Electronic nudge letters to increase influenza vaccination uptake in younger and middle-aged individuals with diabetes: a prespecified analysis of the NUDGE-FLU-CHRONIC trial

Despite evidence demonstrating that influenza vaccination is associated with reduced risk of adverse cardiovascular events and all-cause mortality, influenza vaccine uptake remains suboptimal in persons with diabetes mellitus (DM).

In this prespecified analysis of the NUDGE-FLU-CHRONIC trial we assessed the effectiveness of electronically delivered nudges on influenza vaccine uptake according to DM status.

NUDGE-FLU-CHRONIC was a nationwide, randomized, pragmatic implementation trial among younger and middle-aged (18-64 years) Danish citizens with chronic disease during the 2023/2024 influenza season. All trial participants were eligible for a free influenza vaccination through the Danish governmental vaccine program. Participants were randomized in a 2.45:1:1:1:1:1:1 ratio to usual care (no electronic letter) or one of 6 different electronic nudge letters delivered on Sep 26, 2023. Baseline and outcome data were obtained using the Danish nationwide registries. The endpoint was receipt of a seasonal influenza vaccine on or before Jan 1, 2024.

Of 299,881 NUDGE-FLU-CHRONIC participants, 57,666 (19.2%) had DM at baseline. Participants with DM had a median age of 51.6 years, 43.0% were female, median Hba1c of 53 mmol/mol, median DM duration of 10.0 years, and 51.2% were insulin dependent. During follow-up, 43.0% of those with DM vs. 34.6% of those without DM received the seasonal influenza vaccine (p<0.001). Any electronic letter vs. usual care was highly effective in increasing vaccine uptake in participants with DM at baseline (45.6% vs. 36.5%, difference: +9.1 percentage points [99.29%CI: 7.9-10.3], relative risk ratio [RR]: 1.42, 99.29%CI: [1.39-1.44]). However, DM status modified the effect of any electronic letter such that participants without DM at baseline experienced a slightly greater effect than those with DM (37.3% vs. 25.9%, difference: +12.3 percentage points [11.7-12.8], RR: 1.47 [1.45-1.50]; p-interaction <0.001). For each individual electronic nudge letter, the effect of the intervention was slightly lower in participants with DM compared to those without (Figure 1; all p-interaction < 0.015). When assessing DM subgroups, the effect of any electronic letter vs. usual care was greater in participants with DM and concomitant cardiovascular disease than in those without, greater in non-insulin treated than in insulin-treated individuals, and greater in those with shorter diabetes duration. Effectiveness of nudges were consistent across all other DM subgroups including HbA1c level (Figure 2).