Friday, August 1, 2008

Health care

There's something that's always bugged me about the idea of injecting market incentives into the health care system as a check on patients' consumption of services. As far as I understand it, the idea goes like this: if patients have to bear more of the cost of their health care, they will decide to purchase the combination of services that is most economically efficient for them. People will go to the doctor when they need to, they will stop getting unnecessary tests and treatment, and the overall costs of health care will go down.

The problem is in the assumption that people should consume health care services the way they consume other goods. Take shoes. Say I have two possible shoe insurance plans. With Plan A, I pay a flat fee (copayment) every time I visit a shoe store, but while I'm there I can pick up almost all the shoes I want. For the really really expensive ones, the Manolos whatnot, I have to convince the shoe shop owner that I really need them. But if I can prove that I need them, I can have them, and it'll only cost me a fraction of the price. With Plan B, instead of paying a flat fee, I have a shoe deductible, so that in one year I have to pay for the first $1000 of shoes I buy. But, if I end up needing lots and lots of shoes, the Plan will cover much of the cost over the first $1000. Having to pay that first $1000 out of pocket is going to be a big deterrent for buying shoes, so I'll almost certainly consume a lot fewer shoes on Plan B than on Plan A, and then only the ones that I need, rather than the ones that are just really pretty. Plan B is therefore minimizing my shoe purchasing costs.

With health care, on the other hand, there's a big difference between the cost of primary preventive care (a check-up, an HIV test) and the cost of catastrophic care (pregnancy, cancer). There's also the link between the two, such that consuming the right amount of primary care can reduce your consumption of catastrophic care: getting a patient to exercise and stop eating butter and shrimp when they come in with high cholesterol is much cheaper than hospitalizing him for a heart attack. High-deductible plans put the cost of primary care on the patient but cover the cost of catastrophies. What kind of incentive does that give to the patient? If I have a cold, and I have to pay for my doctor and my meds out of pocket, I'm more likely to wait until I have pneumonia before getting treatment.

The incentives are backwards! We want people to get primary care. We want everyone to go to the doctor at least once a year, to go to the dentist at least once a year. Some healthy people should do more: women should get annual pap smears, pale people should get annual skin check-ups. This care should be free, or at least very very cheap. And I'm sorry, but $500 a month in health insurance premiums is not cheap.


Anonymous said...

The basic point of the pro-market-incentives crowd, I believe, is that since people with health insurance don't pay the marginal cost of their health care, they consume more of it than they would if they did. For catastrophic care, this is a good thing, and this is the point of insurance, but the reason why it works is because the odds of a catastrophe are low, and all of the people not experiencing a catastrophe pay for those that do.

For non-catastrophic care, it's not quite as good a thing. The costs are lower and are much more predictable. It's not clear to me why insurance is really necessary here at all. I'm fine paying the marginal cost of these services. Take the dentist example, I don't really understand why checkups are ever paid through health insurance plans. I go to my dentist twice a year for a checkup. That cost is predictable. So why do I need insurance for it?

With respect to the cold example: "What kind of incentive does that give to the patient? If I have a cold, and I have to pay for my doctor and my meds out of pocket, I'm more likely to wait until I have pneumonia before getting treatment."
That doesn't seem right to me. Perhaps you've just taken an overly extreme example, but I wouldn't want to risk my health like that just to save the relatively small cost of getting my cold treated. And I'm not sure you did pick an extreme example - your point is that catastrophes are paid by the insurance companies, but who is willing to let their health deteriorate to the point of catastrophe?

I'm just trying to think through the issues here, so perhaps I'm wrong on all of this. I understand that the Singaporean system is often lauded as being more market driven than the average health care system. All people are covered by the government for catastrophes. For non-catastrophic health care, people have to pay out of their own pocket, although they are forced to save some of their income in a health savings account, which is tax free, and they also have subisides for the poor and elderly to supplement these savings accounts.

Anonymous said...

Oh, and they throw in the argument that individuals are a better judge of what their non-catastrophic care needs are then either insurance companies or governments. That rings true but I can't really make an informed call.

sai said...

Thanks for the comment! I confess I was hoping for something exactly like this to force me to think through these issues more thoroughly.

You said: "Take the dentist example, I don't really understand why checkups are ever paid through health insurance plans. I go to my dentist twice a year for a checkup. That cost is predictable. So why do I need insurance for it?"

I'll follow your example. First, my assumption is that it's better to avoid cavities than to treat cavities. It's cheaper, takes less time, and preserves your teeth, which is less expensive in the long run (because you can avoid additional expenses like root canals, dentures, and surgery). The cost of prevention, in this case, is less than the cost of treatment.

I'm also going to shamelessly generalize from my personal experience. The years when I went to the dentist regularly, I had no cavities or very small ones that were easy to fix, whereas when I skipped a few years, I ended up with a bigger cavity that required anesthesia to fill. I'll assume, for the sake of argument, that most people are like me.

Now, if I have to pay full price for my twice-yearly dental checkup, I'm going to be less likely to do it than if it's free. If I'm on a tight budget, I'm going to cut this type of preventive care out in favor of spending the money on things I need today, like milk, or gas, or a toy for my kids. I will probably do this because my teeth are fine today; they can hold out for another month or two. So I put it off and put it off and put it off until I have a toothache, which is more expensive to treat. I'd rather not have the toothache; it'll cost me more to fix that than it would to go to my checkup, both in money and time. And yet, here I am.

My point here is that the rational decision in this case would be to pay for my checkup than to wait and risk something serious, and yet I'm not making that decision because I am underestimating the cost of my future risk when I undertake it in favor of paying for my present needs. Health care is unique; I don't think the rules that apply to normal consumer goods are particularly useful in understanding people's behavior in consuming health care services.

Because, overall, it's more efficient for people to get preventive care than to get treatment, we should, as a society, be making prevention as accessible as possible, so the barriers to getting it are reduced. One solution is to make this sort of care free.

Because making preventive care free encourage people to use it, it minimizes the overall cost of catastrophic care. One example is cancers with easy and relatively cheap diagnostic tests, like pap smears and prostate exams. Catching cancers early brings down the cost of treating them and keeps the people who have them healthy and productive for more of their lives. If we can do things to minimize the cost of catastrophic care - which is most of the expense in our health care system - it'll do a lot to bring down costs for everyone.

One argument I hear is that if care is free, people will over-consume. But don't forget that health care services aren't fun: I know I personally don't like being poked and prodded any more than necessary, and if the prevailing wisdom is two checkups a year, I'm not going to get eight.

I really like the Singapore system you described. At first glance it sounds like it's doing lots of good things: it forces people to put aside money for preventive care, which they might otherwise skimp on; it covers catastrophic care, which I think most people agree is a good thing; it makes allowances for people who might not be able to afford care on their own. I'll do some sniffing around to find out more; if it turns out they also cover people with congenital diseases, it might make a believer out of me.

I'll just briefly address your last point about my pneumonia example. You said: "who is willing to let their health deteriorate to the point of catastrophe?" The answer is, lots of people, mostly poor people, who can't afford preventive care. I agree, pneumonia is probably extreme, but certainly a plain ol' cough is not going to force someone to take half a day off work to go to a doctor and then pay for that out-of-pocket if it's likely that the doctor will just say "get some rest and drink your fluids." It's only when the person's daily routine is affected - say, coughing up blood, or not being able to work - that he'll be willing to take the time to get medical help. And at that point "medical help" is usually the emergency room, where the taxpayer picks up the tab - for a chest x-ray, or an overnight observation stay - anyway.

Anonymous said...

Yeah, you have a point. Procrastination means people won't seek as much primary care as they should, and that introduces a role for government intervention, just like 401(k) and social security incentives, especially since the private sector seems to have failed so miserably.

But... a couple things. By free you mean provided by the government and paid for by taxes, which isn't my *usual* definition of free. It's more like the marginal cost is zero because we'd just be paying a lump sum each year.

And if you're allowed to shamelessly generalize, then I reserve the right to baldy state the following: reducing the price of the service will increase its consumption. It doesn't matter that medical services are uncomfortable for the patient; we're talking about the exact same service for a different price. If the goal is to increase the consumption of primary care services, I'm skeptical that the optimal marginal price happens to be zero.

There's probably more informed things to say on the subject. It seems that we underconsume primary care in the USA compared to what is optimal and especially compare to most other developed countries. It's not clear to me, though, whether that's a selection issue -- that 80% of the country are fine but the poorest 20% screw up the statistics -- or what. Because we also overconsume medical services in general -- as evidenced by its larger share in GDP, even after subtracting administrative costs. Your argument would probably be, if I may guess, that we overconsume because too many problems fail to get nipped in the bud. I bet others believe it's because we abuse the system by spending just up to the deductible limit and otherwise departing from what's optimal in favor of what we can get away with. Sounds like an empirical question and I'd be interested in seeing what work has been done on it.

sai said...

You know, I've been trying for a few days to come up with a clever reply, but the fact is I don't have any empirical research up my sleeve, and each line of thought ends with "...but I really just don't know."

I will admit that my inclination toward providing free (or low-cost, flat-rate) preventive primary care services has a lot to do with the kind of society I'd like to live in. I guess I'd prefer my community to provide certain services to its members free of charge, no questions asked: school through 12th grade, access to clean water, two doctors visits a year. If that results in some slight over-consumption of preventive health care services, I think I'm okay with it.

Anonymous said...

Take a look at this article by Krugman and Wells:

In section 4, they come around to what has been focused on here - whether overconsumption due to health insurance is a big issue and how to deal with it. They have a few answers, ordered by how convincing I think they are:

(1) The bulk of health care costs are not on the end of routine doctor visits and preventive medicine. 80% of health care costs are accounted for by 20% of the population. So quantitatively, this stuff isn't the highest priority.
(2) People don't make wise decisions when paying for medicine out of their own pocket. This accords with your intuition, and cites a Rand study for empirics.
(3) The common policy option advocated in these situations is to establish health savings accounts which are tax free. Two specific problems with this proposal are that they were further aggravate the adverse selection problem in the employer-based health care system, since the more healthy would choose to leave the system, and second that this does not help the poor with low incomes very much at all, while it is quite nice for the rich.

I think/help that helps illuminate the issue. Important stuff.