An Open Letter To Kevin Counihan (aka, "Make my job obsolete, please!")

The big news ACA yesterday was that Kevin Counihan, the guy who ran AccessHealthCT (one of the best-run state exchanges) has left that post to become the new "CEO" of Healthcare.Gov, just in time to prepare things for the 2nd open enrollment period.

By all accounts, this is excellent news all around. Counihan has been widely praised for his work in Connecticut--as well as in Massachusetts prior to that (he was one of the guys in charge of the original "RomneyCare" initiative).

So. We now have a new HHS Secretary (Sylvia Burwell, who's been in charge since early summer), a new Healthcare.Gov CEO (the first, actually), and a new Communications Director for CMS (Centers for Medicare & Medicaid Services), Lori Lodes.

Given that we're at a bit of a lull in actual ACA enrollment news, and given that this probably won't last long with the November election fast approaching, here's an open letter to Mr. Counihan primarily, but also to Sec. Burwell & Ms. Lodes as appropriate...honestly not sure who'll be in charge of what:


Dear Mr. Counihan:

First, Mazel Tov & Congratulations on your new position at the HHS Dept! I've heard nothing but the best about your work at AccessHealthCT, and am sure you'll do a great job of keeping Healthcare.Gov a smoothly-operating machine this autumn and throughout the 2nd Open Enrollment period.

I'm writing this on the assumption that you're at least vaguely aware of who I am and what my connection to the Affordable Care Act has been. If not, ask around; I'm sure someone there has at least heard my name tossed around on occasion. With that in mind, I'd like to humbly (ok, not so humbly) make a few recommendations about how to handle publicizing ACA enrollment data going forward.

As you know, during the first Open Enrollment period, from October of last year until April of this year, the HHS Dept. would issue monthly enrollment reports around 40 pages in length, generally coming out around 10-12 days into the following month. These reports broke out the data for the prior month in all sorts of ways, but were also frustratingly stingy with other data. Most significantly, of course, they didn't give any hard numbers about first-month premium payments, since only 2 states (MA & WA) were set up to only report enrollees after they had made their first payment, and the HC.gov "back end" payment system still wasn't set up. Then in May, it was announced that the HHS Dept. was deliberately stopping any further monthly reports during the "off season".

Please note that the following suggestions are not meant to malign those who put them together. Regardless of my criticisms, the actual reports were pretty useful as far as they went, and no doubt took a tremendous amount of work to compile.

1. Start issuing the enrollment reports again!

I don't know if this decision was made by Sec. Burwell upon her arrival or Sec. Sebelius upon her departure, but either way it was a very misguided policy change in my opinion. As I've vented about repeatedly, there are no logical reasons I can think of not to continue issuing the enrollment reports and every reason in the world to bring them back. The only "good" reason I can think of would be if the off-season enrollment numbers were poor, except that a) they aren't (in fact they're pretty good, around 9,000/day nationally by my estimates); b) there never were any widespread "expectations" about off-season enrollments in the media anyway (everyone obsessed about the magic "7 million" number for the on-season, but hardly anyone in the media seems to even realize that millions of people qualify to enroll even now); and c) even if the off-season numbers were poor...well, so be it; sunshine is still always the best disinfectant and all that.

Meanwhile, by going "radio silent" all HHS has done is give opponents a wide opening to further criticize them for a lack of transparency, leading to stories like this one at CNBC.

So...why has HHS/CMS gone radio silent? Not only have you stopped issuing the reports, so far I've been unable to get anyone to even tell me why you've stopped issuing the reports. My advice: Bring 'em back.

2. Make 'em weekly, while you're at it!

No, I'm not expecting a 40-page tome to be fully researched, compiled & edited every week. That would be an awful lot to ask. However, a simple press release with the major numbers (broken out by state, anyway) would do the trick. You don't need to go into Metal Levels, Age/Ethnicity Demographics or any of that for the weekly reports, but at least provide 2 hard numbers for each state (+DC): Total QHP enrollments & Total Medicaid/CHIP enrollments through the exchanges.

I'm not saying they have to be daily, but if MNsure can do so in Minnesota, I don't see why you couldn't do this on a weekly basis.

3. How to handle attrition:

Frankly, the only even remotely "legitimate" reason I can think of as to why HHS stopped the monthly reports is that they were concerned that the number of people being removed from the roles (either due to not paying the first month's premium or dropping their coverage in the 2nd, 3rd, etc. month) would overshadow the number being added. Well, this may be a real concern...but guess what? The media is doing that anyway. I've done my best to point out the flaws in poorly-researched, poorly-reported stories like that one, but it'd be a hell of a lot easier if you provided the real data to work with.

So, how do you handle the "attrition" factor? Simple: Include three more columns to the report: QHPs 1st Month Paid, QHP Enrollees Dropped & Net QHPs. I'm aware that these numbers will be unknown for most states; so be it. Include them for the states that you do have the numbers for.

Now, I'm guessing that in a given week, this will look something like the following:

State QHP Enrollees
This Week
QHP Enrollees
Cumul
QHPs
1st Mo Paid
QHPs
Dropped
Net Paid
QHPs
Medicaid/CHIP
This Week
Medicaid/CHIP
Cumul
XX 500 45,000 40,243 642 39,601 4,000 98,000
YY 300 32,000 ?? ?? ?? 2,300 28,000

See? No biggie.

4. Include how many for each START DATE.

One of the things which drove me nuts about the "How many have PAID???" debacle is that ACA opponents were screaming about this as early as January when millions of QHP enrollee policies didn't even go into effect until February or March. This was even more absurd when it turned out that 2.9 million of the "original" 8.02 million enrollee policies didn't even kick in until May 1st...or over 36% of the total. Most people don't pay their cable or electric bill until it's due; why would anyone expect them to treat their insurance premiums any differently?

The solution to this is to break out (in the monthly reports, not weekly) the enrollments by what their starting policy date is. Vermont already does this (though not on a regular basis); I don't see why this shouldn't be feasible on a monthly basis; just break out everyone who enrolls between the 16th of one month and the 15th of the next.

5. Abandon the "automatic re-enrollment" policy ASAP.

In late June it was announced that the HHS Dept's official policy would be to allow current QHP enrollees to automatically renew their enrollment with certain exceptions. While I understand the desire for convenience, at the time, I questioned the wisdom of this policy, noting that the "certain exceptions" confusion would likely more than cancel out the convenience factor.

This concern was solidified over the past few weeks as it was revealed that the unusual formula used to calculate tax subsidies means that millions of people are likely to see their subsidies either increase or decrease substantially...even if nothing has changed at their end income/dependent/policy-wise, and even if their "full rate" premium doesn't change a dime.

In addition, there's hundreds of thousands of people on some of the state exchanges who will have to re-enroll "fresh" anyway, due to their states overhauling/abandoning the original software platform. This includes all 75,000 or so in Maryland, 38,000 in Nevada, 80,000+ in Oregon and a whopping 400,000 in Massachusetts. That's 600,000 at a minimum. Seriously, I'd strongly recommend just having a blanket policy of manual re-enrollment across the board. If people want to keep their current policy, set up the website so they can just log in and answer a couple of simple questions:

  • Have any of the following changed since the last time you logged in? (Provide a list of life changes as you already do now)
  • If they answer "No" to all of the above, let them see what their current premium/subsidy is next to their 2015 premium/subsidy for the same plan
  • Then ask if they still want to keep the existing plan or if they want to shop around to see the new policies/rates

In most cases they'll still likely stick with whatever they already have...but make them at least log in and affirmitively state as such. That way they won't be hit with a rude awakening at tax time next spring.

Of course, this means that you'll have to...

6. Separate out "Renewed from 2014" enrollments from "Newly Enrolled for 2015"

That way we can keep a running tally of both "YTD" enrollees as well as "Cumulative" ones.

And finally...

7. Don't fear poor numbers.

Remember when everyone was spazzing out about there being no way in hell that the 1st year QHP total would even come close to the "magic 7 million" number? Even I thought that reaching 7 million by March 31st was a "pipe dream" as late as March 23rd (although I was very comfortable with reaching 6.5 million by that point).

The numbers will be decent. Yes, I know the "low-hanging fruit" has been picked already, but there were still long lines at the deadline last spring, so much so that you had to bump out the enrollment period by another couple of weeks. Here's an excellent overview of why the numbers should be pretty good anyway.

And even if the numbers do prove to be poor...well, that's a relative term. There's still a good 10-13 million people who have already gained insurance to date, plus another 10-18 million who have replaced their old insurance with new, ACA-compliant policies. Even if the Year 2 gains are incremental, they'll still be worth talking about.

If you take all 7 of the above suggestions, there will be little remaining need for this website to continue operation. I'd probably keep it chugging along for some light analysis and such, but about 80% of the point of this project would become obsolete...and I could live with that.

Thanks for reading.

--Charles Gaba