Patient-reported experiences with general practitioners: a randomised study of mail and web-based approaches following a national survey.

Objective The standard data-collection procedure in the Norwegian national patient experience survey programme is post-discharge mail surveys, which include a pen-and-paper questionnaire with the option to answer electronically. A purely electronic protocol has not previously been explored in Norway. The aim of this study was to compare response rates, background characteristics, data quality and main study results for a survey of patient experiences with general practitioners (GPs) administered by the standard mail data-collection procedure and a web-based approach. Design Cross-sectional survey. Setting GP offices in Norway. Participants The sample consisted of 6999 patients aged 16 years and older registered with a GP in November 2018. Intervention Based on a three-stage sampling design, 6999 patients of GPs aged 16 or older were randomised to one of two survey administration protocols: Group A, who were mailed an invitation with both a pen-and-paper including an electronic response option (n=4999) and Group B, who received an email invitation with electronic response option (n=2000). Main outcome measures Response rates, background characteristics, data quality and main study results. Results The response rate was markedly higher for the mail survey (42.6%) than for the web-based survey (18.3%). A few of the background variables differed significantly between the two groups, but the data quality and patient-reported experiences were similar. Conclusions Web-based surveys are faster and less expensive than standard mail surveys, but their low response rates and coverage problems threaten their usefulness and legitimacy. Initiatives to increase response rates for web-based data collection and strategies for tailoring data collection to different groups should be key elements in future research.