David Anderson

Yesterday, I took my daughter in for her annual well child check-up.

The nurse practitioner checked her heart and lungs. She talked about puberty and recommended a couple of good books about that before asking about the regular public health checks (tobacco, guns, seatbelts). Finally she did a quick physical exam. My daughter was given a clean bill of health and development. She was also happy; she had no vaccinations due, so we were within thirty seconds of getting out of the appointment without her getting poked by a needle. However the NP said that there was an optional lipid screen that was available if we wanted it. I asked if there were any results that would change her clinical decision making. She said that given all the other measurements she had on my daughter plus her age, there would be absolutely no actionable information from the screening. We declined. My daughter does not like needles, and for once, my dad and health policy geek hats both fit comfortably on my head.

The lipid panel for my daughter would have been no value care. It would have changed nothing in the clinician’s medical judgement. If we had elected to have the blood test, my daughter would have had the cost of dealing with needles and blood, I would have been paying for it, and the clinician would have one more task to acknowledge and manage on a brain that was already hitting cognitive overload.

The claim payment part of low or no value care is easy to measure when there is a moment of low or no value care. Clinician cognitive management costs are harder to measure but they are measurable or at least reasonably estimated. Patient-centric costs of receiving no value care are harder to measure. Speaking as a dad, my daughter’s morning was much better in a no blood draw universe than a universe that had a single, no value blood draw. She would have been stressed out and slightly anxious for at least the entire drive from the doctor’s office back to school.

I wonder if there is an analysis of the non-monetary costs of receiving low or no value care?

The value of avoiding no value carePost + Comments

One of the challenges of choosing health insurance in an open enrollment period when the menu of choices is somewhat lengthy is avoiding dominated plans. A dominated plan is one where on all relevant to the decision maker characteristics it at best ties and typically loses to another option. Choosing a dominated plan means that you are choosing a plan that is objectively inferior to at least one other option. An inferior plan means the insurer will make more money from you compared to an optimal plan choice.

I will show an example below of Plan A being dominated by Plan B on a few characteristics:

Plan B beats Plan A on premium, deductible, coinsurance rate, maximum out of pocket and network. Plan A ties Plan B on PCP co-pay. A full domination plan analysis will probably look at more characteristics than those, but this is the short toy example of what to look for when thinking about a plan being dominated.

In the ACA, the creation of well defined, and mutually (or mostly mutually) exclusive metal bands would make unsubsidized dominated plans a rare event. From 2014-2017, subsidized dominated plans are easier to see as one could expect a company offer two Bronze plans priced significantly below the benchmark Silver plan with significantly different deductibles but otherwise identical cost sharing structures both be priced at zero for a particular subsidized buyer.

From 2018 to present, due to silver loading for CSR funding termination, dominated plan offerings are likely to be more common within a particular company in a particular region as gold plans are frequently priced underneath silver plans. In some regions, gold plans can have zero premiums after subsidies for large populations. As long as the benefit design is not too funky relative to silver and bronze plans, gold plans can readily dominate bronze and silver plans for subsidized buyers in certain regions.

So when you are looking for plans in either Medicare Advantage, Medicare Part D drug coverage or the ACA exchanges, look hard at dominated plans. And if you find one, avoid it.

Beware of dominated plansPost + Comments

Georgia’s governor released a plan for a 1332 waiver.  This waiver has two severable steps. The first is a reinsurance step.  ACA reinsurance has a source of funds that are not derived from premiums cover some portion of claims. The external infusion of money means premiums don’t have to be high enough to cover all claims and all administrative costs. Instead, they just have to be high enough to cover some portion of claims and all of the administrative costs.

Georgia is proposing a reinsurance scheme that varies by geography. The health economists are going to love this, especially in rating area 13 as it provides a great source of variation and clean discontinuities. The policy objective is to bring down gross premiums in rural, expensive regions more than in urban and suburban regions. This is a valid policy objective. Colorado is running a reinsurance scheme similar to the proposed Georgia scheme for the same exact objective.

The other notable aspect of this proposal is the amount of claims that the reinsurance scheme is proposing to eat. The attachment point which is the claim value where reinsurance starts is only $20,000. That is a knee replacement or a somewhat complicated C-section. Many other states won’t start their reinsurance until the claims run to $40,000 or $50,000. And then the reinsurance program runs to $500,000. This is much higher than usual. None of this is bad. It is a lot of reinsurance.

Finally, the Georgia market has some significant premium spreads between the cheapest plan in a metal level relative to the benchmark premium. Big spreads are great for healthy, subsidized folks as they will see cheap net of subsidy plans. Small spreads push healthy subsidized folks out of the market as they are priced out. Regions with large, current spreads and high (80% coinsurance) will see much smaller spreads in 2021 than they do now. This will bring some uninsured and unsubsidized folks into the marketplace while pushing some currently insured and subsidized out. I argued in August that states have to make choices:

All states will face an environment of guaranteed issued, community‐rated, subsidized private market insurance, available Medicaid expansion for working‐age adults at an enhanced federal cost‐sharing and wide waiver authority that is constrained by law and clear administrative procedures. Within this framework, states will need to find ways to smooth the edges of the law and address issues and populations that are not well served by the law. They have the tools to do so, but the decisions to trade‐off the well‐being of some groups for the gains of others will remain a potent political and policy problem.

Georgia is making a choice.

Georgia’s 1332 waiver reinsurance thoughtsPost + Comments

Open Enrollment on Healthcare.gov started this morning.

With the somewhat small exception of the Texas v Azar case that is currently hanging out at the 5th Circuit which, if affirmed, could throw out the entire law, I expect the open enrollment period to be fairly straightforward.

I expect total on-exchange enrollment to decrease by a few percentage points for a couple of reasons.

• Slightly worse pricing net of subsidy as spreads between Benchmark and cheapest plans compressed
• Reinsurance is compressing on-exchange premium spreads
• More Medicaid Expansion moves people who were buying plans and earning 100-138% FPL to Medicaid
  • Maine and Virginia have hit their strides
  • Mountain West expansions states are slowly starting up
• More off-ramps this year with wide spread marketing of short term limited duration plans, cost sharing ministries and association health plans

The biggest thing that could boost enrollment is if the Health Reimbursement Account scheme takes off quickly.  I think that is more of a 2020 thing for plumbing reasons but I could be wrong.

I don’t expect to see much difference in enrollment due to outreach and advertising.  We have baked into the baseline a very low level of advertising and assistance already, this is not new.

There is little to no policy uncertainty during the pricing cycle.  There are no policy shocks.  There is no free, earned media like in the Fall 2017.  This is a business as new usual year.

Now let’s just go back to an earlier post on choosing insurance as that applies here as well:

 

 

• Insurance is complicated

 

… I know my limitations. I am very confident that I won’t make a horrendous choice. I am also very confident that I won’t make the optimal choice. And I’m okay with that. When I have to choose an insurance product, I engage in a strategy that minimizes my regret. I’m okay with sacrificing some perfection as long as I know that my eventual choice is at least good enough. The extra two or three or thirty hours of gaming out every potential scenario is not worth the time nor the stress.

 

• Identify what you value

 

How important is access to a particular set of doctors or hospitals? This will be a mixture of your health status and the learning/hassle costs of switching docs….

How much can you afford in a worse case scenario versus how much can you pay out each month? This is a tough trade-off. All else being equal, a higher monthly premium means more protection in a worse case scenario (that is less likely in the ACA as there are an increasing number of gold and silver plans with the same out of pocket maximum as bronze plans but merely take a lot more time/claims to get there). For a given monthly premium, what are you willing to concede to get a targeted level of protection? How important is a cap on exposure?

 

• Know where your help is

 

Insurance can be complex.

Most people don’t spend several thousand hours a year thinking about insurance markets. Some people do. Use those people.

Good luck, and ask for help in comments as needed.

ACA Open Enrollment has begunPost + Comments

Most projects for the 2021 Senate composition will have both parties at 50 +/-2.  Democrats are likely to lose one seat (Alabama) and Republicans are playing defense in several seats where early fundraising and favorability polling supports plausible Democratic pick-up in Arizona, Colorado and Maine.  There are several other seats up for grabs right now.

There is a thin chance of a 59 seat Republican Senate, and a chance of a 47 seat Republican Senate.  There is no chance of a 60 seat Democratic Senate in January 2021.  Plausible policy pathways for Democratic preferences are constrained by whatever can go through a 51 vote procedure.  This means either getting rid of the filibuster for legislation and equalizing legislation with confirmation standards, or reconciliation.

Sinema is a NO on getting rid of the filibuster

“They will not get my vote … whether I’m in the majority or the minority I would always vote to reinstate the protections for the minority” https://t.co/b2ekoUz66i

— Bill Scher (@billscher) October 29, 2019

Health policy that can go through reconciliation with fifty votes and a friendly Vice President means that we would be living in a world controlled by Senator Murkowski (R-Alaska) or Senator Manchin (D-West Virginia). Those are the senators who are likely to defect from their majority caucus position. This realization should strongly inform what is mechanically plausible after a Byrd Bath.  It should also inform to a somewhat lesser degree what is politically possible.

2021 will be Manchin’s or Murkowski’s worldPost + Comments

ACA open enrollment is starting at the end of the week. The 2020 Open Enrollment Period will be one with more choices than the past couple of years with significant premuium swings. There are no major changes in outreach activities budgeted for by Healthcare.gov. The advertising and navigation budget is very low but not too different from their 2019 Open Enrollment Period budget. Beyond insurer entry and exit and plans being terminated and created, the big source of variation will be in the post-subsidy premiums. Subsidized buyers, and more importantly, the marginal subsidized buyer who is flipping a coin as to whether or not they will be covered next year, are primarily concerned about the premium that they pay every month and little else. This means that they care about the spread between the cheapest plan available to them and the benchmark plan. This is the “Silver Gap” game I’ve been talking about for years now.

There is significant variation. The maps below show the difference in premium spreads between the cheapest Bronze and Silver plans to the county benchmark for 2019 and 2020. Green shows that a county has a cheaper net of subsidy plan now available in 2020. Red shows that the premium spread has shrunk and therefore the cheapest subsidized plan in that metal level for that county is now more expensive net of subsidy. Bronze plans will be the cheapest plans, while silver plans will have high Cost Sharing Reduction (CSR) enhancements added which makes these the overwhelmingly dominant choice for people earning between 100-200 percent of the Federal Poverty Level (FPL).

All else being equal, we should expect to see enrollment overperformance in northern Iowa, Oregon, the Upper Pennisula of Michigan.  We should expect to see enrollment underperformance in parts of Oklahoma, most of Utah, Appalachian Kentucky, Western Tennessee, Southern Ohio, the coast of Wisconsin among other areas.  These are rough directional estimates. There are twelve counties in Oklahoma that have seen their bronze to benchmark gap be reduced by over $150 for a single forty year old.  That would, naively, lead to expectations that enrollment for healthy individuals earning over 200% FPL would crash. However the 2020 bronze to benchmark spread is still large enough to make a zero premium plan available.

However, as a first rule of thumb, big changes in premium spreads between the cheapest plans and the cheapest silver plans relative to benchmark will give us a good intuition as to which counties will see significant changes in enrollment for 2020.

Where to expect enrollment swings on Healthcare.govPost + Comments

Earlier this week, the Food and Drug Administration (FDA) approved a new cystic fibrosis (CF) treatment that can be used against the mutations that 90% of the people who have CF and are over the age of 12 have. This is a major risk adjustment problem.

• Perfect risk adjustment should make insurers indifferent to population characteristics
• Risk adjustment is never perfect
• Most risk adjustment systems don’t deal well with shocks
• Selection and screening are to be expected

Axios reports:

The Food and Drug Administration has approved a triple combination therapy manufactured by Vertex that would treat 90% of patients with cystic fibrosis.

The big picture: The approval of Trikafta came in record time at five months and will be priced at $311,503 annually, or $23,896 per 28-day pack

That is a lot of money. Insurers don’t collect enough in premiums from someone with CF to cover that expense. Instead, they must rely on reinsurance and risk adjustment to help cover the expense.
click below to see the entire post

Technological shocks and risk adjustmentPost + Comments

Yesterday, CMS released the 2020 ACA public use files. I’ve done my regular analysis of the “Silver Gap” between the benchmark Silver plan and the least expensive plan in each metal band for a single, subsidized forty year old non-smoker

• More competition has decreased the number of counties where a Gold plan is less expensive than the benchmark
  • 2017: 5 counties on Healthcare.gov
  • 2018: 594 counties on Healthcare.gov
  • 2019: 1136 counties on Healthcare.gov
  • 2020: 937 counties on Healthcare.gov
• Low to no cost Bronze plans are quite common
• Iowa got smart
• 12 counties in Oklahoma still have zero premium Gold plans

This analysis is only relevant for individuals who make between 100% to 400% Federal Poverty Level (FPL) as this is the group that receives Advanced Premium Tax Credits (APTC). Everyone else pays full price. Subsidized individuals are sensitive to premium spreads while non-subsidized individuals are sensitive to premium levels.
click below to see the entire post

Premium spreads for the 2020 subsidized individual marketPost + Comments

The 2020 ACA premiums were released this morning. The press release headline is as follows:

Today, the Centers for Medicare & Medicaid Services (CMS) announced that the average premium for the second lowest cost silver plan on HealthCare.gov for a 27 year-old will drop by 4 percent for the 2020 coverage year. Additionally, 20 more issuers will participate in states that use the Federal Health Insurance Exchange platform in 2020 bringing the total to 175 issuers compared to 132 in 2018, delivering more choice and competition for consumers. As a result of the Trump Administration’s actions to stabilize the market, Americans will experience lower premiums along with greater choice for the second consecutive year.

The cynical take on this paragraph is that 2018 was a local minima due to massive policy, political and legal uncertainty. 2019 and 2020 are reactions to the re-stabilization of expectations and rule sets which were not present in 2018.

We know that 2018 was massively overpriced. We know that as insurers are paying out very large Medical Loss Ratio (MLR) rebates for the 2016-2018 plan years and are expected to pay out even larger rebates for the 2017-2019 period. If we assume that the final 2020 rates are reasonably rates that “should” be close to actuarially fair plus “normal” profits and admin costs, there are two pathways that we could have arrived at this price level with a starting point of 2017 being priced at a reasonable level high enough to cover claims and normal profits. The first is the path we took; massive rate hike in 2018 and then slight declines in 2019 and now 2020. The other is a slow and steady route.

Both paths get to the same exact endpoint. One path produces the headline of multiple years of premium decreases. The other one produces a headline of steady rate increases.

Slow and Steady or Herky-JerkyPost + Comments