Trimodal skin health programme for childhood impetigo control in remote Western Australia (SToP): a cluster randomised, stepped-wedge trial.

BACKGROUND

Skin infections affect physical health and, through stigma, social-emotional health. When untreated, they can cause life-threatening conditions. We aimed to assess the effect of a holistic, co-designed, region-wide skin control programme on the prevalence of impetigo.

METHODS

The SToP (See, Treat, and Prevent Skin Sores and Scabies) trial is a pragmatic, open-cohort, stepped-wedge cluster randomised trial involving participants aged 0-18 years in nine remote communities of the Kimberley, Western Australia. The trial involves programmatic interventions in three domains: See (skin checks and skin infection recognition training), Treat (skin infection treatment training, sulfamethoxazole-trimethoprim for impetigo, and ivermectin for scabies), and Prevent (co-designed health promotion and environmental health). Four clusters, defined as pragmatic aggregations of communities, were randomised in two steps to progressively receive the activities during ten visits. The primary outcome was the proportion of school-aged children (aged 5-9 years) with impetigo. We adopted an intention-to-treat analysis and compared the intervention with the control (usual care before the start of intervention) states to derive a time and cluster averaged effect using Bayesian modelling. This study is registered with Australian New Zealand Clinical Trials Registry, ACTRN12618000520235.

FINDINGS: Between Sept 19, 2018, and Nov 22, 2022, 915 children were consented and 777 (85%) had skin checks performed on at least one of ten possible visits between May 5, 2019, and Nov 22, 2022. Of the participants, 448 (58%) of 777 were aged 5-9 years at one or more of the visit timepoints and were eligible for primary outcome assessment. A decline in impetigo occurred across all clusters, with the greatest decline during the observational period of baseline skin checks before commencement of the interventional trial activities activities. The mean (95% credible interval) for the conditional posterior odds ratio for observing impetigo in the intervention compared with the control period was 1·13 (0·71-1·70). The probability that the intervention reduced the odds of observing impetigo was 0·33.

INTERPRETATION: A decreased prevalence of impetigo during the observational period before the commencement of trial activities was sustained across the trial, attributable to the trimodal skin health initiative. Although the prevalence of impetigo reduced, there is no direct evidence to attribute this to the individual effects of the trial activities. The wholistic approach inclusive of skin checks collectively contributed to the sustained reduction in impetigo.

FUNDING: Western Australia Department of Health, Australian National Health and Medical Research Council, and Healthway.