In a previous posting I established the need to define a theoretical optimal health insurance system with which to compare to the several legislative proposals now circulating. This posting identifies a source of data, the types of healthcare services covered, how much is spent on them and how many people are truly uninsured.
To estimate the costs of an optimal system, we need data and the latest publicly available data on health expenditures and insurance comes from the 2016 Medical Expenditures Panel Survey (MEPS) produced by the Agency for Healthcare Research and Quality (AHRQ). Using this survey, we can estimate the cost of providing health insurance to those who were uninsured in 2016 and then to those who would have been uninsured in 2016 had the subsidized health insurance exchange market and its individual mandate (aka Obamacare) not existed.
What Kinds of Healthcare Should Be Covered by Insurance?
There are several different kinds of healthcare. There are traditional services such as surgery and office visits and non-traditional services such as acupuncture and hypnosis. Some have proven to be highly effective (or debunked) by randomized clinical trials. Most have never even been tested and continue to be administered simply out of tradition. Should all of them be covered by health insurance?
Establishing an objective standard for health insurance coverage is beyond the scope and capacity of this study. So, I’ll simply include the six categories that are normally covered by health insurance and tracked by MEPS: inpatient stays (including physician fees), office visits, prescription drugs, outpatient visits, home health, and emergency department visits. This excludes other healthcare services like chiropractic and optometry because those kinds of services are not generally covered by health insurance, but they are tracked by MEPS. In any event, the six covered categories comprise 89.9% of the $1.6 trillion spent on healthcare in 2016.
The big three are inpatient stays, office visits and prescription drugs with each accounting for nearly a fourth of total spending. In spite of what you might have read, we spend about the same on inpatient stays as we do on office visits. This happens because although the average hospital stay is very expensive, only a small fraction of the population experiences a hospital stay in a given year. So, controlling the cost of hospitalization is not significantly more important than controlling other types of healthcare.
How Many People Are Truly Uninsured?
Identifying the uninsured population via a public survey might sound like a straightforward task — simply ask the respondents if they have health insurance? –, but the truth depends on what it means to be uninsured, that is, on your own with no other source of payment. According to the 2016 MEPS, an estimated 24.6 million U.S. residents (7.6%) claimed to have no health insurance. These are the nominally uninsured. However, only 13.3 million of these people (a little more than half) had any healthcare expenditures that year and most of their healthcare expenditures were paid by somebody else, like the Veterans Health Administration (VHA) or Worker’s Compensation (WC). To be precise, 82% of the $35 billion in insurable healthcare expenses incurred by the nominally uninsured were paid by entities other than the patients themselves.
So, were all of these 24.6 million nominally uninsured people really uninsured in the true sense of the word? While organizations like the VHA and WC may not provide insurance to individuals and families, they are de facto insurance providers and should not be ignored when estimating the cost of providing health insurance to the nominally uninsured.
Prospectively, that is, at the start of the covered time period, we can only guess how much healthcare expenses will be incurred by a group of people. But since this is a retrospective analysis, we need only consider nominally uninsured people who had healthcare expenditures in 2016 and received no outside help in paying their healthcare bills, in other words, the truly uninsured. This reduces the number of uninsured from 24.6 million down to only 4.0 million people. This core of truly uninsured people incurred healthcare expenses of $5.5 billion of which they were able to pay only $1.4 billion (25%).
The rest of these incurred expenses were not collected by the hospitals, doctors and other providers of healthcare. This effectively amounts to a 75% price discount payed by the uninsured (as a whole, but with a great deal of variation by patient). That percentage discount might sound large, but it should be noted that Medicare and private insurers also pay large discounts on average and the amount of expenses incurred (i.e. billed charges) are much greater than the actual costs incurred by hospitals, etc. for providing care.
So, only 13.3 million of the 24.6 million nominally uninsured actually incurred healthcare expenses in 2016. The remaining 11.3 million had no healthcare expenditures during this time period. How many of them were truly uninsured? In other words, how many would have had to pay the full cost of their healthcare if they had had any? It is impossible to know for sure, but it is safe to say that it is likely that many of them could have counted on the VHA and other payers if they had incurred healthcare costs. The most we can say is that the truly uninsured ranged from 15.3 million to 4.0 million (4.7% to 1.2%) with the exact percentage likely being somewhere near the middle of that range. This is a far cry from the 7.6% who stated they were uninsured in the survey and the $5.5 billion in incurred expenses are a far cry from the $35 billion incurred by everyone who self-identifies as being uninsured.