MethodLogical is now at

MethodLogical is now at

Due to some persistent technical issues we've been having with Blogger, we're now posting at Please update your RSS feeds, etc. For the time being, our new posts will automatically be mirrored here, but you'll have to visit the new site to comment.

Wednesday, November 17, 2010

The Case for Cost-Effectiveness

It appears I’ve engaged in a bit of a bait and switch.  At the end of my last post I promised to address the issue of HIV-related health expenditures dominating global health budgets and the adverse effects of such overfunding.  And that was my intention.  However, as I sat down to write this post, I realized that, in order to make my argument, I had to establish the importance of cost-effectiveness in global health decision-making.  And so, while my next post will tackle the issue of HIV overspending, today I make the case for cost-effectiveness.

Say you’re on a sinking ship.  There are 1,000 passengers and only spots for 500 people on the lifeboats.  You would, of course, be justified in your outrage regarding the lack of safety precautions.  In general, the ratio of passengers to lifeboat spots should be pretty damn close to 1.  However, as HMS Pangea takes on water, you might be better served to determine who should get those spots on the lifeboats.

There are an infinite number of arguments that can be made for who should be put on those lifeboats. We could identify the healthiest people with the longest life expectancy.  Maybe we should save them so our survivors live for the most years.  Maybe we should we put all the kids on first, after all, children are among those least able to care for themselves and deserve succor first.  Maybe we should throw in a few burly men and women to help guide these individuals and ensure that those in the lifeboats make it safely to land.  Maybe the best swimmers should be left out of the boatsthey have the best chance of making it to land on their own.

The arguments go on and on.  Our task today is not to determine who gets on those lifeboats, but to agree that somehow we need to make sure all 500 spots on those boats are filled in a manner that is both fair and rational.

This kid knows a thing or two about cost-effectiveness.
Erroneus! you say.  The urgency of a sinking ship is not comparable to the inexorable fight against poverty and disease.  But much like a sinking ship, health care and health tragedies change lives in a matter of minutes and hours, not months and years.  Waiting for a dose of life-saving antibiotics may be a race against the clock.  The same holds true for those desperately awaiting the next calorie or the man who needs emergent dialysis because his kidneys quit working weeks ago.  And it holds for the laboring pregnant woman with AIDS who needs antiretrovirals to prevent her child from acquiring HIV.  That these tragedies take place far away does not make them any less urgent.

Fine, you say, there is urgency.  But where is the scarcity?  Is it fair to compare efforts to aid the world’s sick to a sinking ship where only half of the passengers can be saved?  No.  It is an unfair comparison.  More than half of the world’s sick will fail to be rescued.

Resources for global health and development don’t have to be limited.  For about the cost of the Iraq War we could halve the global incidence of HIV in under 25 years.  Or perhaps less fashionably (but more cost-effectively), build a network of clinics throughout Africa, roads to connect them to towns and villages, and train doctors and nurses to staff them.

However, the fact of the matter is, there is a limited pool of money in the world and we all make choices about how it should be spent.  If our private boats were sailing near the aforementioned sinking ship, we would surely offer up some space for the stranded passengers.  But would we have donated to a fund to equip the ship with extra lifeboats to begin with?  Would we have elected to pay higher taxes to support the World Health Organization or USAID?  Let’s put it this way, the last time a major presidential candidate ran on a platform promising to raise taxes, he lost 49 of 50 states.

We all want to help, but there are other things we want as well.  Over the next two decades, Americans will spend more than the sum required to halve the HIV incidence on their pets.  The cost of an average wedding in America is more than the median income in dozens of countries.  I'm writing this on a pretty spiffy MacBook.  Resources for global health and development don't have to be limited.  But they are.  And so, the task falls to us to figure out how best to spend that money.

Cost-effectiveness makes us uncomfortable.  It sounds an awful lot like rationing or like valuing lives against each other.  Yet in medicine, we accept this all the time, even if we don't realize that we're doing it.  In transplant medicine—with fewer organs than patients in need—we are willing to offer the transplant to the person less likely to abuse alcohol or die of something else in the immediate future.  On the battlefields of World War I, the French came up with the notion of triage (from the French word for to separate).  In the chaos of war, army medics could not attend to all the injured soldiers so they made three categories: those likely to die regardless of intervention, those likely to survive regardless of intervention, and those that could survive but only with medical care.  It was the last group that got attention first because, with time at a premium, the medics wanted the biggest bang for their buck.  This system is still used in emergency departments throughout the world. 

In the U.S. medical system, where cost-effectiveness is explicitly not used, overuse of technology and resources is rife.  As such, every unnecessary test and procedure increases costs for insurance companies.  To recoup this loss, insurance companies raise premiums.  As premiums rise, families decide they can no longer afford health insurance and the ranks of the uninsured grow.  The cause and effect may be separated by months and miles, but the connection is very real.

Remind you of your local emergency room?

In determining cost-effectiveness, we usually look at an outcome related to years lived.  It is fair to question this outcome, as people's contributions to the world are not necessarily tied to their longevity.  However, we would not necessarily agree to base access to care on one's contributions to society in the first place (or even how to measure those contributions).  Longevity is also a problematic measure as people in some parts of the world have much shorter life expectancies than those in other parts.  Is it less cost-effective to help someone in Somalia rather than in France if she will not live as long after a given intervention?  And does cost-effectiveness diminish as those who are treated and live longer require more costly care in the future?  These are important questions and not easily answered.  In global health, if the money is limited (and it always is), then we have to figure out who to give it to and cost-effectiveness offers us a method.  We may not always agree on all of the variables in the equation, but my hope is that we agree that a calculation—even an imperfect one—is necessary.

The argument for cost-effectiveness does rest on the assumption of fungibility, the property of money that it can be substituted for itself.  By way of example, my grandmother used to give her grandchildren money when she visited.  She’d always tell us to spend it on whatever we wanted, except for alcohol.  As my financier cousin explained, despite being an formidable intellectual, our grandmother didn’t really grasp fungibility.  We could use her $20 gift to pay part of our rent and then take $20 out of our rent money to spend on booze.  Problem solved.  In terms of global health spending, cost-effectiveness is only relevant if funding is fungible.  If a donor (or donor country) will only give money if it is spent on HIV interventions, then it is irrelevant if that money could be more efficiently spent on primary care.  In that case, the cost-effectiveness argument has to be made to the donor, not to the NGO that is spending the money.

While this discussion can feel awfully utilitarian at times, it’s important to remember the gravity of its implications.  Proper regard for cost-effectiveness can help more people live longer and healthier lives.  The converse of this is true, too: disregard for cost-effectiveness can lead to unnecessary death and suffering.  Bioethicist and Obama adviser Ezekiel Emanuel and his colleagues put it eloquently when they wrote, “Because resources devoted to international health aid are inherently limited, seemingly economic considerations about cost-effectiveness actually reflect fundamental ethical principles.  The more cost-effectively resources are used, the more lives can be saved.”

We shop around for the best value on cars, computers, even groceries.  For nearly everyone, money is a finite resource and it is imperative that we spend it wisely.  That this should apply to the health needs of the world’s poorest and most vulnerable is merely a logical—and ethical—extension.


  1. Emanuel says that "seemingly economic considerations about cost-effectiveness actually reflect fundamental ethical principles." He's pointing out the fallacy of thinking those are separate issues, but I'd like to reiterate his point: cost-effectiveness and its close sibling cost-benefit analysis is a proud tradition in normative economics. It rests on a flexible set of ethical assumptions that in some sense define economics. As my undergrad public policy economics professor, Roger Noll, put it, economists are "liberal utilitarian consequentialists", which means we care about outcomes as measured by people's own revealed preferences. It's debatable whether we hold the right position, but many people mistakenly think we don't think about ethics at all.

    I take personal pleasure in the fact that people commonly can't understand the moral theories espoused by actual ethicists, but when I explain my own (let people do what they want and give them the resources so they can actually do it) they get it and tend to agree.

  2. Adam, great article. I would add to this that unfortunately, the cost-effectiveness argument has been used against surgery being included as a part of global health. For many years, global health providers have argued that it is much more cost-effective to offer vaccines and antibiotics than to set up a surgery clinic and start cutting.

    While this is correct, and I am all for C/E to be a part of this equation, I think the fault in this reasoning is that the burden of surgical disease is quite high (people like Atul Gawande have estimated the global burden of surgical disease to be around 5% of all DALYs in the world). Because medical treatments don't touch things like hernias, obstetrical emergencies, and congenital deformities, it is important to consider that surgeons must be a part of the discussion as well - even if the treatment is inherently more expensive.

    While operations like liver transplants have been determined to be in the hundreds of thousands of dollars for just 1 DALY, hernia repairs and other procedures in a pre-existing hospital have been shown to cost on the order of $100-200/DALY.

    Thanks for writing on this and walking the walk.



Note: Only a member of this blog may post a comment.