopian communities previously treated with annual mass azithromycin distributions for trachoma were randomized in equal numbers to (1) annual azithromycin distributions targeted to children aged 0-5 years, (2) annual azithromycin distributions targeted to households with a child aged 0-5 years found to have clinically active trachoma, (3) continued annual mass azithromycin distributions to the entire community, or (4) cessation of treatment. The primary outcome was the community prevalence of ocular chlamydia infection among children aged 0-9 years at month 36. Laboratory personnel were masked to treatment allocation.

Results: The prevalence of ocular chlamydia infection among children aged 0-9 years increased from 4.3% (95% confidence interval [CI], .9%-8.6%) at baseline to 8.7% (95% CI, 4.2%-13.9%) at month 36 in the age-targeted arm, and from 2.8% (95% CI, .8%-5.3%) at baseline to 6.3% (95% CI, 2.9%-10.6%) at month 36 in the household-targeted arm. After adjusting for baseline chlamydia prevalence, the 36-month prevalence of ocular chlamydia was 2.4 percentage points greater in the age-targeted group (95% CI, -4.8% to 9.6%; P = .50; prespecified primary analysis). No adverse events were reported.

Conclusions: Targeting azithromycin treatment to preschool children was no different than targeting azithromycin to households with a child with clinically active trachoma. Neither approach reduced ocular chlamydia over the 3-year study.

Clinical trials registration: NCT01202331.

Keywords: chlamydia; clinical trial; mass drug administration; trachoma.

Figure 1.

Trial profile. Both the original TANA I randomization and TANA II randomization are shown. Separate cross-sectional stratified random samples were selected for trachoma monitoring in each community, with stratification according to 2 age groups: children (0–9 years) and adults (≥10 years). Abbreviations: SD, standard deviation; TANA, Trachoma Amelioration in Northern Amhara trial.

Figure 2.

Longitudinal prevalence of ocular chlamydia infection and clinical trachoma in the 2 targeted treatment arms among children aged 0–9 years. Each thin line represents the prevalence of ocular chlamydia (A) or trachomatous inflammation–follicular (TF) (B) in a specific community over the 4 annual study visits; the thick line is the mean prevalence in the treatment arm.

Figure 3.

Mean longitudinal prevalence of ocular chlamydia infection and clinical trachoma in 4 treatment arms among children aged 0–9 years. Each line represents the mean prevalence of ocular chlamydia (A) or trachomatous inflammation–follicular (B) in a treatment arm over the 4 annual study visits, and the bar represents the corresponding 95% confidence interval. Abbreviations: MDA, mass drug administration; TF, trachomatous inflammation–follicular.