In a recent grand rounds on the future of medicine, the buzzwords were “collaboration” and “managing of the health of populations.” The same day, a group of ten residents were presented with their patient data about cancer screening rates. In both venues, the call to “population health” elicited sighs of exasperation. It’s just another checkbox we are being asked to click off. How can we be assessed on that as well?
In the last five years in medicine, the initiatives to teach quality improvement (QI) have gained money, resident interest, and distinct educational pathways. Most of QI research has focused on inpatient initiatives, partly by virtue of the amount of resident time devoted to inpatient rotations. Consequently, the notion of ambulatory QI and management of populations (cancer screening, immunization, diabetic panels) feels foreign and even invasive to residents.
Why is inpatient QI perceived as more exciting, and what can we do about it? I offer three models.
1. Treatment is more satisfying than prevention. Take two common quality initiatives: initiating a new protocol for treating sepsis and developing a new system for tracking colonoscopies. At the end of the day in the ICU, watching a very sick septic patient improve is extremely gratifying; tracking an increase in colonoscopies less so.
2. The hospital is a neater laboratory than the world. When we looked at our resident data in the clinic, we talked about barriers to patient care: no-show appointments, non-English speaking patients, and lack of translated patient education materials. The social world of patients is messy and requires a non-medical gaze to respond to cultural, socioeconomic, and educational diversity. A medical gaze goes further in the hospital, where most data can be captured and manipulated in the EHR.
3. Inpatient data is more complex and dynamic than ambulatory data. For better or worse, the modern EHR contains a near labyrinthine repository to analyze. In the inpatient setting, there is a near constant flow of new information to decode and analyze. In contrast, the common metrics for ambulatory improvement as few (screening rates, number of visits). Not only this, but the salient variables that impact patient health are harder to put into numbers (i.e., a patients’ readiness to change, their trust of the medical establishment, their ability to pay for next months medications).
What can we do about it?
First, we need to reframe the importance of preventative measures so they are more powerful. If a hospital initiative to treat sepsis concludes “we saved 100 lives this year from sepsis” we need to use the language for preventative care: “In their career, a primary care doctor will save ten lives from colon cancer — will this be you?”
Second, we need to teach medical students and residents how to approach medical problems with a non-medical gaze. This is easier said than done, but requires building upon existing skills and capacities of residents (i.e., as economics majors, multilingual speakers, volunteers, etc.).
Last, we need to develop a scientific curiosity about how to screen and measure social determinants of health. Maybe no-show rates are correlated with fewer interactions with medical office staff, or age, or severity of illness. We currently do not have the informational infrastructure to ask these types of questions. With these changes, we the words “population health” might not exasperate us, but lead to a renewed sense of purpose and scientific curiosity.
Tom Peteet is an internal medicine resident.