Skip to main content
The overall purpose of the research was to answer the question of whether a team-based approach in the healthcare setting was best for quality, cost, patient satisfaction and provider burnout. The authors hypothesized that increasing the ratio of medical assistants (MA), to advance care providers in a ratio of 2:1, integrating advanced practice clinicians, expanding MA roles and including an extended interprofessional team would help achieve this goal in a family medicine or internal medicine practice.
This was a prospective, quasi-experimental, longitudinal study that compared an implementation group to a control group using a difference-in-difference technique. The study was conducted in five different locations, all in the same healthcare system and primarily served a similar patient population within a primary care practice setting. Data was collected using surveys with both staff and advance care providers. We found that this was an important design of the study because it allowed for similar healthcare protocols and reward systems for staff, as well as similar socioeconomical status and insurance plans of the patients their office was serving.
The primary outcomes of the study included team development and wellness scores. Team development was assessed using the team development survey that measured cohesion, communication, team roles and end goals. The secondary outcomes that the study evaluated included quality, cost and patient satisfaction. Quality was assessed using data from hemoglobin A1C, diabetic nephropathy and ACE-inhibitor/ARB medications. Cost was assessed using total cost of labor per each individual patient visit. Lastly, patient satisfaction was assessed using Press-Ganey Surveys which is a well-known analytic company that many healthcare offices use in the United States.
The data analysis revealed several different points. Overall the survey response was higher in the implementation group when compared to the control. When looking at the primary outcomes, the team development data showed that scores increased over time, all of which were statistically significant. Whereas the wellness scores data did not necessarily prove to be statistically significant we discussed that it still could have been clinically significant with regards to provider burnout. Throughout the study, the secondary outcomes (quality, cost and patient satisfaction) remained steady and stable with no real decrease in the implementation group.
Our discussion focused on defining roles within the primary care team. We were interested in learning more about the role of the medical assistant in the implementation group as well as how the team functioned prior to adding an additional team member. We discussed the differences in MA’s that we have observed in our own clinic experience. The expectations of some MA’s may be to get the vital signs and reconcile medication lists and others may include taking an entire focused history of present illness for the advanced provider. We also discussed the differences that arise between teams that have more experience working together (5+ years) in comparison to teams that are relatively new to each other (<1-2 years) and how this might change the overall outcome of the study. Keeping in mind that the primary questions our discussion came back to was “How was the role of each team member defined?”.
Some changes that our group proposed included comparing the same type of practices, defining the role of the MA and exploring the return on investment from implementing a team-based care model. Although the study did compare similar medical office settings, we felt that it would be important to only compare FM to FM and IM to IM offices. Yes, these offices see similar disease processes, but we felt that this would be a relatively easy factor to control that would help improve the validity of the study. Again, our group felt very strongly about clearly defining the role of the added medical assistant. This would be beneficial for other medical offices to be able to replicate this model and assess outcomes. Finally, the study does mention that the labor cost was lower with the implementation group, however they used advanced providers such as nurse practitioners and physicians associate rather than comparing MD/DO’s. Our group would have liked to see the comparison of a MD/DO with less MA’s or if advanced practitioners with more MA’s was superior. This could have been done in the control groups, but it was not specified.
There are some limitations that our group discussed including the survey response, amount of experience, type of clinic, type of healthcare system and data outcomes. In the study, the implementation group had a higher survey response, which could have made the data more favorable for the primary care 2.0 model. We felt that the study did not control for the amount of experience that each MA had in practice and felt that this was a limitation of the study. Our group also felt that the data from the study would only apply to a family/internal medicine practice because specialty clinics may have unique characteristics that will result in different outcomes. Overall the study was not generalizable because it was done in the same health care system. It did help to control for other confounders but made it less likely to apply this information to other healthcare systems that likely work in a completely different way. Lastly, the data that was concluded was not statistically significant, however it could still be considered clinically significant.
In conclusion, this study could have shown stronger evidence of the primary care 2.0 model being statistically significant if the data would have continued for a longer period of time without the disruption in MA staffing which caused the implementation clinic to revert back to the original set up. Our group shows particular interest in learning about the team members roles and would have liked to see them directly outlined. This could make the study reproduceable and allow similar effects across other health care offices that want to carry out this approach.