Development and validation of the knowledge and attitudes regarding antibiotics and resistance (KAAR-11) questionnaire for primary care physicians.

The Journal of antimicrobial chemotherapy

PubMedID: 27353468

López-Vázquez P, Vázquez-Lago JM, Gonzalez-Gonzalez C, Piñeiro-Lamas M, López-Durán A, Herdeiro MT, Figueiras A, GREPHEPI Group. Development and validation of the knowledge and attitudes regarding antibiotics and resistance (KAAR-11) questionnaire for primary care physicians. J Antimicrob Chemother. 2016;.
OBJECTIVES
The aim of this study was to develop a novel, self-administered questionnaire to identify primary-care physicians' knowledge and attitudes regarding antibiotics and resistance (KAAR).

METHODS
The study population comprised primary care physicians. The study was conducted in five phases. Phase I consisted of a systematic review and qualitative focus-group study (n?=?33 physicians), in which items were formulated so as to be measured on a continuous, visual analogue scale (VAS); in Phase II, content validation and face validity were evaluated by a panel of experts, which reformulated, added and deleted items; Phase III consisted of a pilot study on a population possessing similar characteristics (n?=?15); in Phase IV, we analysed reliability by means of a test-retest study (n?=?91) and calculated the intraclass correlation coefficients (ICCs); and in Phase V, we assessed construct validity by applying the known-groups technique, measuring the differences between contrasting groups of physicians formed according to antibiotic prescription quality indicators (group 1, n?=?156 versus group 2, n?=?191).

RESULTS
Following Phases I and II, the questionnaire contained 16 knowledge and attitude items. Participants in the pilot study (Phase III) reported no difficulty. The test-retest study (Phase IV) showed that 11 of the 16 initial knowledge and attitude items yielded an ICC?>?0.5, while analysis of known-groups validity (Phase V) showed that 13 of the 16 initial items which assessed knowledge and attitudes discriminated between physicians with good and bad indicators of antibiotics prescription.

CONCLUSION
The final 11 item KAAR questionnaire appears to be valid, reliable and responsive.