Large coverage expansions under the Affordable Care Act have reignited concerns about physician shortages. The Association of American Medical Colleges (AAMC) continues to forecast large shortfalls (130,000 by 2025) and has pushed for additional Medicare funding of residency slots as a key solution.
These shortage estimates result from models that forecast future supply of, and demand for, physicians – largely based on past trends and current practice. While useful exercises, they do not necessarily imply that intervening to boost physician supply would be worth the investment. Here are a few reasons why.
1. Most physician shortage forecast models assume insurance coverage expansions under the ACA will generate large increases in demand for physicians. The standard underlying assumption is that each newly insured individual will roughly double their demand for care upon becoming insured (based on the observation that the uninsured currently use about half as much care). However, the best studies of this – those using randomized trials or observed behavior following health insurance changes – tend to find increases closer to one-third rather than a doubling.
2. A recent article in Health Affairs found that the growing use of telehealth technologies, such as virtual office visits and diagnoses, could reduce demand for physicians by 25% or more.
As the Affordable Care Act’s (ACA) exchanges open and Medicaid expansion takes effect, millions of uninsured Americans will gain new coverage. This raises a key question: how are we possibly going to meet the demands of all of these new individuals entering the system? The physician workforce is growing slowly, at best, at a time when an aging population is increasing demand for care.
Predictions include long lines for everyone, rising prices and premiums as physicians are able to command greater market power, and reduced quality of care. Some have recommended additional government funding to help train more medical residents as a response.
But while studies predict ACA implementation will prompt an increase in demand for medical services, there is evidence that the increase in demand will not be as great as the raw number of newly insured Americans might suggest.
The latest CBO forecast projects the reduction in the number of uninsured Americans under the ACA will be 11 million people next year and 24 million by 2016. That’s an increase in the percentage of Americans with insurance of roughly 5% in 2014 and 12% in 2016. If the uninsured used zero health care today, but upon becoming insured used the same amount as a typical insured person, then the increase in demand for care would be the same as the increase in coverage.
In reality, the uninsured use substantial amounts of health care – but only about half the care that the insured use today. One reason is because they are uninsured – paying full prices for care rather than a small copay discourages use. Another reason also explains why many (but not all) are uninsured in the first place: they are healthy and don’t anticipate needing or wanting medical care.
When the uninsured gain coverage, demand does increase, but not dramatically, studies show. Evidence from the Oregon Health Insurance experiment, in which a funding cap forced the state to grant Medicaid coverage to some applicants but not others using a lottery-type system, found that those who did gain coverage increased their use of both hospital and physician care by about one-third relative to controls.