Reference pricing is a common payment reform for commercial insurance. Common, non-urgent, deferrable procedures are prime candidates. Hip replacement is the classic example in California. There, CalPers sets a flat fee for hip replacement. If a covered individual chooses a hospital that has a price below the benchmark, CalPers pays the full benefit minus any normal cost-sharing. If the hospital price is above the benchmark, the individual pays normal cost sharing plus the gap between the reference price and the total price.
It is a scheme that delivers pain to non-preferred behavior. Behavior has changed. Hospitals over the benchmark reduced their prices and people shifted hospitals.
It works partially because employer sponsored insurance has built in expectations of high deductibles and co-payments and more importantly, employer sponsored coverage is for people who have some income and probably some assets that can cover a one time expense.
Reference pricing as it is currently built is problematic for Medicaid. Medicaid covers people who don’t have significant assets nor income. Furthermore, Medicaid is strictly limited in the out of pocket costs it can impose on individuals. Non-waivered Medicaid may have only de minimis cost sharing. Waivered Medicaid limits cost-sharing to no more than 5% of a family’s income. Traditional reference pricing can’t work if the total incremental cost of going to a high cost provider is capped at an extremely low level.
However, there is a potential tweak that could make this alternative payment system work well for Medicaid as well as incorporate my argument for three way gain sharing.
Medicaid can not use paid as a tool for cost minimization. However, it is possible for Medicaid to use gain as a motivator.
What if Medicaid used reference pricing but inverted the incentive structure?
For instance, a Medicaid managed care organization could have a network of fifty hospitals. Twenty of the hospitals have agreed to take a reference priced bundle for knee replacement surgery that is slightly more than the median bundle payment for knee replacements but less than the regional average knee replacement bundled price. So far, this is the same scheme as commercial reference pricing.
The difference is motivating factor. A commercial plan would make a member pay more if they went to a high cost provider. Under my plan, the Medicaid beneficiary would receive a bonus check for a good choice. The check would be a portion of the difference between the bundled reference price and the regional average price for the knee replacement. For instance if the bundled price for a Medicaid knee replacement is $15,000 and the regional average price is $18,000, the patient could get a check for $1,000 for choosing a low cost, high quality facility. If the patient chooses to have their surgery at a high cost facility, they would not receive a bonus check. The check, in an ideal scenario, would not count against Medicaid eligibility income limits for the next re-determination period.
The goal is the same, move people to low cost but highly effective facilities and either make high cost facilities lower their prices in order to compete, or significantly reduce the number of procedures performed at high cost facilities.
There are a couple of potential problems with this scheme. The first is that Medicaid tends to pay low, so a hospital may only allocate a fixed number of slots for a certain procedure to Medicaid. This is not a problem for commercial payers as commercial tends to pay much higher rates. If the gain sharing system brings down total system costs, the average payments to participating reference priced providers could increase. We saw that providers were willing to accept more Medicaid patients when primary care payments increased to Medicare levels, so increasing surgical reimbursement for select providers on select service bundles would also lead to more availability.
Secondly, the political optics are a tough sell. The scheme may save the entire system money while delivering the same care at the the same or higher quality at lower costs. That is very attractive from a health wonk point of view. Health wonks are not a decisive voting block. The scheme would be easy to demonize on a thirty second ad as poor people would get checks for having surgery while middle class taxpayers have to pay their deductible and perhaps more for the same surgery. Building and sustaining political support would be difficult. This is a scheme that would have to be run in a very Blue state that is open to healthcare experimentation.
What other problems do you see with this idea?