You’ve probably read about the nearly 50,000 Arkansans who have either lost health insurance coverage or are set to lose coverage at this end of this month because they didn’t respond to an income-verification letter within 10 days. These are Medicaid beneficiaries (including beneficiaries of the state’s private option) on the chopping block as the state finally implements its troubled verification system.
The state is mandated by federal law to do annual income verifications and of course Medicaid benefits should only go to those who are actually eligible. But a key point that is sometimes missed in this kerfuffle (and that isn’t addressed at all by the governor’s moratorium on termination notices): many, perhaps most, of the people being kicked off of coverage are actually eligible – demonstrably so by the very data that the state's Department of Human Services is using to send out the letters to begin with. DHS is making no claim whatsoever that the folks being targeted by this letter have had their incomes rise above the eligibility line. Here’s some important context as this story unfolds.
The letters didn’t go out to people DHS believes are no longer eligible; they went out to any beneficiary whose income went up or down by 10 percent.
***UPDATE (8/7): The information below accurately reflects what DHS officials told the Times earlier this week. However, they have sense contacted us and provided different information about who was flagged: not all beneficiaries whose income changed by +/- 10 percent, but rather beneficiaries with a +/-10 percent change in income that moved them above/below an eligibility line for one of the Medicaid categories AND any beneficiary with an income of zero. See all of the details in this post. The DHS clarification does not impact the information on the 10-day window.
This has been reported elsewhere but it’s often buried late in stories and has tended to get lost in the shuffle. The state’s new automated eligibility verification system is based on wage data for the previous quarter via the state’s Department of Workforce Services. If a beneficiary’s income changed by 10 percent (again, even if the beneficiary’s income went down) versus the verified income at the time of the original application, the process begins that has led to termination of coverage for tens of thousands. The verification letter is automatically generated, demanding that the beneficiary respond within 10 days with pay stubs or a letter from an employer, etc.
Let’s think about this for a moment. If you’re a beneficiary who makes $10,000 a year and your income fluctuates—let’s say you’re now on pace to make $11,000, or $9,000—you got the letter. This despite the fact that the Department of Workforce Services data itself shows that you are eligible for Medicaid. If you don’t notice the confusing mailing, or you don’t realize that it demands immediate action, or you cannot make contact with DHS despite good-faith efforts, or you can't quickly come up with the necessary paperwork (such as multiple pay stubs), or you happen to be out of town, or you have moved … the state will kick you off of your health insurance. All because of data which itself would seem to stand as proof that you are Medicaid eligible.
“What we know is that their income has changed,” said DHS spokesperson Amy Webb. “We have never said that they are no longer eligible at that point.”
This is a key point that seems to be a common point of confusion. John Brummett has a good column this morning on this scandal, but he’s off here: “The state has sent out 60,000 or so letters telling people it looks like they've improved their financial lot in life sufficiently to be purged from the rolls of the private option.” In fact, the state made no effort whatsoever to ascertain whether the change in income was sufficient to become ineligible. It simply sent this letter to anyone with any change in income of +/- 10 percent even if that didn’t change their eligibility. And again, the hatchet fell not just on people making a bit more money, but also people making a bit less money. This amounts to purging the rolls of tens of thousands of people who are eligible for the program according to the very data that the state is using to send out these letters in the first place.
There is reason to believe that most of the people being kicked off are in fact Medicaid eligible.
: This is still true, but for different reasons than stated below—based on the new information from DHS, likely many or most of those kicked off are Medicaid-eligible unemployed people. Again, see this post for details.
If the trigger is +/- 10 percent, it's hard to believe that the data verification system isn't flagging a huge number of people who are Medicaid eligible. First of all, anyone whose income went down by 10 percent or more is still income-eligible. Secondly, many of the folks whose income increased by 10 percent or more are still Medicaid eligible. Taking the private option as an example, 80 percent of beneficiaries are below the poverty line—someone right at the line could see income increase by 33 percent and still be eligible; someone below the line could still be eligible with an even larger increase. Some of the people flagged really have transitioned out of eligibility, a normal part of any social safety net program. But based on the criteria used, it's reasonable to wonder whether most of the folks who got these letters were Medicaid eligible according to the updated Department of Workforce Services data. Remember, 97 percent of the folks that got the letter were kicked off the program because they didn't respond in time or DHS didn't process their response in time, so no verification was ever done one way or the other. So how many of those kicked off are actually eligible? Well...
DHS doesn't know how many of those losing coverage are actually eligible according to the very data they are using for verification. They aren’t tracking this at all. Now, they say, they’re too busy to check.
Now, it seems at least a little bureaucratically inefficient to rush to terminate the coverage of tens of thousands of people who are probably eligible, setting up another layer of bureaucracy for them to re-enroll. But that's just me, and I suppose it's water under the bridge. But given the mess we're in, it seems pretty important to know how many of the folks that have been kicked off are actually eligible! I asked DHS for numbers on this since, you know, the letters were triggered by data that could also be used to determine eligibility. "A person's reported income is in each file," Webb said. But, she said, "in order to give you a report of how many had an income changed that would seem to show continued eligibility, we’d have to run a report that I’m told is rather involved. At this time, our focus is on processing redeterminations and getting that verified income."
To be clear: according to DHS the system stores/saves the change in income, but not the income itself, so this plausibly explains why it's not immediately available now. But this certainly seems like a flaw in a system designed to check for income eligibility. I've asked DHS whether there is some legal or technical reason for this and am still awaiting response.
Flagging people for a 10-percent change in income rather than for income eligibility appears to be a policy choice, and it was a bad one.
: Again, the information below accurately reflects what DHS officials told the Times earlier this week. However, they have sense contacted us and provided different information about who was flagged: not all beneficiaries whose income changed by +/- 10 percent, but rather beneficiaries with a +/-10 percent change in income that moved them above/below an eligibility line for one of the Medicaid categories AND any beneficiary with an income of zero. There are still questions about the state's policy choices; see all of the details in this post.
"They could have targeted it quite a bit better," said Tricia Brooks, Senior Fellow at the Center for Children and Families at Georgetown University
, an expert on state-to-state Medicaid eligibility and enrollment practices. Many states have struggled with developing an eligibility verification system, Brooks said. But she said that flagging the 10-percent change in income rather than income eligibility was an unusual approach that she was unaware of any other state doing, and appeared to be administratively counterproductive and run counter to the spirit of the law:
If they are able to collect data and that data shows someone still being Medicaid eligible, then unequivocally they should be automatically renewing them. They should not be sending a notice to anyone whose eligibility is clearly established by the data they're checking.
It seems to me that the logic they are applying here is not consistent with what they should be doing, even given the place that they are in system development.
There is no requirement under the law to do verification based on minor changes in income. The law requires the state to check electronic sources regarding income, if available, and if those sources show that a beneficiary has ongoing eligibility, renewal should happen automatically. If the state has evidence that a person is no longer
income eligible, the law requires contacting the beneficiary—and requiring income verification to maintain coverage if the the beneficiary is in fact still eligible. But the typical standard for sending the letter is evidence that the beneficiary's income is now above the eligibility line, not simply that the beneficiary's income has changed at all. It is particularly baffling that the state is flagging people whose income has gone down. Imagine someone who has a low-paying job and then loses that job. The system would flag that person for an eligibility check. Why?
The particular data system that Arkansas has implemented—flagging based on change in income—has created additional burden and hassle for beneficiaries, led to eligible beneficiaries being kicked off of their coverage, and led to a bureaucratic nightmare as DHS now has to handle re-enrolling eligible beneficiaries.
Again, perhaps the state had a legal or technological reason for pursuing this path, but it seems curious that other states have not done so. Thus far DHS has only been able to tell me that the system is only capturing/storing that there was a change in income, not the actual income. I am still awaiting word on why this is, or why this was the basis for flagging income verification.
The 10-day response demand was a policy choice, and it was a bad one.
This is the crux of the issue. Ten days is not enough time to get responses to a confusing letter from tens of thousands of beneficiaries enrolled in a brand new program. (Can you imagine a private insurance company post-dating a letter which arrived in the mail two or three days later with a deadline a week away — send us this specific paperwork or your coverage ends?) It's hard to deny the problem at this point: 97 percent of beneficiaries who received the letter did not reply in time (well, some of those actually did reply and DHS couldn't handle the volume; more on that in a moment).
Other states have been through this ordeal already and developed best practices, such as sending up followup notices if a reply doesn't come to the first letter (impossible with a 10-day deadline) or using community-based organizations to communicate with beneficiaries (again, massively harder with a 10-day deadline)—and of course, thanks to the Nate Bell amendment
, Arkansas rejected federal funding for Obamacare guides
, who would have come in handy in helping beneficiaries through this process.
DHS officials have noted that they have a 10-day response process for other benefits such as SNAP (food stamps), but by their own admission, those are long-established practices in a system beneficiaries are used to as opposed to a massive new program.
Brooks said that states are within their rights to use a 10-day limit [UPDATE
: after further research, Brooks concluded that Arkansas was in fact violating federal regulations by using a 10-day response limit for what amounts to renewals
] but it's a poor policy decision. "The thing you have to appreciate here—many of these people have moved into coverage for the very first time," she said. "They have never dealt with [this]. They don’t even know they have to do renewal. And then they get this confusing letter and they have no idea what they’re supposed to be doing. Anticipating all of this – what’s distressing is that I don’t think Arkansas has been paying attention to the lessons learned in the other states for the renewal process and picked up some of the strategies that we know soften the blow when all of this hits."
Brooks noted that there is no maximum limit
under the law in terms of response time—states have to verify eligibility annually but they can give people plenty of time to respond to verification letters (Webb earlier this week told the media that the maximum time allowed by federal law for response was 30 days — but she told me today that she misspoke).
There has been some controversy over whether the 10-day limit was Hutchinson's decision or simply a policy he inherited from the previous administration. That hardly matters at this point: it's Hutchinson's choice now, and he is vigorously defending the 10-day policy even at is it is proving to be an unmitigated disaster.
DHS was not ready for the immediate volume created by the 10-day response demand.
The insistence by the Hutchinson administration on the 10-day response demand is particularly troubling since DHS is clearly not prepared to handle the volume of responses. By their own admission, some people have been automatically terminated even though they sent in documentation, simply because DHS was not able to process them in time. (Webb said that DHS has been alerted about such cases but she does not know many there are.)
Under the circumstances, you might think that Hutchinson would be open to a longer response period as the kinks get worked out. But nope. The moratorium temporarily stops terminations, but does nothing for those whose coverage has already been eliminated (including both people are who are in fact eligible and people who replied to the verification letter but didn't have their information processed in time by DHS). Once the moratorium is lifted, nothing will be done to slow down a system that is kicking tens of thousands of eligible beneficiaries off of their health insurance. The governor has hired new staff—a wise (if late) move, but many are understandably wary of DHS being able to handle these verifications going forward given this fiasco. And note that beneficiaries have, under federal law, 90 days to send in proof of eligibility after their coverage is terminated—if they do, coverage must be reinstated. That means DHS will be dealing with even more volume soon. Anyone have faith that everything will go smoothly?
"If you want to stick to the 10 days, then you need to have a mechanism in place to handle the volume so that somehow if you aren’t able to process all of those renewal responses on a timely basis, you at least flag them so that they won’t be automatically terminated," Brooks said.
The governor's moratorium simply delays bad policy. He might instead commit to fixing a system that is, by design, bound to kick tens of thousands of eligible people off of their health insurance, creating a bureaucratic nightmare and disrupting the lives of the state's most vulnerable citizens.