The Blueprint

Benchmarking clinical support staff in primary care sites

by Amanda Berra

We receive many requests for primary care clinical support staff benchmarks specific to certain types of PCMH (or general primary care) sites, for example, integrated delivery systems, independent practices, large practices, small practices, high-capitation markets, etc.

Here, a current set of benchmarks for all medical home site types, along with answers to questions about benchmarking by site type.

Staff benchmarks for all medical home types

Though the Primary Care/Medical Home benchmarking initiative, we have collected support staff benchmarks for sites self-identifying as medical homes. We're about to update this data again with additional analyses and a larger sample base, but here are the benchmarks as of March, 2011:

Support Staff, Per Physician, in Primary Care Sites

(Self-identified medical homes, n=48-53 (depending on job type)

MA: 1.4

LPN/LVN: .3

RN: .4

PA: .1

NP: .3

Total clinical support staff per physician: 2.6*
*Due to rounding

Benchmarking by specific site types

Since we collected data on site characteristics, such as whether reporting sites were hospital-owned, independent, rural, small, large, etc., we had planned to provide data to Medical Home Project participants cut by subgroups. But it turns out, we can't--because when we tested the differences between the groups, we found no statistically significant patterns by site or market context.

Yes, it does seem odd that there would be no consistent differences or models. Conventional wisdom would strongly suggest many patterns, such as hospital-owned (and subsidized) practices being more heftily staffed than independent sites.

But the lack of pattern actually backs up findings from our qualitative research into emerging staff models in primary care, which says that:

  • Primary care/medical home staff models are tremendously variable based on a range of difficult-to-measure factors, such as culture, training, workflow, and availability of different support staff types in a geographical area.
  • Primary care staff models are in flux industry-wide; the advent of the medical home model, among other factors, has spurred many organizations to experiment with different clinical support staff configurations.
  • The most telling variable in the staff model equation is not how many/what type of support staff FTEs are present in the practice. It's what role the team members are playing on the team. (More information on this in the Resources section at end of post).

The search for staff model right answer

Staff model experiments are ongoing, and nowhere near settling at any correct model anytime soon, if ever. That being said, the experiments we hear about these days all share a common direction: more emphasis on expanding and elevating the role of team members with lower-level clinical credentials (or no clinical credentials). Thus we field increasing questions along the lines of "What models exist for NP, PA, or RN-led clinics or teams?" These questions make perfect sense in the context of organizations searching for ways to use staff model transformation to improve patient access, lower site-level cost of care, and maximize the contribution and impact of each team member, up to and including physicians, who must be used as a scarce and valuable resource.

We predict continued movement toward more clinical (and non-clinical) support staff per physician on each team, more focus on team-based workflows, and more independence/higher-level roles for support staff.

More information and resources

  • The Primary Care/Medical Home Benchmarking Initiative is still open; we wanted to give organizations a little more time to participate. Customized, 35-page benchmarking reports now available for each site that responds to the survey. There is no cost for Medical Home Project participants to do this, it's part of the membership. More information here.
  • Do you have a theory about a factor that would predict staffing model differences among primary care/medical home sites? (E.g., what market or site-level factors would increase or decrease the likelihood of having more total clinical support staff per physician)? We're re-visiting and expanding our analysis now, so e-mail me directly if you have a hypothesis you would like our data team to test. (My email: berraa@advisory.com).
  • Re: the importance of analyzing not just how many clinical support staff there are, but what they are doing: Through the survey results, we do have data on this too--for example, what percentage of medical home sites give primary ownership of, say, pre-visit chart review to MDs vs. NPs, PAs, RNs, LPNs, MAs, or other. E-mail me for more information on these findings, or submit data to the survey to receive a comparison of roles at your site versus medical homes in the sample.
  • See also the Medical Home Project Innovator Spotlight webconference, "Elevating Staff to Top of License at Kaiser Permanente NW", which illustrates how one Kaiser region systematically retrained and repositioned its clinical support staff to work more robustly as a team. Slides and audio available here
  • The most up-to-date full study from our group on PCMHs, in the context of trends in all of primary care, is Transforming Primary Care: Building a Sustainable Network for Comprehensive Care Delivery.  Download or order hard copies (no cost to Medical Home Project Participants) here.