The process for identifying community pharmacist-physician pairs engaged in effective CWRs was fruitful. Despite not providing experts with a clear case definition of “effective” or “successful,” the pharmacist PPCI scores were comparable with the highest scores reported in earlier studies, indicating high levels of collaboration among the identified sample.18
In addition, the physicians’ PPCI scores were higher across each domain compared with previously reported scores among a large, cross-sectional sample of primary care physicians.16,17
The qualitative exploration of the CWR exchange domains revealed several exchanges that occurred among the professionals when relationships were initiated, trust was established, and professional roles were clarified. Many of these findings (eg, pharmacist as initiator, the importance of communication at early stages of the relationship, and the emphasis on high-quality pharmacist contributions) support the CWR model proposed by McDonough and Doucette9
and provide opportunity for future study.
Specifically, the role of the pharmacist as relationship initiator has been described in earlier work examining these collaborations.20
The study’s qualitative findings of the pharmacist as the primary relationship initiator likely influenced the quantitative results for this exchange domain. On the relationship initiation domain of the PPCI, physicians scored higher than pharmacists. Recognizing that relationship initiation depends largely on their actions, pharmacists may have been critical of their actions, resulting in lower scores on these items. On the other hand, the physicians who recognized that they would not have taken the initiative for relationship development were pleased with the approaches taken by pharmacists. Notably, the mean physician PPCI score for this domain was 20.3 out of a maximum possible score of 21, suggesting high physician satisfaction with the specific initiating behaviors described by the pharmacists in this study. However, other authors who studied CWRs from the pharmacist perspective have not found a relationship among exchanges within this domain and successful collaborations.21
Therefore, more work is needed to determine whether pharmacists initiating relationships in the manner described by participants have greater success in developing collaborations than those using other methods.
Both the qualitative and quantitative findings suggest that community pharmacists and physicians engaged in highly collaborative relationships view trustworthiness in a similar fashion. Both professionals scored similarly on the PPCI and emphasized similar characteristics of the relationship that resulted in a high level of trust. In particular, the importance of establishing a “track record” through consistent, high-quality contributions (by the pharmacist) to patient care was emphasized by both types of professionals.
For role specification, the qualitative findings suggested congruence between how each professional viewed his or her respective roles. However, the discrepancy in PPCI scores between the pharmacists and physicians imply that pharmacists feel less strongly than physicians that both professionals are mutually dependent on each other and that pharmacists are able to successfully negotiate their role in patient care. This apparent discrepancy between the qualitative and quantitative results of this exchange domain warrant further study, because other authors have found that role specification is an important aspect of successful CWRs.16,21
To the authors’ knowledge, this is the first study to triangulate PPCI scores with qualitative perspectives from both professionals. Several of the findings support the work of Brock and Doucette, 20
who conducted an in-depth case study with 10 pharmacists engaged in varying levels of collaboration with physicians. The purpose of the inquiry was to identify variables that distinguish between highly collaborative and less collaborative relationships. Similar to the current results, these authors found that the pharmacist was the relationship initiator, that face-to-face communication was important, and that the relationships developed over time.20
However, they did not find that initiating behaviors, trust, conflict resolution, an assessment (by the physician) of the pharmacists’ competence, or a history of a prior relationship among professionals differentiated between the pharmacists in high-level collaborations versus those at a lower stage of the CWR. In the present study, each of these exchanges was discussed by both professionals (across the pairs) during the qualitative interviews. This warrants further study, because Brock and Doucette20
collected data only from the pharmacist perspective, whereas the current study elicited both perspectives. Nevertheless, differences in perspectives of pairs engaged in high-level versus lower-level collaborations were not assessed. Consequently, more work is needed to understand the importance of these exchanges on developing relationships from the physician perspective.
This study had a relatively low response rate; 26% of the “experts” responded to the request for pharmacist identification, and fewer physicians than pharmacists agreed to participate in the online surveys and interviews. This may be because of “expert” misinterpretation of study-inclusion criteria. For example, the first author received several e-mails from “experts” stating that they could not identify a pharmacist, because the state they reside in does not allow legal collaborative practice agreements. Furthermore, a lower physician response may be because of the methodology used for recruitment, and/or the lack of compensation for time associated with participation, because the interviews averaged 30–60 minutes. Although an attempt was made to increase participant enrollment through follow-up contacts with identified pharmacists and physicians, there was no attempt to identify and recruit new pharmacist-physician pairs. This approach could have resulted in a greater sample size.
Although analysis of the qualitative data revealed repeating themes, it is unknown if additional themes would have emerged with a greater number of interviews. More interviews also may have provided greater insights into the qualitative and quantitative discrepancies we found for the role specification domain. Furthermore, the design and sampling strategy for this study only included participants with high levels of collaboration. This study did not explore outliers who do not collaborate or struggle to collaborate. Future studies among a group of low collaborators may provide additional information important in developing relationships. In addition, given the small sample size, inferences from the quantitative data are speculative. Future studies with larger sample sizes should explore variations in the magnitude of the difference between pharmacist and physician PPCI scores. Finally, although some of the participants have been informally exposed to the current findings after participation in the interviews, a formal process for confirming the authors’ interpretations of the qualitative findings with study participants and allowing participant commentary may have strengthened the study results and enhanced confidence in the authors’ interpretations of the qualitative data.