Medicare Advantage and provider pricing

There is a really nice little paper in this week’s JAMA that looks at what a single major Medicare Advantage insurer pays its providers.**

physician reimbursement in Medicare Advantage was more strongly tied to traditional Medicare rates than to negotiated commercial prices, although Medicare Advantage plans tended to pay physicians less than traditional Medicare. However, Medicare Advantage plans take advantage of the commercial market’s favorable pricing for services for which traditional Medicare overpays, including laboratory tests and durable medical equipment….
mean MA reimbursement ranged from 91.3% of TM for cataract removal in an ambulatory surgery center (CPT 66984; 95% CI, 90.7%-91.9%) to 102.3% of TM for complex evaluation and management of a patient in the emergency department (CPT 99285; 95% CI, 102.1%-102.6%)….
for laboratory services and durable medical equipment, where commercial prices are lower than TM rates, MA plans take advantage of these lower commercial prices, ranging from 67.4% for a walker (HCPCS code E0143; 95% CI, 66.3%-68.5%) to 75.8% for a complete blood cell count

This is a big, comprehensive study with very tight error bands. I find it persuasive with one caveat that I’ll get to in a minute. The big result is that the administrative prices set by Medicare Fee for Service matter a lot on procedure codes. Medicare probably currently overpays for labs and durable medical equipment.

I think this is good evidence that Medicare Advantage (MA) is able to get a better deal than Traditional Medicare (TM) because they have the ability to say no to providers. Medicare Advantage can create networks that are only enforced by cost sharing arrangements where they try to drive their members away from low value providers. From a policy perspective, this is a win as we want to force low value providers to stop being low value providers. That change is that they either go out of business or they get better or less expensive.

From a political angle, this finding along with other evidence that Medicare Advantage providers are getting fairly paid by the Federal government makes the long run case of a Medicare Advantage for all style universal coverage schema more plausible.

I do have one reservation about this study and it is the generalization of the findings.

The data is impressive. It is national in scope.

We analyzed claims data for MA and commercial enrollees from a large national insurer operating in both markets from 2007 through 2012. In 2012, the insurer held 17% of nationwide MA market share and offered 1 or more MA plans in 98% of counties, in which 94% of Medicare beneficiaries lived. The data include the full set of adjudicated and paid claims for all enrollees; enrollment increased from 1.7 million in 2007 to 2.6 million in 2012.

It is from a single insurer.

They acknowledge this limitation but I think they could be stretching it a bit far:

Our findings have some important limitations. First, the MA and commercial data included in our study are only from 1 insurer and therefore may not necessarily be representative of the experience of other private insurers or of geographic locations not served by this particular insurer. However, the insurer has a large presence in the MA market and thus the findings are reflective of common pricing patterns in the MA market.

Roberts, Chernew, McWilliams ^^ in January 2017’s Health Affairs has an article that I’ve gone back to a few times which looks at the influence of market power on pricing:

Using multipayer claims for physician services provided in office settings, we estimated that—within the same provider groups—insurers with market shares of 15 percent or more (average: 24.5 percent), for example, negotiated prices for office visits that were 21 percent lower than prices negotiated by insurers with shares of less than 5 percent.

Market share matters. The data for the JAMA article is from a medium to high market share insurer. They should be able to get a good rate from most providers. They should be able to get a better rate than typical in most markets than most insurers. This is merely a quibble. I think the general direction of the finding is real. It is in concordance with the CBO research using HCCI data.

Here is where the Health Care Cost Institute data set is hyper valuable. They actually have a majority of the Medicare Advantage claims universe from their data contributors. I am assuming the JAMA data set is not an HCCI contributor so combining these data sets gives us 70% or so of the Medicare Advantage universe. At that point there is deepness and richness to exploit.

** Trish E, Ginsburg P, Gascue L, Joyce G. Physician Reimbursement in Medicare Advantage Compared With Traditional Medicare and Commercial Health Insurance. JAMA Intern Med. Published online July 10, 2017. doi:10.1001/jamainternmed.2017.2679

^^ Roberts, E. T., Chernew, M. E., & Mcwilliams, J. M. (2017). Market Share Matters: Evidence Of Insurer And Provider Bargaining Over Prices. Health Affairs, 36(1), 141-148. doi:10.1377/hlthaff.2016.0479