John Murray Last was born in 1926 in Australia, and grew up in Adelaide. He studied medicine at the University of Adelaide, graduating in 1949. After 5 years of hospital training, 5 years in general practice, and international voyages as a ship's surgeon, he trained in epidemiology and public health at the University of Sydney and as a visiting fellow in the Medical Research Council Social Medicine Research Unit in London. After faculty appointments in Sydney, Vermont and Edinburgh, he joined the Department of Epidemiology and Community Medicine at the University of Ottawa in 1969, which he chaired between 1970 and 1978 and where he has been professor emeritus since 1992. He is known especially for his leadership in the Dictionary of Epidemiology, now in its fifth edition and translated into 14 languages, and of the International Epidemiological Association's initiative to develop guidelines on ethical conduct of epidemiologic research, practice, and teaching. Currently, his principal interests are studies of the interactions of ecosystem health with human health, and ethical problems arising in public health sciences and practice.
JL: Few people start with the intent of becoming an epidemiologist. What did you want to be when you grew up?
JML: When I was about 14, I wanted to be a doctor because I thought it would be a useful way to gather material to write novels. However, the dialogue and character development that make creative fiction possible were beyond me. I did a conventional medical course at the University of Adelaide without serious blemishes, and graduated in 1949. I had a year of rotating internship, then 6 months in country general practice, which revealed that I needed more training. I spent the next 3 years doing hospital jobs and medical residencies in London—where I had arrived in July 1951, in time to be included in Richard Doll's cohort study of British doctors' smoking habits.
JL: Were you in London during the terrible fogs?
JML: Indeed. I was interviewed for a residency on the day that the Great Smog started (5 December 1952). It lasted 5 or 6 days. I remember walking along Piccadilly and being unable to see my feet. You could smell the sulfur dioxide, a contaminant from the coal fires. People with pre-existing chronic lung diseases had excess mortality: 7000–14,000 deaths.
JL: Did you see people with chest symptoms?
JML: The chest diseases I saw most were chronic obstructive pulmonary disease and lung cancer, which was beginning to approach epidemic proportions. At the Whittington Hospital, I had charge of a male ward with 30 beds; sometimes 24–25 of those beds were occupied by men with lung cancer.
JL: For the next step of your path into epidemiology, you went back to Australia?
JML: I went back to Adelaide in 1954. In those days we traveled by sea, and I hitched a ride as a ship's doctor on a freighter carrying 12 passengers and a crew of 80 or 90. Doctors got free passage, and it was a popular way for young doctors to travel. In medical school, I had become quite good at extracting teeth because the dentist who supervised us had a beautiful assistant. When at sea, the chief engineer broke a tooth. He was in agony and facing 4 weeks without a port, and my dental skills were put to good use. But most of my experiences were treating venereal disease, and the bruises, muscle strain, and lacerations that commonly occur working on a ship.
After getting back to Adelaide, I went into general practice. Several interesting things happened. First and best of all, I got married. Second, I became interested in the different ways in which people reacted to illness, which seemed to relate to cultural background. The practice served a mixed population of “old” and New Australians, the latter being displaced persons from Eastern and Northern Europe, and voluntary immigrants from Southern Europe. It was striking that some seriously ill people carried on working, whereas others who were not very ill required long periods of sick leave. Fundamental questions about perceptions of health and sickness began to interest me. I began to keep records that enabled me to count and classify my patients – primitive descriptive epidemiology. The third important experience was a very serious epidemic of Asian influenza in 1958. The deaths of 2 people my own age whom I knew well deeply affected me. Then I got sick myself, developing a life-threatening pneumonia.
During a long convalescence, I realized that I needed to learn more about how to stop people getting sick rather than waiting until they got sick. This epiphany led me to leave the practice where I'd been financially secure, and launch myself and my family into a precarious future in public health research. We went back to England, and I had a year with Jerry Morris in the MRC Social Medicine Research Unit, which was a mind-blowing experience. I met many wonderful and interesting people in that year, all the stars of British social medicine and medical sociology.
JL: What do you think are the biggest changes between epidemiology then and epidemiology as you see it now?
JML: Epidemiology has become more rigorous. It has evolved from simple descriptive studies, mainly directed toward investigating epidemics. Rigorous case-control and cohort studies and randomized trials blossomed in the 1950s. A consequence has been that sometimes statistical methods are emphasized more than the possible benefit for people, with statistically significant associations sometimes found which don't make clinical sense. Second, because epidemiologic findings were getting increased publicity, the field began to attract scientists from diverse backgrounds. Before the 1950s, epidemiology was primarily a medical specialty.
JL: Is epidemiology a set of tools or has it a theoretical framework?
JML: Obviously epidemiology has a theoretical framework. I think epidemiology is early in its evolution. It had a paradigm shift in the 1950s with the blossoming of rigorous study designs, and now I think we need another. It has begun with molecular and genetic epidemiology and a marriage of epidemiology with other domains in ways that transcend disciplinary boundaries. Epidemiology has always seemed to me to transcend disciplinary boundaries to a greater extent than other biomedical sciences.1 That is one of its fascinations.
JL: What person would you single out as having most strongly influenced your career?
JML: Jerry Morris had the most profound effect on my life and career.2 I read his paper on the “Uses of Epidemiology”3 and his book with the same title4 when my ideas about classifying and counting patients in my practice were beginning to crystallize. I learned a lot from the leaders in public health sciences and epidemiology in the UK whom I met through Jerry, but Jerry himself had the greatest impact. I value enormously having learned from him how to make the maximum use of available data, without having to do expensive research studies with a large staff and costly equipment.
JL: Collaboration is essential to successful epidemiologic studies. What have you found to be the most important ingredients of a good collaboration? With whom have you had your best collaborations?
JML: I've always gotten along with people by regarding them as equals. I think that's why I've gotten along well with students. My collaborative work on the Dictionary of Epidemiology 5 was the most enjoyable work I've ever done. This started in 1978 when Anita Bahn at the University of Pennsylvania got funds from the National Library of Medicine to compile a glossary of epidemiologic terms. I worked with her, and then, after her death, the International Epidemiological Association (IEA) took over and invited me to take the initiative forward. We posted notices in the IEA Newsletter, The Lancet, New England Journal of Medicine, and other journals. Many people offered to help. Michel Thuriaux wrote wonderful witty and thoughtful letters laced with quotations from the lesser-known works of Lewis Carroll6 and became a lifelong friend. Among the contributors to the first edition were epidemiologists from Syria and Israel, countries then at war. A passion for precise terminology trumped international hostility. Mervyn Susser and Sander Greenland had a polite but rather acerbic exchange of views through me about the best way to define terms like “rate,” “case-control,” and “cohort studies.” A dozen or more basic definitions gave us trouble, much of which was resolved at a 5-day meeting that Kerr White convened in New York. There we drafted a preliminary definition of “epidemiology” (among other terms) but it took a 3-hour journey in a Volkswagen bus from Calcutta across the plains of Bengal with Zbigniew Brzezińnski to finalize this.7 My most important collaborator has always been my wife, Wendy, a true life-partner.
JL: Whom would you regard as the 2 or 3 most important epidemiologists during your lifetime?
JML: This is hard! I hate small lists when there are so many illustrious men and women who merit mention. If I name 3, they are Jerry Morris, Richard Doll, and Archie Cochrane. All have been personal friends and have had original, creative minds. Jerry Morris demonstrated how to connect seemingly unrelated items of information. Richard Doll led epidemiology into new realms of rigorous science. Archie Cochrane gave birth to clinical epidemiology, merging clinical medicine and epidemiology in a fruitful partnership.
JL: Which do you think has been your most influential paper?
JML: I suppose “The Iceberg: Completing the Clinical Picture in General Practice”8 is the best scientific paper I've ever written. I think it gave a useful model of what the average family doctor sees, and what's submerged below the surface in that same population if you could find the affected people. Some of what's below the surface is serious and important in that, if detected early, it can be treated effectively, but if not detected early will eventually cause serious trouble – and cost much more in medical expenses and shortened lives. Sir Theodore (Robbie) Fox, editor of The Lancet, required just one change: the addition of “The Iceberg” to the title I'd originally given the paper.
JL: Which has been your most neglected paper?
JML: What has disappointed me most has been the indifferent response to papers on the health impact of climate change that I began to think about in the mid-1980s. My first paper on this theme was rejected by several journals, but eventually published in the Canadian Medical Association Journal in 1989.9 It sank without trace. My worrying thoughts about the impending climate crisis were too far ahead of medical and public opinion back then.
Even some of the most extreme scenarios have underestimated the extent of current observations, for example of polar and alpine ice-melt. Even a sea level rise of about a meter (predicted in the Intergovernmental Panel Climate Change report in 200110) would incapacitate facilities in many seaports and cause great population displacement. Perhaps half a billion people could be displaced because the low-lying fertile lands on which they live (eg, in south China, parts of India, the Indonesian Archipelago, parts of Latin America and Florida) could be submerged. Hurricane Katrina displaced about 300,000 people and caused the USA major economic and social problems. Imagine the situation if there was a need to relocate several hundred million people over a rather short period. Moreover, when there are rapid, massive environmental and ecologic changes, especially in our increasingly crowded world, there is a high risk of violent conflict. Turbulent refugee movement increases risks of epidemics. More epidemiologists must get interested in these problems and their solutions. If I had a wake-up call, that's the most important thing I've got to say.
JL: You mentioned the power of using existing sources of information. What do you think about the increased attention to privacy legislation?
JML: In public health, utilitarian ethics—the greatest good for the greatest number—can be more important than the rights and dignity of individuals. That's a perennial dilemma in public health but rarely an issue in clinical medicine. In European countries that endured totalitarian regimes during World War II, such as Holland and Germany, there is an obsession with privacy and confidentiality to the exclusion of other social concerns. That has led to crippling limitations on valuable social instruments such as a cancer registry or even a census.
The intervention of some ethicists in the development of guidelines for research in epidemiology troubles me because of a tendency to treat guidelines as though they were the law. If there are too many constraints, for example on requirements for informed consent in a record-linkage study, this can impede population studies because getting informed consent from very large numbers of people is not feasible.
JL: What you might think of doing in the future may be very different from what you might think of now?
JML: Exactly. So there is an issue in preserving personal medical records, as well as biobank specimens. In some jurisdictions, when a person dies, by law their records are destroyed. That's unfortunate because sometimes years later their records could be a key resource in epidemiologic studies. Destroying records because the person is no longer alive makes no sense.
JL: How have you chosen the research questions you want to address?
JML: Under Jerry Morris's influence, I've always preferred questions that can be answered by using available sources of information in new ways—it's elegant, usually quick and cheap, and can often provide answers as useful as more costly studies. I've done a little bit of a lot of different things—theoretical models such as the “iceberg” of disease; demographic and behavioral analyses of supply and recruitment to health professions; case-control and cohort studies of selection, recruitment, and careers in medicine; meta-analysis of climate and other environmental factors affecting ecosystem and human health; cancer causes; HIV/AIDS, among other things. And, of course, for the past quarter century I have spent a lot of time combing through books, journals, and manuscripts, learning from the context in which they are used about the meanings of words and phrases in epidemiology and public health.
JL: Which of your contributions to the field would you most like to be remembered for?
JML: The work on compiling the Dictionary of Epidemiology,5 a very collaborative effort, has been the most fun, the most useful, and the most likely to outlive me.
JL: And the Dictionary of Public Health 11 followed. What prompted you to work on that?
JML: I thought it would be useful to have something that spanned wider horizons. And there's much less disagreement about terminology in most other areas of public health than in epidemiology.
JL: What are your other passions, apart from epidemiology?
JML: I have mentioned my commitment to ethical issues in epidemiology and public health, which stems from an abiding concern for human rights, social justice, and fairness in all aspects of society. I was a serious collector of antiquarian books in the medical and natural sciences until old books got out of my modest price range. I'm an eclectic reader: I read books on theoretical physics, political and literary biographies, serious modern fiction and the classics, and ancient and modern history.
Listening to music and biking (until a small stroke a few years ago made my balance a bit unsteady) have always occupied a good deal of leisure time. We love to see our children and grandchildren. Reading aloud to our children when they were young and making up occasional stories for them was perhaps the most pleasurable pastime of my life.
JL: In your opinion, what has been epidemiology's most important contribution to society?
JML: Epidemiology has made enormously valuable contributions since the middle of the 19th century, leading to important shifts in societal values, behaviors, and levels of health. Epidemiology connects the dots, the isolated bits of information that begin to form a coherent pattern when put together in the right way. What we learn is passed on to society through concerned citizens, media, and eventually (often lagging far behind) policy-makers, our elected leaders.
I think there are 5 features that are essential to solve any public health problem. First, awareness that a problem exists. Second, understanding of what causes it. Third, having (or developing) the capability to deal with the problem, to solve or to reduce it, based on understanding its causes. Fourth, a sense of values that it's worthwhile to deal with the problem. Finally, the political will to do what's necessary. These 5 ingredients led to the sanitary revolution and have led in my lifetime to better control of tobacco smoking, impaired driving, child abuse, domestic violence, environmental lead poisoning, and various occupational diseases. We urgently need to apply this set of 5 ingredients to control the very dangerous problem of global climate change.
JL: What is your assessment of the current state of health of epidemiology?
JML: The field is thriving. It's ripe for further expansion and advances along several fronts, including large computer-assisted population-based studies, molecular and genetic epidemiology, modeling and forecasting future disease, and devising ways to minimize their impact. A minor problem is that some epidemiologists risk getting bogged down in what the late and much-missed Stony Stallones called mathematical trivia.12
The ripest and richest opportunities are in environmental epidemiology, a complex field with daunting challenges. Here more than elsewhere our efforts are endangered by powerful vested interests, for instance in the chemical, pharmaceutical, and energy industries. Hired “experts” have been known to conceal blatant conflicts of interest while accusing honest scientists of “junk science.” This is a reprise of the actions of tobacco companies in the mid-20th century, and owners of private water companies, mines, factories, and tenement housing in the late 19th century. History shows that the truth eventually overcame denial and obfuscation. For the sake of our children, grandchildren, and those who come after them, let's hope that honest scientists' efforts overcome the comparable forces at work in the early 21st century.
JL: Do you have any predictions about what the future might hold for our field?
JML: My past predictions have more often been wrong than right so I'm cautious. However, I believe it is safe to predict a period of dangerous environmental challenges to human health, indeed to the health of many living things. As Tony McMichael and others have emphasized,13 epidemiologists have an important role in identifying cause-effect relationships in environmental disease and in preparing for the challenges of unsustainable environments and stressed ecosystems. These include catastrophic epidemic diseases due to new and re-emerging pathogens, adverse health impacts of climatic and other environmental change, and violent armed conflicts over shrinking resources of fresh water and food supplies.
JL: What advice would you give to young epidemiologists?
JML: The one piece of advice I always give is: Keep your options open. Don't get locked into a narrow, specialized career path. Keep broad intellectual and cultural horizons throughout your life. In that way you won't be a crashing bore at parties because you don't have just that one specialized thing to talk about. And you'll have a lot more fun.
We thank Jamie and Sylvie Desrochers for making possible the video recording of this interview, and Valery L'Heureux for secretarial assistance.
ABOUT THE INTERVIEWER
JULIAN LITTLE holds the Canada Research Chair in Human Genome Epidemiology, and is Chair of the Department of Epidemiology and Community Medicine in the University of Ottawa—the department John Last established. Little's knowledge continues to benefit hugely from the Dictionaries5,11 while the shortening of the winter sports season in Ottawa is a constant reminder of the effect of global warming.