Understanding preferences for self-sampling in a national cervical screening programme: a protocol for a discrete choice experiment.

INTRODUCTION

The National Health Service Cervical Screening Programme (NHSCSP) currently involves a healthcare professional collecting a cervical sample in a healthcare setting. This method of screening has barriers associated with access to screening appointments and the poor acceptability of the speculum examination. Primary screening through HPV testing has led to the development of self-sampling screening methods including vaginal and urine self-sampling, with many UK studies comparing these screening methods with the current NHSCSP. It is not known what features of self-sampling influence individuals' preferences and cervical screening uptake. To understand these preferences, we plan to undertake a discrete choice experiment (DCE). This protocol aims to describe the steps taken to design the DCE and the proposed approach to fielding the DCE to identify preferences for different sampling approaches in cervical screening.

METHODS AND ANALYSIS: An online survey comprising a DCE was designed to understand preferences of individuals for self-sampling methods within the NHSCSP. Attributes and levels for the DCE were generated through an iterative process including a literature review of qualitative studies about self-sampling cervical screening methods, input from cervical screening clinical experts and a patient and public involvement group (n=6). A D-efficient design was used to create choice sets for the DCE survey. Regression-based analysis will be used to estimate the impact of each attribute and level on individual choices.

ETHICS AND DISSEMINATION: This study has been approved by The University of Manchester Proportionate Research Ethics Committee (2024-20767-37669). The results of the DCE will be submitted for publication in a relevant peer review journal and the results will be presented at national and international conferences.

DATA STATEMENT: There are no data associated with this protocol. The data produced by this study and analysis scripts will be made available in a public repository following publication of the study.