Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Nov 26, 2023
Date Accepted: Jun 12, 2024
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Effect of Differential Incentives on Response Rate Using an mHealth Platform Among Those in HIV Care in Lusaka: A pilot Study
ABSTRACT
Background:
The ubiquity of mobile phones, especially in resource limited settings, offers a wide range of options for public health service delivery interventions and measurement. However, there are limitations to mobile health (mHealth) applications including a relatively large proportion of basic mobile devices (i.e., not smart phones) and frequent switching of phone numbers which may limit utility. We developed a survey instrument leveraging the unstructured supplementary service data (USSD) interface that runs on all mobile phones to systematically capture the participant experience at a recent clinic visit among participants receiving HIV care services. We offered varied financial incentives, in the form of mobile network credit, for survey participation and analyzed response rates by incentive level to determine optimal incentives to maximize survey participation within feasible financial limits.
Objective:
Assess the impact of incentives on survey response success for a health experience survey at two government supported HIV care facilities in Lusaka, Zambia.
Methods:
Voluntary survey enrollments were conducted among people exiting two study clinics, one peri-urban and one rural. Two rounds of survey invitations were sent to those enrolled in the study. The first invitation was distributed between 1 November 2018 and 25 June 2019 with the invitation to the second survey round following three months after the first (1 February 2019 to 23 September 2019). Three incentive levels were randomly assigned by participant and survey round 1) no incentive, 2) two Zambian Kwacha (ZMW) (~0.16 USD), and 3) five ZMW (~0.42 USD). We assessed the association between incentive and response to the survey using a mixed effects Poisson regression model allowing random effects at the individual and facility level. Probability plots for survey completion were generated based on language, incentive level, and survey round. We projected the cost per additional response for different incentive levels.
Results:
A total of 996 participants were enrolled of which 72.1% occurred in the urban HIV care facility and 62.2% requested survey be sent in English. We had a total of 1,992 potential survey responses for both rounds. We found an overall completion rate across both surveys at 32.1%, with significantly different response rates between the first (40.5%, 95% confidence interval [CI]: 37.4, 43.6%) and second (23.7%, 95% CI: 21.1, 26.4%) round surveys. Implementing a 5 ZMW incentive significantly increased adjusted prevalence ratio (aPR) for survey completion compared to those that received no incentive (aPR: 1.35; 95% CI: (1.11, 1.63). The cost per additional response highest at 5 ZMW (72.8 ZMW).
Conclusions:
The study findings emphasize the significance of financial incentives in survey participation. Implementing a moderate incentive, effectively increased response rates. Tailoring survey strategies to accommodate language preferences and providing appropriate incentives can optimize response rates and improve data quality. Future research should explore methods to maintain and improve response rates in longitudinal surveys beyond the initial engagement and investigate alternative incentive levels and innovative approaches to further enhance participation. Clinical Trial: The study is registered with Pan-African Clinical Trials Registry (PACTR202101847907585).
Citation
