• Luke T A Mounce, Heather E Barry, Raffaele Calitri, William E Henley, John Campbell, Martin Roland, and Suzanne Richards.
• Collaboration for Academic Primary Care (APEx), University of Exeter Medical School, University of Exeter, Exeter, UK.
• BMJ Qual Saf. 2016 Nov 1; 25 (11): 842-850.

BackgroundA 2014 national audit used the English General Practice Patient Survey (GPPS) to compare service users' experience of out-of-hours general practitioner (GP) services, yet there is no published evidence on the validity of these GPPS items.ObjectivesEstablish the construct and concurrent validity of GPPS items evaluating service users' experience of GP out-of-hours care.MethodsCross-sectional postal survey of service users (n=1396) of six English out-of-hours providers. Participants reported on four GPPS items evaluating out-of-hours care (three items modified following cognitive interviews with service users), and 14 evaluative items from the Out-of-hours Patient Questionnaire (OPQ). Construct validity was assessed through correlations between any reliable (Cochran's α>0.7) scales, as suggested by a principal component analysis of the modified GPPS items, with the 'entry access' (four items) and 'consultation satisfaction' (10 items) OPQ subscales. Concurrent validity was determined by investigating whether each modified GPPS item was associated with thematically related items from the OPQ using linear regressions.ResultsThe modified GPPS item-set formed a single scale (α=0.77), which summarised the two-component structure of the OPQ moderately well; explaining 39.7% of variation in the 'entry access' scores (r=0.63) and 44.0% of variation in the 'consultation satisfaction' scores (r=0.66), demonstrating acceptable construct validity. Concurrent validity was verified as each modified GPPS item was highly associated with a distinct set of related items from the OPQ.ConclusionsMinor modifications are required for the English GPPS items evaluating out-of-hours care to improve comprehension by service users. A modified question set was demonstrated to comprise a valid measure of service users' overall satisfaction with out-of-hours care received. This demonstrates the potential for the use of as few as four items in benchmarking providers and assisting services in identifying, implementing and assessing quality improvement initiatives.Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.

20 keywords...

hide…