Effects of Ownership Text Message Wording and Reminders on Receipt of an Influenza Vaccination: A Randomized Clinical Trial

Abstract

Importance: Despite the availability of safe and effective vaccines, many people fail to get vaccinated. Messages using behavioral science principles may increase vaccination rates. Objective: To determine the effect on influenza vaccination rates of a text message telling patients that an influenza vaccine had been reserved for them. Design, Setting, and Participants: As part of a larger influenza vaccine messaging megastudy, in this randomized clinical trial, 11188 patients in 2 large health systems were assigned to receive a text message that stated "a flu shot has been reserved for you," a text message that stated "flu shots will be available," or no text message. Both messages included the option to reply yes (Y) or no (N) to indicate that the patient wanted to get vaccinated. Patients 18 years or older were included if they had new or routine (nonsick) primary care appointments scheduled from September 20, 2020, through March 31, 2021. Interventions: The evening before the scheduled appointment, patients in the 2 message conditions were sent 3 back-to-back SMS messages containing the study wording. Patients in the usual care control group did not receive any study messages. Main Outcomes and Measures: Receipt of an influenza vaccine on the date of the patient's scheduled appointment. Results: A total of 11188 patients were randomized to the reserved or the available message conditions or to usual care. The 10158 patients analyzed in the study had a mean (SD) age of 50.61 (16.28) years; 5631 (55.43%) were women; and 7025 (69.16%) were White. According to health records, 4113 (40.49%) had been vaccinated in the previous influenza season, and 5420 (53.36%) were patients at Penn Medicine. In an intent-to-treat analysis, changes in vaccination rates in response to the reserved message did not reach statistical significance (increase of 1.4 percentage points, or 4% [P =.31]) compared with the message conveying that influenza vaccines were available. Relative to the usual care control, the reserved message increased vaccination rates by 3.3 percentage points, or 11% (P =.004). Patients in the reserved message condition were more likely to text back Y (1063 of 3375 [31.50%]) compared with those in the available message condition (887 of 3351 [26.47%]; χ2= 20.64; P