Shifts in routine vaccine confidence during the COVID-19 pandemic in Kinshasa Province, DRC: A mixed-methods approach

Public Deposited
Analytics
No preview available

Download the file

Request Version for Screen Reader
Last modified date
  • January 15, 2024
Creator
  • Boisson-Walsh, Alix
    • ORCID: aboisson
    • Affiliation: School of Medicine
Abstract
  • The COVID-19 pandemic negatively impacted routine immunizations worldwide, decreasing confidence in vaccination programs. Leveraging a previously validated Shift in Vaccine Confidence (SVC) tool, we measured changes in routine HBV vaccine confidence due to the pandemic among Kinshasa-based participants and factors influencing vaccination confidence, uptake, and intention. We verbally administered SVC surveys in person to HBV-negative adults in Kinshasa who were ≥18 years of age, exposed to HBV in the household, and eligible for HBV vaccination. We measured HBV vaccination status in three ways: vaccinee, willing to receive HBV vaccination, or refused. We interviewed participants at one time point, during which they responded to prompts in the framework of (1) before and (2) during the pandemic. To measure shifts in vaccine confidence, we compared vaccination confidence before versus during the pandemic using Chi-square tests. We also coded open-ended responses to explore context-specific vaccine uptake and perception determinants. From April 2022 to February 2023, we administered the SVC tool to a purposive sample of 41 participants: 7 vaccinees, 23 willing to receive HBV vaccine, and 11 refusers. We observed statistically significant declines in opinion across all five vaccine confidence domains when comparing responses during versus before the pandemic (p<0.01). The most significant shift in perspective was that 80.5% of participants believed that vaccines were safe before the pandemic, compared to 46.3% during the pandemic (p<0.01). Qualitative analysis identified four emergent domains impacting uptake decisions: vaccine confidence, knowledge, risks, and external influences. Rising uncertainty about efficacy, safety, and distrust in the COVID-19 vaccine undermined vaccine confidence among our participants. Factors such as distrust in manufacturers and government, fear of side effects, perceived low illness risk, and inconvenient healthcare access contributed to low vaccine uptake. These insights underscore the pandemic's impact on routine immunization and emphasize the need for consideration in future vaccination campaigns.
Methodology
  • Study setting and survey tool This study was an ancillary study to the "Horizontal and Vertical Transmission of HBV" (HOVER-HBV) study, focused on investigating modes of HBV transmission within households in February 2021 in Kinshasa.11 Within the HOVER-HBV study, we observed low uptake of the HBV vaccine when offered free of charge to adults who were HBV-negative and exposed to HBV in the household.11 In response to the decreased uptake, we developed and validated the content of the SVC tool10, a standardized tool to assess changes in routine vaccine confidence due to the COVID-19 pandemic. The survey items fall within four categories: vaccine confidence, vaccine uptake, vaccine intention, and sociodemographic status. The items included in the tool had a three-point rating scale (yes, don't know, no) and open-ended questions. No changes were made to the wording of the items after content validation of the tool; see Boisson et al. for more information.10 Design and population In this cross-sectional mixed-methods study, we employed the SVC tool to survey a convenience sample of adults eligible for HBV vaccination within the parent HOVER-HBV study.10 Eligible participants were HBV-negative, exposed to HBV in the household, and ≥18 years of age. DRC did not introduce the 3-dose infant HBV series to the national immunization schedule until 2009;12 therefore, this adult population was not offered HBV vaccination until our study. We measured HBV vaccination status in three ways: vaccinee, willing to receive HBV vaccine, or refused. Participants were approached by phone for this sub-study if they were offered an HBV vaccine in the parent HOVER study. A purposive sampling approach guided the selection of HOVER participants to approach for enrollment. If individuals consented to participate in the study, a research assistant would travel to the participant’s home at a pre-specified date and time to interview in person. Data collection Two research assistants (JM, SN) with previous experience in conducting interviews in the field and who are native to the region received training on the use of the SVC tool. Interviews were conducted in Lingala (the local language) in a private space within the home and lasted between 45 and 60 minutes. We administered the interviews in two parts. The first was a structured survey format with three response options (yes, I don’t know, no). The second was a semi-structured interview format with open-ended questions. The interviewers received training to probe participants on determinants such as vaccine decision-making, uptake, and perception. Interview responses were captured on a tablet using the REDCap (Research Electronic Data Capture) software. They were simultaneously audio recorded. A bilingual (French and Lingala) third-party service, 'Centre Afro-Linguistique de Traduction, Interprétariat de Kinshasa,' then translated and transcribed the audio recordings of the interviews to French for analysis. Data analysis Quantitative analysis The SVC survey tool included five validated domains of vaccine confidence: vaccines prevent against diseases, vaccines are important for a child’s health, vaccines are important for one’s own health, vaccines are safe, and new vaccines do not pose higher risk than existing vaccine. We interviewed participants at one timepoint during which they responded to interview questions in the framework of (1) before the pandemic and (2) during or currently. We then compared “before” and “during” pandemic responses to measure shifts in vaccine confidence using Pearson’s chi-square test. The responses were coded 0 for “No”, 1 for “I don’t know” and 2 for “Yes”. We reverse-coded one item’s response options (do new vaccines pose more risk than old vaccines?) so that higher scores indicated less confidence on all items. We used Stata/SE 16.1 to conduct all quantitative analyses.
Date of publication
DOI
Kind of data
  • Numeric
Resource type
  • Dataset
License
  • CC0 1.0 Universal
Language
  • English

Relations

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details VHS_variables.dta 2024-01-15 Public Download
file details VHS_variables.dta 2024-01-15 Public Download