Dissemination and implementation of SMS mobile technology to improve child vaccinations in Guatemala
BACKGROUND: Despite immunizations being one of the most successful and cost-effective public health interventions, vaccine coverage in low and middle income countries (LMICs) has remained a significant challenge with nearly 20 million infants underimmunized and at risk for unnecessary morbidity and mortality. Patient reminder systems have proven to be an important mechanism for improving childhood vaccination coverage and can be adapted to mobile health technologies, such as Short Message Service (SMS) texts, which have tremendous and untapped potential for disease management in LMICs where the majority of mobile subscribers reside.
GAP: Despite the positive results of many research studies, the translation of research findings to clinical practice and policy has been historically poor, and there remains little understanding of the barriers and facilitators to program adoption and more widespread dissemination.
HYPOTHESIS: This study adds a mixed methods design to complement a concurrent randomized controlled pilot trial evaluating the effect of SMS reminders on increasing immunization timeliness and coverage for the infant vaccination series in a rural and urban site in Guatemala. Understanding the real-world context in which an SMS immunization reminder system is implemented will be imperative for successful program adoption, replication, and scale-up.
METHODS: This qualitative study aimed to evaluate the SMS immunization reminder system in terms of perceived facilitators and barriers to the program through (1) focus group (FG) discussions with participating parents in both the rural and urban study settings and (2) key informant interviews (KIIs) with at least two representatives from each of the following stakeholder groups in both the rural and urban study populations: nurses and physicians, clinic administrators, Information Technology Systems directors, and Ministry of Health personnel. Interview data were de-identified and imported into NVivo software. One analyst coded all transcripts in Spanish to preserve cultural nuances. A second analyst randomly selected 10% of the translated transcripts in English to code independently and assess inter-rater reliability. The analysts met to discuss disagreements, determine shared protocol for coding, and then recoded segments where there were disagreements on coding. After this process, inter-rater reliability was evaluated with a mean Kappa statistic of 0.72. Identified themes were then organized into an overall framework describing the barriers and facilitators of the SMS reminder system.
RESULTS: Nine FGs and 14 KIIs were conducted between February – September 2017. Thematic analysis revealed perceived facilitators and barriers of the intervention. For the FGs, facilitators included the general acceptability of text messaging for vaccine reminders, the perception that these messages are helpful, and the opportunity to report and receive counseling regarding adverse reactions via bidirectional text messaging. Barriers included a general dissatisfaction with the public health system and the common experience of adverse reactions to vaccines. For the KIIs, facilitators included an overwhelming consensus that text message reminders would be both acceptable and helpful at the clinic and that timely vaccination is important. Barriers included children not getting vaccinated on time due to structural barriers such as vaccine shortages commonly experience by the clinics.
IMPACT: Developing, implementing, and disseminating an SMS system that is generalizable, scalable, and effectively increases childhood immunization rates in resource-limited settings would have significant and timely implications towards improving worldwide vaccine coverage and ultimately reducing childhood morbidity and mortality. This study highlights the potential for SMS vaccine reminders in a LMIC and describes facilitators and barriers identified from participants to improve further program dissemination and implementation.