Business Week’s Get Healthy - or Else shows how private profit incentives in the US can improve health outcomes – health insurance costs can be reduced and health outcomes improved. The same incentives will work with our private health schemes in Australia.
Firms in the US often pay a proportion of their employee health care costs. As health is expensive this is a rising cost and firms have encouraged employees to bear a wider range of costs themselves through higher premiums and co-payments. This doesn’t work well because health costs are still shared threough pooling of health risks. Hence firms are starting to force employees to take preventative measures to improve their health and even firing them if they don’t. Firms provide discounts on health club memberships, free weight-loss and smoking-cessation programs, gratis gym memberships, counselling for emotional problems and prizes for health achievements like vacations or points that can be redeemed for gifts. These schemes prevent employees from driving up costs simply because they are unwilling to address their health problems. The schemes save firms money and employees who respond to such incentives get healthier. It is a good deal all round and should be duplicated here in Australia where health insurance firms and the State, rather than private firms, provide most health cover.
It should be possible to lower health care costs in Australia by encouraging consumers to signal to private health funds and to Medibank that they are taking measures to improve their health or to prevent its deterioration. This is not the same thing as offering reduced premiums to those with good health – I am not suggesting free competition for clients and indeed, I favour the current universal health care system. What I am proposing is that individuals be offered incentives to encourage them to take care of their health.
For example many chronic conditions such as Type 2 diabetes can be largely self-regulated by controlling diet and by exercising. One way of encouraging self-regulation is to first help people understand that they have the condition – screening is very inexpensive requiring only a glucose tolerance test. The main costs associated with the condition are associated with not dealing with it, rather than treating it, and thereby incurring catastrophic conditions such as blindness, limb amputations, heart attacks or strokes. As I have previously reported:
It would be sensible for the Australian private health funds and perhaps for Medibank to subsidise testing for such conditions and, where relevant, to offer health insurance rebates to those who do successfully manage the condition by regulating diet and by taking exercise. It would be a sound investment that would encourage better health care practices and reduce our health bills. Moreover, this would in no way threaten the system of universal cover for health costs that we enjoy.‘Average annual health costs/per person of the disease (diabetes) were $5360 in direct costs. Health care costs contributed 79% of costs with medications accounting for only 30%. The complications mentioned, which occur only in a relatively small number of DM sufferers, were the main driver of diabetes costs. Annual costs without complications were $4025, $7025 for those with only microvascular complications, $9055 with macrovascular complications and $9645 for those with both.’
I have taken the chronic Type 2 diabetes condition as an example but the general approach would apply to other conditions such as obesity. Private health schemes should be able to provide financial rewards to those who take care of themselves and offer subsidies for screening abnd other forms of treatment where such incentives reduce health insurance costs and improve health outcomes.
Posts
8 comments:
Why not offer reduced premiums for those in low risk groups ? I don't see why this logically different to, say, charging more beause of smoking, and it doesn't seem as unfair as, say, charging 18 year old males more for their car insurance (for which they have no control whatsoever), or charging old people who only buy helath insurance very late an extra premium.
Interesting Harry. But is there evidence of this moral hazard effect with subsidised / socialised health care? I wouldn't have thought there'd be too many people who would conclude that they weren't too concerned if they got diabetes etc because they wouldn't pick up all of the tab for treatment. Ie, the risk of getting sick is sufficient incentive to look after yourself regardless of who pays. The cost would seem secondary to me.
You've previously talked about self-control problems and isn't this more a case of that rather than a rational weighing up the risk of illness and the financial costs that flow from that against the benefits of another burger?
That said, if employers or insurers have a sufficient financial incentive to take an interest in individuals' health, then that could overcome some of the self-control problems - you might have difficulty, eg, quitting smoking by yourself but it might be easier if your employer is supporting you - or forcing you.
Harry,
I don't think that there is a moral hazard problem of the type you describe (well in any significant way)
I don't think that having health insurance or being guaranteed health care induces negligent individual behavior.
No one continues smoking because they know that should they develop emphysema they would get top notch medical treatment.
In fact, I suspect that people in the US without private health cover engage in more risky behavior than those with private health cover.
The main problem for health insurance is deciding which expensive procedure to administer to sick individuals.
Economic theory suggests there will be ex ante moral hazard though the evidence on the existence of moral hazard is mixed. See
http://www.nber.org/papers/w12764.pdf.
One of the reasons is that taking out insurance brings you into contact with physicians who may counsel you to better take care of your health. This may offset some of the effects of taking less care because your costs are reduced.
But once you allow for both effects there does seem to be evidence.
In my view this is unsurprising - if your long-term health costs are guaranteed you are plausibly less likely to take care, to exercise and so on.
Rabee earlier I posted evidence showing that educated people smoke more if they know that quit technologies exist. I think for smokers it : is plausible to suppose they will smoke more if they are health insured for the same sorts of reasons.
Harry
Harry, the models in economic theory are anemic
Consider this simple example of partial insurance (the doctors are not God but bastards model)
An agent has a choice be fat or slim. There are two possible outcomes a curable heart attack (CH) and a not curable heart attack (NCH). We can cure the CH but not the NCH.
1) If he stays fat, there is a 50% chance of getting a NCH and 50% a CH.
2) If he stays slim, there is a 90% chance of getting a CH and 10% of NCH.
If he is uninsured NCH and CH are indistinguishable since he can't get care. So let's say the cost of all heart attacks is $2 for the uninsured agent, because he does not receive treatment.
If he is insured, then the cost of CH reduces to $1, because he receives treatment.
Insured agents will slim down.
Rabee, An interesting model - here the cost of dealing with disease falls if you insure provided you take a costly action (slim down).
In the more usual case I am considering the cost of remaining fat falls when you insure.
Harry
Harry,
In the example the expected cost of staying fat decreases if you are insured. But the cost of going slim decreases even more if you are insured.
Access to health care through insurance induces an incentive to take actions that will avoid incurable diseases. While without access there is no such incentive.
Access to treatment can result in a positive change in your behavior.
(I hate the visual verification you're using on your blog)
Rabee, The visual ID is a pain-in-the-ass I agree. Since I inserted it I have avoided spam completely.
Before that I was spending 10-20 minutes per day manually deleting spam. Let mre know if you have problems logging on.
I am still thinking about the generality of the example you provide.
Post a Comment