Development of an innovative theory-based instrument to assess the impact of continuing professional development activities on clinical practice

Article type
Authors
Légaré F1, Borduas F1, Jacques A2, Drolet R3, Godin G4, Luconi F5, Rousseau M6, Freitas A3
1Université Laval, Canada
2Collège des médecins du Québec
3Centre de recherche du CHU de Québec, Canada
4Université Laval, Canada
5McGill University
6Université du Québec à Trois Rivières, Canada
Abstract
Background: A group of continuing professional development (CPD) decision-makers met with knowledge-translation (KT) researchers. They identified needs for high-quality evaluation methods to assess the impact of CPD activities.

Objectives: To present the development of a theory-based, reliable instrument to assess the impact of accredited CPD activities on clinical practice.

Methods: After a systematic review and analysis of existing instruments assessing healthcare professionals’ intentions and behaviours, an inventory of instruments based on social cognitive theories was created. Items most relevant to the constructs of an integrated theoretical model were selected from this inventory to devise a new tool. An e-Delphi study with experts from different domains was conducted to check its face validity and likely acceptability in CPD settings. A test-retest validation was done with end-users.

Results: We identified 47 eligible instruments with 1218 items. These items were reclassified into the eight constructs of an integrated theoretical model for the study of healthcare professionals’ intentions and behaviours. Through an interactive process, 61 items were selected to compose the preliminary tool. Following an e-Delphi process, a generic questionnaire with 40 items was created. By completing the 40-item questionnaire at the end of a CPD activity (test), 138 physicians indicated their agreement to participate in the validation study. The same questionnaire was completed 2 weeks after by 121 participants (retest). Exploratory factorial analysis allowed an item reduction process resulting in a 12-item questionnaire. The Cronbach α coefficient of a global score was determined for each construct of the integrated model. Values varied from 0.77 to 0.89.

Conclusions: We propose a new instrument for assessing the impact of CPD activities on physicians’ clinical practice. Before its implementation on a large scale, further studies are needed to validate its ability to predict intentions and behaviour and its sensitivity to change in response to CPD activities.