Owen Heisler discusses ethical resource allocation in this thought experiment

Owen Heisler, MD, MBA, FRCSC, FCCHL, is the Assistant Registrar of the College of Physicians and Surgeons of Alberta. This piece discusses the ethical allocation of scarce healthcare resources and the challenge that there is no consensus among stakeholders as to what ‘ethical’ is or should be.  

Introduction

For allocation decisions, ethical theories facilitate asking better questions, not providing definitive answers.  Varied ethical frameworks often provide conflicting answers.  Ethical theories are better considered lenses to enhance developing solutions that respect and reflect the context framing a problem.   A return to basics by considering fundamental ethical theories promotes better understanding of ethical frameworks and how they apply to current leadership frameworks such as LEADS.1  A hypothetical allocation problem provides such an opportunity.

Scenario

Consider the following: Imagine an entity responsible for the healthcare needs of 4 million people – it might be a state, province, health trust, or health region.  The responsibility is inclusive; there are no other providers in the health care sector.  Your population pays the taxes supporting health programs plus they have variable individual responsibility.  You spend $20 billion on healthcare annually; $1.5 billion of this is for prescription drugs inclusive of $500 million for seniors.  While most of your drug spend is for inpatient care, there is a senior’s co-pay outpatient program.

An additional $100 million is available in the budget for new health-related services.  After much discussion, the options have been narrowed to three.  Your team has decided to convene a non-binding focus group to provide direction in making the final choice.  This focus group includes clinical experts, advocates for each of the three options, government, and media.  The government is going to the polls in 18 months for what is expected to be a hotly contested election, centering on healthcare issues.  The media has been antagonistic towards the government healthcare agenda – there have been many critical reports, often identifying individuals who have been poorly served by the current system.

You have developed comprehensive business plans for each of three options:

  • Fund a new Cystic Fibrosis (CF) drug (like ivacaftor) that improves symptoms for a small subgroup of CF patients.2 There are currently 200 patients eligible for this drug. The additional cost per year per patient treated would be $500,000. The current cost for treating CF patients is $10,000 per patient per year.
  • Fund a new Hepatitis C drug (like sofosbuvir) that can cure the disease in a large proportion of treated patients.3 There are 20,000 eligible. While the direct cost of the drug is $50,000 per patient, the marginal cost per patient treated would be $5,000 as other costs will be eliminated reflecting significant cost avoidance.
  • Increase home nursing care funding for all Seniors. The 400,000 seniors over the age of 65 (expected to be 600,000 by 2021) are currently eligible for $200 per family per year. Given current utilization, the extra $100 million would increase available funding to $500 per year. There has been increasing demand for home services as decreased funding for acute care has led to earlier discharge from the hospital, a significant issue for the senior population.The above information is transparently available to the focus group.

However, there are other factors known only to subsets and not shared widely:

  • Government polling suggests to be re-elected, it is imperative to carry the seniors vote. The government directed their focus group members to support initiatives to attract these voters.
  • The Media have a strong, respected leader who has a child with CF. The child would likely be a candidate for the new drug. The leader regularly solicits donations from colleagues in support of CF research. However, the leader is not aware a new CF drug is under consideration.So, how do we move forward? What principle(s) should be applied? Should we fund for cure or for symptom control? Seniors or children? Do we owe anything to anyone? Should we marginally help the many or maximally help the few? Does it matter who pays the taxes? Do we owe more to those we have disadvantaged because of other choices? Should cost avoidance be an important consideration? Do sunk costs count? What is just? What role will hidden agendas, direct or indirect, play? Do you think this focus group will go well? Do you think it will come to a consensual recommendation.

Ethical theories

This scenario pits ethical theories against each other.  Before considering their application, it is worthwhile reviewing the frameworks themselves.  This case is about normative ethics – the questions about what one ‘should’ or ‘ought to’ do.  Normative ethics can be grouped into two large categories, either considering the alternative chosen (the ‘act’) or the person making the choice (the ‘actor’).  Emphasizing the actor is virtue ethics, classically advanced by Aristotle.4  Aristotle believed the actor to be more important than the action.  The primary goal is therefore to have virtuous decision makers who will ‘naturally’ do the right things; their actions will be natural outcomes of their character and therefore right.  Aristotle believed being virtuous is a learned behavior that can be taught but requires considerable practice.  Considering a virtue as the ‘golden mean’ between two vices, virtuous individuals consider alternatives, deliberate about them and, as a result of their training, voluntarily choose the correct action based upon the nature of humans to aim for eudaimonia.  Eudaimonia represents the state of happiness and well-being that, according to Aristotle, is fixed in human nature.

Normative theories that concentrate on the act codify ‘right’ and ‘wrong’ actions.  The two major theories are utilitarian (outcome based) and deontologist (based on duty).  Utilitarianism, as originally proposed by Bentham and Mills, is consequence driven.5  The rightness of an action relates entirely to an algebraic determination of the production of overall happiness.  The right act is that act which maximizes happiness no matter what it takes to get to this point – the end justifies the means.  Deontological ethics, on the other hand, argue adherence to duty, rather than consequences, is most important in analyzing the rightness of an action.  Immanuel Kant, classically tied to this theory, advanced general rules or ‘maxims’ to evaluate the rightness or wrongness of actions.6  Actions are right, he said, if they 1) are able to become universal laws; 2) treat humanity as an end and never only a means; and 3) treat others as autonomous agents.

Virtue ethics dominated healthcare until the middle of the 20th century.  People trusted those in charge of the system to make the right decisions.  Physicians were trained to do the ‘right thing’ and patients trusted they would do so.  However, the lack of consistency in evaluating actions and the paternalistic nature of this framework were challenges.  There was increasing interest in developing guidelines and standards to evaluate and guide the actions of providers.  Beauchamp and Childress’s groundbreaking work applied deontological ethics in health care where the maxims evolved to four basic principles of beneficence (the Golden Rule – do good), non-maleficence (the Silver rule – don’t do bad), autonomy, and justice.7

Scenario Analysis

This scenario is a meta-allocation problem.  The macro-allocation decision on funding is complete – we have $100 million.  Further, we are not considering micro-allocation issues as to which individual patient in each group receives care.  Part of the ultimate answer (with another set of ethical issues) may be to have a bigger pot of money (e.g. reallocating money from education funding) or restricting services to only certain individuals within each group (e.g. only those with hepatitis from tainted blood transfusions).  While these discussions need to be convened, the challenge for the chair will be to focus on this meta-allocation decision.  Participants have a tendency in such discussions to lose focus and defer solutions to alternate allocation levels, blurring issues into what can become a circular, frustrating discussion.

Utilitarian and deontologist normative frameworks are most applicable to this scenario.  Both need to be considered in the context of individuals and groups.  It is instructive to develop a two-by-two diagram where on one axis we recognize the ethical theory and the other the context (See Figure 1).

Figure 1: Act based normative ethical theories

Utilitarian(outcomes based) Deontological(duty based)
Individual Hippocratic Respect for persons
Population Social utility Social justice

Physicians and other caregivers are classically positioned in the upper left quadrant reflecting their fiduciary relationship with patients. The basis of the Hippocratic Oath is to forego one’s own needs to advance the needs of one’s patient.  In this scenario, the advocates for each of the interest groups can be expected to advance stories about patients they are caring for.  There will be real names, real outcomes, pictures, and tragedies.  Their support will likely be strongest for the CF and Hepatitis patients, the more defined and granular groups.  They will say it is ethically right to look after the young and disadvantaged.  “We must do good.”

Media reflect the top right hand quadrant, respect for persons.  Optimization in this ethical framework is a strategy of complete equality – limit what is provided to only those things that can be provided to all.  Advocates for this strategy will also be advancing individual stories, in this case reflecting on where principles failed.  As the decision approaches its end stages, they will be the voice of any segment not chosen given there are principles why each group should be advanced; this risks delaying resolution through consensus. They will be advancing the negative version of the fiduciary argument; it is not ethically right to ignore any groups.  “We must not do bad.”

Both upper quadrants reflect the great weight placed in health care communication on stories.  Healthcare communication is the slave of the anecdote, a reality that can be an impediment to adopting evidence and population based protocols.  For example, immunization programs proven to enhance overall public health by avoidance of disease are hampered in their implementation with stories of individuals who had an unfortunate outcome.8  The unfortunate reality is that a policy which might result in avoidance of negative outcome to ten while directly causing a negative outcome to one (net positive of 10 – 1 = 9), is doomed to failure since we cannot identify by name the ten who avoided disaster and can easily identify ‘Mary’ with the bad outcome.

Governments are motivated towards the bottom left quadrant, social utility – the ‘happier’ a population, the better the chance of getting elected.  The challenge is both promotion of ‘free riders’ who get benefits without contributing and the unfortunate reality that maximal happiness for a large group occurs when there is a significantly disadvantaged minority group; the majority have more resources to share and paradoxically are ‘happier’ because they are not part of the disadvantaged minority.  In this scenario, if the government is confident the home care support program is perceived as positive by the senior population they would support this program by arguing society has an ethical duty to the elderly who have provided so much to get us to where we are.  “It is clearly just and fair.”

So who is going to populate the bottom right quadrant, social justice?  It should be the goal of all healthcare leaders to strive for principle based allocations optimizing health for an entire population, the population that determines and pays the taxes supporting the system.   These might be the healthcare leaders that wonder about the Hepatitis C population and how many are in this group because of societal challenges, be it in tainted blood transfusions or less than ideal management of addicted populations.  These might be the healthcare leaders that wonder if supporting home care in the current manner is the right use of not only new but existing dollars.  These might be the leaders that wonder how to better support all CF children and not just a sub-population.  These might be the leaders that wonder not only about rights in a society but also responsibilities.  Is it you?  Should it be?  Could it be?

Discussion

In this scenario, there are principle based reasons one could choose each option.  Ethically, each of the decisions could be supported, depending on perspective.  It is likely during debate that the ‘ethic card’ will be used to advance a position.  The challenge is when individuals say something is ‘ethically’ or ‘morally’ right this is usually based on their unique perspective, shared within their local culture.  A position so held is often assumed to be a universal, a tool to solicit support when attempting to capture a moral high ground.  Participants in healthcare debates often overlook the multiple, often disparate, perspectives that should be considered – the ‘walk in the other shoes’ so many do not take because of the comfort of their current familiar ‘pair of shoes’ (even if the shoes are old, weathered and torn!)

There is no hard and fixed rule as to what to do when principles collide.  The best decision is the one that considers all points of view fairly and transparently; everyone should be heard.  The value of ethical frameworks is as a tool to assist in considering all perspectives, not to provide the ‘right’ answer.  Those who suggest reliance exclusively on principles to make these trade-off decisions only demonstrate in their analysis the exponential challenges as different perspectives are layered into decisions.9

In order to move forward and bring the scenario in this case to a fair conclusion, it requires leadership.  It requires understanding the people, the culture, and the subtleties that are not in the scenario but rather in the hearts and minds of the people mutually struggling to make the best use of limited resources.  Such decisions are not made in a vacuum; the conclusion needs to respect a risk analysis of the ‘pebbles in the pond’ that invariably flow from it.

The current leadership paradigm places heavy emphasis on training to amass content knowledge and apply business standardization principles such as process redesign and elimination of waste.  However, the business literature does not suggest such standardization is the answer to all problems.  Rather, when inputs are variable (as each patient is variable) and when customers value variation in outputs (individualized medicine) processes must be less about standardization and more about investing in employee skill development, judgment, and cultural appreciation.10

Leadership starts with a respectful environment wherein issues can be transparently and safely considered at a decision table.  The challenge is when leadership stops at this point, a cycle of unending meetings and frustrated participants accrues.  What is needed are leaders that move forward with a multi-faceted approach including measuring performance by patient outcome, applying both financial and behavioral incentives, optimizing processes, and re-engineering current dysfunctional cultures – leaders who organize care delivery around the needs of patients rather than providers.11

So how do we develop such judgment and cultural appreciation in leaders?  In health’s value-based environment, the answer is a return to a virtue ethic, the ethic of caring.12  It is possible, as Aristotle noted many centuries ago, to train leaders to do the right thing; this is ‘leading self’ from which the LEADS framework builds.  The LEADS framework at its very heart is about ‘what’s old is new again’.  It is about leaders taking the time and energy to develop judgment, build networks, look at the big picture, and utilize all available tools to optimize decisions.  It takes the four cardinal virtues of prudence, justice, temperance, and courage.  It is about the art of medicine rather than the science.  It is about working with others, reading around things rather than just about things.  It is expanding rather than focusing.  It takes time.  It takes mentorship.  It is possible – with the downward spiral of the current paradigm, it has to be.

References

  1. Canadian College of Health Leaders. LEADS framework. Available at: <http://www.leadersforlife.ca/;eads-framework&gt;.  Accessed October 7, 2014.
  2. Bell SC, De Boeck K, Amaral MD. New pharmacological approaches for cystic fibrosis: Promises, progress, pitfalls. Pharmacology & Therapeutics. 2014 Jun 14. doi: 10.1016/j.pharmthera.2014.06.005. [Epub ahead of print]
  3. Kim DY, Ahn SH, Han KH. Emerging Therapies for Hepatitis C. Gut and liver. 2014;8:471-9.
  4. Aristotle, Sachs J. (trans). Nicomachean Ethics. Newburyport MA: Focus Publishing; 2002.
  5. Sen A, Williams B. Utilitarianism and Beyond. Cambridge University Press; 1982.
  6. Sullivan RJ. An Introduction to Kant’s Ethics. United Kingdom: Cambridge University Press; 1994.
  7. Beauchamp TL, Childress JF. Principles of Biomedical Ethics 6ed. USA: Oxford University Press; 2008.
  8. Schonberger LB, Bregman DJ, Sullivan-Bolyai JZ, et al. Guillain-Barre syndrome following vaccination in the National Influenza Immunization Program, United States, 1976-1977. Am J Epidemiol. 1979;110:105-23.
  9. Levitt, D. Ethical decision-making in a caring environment: The four principles and LEADS. Healthcare Management Forum. 2014; 27:105-7.
  10. Hall JM, Johnson ME. When should a process be art and not science. Harv Bus Rev 2009, March:58-65
  11. Lee TH. Turning doctors into leaders. Harv Bus Rev 2010, April:50-8.
  12. Faith, KE. The role of values-based leadership in sustaining a culture of caring.  Healthcare Management Forum. 2013; 26:6-10.