Which factors drive the choice of the French-speaking Quebec population towards a COVID-19 vaccination programme: A discrete-choice experiment.

Abstract

The aims of this study were to elicit preferences about the coronavirus disease 2019 (COVID-19) vaccine campaign in the general French-speaking adult Quebec population and to highlight the characteristics of the vaccine campaign that were of major importance. A discrete-choice experiment (DCE) was conducted between April and June 2021, in Quebec, Canada. A quota sampling method by age, gender and educational level was used to achieve a representative sample of the French-speaking adult population. The choice-based exercise was described by seven attributes within a vaccine campaign scenario. A mixed logit (MXL) model and a latent class logit (LCL) model were used to derive utility values. Age, gender, educational level, income and fear of COVID-19 were included as independent variables in the LCL. A total of 1883 respondents were included for analysis, yielding 22,586 choices. From these choices, 3425 (15.16%) were refusals. In addition, 1159 (61.55%) individuals always accepted any of the vaccination campaigns, while 92 individuals (4.89%) always refused vaccine alternatives. According to the MXL, relative weight importance of attributes was effectiveness (32.50%), risk of side effects (24.76%), level of scientific evidence (22.51%), number of shots (15.73%), priority population (3.60%), type of vaccine (0.61%), and vaccination location (0.28%). Four classes were derived from the LCL model and attributes were more or less important according to them. Class 1 (19.8%) was more concerned about the effectiveness (27.99%), safety (24.22%) and the number of shots (21.82%), class 2 (55.3%) wanted a highly effective vaccine (40.16%) and class 3 (17.6%) gave high value to the scientific evidence (42.00%). Class 4 preferences (7.4%) were more balanced, with each attribute having a relative weight ranging from 1.84% (type of vaccine) to 21.32% (risk of side effects). Membership posterior probabilities to latent classes were found to be predicted by individual factors such as gender, annual income or fear of COVID-19. Vaccination acceptance relies on multiple factors. This study allowed assessment of vaccination-specific issues through a choice-based exercise and description of factors influencing this choice by segmenting the sample and drawing profiles of individuals. Moreover, besides effectiveness and safety, a major point of this study was to show the importance given by the general population to the level of scientific evidence surrounding vaccines. A small group of citizens was involved in the conception, design and interpretation of data. Participants of the DCE were all from the general population.