Life, Not Death, is Focus of New Health Metrics

This maverick researcher wants to replace conventional death statistics with data on how well we live.

By Jeremy Smith|Wednesday, December 11, 2013

IHME Director Christopher Murray drills into the math of life and death in his Seattle office. One key find: Traffic accidents cost us more years of healthy life than lung cancer.

Erika Schultz/The Seattle Times

It’s Monday morning in Seattle, and Christopher Murray and I are glued to his computer screen, watching how the world dies. A drop-down menu offers choices from measles and cirrhosis to eating disorders and self-harm. I choose road traffic injuries and see a map divided into 187 countries, each color-coded by incidence: blue for safest, fire engine red for most deadly. Commuting in Sweden or Singapore? Go ahead, gas up. Traversing Gabon or Angola? Buckle up and pray.

The maps represent the ascendance of a new accounting of life and death led by Murray, a still-boyish-looking salt-and-pepper-haired 50-year-old who is the director of the Institute for Health Metrics and Evaluation. Rather than focus only on how many people were killed by a particular malady, as others measuring death and disease do, IHME emphasizes the number of years of healthy life lost.

Take the traffic maps in front of me. Using data released in December 2012, Murray and his colleagues have calculated that road accidents in 2010 cost humanity 75 million years of healthy life — more than measles, lung cancer, tuberculosis or iron-deficiency anemia. “If someone dies of cancer at age 75, maybe the disease has taken five years of his life,” explains Murray, whose group’s massive study, entitled “Global Burden of Diseases, Injuries, and Risk Factors Study 2010,” shares the tolls by age, sex and country for 291 different health problems and 67 risks. “If he dies of a car crash at age 25, though, that has taken 55 years. And if he survives the car crash but suffers a severe spinal injury and then dies at age 60, that is both 20 years of life lost and 35 intervening years lived with disability.”

All of a sudden, asphalt might be worse for you than cigarettes.

Out of Africa

Murray was all but born to a career in international public health. When he was 10 years old, his family raised money, bought a couple of Land Rovers and moved from Minneapolis to eastern Niger for a year. There, his father, a professor of medicine at the University of Minnesota, and his mother, a microbiologist, took over an empty hospital. “There was just us,” Murray says. “Dad was chief — and only — physician. My mother and 14-year-old sister ran the inpatient nursing services. My 18-year-old brother ran the lab and helped my dad with surgery. I was the pharmacist and run-around boy.”

Niger was, and remains, one of the poorest places on Earth, and the 10-year-old witnessed locals carrying sick boys and girls his own age in from the desert, wrapped up in brightly colored cloth. These children had their ribcages sticking out. Some screamed. Others didn’t have the energy even for that. Then there were the adults.

“I remember a guy who walked out of the hospital,” Murray says. “He was very proud. He didn’t want to show people he was sick. But I was a kid. He dragged me over and showed me where he had vomited blood in the sand.”


Murray (foreground) and demographer Alan Lopez (behind him) challenged the way the U.N. determined mortality rates.

Institute for Health Metrics and Evaluation/University of Washington

The direct medical work made a difference, Murray recognized. It saved lives. But his father was the kind of academic who encouraged his children to be skeptical of conventional wisdom. Year after year, the family returned to Africa. As Murray grew older, he questioned whether saving someone’s life and sending him back into the desert, still vomiting and sick, was the best science could do.

Murray went to Harvard for college. He became a Rhodes scholar and earned a doctorate in international health economics. While writing his dissertation at Oxford in the mid-1980s, he saw residents of some low-income countries — China, Costa Rica, Sri Lanka and the state of Kerala in India — repeatedly trumpeted as healthier than others and, in terms of mortality rate improvement, doing even better than those in many wealthy Western nations. 

For instance, the Rockefeller Foundation published a highly influential report, “Good Health at Low Cost,” promoting China’s patriotic health campaign, Costa Rica’s universal health insurance, Sri Lanka’s land reform movement and Kerala’s rural nurse-midwives as reasons residents of these areas were living better. Hard data was sketchy, however, and largely limited to average life expectancy. Who was really extraordinary, and what specific interventions did the greatest good?

“There was incredible expertise in the world on malaria or tuberculosis or any other given disease or problem, but there was nobody standing back and saying, ‘What is the landscape?’ ” Murray says. Say you believed the data and wanted to help other countries follow the examples set by China, Costa Rica, Sri Lanka and Kerala. Should you focus limited time and money on safer birth conditions or lower hospital copays? On improved nutrition or better drinking water? What if none of the above is as important as the number of years young women spend in school — or, returning to road traffic, how safe it is to cross the street? 

“If you don’t have the big picture,” says Murray, “it’s incredibly easy for groupthink to lead you to focus on a limited number of things and miss what’s really important.”

The ideal life not only lasted the longest, he thought, but also experienced the least illness. “Obviously, being in good health is avoiding dying, but it’s also being able to move around well, being able to see and hear, being able think clearly and not being in pain, not suffering from anxiety and not being depressed,” Murray says. “These things really matter to how you live your life. But if you just focus on death, you miss them.”


Alan Lopez

University of Queensland

Dueling Deaths

When it comes to advocating his ideas, Murray has always been brash. Demographer Alan Lopez, Murray’s close collaborator for three decades, recalls meeting him in 1984 at the World Health Organization in Geneva: “A young man knocked on my door. He said, ‘Is this Alan Lopez?’ I said yes. He said, ‘My name is Chris Murray, and everything you’ve written about mortality in Africa is wrong.’ ”

Murray’s interest and intellect thrilled Lopez, himself a young demographer taking piecemeal mortality information from around the world and trying to estimate what killed whom where. An Australian trained in statistics and epidemiology, Lopez recognized not only the limits of his own mortality measurements, but those of the WHO as a whole. 

Every year, researchers in each program of the organization calculated the annual number of deaths attributable to their specialty: malaria, pneumonia, diarrhea and the like. When Lopez added them all up for the first time, the total was over 50 percent higher than U.N. estimates of overall mortality — akin to saying a family with four children has three girls and three boys. It meant that the leading global authority on life and death was either missing or misdiagnosing tens of millions of sick people a year.

“These were well-meaning advocates doing the calculations, but because there was no central oversight, they were using different methods, different degrees of rigor and databases of different quality,” Lopez remembers. “When they made their estimates, there was no constraining them. No one was saying, ‘Hold on a minute, you’re saying a million deaths from diarrhea. The people down the corridor are saying 5 million deaths from pneumonia. Are any of these the same deaths?’ ”

Such double-counting could be deadly. “The whole purpose of doing the numbers is to correctly describe the comparative importance of different diseases,” Lopez says. Diarrhea and pneumonia were the two leading causes of child deaths in the 1980s. They have completely different treatments. 

“So if you had 9 million deaths between the two of them and you said there were 4 or 5 million each, but in reality there were 3 million from diarrhea and 6 million from pneumonia, you would end up with a lack of pneumonia treatment,” says Lopez. “That potentially could cause many more child deaths.” When he brought his concerns to colleagues, most just disregarded him and continued to promote the numbers that they themselves had produced. 

Murray argued for change.

First, rather than rely on researchers studying one specific ailment or area, decision-makers needed an independent team consistently and comprehensively tallying how bad each disease was, for whom, everywhere. Second, it wasn’t fair to concentrate only on mortality. How many 10-year-olds from eastern Niger ended up as Oxford grad students? How many ended up injured by a car accident? How many didn’t have enough to eat? How many would suffer an obstetric fistula giving birth? How many would cough blood? “At that time, the global public health community was very much focused on child survival,” says Lopez. “He was one of the first people who started to think that mortality and health actually mattered for adults as well.”

Murray and Lopez began corresponding and kept in contact even after Murray returned to the United States for medical school. While still a resident in internal medicine at Brigham and Women’s Hospital in Boston and a junior faculty member at the Harvard School of Public Health, the economist-turned-doctor researched various attempts since the 1960s to create a combined measure of impairment, illness and death. “The original idea was you want a metric that can also be used in economic studies: If you spend X amount of money, this is how much health you’ll get.” In 1991, he and Lopez started working on the global burden of disease.

Change in Death and Disability Risks, 1990 to 2010

A Simple Sum

Murray defined burden with a two-part sum remarkable for its simplicity. The first part was about everything that kills people. Let’s say, as was true in 1990, that men and women in the healthiest places on Earth could expect to live about 80 years on average. Then if you died at any age short of 80, you had “lost” that many years of healthy life — at least compared with the ideal. Demographers called this shortfall potential years of life lost (YLL). 

The second part of the sum concerned nonfatal health problems, weighting each on a scale from 0 (perfect health) to 1 (death) — blindness being worse than deafness, moderate dementia being worse than autism, neck pain being worse than infertility and so on, as rated originally by a panel of experts and now by statistically representative surveys of the general public around the world. 

If you think of blindness as a fifth less healthy than perfect health — 0.2 on the 0-to-1 scale — then living with 10 years of blindness could be thought of as equivalent to dying two years early. In the same manner, living a decade with mild neck pain was roughly equivalent to dying one year early, and living a decade with moderate depression was about the same as losing four years of healthy life. Multiply each condition’s weight by the number of years people suffer it, and you have what Murray called years lived with disability (YLD).

The beauty of his formulation for death and infirmity is that both parts of the sum share the same unit: years of healthy life lost. Add them up, and you have what Murray termed the number of disability-adjusted life years, or DALYs, attributable to any health problem. Expressed as an equation, DALYs = YLL + YLD

In plain English, the number of disability-adjusted life years (DALYs) is the sum of years of life lost because of premature death (YLLs) and equivalent years lived with disability (YLDs). DALYs, aptly enough, rhymes with tallies, and the metric allowed important new comparisons: not only between what kills you and what merely makes you sick, but also between, say, life in Latin America and the Caribbean; between homicide and heart disease victims; and between the relative risks of iron deficiency and alcohol abuse. 

At a glance, given the right background data and analysis, one could say that in 2010 premature deaths were responsible for two-thirds of all DALYs lost and disabilities for one-third; that 25 percent of all DALYs lost worldwide were due to death and disability in children younger than 5; and that sexually transmitted diseases and tuberculosis added together contributed 5.7 percent of the global burden of disease. 

“That,” comments Lopez, “is an extraordinarily beautiful and extraordinarily useful policy tool.”

Murray and Lopez brought the idea to the World Bank and then the World Health Organization, which together published the first preliminary “Global Burden of Disease” in 1993.

Death and Disability by Cause

Recalculating Risk

Young, influential, confident and contrarian, Murray and his growing research unit at Harvard attracted critics. Contention only increased in 1998, when the team shifted headquarters to the WHO. As a U.N. agency, the WHO is inherently political. When member nations didn’t like what was said about their country, they made protests to the executive board. 

Pakistani and Brazilian leaders, for instance, were upset when the 2000 World Health Report, overseen by Murray, estimated their health systems’ performance to be much lower than they thought was the case. In response, an independent scientific committee was established to review
Murray’s results. “There were very diverse opinions, a lot of emotions, a lot of anger,” Lopez recalls.

Likewise, aid organizations and WHO staff working to control individual diseases like pneumonia and malaria didn’t necessarily appreciate having their numbers challenged. “For every disease and every risk factor, there’s a constituency out there,” Murray says. Altering the story an advocacy group tells can create cataclysmic change. 

With such powerful tools, policymakers soon followed “Global Burden’s” techniques. At the vanguard was Julio Frenk, the current dean of the Harvard School of Public Health. From 2000 to 2006, Frenk served as the minister of health in Mexico, and that nation performed the first country-level burden of disease assessment. “It completely changed the perception and sense of priorities in Mexico,” he says. “Before DALYs, we assessed the importance of health problems by the number of deaths. Obviously, there are a lot of diseases that don’t kill people but produce a lot of disability. That’s the case with mental illness.” 

For the first time, not only depression, but osteoarthritis, arthritis and lower back pain were seen as vital state concerns, each nonfatal yet among the 10 biggest DALY-ranked health problems facing adults in Mexico. Also newly emphasized were road traffic accidents for men and breast and cervical cancer for women. These struck young people disproportionately, costing more life years.

Frenk used the analysis to drive national health policy in the Congress of Mexico. “For what was given priority, and in what order, we used the national burden of disease,” he says. Medications for breast cancer and cervical cancer and for arthritis are covered in the new national insurance plan, for instance. Where emergency care after a car accident was once an out-of-pocket expense, now it, too, is part of the insurance. So is treatment for mental illness. “I doubt such comprehensive reforms would have been approved if we hadn’t had that evidence,” Frenk says.

Country-level burden studies for Mauritius, Colombia, Chile, Uruguay, Japan, Australia and India’s Andhra Pradesh followed Mexico’s. Murray found himself invited to the World Economic Forum in Davos, Switzerland, and to social events with U2 rocker-turned-humanitarian Bono. When Bill and Melinda Gates, the world’s richest couple, made global health their family foundation’s top priority, the Microsoft founder invited Murray to dinner at his house. “He was incredibly well-read and very detail-oriented,” Murray says of the initial 1999 encounter. 

One extreme numbers enthusiast had met another. DALYs, Murray was told, would guide much of what the Gates Foundation funded, and how they determined if it worked. “We use global burden estimates to prioritize not only work in the field, but also research and development: Where should we invest?” says Stefano Bertozzi, who leads the foundation’s efforts to prevent, treat and cure HIV. “For the pharmaceutical industry, it’s expected return on investment, measured in dollars. For us, it’s expected return on investment, measured in improvement in global health.” DALYs had changed the debate. 

In 2007, the Bill and Melinda Gates Foundation and the state of Washington offered Murray $125 million to fund the creation of IHME. The first task before him was building the skilled team necessary to complete a new, far more detailed and nuanced version of the “Global Burden of Disease” study.

Emmanuela Gakidou, Murray’s wife and colleague, was crucial to the effort, helping to hire faculty and overseeing a fellowship program where recent college graduates can earn a master’s degree in public health. Today the institute employs 139 people, including 24 faculty, 35 fellows and 14 data analysts. 

Staff members are interdisciplinary and international. The director of data development is a former media executive from Germany. An American mathematician designed the software program to model levels of impairment from each illness or injury. Rafael Lozano, lead author of the original Mexican burden of disease study, leads the global cause of death analysis group. 

Official “Global Burden” results began appearing in the British medical journal The Lancet late last year in an unprecedented triple issue of the weekly publication, the first in its 190-year history. “In total, there are 650 million individual results to be reported,” Editor-in-Chief Richard Horton, not exaggerating, told Twitter followers. If he tried to run them all, he joked, “that would be six years of The Lancet with nothing but the Global Burden of Disease.”

The publication was celebrated with a daylong scientific symposium at the Royal Society in London, open to the public and attended by many of the study’s 486 co-authors from 302 institutions in 50 countries. The audience included Christine Kaseba-Sata, the first lady of Zambia; Jane Halton, secretary of the Department of Health and Ageing in Australia; Mickey Chopra, chief of health for UNICEF; and Mark Dybul, executive director of the Global Fund to Fight AIDS, TB and Malaria. Murray and colleagues discussed a few of the startling findings filling the global burden ocean. 

Around the world, for example, chronic obstructive pulmonary disease claimed roughly twice the number of lives as HIV/AIDS in 2010, but HIV/AIDS was much more fatal to young people, and therefore appears higher in the DALY ranking. Likewise, as risk factors, not eating enough fruit out-rivals illicit drug use. For years, experts have said that most of the world’s major childhood diseases could be eliminated with clean water. What global burden suggests is that while lack of access to water and sanitation is a concern, five times worse for the world is indoor smoke from cookstoves, a major contributor to respiratory illnesses, communicable diseases, cardiovascular problems and cancers. 

The concept of global burden upsets entire categories of thought, not the least of which is the old distinction between “developing” and “developed” nations around the world. For example, high blood pressure is the top risk factor in nations as diverse as Germany, Brazil, Indonesia, Egypt, China and Russia. Conversely, the No. 1 cause of ill health varies widely, even among close neighbors. “In Venezuela, it’s homicide. In Mexico, it’s diabetes,” says Murray. “Tuberculosis is big in Indonesia; in Malaysia and Thailand, it’s ischemic heart disease, same as in Western Europe.” 

“What is exciting about this is that it’s the most up-to-date information,” says Lisa Cohen, executive director of the Washington Global Health Alliance, representing many of the world’s leading global health nonprofit organizations, including the Bill and Melinda Gates Foundation. “Where should we put our program resources?” In some nations, health spending may be as little as $2 or $3 per person. “That is infinitely precious money,” says Cohen. And with guidance from global burden data, government officials can allocate money to save and improve the most lives. 

Simple comparisons suggest tomorrow’s priorities. Back pain costs men more years of healthy life than malnutrition; depression and suicide hurt women more than tuberculosis and cirrhosis; glaucoma was one of the fastest-growing contributors to health loss between 1990 and 2010. While Swiss males outlive even the Japanese, life expectancy in the United States has slipped by comparison with other nations of wealth. 

Depending on the county, men and women may have the same average life expectancy in Indiana and Panama, Nevada and Vietnam, Michigan and Syria. “What we have found is that America has wonderful health care services for those who enjoy them,” says Lopez, today an IHME affiliate professor and the Rowden-White Chair of Global Health and Burden of Disease Measurement at the Melbourne School of Population and Global Health at the University of Melbourne in Australia. “But you have large proportions of the population who don’t have access to those services.”

What countries do with the data is up to political leaders, not IHME. But Murray is committed to making global burden datasets a constantly updated, freely available public resource rather than a once-every-other-decade publication. In January and February of 2012, even as he and his team raced with collaborators to ready the entirety of their research for publication, Murray traveled to Washington, D.C., Brasilia, Dhaka and Beijing to share preliminary findings and demonstrate interactive visualizations. And this year, IHME has put the same tools online for use by anyone in the world. Murray’s energy and impatience to measure life and death just may prove infectious.

Back in Murray’s office, we click again at the computer, and the display shifts to a series of nested rectangles showing the top global causes of DALYs in 2010. The bigger the rectangle, the worse the problem; the darker the rectangle, the more it has changed recently in percentage terms. Especially dark is “exposure to forces of nature.” 

Murray swivels in his chair to face me. “That’s Haiti,” he says.

I ask if numbers can tell the whole story. 

We already fit vast human tragedies into little rectangles, Murray responds. “People walk around with a mental map that’s different for every one of us. A real map has got to be a better guide.” 

Next Page
1 of 4
Comment on this article