Assessing the Response Results of an mHealth-Based Patient Experience Survey Among People Receiving HIV Care in Lusaka, Zambia: Cohort Study

Abstract

Background: This pilot study evaluates the effectiveness of mobile talk-time incentives in maintaining participation in a longitudinal mobile health (mHealth) data collection program among people living with HIV in Lusaka, Zambia. While mHealth tools, such as mobile phone surveys, provide vital health feedback, optimal incentive strategies to ensure long-term engagement remain limited. This study explores how different incentive levels affect response rates in multiple survey rounds, providing insights into effective methods for encouraging ongoing participation, especially in the context of Zambia's prepaid mobile system and multi-SIM usage, a common practice in sub-Saharan Africa.

Objective: This study aimed to assess the response rate success across multiple invitations to participate in a care experience survey using a mobile phone short codes and unstructured supplementary service data (USSD) model among individuals in an HIV care setting in the Lusaka, Zambia.

Methods: Participants were recruited from 2 study clinics-1 in a periurban setting and 1 in an urban setting. A total of 2 rounds of survey invitations were sent to study participants on a 3-month interval between November 1, 2018, and September 23, 2019. Overall, 3 incentive levels were randomly assigned by participant and survey round: (1) no incentive, (2) 2 Zambian Kwacha (ZMW; US $0.16), and (3) 5 ZMW (US $0.42). Survey response rates were analyzed using mixed-effects Poisson regression, adjusting for individual- and facility-level factors. 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 1006 participants were enrolled, with 72.3% (727/1006) from the urban HIV care facility and 62.4% (628/1006) requesting the survey in English. We sent a total of 1992 survey invitations for both rounds. Overall, survey completion across both surveys was 32.1% (637/1992), with significantly different survey completion between the first (40.5%, 95% CI 37.4-43.6%) and second (23.7%, 95% CI 21.1-26.4) invitations. Implementing a 5 ZMW (US $0.42) incentive significantly increased the adjusted prevalence ratio (aPR) for survey completion compared with those that received no incentive (aPR 1.35, 95% CI 1.11-1.63). The cost per additional response was highest at 5 ZMW, equivalent to US $0.42 (72.8 ZMW [US $5.82] per 1% increase in response).

Conclusions: We observed a sharp decline of almost 50% in survey completion success from the initial invitation to follow-up survey administered 3 months later. This substantial decrease suggests that longitudinal data collection potential for a care experience survey may be limited without additional sensitization and, potentially, added survey reminders. Implementing a moderate incentive increased response rates to our health care experience survey. Tailoring survey strategies to accommodate language preferences and providing moderate incentives can optimize response rates in Zambia.

Trial registration: Pan African Clinical Trial Registry PACTR202101847907585; https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=14613.

Keywords: HIV; HIV care; USSD; Zambia; airtime; incentives; longitudinal; mHealth; mobile; mobile health; mobile phone; patient feedback; pilot study; public health service; regression model; rural; survey; unstructured supplementary service data; urban.

Figure 1

Participant response outcomes flow diagram by incentive level and survey round. Round 1 invitation to participate occurred within 14 days of enrollment in the study and round 2 occurred 3 months after round 1. ZMW: Zambian Kwacha.

Figure 2

(A) Marginal probability of completing survey by language and survey round. (B) Marginal probability of completing the survey by incentive level and survey round. Round 1 occurred within 7 days of enrollment and round 2 occurred 3 months after the initial survey invitation of round 1. ZMW: Zambian Kwacha.

Figure 3

Marginal probability of responding to the survey by incentive level and language. ZMW: Zambian Kwacha.