Update: On January 5, 2015, Stephen Brill published America’s Bitter Pill: Money, Politics, Back-Room Deals, and the Fight to Fix Our Broken Healthcare System, expanding on his popular Time article.
It is an interesting book, not because I agree with Mr. Brill’s data but because it happens to be a useful exercise in how data sometimes can be applied/misapplied to streamline a compelling narrative.
Below is a shameless self-reblog from my response to the original Brill article focusing on how data from a sub-sub-sub-specialized hospital should not be used to comment on an entire set of hospitals, some of which struggle to serve as the safety net for a vulnerable population.
In an article titled “Bitter Pill: Why Medical Bills are Killing Us,” Stephen Brill outlines a well-researched investigation on hospital over-billing. In the article, Brill begins by highlighting the unreasonable mark-up MD Anderson places on every medication, service, and imaging that it provides. He argues that this “hard-nosed approach pays off,” earning MD Anderson $531 million operating profit in 2010, and that this comprises a 26% operating margin. $1.8 million of that went to the pockets of Ronald DePinho, the president of the cancer center.
Although Brill never outright states the connection, his implication is clear: general hospitals are the oft ignored mammoth in the health care debate, operating under the veil of legitimate non-profit business. A general hospital funds its astounding operating income by making the uninsured and under-insured suffering patients an offer they cannot refuse – a markedly inflated bill. It then funnels this unfairly earned profit into the pockets to the Godfather of the organization.
The logic is sensational, if only it were correct.
Putting MD Anderson on the Map
As a practicing physician in a general hospital, I was appalled by the possibility that my employer – a nonprofit organization like MD Anderson – could be part of this incredible scheme to squeeze the patients whose medical interests I swore an oath to protect. In fact, if MD Anderson is representative of a typical hospital, it would mean that general hospitals in America are so unfairly profitable that Barack Obama had been chasing after the wrong villains all along.
I had to make extra sure. Exactly how profitable is the typical general hospital?
The commonwealth of Pennsylvania has a Health Care Cost Containment Council (http://www.phc4.org/) where financial information for all general hospitals in the state is collected. It contains a wealth of information divided by region, and offers data such as total revenue, operating profits and net income for each. Despite the state-to-state differences in health care economics, I chose to use Pennsylvania instead of Texas because (1) PA makes its health finances data very easy to access, and (2) there is no freely available, easily formatted national hospital database in my possession, and (3) I care about PA because I live in it.
Because I am probably unconsciously biased towards general hospitals as a physician that works in one, I will do my best to be as transparent as possible. All the data I used for the calculation is publicly available here. I used the 2010 data because it is more comparable with Mr. Brill’s data on MD Anderson.
Descriptive Statistics for hospitals in Pennsylvania:
- Number of Hospitals Included: 161
- Average Operating Revenue: $222 million
- Average Operating Expense: $212 million
- Average Operating Income: $10 million
- Average Operating Profit Margin: 2.5%
The hospital with the largest operating revenue in Pennsylvania is Hospital of University of Pennsylvania, with revenue of $1.9 billion, operating expenses of $1.75 billion, making an operating profit margin of 9.7%.
The most profitable (highest operating profit margin) hospital in Pennsylvania is Easton Hospital (Easton, PA), with revenue of $213 million, operating expense of $159 million, comprising an operating margin of 25.3%.
Hospitals with the highest operating incomes:
- Hospital University PA ($188.7 million)
- Children’s Hosp Phila ($108.8M)
- UPMC Presby Shadyside ($96M)
- Lancaster General ($61M)
- Reading ($60.5M)
- Milton S Hershey ($57.7M)
- Thomas Jefferson Univ ($56.2M)
- York ($55.4 M)
- Easton ($54M)
- Geisinger/Danville ($53.6M)
- Magee-Womens UPMC ($49M)
Finally, I created two graphs, one for operating income (in raw dollars), and one for operating margin.
The data from the financial data of Pennsylvania hospitals suggest that MD Anderson not only makes more money than the largest hospital in Pennsylvania by almost a factor of three ($531M vs $189M). Indeed, MD Anderson would be literally off the chart if plotted. It is also more profitable than every hospital in Pennsylvania.
The 2010 data also reveals that many hospitals in PA make very low margins. In fact, one third of the hospitals are money-losing propositions. While some may argue that MD Anderson’s prestige and sheer size explains its profitability, some of Pennsylvania’s largest, most prestigious hospitals are losing money. In 2010, University of Pittsburgh Medical Center children’s hospital was among the least profitable hospitals in the state, as was Temple University Hospital. Although no meaningful data exists to directly measure quality of care for an entire state, it is reasonable to assume that hospitals that deliver quality care can be money-losing businesses as well.
Mr. Brill’s article is well-researched and backed by strong story-telling. However the causal connection behind over-billing and profitability is far more nebulous. Given that the “average hospital” operates at a thin operating margin of 2.5% – compared to other service industries such as hotels (12%, according to Value Line’s Industry Analysis) and restaurants (12%) – one may conclude that the average hospital is making a thin margin despite over-billing its patients.
Before we point fingers at general hospitals for over-billing the uninsured and underinsured patient, it is worth pondering whether this may be an effect rather than the cause of a highly dysfunctional reimbursement system.
[ Part 1 | Part 2 ]