What good is hospital pricing data anyways?
As most in the healthcare world know, CMS’ rule requiring hospitals to post on the internet in machine-readable format their charge data went into effect on January 1, 2019. I was surprised that many of my friends who aren’t in the healthcare industry were posting news articles about the requirements in the last few weeks, and I will admit that my first inclination was to be dismissive about the requirement and warn my friends it doesn’t mean what they think it means. After all it’s basically the chargemaster which isn’t the price you pay anyways, assuming you have insurance. Plus I’ve read some of the research that evaluates patient decisions when they are provided with pricing tools, and it turns out that making prices more transparent doesn’t necessarily affect the choices patients make when seeking care; nor do patients always have the option of where they go for care because of narrow networks their insurer has. I think we have a tendency to believe that if something costs more, it’s of higher quality. So a pricing tool may just indicate to the general public that a hospital is providing really good care if it’s prices are higher, and I think we all know that isn’t true. Further, can the average patient even find this data on the internet as the requirement states and know how to interpret it? Spoiler: I tried on January 1, 2019 to find this data online for major health systems in my area, and the best I could find was one or two pdfs with the average price for the top 10 services. I could not find the elusive machine readable data that they’re required to post, and if I can’t find it when I know what I’m looking for, I’m guessing the average patient isn’t going to fair very well.
As I noodled on the price transparency requirements from CMS and my friends’ interest in the requirement, I started to think about what you could do with the data, particularly when you mash it up with other healthcare data, that would in fact help individuals make better choices. What if instead of allowing individuals to compare the prices of hospitals and select where they seek care, a mashup of a few different data sets allowed individuals to compare health plan options and make choices about which plan they enroll in?
How would that work exactly? Let’s start with Medicare, since Medicare data is often the most readily available. Health plans negotiate prices with each hospital/health system based on the chargemaster. When a member seeks care, the health plan processes the claim and determines what they will pay based on the negotiated rate and the individual’s out of pocket cost. Keep in mind that patients often pay not only their co-pay but a certain percentage of the bill, particularly if they have a deductible they must meet. These numbers are of course wildly different across health plans and often very opaque to patients (unless they happen to read their explanation of benefits, which let’s be honest, few do). Even with the explanation of benefits, individuals have no idea if they would paid less if they had a different insurance plan. There are of course a few price comparison tools when choosing your health plan, but most of the pricing is limited to prescription medications, and you'll know your deductible and your co-pays but you won't know the actual costs you'll be paying for care. There’s really no good way to see which plans have negotiated the best pricing with the local hospitals and healthcare providers for all other services. You certainly can look at things like what percentage of a hospital stay you’ll be responsible for, but if Healthplan A has negotiated a lower overall rate than Healthplan B, you'll still be paying more for care with Healthplan B.
CMS has started releasing Medicare Advantage encounter data. With hospitals having to make available machine-readable charge data, organizations with access to the Medicare Advantage data could combine these data sources together and identify which plans are negotiating the best rates for their members. CMS would of course need to release data that’s closer to a year old and not nearly 4 years old, but it could be done.
How would that help individuals? I have had the pleasure/responsibility of helping both my grandparents and parents enroll in Medicare Advantage plans for the last few years. Because they basically live off of social security, finding a plan that has the lowest possible premiums, while also covering their medical needs at the lowest possible price is vitally important, and nearly impossible to do with the current plan compare tools. In other words, if the premium is low, but their out of pocket costs for some chronic conditions are high, then the plan won’t work for them, but if you knew that Health Plan A negotiated a much lower rate than Health Plan B (even if it was a simple percentage difference), than you could make a much more informed decision about which plan to pick. Imagine seeing a compare tool where you could know quickly which plan was the best negotiator overall. Such data mashups could also be done for Medicaid and CHIP, using the T-MSIS data, which includes Managed Care Organization data, though it’s unclear if such a data mashup could be done with private plans since their pricing data is heavily guarded as a trade secret. States do have all payer claims databases which contain private payer data; however, many of state legislatures limit what you can do with the data. Some states may already allow this type of analysis by approved organizations.
Obviously, this isn’t a simple data mashup to do and will be limited by the data available, though with CMS’ commitment to price transparency perhaps they’ll be willing to make more cost data available. It could also have unintended consequences to the overall market (i.e. making negotiations harder for some parties, and potentially driving up the chargemaster prices), but it would be interesting to see someone try to make use of data that is otherwise, unhelpful to the average consumer.
The goal of price transparency is ultimately to lower costs for individuals and the system as a whole. It’s not clear that price transparency leads to individuals making different decisions about where they seek care. However, price transparency data that helps individuals choose health plans more wisely could still lead to a lower cost of care, particularly for individuals. It’s easy to look at the data being made available and complain it’s not enough (is it ever really going to be enough?), rather, let’s all think outside of the box about how we can make the most use of that data with the tools we have.