Long continues the debate of what impact hospitalists have on inpatient outcomes. This issue has been playing out in the medical literature for 20 years, since the coining of the term in 1997. In the most recent iteration of the debate, a study was recently published in JAMA Internal Medicine entitled “Comparison of Hospital Resource Use and Outcomes Among Hospitalists, Primary Care Physicians, and Other Generalists”. The study retrospectively evaluated healthcare resources and outcomes from over a half million Medicare beneficiaries hospitalized in 2013 for 20 common DRGs, by type of physician provider (hospitalist, their primary care physician, or other generalist). The study found that non-hospitalists used more consultations and had longer lengths of stays compared to hospitalists. In addition, relative to hospitalists, PCPs were more likely to discharge patients to home, had similar readmission rates, and lower 30-day mortality rates; but generalists were less likely to discharge patients home, had higher readmission rates, and higher mortality rates).
This study makes a compelling argument that longitudinal contact with patients may translate into different care patterns and outcomes (e.g. length of stay, discharge disposition, and even mortality). Importantly, this study was the first to distinguish between PCPs familiar with patients versus generalists without prior familiarity in the outpatient setting. However, the authors do acknowledge that, as with any observational study design, unmeasured confounders could contribute to the results, and they call for further research to understand the mechanisms by which PCPs may achieve better outcomes. Given that this patient population was Medicare (and the average age was 80 years old), it may very well be that having deep historical knowledge of such a patient population is required to produce better outcomes.
As hospitalists, we need to understand and acknowledge that most of our patients are “brand new” to us, and it is paramount that we use all available resources to gain a deep understanding of the patient in as short a time as possible. For example, ensuring all medical records available are reviewed, at least as much as possible, including a medical list (including a medication reconciliation). Interviewing family members or caregivers is also obviously a “best practice”. As well, having the insight of the PCP in these patients’ care is unquestionably good for us, for the PCP and for the patient. With good communication processes and an eye for excellence in care transitions, hospitalists can and should achieve the best outcomes for all of their patients. I look forward to more research in this arena, including a better understanding of the mechanisms by which we can all reliably produce excellent outcomes for the patients we serve.
Been on both sides of the fence, a hospitalist for the VA, 1980-88 out of residency, before there were hospitalists outside the VA, then fellowship, then endocrine practice where for a while I admitted my own patients to myself as attending, finally opting for convenience, maybe efficiency and controversial outcome in the late 1990’s as hospitalists took over the management of the hospital patient. They scheduled diagnostics more efficiently, took paperwork in stride without complaining about discharge summaries and suspension lists, and unfortunately perhaps did not get bogged down in detail like which endocrinologist has been taking care of that diabetes for the last ten years.
As I look at outcome reports, they focus on the few days in the hospital. And from that perspective, sick people have an advantage in having their doctor around during the days they are ill. However, a chronic illness that deteriorates or a new illness amid a chronic illness is really like walking into the middle of a movie without all that much thought as to what went before or what comes next. The PCP’s admitting their patients or covering a colleague seemed to do a lot better at history taking. Now as a specialist, I work the hospital as a consultant and see those patients back in the office. Unfortunately, I too often see patients back in the office whose diabetes was neglected while they were in the hospital or for some dubious reason their key medicines were stopped with no reasoning as to why in the discharge summary, and often not in the progress notes either.
I think one of the faults of hospitalist performance analysis may be that it stops at discharge while the decisions made during that hospital stay carry forward probably to the second specialty visit, the first spend figuring out and cleaning up, the second to see if the office decisions restored that patient to where they were before the time in the hospital.
It is much harder to analyze performance that way, of course, but it would be a far better measure of what seems a very mixed result to those of us who still work both the office and the hospital