This week in Economics Society, Miranda Rogers gave a talk about her waynflete project. In her waynflete project, Miranda asked the question “What can the NHS learn from ‘The Oregon Experiment’ – Oregon’s attempt to ration healthcare?”.
In 2008, the UK entered a state of recession, and the UK’s economy is still far from how it was before the global crash occurred. In the third quarter of 2007, the rate of Gross Domestic Product (GDP) growth for the UK was 1.2%, but the rate of GDP growth in the most recent data published, for the first quarter of 2013, is only 0.3%. This has led to an increase in the budgetary pressures on the economy. These will carry on into the foreseeable future and have affected all areas of government spending. Thus, there is a question as to whether we will be able to fund the healthcare budget, which is growing at an exponential rate. Many public services have already seen a cut in spending and it is likely that this will affect the NHS too, which will have a dramatic effect on healthcare.
One of the major attempts to help our economy to recover has been through budget cuts, and although the NHS has been relatively safe from these so far (as its funding has been ‘ring-fenced’), its budget won’t be increasing anytime soon. The cost of our healthcare service has been rising dramatically over the years, and shows no sign of stopping. Even before the recession, the NHS needed to face up to the fact that demand for healthcare is outstripping our finite resources. This exponential increase in healthcare costs is due to the fact that in healthcare, advances in technology have led to new and improved medical treatments and surgical techniques, which are more expensive and can be used on more people. It is also because of changes to society – we are now treating patients at extremes of ages who would not have received treatment in the past, for example a lot of money is spent on operating on ninety year olds who wouldn’t have been considered for surgery a generation ago. Thus, inflation rates in healthcare are far higher than general inflation rates. In 2011, the King’s Fund predicted that the NHS would experience a £35 billion gap in their spending by 2014/15. The NHS is going to need to make some drastic economic changes if it is going to survive this shortfall, and the State of Oregon in the United States of America might offer some lessons to help us in the UK.
The Oregon Experiment – The Case of Coby Howard
In the USA, healthcare is market led – individuals pay for health insurance the way UK citizens pay for car insurance, whereas in the UK healthcare takes the form of a command economy – the allocation of healthcare resources is controlled by the government in the form of the NHS. However, America does have a form of ‘federal safety net’, and this is known as Medicaid. This is a scheme in which the government pays for the health insurance of very low income families, but each state can alter the programme as they wish, and this is exactly what Oregon did.
When Oregon’s radical healthcare reform, known to many as ‘The Oregon Experiment’, was put in place in 1989, it attracted worldwide attention. The goal was to secure universal healthcare coverage to everyone in the state who did not have healthcare insurance by providing (through the economic method of rationing) a free, basic level of healthcare to everyone who had an income that was below the federally established poverty level – this being, for a family of four, an annual income of $23,500 (roughly equivalent to £15,400). The idea of ‘coverage for all’ came about in 1987, after a 7-year-old boy named Coby Howard, who lived in Oregon, was diagnosed with Leukaemia. The state would not fund a bone marrow transplant for him as it did not come under his Medicaid insurance, and he died a year later. This resulted in extensive publicity and public speculation over the morals and ethics of denying a child such a treatment. John Kitzhaber, an emergency room doctor and President of the State Senate, who would later become Oregon’s Governor, realised that, like Coby, many Oregonians were uninsured – around 120,000 of the state population, many of whom had incomes below the federally established poverty level – and proposed that this problem could be solved if healthcare was rationed to a basic level.
The case of Coby Howard was the inspiration for the original theoretical underpinnings of the Oregon Experiment – the idea of health services for all being provided within the state budget. Kitzhaber knew it was possible to do, but the system was going to have to be radically different to anything seen before. The emergence of the use of economic theory in other areas of life prompted the idea to use rationing to create an effective healthcare system. Rationing is the allocation of scarce resources by the government, as an alternative to using the price mechanism, which is when goods and services are allocated to those who are most willing and able to pay for them. It is both an economically efficient method and an ethical method, as all people have access to the same services, whatever their income. These two advantages could explain why Oregon chose to use rationing to allocate their Medicaid healthcare resources. However, in practice, both these advantages are not always apparent. Rationing is not often used in healthcare, for reasons that I will discuss later, but it became a major part of the war efforts in both World War I and II. It is also used in times of famine and national emergency. There is also discussion about its use in the future with fuel, especially gas. The historical use of rationing in desperate situations such as war and famine served to make the Oregon Experiment controversial, as it was creating a common use of rationing.
The basic level of healthcare that was to be provided came in the form of Medicaid – America’s free health insurance for the poorest members of their country, but the form of Medicaid provided to everyone uninsured in Oregon was more limited than traditionally. This was simply due to the fact that the state budget wasn’t large enough to provide everything on the traditional Medicaid programme to all of the uninsured, and thus the healthcare provided had to be rationed to a basic level. However, deciding on what counted as ‘a basic level of healthcare’ brought up as many issues as it had solved. Thus, ‘The Prioritised List of Health Services’ was created. The intention was that this list, as well as all other methodological aspects of the scheme, would be transparent to the public – they should be able to see and contribute to every decision made about the scheme.
The Prioritised List of Health Services
Oregon created a list ranking 710 condition/treatment pairs (CT pairs) – medical conditions that were felt to be basic with their related treatments – in order of 13 different criteria. These included their effectiveness, cost-benefit analysis, value to society, public preferences, probability of disability or death with treatment and without treatment and number of Quality-Adjusted Life Years (QALYs) the average patient gains from the treatment. QALYs are a quantitative measure of how much a patient benefits from a medical treatment and take into account both how much longer they live as well as the quality of life of the extra years that the treatment buys the patient. One QALY is a year in perfect health, zero is death. However, the way in which some of the other criteria were calculated (such as value to society and public preferences) is not clear. A budget was then allocated by the state for the ‘experiment’ and then a line was drawn across the list when they had exceeded the estimated expenditure to fund any more treatments. The CT pairs above the line would be paid for by the state if a Medicaid recipient needed them, but anything below the line would not be provided for. Ergo, the Oregon Health Plan (often shortened to OHP) was put in place, and it was declared to be a pioneering innovation in healthcare by the general media.
An equally important criterion used for the OHP’s prioritised list was a cost-benefit analysis. Cost-benefit analysis is an assessment of the social and economic costs and benefits of a treatment. Social costs and benefits include, for example, the reduced offending rates of patients treated for drug addiction, and economic costs and benefits include the monetary cost of placing someone in a nursing home or the benefit of having a well patient return to work. QALYs are a quantitative way of assessing health benefits to a patient and can also be considered in the context of a cost-benefit analysis. The number of QALYs gained from a specific health intervention can be calculated using the formula:
QALYs gained=∑_(t=a)^(a+L^i)▒〖(Q_t^i)/〖(1+r)〗^(t-a) -∑_(t=a)^(a+L)▒Q_t/〖(1+r)〗^(t-a) 〗
In this, Qi is a vector representing the predicted weights of quality of life that is health-related, t is the time period after the treatment, L is the disease duration, Li is the duration over which the individual experiences the treatment benefits, a is the age of the individual, and r is the discount rate. Thus, using this formula, the QALYs for every CT pair can be calculated, so an apparently objective quantitative result can be obtained for comparison. However, QALYs are not without their flaws. They are, in practice, subjective, can easily experience bias, and only measure the health benefits experienced directly by the patient as opposed to the benefits to family and society as a result of improved health status, which possibly explains why 13 criteria were chosen instead of just QALYs.
Benefits of the OHP
The OHP quickly became a topic of discussion in medical circles across the globe for managing to ration healthcare, yet still create a fully functioning system. It managed to do what it intended to do. It provided acceptable healthcare for almost everyone who was uninsured in Oregon, as more than 100,000 people acquired healthcare through the programme. In fact, the new system was actually more generous than the old traditional Medicaid one, as it actually ended up covering more services, such as in the area of mental health. Furthermore, the number of services that were not covered was smaller than expected – it covered 530 of the 710 CT pairs listed on the original prioritised list, as of January 1st 2006.
In addition, the OHP was one of the first radical attempts to create a healthcare system for the uninsured. There had not been another scheme anything like it – it was controversial, ambitious, and original. It ignited a change in thinking and approach to the allocation of healthcare resources. The OHP is still in operation twenty four years later, which shows that the scheme must have had enough benefits to the community of Oregon to be still in existence. However, the system is not without its ethical, legal, economic, methodological and political flaws.
Problems raised by the OHP
One of the sources of problems with the OHP was the Prioritised List itself. When it was first created, the rankings showed many odd results: for example, treatment for an open thigh fracture was ranked below cosmetic breast surgery. I believe that it is unlikely that the designers of OHP had intended this to be a consequence, or that they would wish to support cosmetic surgery above treatment for a broken bone. This is probably due to an oddity in the methodology, and is a concept known as ‘The Law of Unintended Consequences’ – that is that the more complicated you make a scheme, the harder it is to predict the outcomes of it, which is exactly what the OHP experienced.
A further, more complex ethical issue arose when patients with liver failure and cirrhosis from over-consumption of alcohol who were in need of a liver transplant were able to get treatment, that is that their CT pair was above the cut-off point. Yet patients with liver failure and cirrhosis from hepatitis who were in need of a liver transplant were not covered by the healthcare package. At a first glance, this situation appears to be immoral – it’s not just that one person is getting a transplant and not the other, but also that the person that is receiving the transplant could be considered to have ‘brought the situation upon themself’, whereas the patient with hepatitis contracted it ‘through no fault of their own’. The reason for this was that liver transplant in alcohol-related liver disease was more likely to be successful than in cirrhosis related to infectious hepatitis, and thus the QALY outcome was higher in alcohol-related disease. This emphasises to us the difficulty in measuring all costs and benefits of a treatment, as QALYs only take into account the health benefits (or health costs) directly experienced by the patient. It is very difficult to measure all the benefits and costs, especially in the presence of uncertainty and subjective judgements, both which the OHP experiences.
The ethical standard for doctors in the United Kingdom is ‘Good Medical Practice’, a booklet that gives guidance for doctors, which states that “You must not refuse or delay treatment because you believe a patient’s actions have contributed to their condition”. If you apply this piece of guidance from the General Medical Council (also known as the GMC) to the situation at hand, the outcome is that both patients should be treated equally – the patient with the greatest need for the transplant should receive it first – as the GMC says, “You must give priority to the investigation and treatment of patients on the basis of clinical need” . Thus, we can see that the OHP decision is still immoral, but not for the reason we first assume. The issue of the second patient receiving treatment and not the first is not unethical because one cause of disease is favoured over the other, but instead the ethical concern is that one patient is getting treatment and the other is not. This then causes you to question the whole idea of the basic package and the list itself, since some people are getting treatment for their problems and not others. Yes, the diseases are not identical as in the example, but then again, no two diseases are ever carbon copies of each other, and each must be treated individually, just as a doctor must “treat each patient as an individual” . This shows us that creating a basic level of healthcare is nowhere near as simple as originally thought, and complex ethical issues that arise lead you time and time again to the ultimate question: Can rationing healthcare ever be moral? However, although the Good Medical Practice’s non-discriminatory approach is ethically just, it gives no sense of an economical point of view, emphasising that it is difficult to strike a balance between economic efficiency and ethics.
For Oregon’s governor, Kitzhaber, the answer to these ethical issues was simple: adjust the line until the issues go away! This is certainly one approach, but there may be better ways to solve the problems.
When the list was created, no ethical considerations were put in place, and from a purely economic viewpoint this would be fine – the ethical problems previously discussed would occur, but these would just be minor disadvantages compared to a much larger economic gain: universal health coverage. However, one can never look at any situation from just one viewpoint, especially not a situation which concerns the health and wellbeing of an entire state. As human beings, our motivations, behaviour and beliefs are complex and hard to quantify and predict. Morality is important to us – the concepts of ‘fairness’ and ‘doing the right thing’ are taught to us from a very early age. The fact is that the list didn’t originally achieve its ethical or financial goals because we aren’t simply theories or diagrams – moral judgement is also needed, as the scheme is about people’s lives. As the politician Kitzhaber realised, if the ethical issues are not taken account of, the trust and support of the people will be lost to what they consider to be an immoral scheme. In addition, it was realised that the balance of ethics with other factors (such as economics) in one domain, for example healthcare, could alter the balance of ethics in other domains, for example, the law or education. The fact is that you can’t separate healthcare from the other aspects of society – changing one aspect has ‘ripple’ effects on all the others.
If we consider this further, we have to decide upon whose ethics the ethical decisions should be based. As we live in an open capitalist society that supports individuality, opinions on all topics vary greatly from person to person. Thus, different people will have different ethical priorities. Whoever makes the decision on what counts as ethically fair and unfair is going to be affected by their own experiences and moral code. One way of tackling this problem is to look at it democratically and have the public vote on the ethical criteria, which would fit in well with the desire for public opinion to be a major part the OHP. However, it could be argued that the opinion of an appointed expert or and experts would produce a better set of criteria than the less-informed public. On the other hand, experts may be biased, for example experts often neglect data (intentionally or unintentionally) that disproves their opinion, but at least an expert may have a greater understanding of the complex issues involved.
A hybrid solution would be to use randomised sampling to obtain a sample of the population and gather subjective data from them on their ethical priorities, and combine this data with data gathered from a panel of experts across all areas of work (economists, doctors, lawyers and so on).
Another issue occurred with keeping the limited services truly limited. Many of the recipients of the OHP received medical treatments although they had conditions that were ‘below the line’ because physicians discovered a loophole in the system that allowed them to provide medical coverage for conditions that weren’t ranked high enough to be in the OHP package. This loophole occurred because the OHP pays for any visits that are diagnostic, even when it wouldn’t pay for the treatment of the condition. This allowed doctors to provide treatment for people whose medical issues would not have been covered under the OHP otherwise, either by treating them at the diagnostic visit (for simple problems like bronchitis) or by labelling the patient with a different diagnosis, which was above the line, and enabled the provision of the treatment the patient needed. The dishonest, but well-meaning behaviour of the doctors in this situation shows that the system has not been implemented well. The doctors were breaking the law, but because they were all doing so they couldn’t all be banned from practice, otherwise there would be no doctors left to treat the patients!
The intention of the OHP was to both save money and distribute resources more effectively. The OHP should have saved the state of Oregon a significant amount of money, since many of the very costly CT pairs were not found above the line on the list, yet this does not seem to have been the case. The list saved 2% of the previous expenditure on healthcare in the first 5 years, which, even taking into account the ever-increasing costs of healthcare year on year, is far lower than expected. One explanation may be the ‘re-labelling’ of the patients and conditions by doctors to enable their treatment to be ‘above the line’.
The savings accrued were not enough to make the scheme worthwhile, thus, in order to raise revenue, Oregon introduced a new indirect tax on tobacco. It is not clear why they chose tobacco, but it seems an obvious choice as tobacco is a demerit good – that is a good that consumers overvalue due to imperfect information, a good that is socially undesirable. This is shown by the fact that there is proof that tobacco consumption is a cause of disease and thus is perceived to increase healthcare costs. However, smoking reduces life expectancy, so paradoxically, it may lead to reduced overall costs, as the smokers will receive fewer pension payments and may require less healthcare in old age if they die younger.
The money from the tax can be used to provide treatments for the diseases tobacco causes, such as lung cancer and asthma. However, this assumes firstly that smoking-related diseases are not already on the list, and secondly that all the money raised from the tax will go towards treatments for these diseases, which may not necessarily be the case. Thus, the effectiveness of using this tax as a redistribution mechanism in this scenario is debatable.
Before the tax was introduced, the equilibrium for cigarettes was at P1Q1. When the tax was introduced, the supply of tobacco and thus cigarettes decreased, so the supply curve shifted from S1 to S+ Tax. This caused the price of cigarettes to increase from P1 to P2, and the quantity to decrease from Q1 to Q2. Thus, the revenue gained was larger than the revenue lost, so the government gained money.
The tax was particularly effective because the demand for cigarettes is inelastic, as nicotine is an addictive product. This means the responsiveness of demand to a change in price is low, maximising revenue.
The problem of liver transplants for certain conditions being above the line whereas other liver transplants for other conditions being below the line brings up methodological questions as well as ethical ones. This problem can be seen as evidence that the criteria for the list were not calculated well, meaning that you could consider the whole methodology of the scheme to be flawed.
There are also issues with the choice of criteria used to create the prioritised list. Firstly, many of them were subjective, such as value to society or public preferences. It is very difficult to see how could be made objective and quantifiable in order to produce valid results.
Another issue with producing valid rankings is that both effectiveness and a cost-benefit analysis were included in the criteria. However, a cost-benefit analysis looks at effectiveness, thus effectiveness would have been covered twice, and so this would mean more weighting than planned was given to the effectiveness of the treatments, and thus skewed the results of the rankings.
One of the key parts of the OHP was the inclusion of public opinion and debate in the scheme. However, this did not produce the positive outcome that the state leaders hoped for. Instead, it created “the political paradox of rationing” – that as decisions about rationing become more public, it becomes harder to ration the services. This is because the legislators had to confront the public on every cut made to the OHP, putting them under pressure to reduce cuts and thus the amount of rationing was less than originally planned. This is because it is in politicians’ interests not to disappoint the public, since the public have control of their futures as politicians. A parallel can be seen with the current day issue in the UK of closing under-performing hospitals.
The concept of impressing the public to win votes can also be used to explain why there is no published data on how any of the 13 criteria that contribute to the list are calculated, nor is there data on the weighting of each criterion. This is surprising, seeing as the OHP had put a lot of emphasis on having a transparent system that the public were very much involved in. Ergo, the lack of transparency in the methodology may indicate that the organisers of the OHP used the calculations as a political tool, that is they could manipulate the system by adjusting the numbers and weightings of each criterion to create a ‘better’ list around the time of elections. Thus, the OHP created an illusion of transparency.
Are there any methods or lessons that can be applied to the NHS?
Although it is not explicit, the NHS does in fact have its own form of rationing scheme – NICE, which stands for the National Institute of Clinical Excellence. Established in 1999, NICE aims to provide guidance on technological appraisals, clinical medicine, interventional procedures and public health, and prevent a situation of a ‘postcode lottery’ in healthcare. Its main role is to decide which services should be provided by the NHS by looking at many different factors of each new drug, device or procedure, just as the prioritised list did – every new drug, device or procedure has to be approved by them. However, NICE does not lay down a set of rules and then apply these rules to every service available. Instead, it uses a piecemeal method, looking at each treatment individually. This allows for variation, as it can set new criteria for each new treatment, and doesn’t have to apply these to the treatments already accepted. Although this means each new technology is looked at individually, it also creates bias, whether it be intentional or not. Furthermore, NICE is not explicit in its method for deciding which technologies to accept and which to reject, meaning that it is able to ‘fudge’ the mathematics it uses when considering the ethical dimensions. The range of acceptable cost-effectiveness is the threshold of the cost of the treatment per QALY gained at which NICE stops accepting new treatments into the NHS. Although NICE’s quoted figure for this measure is between £20,000 and £30,000, calculations by independent researchers has shown that the threshold is actually much higher than this – around £46,000.
Another difference between NICE and the OHP is that where the OHP used 13 criteria to decide on which services to provide, NICE uses just one – a cost-benefit analysis. Although at first one would think that this makes the decisions as to what services to provide less equitable, when one looks more closer at the situation one can see that a cost-benefit analysis actually encompasses most, if not all, of the 13 criteria used by the OHP. In this way, the methodology of NICE can be seen to be more solid than that of the OHP, and due to the Law of Unintended Consequences (which I discussed in the ethical issues section), one would expect NICE to experience far less adverse results or outcomes. This suggests that the NHS has already learnt a number of lessons from The Oregon Experiment.
As of April 1st 2013, NICE has undergone a major reform – it is now the National Institute for Health and Care Excellence. The name change reflects the main change from the reform – the new NICE will be looking into social as well as clinical care, emphasising the changes in our need as a population for social care as part of our healthcare system.
The OHP is not the only example of healthcare systems that use rationing– it may have been one of the first, but it is certainly not the last. Another possible answer to the NHS’s economic woes is a voucher scheme. This was originally suggested by the famous economist Milton Friedman, in the context of education. He suggested that parents were given vouchers redeemable for a specific sum to spend on approved educational services of their choice. This concept could applied to solve the problems we face in the provision of healthcare. This means that everyone would be given a certain sum to spend how they please on their healthcare, but clearly issues would arise if someone ran out of vouchers but still needed treatment. A scheme very similar to this operates in Singapore, where all employees with healthcare insurance (Medisave) have compulsory savings which create a pool of money that can be used for themselves or their families for their healthcare needs, at a reduced cost through government subsidies.
A further idea, being employed in the United States of America, is ‘Obamacare’, otherwise known as the ‘Patient Protection and Affordable Care Act’. This was introduced by President Obama in March 2010, and includes two major reforms: all insurance companies must cover preventative services, and all insurance companies are no longer able to end contracts if the customer falls ill. The goal was to provide healthcare that was universal and affordable, and thus reduce healthcare spending growth, so it had similar aims to that of the Oregon Experiment. If the UK were to change over to an insurance healthcare scheme, rather than the NHS, Obamacare may be the way forward.
The NHS is undergoing major reform – the changes to NICE demonstrate this – but it’s going to need to change a lot more, and soon. An exponential growth in health care costs will create an increasing need to ration services. The present economic conditions have only accentuated this inevitable problem. So could a prioritised list of rationed services be the answer to the NHS’s current and increasing spending gap? Personally, I think not. Although the OHP accomplished its goal, the negative consequences of rationing healthcare mean that the scheme created more problems than it solved, especially those of an ethical nature. I don’t think that markets should be left unfettered, particularly when it is a healthcare market, as, in the provision of health services, morality matters, and markets do not take morals in to account. The OHP serves to prove that whilst rationing healthcare is economically efficient, it is unable to achieve the level of morality required to form the basis of what needs to be an ethical healthcare system. Although the OHP was successful in economic terms, the problems that the OHP uncovered demonstrate the importance of consistency and transparency of approach, and in particular the importance of addressing ethical and moral issues. In the UK, NICE is already tackling an economic analysis of procedures and treatments, although its methodology is not always explicit. It is important to appreciate, however, that healthcare involves more than just economics: people’s lives are at stake, and an ethical framework for decision-making and transparency are vital to ensure that any system of rationing is not only fair, but is to be fair.