Skip to main content
SearchLoginLogin or Signup

Review 2: "Evaluation of Four Interventions using Behavioural Economics Insights to Increase Demand for Voluntary Medical Male Circumcision in South Africa through the MoyaApp: A Quasi-Experimental Study"

Overall, reviewers had substantial concerns about this preprint ranging from the accuracy of the conclusions drawn from the stated results to concerns about ambiguous phrasing in the methods section.

Published onMar 19, 2024
Review 2: "Evaluation of Four Interventions using Behavioural Economics Insights to Increase Demand for Voluntary Medical Male Circumcision in South Africa through the MoyaApp: A Quasi-Experimental Study"
1 of 2
key-enterThis Pub is a Review of
Evaluation of four interventions using behavioural economics insights to increase demand for voluntary medical male circumcision in South Africa through the MoyaApp: A quasi-experimental study
Evaluation of four interventions using behavioural economics insights to increase demand for voluntary medical male circumcision in South Africa through the MoyaApp: A quasi-experimental study
Description

Abstract Background While voluntary medical male circumcision (VMMC) reduces the risk of HIV transmission by 60%, circumcision coverage falls short of the UNAIDS 90% target. We investigated whether behaviourally informed message framing increased demand for VMMC.Setting Adult users of the MoyaApp, a data free application in South Africa, who viewed a form designed to generate interest in VMMC during August-November 2022.Methods A quasi-experimental study was conducted to evaluate four MoyaApp VMMC intervention forms against the Standard of Care (SOC) form. All forms enabled users to provide contact details for follow-up engagement by a call centre. The primary outcome was the proportion of forms submitted. Secondary outcomes included successful contact with the user, VMMC bookings/referrals and confirmed circumcision. Multivariable ordinary least-squares regression was used for the analysis.Results MoyaApp VMMC form viewers totalled 118,337 of which 6% submitted a form. Foot-in-the-Door form viewers were more likely (+1.3 percentage points, p<0.01) to submit a form compared to the SOC group (6.3%). Active Choice (-1.1 percentage points, p<0.01) and Reserved for You (-0.05 percentage points, p<0.05) form viewers were less likely to submit a form compared to SOC. Users submitting on Foot-in-the-Door were less likely to be booked/referred compared to SOC (-5 percentage points, p<0.05). There were no differences between the intervention and SOC forms for successful contact and circumcisions.Conclusions Message framing using behavioural insights was able to nudge men to engage with VMMC services. However, more work is needed to understand how to convert initial interest into bookings and circumcisions.Trial registration South African Clinical Trials Registry DOH-27-062022-7811Pan-African Clinical Trials Registry PACTR202112699416418

RR:C19 Evidence Scale rating by reviewer:

  • Misleading. Serious flaws and errors in the methods and data render the study conclusions misinformative. The results and conclusions of the ideal study are at least as likely to conclude the opposite of its results and conclusions than agree. Decision-makers should not consider this evidence in any decision.

***************************************

Review: The main claim can be summarized as: message framing using behavioral insights was able to nudge men to engage with VMMC services.

This is essentially a study with null results. Two out of the three messaging options piloted decreased form submissions while one increased it by a non-clinically impactful 1.3% with no impact on successful circumcisions. Yet the authors state: “Message framing using behavioral insights was able to nudge men to engage with VMMC services.” This does not seem supported by the evidence presented, particularly the very small impact in one intervention which is not clinically different from a null result. I suggest greatly tempering the conclusions and highlighting the limitations of the study design, which are non-randomized. I think the main finding is that mobile messaging is not successful at increasing VMMC and other interventions should be evaluated.

The proportions presented in the results section are misleading as the denominator continues to change. All outcomes should be presented with the denominator of form viewers. Yet the % contacted and booked referred are presented out of those who submitted a form and the circumcised (48/118k) are presented as 1.2% of the sub-sample—which is not defined.

The discussion does not present this as a null finding. No mention is made of the similar circumcision rates in all groups but rather a large focus is placed on foot-in-door traffic, which is deemed an effective intervention.

Comments
0
comment
No comments here
Why not start the discussion?