Does delivery matter? Examining pandemic vaccination preferences across time and countries using a discrete choice experiment

• Published in: Social science & medicine (1982) 2025-02, Vol.366, p.117637, Article 117637
• Main Authors: Kong, Qingxia, de Vries, Harwin, Poyraz, Dursen Deniz, Kayyal, Abed
• Format: Article
• Language: English
• Subjects: Discrete choice experiment; Last-mile vaccine delivery; Pandemic preparedness; Public vaccination preference
• ISSN: 0277-9536; 1873-5347; 1873-5347
• DOI: 10.1016/j.socscimed.2024.117637

After a pandemic outbreak, designing and implementing an effective vaccine supply chain is key. The last-mile vaccine delivery (LMVD) – the last stage of the vaccine supply chain – is of particular importance. LMVD design choices, such as the delivery mode (e.g., mobile team or PODs) and the number and locations of PODs, affect “operational attributes” like travel time and waiting time. This paper aims to understand how operational attributes of LMVD systems impact vaccination preferences across different priority groups, pandemic stages, and countries. Adults in The Netherlands (N=473 in May 2021 and N=385 in January 2023), Italy (N=302 in March 2023), and Poland (N=303 in March 2023) completed an online discrete-choice experiment. Participants chose between two hypothetical LMVD systems (and opt-out) with five attributes: appointment flexibility, appointment delay, in-facility waiting time, travel time, and familiarity with personnel. Mixed logit models were used to estimate coefficients, based on which relative importance (RI) and vaccine uptake percentages were calculated. Operational attributes have a significant impact on estimated uptake. Travel time was the most important attribute in all four studies. Appointment delay and in-facility waiting consistently ranked second or third. The impact of the choice of an LMVD system on uptake varies over time and by country. In The Netherlands, it was strong in 2023 (23.4%) but moderate in 2021 (12.6%). Similarly, it was strong in Italy 2023 (21.8%) and moderate in Poland 2023 (11.4%). The estimated impact also varied by priority group, and was minor (6.1%) for the elderly with poor health in The Netherlands 2021 study. Design choices for the LMVD system significantly impact vaccination preferences. Adapting these systems to different countries, time periods, and population groups is key for their effectiveness. •First to study the impact of vaccine distribution on vaccination preferences.•Travel time, in-facility waiting, and appointment delay are most important in all studies.•Significant impact of vaccine delivery system over time and across countries.•Minor impacts on elderly/sick in Netherlands 2021.