Why does this study matter?
Increasingly, outpatient care is delivered with a team-based approach to care, often called the patient centered medical home. A key principle of this model is to deliver the right care, at the right time, in the right place, using the right provider, which is not necessarily the MD.
Team-based care benefits from a different physical layout, most often associated with co-located staff or decentralization. This can be compared to the traditional care model that is physician-centric, hierarchical, office-visit based, and high volume and designed with private offices for the doctors. (And we all know how important that can be when we’re discussing design.) We have research in this area, but it’s focused on communication and some level of collaboration, not the physical configuration is it pertains to team development.
How was the study done?
This was a quasi-experimental design conducted using 13 clinics within the organization, most in rural communities with small populations. Some clinics had transitioned several years earlier to the Patient-Centered Medical Home model with space specifically designed with co-location strategies to support team-based care. They used 4 of these purpose-designed team-based clinics and compared them to 9 traditional clinics in the same system. Teams were considered co-located if they met two conditions. 1) multi-disciplinary team members shared work space and 2) private provider offices were eliminated. Teams were considered non-collocated when team members were in different locations, most often physicians in private offices.
Care teams included providers (physician, nurse practitioner, or physician’s assistant), registered nurses, and licensed practical nurses or medical assistants. Pharmacists and social workers were also considered team members, but they were defined as specialists. They were shared resources and might only be at a site 1 or 2 days each week.
The measures included a validated measure of team development through the Team Development Measure (TDM) and the Press Ganey patient satisfaction tool. Statistical analysis was conducted to establish group differences looking at mean TDM scores between the two categories of co and non-collocated. Analysis was also done to see if there were differences among the clinics in each group, establishing whether the co-location was the issue, or something else.
TDM scores of the two groups were also compared for each role within the care team, so there was an overall comparison of each role within each group, and then for each role within its own group to see if there were differenced among the clinics in each group. Patient PG scores were compared between four collocated and nine non-collocated clinics.
So what do we learn from the study?
The TDM consists of 4 constructs: [CLICK] communication, cohesion, primacy of the team, and role clarity. Role clarity wasn’t included, as the researchers felt this wouldn’t be influenced by the clinic design. The overall mean scores of all staff within the collocated clinics on the three constructs were higher than mean scores of all staff in non-collocated clinics, and the analysis confirmed these differences were statistically significant. The analysis also confirmed that clinics within collocated groups and clinics within non-collocated groups didn’t significantly differ on their scores.
Additionally, when looking at roles, the mean values of communication, cohesiveness, and team primacy were higher for providers, registered nurses, and licensed practical nurses in collocated clinics than those in non-collocated clinics. This matches the overall findings, but there was one role that did not follow this pattern. The mean values were lower for specialists in collocated clinics than in non-collocated clinics.
That’s the trend, but were these differences statistically significant? Most were, but there were a few exceptions. First, there was no statistical significance on communication for providers between the two location types, so nothing to say it was more than a random difference. There was also no statistically significant differences in scores for any constructs for specialists – again, nothing to suggest it was a meaningful difference between the two types. For patient satisfaction, the only significant difference found was for “moving through the visit,” where more patients actually selected the top box option in non-collocated clinics, so traditional offices.
Can we say the results are definitive?
This was a small sample size and it was geographically limited. The work did not directly address the staff satisfaction with their space. They did not measure the impact of the built design on performance or patient outcomes (other than patient satisfaction). The data was analyzed at an individual level, not a work unit level, so this may be an addition in future study. Patient satisfaction data from two sites needed to be excluded due to low sample size. Importantly, we don’t have any plans, so we can’t see what other influences may have been at play here - layout, age, context.
What’s the takeaway?
The elimination of private offices can be controversial, and in this study, we start to develop a better understanding of some empirical data that might be used with staff when presenting shared work space as a design strategy. With studies like this, we might be able to develop open minds, not just open plans.
Interested in the topic? Visit The Center for Health Design Knowledge Repository for more.
Stroebel, R. J., Obeidat, B., Lim, L., Mitchell, J. D., Jasperson, D. B., & Zimring, C. (2021). The impact of clinic design on teamwork development in primary care. Health Care Management Review, 46(3), 257–264. https://doi.org/10.1097/HMR.0000000000000259
Our slidecasts are an outcome of the popular Research Matters presentations at the annual Healthcare Design Expo & Conference. Our research team picks papers that have some significance to the healthcare design community and distill the study down into a 5-minute summary of how the study was done, what was learned, the limitations and the takeaway. The slidecasts bring research to you in digestible format. Just five minutes, and you’ll know more.