• The cost of chronic illness

    One route to improving health outcomes and potentially reducing health care spending is to focus on individuals with chronic conditions. It’s possible that reducing cost sharing for certain health goods and services for those with chronic illnesses could be cost effective and possibly cost saving.

    That’s one policy conclusion suggested by a new NBER publication by Jean Abraham, Anne Royalty, and Thomas DeLeire. It explores the generosity of employer-sponsored insurance for those with and without chronic health conditions. Using nationally representative Medicare Expenditure Panel Survey data from 1997-2007 the authors find that

    the chronically ill have less generous insurance coverage than the non-chronically ill. Additional analyses suggest that the reason for this less generous coverage is not that households with a chronically ill member are in different, less generous plans, on average. Rather, households with a chronically ill member have higher spending on certain types of medical services (e.g., pharmaceutical drugs) that are covered less generously by insurance. Given recent work on value-based insurance design and coinsurance as an obstacle to medication adherence, our findings suggest that the current design of health plans may put the health and financial well-being of the chronically ill at risk.

    In the following figure from the paper, total spending is on the horizontal axis and median out-of-pocket spending is on the vertical axis. Relative to those with no members with a chronic condition, households with a member with a chronic condition (cancer, diabetes, heart disease, asthma, and anxiety or depression) have higher median OOP spending when total spending is above about $5,000. At the high end of the spending distribution, the gap in median OOP spending is over $1,000.

    Most of the difference in OOP spending is due to prescription drugs, as shown in the following set of figures.


    The authors explain,

    Specifically, we find that it is greater coinsurance for prescription drugs, controlling for total healthcare spending that appears to be responsible for the less generous coverage of the chronically ill. That is, the specific services used most by the chronically ill – prescription drugs—are, by design, reimbursed at a lower rate. This is not due to the higher overall expenditures on average of the chronically ill, since we control for total spending in all of our analyses. […]

    [P]olicymakers concerned about health outcomes as well as cost control may look to recent evidence on coinsurance as an obstacle to medication adherence and question whether it is desirable to deter those with chronic conditions from taking needed medications. Additionally, some recent work showing that health care cost savings generated by lower medication adherence lead to higher spending on other services such as inpatient and outpatient care suggests the possibility that higher coinsurance for this group could lead to increases in overall healthcare costs.

    One might also ask if it is fair that those with chronic illness pay more than those without. That is, to what extent should insurance be a transfer from the healthy to the sick? Does your answer depend on whether you believe the sickness is related to individual behavior and lifestyle choices? To what extent is your answer based on your personal experience with chronic illness, either your own or that of a family member or friend? Should our opinions about policy be driven by such things?

    • Since no one else has posted yet, I thought I’d get this conversation started:

      Well, if it wasn’t transferring from the healthy to the sick, then it wouldn’t be carrying out the purpose of insurance very well, right? I think the real question is who decides which healthy to transfer from and how much to transfer.

      Some people (usually conservative) advocate it as the healthy’s individual choice, both in deciding to participate in an insurance plan and, if they do, in deciding which illnesses to pay for and how much to pay. In my reading of Internet forums, such people seem to almost always think that the vast majority of chronic illnesses are completely preventable by behavior choices (i.e., more like diabetes and heart disease). And they think that the rare exceptions (like cancer, asthma, and depression) could be effectively addressed through private charitable initiatives.

      Other people (usually liberal) think of illness and injury as more of a “there, but for the grace of god(or goddess or FSM or whatever), go I” situation. They see chronic illnesses that could be prevented by lifestyle as maybe the majority or close to the majority, but definitely not the vast majority, while chronic conditions having nothing to do with behavior choices are certainly not rare cases. And thus they think that all people should participate in (probably government-organized) insurance, since it could be anyone who becomes sick, and trying to address everyone through private charity or individual savings accounts would be very inefficient (given human biases to distribute charity to those who are charming vs. who could actually benefit from it, and from human biases towards immediate rewards from not saving vs. deprivations in planning for far-off futures).

      How much our opinions are shaped by our experiences with chronic illness of ourselves and family/friends I can’t answer for everyone, but I can answer for myself. I have several chronic conditions, some common and some exceedingly rare, and some dependent on my past lifestyle choices and some having nothing to do with my lifestyle. I have family members who are in the same situation, with a mix of health problems. For me, my worse condition is genetic, really really rare, and somewhat expensive, and I believe in a much more liberal POV. (BTW, I’ve had that POV my whole life, long before I got this particular illness, but I think I feel it more strongly now.) Another sibiling, who I believe also tends toward the liberal POV (on health care at least), has a common condition from birth that is a little bit costly. My other family members have their most problematic chronic conditions tied to lifestyle factors, and they all take a more conservative POV. (I had the fun of hearing my mother say that she would rather I or one of my sibilings die needlessly, should we get some new illness that society disapproves of and won’t donate to charity for, than anyone in the top tax bracket should pay one more dime in taxes.)

      The most frustrating thing about the health care debate so far is that, as far as I’ve been able to research, no one has actually studied the actual facts behind the two different POVs. Our policy shouldn’t be driven by such diverse “opinions” when one or the other could probably be established as closer to reality. (How many people in the US have what is classified as a “chronic illness”? Of those, what percentages have illnesses that have strong/weak/no/unknown ties to behavior choices? For an otherwise healthy baby/young adult/65-year-old/etc., what are the chances of them developing a chronic condition no matter what lifestyle choices they make? And how much of health care costs go to such things, as compared to one-time random accidents (like car crashes) or to routine care that everyone should expect (like vaccinations and cholesterol checks)?) If anyone could point out such studies to me, I would appreciate it; hopefully they will help me argue more intelligently with my mother the next time we start talking politics. 🙂

    • I have great respect and admiration for Lisa Copen. I live with terrible chronic pain and you are a great inspiration to me. Thanks for your leadership and drive. I will pray for you today and thank God for your encouragement to others.


    • I think that most people would not see subsidizing chronically ill but rich people as desirable. Those people could in fact shop for price and drive down costs or provide funds for cures.

      So I see it as a more general question of who should subsidized and I think that only the poor should be subsidized. If you judge poorness by looking income minus medical spending you should subsidize those that land well below average on that measure. It seems to me impossible to subsidize the middle class.

      • @Floccina: Thanks for your thoughts. I can see the appeal of aiming the subsidies at only the people who are “poor” after paying for their chronic medical costs out of pocket. But I am wondering what your thoughts would be on the people who just miss the cutoff…in particular, would it matter to your plan if their chronic condition arose from lifestyle choice or not?

        I’ll admit, this is a personal interest of mine. Often, I have heard a proposal everyone should have a high-deductible policy, and subsidies should go only to people who pay more than 10% of their income on health costs each year. Since I’m considering entering a field where the average salary is $50k, 10% would basically be the $5k deductible.

        In addition, the drugs and tests for some of my chronic conditions comes to about $4k per year. So basically, if the proposal I often hear about comes true, I would have to expect to pay $4k per year for health costs year after year after year, due to really random chance. (I’m only counting the costs for my chronic conditions that are _not_ related to lifestyle choices…one is a common condition that affects about 1 in 10 Americans, but no research yet has connected it to lifestyle factors, and the other is a really really rare genetic condition. I have done a little comparison shopping on the first one, but frankly the second is so rare that there isn’t much chance for any test provider to get volume discounts.)

        Compared to someone else who has the same education, the same job prospect for salary, the same money-management skills, etc. but who lucked out in the lottery of health, I will be $4k in the hole monetarily for the rest of my life, because I would be under the deductible each year but not low enough to be subsidized. The healthy person would be able to save more for retirement, take more nice vacations, etc every year. For me, it is really apparent that the push to _not_ spread the risk up to a certain point ($ amt or % of income) in order to give “customers” an incentive to shop around is going to really penalize some segment of the chronically ill, even those ill through no fault of their own.

    • My husband is a surgeon and I am psychiatrist, so I write from the perspective of an inside look at the healthcare dilemma.

      What many people do not realize is that when an unfunded patient comes to the Emergency Room, they get the same care as a funded patient — hospitals are required to do this, and humanitarian values require it. Those costs are shifted onto the bills of those with insurance, which is why hospital charges are so exorbitant, even at not-for-profit hospitals. Certainly many indigent patients cannot afford insurance, but there are many middle class and even wealthy people who choose to “go bare” to save money. Therefore, it strikes me as odd that conservatives are against mandated insurance, when they often end up paying not just for the indigent, but for true free loaders. Remember, all hospitals must end up in the black, so the deficit caused by the unfunded HAS to be shifted onto those with insurance, causing premiums to go up.

      A second point is the terrible individual problem when one cannot get health insurance b/c of a pre-existing condition. That means that once a member of your family has a chronic condition . . . diabetes, depression, breast cancer . . . the wage earner is stuck in their current job unless they are fortunate enough to find another company with a group plan. If they are in a miserable marriage and one party has a chronic illness, they are also stuck . . . divorce means no insurance. We cannot fix the pre-existing condition problem without requiring health insurance. Otherwise people will just wait until they get sick and then get insurance,and the system will go broke.

      So, while the Affordable Care Act has many flaws (especially in not addressing pharma) at least it is a start, and one that is long overdue.