Closing the Gap in Provider Networks

Much attention has been given to the nation’s physician shortage. The often-quoted annual AAMC study suggests a shortage of 37,000 today that could grow significantly over the next 10 years. To reinforce the point, multiple sources report that the average wait time for a new patient appointment across the US is almost a month and, for certain specialties or underserved markets, the wait can be much longer than that. Innovation in our collective approach to network development is required as health systems seek to close the provider gap. Consider the following:

1) Health systems need to think like Starbucks

How does Starbucks know where to put the next Starbucks? Their models are based on the concept of “gravity.” Gravity is a calculation that weighs time and distance against the likelihood to travel for retail-sensitive products and services. Retailers also consider block groups instead of zip codes; and detailed consumer segmentation models (e.g. Mosaic) instead of payor mix or household income as a more sophisticated means of determining service placement. The best planning models use these retail concepts to plan their provider networks.

2) Using historic assumptions in provider network planning will guarantee historically poor access

The long-held industry standard of 275 doctors per 100,000 people is insufficiently blunt for market planning because all markets are different. The difference lies in utilization rates for ambulatory encounters. We know that those over age 65 use 3x as much healthcare as those below age 65, the uninsured use only one-third of the healthcare of an insured population, and rural populations use less than 30% less care of their urban counterparts. Equitable provider network planning requires a prospective look at where use rates are heading—not where they have been.

3) What is the “optimal network pyramid?”

Understanding the right mix of specialties is key to creating a durable provider network. There are approximately 186,000 primary care physicians (PCPs) in the US, but only 5,300 neurosurgeons, a ratio of 35:1. We would suggest that these ratios of physician supply represent a starting point for the “optimal network pyramid.” In this way, ” bottom-up” provider planning starts with targeting of primary care market share across a system’s geography and a pyramidic blend of specialists can be inferred. “Top-down” provider network planning uses the same physician ratios but works the other way. If a system seeks to add 150 open heart surgeries, we know that will support one additional cardiovascular (CV) surgeon. However, we also know that for every CV surgeon, a system should have 6-8 cardiologists and a corresponding group of 40-50 PCPs. Either way, the “optimal network pyramid” works.

4) Adequacy is not the same as access

Most planners are familiar with the Medicare Advantage requirements for network adequacy. These guidelines suggest that a network must have providers within a certain time and distance from a population center to be deemed ”adequate.” What some may not know is that the calculation as prescribed in the legislation suggests that if even ONE provider is located in the right place, adequacy is met regardless of how many people may live in the proximate area. Moreover, adequacy calculations make no allowance for telemedicine, which is an increasingly critical means for the provision of care in undersupplied specialties and rural areas. For this reason, industry-leading planners recognize that “Adequacy” is not the same as “Access,” and plan accordingly.

5) What is the impact of physician loyalty on your network?

According to the Physician Advocacy Institute, 75% of physicians are now employed, but only one-half of those are employed by health systems. For years, health system executives have suspected leakage of services to competitors—both from their employed providers as well as closely held independents. A recent review of claims data suggests health system provider networks are indeed less loyal than most systems think. Understanding the system affiliation and relative loyalty of providers in a market is critical in provider network planning.

6) The impact of physician productivity on provider network planning

Health system executives often lament, “how can you say I need more cardiologists, the ones that I have are not busy?” Our experience suggests that, in most cases, low productivity in provider practices is due to operational issues and not provider supply. Broken processes can include scheduling templates, overburdensome referral rules, ineffective use of IT systems to reduce the time required for documentation, processes to fill no-show gaps, and other factors. Accordingly, health systems need to first ensure that operations promote optimal provider productivity and that decisions to recruit new physicians are not based on the perpetuation of inefficient processes.

7) Advanced practice providers (APPs) are critical to closing the provider gap

Ten years ago, there were approximately 30,000 APPs across the country. Today, there are more than 350,000. Other studies suggest that APPs provide almost 25% of all US health visits. Despite these facts, most provider workforce planning models do not account for use of APPs in the practice. The nation’s best healthcare systems have worked to ensure the impact of APPs on the provider workforce is accounted for in provider network planning.

8) Household income and payor mix are things of the past

Historic “market attractiveness” measures were always heavily weighted to growth, household income, and payor mix. As payment rates norm and the impact of cross-subsidization is eliminated, the new metrics of “net need,” low use rates, and low access become the new planning currency.

9) Panel size is a complex question

There has been much discussion about the appropriate panel size for a primary care provider. A recent study found that more than 40% of PCP panels were closed to new patients. So, what is the right size for a primary care panel? The answer is: “it depends.” The most important dependency is the acuity of the patients. Equally impactful on panel size is the “model of care” that is deployed to the panel. Patient panels with a large support team (APPs, nutritionists, pharmacists, psych, remote patient monitoring, etc.), enable physicians to care for much larger patient panels. Gaining an in-depth understanding of these factors will help better gauge optimal panel size.

Sophisticated health system planners understand that provider network planning requires advanced tools interpreted by experts to develop strategies that can have a meaningful impact on patient access.

References

1 “The Complexities of Physician Supply and Demand: Projections from 2021 to 2036,” AAMC, March, 2024

2 “Survey of Physician Wait Times”, Merritt Hawkins, 2023

3 Huff, David L. “A probabilistic analysis of shopping center trade areas. Land Economics” 39, 1963

4 National Ambulatory Medical Care Survey (NAMCS), www.cdc.gov

5 Number of People by Active Physician Specialty, 2022; AAMC, March 2024

6 42 CFR 422.116, and “Medicare Advantage Network Adequacy Criteria Guidance,” www.cms.gov

7 “PAI Report on Physician Employment Trends and Acquisition of Medical Practices 2019-2023.” www.physicianadvocacyinstitute.org

8 Bureau of Labor Statistics - www.bls.gov

9 BMJ 2023: 382:e073933

10 Primary Care Need in New Mexico, Report to the New Mexico Legislature, March 2024.