What's wrong with the US Healthcare system?
A few key insights from "Priced Out" - the late Uwe Reinhardt's new book
I had always sort-of known the American healthcare system was problematic, but never really knew exactly how or why this was the case. As a result, I’ve always felt rather uninformed about the healthcare debate.
Uwe Reinhardt’s book Priced Out has changed that a bit for me - it offers an extremely clear, data-driven, and politically agnostic exploration of the American Healthcare system. In this post I’ve decided to summarize a few of the key points of the first few chapters of the book, which give a high level overview of the cost-effectiveness of our current system:
[1] Healthcare in the US costs way more than it should
Healthcare services are on average twice as expensive in the US as they are in other developed countries, and it’s consistently getting even more expensive (relative to income) in the US year over year.
Across any reasonable metric, the US spends significantly more (about double) on healthcare compared to other developed nations.
This is true on per-capita purchasing power parity adjusted dollar terms (US spends ~$10k per capita while other nations spend ~$5k), true on overall percentage of GDP spent (US spends 17% while other developed countries spend 11% or less), and true when you look at percentage of median household income spent per person (accounting for taxes + consumer out of pocket/premium payments). These stats indicate our spending is way higher than other countries even relative to our higher income.
Furthermore, the data indicates that this high level of healthcare spending isn't because the US population is older or sicker than other countries (it isn’t), or because the US population consumes more healthcare services than other countries (we don’t), or because the US healthcare services on average are superior/of higher quality (they aren’t).
The truth is that the same exact healthcare services, whether it’s a knee surgery or a prescription to a specific drug, cost significantly more in the US than they do in other countries.
[2] Why so expensive? It boils down to higher administrative costs, and lack of downwards pressure on prices
Why is our healthcare so much more expensive? While Reinhardt points out many different issues, in my opinion most of his points boil down to two fundamental reasons that are intertwined:
[a] high administrative costs (ie a burgeoning of non-healthcare producers who work in and are now vested interests in the healthcare system) that have blown up due to a number of reasons, and
[b] absence of demand-side or regulatory downwards pressure on prices (due to consumer inability to “shop” around), which enables perverse incentives on the supply side (healthcare providers charge as they see fit and/or recommend unnecessary services)
Higher Administrative Costs
Between 1990 to 2012, the number of workers in the US healthcare system grew by 75%, but 95% of this growth was in non-doctor workers. For every doctor in today’s healthcare system, there are 16 non-doctor workers, of which only 6 have clinical roles (ie nurses, care coordinators, aides, etc). It’s estimated that the average American physician spends $80k per year (in employee hours) interacting with health insurers. One specific illustration of this - Duke University has 957 beds and 1600 billing clerks.
The data shows that these absurdities are simply not the case in the healthcare systems of other developed nations, which have significantly lower non-healthcare producers employed.
There seems to be a few different reasons for the spike in administrative costs. A big one lies in the structure of our insurance system - the US healthcare insurance system has become an unnecessarily fragmented “system of systems” over the past 30 years, each with its own administrative procedures and overhead - all of which is largely unnecessary. Some examples:
Employed individuals are offered community-rated insurance plans from private insurance companies via their employers.
Small business owners/non-employed individuals can purchase plans directly via independent exchanges (ie Obamacare).
For the old and the poor, there are the government subsidized/negotiated plans of Medicare and Medicaid. Furthermore, under Medicaid there’s a separate supplemental government-subsidized insurance program for children (CHIP), which is administered by states.
Veterans are served separately by the VA Hospital System, the only socialized medical program in the US.
You might read about multiple insurance systems and wonder how this is a bad thing - more competing systems means increased efficiency and better results for everyone, right? Well, as I discuss in the next section, healthcare consumers aren’t really in a position to shop around, so these different entities aren’t really competing with each other.
Each program has its own way of doing things, its own administrative costs, and its own set of employees/stakeholders who have a vested interest in keeping it alive. For example, each insurance company and government program has different negotiated rates and different systems for paying providers, which means that providers require their own armada of administrators simply to handle interfacing with each of the insurance systems. Here was a particularly illuminating quote:
“Doctors and hospitals contract with armies of coding specialists to help them code their work in a way that extracts the maximum revenue from the rest of society (private and public insurers as well as patients). For their part, private insurers contract with equally adept coding experts to help them fend off up-coding by providers of care”
Essentially, we’ve had a zero-value generating administrative arms race between insurance companies and providers, that’s ultimately being paid by taxpayers and healthcare consumers.
There is also a negative cyclical relationship between administrative complexity and incremental healthcare reform. Well-intended incremental regulations often end-up adding an administrative “tax” on the system that can be significantly higher than the benefit provided (if any) by the regulation. One example is the introduction in the early 2000s of HSAs, which like 401ks allow employees to contribute a certain amount of pre-tax income directly to a fund that can be spent only on out of pocket medical costs. While they have become popular over time, the benefit of HSAs is quite questionable, as they are a regressive subsidy - those who earn more get the largest tax subsidies for healthcare spending. However, what’s worse is that the growth of HSAs has created a mini-industry of HSA account providers and administrators who, as Reinhardt likes to put it, “feed themselves at the healthcare trough,” since their salaries are ultimately paid by us via higher insurance premiums and deductibles.
Little Downwards Pressure on Prices
In the US, pricing for healthcare services is almost entirely opaque. It’s basically impossible to know what you’ll end up paying when you visit a hospital - care providers aren’t obligated to tell you how much a service might cost, even if you demand to know.
It actually turns out there really is no consistent “price” for a service. Each hospital or primary care provider has an internal charge master (not shown to patients) which is a document that contains, for every service offered:
[a] the “list” price for that service that someone uninsured would pay - typically obscene
[b] the “negotiated” in-network price for each insurance company and each insurance plan, and
[c] a negotiated “Medicare” charge price. Each hospital will have its own charge master, the final product.
Just another example of the “value-generating” work the hospital and insurance administrative staff are doing.
Practically, this means that two hospitals right across the street from each other could charge you wildly different amounts for a certain procedure, or a single hospital could charge two patients who have similar health insurance but from different insurance companies wildly different amounts for the same service.
When you combine the fact that prices for services are undisclosed and highly variable, with the fact that often times medical services are critical and time-sensitive, it starts becoming clear why healthcare consumers (patients) have such little pricing power against the supply-side. The additional intermediary of the insurance companies only further obfuscates pricing - as a patient I might willingly undertake unnecessary services that my doctor recommends without knowing their cost-effectiveness, as long as I believe my insurance company will pay for it.
Thus, unlike properly functioning industries where prices are determined jointly and roughly equally by the supply-side and the demand-side, healthcare is currently one where the “demand” side’s pricing power is currently significantly weaker.
Doctors and providers today have little personal incentive to keep costs down for their patients. They can largely prescribe whatever services they’d like and charge fees as they wish (as long as someone on their side is making sure that someone on the other side will pay). Atul Gawande profiled the discrepancies in how healthcare services are charged and prescribed in his famous 2009 New Yorker piece, The Cost Conundrum.
As Charlie Munger famously said, “show me the incentive and I will show you the outcome.” While there are exceptionally principled doctors and hospital systems that strive to keep costs down for patients, such as the Mayo Clinic, it’s not the norm. And we can’t expect it to be given the current incentives for providers. Overprescribing services has become the rational thing to do, even for well intentioned providers.
For example, suppose I were a rational self-interested doctor considering whether I should recommend a CAT scan for a patient when I know there’s a 99.5% chance it won’t reveal any useful information. What reason would I have NOT to recommend the scan? I’m not paying for it. And if I don’t refer the patient for a CAT scan and it later turns out that the scan would have revealed some critical information, there’s a risk the patient could sue me. And of course, referring the patient to my radiologist friend means I’ll get a kickback of some form. There is no incentive for me not to recommend the scan (which is the sensible, cost-effective decision for the patient).
A few final thoughts:
One silver lining is that the aforementioned problems can, at least, be thought about independently of the tougher, more fundamental healthcare debate around the equitable distribution of healthcare costs.
Uwe Reinhardt maintains that the heart of the healthcare debate in the USA is the argument around the distributive ethics question:
“To what extent should the better-off members of society be made to be their poorer and sicker brothers’ and sisters’ in health care?
However, I’d point out that regardless of our positions on the above question, we should all agree that the inefficiencies that cause us to collectively pay twice as much as most other countries for healthcare need to be addressed. The unnecessary administrative armadas and the lack of downwards pricing pressure on providers need to be addressed.
In a future post, I plan to list out some brainstormed ideas on how we could address these problems.