Predictors of COVID-19 vaccination among OVC and AGYW in Eswatini

Abstract

Las tasas de vacunación de COVID-19 en Eswatini se han mantenido por debajo del umbral esperado. Comprender los predictores de aceptación de la vacuna COVID-19 es esencial para diferenciar los mensajes dirigidos a su población. Identificamos la prevalencia y predictores de la vacunación contra COVID-19 entre los beneficiarios del Proyecto Insika Ya Kusasa en Eswatini. Estudio de cohorte retrospectivo de datos secundarios. La exposición principal fue la oferta de vacunas contra COVID-19 desde abril 2021 y el resultado principal fue vacunación completa hasta junio 2022. Realizamos análisis univariado para tasas de vacunación por tipo de beneficiario y regresión multivariada ajustada por factores de confusión. En 6 562 beneficiarios, la mediana de edad al final del seguimiento fue 18,2 (RIC: 15,8-22,1), 78,3 % mujeres y 51,1 % vacunados contra COVID-19. De los vacunados, 32,9% tenían entre 15 y 17 años y 35,6% eran OVC. En el modelo final, embarazo y residencia urbana fueron predictores positivos de vacunación contra COVID-19. Cuidadores vacunados contra COVID-19 y situación económica del hogar inestable tuvieron efecto contrario, evidenciando la carga social de la pandemia del COVID-19 en poblaciones vulnerables. Diferentes mensajes dirigidos a esta población específica pueden mejorar la aceptación de la vacuna COVID-19 en Eswatini.

Background: COVID-19 vaccination rates in Eswatini have remained under the expected threshold. Understanding the COVID-19 vaccine uptake predictors is essential to differentiate the messaging targeting its population to meet the national vaccination goals through a massive COVID-19 vaccine campaign. We identified the prevalence and predictors of COVID-19 vaccination among the beneficiaries of the Insika Project in Eswatini. Methods: Retrospective cohort study from secondary data of beneficiaries enrolled in the Insika Project in Eswatini since 2018. The main exposure was the registry of COVID-19 vaccine offer through the Insika vaccination campaign since April 2021 and the main outcome was the complete COVID-19 vaccination until the end of June 2022. We conducted univariate analysis for vaccination rates by type of beneficiary and multivariate regression analysis adjusted for confounders; the final model was built backward stepwise starting with all variables theoretically important and with p-value ≤ 0.05 in bivariate analyses. We present adjusted Odds Ratios (aOR) with 95% confidence intervals (CIs). Results: In 6 562 beneficiaries, the median age at end of follow-up was 18.2 (ICR: 15.8-22.1), 78.3% were female, and 51.1% were vaccinated for COVID-19. Of the vaccinated ones, 32.9% were 15-17 years old, 76.5% were female, 35.6% were OVCs, 40.5% were from Lubombo region and 33.4% were from Bantwana IP. The final model included region, IP, pregnancies, caregivers’ COVID-19 vaccination, household socioeconomic status, GBV cases, receiving protection services, receiving health services, and life mentorship beneficiaries; pregnancies, urban region residents, and Bantwana, Cabrini, and Young Heroes IPs were independently and significantly associated with COVID-19 vaccination. On the opposite, caregivers’ COVID-19 vaccination, caregivers’ self-employment, and receiving protection and life mentorship services were independently and significantly associated with no vaccination. The predictors differ by type of program when the model was segregated for each of them. Conclusion: Among the Insika beneficiaries in Eswatini, pregnancy and urban places of residence were positive predictors of COVID-19 vaccination. Having caregivers vaccinated for COVID-19 and unstable household economic status had the opposite effect, evidencing the social burden of the COVID-19 pandemic in a vulnerable population. Different messaging targeting this specific population may improve COVID-19 vaccine uptake in Eswatini.