Social group and health-care provider interventions to increase the demand for malaria rapid diagnostic tests among community members in Ebonyi state, Nigeria: a cluster-randomised controlled trial

Abstract

Background The rate of diagnostic testing for malaria is still very low in Nigeria despite the scale-up of malaria rapid diagnostic test (MRDT) availability, following WHO's recommendation of universal diagnostic testing in 2010. We investigated whether a social group sensitisation and education intervention (social group intervention) and a social group intervention plus health-care provider training intervention would increase the demand (use or request, or both) for MRDTs among community members in Ebonyi state, Nigeria. MethodsWe did a three-arm, parallel, open-label, stratified cluster-randomised controlled trial in Ebonyi state, Nigeria, to evaluate the effects of two interventions compared with a control. We randomly assigned geographical clusters that were accessible (close to a road that was drivable even during the rainy seasons) and had at least one eligible public primary health facility and patent medicine vendor (those that offered MRDT services) in a 1:1:1 allocation to the control arm (receiving no intervention), social group arm (receiving sensitisation and education about MRDT), or social group plus provider arm (receiving the social group intervention plus provider training in health communication about MRDT). Investigators, participants (social groups, providers, respondents), and interviewers could not be masked to group assignments. The primary outcome was the proportion of children younger than 5 years with fever or malaria-like illness, in the 2 weeks preceding a household survey, who received an MRDT, and the coprimary outcome was the same outcome but among children aged 5 years and older (ie, up to and including 17 years) and adults (excluding pregnant women). The outcomes were measured at an individual level via household surveys before the interventions and 3 months after the end of the interventions. All analyses were done using a cluster-level method on an intention-to-treat basis. This trial is registered with ISRCTN, number ISRCTN14046444. FindingsWe carried out eligibility screening and recruitment of participants (clusters, social groups, and providers) between July 2 and Sept 27, 2018. 34 clusters met the eligibility criteria and 18 were randomly selected to participate and randomly assigned to arms (six clusters per arm). A mean proportion of 40•6% (SD 14•5) of eligible children younger than 5 years in the control arm received an MRDT, versus 66•7% (11•7) in the social group arm (adjusted risk difference [aRD] 28•8%, 95% CI 21•9-35•7, p<0•0001) and 71•7% (19•8) in the social group plus provider arm (aRD 32•7%, 24•9-40•5, p<0•0001), with no significant difference between the social group arm and the social group plus provider arm. A mean proportion of 36•3% (18•5) of eligible children aged 5 years and older in the control arm received an MRDT, versus 60•7% (14•0) in the social group arm (aRD 25•6%, 16•8-34•4, p=0•0004), and 59•5% (18•3) in the social group plus provider arm (aRD 28•0%, 19•5-36•5, p=0•0002), with no significant difference between the social group arm and the soci...

Improving malaria case management with artemisinin-based combination therapies and malaria rapid diagnostic tests in private medicine retail outlets in sub-Saharan Africa: a systematic review

“…Reflecting very low routine RDT provision by PMRs in most settings, nine studies involved introduction of RDTs to PMRs where they were not previously used, while only 4 explored interventions to strengthen existing RDT use by PMRs [45,48,49,51]. The intervention package also involved ACT subsidies in 6 studies [47,48,52,[54][55][56], while in 5 of the remaining 7, AMFm or CPM subsidised ACTs were available in the market [44-46, 50, 53].…”

“…The intervention package also involved ACT subsidies in 6 studies [47,48,52,[54][55][56], while in 5 of the remaining 7, AMFm or CPM subsidised ACTs were available in the market [44-46, 50, 53]. Eight studies tested an intervention package with PMR staff conducting the RDTs [44][45][46][47][48][49][50][51]; while in 5 studies the RDTs were conducted by research team members stationed at the PMR [52][53][54][55][56]. Of the latter group 3 studies generally had relatively short durations of 4 months or less (S5 Table ).…”

“…Five studies included communications activities: Ansah et al, Mbonye et al and Soniran et al conducted community meetings [44,47,51]; Omale et al tested sensitisation and education of women's and men's social groups [49]; and Onwujekwe et al included a school intervention [50]. In 5 studies RDTs were provided directly to PMRs by the study team [44,47,48,50,51], in 2 studies PMRs purchased RDTs through designated wholesalers [45,46], and in one study RDTs were obtained through regular distribution channels [49]. In one study RDTs were provided free to users [44]; in most others RDTs were subsidised, except for Omale et al and one intervention arm in Maloney et al and Dieci et al where no subsidy was provided [46,48,49].…”

“…In 5 studies RDTs were provided directly to PMRs by the study team [44,47,48,50,51], in 2 studies PMRs purchased RDTs through designated wholesalers [45,46], and in one study RDTs were obtained through regular distribution channels [49]. In one study RDTs were provided free to users [44]; in most others RDTs were subsidised, except for Omale et al and one intervention arm in Maloney et al and Dieci et al where no subsidy was provided [46,48,49]. The number of providers included varied from 7 [44] to 262 [46], and the duration in months from 3 [49] to 17 [44].…”

“…In one study RDTs were provided free to users [44]; in most others RDTs were subsidised, except for Omale et al and one intervention arm in Maloney et al and Dieci et al where no subsidy was provided [46,48,49]. The number of providers included varied from 7 [44] to 262 [46], and the duration in months from 3 [49] to 17 [44]. All the studies were conducted in rural areas, while 3 also included urban areas, and all in situations of moderate or high malaria transmission, except Dieci et al (mainly low transmission) [48].…”