Development and validation of a measure and a model of general practitioner attitudes toward collaboration with pharmacists

Background: Collaboration between general practitioners (GPs) and community pharmacists has been shown tobe effective in improving patient outcomes. However, little is known about GP attitudes toward collaborating with their pharmacist counterparts and variables that influence this interprofessional collaboration. Objectives: To develop and validate, in the context of primary care in Australia 1) an instrument tomeasure GP attitudes toward collaborating with pharmacists and 2) a model that illustrates how GP attitudes (and other variables) influence GP-pharmacist collaborative behavior. Methods: The "Attitudes Toward Collaboration Instrument for GPs" (ATCI-GP) was developed to measure GP attitudes toward GP-pharmacist collaboration based on existing literature and qualitative interviews with GPs and community pharmacists. The ATCI-GP and a previously validated behavioral measure "Frequency of Interprofessional Collaboration Instrument for GPs" (FICI-GP) were included in a survey and administered to a sample of 1145GPs in 12 divisions of general practice across Australia. Principal component analysis (PCA) was used to assess the structure of the ATCI-GP. Structural equation modeling was used to determine how attitudes (measured by the ATCI-GP) and other variables, influence collaborative behavior (measured by the FICI-GP). Results: Three hundred and seventy-six surveys were completed and returned for a response rate of 33%. PCA of the ATCI-GP suggested a two factor ("interactional and practitioner determinants" and "role for pharmacist in medication management") solution accounting for 66.2% of the variance. The model for GP-pharmacist collaboration demonstrated adequate fit (χ2/df=2.27, CFI=99, RMSEA=060, 90% CI [.052-069]). Factors found to predict collaboration included: 1) Interactional and practitioner determinants 2) environmental determinants and 3) GP perception of the pharmacists' role in medication management. Conclusions: The study provides evidence for the validity of the ATCI-GP for measuring GP-pharmacist collaboration from the GPs perspective and supports a model for collaboration in which collaborative behavior is influenced by a number of variables.