Fat Stats
Do obese people impose a disproportionate cost on the health system? Being over-weight is certainly strongly associated with high blood pressure and heart attack, and these diseases certainly impose high costs. At first glance this appears to imply that such people cost more. Can you spot the flaw?
People are always going to die of something. From the point of view of a health economist, the question is whether it is quick and cheap or long and expensive. So a health nut who lives to 100 but spends the last 10 years in a nursing home with Alzheimer’s might cost a lot more than a chain smoker who keels over on the train in middle age.
It is an empirical question, and one that has been answered – at least in Holland. This paper published in the Public Library of Science Medicine ran a numerical model inputting their best estimates of demographic and epidemiological parameters for the relevant diseases. Not surprisingly they found that
Until age 56 y, annual health expenditure was highest for obese people. At older ages, smokers incurred higher costs.
But the story was different for total lifetime cost.
Because of differences in life expectancy, however, lifetime health expenditure was highest among healthy-living people and lowest for smokers. Obese individuals held an intermediate position. Alternative values of epidemiologic parameters and cost definitions did not alter these conclusions.
Get that? Even smokers actually cost the health system less! I expect that amateur bungie jumpers cost the health system even less again. And Russian roulette players are a veritable bargain. Dangerous and unhealthy behaviour is not necessarily costly to the health system.
There are still great reasons to encourage or cajole or (maybe) even tax people to not smoke and stay thin. They will enjoy many more years of active life. But arguing that these impose costs on society, implies that society has the right to impose conditions on them. In extreme cases this can lead to rather ridiculous and draconian calls such as the recent bill Bill in Mississippi that would have made it illegal for restaurants to serve the obese. Too bad for Elvis if he wanted to eat out.
Of course, there are other indirect costs of being obese – such as lost productivity - which perhaps account for half the total costs, see for instance a recent Access Economics report. But it seems to me that productivity costs, at least, are also taken into account by employers who will presumably not promote low productivity workers or pay them less. So most of the loss of low productivity is borne by the obese person (assuming it is even true that the obese are less productive) rather than the community at large.
You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
February 28th, 2008 at 10:47 am
But there is flaw in this argument… There is an assumption that the cost of the health system is borne by some disconnected third party (unfortunately the link to the paper doesn’t work and I can’t check whether this is addressed). This is not the case (at least in Australia).
I contribute to my own health costs both directly through health insurance and bills and indirectly through medicare. If I am alive longer I contribute more and therefore my overall health cost (as opposed to the health system’s expenditure) is reduced. This is exacerbated by the fact that I will pay most tax in the last few years of my employment when I have reached the ceiling of my earning potential. This is not true of the smoker who dies at 45 or the obese person who dies at 50. Their outlay on health care may be exhausted much more quickly than mine.
February 28th, 2008 at 10:59 am
I fixed the link Jason.
You make a good point. From the abstract, it does not appear that they are wanting to measure the lost tax of those dying early.
But even if it turns out to be true that dying early loses money for the state coffers, the argument that we should force people to be thin so that they can pay more tax is not nearly as seductive as the argument that we should force people to be thin so that they do not cost the health system money.
February 28th, 2008 at 11:05 am
This is an example of the fallacy of composition. That is, that something that’s true of a part is also true of the whole. Taken to its extreme conclusion we might conclude its better not to have anyone alive because this would minimize the lifetime cost of health expenditure! The research needs to look more broadly at what the objective is which could be to maximize productivity/capita, part of the cost of which is lifetime health expenditure. They might still get the same answer as they did with the total lifetime cost of obesity, but the approach and results would be more meaningful.
February 28th, 2008 at 2:04 pm
Fair enough David but that is much harder to study.
I think that the fact - if it is true - that obese people or smokers do not impose a larger direct cost on the health care system is an interesting and pertinent one. Especially when the assumption that these people impose a direct costs is used to justify beating them around the head.
February 28th, 2008 at 6:19 pm
Well, this is a very cheerie discussion for me. The argument seems to be that obese people do not impose a disproportionately high cost on the health system - because they kark it so much sooner! I should explain - I am, by any reckoning, obese. There, my secret is out, what a relief. No more hiding it and pretending to be thin when in polite company.
To be a bit more serious, what strikes me about this often raised issue is the suite of hidden assumptions, the main one being that all obese people are lazy and entirely to blame for their condition. If you make the opposite assumption, equally untrue, that fat people are victims of something beyond their control, then it becomes like asking the question “Are sick people more of a cost on the health system than others?” The answer suggested by this study is, curiously, ‘probably no’, by the same cold and dispassionate reasoning.
The fact is that smokers, and drunks, and druggies, and fatties, and a few other marginalised groups have two characteristics in common. They are easily identified and therefore become natural categories on which to condition, (in the conditional probability sense). Secondly, from the outside they are easy to see as homogeneous groups, sharing a common character defect, for which the individuals involved are entirely responsible and culpable. Hence the question arises in the first place. Note that if you only require the first of these two as a basis for conditioning, you get all sorts of prejudice such as racism, sexism, homophobia, anti-left-handed-ism and so on.
The fact is conditions like obesity are very complex. In the individuals involved it ranges from being a medical condition over which they have little control to being a simple character flaw over which they have every control. They also range from being decrepit and helpless individuals to being among the most productive and prolific individuals around. (My favourite examples of highly productive, highly obese individuals are G. K. Chesterton and St Thomas Aquinas, the first writing a landmark book on the second. St Thomas was also the one who actually introduced the notions of ‘Gluttony’ and ‘Sloth’ into the Western moral discourse.) The tacit homogeneity assumption on which the rationale of conditioning depends is, frankly, a lie.
The honest and practical thing to do is to marginalise over such heterogeneous categories (rather than marginalise the individuals) when we reckon health costs, just as we do so many others which are not so easily identifiable, but also support public health measures to combat the condition, just as we do measles, polio, tuberculosis and HIV/AIDS for humanitarian as much as economic reasons.
Bill Venables.
March 5th, 2008 at 8:02 am
It does put a huge strain on the health system and it seems only to be getting worse. It needs to be dealt with at childhood by promoting healthy eating habit and exercise.
March 16th, 2008 at 11:18 am
Here in The Netherlands we have the same discussion. Giving fines through insurance fees for people who are too fat, or life unhealthy. And discounts for those who are going to the gym.
The idea is not that bad, but stretch it then completely or it won’t make sense. So you drive too fast? Higher insurance, you have higher risks!
You never go to a doctor for bloodpressure check after 50 years? Higher insurance fees!
March 22nd, 2008 at 10:30 am
It’s very interesting to read how someone’s good or poor habits can determine the overall economic cost up to the time of his/her death.
July 18th, 2008 at 2:35 pm
thanks for the great information…
September 2nd, 2008 at 1:20 pm
See similar discussion about smokers at
http://www.analyticbridge.com/forum/topic/show?id=2004291%3ATopic%3A22778
In the lawsuit against cigarette manufacturers, it was argued by the defendants, after long statistical investigations by world class experts and University professors, that smokers actually cost less to society in general, because their increased health expenditures costs (lung cancer etc.) are more than offset by premature death, and thus the government does not have to pay much to them in terms of retirement benefits, medicare etc.
Is this true?