As provisions of the ACA are implemented — if they are implemented — there will be a lot of competing claims about the extent to which they are working. How should we judge the effects of the law?
Robert Brook, in JAMA, offers four principal metrics, though admits there are many other sensible ones as well.
- “The first measure of the law’s success should be how much preventable mortality that is due to the health care system will be eliminated.”
- “The second goal of health care reform should be to drive the avoidable hospitalization rate from ambulatory-sensitive conditions to zero.”
- “The third measure of the success of health care reform should be whether it increases to 100% the number of US residents who have access to a system of care.”
- “If health care reform is successful, the growth rate of health care costs for all US residents enrolled in a health care system should be reduced to the growth rate of the gross domestic product or less.”
Doesn’t that all sound perfectly reasonable? It would, indeed, be a wonderful world if success were obtained on all four of these metrics. Yet, I think they’re unfair metrics for the ACA and set it up for failure. Let’s take them in turn.
- I agree that there is a relationship between health insurance and mortality, and a vast body of evidence backs me up. However, the relationship is not an immediate one. Many of the uninsured are young and healthy, though some are not. Most of them, thankfully, won’t die for decades even without insurance. Though good coverage will improve their health and increase their odds of living longer, that won’t be measurable for a long, long time. It may one day be possible to credit the ACA with changes in mortality, but we cannot do so in the near term. If we look for mortality effects in the first years or even decade we will be disappointed.
- I suppose if by “drive” to zero one means move toward zero, then I’m OK with #2. What I want to caution you about is that we should not expect the rate of “avoidable hospitalizations from ambulatory-sensitive conditions” to actually be zero. These are conditions that, according to administrative and billing data, should be addressed without hospitalization. However, it is well-known that we can’t see everything relevant from administrative and billing data. Some individuals with conditions so labeled really do require hospitalization for reasons unobservable to us. Zero is too low a target. I’d look for a decline, though.
- My concern about #3 is that the law isn’t really set up to achieve universal access to a “system of care” (which is more than emergency department care). No question that it should increase such access, but 100% is too high a target. Some classes of individuals — like undocumented immigrants — will not have the same type of access to insurance subsidies and care as others. There will be regions of the country where access is more limited. That should be improved, but the ACA is not designed for improvement to the 100% level.
- Though it would be nice to see health care spending achieve a GDP or lower growth rate, I’m not convinced that should be or is a goal of the latest reform. The Medicare target in the law is GDP+1%, after all. Moreover, it is not unreasonable for a nation to spend a greater share of wealth on health as that wealth increases. Naturally, that cannot be maintained indefinitely, but GDP+0.5% or GDP+0.25% can be maintained for a very long time. Perhaps the right level of spending on health as a proportion of GDP is much higher than we see today. That’s not to say we don’t have waste in the system. We absolutely do! It’s just to note that expecting a GDP or lower growth rate is a very ambitious and possibly inappropriate goal.
We should definitely establish goals for reform and continue to make reforms until our goals are met. But it is important that those goals be sensible and reasonable. It’s crucial they don’t set up the reform for failure. It is also important that metrics for them are consistent with the intentions of the reform. Not all of the ones suggested by Brook are.
I’ll bet you can think of some goals for the ACA? How will you judge whether it is successful or not?
AF
