Monday, June 6, 2011

Why health care cannot not be rationed?

One of the central themes of the debates about health insurance in the United States is the "rationing" of health care. Conservatives believe in the provision of unlimited access to all health care, irrespective of its costs and benefits, and oppose all attempts to ration health care on any cost-effectiveness criterion. On the contrary, liberals argue that the steep rise in health care costs leave governments with little choice but to restrict the types of care covered.

Governments across the world face a two-fold challenge with their health care budgets. On the one hand, a large number of public spending needs compete for scarce resources, while on the other hand aging populations and increasing cost of medical treatments put upward pressure on health care budgets. In this context, the rationing debate is all about whether governments should finance unlimited health care or should cap expenditures at some reasonable level. This essentially boils down to putting a value for human life. This value would vary, decreasing with age.

There are four fundamental challenges with health care that inevitably leads us down the path of rationing.

1. With terminal care patients, especially given the advances in medical technology, it may be possible to prolong lives, albeit at a massive treatment cost. In fact, end-of-life care is taking up an increasing share of health care budgets (approximately a quarter of the $450 billion that Medicare spent last year went to pay for care in patients’s last years of life). For example, many oncology treatments on patients with solid tumors are extremely expensive and prolong life by just a few weeks.

Should health insurance schemes cover these cases? More specifically, should government health insurance cover treatments that have no curative potential and are merely life-prolonging? Should governments refrain from aggressive therapies which have no chance for cure or recovering function? More generally, should governments cover all the possible treatment options?

The problem with rationing here is that distinguishing severely ill patients who are treatable from those who are terminal is not always simple. There is also the issue of treatments that while not curative, can improve the quality of post-treatment life.

2. The perennial problem for health insurance has been about managing the issue of over-treatment. Incentives of doctors, health service providers, and even patients are aligned to over-treat. Doctors and service providers get paid more if the patient undergoes more diagnostic tests and surgical interventions. Subjecting the patient to the full spectrum of diagnostic tests also protects doctors from expensive legal suits if something goes wrong with their treatment. And finally, patients too are comfortable and increasingly demand that all possibilities be ruled out during their treatment.

It therefore becomes critical for an insurance provider (and therefore governments which subsidizes the insurance scheme) to place restrictions that would contain over-treatment. This invariably results in some form of rationing.

3. Then there is the issue of over-diagnosis, especially for the elderly. For example, an "MRI will find nasty looking knee and spine abnormalities in many Medicare-aged patients who don’t (and won’t) suffer from serious knee or back pain. Most men over 80 have a few abnormal prostate cells that will never make them sick (and won’t be helped by treatment) but can be profitably labeled 'cancer'." Similarly, most old people will display evidence of heart disease if one looks hard enough.

Knowing when it is prudent or necessary to start treatment is also essentially a rationing decision. Insurers will have to take judgement decisions to either agree or disagree for certain treatments, though the patients may demand and doctors may want it.

4. Finally, there is the issue of ineffective treatments. The BMJ has classified more than 3000 treatments as either of unknown effectiveness (51 percent), beneficial (11 percent), likely to be beneficial (23 percent), trade-off between benefits and harms (7 percent), unlikely to be beneficial (5 percent) and likely to be ineffective or harmful (3 percent). In Britain, the National Institute for Health and Clinical Experience (NICE) is entrusted with the responsibility of studying treatments and declaring them ineffective or not and whether to be included in the NHS treatments.

Any attempt to restrict treatments whose effectiveness is largely questionable is another form of rationing. In the absence of such restrictions, pharma companies and health service providers have an incentive to push through treatments of doubtful value.

In any case, it is indeed surprising that conservatives in the US, who are all ardent free-market advocates, take umbrage at rationing of health care. After all, free-market itself is the ultimate rationing mechanism - one that rations scarce resources among competing interests in the most efficient manner. In other words, the debate should not be over whether health care should be rationed, but about how efficiently and fairly can this rationing be done.

Update 1 (22/8/2012)

Very good article on the need to ration healthcare by Eduardo Porter in the Times. 


Rajesh said...

Your thoughts on health care has raised many questions and left it short of suggestions to effectively ration health care. Your views strengthen my view that in Indian context it is not the time to invest or focus on health insurance as we have not done much on prevention of diseases.

I think there can be either no ration or ration and " effective rationing " will be unachievable since what is involved is lives which can never be valued by a administrative set up.

I think better will be to spend our resources on preventive health and improve personal hygiene and cleanliness of the surroundings as a clean environment itself will bring down lot of incidences of malaria / diatribes etc., in a country like india. So,

gulzar said...

thanks Rajesh. I agree with your points. but there are two issues here - preventive health care and curative treatments.

the former, and an effective one at that, is an undoubted pre-requisite for any half-decent health care system. but the second is more complex.

at a time when an increasing share of even government employees are relying on private hospitals to treat their diseases at government expense, the utility of an insurance model assumes relevance. it will become more so as the reach of private health care providers expands.

the challenge is to design a health insurance model that delivers bang for the buck - leverages the considerable existing government health care facilities and gets private providers to complement it. can the design provide incentives to get the government entities compete with private providers and improve their competitiveness?