Community control strategies for scabies: A cluster randomised noninferiority trial

Abstract

Background: Scabies is a neglected tropical disease hyperendemic to many low- and middle-income countries. Scabies can be successfully controlled using mass drug administration (MDA) using 2 doses of ivermectin-based treatment. If effective, a strategy of 1-dose ivermectin-based MDA would have substantial advantages for implementing MDA for scabies at large scale.

Methods and findings: We did a cluster randomised, noninferiority, open-label, 3-group unblinded study comparing the effectiveness of control strategies on community prevalence of scabies at 12 months. All residents from 35 villages on 2 Fijian islands were eligible to participate. Villages were randomised 1:1:1 to 2-dose ivermectin-based MDA (IVM-2), 1-dose ivermectin-based MDA (IVM-1), or screen and treat with topical permethrin 5% for individuals with scabies and their household contacts (SAT). All groups also received diethylcarbamazine and albendazole for lymphatic filariasis control. For IVM-2 and IVM-1, oral ivermectin was dosed at 200 μg/kg and when contraindicated substituted with permethrin. We designated a noninferiority margin of 5%. We enrolled 3,812 participants at baseline (July to November 2017) from the 35 villages with median village size of 108 (range 18 to 298). Age and sex of participants were representative of the population with 51.6% male and median age of 25 years (interquartile range 10 to 47). We enrolled 3,898 at 12 months (July to November 2018). At baseline, scabies prevalence was similar in all groups: IVM-2: 11.7% (95% confidence interval (CI) 8.5 to 16.0); IVM-1: 15.2% (95% CI 9.4 to 23.8); SAT: 13.6% (95% CI 7.9 to 22.4). At 12 months, scabies decreased substantially in all groups: IVM-2: 1.3% (95% CI 0.6 to 2.5); IVM-1: 2.7% (95% CI 1.1 to 6.5); SAT: 1.1% (95% CI 0.6 to 2.0). The risk difference in scabies prevalence at 12 months between the IVM-1 and IVM-2 groups was 1.2% (95% CI -0.2 to 2.7, p = 0.10). Limitations of the study included the method of scabies diagnosis by nonexperts, a lower baseline prevalence than anticipated, and the addition of diethylcarbamazine and albendazole to scabies treatment.

Conclusions: All 3 strategies substantially reduced prevalence. One-dose was noninferior to 2-dose ivermectin-based MDA, as was a screen and treat approach, for community control of scabies. Further trials comparing these approaches in varied settings are warranted to inform global scabies control strategies.

Trial registration: Clinitrials.gov NCT03177993 and ANZCTR N12617000738325.

Fig 1. Map of study sites, village locations, and treatment allocation.

IVM-1, one-dose ivermectin-based MDA; IVM-2, two-dose ivermectin-based MDA; MDA, mass drug administration; SAT, screen and treat with 1-dose permethrin to index cases of scabies and their household contacts. Made with reference to Natural Earth (naturalearthdata.com) and The Pacific Community (SPC) Statistics for Development Division (pacificdata.org/data/dataset/2017_fji_phc_admin_boundaries).

Fig 2. Trial profile detailing village cluster randomisation, enrolment, and treatment at baseline and enrolment at 12-month follow-up.

IVM-1, one-dose ivermectin-based MDA; IVM-2, two-dose ivermectin-based MDA; MDA, mass drug administration; SAT, screen and treat with 1-dose permethrin to index cases of scabies and their household contacts. ^aLess than 5 years old and weight equal to or greater than 15 kilograms. ^bTreatment violation due to misclassification of participant’s resident village.

Fig 3. Scabies prevalence risk difference and 95% CI between any 2 groups at 12 months.

CI, confidence interval; IVM-1, one-dose ivermectin-based MDA; IVM-2, two-dose ivermectin-based MDA; MDA, mass drug administration; SAT, screen and treat with 1-dose permethrin to index cases of scabies and their household contacts. Whiskers represent 95% CI around risk difference. ^aReference treatment group.