Initial reaction to CMS 9929P

The Center for Medicare and Medicaid Services (CMS) released a proposed rule for the PPACA Exchanges this morning. These are my initial thoughts that I will most likely change as more information comes in and points are clarified. This rule is fundamentally a technical corrections rule that seeks to make the Exchanges work while favoring insurer interests. It is not a rule that will blow up the Exchanges.

The rule can be broken into two broad segments. The first segment is premium decreasing measures through plan design. This segment seeks to lower the cost of premiums (and incidentally, the cost of premium tax credits) by slightly dropping minimum requirements of coverage. This will benefit healthier individuals who do not receive subsidies and it will disadvantage sicker and more expensive individuals who are subsidized.

The second and far broader portion of the rule is risk pool management functions. The rule making assumes that there is significant intentional gaming of the various enrollment rules which have led to an adversely ill and expensive risk pool as healthy people are finding ways to avoid paying for coverage until

Larry Levitt at the Kaiser Family Foundation has a good general take on the rule making direction.

Now let’s dig into the details. Read more of this post

A timeline to run

The Republican Party does not have the luxury of time to get their act together on health policy. They have maybe fifty days to get a coherent plan with the possibility of passage before external actors move to foreclose on the 2018 policy option space:


Below is the Center for Medicare and Medicaid Services plan filing timeline.

Read more of this post

What Should Health Insurance Cover?

This is a basic health reform question, and one knock against the ACA has been that the benefits are too generous, which drives up the cost of premiums. This WSJ piece provides a good overview of the issue (and tradeoffs), and has a nice graph showing what proportion of individually sold insurance plans included benefits that are required by the ACA. Changes to the benefits required in the ACA could be made via rulemaking (the law describes categories of benefits that were detailed via rulemaking), Congress could reduce benefits in some “replace” or “tweak” of the ACA, or the issue could be devolved to the states, also via a replace or tweak bill. It is easy to criticize in health reform, but improvements are hard, because of the tradeoffs.




Subsidized market stability in 2018 amidst policy uncertainty

Last Thursday, CNBC highlighted a study by the Urban Institute and Robert J Woods Foundation about insurers worrying about political and policy uncertainty for their 2018 plan offerings for the Exchanges. As I see it there are three major sources of uncertainty. Will the Cost Sharing Reduction (CSR) subsidies be paid? If not, the insurers are gone in 2017. Will the individual mandate be enforced and if not, what will that do to risk pool size and composition? What other policies are coming down the pipe?

CSR is easy. If it is yanked, the insurers are gone at the end of the month it is yanked. That is not confusing me.

But the other two parts have me scratching my head a bit:

If Republicans repeal the Obamacare individual mandate without a concrete replacement plan, we could see another year of big price increases and insurer withdrawals for open enrollment in 2018, according to a new report. “The greater the uncertainty, the higher the rates,” said Sabrina Corlette, a Georgetown University research professor and one of the authors of the new report, which was published by the Urban Institute and the Robert Wood Johnson Foundation. “At a certain point you can’t even price high enough to account for the uncertainty and it’s at that point that the carriers say, ‘we have to get out,'” said Corlette.”

Here is where I am getting confused. I understand the model well enough. The combination of no pool participation mechanism through the granting of unlimited hardship waivers and policy uncertainty will drive many healthy individuals out of the market in the models that the actuaries run. The only people who will stay in the market for guarantee issued, community rated, no lifetime cap insurance are the people who really need the insurance because they are sick as hell. This is the classic death spiral. And insurers will pull out before they are on the tab for billions in losses as they set their premiums too low. This is the Washington State in the 90s story; this is the pre-ACA New York individual market story. I get the fear.

The big difference in the ACA story is the subsidies are attached that lower the consumer facing costs. I have been giving the exchange mechanics significant thought in better times and I want to revive that post It is very hard for a single carrier to lose money in this market structure when it offers the right configuration of products.

The basis of my response is that the federal subsidy structure makes it very hard for a carrier to continually lose money in a state if it is the only carrier. Let’s look at a few scenarios below the fold.
Read more of this post

Distributional impacts of Cassidy-Sessions 2015

A very smart analyst that I respect asked me to think a bit more on the Cassidy bill from 2015 that I was writing about last week. So I’m working on that. Now I want to look at the distributional impacts of the Cassidy-Sessions bill. I’ll be using a pair of 27 years olds as I’m working my way through a couple of different scenarios. The Cassidy scenario is the state run and regulated HSA option #3.

The first thing we need to review is the individual level funding mechanism in Cassidy as money illustrates values. Cassidy Option #3 gives a flat per capita deposit to HSA that is determined by the funds that the eligible but not insured population would have received on the PPACA as is. A 5% discount from the PPACA money pool would be applied. This applies to states that have expanded Medicaid and also those that did not expand Medicaid. Non-Expansion states would see 95% of the theoretical federal Medicaid Expansion funds also go into the pool for the HSA’s.

So let’s look at a simple chartof who benefits and we’ll go through some of the scenarios below the fold:

Read more of this post

Advance Care Planning in Medicare: Year One

On January 1, 2016, Medicare began reimbursing physicians for Advance Care Planning (ACP), using the following CPT Codes:

CPT Code 99497 – Advance care planning including the explanation and discussion of advance directives such as standard forms (with completion of such forms, when performed), by the physician or other qualified health care professional; first 30 minutes, face-to-face with the patient, family member(s), and/or surrogate


CPT Code 99498 – each additional 30 minutes (List separately in addition to code for primary



According to the AMA’s CPT Assistant, advance care planning “involves learning about and considering the types of decisions that will need to be made at the time of an eventual life-ending situation and what the patient’s preferences would be regarding those decisions.”

Paying for ACP incentivizes health providers to have these conversations with their patients and clearly demonstrates the value that Medicare places on this service. According to the Institute of Medicine’s pivotal report, Dying in America, patients often fail to obtain the care they need at end of life, and an uncoordinated and inefficient delivery system incur unsustainably high costs. Broader delivery of advance care planning could help address two key priorities from the report—ensuring that patients’ needs and goals of care are met, while reducing unnecessary costs resulting from uncoordinated care.

What’s remarkable about Medicare’s implementation of ACP reimbursement is how unrestrictive it is for billing providers. There are no limits set on appropriate use of the code related to place of service, diagnosis, frequency, or provider type (any primary care or specialist physician, nonphysician practitioner, or other staff under a treating physician can deliver the service, with no required training), and it can be billed independent of and on the same day as E/M services. The ACP conversation does not even need to be delivered directly to the patient, but can be had with a family member or surrogate instead.

As ACP claims are now being adjudicated, in practice, Medicare has not placed any additional limitations on the general CPT description. The rule authorizing payment of the new codes included no national coverage determination, which means that the regional Medicare Administrative Contractors (MACs) “are responsible for local coverage decisions in the absence of a national Medicare policy.” In early 2016 we contacted the MACs responsible for each of the 10 jurisdictions and learned that none had a specific local coverage determination in place, nor did any have plans to develop one.[1] Since then, CMS released an FAQ document in July explicitly clarifying that qualified providers may bill for ACP with no limitations related to place of service, diagnosis, or frequency, and that while providers should refer to their MAC for minimum documentation requirements, completion of an advance care directive is explicitly not required to bill for the service. Shortly after implementation, one of the MACs was denying ACP claims based on place of service codes, but quickly attributed the problem to system error rather than a deliberate coverage determination and reversed those claims.

For comparison’s sake, consider smoking cessation counseling (CPT codes 99406 and 99407), which CMS made a covered benefit in 2010. It’s similar to ACP since it’s a conversation delivered by a provider and aimed at improving patient health and reducing future costs of smoking-related illness. However, some  reasonable and intuitive limits are placed on billing of the code—Medicare requires that the patient be a smoker with corresponding diagnosis codes reported, cessation counseling attempts can only be billed twice per year, and although some state Medicaid programs permit billing for cessation counseling with the parents of patients with child eligibility coverage, CMS otherwise only pays for counseling directly to patients.

The lack of restrictions on ACP reimbursement is not a simple oversight by CMS, but seems to be a conscious decision to encourage widespread provision of the service. In the Federal Register for the 2016 Physician Fee Schedule that establishes ACP as a covered benefit, CMS acknowledges the comments it received with concerns about lack of a national coverage determination and potential for abuse of the code, but provided the following response:

We believe it may be advantageous to allow time for implementation and experience with ACP services, including identification of any variation in utilization, prior to considering a controlling national coverage policy through the National Coverage Determination process (see 78 FR 48164, August 7, 2013).


As the first full year of ACP billing in this laissez faire environment concludes, the claims record will start to reflect any increase and/or variation in utilization of the code. A national survey of both primary care and specialist physicians conducted at the start of the year found that 75% of respondents reported being either much more likely or somewhat more likely to talk with patients about advance care planning thanks to the new Medicare benefit.

If utilization of ACP does in fact increase, it may still be years before robust claims and medical record data indicate whether these conversations impacted end-of-life care decisions and reduced cost to the Medicare program. However, as a final point, it’s worth considering reimbursement in the context of the payment reform transformation currently underway as Medicare and other payers move from fee-for-service models to value-based arrangements.

In the prevailing fee-for-service environment, ACP has been a service with obvious potential benefits for patients and payers, but not necessarily for providers, who may have recognized the importance of ACP conversations but encountered barriers to delivery and have to balance meeting the many demands of their practice and patient panel. In a value-based payment future—where providers will be accountable for quality of care but also able to share in savings realized through better coordination of end-of-life care—patient, provider, and payer will all be organically aligned to place high value on delivery of ACP. Given the high costs and unique demands of end-of-life care, health system embrace of ACP could prove to be a key driver for achieving the triple aim of improved quality of care, patient satisfaction, and reduced costs. Medicare’s addition of the ACP benefit is a critical first step for helping orient providers towards this future.

[1] Thanks to stellar Research Assistant, Sara Constand, for gathering this information.

Welcome Dave Anderson

Dave Anderson has joined our team in the Duke-Margolis Health Policy Evidence Hub at Duke University, and he will also be blogging here at freeforall. His first post went up yesterday.

You may know Dave under his pen name of Richard Mayhew, a prolific blogger on health policy at Balloon Juice who has forgotten more about the intricacies of ACA insurance market reform and subsidy details than I ever knew. He will keep blogging there and a subset of his posts will appear here. The exact focus will be worked out over time.

At Duke, we are thrilled to have Dave’s health policy-infused data scientist/counterfactual thinker perspective applied to our ongoing work on the development of Palliative Care payment changes, and end of life policy generally. Further, we are hoping to build the go-to shop (policy analysis, some evaluation work in conjunction with partners) for health reform in States that haven’t yet expanded Medicaid in the belief that action in these States is likely to accelerate, and evidence based policy analysis is needed. Duke is in the South  you know (don’t tell anyone), and we want to help provide this across the nation, but particularly in the South.

There will be some other Duke colleague voices joining the blog as well.Look for more on that later this week.