Personalized Text Messages and Automated Calls for Improving Vaccine Coverage Among Children in Pakistan: Protocol for a Community-Based Cluster Randomized Clinical Trial

Abstract

Background A major reason for poor childhood vaccine coverage in developing countries is the lack of awareness among parents and caregivers regarding the need for immunization and the importance of completing the entire series of vaccines. Short message service (SMS)–based interventions have been quite effective in different programs such as smoking cessation, treatment adherence, health care scheduled appointment attendance, antenatal care attendance, and compliance to immunization. However, there are limited data from low- and middle-income countries on the role of SMS and automated call–based messages and interventions to improve routine immunization (RI) coverage. Objective The primary objective of this study is to evaluate whether automated mobile phone–based personalized messages (SMS or automated call) can improve RI uptake at 6, 10, and 14 weeks of age per the expanded program immunization schedule, compared with a usual care control group. Secondary objectives include assessing the effects of different types of automated SMS text or calls on RI coverage at 20 weeks of age. Methods This is a mixed methods study using a clustered randomized controlled trial with 4 intervention arms and 1 control arm, augmented by qualitative interviews for personalizing the message. The study is being conducted in Pakistan (an urban site in Karachi and a rural site Matiari). In Karachi, 250 administrative structures are taken as 1 cluster, whereas in Matiari, a catchment area of 4 Lady Health Workers is considered as 1 cluster. The intervention targets families to receive weekly 1-way or 2-way (interactive) personalized automated SMS or automated phone call messages regarding vaccination. Possible barriers to vaccination are assessed in each family at the time of inclusion to determine the type of personalized messages that should be sent to the family to increase the chance of a positive response. Finally, in-depth interviews using purposive sampling are conducted before and after the trial to determine the family’s vaccination experience and related factors. Results All study participants for the cluster randomized trial were enrolled by January 14, 2019. Study exit interviews at 20-weeks follow-up visits will be completed by June 2019. Conclusions The results of this study will be useful to understand the respective effects of SMS text messages versus automated phone–based communication to improve RI coverage and timelines. Moreover, information regarding families’ perceptions of vaccination and the daily life challenges for timely visits to the vaccine clinic will be used for developing more complex interventions that use mobile phone messages and possibly other approaches to overcome barriers in the uptake of correct and timely immunization practices. Trial Registration ClinicalTrials.gov NCT03341195; https://clinicaltrials.gov/ct2/show/NCT03341195 (Archived b...

Texting Parents about Early Child Development: Behavioral Changes and Unintended Social Effects

“…This paper complements a broader literature showing the potential of text messages as nudges for health-related behaviors such as prevention and disease management (Cole-Lewis and Kershaw, 2010), treatment adherence (Pop-Eleches et al, 2011), child eating behavior (Chai et al, 2019), or vaccination (Kazi et al, 2019). It links to evidence of text messages influencing pro-social behaviors (Konrath et al, 2015).…”

An Artificial Intelligence–Based, Personalized Smartphone App to Improve Childhood Immunization Coverage and Timelines Among Children in Pakistan: Protocol for a Randomized Controlled Trial

“…The app will have features and capacity for text, voice, pictorial, and video messages. The content of the messages will be based on the previous study that utilized automated SMS text messages and calls in various local languages to improve RCI coverage [ 25 ].…”

“…AI is being applied in this study in two different manners. First, AI is being applied to the baseline data and RCT results from a previous study, which is based on an automated SMS text message and voice call intervention [ 25 ]. Paigham-e-Sehat is a recently completed study that includes data from 3300 children, in which RCI-related messages were sent to caregivers via SMS text messages or automated calls to improve their children’s vaccine coverage at 6, 10, and 14 weeks of age.…”

“…This system will be tested during this study by validating the system’s prediction ability against the actual outcome of this study. Further, the important predictors and barriers identified in the previous study will be used for extending the work on data obtained from this study [ 25 ] . After the availability of the study results, factors related to the effect of users’ app behavior on the vaccination outcomes, including their preferred modes, timing, and interactivity with the app, will be analyzed through pattern recognition using a mixture of hidden Markov models.…”

Operability, Usefulness, and Task-Technology Fit of an mHealth App for Delivering Primary Health Care Services by Community Health Workers in Underserved Areas of Pakistan and Afghanistan: Qualitative Study

“…They fear a graveyard of poor-quality, unproven apps contributing to a fragmented mHealth landscape [ 22 ]. Some experts recommend assessing the barriers and challenges faced by the targeted end users before designing the app to be able to provide mHealth as a good-fit solution [ 20 , 23 ]. A study attributed the failure of technology to its lack of regard for user requirements and experience [ 24 ].…”