Evaluation of Mechanisms to Improve Performance of Mobile Phone Surveys in Low- and Middle-Income Countries: Research Protocol

Dustin G Gibson; George William Pariyo; Adaeze C Wosu; Abigail R Greenleaf; Joseph Ali; Saifuddin Ahmed; Alain B Labrique; Khaleda Islam; Honorati Masanja; Elizeus Rutebemberwa; Adnan A Hyder

doi:10.2196/resprot.7534

Evaluation of Mechanisms to Improve Performance of Mobile Phone Surveys in Low- and Middle-Income Countries: Research Protocol

JMIR Res Protoc. 2017 May 5;6(5):e81. doi: 10.2196/resprot.7534.

Authors

Affiliations

¹ Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States.
² Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States.
³ Department of Population, Family and Reproductive Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States.
⁴ Berman Institute of Bioethics, Johns Hopkins University, Baltimore, MD, United States.
⁵ Institute of Epidemiology, Disease Control and Research, Dhaka, Bangladesh.
⁶ Ifakara Health Institute, Dar es Salaam, United Republic Of Tanzania.
⁷ Makerere University School of Public Health, Makerere University College of Health Science, Kampala, Uganda.

Abstract

Background: Mobile phone ownership and access have increased rapidly across low- and middle-income countries (LMICs) within the last decade. Concomitantly, LMICs are experiencing demographic and epidemiologic transitions, where non-communicable diseases (NCDs) are increasingly becoming leading causes of morbidity and mortality. Mobile phone surveys could aid data collection for prevention and control of these NCDs but limited evidence of their feasibility exists.

Objective: The objective of this paper is to describe a series of sub-studies aimed at optimizing the delivery of interactive voice response (IVR) and computer-assisted telephone interviews (CATI) for NCD risk factor data collection in LMICs. These sub-studies are designed to assess the effect of factors such as airtime incentive timing, amount, and structure, survey introduction characteristics, different sampling frames, and survey modality on key survey metrics, such as survey response, completion, and attrition rates.

Methods: In a series of sub-studies, participants will be randomly assigned to receive different airtime incentive amounts (eg, 10 minutes of airtime versus 20 minutes of airtime), different incentive delivery timings (airtime delivered before survey begins versus delivery upon completion of survey), different survey introductions (informational versus motivational), different narrative voices (male versus female), and different sampling frames (random digit dialing versus mobile network operator-provided numbers) to examine which study arms will yield the highest response and completion rates. Furthermore, response and completion rates and the inter-modal reliability of the IVR and CATI delivery methods will be compared.

Results: Research activities are expected to be completed in Bangladesh, Tanzania, and Uganda in 2017.

Conclusions: This is one of the first studies to examine the feasibility of using IVR and CATI for systematic collection of NCD risk factor information in LMICs. Our findings will inform the future design and implementation of mobile phone surveys in LMICs.

Keywords: Bangladesh; CATI; IVR; Tanzania; Uganda; mHealth; mobile phone survey; noncommunicable diseases; survey methodology.

©Dustin G Gibson, George William Pariyo, Adaeze C Wosu, Abigail R Greenleaf, Joseph Ali, Saifuddin Ahmed, Alain B Labrique, Khaleda Islam, Honorati Masanja, Elizeus Rutebemberwa, Adnan A Hyder. Originally published in JMIR Research Protocols (http://www.researchprotocols.org), 05.05.2017.