Have you ever read an article in which the writer compares the incomes of the top 1% to the bottom 99% over the last decade, say? Or the comparison might contrast the top 10% to the bottom 90%?

The problem: the author is encouraging you to think that the people in the top 1% at the beginning of the decade are the same people who are in the top 1% at the end of the decade. But they aren’t. People move in and out of this category with surprising frequency. Yet if they aren’t the same people, what’s the point of the comparison?

A similar thing happens in health care. I frequently see writers say that a small number of people spend most of the health care dollars. True. But the small number this year are not the same people as the small number last year, or the year before.

As in the case of the income comparisons, readers can be misled into thinking that our health care problems boil down to how to take care of a small number of people. Not so.

“My momma always said life is like a box of chocolates…
You never know what you’re gonna get.”

– Forrest Gump

A new study by the Agency for Healthcare Research and Quality shows how much fluidity there is among the categories of patients that spend the most health care dollars:

  • In 2008, 1% of the population accounted for about one-fifth of all health care spending. Yet the following year, 80% of these patients dropped out of the top 1% category.
  • The top 5% of the population accounted for nearly half of all health care spending. Yet 62% of these patients dropped out of this category the following year.
  • Although the top 10% spent 64% of all health care dollars, the following year fewer than half of these patients were still in this category.
  • At the other end of the spectrum, the bottom half of the population spent only 3% of health care dollars. Yet one of every four of these patients moved to the top half the following year.

Here is something else that’s interesting:

  • The top 10% are spending almost two-thirds of all health care dollars in any one year.
  • Of those who remained in this category for both years, 43% were elderly.
  • Another 40% were under 18 years of age.

In other words, the persistently sick tend to be young or old. Among the adult, nonelderly population who were in the top 10% the first year, almost three of every four were in the bottom 75% of spenders the second year.

Why is this important? If a small number of people spent most of the health care money and they were the same people year after year, there would not be much point in having a real market for health insurance.

Consider fire insurance. This makes sense only if fires are largely unpredictable and could happen to any homeowner. But suppose that the small percent of home owners who experience a fire in any one year are the very same people who experience a fire every year. In such a world, fire insurance would not be very practical.

The same thing is true in health care.

Most people in health policy view health insurance as just a way to pay medical bills. In fact, I am probably one of the very few people you interact with who believes in real health insurance and who believes there is a social need for it. I am also one of the very few people you interact with who believes we need a real market for health risks in order to determine what is the best way to insure against them and to determine what is the best way to partition insurance products between self-insurance and third-party insurance.

It’s always nice to have one’s view of the world confirmed by the evidence.

Comments (16)

Trackback URL | Comments RSS Feed

  1. Paul H. says:

    Excellent post. Excellent points.

  2. Vicki says:

    I like the Forrest Gump video.

  3. Stan Ingman says:

    John .. Good point!

  4. Brant Mittler says:

    The real purpose of the stats that 1% of the population costs us 22% of health care costs is to drive doctors and health plans to give those sick elderly “palliative care” and thereby hasten their deaths to save money. It doesn’t matter who are in the costly 1% or the 5% — they have to go so America can be globally competitive — to buy more F-35 jets or bigger bunker busting bombs, or longer bridges to nowhere. Our government and the insurance industry and the editors of both the NEJM and JAMA (see their articles which have appeared in recent weeks advocating using (flawed)estimators of life expectancy to “counsel” the elderly ) to jump on this bandwagon that has decided that sick costly elderly patients don’t have lives we can afford to support anymore. See also Ezekiel Emmanuel’s, Rahm;s brother, and annointed columnist for the NY Times, writings about the “complete lives system.” The insurance industry could care less about your parsing these data, John. They just want these folks dead. This is a not so subtle campaign of targeted genocide against the sick elderly.

  5. Harry Cain says:

    John, your points also underscore the high difficulty of making risk adjustment systems work as well as we would like. It’s so difficult to predict who will be, and will not be, this year’s expensive cases based on the last two years’ medical and hospital records.

  6. Devon Herrick says:

    In health care, the best predictor of future spending is past spending. Yet, this report illustrates how loose this relationship is. The good news is: if someone’s future medical expenditure in any given year is unknown, it is relatively easy to underwrite the risk. However, if someone is known to have a problem (e.g. hemophilia), it is not a risk, it’s a certainty. Insurance is ideally suited to underwrite risk; but not suited to subsidize existing problems where healthy people are required to pay more than their expected costs to offset the costs of those whose conditions are unaffordable. In addition, if someone’s medical spending fluctuates year-to-year, they care in a good position to couple self-insurance (e.g. health savings accounts) with high-deductible insurance to offset the risk.

  7. Wonderful uncommon common sense!

  8. Larry says:

    One of my favorite observations – ‘Health care is our destiny’

  9. Robert A. Hall says:

    Unfortunately, Dr. Goodman, the numbers on income or healthcare are not presented to educate the public with facts to inform the debate, but to mobilize voters who pay little attention who can be won over with spin. We are “spinning” ourselves to destruction. Thanks for keeping presenting common sense—it’s so uncommon. I will link to this from my Old Jarhead blog.

    Robert A. Hall
    Author: The Coming Collapse of the American Republic
    All royalties go to help wounded veterans
    For a free PDF of my book, write tartanmarine(at)

  10. Gail Wilensky says:

    Good piece. Important points of information that many do not understand. There is, however, a complication that the concentration in spending raises that complicates efforts to improve the value of care provided to the top 1-10
    % of users. That complication is that these high users will always be spending “someone else’s” money, whether they have private insurance or public insurance, and that means that making sure that the right incentives exist for the providers of care, the payors of care and the users of care is more complex than in other arenas. This is in addition to the problem you reference when people start using insurance to prepay routine and expected health care expenses.

  11. Brian says:

    I never knew that there was such a great deal of change among health care spenders from year to year.

  12. Greg Scandlen says:

    Mark McClellan did some interesting work on this before he went to FDA. The lack of “persistency” is an important aspect to HSAs — give people a chance to save up for those rare years when they need the funds. There remains a population who are moderately high users year after year. This is the group that is financially disadvantaged by HSAs, but it is also where the savings can be found. People who use few services can’t save the system much money. People who use a whole lot are beyond cost containment. The moderately high users are the ones who spend enough to make a difference and are also healthy enough to do it.

  13. wanda j. jones says:

    Let people with health savings accounts cover their own primary care, and design primary care according to distinct customer groups.

    Let insurance cover unpredictable events of care, then

    Let people with the 1 – 7 chronic conditions purchase chronic disease management policies that contain incentives to maintain their end of a life plan and a care protocol and make these “mutual” so everybody with these rider policies have a stake in supporting improvement by everyone. As seniors will age and will heve additional risks, the earlier people enter the system, the better, if providers and health plans take a “whole person” approach to both care and money flow.
    Right now, Medicare will pay only for one disease per office visit!

    Our biggest payoff will not come from pressing down on the terminal 1% but from aiming at boomers from 45 to 64 not yet in the government program and having them push back on those factors that will mean expensive medical care later. No one wants “economic euthan-asia” but we have set ourselves up for it as the government cannot run Medicare with the right forethought.

  14. frank timmins says:

    We have had the argument of trending “reliability” with health underwriters for years (without much success I would add). Insurance companies are mortally afraid of “chronic” illness especially since there has been so much shifting of expenses from facilities to out patient life extending drugs drugs.

    But if the data John Goodman presents is close to accurate (which I believe it is), then the health underwriters over rely on the “experience data” in evaluating risks. In my way of thinking if follows that insurance carriers are overly discriminating for health risks and could ease underwriting restrictions substantially without doing themselves harm.

    What’s the point you may ask? If we are successful in ridding ourselves of the ACA we are going to have to be talking alternatives. And we will still be having to deal with that “uninsured” issue. It’s really too involved to get into here, but health underwriting is going to be a key issue in reforming the healthcare system in a positive manner.

  15. Tania says:

    If you are generally hlahtey, don’t take or need prescriptions or ongoing care, you will save money by taking out a larger deductible plan. You just need to make sure you have savings set aside to cover the remaining balance of the deductible, should worse come to worse.Make sure the plan has a doctor copay that you can afford. I wouldn’t recommend a high deductible plan that does not have a copay.The site below can assist in working out the plan and deductible you should choose.

  16. Eva says:

    Another way that life insurance death give benefit to proceeds can
    become part of an estate is when the insured’s spouse is named the inheritor of lifestyle insurance rules. When your girlfriend receives an death help proceeds, associated with funds perhaps may be paid completly and enlist with your spouse’s liquid strategies.
    Then, they both be paid or not used. When that husband passes away and personal
    assets really are transferred to their estate, some death
    reward proceeds will be part of their total estate and may be subject on to federal real estate taxes.
    The easiest way to eliminate this might be to leave your new death
    bonus in most of the hands of a trust.

    Also visit my web site; term life companies