Electronic nudges to increase influenza vaccination uptake in Denmark: a nationwide, pragmatic, registry-based, randomised implementation trial

Background: Influenza vaccination rates remain suboptimal despite effectiveness in preventing influenza infection and related complications. We investigated whether behavioural nudges, delivered via a governmental electronic letter system, would increase influenza vaccination uptake among older adults in Denmark.

Methods: We did a nationwide, pragmatic, registry-based, cluster-randomised implementation trial during the 2022-23 influenza season in Denmark. All Danish citizens aged 65 years or older or turning 65 years by Jan 15, 2023 were included. We excluded individuals living in nursing homes and individuals who had an exemption from the Danish mandatory governmental electronic letter system. Households were randomly assigned (9:1:1:1:1:1:1:1:1:1) to usual care or nine different electronic letters designed on the basis of different behavioural nudging concepts. Data were sourced from nationwide Danish administrative health registries. The primary endpoint was receipt of influenza vaccination on or before Jan 1, 2023. The primary analysis assessed an analytical set of one randomly selected individual per household, and a sensitivity analysis included all randomly assigned individuals and accounted for within-household correlation. The trial is registered with ClinicalTrials.gov, NCT05542004.

Findings: We identified 1 232 938 individuals aged 65 years or older in Denmark and excluded 56 436 (4·6%) individuals living in nursing homes and 211 632 (17·2%) with an exemption from the electronic letter system. We randomly assigned 964 870 (78·3%) participants across 691 820 households. Compared with usual care, influenza vaccination rates were higher in the group receiving an electronic letter highlighting potential cardiovascular benefits of vaccination (81·00% vs 80·12%; difference 0·89 percentage points [99·55% CI 0·29-1·48]; p<0·0001) and the group receiving repeated letters at randomisation and at day 14 (80·85% vs 80·12%; difference 0·73 percentage points [0·13-1·34]; p=0·0006). These strategies improved vaccination rates across major subgroups including those with and without established cardiovascular disease. The cardiovascular gain-framed letter was particularly effective among participants who had not been vaccinated for influenza in the previous season (p_interaction=0·0002). A sensitivity analysis of all randomly assigned individuals accounting for within-household clustering yielded similar findings.

Interpretation: Electronically delivered letters highlighting potential cardiovascular benefits of influenza vaccination or sent again as a reminder significantly increased vaccination uptake across Denmark. Although the magnitude of effectiveness was modest, the low-touch, inexpensive, and highly scalable nature of these electronic letters might be informative for future public health campaigns.

Funding: Sanofi.