By Robert Rohatsch, MD
Throughout my career as an emergency medicine physician and as an entrepreneur in urgent care medicine, I have had the pleasure of working with many nurse practitioners (NPs) and physician assistants (PAs) in various staffing models. In the emergency department (ED) during residency, they were typically assigned to a section of the department where lower acuity cases were triaged. In my program, they were almost exclusively NPs. They functioned fairly independently from the physicians but had full access when they had questions. Their scope of practice in the ED was controlled by the triage nurse with protocols established based on the chief complaint of the patient. There are some studies that suggest nearly 15% of all patients seen in an ED are seen by either a PA or NP. After residency, as an attending physician in emergency medicine, I worked with primarily PAs under two different staffing models. In one large urban ED, I worked with PAs assigned to a ‘fast-track’ area of the ED and had patients triaged to them in a similar manner to my residency experience. In a smaller ED I worked in, I worked along side a PA and we saw patients in an alternating fashion with no triaging based on acuity. My experience as an ED physician indelibly etched in my mind their value in an acute care setting.
The US health care system is the most costly in the world, accounting for an estimated 20% of the gross domestic product by 2020. In an attempt to mitigate this alarming statistic, the Institute for Healthcare Improvement (IHI) developed the Triple Aim framework in 2007.
Improving the patient experience
Improving the health of populations
Reducing the per capita cost of health care
So how can we as operators of walk in healthcare services work towards the IHI goals? Can something as simple as a staffing model change yield improvements in all three areas? How do the policies of payers, government agencies, and other competing interests confound the ability to make a strategic decision on provider staffing models? A review of the literature on the subject provides some interesting results...