Newswise — The old adage about a frog that gets put in a pot of cold water on a stove, and doesn’t leap out even as the heat slowly climbs to boiling, might seem like a strange metaphor for primary care.

But for many primary care providers around the country, it might feel like an increasingly apt one.

And they feel like the frogs.

The heat that has gradually increased under them in recent years came from a range of sources:

  • Insurance companies that make them the gatekeeper for patients who need specialty care.
  • The electronic health record systems that eat up hours after the ‘work day’ is done, to document diagnoses, orders and treatment authorizations.
  • The increased expectations from patients and specialists that they’ll respond to messages, test results or requests instantly.
  • The performance metrics that come with a threat of financial penalty.
  • The growing number of medical decisions that demand in-depth discussions with patients.
  • And now the rapid rise in telemedicine visits due to COVID-19.

And all of this with little or no increase in the time they have to get it all done, or reduction in the number of patients assigned to them.  

Two teams of primary care providers from Michigan Medicine, the University of Michigan’s academic medical center, have just published papers in the Journal of General Internal Medicine looking at issues facing primary care practitioners in the third decade of the 21st Century.

More demands, same amount of time

One of the papers documents the electronic medical record demands faced by general internists at Michigan Medicine, U-M’s academic medical center.

Each one reported an average of 390 “in-basket” tasks each week, and reported spending an average of 20 hours each week handling the tasks and the additional work that most of them resulted in. That’s far more than the eight hours of administrative time per week allowed for each full-time clinician. 

The findings build on other studies showing that primary care providers nationwide face the highest number of incoming tasks on EMR systems.

Lead author Laurence McMahon, M.D., M.P.H., chief of general medicine at U-M, notes that the burden has driven many primary care providers across the country to cut back their clinical schedule, just to make their work hours manageable.

“Fully 70% of Michigan Medicine’s general medicine faculty now work ‘part-time’, because it is the only way they can manage the explosion in work – in essence they are taking a pay cut in order to deal with what has become a full-time body of work,” says McMahon.

Some have decided to go into hospital medicine, which comes with defined shifts providing general care to hospitalized patients. The new paper’s senior author, Vineet Chopra, M.D., M.Sc., leads Michigan Medicine’s Division of Hospital Medicine.

“If we are the frogs in the story, I was there when they put the pot on the stove,” says McMahon, who has practiced medicine for more than 30 years. “I’ve had the opportunity to live through all this change, and I remember when each day would mean you’d finish your work and be done. This is no longer the case, and I fear that it will mean fewer physicians choosing primary care careers at a time when our country needs them more than ever. We need a rational way of dividing up that workload so that it’s supportive of patient care and of the workforce that’s delivering it.”

McMahon is quick to point out that the current situation didn’t come about all of a sudden, or on purpose.

“This has happened by accident, because of incremental steps,” he explains. “Each time we just altered how we practice, and what we do outside our official work hours, to accommodate the additional tasks. But for many, this has made it a job that’s not doable within the time we have to see the patients we need to see.”

It’s no wonder, he says, that some of the “frogs” have chosen to leave the pot, or climb the sides by reducing their hours.

He and his colleagues call for a serious examination of the appropriate size and composition of the panels of patients that primary care providers are expected to maintain, the ability to customize EMRs for primary care, and attention to the actual demands on providers’ time. Because of the importance of lifestyle and work-life balance to the current generation of new physicians, time is of the essence to make this change.

Sharing decision making – express style

The other new paper, by U-M and VA Ann Arbor Healthcare System internists Tanner Caverly, M.D., M.P.H., and Rodney Hayward, M.D., focuses on one of the expectations that has arisen in recent years for primary care providers: shared decision making.

They put forth a framework for an abbreviated version that busy providers can practice, to help patients play a larger role in deciding what’s right for them and avoiding the patriarchal approach of yesteryear. They call it “everyday shared decision making.”

It takes a concept originally developed for high-stakes decisions about cancer treatment or major surgery, and adapts it for the kinds of decisions made at the primary care level.

“An emphasis in recent years has been on improving discussions between the physician and patient—about medical evidence, personal preference and overall goals. But the picture is different in primary care, where most decisions are not high-stakes. Primary care providers most often guide patients about lower-stakes decisions such as whether and when to get screened for different diseases – and only have a minute or two within a clinic visit to make those decisions,” says Caverly.

In fact, some studies have shown that as a result of the short time allowed for primary care encounters, many patients aren’t getting a true chance to share in the decisions about their health. Efforts to involve other clinic staff in the shared decision making process aren’t working either.

In the framework that the authors and their colleagues have proposed, the primary care provider must be able to quickly use information about the individual patient’s risk factors to formulate a personalized recommendation about the potential positives and negatives of the decision at hand.

But instead of trying to go over all of the odds ratios and statistics in the minute or two they can allot for such discussions, they instead give the patient their evidence-based, personalized recommendation, and then put the ball in the patient’s court by saying they’ll support whatever they decide. If the patient asks for more details, they can provide them.

In order to make this approach possible, primary care providers will need more information at the time of care about how much any given treatment, screening or other intervention is likely to help that patient in particular, given their health history and risk factors.

Such information needs to be at their fingertips, including in the computer systems they use during appointments, and a good sense of how much patient preference varies for each decision. This could flag for providers which patients fall into the “preference sensitive zone” for any given decision, so the provider can let them know the decision isn’t clear-cut and inform them about key factors that affect the decision.

Caverly and Hayward give as an example a website created to help providers determine quickly which patients will get the most benefit from a CT scan to screen for lung cancer,

They note that in their future efforts to test the everyday SDM model, a main task for providers is to frame the decision for that individual patient and then let them decide. For instance, for one patient, they might say about a lung cancer screening, “For someone like you, I think the potential benefits outweigh the harms” but for another patient they might say “This is very clear-cut, and the benefits far outweigh the potential harms.”


McMahon, Chopra, Caverly and Hayward are all members of the U-M Institute for Healthcare Policy and Innovation. In addition to McMahon and Chopra, the paper on primary care tasks is co-authored by Kim Rize, M.B.A., and NiJuanna Irby-Johnson, M.D., M.B.A.



Journal Link: Journal of General Internal Medicine. DOI: 10.1007/s11606-020-06077-6 Journal Link: Journal of General Internal Medicine. DOI: 10.1007/s11606-020-06043-2