The effect of airtime incentives to improve participation in noncommunicable disease interactive voice response surveys: randomized controlled trials in Colombia and Tanzania

Abstract

Abstract Objective: We assessed whether airtime incentives can improve cooperation and response rates for a noncommunicable disease interactive voice response survey in Colombia and Tanzania. Methods: Participants were randomized to four arms: (a) no incentive; (b) 1X incentive, where X equals to 5000 Colombian Pesos (US$1.35) or 3000 Tanzanian shillings (TSh; US$1.29); (3) 2X incentive; or (4) lottery incentive of 50 000 COP (US$18.90) or 50 000 TSh (US$21.50), where the odds of winning the lottery were 1:20. Adults aged 18 years and older who possessed a functioning mobile phone were sampled using random digit dialling. We analyzed the primary outcomes, cooperation and response rates using a log binomial regression model as computed by the American Association of Public Opinion Research. Results: In Colombia, between October 15 to November 13, 2018, 125 745 phone calls were made. In Tanzania, 67 800 random digit dial phone calls were made from August 9 to 28, 2018. In Colombia, we observed significantly higher cooperation rates in the 1X, 2X and lottery incentive groups compared to control. Additionally, response rates were significantly higher in the 1X and 2X incentive groups but were significantly lower in lottery group compared to control. In Tanzania, both cooperation and response rates were significantly higher in the 1X, 2X and lottery incentive groups compared to control. Conclusion: Except for the lottery incentive arm in Colombia, which yielded a response rate lower than the control, the introduction of airtime incentives significantly improved cooperation and response rates in Colombia and Tanzania, with no notable variations between the incentive arms.