This is Chapter 10 of 50 in a summary of the textbook Handbook of Healthcare Delivery Systems. Go to the series index here. Listen on YouTube Playlist, or search your podcast app: Gregory Schmidt
Chapter 10 Summary
Managing Physician Panels in Primary Care
Chapter Authors
Hari Balasubramanian - University of Massachusetts at Amherst
Brain T. Denton - North Carolina State University
Qi Ming Lin - EI nk Corporation
Some Comments
After many very vague and general chapters, the textbook starts to get interesting again with some more practical suggestions on addressing performance and process improvement in healthcare.
1. Introduction
Timeliness
Timeliness is one of the most important factors in primary care. When patients are unable to get a timely appointment, they may delay care, or seek care in the ER.
If they pursue care outside of their own provider, they may incure higher resource usage because of unfamiliarity of a new provider with their medical history or personal preferences.
One in three patients identify the inability to get a timely appointment a “significant obstacle to care”. If you live in Canada, this number I suspect is significantly higher. Significantly.
Continuity of care & Appointment Scheduling
Continuity of care is important, because it leads to higher patient satisfaction, increased medication compliance, improved diagnosis, and reduced hospitalization.
In order to provide timely care with the same physician (or group of physicians) a we must look at appointment scheduling. Appointments can be broadly divided into ‘acute and urgent’ and ‘nonurgent’.
The traditional scheduling technique was to book nonurgent appointments into the distant future, and schedule urgent appointment sooner. Inevitably there comes a steady state where there is no room to schedule urgent appointments in a timely matter (aka: welcome to the Canadian Healthcare System).
Open Access
To address the problem of traditional booking systems a new system of ‘advanced access’ or ‘open access’ was popularized. The goal of open access is that patients are able to book an appointment the same day - regardless if it is urgent or non-urgent.
Given how untimely booking primary care visits in Canada is, I was surprised to read that open access “has been adopted by practices nationwide [in the United States]”.
In order to provide open access appointments it is necessary to study the effects of the size of physician panels and the effects of patient queues.
2. Physician Panels
A physician panel is the number of patients a physician has in their practice. Typically this is 1500 - 200 patients.
It is important to consider the ‘case mix’ or ‘patient classification’ (as the chapter calls it) that comprises a physician panel.
Patient classifications in this panel may have higher or lower levels of ‘complexity’. This may be associated with requiring more or less appointments, or different urgency for appointment booking. Many other factors can be taken into consideration in the mix and utilization simulations such as patient age, gender, and past appointment urgency history.
To model a physician practice, the chapter assumes that a primary care physician can see a maximum of 120 patients or less a week. This is based on reimbursement for primary care in the United States at 20 minute appointment increments. 8hrs a day means 24 appointments daily.
3. Patient Queues
Patient queues are not intuitive. To start, the rate of no shows increases as patient queues increases. It seems that once patients have waited too long, perhaps they forget to show up, or forget to cancel the appointment. This leads to situations where there are long backlogs, and low utilization in clinic (because patients are not showing).
As a physician panel size increases, the length of the patient queue increases NON-LINEARLY. The simulations described in the chapter demonstrate that as the panel size increases the chance that a physician is able to book 120 or less patients as week (the assumption of full capacity) decreases. The simulations estimate the number of appointments booked each week with a mean, and standard deviation.
In the simulation with a panel of 1200 patients, there is a 99.4% chance that the expected number of weekly visits will be under 120. AKA everyone can be seen without a queue.
At a panel size of 1400 the expected number of weekly visits under 120 remains 94%. However at a panel of 1600, it drops to 54%, at 1800 drops to 5%, and at a panel of 2000 patients there is a 0.36% chance that the physician will be able to book under 120 patient visits that week. This inevitably means the cue is forming at a rapid rate.
This patient-physician booking problem is an optimization problem. It can be greatly improved by moving a physician out of a solo practice and into a group practice. Another simulation was run to study this and showed reduction in the wait time, as well as the total number of redirections to physicians other than their primary one, that patients underwent.
Emerging Trends in Primary Care
Team Care
In the United States the lower salary that primary care physicians make relative to specialists has driven many medical school graduates to pursue a career other than primary care. This has led to a significant shortage of primary care physicians.
As a result of the shortage of physicians, new models have emerged to provide care. This includes incorporating physician assistants (PAs), nurse practitioners (NPs), clinical assistants (CAs), licensed practical nurses (LPNs), patient appointment coordinators (PACs), and receptionists in the care of the patient.
Patient-Centered Medical Home (PCMH)
The patient-centered medical home is a model designed to reduce fragmentation. The primary care physician and his/her practice is the center of a patient’s care and remains in-the-loop as the patient receives care elsewhere in the healthcare system. The model incorporates non-physician caregivers and non face-to-face communications.
From this chapter, it is not very clear what actually makes a PCMH different than a ‘well run’ primary care practice. Reading the front page of several PCMH websites doesn’t really explain this difference either.