SMS text message reminders versus usual care for cervical screening attendance in London, UK: a retrospective secondary analysis of routinely collected data.

Background: Previous randomised controlled trials have shown that general practitioner (GP)-endorsed text message reminders can increase cervical screening attendance. Following the implementation of these reminders across routine primary care services in London (UK), we aimed to assess their effect on likelihood of cervical screening attendance when implemented at scale. Methods: In this retrospective secondary analysis of routinely collected data, all individuals aged 24-64 years who were eligible for cervical screening within participating GP practices in London between June 21, 2021, and March 11, 2024, were included. People who had opted out of the screening programme or the text reminder service, had had a cervical screening before the initial invitation letter, or were out of the age range at the time of the initial invitation letter (ie, aged <24 years or ≥65 years) were excluded. Attendance was compared between individuals who received a GP-endorsed SMS reminder and those who did not. The main outcome was odds of cervical screening attendance within the screening cycle. Routinely collected data were obtained from the text message provider and the National Health Application and Infrastructure Services (now the UK National Health Service Cervical Screening Management System). We used binary logistic regression to examine associations between SMS reminder delivery and attendance, adjusting for previous screening behaviour, age, and deprivation. Data on sex, gender, and ethnicity were not possible to include. Findings: Of the 3 530 557 individuals assessed for eligibility, 3 127 759 were included and 402 798 were excluded. 2 112 003 (67·5%) of the 3 127 759 individuals received an SMS reminder for cervical screening and 1 015 756 (32·5%) did not. 1 091 902 (51·7%) of the 2 112 003 individuals who received an SMS reminder attended a screening, which was an increase of 1·7 percentage points since the last screening cycle (1 055 287 [50·0%] of 2 112 003), whereas 253 556 (25·0%) of 1 015 756 individuals who did not receive an SMS reminder attended a screening, which was a decrease of 2·8 percentage points since the last screening cycle (282 484 [27·8%] of 1 015 756). Successful delivery of an SMS reminder resulted in an increased likelihood of cervical screening attendance (odds ratio 3·36 [95% CI 3·31-3·40]; p<0·0001). Interpretation: The findings indicate the potential effect of SMS reminders as a scalable component of cervical screening systems while emphasising the importance of digital reach, contact-data quality, and equitable implementation. Further research should evaluate differential effects considering demographic characteristics, including ethnicity and socioeconomic status. Funding: National Institute for Health and Care Research and the National Recovery and Resilience Plan of the Republic of Bulgaria.