(Updated) Medicaid: Should "Woodworkers" count as "Due to the ACA" or not?

"Woodworkers", in ACA Medicaid parlance, refers to people who were already eligible for Medicaid coverage prior to the ACA expansion (in states which have done so),  but who didn't actually enroll in Medicaid until after October 1st for a variety of reasons, ranging from not being aware that they qualified, to finding the enrollment process too difficult, to not even knowing what Medicaid was before the ACA exchanges launched.

In other words, these are people who already qualified, but were "brought out of the woodwork" to sign up since the exchanges launched in October.

I've been struggling with the question of whether to include "woodworkers" in the Medicaid/CHIP tally or not since I first learned of the distinction between them and the far stricter definition of who should "count" (ie, limiting it to "people who only legally qualify for Medicaid/CHIP due to expansion provisions within the ACA".

There's really two questions here--a philosophical one (should they be included) and a data-driven one (if you do include them, how many people fall into this category?).

The first question depends on your perspective. The launch of Medicaid expansion (in the 24 states, plus DC, which have done so...I'm not including Michigan since our expansion doesn't kick in until April 1st anyway), along with the launch of the exchanges, has included a massive public education/outreach/PR campaign alongside it, to explain to people what Medicaid is, how they can apply for it and how to tell whether they qualify or not. In addition, the law has also massively streamlined the enrollment process in many states, either via web-based enrollments (for the ones which aren't still having technical issues, anyway) or by giving resources to states like Oregon and West Virginia to auto-enroll massive numbers of people by using existing food stamp or welfare data to identify them.

Consider the following: If Pepsi launches a new ad campaign for 6 months, and their sales go up 50% during that time as compared with the same period a year earlier, is it accurate to say that the 50% sales increase is "due to" the new ad campaign? Yes...assuming that there weren't any other significant events that also transpired during that time which would impact sales. If Coca-Cola had some sort of PR disaster or embarrassing product recall during the same period, that could be part of Pepsi's good fortune. If Pepsi also happened to sign some unrelated exclusivity deal (separate from the ad campaign) with a large restaurant chain at the same time, that could be given credit as well.

In the case of Medicaid enrollments, there's ample evidence that at least some of the additional enrollments are specifically due to the "woodworker" effect; South Carolina--one of the states which is most definitely not expanding Medicaid--expects their enrollments to increase 16% this year, primarily due to the ACA's awareness campaign, as well as the requirement that everyone get covered included in the law.

So, whether this should count is an open question.

The other tough question is how many should be counted. Unfortunately, so far, only one state, Washington, has broken out their Medicaid numbers enough to get a sense of the "woodworker" effect. As of their most-recent press release, it stands at:

  • 184,783 Expansion ONLY
  • 91,480 Woodworkers*
  • 280,766 Redeterminations (renewals...not counted by me at all)

So, for this one state, about 1/3 of the total new enrollments are "woodworkers". If this applied across all 50 states (+DC), it would make things easy to calculate.

Unfortunately, it's not that easy. Washington may or may not be representative; it could easily be an outlier. In addition, the impact of "woodworkers" is going to be very different in states which have expanded Medicaid vs. those which haven't.

If I just went with South Carolina as a base for woodworkers in non-expansion states (16% of total new enrollments) and Washington as a base for woodworkers in expansion states (33%), the numbers would be fairly easy to calculate:

  • NON-expansion states: Total new enrollments = about 2.31 million; 16% of those = 370,000 "woodworkers"
  • EXPANSION states: Total new enrollments = about 4.81 million; 33% of those = about 1.59 million "woodworkers"

Add these up and you get about 1.96 million "woodworkers" nationwide.

*However, there's another problem: "Baseline Churn". Note that in the non-expansion states, when you remove the 16% woodworkers, you still have abou 1.94 million new Medicaid enrollees, even though they haven't expanded Medicaid. Those are the people who are always moving onto and off of Medicaid from month to month as their circumstances change--they lose a job, get a job, move from one state to another, have a child, get married or divorced and so on. The actual number of these is constantly changing, and is therefore difficult to quantify.

This is why I had to switch to a "range" style listing. Based on Avalere's estimates from last month, along with additional post-12/31 data which wasn't included in their report, I've low-balled the strict "Expansion ONLY" number to be around 2.6 million, with the "expansion ONLY + woodworkers" number coming in at around 4.8 million as of today.

Hopefully I'll be able to revise/update this when the next CMS monthly report comes out, which should be sometime this week...but that still doesn't answer the original question: SHOULD woodworkers be included at all? That's up to you.

UPDATE: Several people have called my attention to a story out of Maryland yesterday in which it's revealed that the "official" enrollment goal for Maryland (150,000 private QHPs, 110,000 Medicaid/CHIP) was actually way off-base due to an error on the part of the firm hired to come up with the projection. Apparently they accidentally counted two years' worth of data instead of one, and the correct projection should have been 70,000 private QHPs and 90,000 Medicaid/CHIP enrollments (160K total).

While this is a bit embarrassing for Maryland, there's actually one tidbit in the article which stands out for me: 

...during the first six months of Medicaid enrollment, 90,639 would automatically be enrolled and 11,046 would come out of the woodwork, not knowing they were Medicaid-eligible before contacting the exchange.

In other words, in Maryland, at least, the number-crunchers predicted that roughly 11% of the newly-added Medicaid enrollments would be "woodworkers". While this doesn't answer the question of "how many woodworkers are there", it at least gives us something to chew on...and also proves that "woodworkers" were indeed taken into account for at least some of the early projection numbers...due to the ACA.