SMS Messaging for Invitation in the Cervical Cancer Screening Programme

Overview

A study on two different methods of invitation to participate to the cervical cancer screening programme will be conducted within a demonstration project to switch from cytology-based screening to HPV-based screening using self-sampling delivered through the network of pharmacy offices among regular screening attendants in the Barcelona Metropolitana Sud Area, in Catalonia. At the moment, eligible women are invited to participate to cervical cancer screening via a telephone call invitation explaining the new self-sampling method. Invitation via SMS containing a link to a webpage with information on most frequent questions might be an adequate alternative method that would save costs and workload on human resources. The aim of this study is to assess the impact on cervical cancer screening participation of an invitation method based on text messaging (SMS). The invitation method will be evaluated through an interventional trial, in which we will compare the invitation to cervical cancer screening using SMS versus a telephone call invitation explaining the new self-sampling method.

Main Hypothesis: The use of direct SMS without an explanatory call for cervical cancer screening invitation is an acceptable option to invite women if the participation obtained is more than 70% and not less than 15% of the call (current practice). Specific Hypothesis: * Participation in cervical cancer screening programme using SMS invitation is non-inferior to 70%. * Participation in cervical cancer screening programme using telephone call is non-inferior to 70%. * Participation using SMS invitation is non-inferior to participation using telephone call invitation, considering that an absolute difference less than 15% may not be clinically relevant. Sample size calculation: A one-sample non-inferiority test for calculating sample size was performed, assuming a participation rate of 80% using SMS invitation and 85% using telephone call invitation in the cervical cancer screening programme. To detect a participation rate using SMS or telephone call invitation non-inferior to 70%, with a margin of 5%, accepting an alpha risk of 0.025 and a beta risk of 0.2, and 5% of loses of follow-up, we will need 27 women in the telephone call group and 59 women in the SMS group. To observe this non-inferiority by age group (≤40 years, 41-54 years, ≥55 years), 27 and 59 women will be needed for each age group. To compare both study arms, a two-sample non-inferiority test was performed assuming that using SMS messages to invite women to participate in cervical cancer screening programme will be 5% lower compared to participation when invitation is performed by a phone call (80% versus 85% respectively), although this difference will be non-inferior to 15-15.5%. Thus, if there is a true difference in favor of the telephone call group of 5%, assuming 5% of loses of follow-up, then 450 women will be required (225 in each arm) to be 80% sure that the upper limit of a one-sided 97.5% confidence interval will exclude a difference in favor of the SMS group of more than 15-15.5%. Randomization procedure: Women will be randomized in two study arms: intervention (SMS message invitation) and control (telephone call invitation). Randomization will be performed based on permuted blocks of 6 women to ensure equal distribution among age groups.