Behavioural nudges increase COVID-19 vaccinations

Abstract

Enhancing vaccine uptake is a critical public health challenge¹. Overcoming vaccine hesitancy^2,3 and failure to follow through on vaccination intentions³ requires effective communication strategies^3,4. Here we present two sequential randomized controlled trials to test the effect of behavioural interventions on the uptake of COVID-19 vaccines. We designed text-based reminders that make vaccination salient and easy, and delivered them to participants drawn from a healthcare system one day (first randomized controlled trial) (n = 93,354 participants; clinicaltrials number NCT04800965) and eight days (second randomized controlled trial) (n = 67,092 individuals; clinicaltrials number NCT04801524) after they received a notification of vaccine eligibility. The first reminder boosted appointment and vaccination rates within the healthcare system by 6.07 (84%) and 3.57 (26%) percentage points, respectively; the second reminder increased those outcomes by 1.65 and 1.06 percentage points, respectively. The first reminder had a greater effect when it was designed to make participants feel ownership of the vaccine dose. However, we found no evidence that combining the first reminder with a video-based information intervention designed to address vaccine hesitancy heightened its effect. We performed online studies (n = 3,181 participants) to examine vaccination intentions, which revealed patterns that diverged from those of the first randomized controlled trial; this underscores the importance of pilot-testing interventions in the field. Our findings inform the design of behavioural nudges for promoting health decisions⁵, and highlight the value of making vaccination easy and inducing feelings of ownership over vaccines.

Fig. 1. Timeline, assessment for eligibility and randomization of two sequential RCTs.

Timeline, eligibility for enrolment, the total number of participants excluded from the analysis, the total number of participants included in the analysis, and the number of participants who were randomized into each condition and included in the analysis are displayed here for the first and second RCTs. t is the date on which participants received the initial invitation to take up a COVID-19 vaccine from UCLA Health. The first reminder date fell on the first weekday after the initial invitation was sent, and the second reminder date fell on the first weekday after the eighth day following the initial invitation. Exceptions were that participants who received the initial invitation during 19–29 January 2021 were enrolled in the first RCT on 1 February 2021 and the second RCT on 9 February 2021, owing to the delay in setting up the infrastructure needed to run the RCTs. In the first RCT, 38,983 participants were sequentially excluded from the analysis, including (1) 33,533 individuals who obtained the first dose before the first reminder date according to the vaccination records UCLA Health could access on 25 May 2021; (2) 5,392 individuals who made the first-dose appointment at UCLA Health before 15:00 h on the first reminder date; and (3) 58 individuals who were under 18 years old. In the second RCT, 35,583 participants were sequentially excluded from the analysis, including (1) 35,127 individuals who obtained the first dose before the second reminder date according to the vaccination records UCLA Health could access on 25 May 2021; (2) 408 individuals who made the first-dose appointment at UCLA Health before 15:00 h on the second reminder date; and (3) 48 individuals who were under 18 years old.

Fig. 2. Appointment and vaccination rates at UCLA Health by condition for the first RCT.

a, b, Proportion of participants in each condition who scheduled an appointment for the first dose of the COVID-19 vaccine at UCLA Health between 15:00 h on the first reminder date and 23:59 h on the fifth day following the first reminder date (a) and the proportion of participants in each condition who obtained the first dose of the COVID-19 vaccine at UCLA Health within four weeks of the first reminder date (b). Error bars represent ± 1 s.e.m. The number of participants in each condition (from left to right in each panel) is 18,629, 18,592, 18,757, 18,627 and 18,749.

Fig. 3. Kaplan–Meier curves reflecting the proportion of participants who had obtained the first dose at UCLA Health by a given day after the first reminder date in the first RCT.

Kaplan–Meier curves tracking the percentage of participants in the holdout arm (blue) (n = 18,749) versus the follow-through reminder arm (red) (n = 74,605) of the first RCT who had obtained the first dose of COVID-19 vaccine at UCLA Health by a given day from the first reminder date (0 on the x axis) onward. All participants were right-censored at 55 days after the first reminder date. The solid horizontal line indicates that 18.38% of participants in the holdout arm had obtained the first dose at UCLA Health by the end of 55 days after the first reminder date.

Fig. 4. Regression-estimated increase in appointments and vaccinations induced by reminders.

a, b, Regression-estimated increase in appointment rates at UCLA Health within six days of the first reminder date (left panel in a, b) and vaccination rates at UCLA Health within four weeks of the first reminder date (right panel in a, b), induced by receiving a reminder (versus holdout) (a) and by receiving a reminder with ownership language (versus one without) (b) across participant subgroups in the first RCT. The full sample referred to 93,354 participants included in the analysis of the first RCT. ‘White’, subsample including 49,909 participants who identified as white (excluding Hispanic or Latino individuals); ‘minority’, subsample includes 29,784 participants who identified as Asian, Black, American Indian or Alaska Native, Native Hawaiian or Pacific Islander, other race (excluding participants whose race was unknown), and/or Hispanic or Latino. The ‘≥65 years old’ subgroup includes 84,075 participants who were at least 65 years old; the ‘<65 years old’ subgroup includes 9,279 participants under 65 years old. The ‘influenza vaccine’ subgroup includes 46,757 participants who received the influenza vaccine in either of two recent influenza seasons; the ‘no influenza vaccine’ subgroup includes 46,597 participants who did not receive an influenza vaccine in two recent influenza seasons. Extended Data Table 2, Supplementary Tables 3, 5, 6, 10, 11, 13 provide complete OLS regression results graphed here and the corresponding sample sizes. Error bars represent 95% confidence intervals of estimated increases.