Intention to receive new vaccines post-COVID-19 pandemic among adults and health workers in Lusaka, Zambia

To estimate intention to receive newly introduced adult vaccines among community members and healthcare workers (HCWs) in Lusaka, Zambia in the context of previous COVID-19 vaccine uptake and perceived disease threat and, identify trusted sources of vaccine information. We conducted a cross-sectional survey among a random sample of community members and a convenience sample of HCWs from 13 November to 15 December 2023. We evaluated future vaccination intentions by self-reported COVID-19 vaccine uptake, community role, vaccine type (COVID-19 booster, HIV, tuberculosis, malaria, pneumonia, diarrheal disease), and source of information using adjusted, mixed effects Poisson regression and adjusted probability models. We enrolled 395 (79.2 %) community members and 104 (20.8 %) HCWs (N = 499). There was high intention to receive new vaccines among community members (mean score = 83.6 %) andHCWs (mean score = 86.0 %), though intentions varied by vaccine type. Prior COVID-19 vaccine uptake (0, 1, 2+ doses) impacted intentions to receive a novel COVID-19 vaccine among community members (43.3 %,62.8 %, and 79.7 %, respectively) but were not associated with any other vaccine types. Intention to receive a vaccine was strongly associated with perceived disease severity and susceptibility as well as age, sex, education, and household income. Social media as a vaccine information source was associated with lower overall vaccine intention among community members, while health system and community sources were associated with higher overall intention to receive new vaccines. The government was a highly trusted source of vaccine information among all participants. Prior COVID-19 vaccination uptake did not predict future non-COVID-19 vaccine intention in Zambia. Perceived threat and select socio-demographic factors were key predictors, suggesting the need for rapid research to design communication strategies and identify trusted sources per target population.

Methods
Study design

We conducted a cross-sectional survey in Lusaka, Zambia, recruiting community members from randomly selected households in four urban communities with relatively low COVID-19 vaccine uptake and healthcare workers (HCWs) at ten healthcare facilities chosen for diversity in size and geographic location. All participants were recruited from November to December 2023, long after the major COVID-19 waves in southern Africa, including the 2021 Omicron wave that was first reported in South Africa and Botswana.

Participants and recruitment strategy

Adults aged 18 years and above who resided in Lusaka Province and could provide consent were eligible to participate in the study. For HCWs, eligibility was restricted to those providing direct clinical services to patients or community-based health services.

We purposively selected 10 healthcare facilities to recruit HCWs and four high-density, low-income areas (compounds) with low COVID-19 vaccination rates to recruit community members. This targeted selection aimed to improve the representation of unvaccinated individuals. By focusing on areas with lower uptake, we sought to identify early signals where pre-emptive engagement strategies may be needed and to understand acceptance in settings that might represent more challenging scenarios for vaccine introduction.

Healthcare facilities included one University tertiary/specialist care teaching hospital, five first-level hospitals providing non-emergency in- and out-patient care, as well as three Urban Health Centres and one Rural Health Centres offering primary healthcare. Some of these health facilities served as COVID-19 referral treatment centres during the pandemic. We collected data from a convenience sample of HCWs drawn from five (5) distinct departments in 10 public health facilities including Antiretroviral Therapy and Treatment (ART), Mother and Child Health (MCH), TB, Adolescent, and Outpatient (OPD) Departments.

Each of the four communities has 10 administrative units (zones), which are divided into 10 neighborhoods of approximately 100 households. Each neighborhood has a Neighborhood Health Committee (NHC) consisting of community members who liaise with the healthcare facility. Per community, we randomly selected one zone and sensitized the participating health facility and 10 NHCs to the study. We employed systematic random sampling to select households from which to draw individuals for data collection.

Two study teams, each consisting of a field supervisor and four trained Research Assistants (RAs) accompanied by an NHC member and led by a study coordinator, collected data. Teams enrolled from 10 households per neighborhood and visited two neighborhoods each day. In each neighborhood, the team used the randomizing technique, "spin the bottle", to determine the initial direction and household and then approached every fifth household to the right. At each selected household, the RA and accompanying NHC member introduced the study to all household members, assessed interested household members for eligibility, and enrolled the first eligible member per household.