Physician practice patterns and network design

Walid Gelad and many others have a recent article in the Journal  of General Internal Medicine** that looked at the different rate of emergency department physicians prescribing opioids in the Veterans Administration.  This is a replication of Barnett’s excellent work from 2017## that examined emergency department prescribing patterns to Medicare beneficiaries who were opioid naive.##

Walid highlights this graph as a key take-away.  There is huge variation in physician behavioral patterns when faced with similar patients.

I want to go in a slightly different direction than an opioid analysis with this graph. I want to go into network design. The Veterans Administration is a singular entity under, theoretically, singular management control. They have a good to excellent electronic medical record system. The opportunities for integrated care are fairly high. They have a much longer shadow of the future than most health financing entities. There are numerous opportunities present in the VA that most other systems in the United States don’t have. And they still have tremendous variation.

Insurers build narrow networks based on two primary objectives: price per unit control and total cost control. Some insurers will leverage reasonable competitive local clinician and hospital markets to get a dirt cheap rate per unit. The objective is to minimize price per unit without caring too much about how many units are paid for.

The other system is not as sensitive to the price per unit. Instead, the network is designed around variation in provider practice patterns on similar patients. The network if it was optimizing on minimizing opioid exposure in the ER would be built around the docs in the first quartile or at least built to avoid the docs in the most frequently prescribing quartile. This takes advantage of the fact that medicine for common conditions is still a folk art. There can be tremendous variation within the same office much less the same town on following evidence based recommendations. Building networks around providers who practice in a particular way that could conceivably lead to higher quality and lower total costs is a viable strategy.

Communicating the value proposition of a higher per unit cost but lower net cost is a challenge in the ACA domain given the subsidy structures. I think this type of network design choice could work far better in Medicare Advantage and large employer self-insured groups.
Variation in practice is widespread and common and it can be targeted.

** Barnett, M.L., Zhao, X., Fine, M.J. et al. J GEN INTERN MED (2019).

## N Engl J Med 2017; 376:663-673 DOI: 10.1056/NEJMsa1610524


Another co-op acting oddly?

Charles Gaba is doing his regular job of collecting all initial rate filing requests. He looks at New Mexico and something is odd in the filing. The largest individual market insurer (Molina) and two smaller carriers all initially filed effectively flat rates. To me, this implies that the 2019 premiums were slightly overpriced and we should expect to see Medical Loss Ratio (MLR) rebates of decent to significant size in September 2020. However, the second biggest carrier by enrollment in the ACA individual market, New Mexico Health Connections (NMHC), a co-op, filed for a 30% rate increase. accessed 6/12/19

I wonder if NMHC will be a going concern in 2021.

My question comes from a pricing perspective. Molina has been an aggressive spread strategy player in New Mexico. They have a low cost, Medicaid-esque network that allows them to significantly underprice insurers that pay their provider networks something closer to standard commercial rates instead of significantly under those rates. Molina’s pricing advantage is especially notable for their Bronze plan in 2019. Their 2019 Silver plan is only $8 below benchmark for a single 40 year old non-smoker so it is price superior but not price dominant for the 138-200% FPL group that makes up a huge chunk of the individual market. NMHC has a slight pricing advantage on Gold compared to Molina. Blue Cross and Blue Shield priced 2019 a bit above NMHC.

That all changes in 2020

I am making some very quick estimates of the premium spread changes in Bernalillo County, New Mexico, the largest county in the state. I am assuming no change in plan offerings by any current 2019 carrier and flat rates for all non-NMHC carriers.

Assuming no other changes, Molina has a dominant pricing position. They would have the cheapest Bronze, Silver and Gold plan by a significant margin. A significant portion of the subsidized population would see a Molina plan that has a zero dollar premium. That is likely to be important for both attraction and retention as it is hard to terminate a policy for non-payment of premiums when there is no consumer facing premium. Gold plans from Molina will have a very significant pricing advantage over other insurers’ golds. NMHC will be priced over Blue Cross so if people are looking for a non-Molina product, NMHC is a second choice on price after an initial quality/attribute assessment.

I am assuming that the rate requests are approved as is. That is implausible but a realistic starting point. I am assuming that the other insurers don’t alter their product offerings. That, too, is a large assumption. If I was running Molina, I would introduce a 59% AV Bronze plan that I could price $30 or $40 below the current Molina Expanded Bronze offering. Doing that would pull in even more people to a zero premium plan. I am assuming that the NMHC currently covered population is no more intrinsically sticky than the rest of the ACA market. I think that is reasonable.

If these assumptions are true (enough), then NMHC has problems as they will be losing a good percentage of their current, low cost members while perhaps holding onto a morbid pool of high cost patients whose expenses may not be appropriately compensated by risk adjustment. At this time, I think that keeping an eye on one of the few operating co-ops is worth the time and attention. I don’t anticipate much news until the summer of 2020 but there is a possibility of failure if the current pricing is approved in magnitude if not in exact number.

PrEP approved as a no cost sharing service

Earlier this week, JAMA ** published the recommendation of the US Preventive Services Task Force that Preexposure Prophylaxis be a no cost sharing preventive service for some high risk populations.

Findings The USPSTF found convincing evidence that PrEP is of substantial benefit in decreasing the risk of HIV infection in persons at high risk of HIV acquisition. The USPSTF also found convincing evidence that adherence to PrEP is highly associated with its efficacy in preventing the acquisition of HIV infection; thus, adherence to PrEP is central to realizing its benefit. The USPSTF found adequate evidence that PrEP is associated with small harms, including kidney and gastrointestinal adverse effects. The USPSTF concludes with high certainty that the magnitude of benefit of PrEP with oral tenofovir disoproxil fumarate–based therapy to reduce the risk of acquisition of HIV infection in persons at high risk is substantial.

Conclusions and Recommendation The USPSTF recommends offering PrEP with effective antiretroviral therapy to persons at high risk of HIV acquisition. (A recommendation)

This is good news for individuals who will benefit from PrEP. It is good news for future individuals who will never be exposed to HIV because their counterfactual exposing partner never contracts HIV. It is good news.

One of the challenges of the recommendation is financial. PrEP, in the United States, is expensive and will not be off patent for several (or more) years. Expensive treatments that can be applied to many people are a technological or policy shock for insurers. We saw this with Solvaldi and the other Hep-C anti-virals in 2013-2014 where many insurers and other payers such as state Medicaid agencies attempted to pay for as few courses of treatments at high price levels while they waited for near substitutes to enter the market and bring down the effective price level.

The big problem for ACA regulated plans is that PrEP is not risk adjusted. The economic cost of PrEP is supposed to be baked into the underlying premiums with no transfers from insurers that cover proportionally few people with PrEP to insurers. But the problem is that the potential PrEP using population is not randomly distributed nor completely unpredictable. This leads to potentially bad incentives:

Right now, someone under the age of 40 who is likely to use zero cost-sharing PrEP is a guaranteed money loser as their premiums won’t cover their PrEP costs net of no risk adjustment. Insurers don’t like to cover guaranteed money losers. They will cover folks with high cost conditions enthusiastically if the premiums are high enough or the combination of reinsurance and risk adjustment makes the proposition of covering a high cost individual not a money loser.

There are several ways insurers can be in a situation where they have to pay for PrEP but don’t want to pay for any more PrEP. The easiest one is to selectively pull out of some counties that are likely to have large potential PrEP populations. This makes the costs and increased premiums some one else’s problem. Another obvious way is to minimize PrEP prescriptions is to minimize the network of prescribers while also throwing up a paperwork minefield of pre-authorization, re-authorization and and random verification. There are other ways insurers in competitive markets will try to avoid getting hit with the new price/policy shock that feeds straight into premiums but these are the obvious methods.

PrEP is a good public health measure but it presents technocratic challenges on getting the risk adjustment right.

** JAMA. 2019;321(22):2203-2213. doi:10.1001/jama.2019.6390

Everything is awesome until there are trade-offs

Early this year, Kaiser Family Foundation released a poll showing Medicare Advantage was popular:

(56 percent) favor a national Medicare-for-all plan

There is a fairly typical partisan split on all of these proposals. Buy-ins are more popular than replace the entire system plans. Incidentally that is why I think a Medicare/Medicaid buy-in approach is what will be on a Democrat’s agenda if there is a blue trifecta.

However, Kaiser then did something interesting. They offered trade-offs and support for Medicare for all sank:

This is the political problem for Medicare for All proponents. It is a massive change to the system which will produce winners and losers. It upsets the status quo bias that worked against the ACA 2009-2016 and it upsets the status quo bias that worked for the ACA in 2017 to present.

Managing the transition so that there is no fear that anyone could, in anyway, be worse off or perceive themselves to be worse off is impossible. The challenge is minimizing both the number and the power of people who are worse off or perceive themselves to be worse off. And that is a tough challenge if one assumes that healthcare is, to some degree, a positional good.

Trade-offs are tough. Passing out free ice cream is easy.

Expanding access and subsidies in California

Last night, the California budget deal was announced. It contains several major health policy provisions. The biggest ones in terms of coverage include expanding Medicaid with state funds for young adults whose immigration status would not allow them access to Medicaid. Other elements of Medicaid were also beefened up. The big policy changes of interest to me is on the individual market.

Covered California already has one of the more unique markets in the country. It is a very active purchaser model which severely restricts the space insurers have to play games with premium spreads. Now it will have significantly more power to offer low and no premium plans.

Conference Compromise
1) Approve Governor’s proposed subsidy levels.
2) Additional $450 million General Fund over three
years, as follows: $133.4 million in 2019-20, $149.4
million in 2020-21, $167.3 million in 2021-22.
3) This funding will be used as follows:
(a) Approximately $10 million per year to provide
state subsidies to individuals below 138 percent of the
federal poverty level (FPL). This augmentation will
fully cover the cost of the standard premium for this
group of individuals.
(b) The remaining funding will be used to provide
additional subsidies to individuals between 400 and
600 percent of the FPL.
4) Adopt placeholder trailer bill language with the
following changes to the Governor’s language:
(a) Sunset exemption from the Administrative
Procedures Act after three years
(b) Require the Franchise Tax Board to report
statistics on the mandate penalty
(c) Clarify exemptions from the mandate penalty

So the short version of this is that California will adopt a state wide individual mandate. Those funds will support enhanced premium subsidies for three distinct groups. The first group will be for the small percentage of Californians earning between 100% to 138% Federal Poverty Level (FPL) who are on the Exchange for a Cost Sharing Reduction (CSR) Silver plan and not on Medicaid. Their premiums will be zero dollars after the new subsidies. This is a fairly low cost proposal as the benchmark premium for a single individual earning 138% FPL is $22 per month. The next group of subsidies are small ($10-$20 per member per month) subsidies to people earning between 200% to 400% FPL. These are “top-off” subsidies to people already getting federal premium support. The final group is the 400% to 600% FPL group. Right now, this cohort pays full premium. These subsidies could be worth several hundred dollars per member per month.

The goal of these actions are to bring in healthier risk and make coverage more affordable to more people. It applies both a carrot (enhanced subsidies) and a stick (a modest mandate) to change the cost calculation for coverage.

These are the steps of a state that is actively trying to get their market to work.