Funding the Health Sector


<> Paper to the ‘Funding in the Health Sector Conference’: 28 September, 2009

<> 

<>Keywords: Health;

<> 

<>Why is health funding a problem? The easy answer – the answer popular in the public policy debate – is to talk about the size of the health services sector, the proportion which is publically funded, especially in a context of the totality of government funding. Projecting those figures out, we find that if trends continue, as they are likely to, spending will rise with the likelihood that taxation levels will have to rise to fund public health care.

<> 

<>The Report of the Ministerial Review Group – the Horn Report – is a recent example. In my view it was misleading. The report says New Zealand spends a high proportion of its national income on health, It is higher than the OECD average and — with the exception of the US, Switzerland, France, and Germany — it is not materially different from the highest in world. The latest Ministry publication is only up to 2005. It reports that our health spending relative to GDP was 11th in the OECD (out of 29) so there is another five ahead of us as well as the ones mentioned. (In any case the international figures are not too accurate or comparable.)

<> 

<>Spending on health has also been growing much faster in New Zealand than it has in other countries, especially when compared with income growth. Since 1995, growth in health spending has exceeded growth in national income by 30% in New Zealand versus an OECD average of 18%. Notice the starting date of 1995. Health spending had been heavily cut back under the previous National government, so some of the subsequent growth in spending was no more than a catch up. Nor is it entirely clear that out health care spending has grown faster than everyone else in the OECD. As a proportion of GDP between 1996 and 2005 (the Ministry tabulation period), the following countries had a greater health expenditure growth measured by change in share of GDP: Belgium, Greece, Luxembourg, Portugal, Switzerland and the United States. Britain was the same. That makes us 7th equal.

<> 

<>Why Health Spending Rises

<> 

<>In summary then, we were a highish spender on health care relative to GDP, but we we are not an extreme. The problem of growing health care spending is not a peculiarity of New Zealand – the problem applies throughout the rich world. The reasons are much the same.

<> 

<>First, health is labour intensive so it does not obtain the productivity gains that a goods-based industry does. The cost pressures are exacerbated because (especially) medical skills are internationally very mobile so we have to pay near-US rates for skilled medical personnel. Under the Obama expansion the rates are likely to go higher. Technological innovation is another factor. While sometimes the new methods are cost saving, more frequently they extend the quantity and quality of life, so they increase demand for health care.

<> 

<>Those factors add to the main reason for the strong health care growth. Too much of the health funding debate has a narrow focus which seems to imply that the problem we face is that individuals making unreasonable demand for health care services. We need to see the demand for health care in a wider context rather than blaming the public.

<> 

<>It is a perfectly rational to want more health care services. Increased material consumption in affluent societies does not add to individual happiness. The longest available series is from the United States. It  shows that happiness has not risen at all over the last 60 odd years, and yet material consumption has doubled. There are various ramifications; today I focus on the implications for health care.

<> 

<>First, the sick are less happy than the well. For instance, based on some work that Ryan You and I did at the Massey Centre for Social and Health Outcomes, Research and Evaluation, a person unable to work because they are sick is 15 percent less happy on average than someone who is employed, retired or is a mother looking after children. Reduce sickness and we increase happiness. Second, while we may not be able to increase the average level of happiness, we may be able to prolong it. Life expectancy has been rising – it now averages over 80 years at birth, up from 70 years half a century ago. That is another 14 percent additional happiness per average life.

<> 

<>So while increased material consumption may not add to our happiness, diverting resources from it to effective medical care may reduce mortality and morbidity, which will add indirectly to happiness. It is a perfectly rational strategy to increase spending on health services.

<> 

<>Moreover it is not unusual for some component of our overall spending to increase faster at the expense of the rest – automobiles were an example in the middle of the twentieth century. The ordinary economic mechanisms of the market coped reasonably well, because in order to buy the cars we desired, we cut back on purchasing other things. (Of course the cars required the government to provide roads for our cars. The government is never far away.)

<> 

<>So what’s the problem with health services? The difficulty arises because particular health care needs – in some objective medical sense – vary from person to person. We want a high quality health care system, but hope that we never have to use it, comforted by the thought that it is there if we need it or, indeed, if people we care about need it,

<> 

<>There are a couple of almost equivalent means of evening out the erratic incidence of health , ensuring the facilities are there if we ever need them. One is to raise taxes to pay for the services, the other is to use some sort of comprehensive insurance scheme. They have a common problem.

<> 

<>When you buy a car, you use your own funds and thereby reduce what else you can purchase. Thus your spending is constrained and you choose a car you can afford. You may want a Rolls Royce, but it is probably too expensive. It would be near enough true to say that a society purchases the cars it can afford.

<> 

<>For either taxpayer or insurance funded health care the link between an individual’s consumption and the individual paying for the services is broken. This is inevitable if we allocate the care on the basis of medical need rather than ability-to-pay. I am not uncomfortable with that principle, but the neat market mechanism which links the affordability of consumption to the amount which is supplied is broken. Were we to purchase cars the way we fund health services, every one would have a Rolls Royce. Not surprisingly, given the funding arrangements, everyone demands a Rolls Royce health service when they are in need.

<> 

<>Just about everywhere, once a certain level of affluence is reached, the health system respects medical need over ability-to-pay. Now there is no automatic mechanism constraining the total demand health services to what the community can afford. Once the pure market solution is  abandoned – and we accept that medical need should have a role in the provision of health care – there is the problem of how to constrain the provision of services to within the capacity of the community to provide them. We need to find another rationing mechanism. Otherwise everyone will have Rolls Royces, and the taxpayer or insurance fund will have gone broke paying for them.

<>No country has found a satisfactory alternative. All their solutions have weaknesses which the market mechanism overcomes, except the market mechanism has the overwhelming weakness that it relies on ability-to-pay.

<> 

<>When proponents of the conventional wisdom grumble about rising demand for health care they are avoiding the embarrassment that the market mechanism does not work. So they push their failure to design a market based rationing system onto people’s failure to perfectly reasonably demand a medical need as the criterion for access.

<> 

<>Recall that America is the greatest capitalist economy in the world, that its single largest industry is the health sector, that no other affluent country has pursued market solutions more vigorously and, yet, America has considerable government regulatory intervention and public funding. Even more damningly, it is almost universally acknowledged that America’s health care is extremely expensive and inefficiently delivered – and inequitable. Despite having some of the most impressive medical centres in the world, the average American gets a much poorer health care deal than in any other rich country with which mortality or morbidity comparisons can be made. The other countries do better, but they do not have a perfect solution; that includes New Zealand.

<> 

<>Isnt medical insurance a market solution? That is a question about a comprehensive system where everyone has medical insurance which funds (just about) all the provision of health services. New Zealand medical insurance operates differently; it assumes there is a taxpayer funded health service and tops up the constrained public provision for those who purchase the private insurance.

<> 

<>In a comprehensive national health insurance system with total coverage so that nobody is left out there remains a disconnect between the funder of any individually provided service and the individual paying for it. The pure market’s reconciliation of demand and the funding is still lost.

<> 

<>Can We Get A Better Funding System?

<> 

<>The best we seem to be able to do is incrementally improve the health delivery and funding system leaving the rationing system to various ad hoc interventions. This rules out the switch from taxpayer funding to a comprehensive national insurance system as was overwhelmingly rejected by the public in the early 1990s.

<> 

<>In any case, I cannot see a lot of difference paying for health care through taxes and paying for it through a system of national health insurance. The purest may think the former is a drag on the taxpayer, but the latter still involves compulsory levies. A purist might say they are not taxes; the general public does not see much difference. Differences could be teased out but the time can be better used by exploring funding possibilities in a taxpayer dominated system, and then looking at the problem of constraining public expenditure.

<> 

<>Should we require the sick to fund some of the health care services themselves, thereby relieving the taxpayer? But might the charges be so onerous that some people – especially the poor – do not take the treatments up? Surely we do not want individuals failing to use necessary health services because they cannot afford it.

<> 

<>There are various places in our system where the sick are expected to pay. Nobody, for instance, thinks aspirins should be paid for by the national health. Over-the-counter pharmaceuticals enable one to avoid going to the doctor thereby wasting her or his and the patient’s time and minimizing the financial cost to the public funder too. It also reflects a view that the individual should have the maximum control over their treatment.

<>There are also co-payments including for general practice services and pharmaceuticals. (We try to protect the poor’s pockets with a community services card.) Typically these co-payment type charges are at the simple end of the medical spectrum, and low. They probably save the funder some money, but not much. To what extent it discourages some from taking up necessary medicine, to the detriment of their health, and sometimes resulting in higher health expenditures later is not known.

<> 

<>There are also a group of treatments which are not deemed sufficiently medically necessary to be on the public health. They include alternative medicines and also some treatments by the medical profession which are thought discretionary rather than for medical need – most cosmetic surgery for instance. In such cases the treated pay.

<> 

<>Another cause of payment is where it is accepted the public funder should provide, but perhaps because of timeliness or convenience, or even to have the doctor of one’s choice, the patient purchases the treatment in the private sector. Often the treatment is paid by supplementary private insurance.

<> 

<>Additionally there are those treatments which are not provided by the public funder because they are not considered cost effective enough. The public or insurance funder has to have a set of core services they will fund. However, with the exception of pharmaceuticals available to general practitioners, this set of acceptable treatment is not listed or, if there is a list, it is not readily publicly accessible. The set of permitted treatments may even have fuzzy borders with different DHBs – and even different clinicians – having differing views of what it may contain.

<> 

<>It has been proposed the list of what is (or should be) available be codified. The Horn committee is the latest to recommend this, while there was a major attempt in the 1990s proposals to do this, although the core services committee abandoned the attempt to the extent of renaming itself. There are two major problems.

<> 

<>First, what can be effectively delivered keeps evolving, so any codification would have to be continually updated. Second, any list generates difficult-to-contain political pressures. We see this in the vigorous lobby for a new drug by the drug company and patients when Pharmac is consider whether to add it to the list.

<> 

<>Yet the implicit list of what the public funder will provide – the explicit rationing system – is critical to the way the public health system works. Which explains the demand for its codification.

<> 

<>Reprise

<> 

<>Let me summarise where we have got to.

<> 

<>We have a funding problem which arises because the public understandably and legitimately want to spend more on health care in pursuit of a better quality of life. We cannot rely on the usual market mechanism of user pays, because of the erratic incidence of the demand for treatment and an unwillingness to accept ability-to-pay as a rationing device.

<> 

<>Co-payments provide limited funding possibilities, but they shift the balance back towards  an ability-to-pay with the possibility that some potentially necessary treatments will not occur.

<> 

<>Instead there has to be some public funding agency – based on taxes or (probably compulsory) insurance – to fund the necessary treatments. Since an ability-to-pay rationing system does not apply, it has to be replaced by some other rationing mechanism. Since the agency cannot fund everything, there has to be a list of what they will fund. Its size determines the cost to the funder. Generally that list will be implicit and fuzzy,

<> 

<>The expectation is that the list will increase over time. Together with demographic change and the rising relative costs of health services, health care outlays will probably rise relative to other expenditures. That means that the taxes or insurance payments to pay for them are going to rise over time – probably faster than incomes. For a taxpayer funded system that means rising tax rates to pay for health.

<> 

<>So the rationing list, or at least its contents, is at the heart of the funding challenge. Can we restrain what is on it? That is what most of today’s conference papers are about.

<> 

<>Restraining the List

<> 

<>I used to challenge my fifth year medical students by leaving on the board for the entire lecture ‘either clinicians take responsibility for the resources they use, or some one else will make the treatment decisions’. The students would be very uncomfortable. Were they not going into a career whose ethic was to give the best possible deal for each patient? Why should they take into consideration the cost of the resources they were using. Yet if they did not, a manager would decide whether the resources would be available for the treatment. That is the story of the last twenty years, as managers have been imposed in order to restrain expenditure of health care.

<> 

<>The students were not greatly mollified when I explained that using resources on one patient took away resources from other patients. The medical professional’s focus is on the patient – as it should be. But the treatment takes place in a wider context. If clinicians dont think about the wider context, someone else will do it for them.

<> 

<>I would explain how the Pharmac system worked. Then Pharmac had a capped budget. It chose to fund only those drugs which in total value came under the cap. Rationality required that some criteria had to be used to decide which drugs were included on their list. The exact criteria are complex, but an important element is that the spending has to generate enough medical welfare per unit of input. If a new medicine came along which the million dollars could result in 300 quality adjusted life years being saved, then the drug is likely to be adopted; if a million dollars spent only gave an extra 100 quality adjusted life years, then it is unlikely to be put on the schedule of drugs which GPs could prescribe. (In practice Pharmac might try to beat down the price of the drug – this is an aspect of the system we dont need to pursue today.)

<> 

<>If a drug meets the criteria it goes onto a Pharmac list, and a GP can use it without the patient paying (or paying much). Thus doctors do the best they can for their patient within the confines set by Pharmac.  A similar approach is proposed for medical equipment, although the gains will not be as big as for pharmaceuticals.

<> 

<>The arrangements inside hospitals are different. Since the 1970s most hospitals have a preferred drugs list, which the specialists can use without restraint. They may only use pharmaceuticals which are not on the list with the approval of senior clinicians. Thus decisions of the individual doctor are subject to a collegial review.

<> 

<>The preferred drugs list led to major savings; I cant help wondering whether a similar practice could be introduced for all treatments in hospitals – that is the most expensive can only proceed after collegial review. The effect of the medical protocols is much like this, but protocols are not universal and perhaps those that exist could take more notice of the costs of treatment. You will observe that the approach does not rule out any available treatment; it uses a collegial oversight to control resources: better than an accountant, economist or manager.

<> 

<>Clinical responsibility has got lost in the managerial turmoil of the past few decades. Even the Horn review was too top-down. How can we tackle a health problem unless the clinicians own it? The bureaucrats want to rationalise their safety committees. It is far more important that those interfacing with their patients work in, and promote, a safe cultural environment than having a committee hidden in the bowels of hospital bureaucracy or in Wellington. The Horn report does encourage ‘clinical leadership’ but every clinician has to be involved – to own the problem.

<> 

<>Another weakness in the current system that have to be addressed is over-treatment. It comes in a variety of ways. GPs still tell me they sometimes feel pressured by their patients to prescribe unnecessary drugs. Perhaps the prescription charge is a means of reducing the pressure.

<> 

<>A bigger problem may be ‘supplier induced demand’ or SID. That occurs where the clinician persuades the patient they need some treatment which an objective panel of colleagues would say is unnecessary. That usually requires patient ignorance – a general problem. Second, it probably requires that there is a third party funder – the taxpayer or insurance. If patients have to pay themselves, they are likely to be more cautious than if someone else pays the bill. Third, it requires an incentive structure which encourages clinicians to supply the unnecessary treatment. Among the possible incentives are professional status from doing advanced – albeit ineffective – treatments. However probably the most common reason for SID is when the clinician’s income increases as a result of the treatment. This is said to often happen in the American health system, where supply is more profit driven. Does it it happen in New Zealand among some of our private providers where remuneration includes a profit element? The SID problem may belong more to the private insurance industry than the taxpayer.

<> 

<>The third over-treatment problem is where the patient chooses to have less treatment than is the medical standard, but the doctors insist otherwise. There is some evidence that the elderly want less treatment than they did when they were younger; they may be more interested in care and comfort than prolongation of a poor quality life. Sometimes clinicians may press on them treatments which they do not really want. An anecdote involving a younger patient will illustrate. Bridget, a university teacher, was a feisty woman who got terminal cancer in her forties. Her chemotherapy was perhaps offered a bit too persistently. Towards the end Bridget had to say to her doctor ‘all my life I have prided myself at my ability to convey my ideas. I do realise that I may be losing it a little, but what part of ‘NO’ do you not fucking understand?’ This is a commendable example of a patient owning the problem; one of the strongest reasons for patient consent and involvement in treatment is that they then can own it.

<> 

<>Using ineffective treatments is over-treating. Most treatments do not go through the rigorous evaluation that Pharmac applies to pharmaceuticals. It is quite likely that some regularly provided treatments dont work – at least in certain circumstances. Frequently the clinician proposing the treatment does not know it doesnt work. We need more attention to evidence-based medicine.

<> 

<>Conclusion

<> 

<>I have looked at some of the ways we can improve the effectiveness of health care provision. And we should. But there is a gloomy conclusion to this paper.

<> 

<>Suppose the result of these measures was to represent a ten percent reduction in health care outlays for the same health outcome. (Ten percent is the sort of target you might expect given the Pharmac experience.) That would be a significant saving to the taxpayer, but – a large BUT – in the long run the gains will be overwhelmed by the rising demand for health care and the rising relative cost of the services provided. So the measures  delay but do not resolve the pressures that the taxpayer (or a national health scheme ) is under. I am confident I have not solved the problem. Neither has anyone else.

<> 

<>Even so, we should vigorously pursue a strategy of ensuring that every treatment is effective and desired, while we seek to limit public funding without shifting the back to an ability-to-pay basis.

<> 

<>Endnote: I am grateful for comments on an earlier draft by Rob Bowie and Elizabeth Caffin.

<> 

<>Go to top