Interview with Theo Vos

Theo Vos, MD, MSc, PhD, is Professor of Global Health at the Institute for Health Metrics and Evaluation (IHME) at the University of Washington. He is a key member of the research team for the landmark Global Burden of Disease (GBD) study, which is coordinated by IHME. In this role, he is working to improve the GBD methods, update sources of data, and develop partnerships with countries and disease experts to produce GBD estimates that are most relevant to policy decision-making. He is also focused on linking the epidemiological estimates from GBD to information on health expenditure and cost-effectiveness.

Dr. Vos received his PhD in Epidemiology and Health Economics from Erasmus University and his medical degree from Groningen University, both in the Netherlands. He also studied at the London School of Hygiene and Tropical Medicine, where he obtained an MSc in Public Health in Developing Countries.

From: http://www.healthdata.org

Why tropical medicine?

‘I had the idea of becoming a doctor when I was a child. I was already fascinated by the tropics by that time. This was probably because it was very common at home to have guests around from all over the world, including people from low- and middle-income countries (LMIC). So I was used to hearing and thinking about those parts of the world. My own first experience with global health was in a school-building-programme in Kenya. Thereafter, at the end of my medical study, I did a 3-month internship in the same country. After I had finished my internship, my girlfriend (who I can call my wife nowadays) came over and we travelled through Africa for six months. From my side, this was the trick to lure her to Africa; she quickly smelled a rat!

After my surgical and obstetrical training, we left for Lesotho in 1984. I was working as a bush doctor, and she as a bush psychologist. We worked there for four years and then we moved to Zimbabwe where we stayed for another five years. It was a really great experience. I am very grateful for the experiences of working in a government health system in LMICs, because I use those experiences in my current job.

I have experienced that it helps to understand the context of health problems in these settings. Often few data are available in LMICs, which raises problems in estimating burden of disease. For example, I never diagnosed ischemic heart disease (IHD) in Africa, while the statistical models based on scarce data give a higher incidence of the disease in that area. It is possible that things have changed since the 80s, but it illustrates the very different cause pattern of diseases.’

Out of Africa?

‘After many years of clinical work, the question arose what to do next. I didn’t aspire pursuing a long career in Zimbabwe, because I would always feel like a foreigner. Towards the end of that period, I was already more involved in public health work, so the next step was the Master of Public Health course at the London School of Hygiene and Tropical Medicine (LSHTM). During the master programme, I was asked to stay as a faculty member, and one week later I was asked to do a project in Mauritius. I said: “I have no idea what it is, but I’ll do it!”

That was my first introduction to the burden of disease concept – measuring the health of a whole population. At that time, there were no records of methods whatsoever. Yet, I was sent there as an expert. We had to build everything from scratch, which was a very interesting lesson. Mauritius was a great place to do that, because the data were quite good. They had complete records of deaths, causes of death, and hospital admissions – a far better system than in similar countries in the early 90s. Mauritius is a very interesting country as well, having a very mixed island population with people from India, China, the African continent and France. Like many other beautiful tropical islands, a lot of people are obese. As a consequence, diabetes and cardiovascular disease are very prominent. There were remnants of poverty-related diseases, like diarrhoea and pneumonia, but almost no malnutrition and way better birth outcomes than I had experienced before in Southern Africa. The example of Mauritius in the 1990s is playing out in many other LMICs. We call this the epidemiological transition: a drop in fertility, better education and increasing wealth in a country typically lead to a massive reduction in childhood deaths and a shift from communicable to non-communicable diseases. So yes, Mauritius was a very interesting place. And… a perfect place for snorkelling!’

Down-under

‘After 4 years in London and frequent commutes to Mauritius, I moved to Australia, partly to escape the poor quality of life in London, but to do similar work. I worked there as a government bureaucrat in a State Health Department but in the privileged position in which they basically let me do what I thought was important. A lot of ‘burden’ work for Australia, but there I also focused on cost-effectiveness analyses. It was a natural step from mapping health problems to looking at possible solutions, how effective they are and what they cost. Within Australia, I moved to Queensland for an academic position and ran a centre of burden of disease and cost-effectiveness on a larger scale. We helped many Asian and African countries do their burden of disease analyses and economic evaluations.

But this time around, we did it with much better methods of course! Ten years before, we still did much small-scale analyses, with a rather big influence by the researcher picking the one data source that best represents a population. Nowadays, we combine all the information that is known about a specific disease, and with statistical models we analyse the burden of it, including the predictors and the level of uncertainty – a much more sophisticated approach! When I came to the Institute for Health Metrics and Evaluation in Seattle four years ago, there were just over a hundred people working on the global burden of disease. Nowadays, there are over 300 and we cooperate with thousands of experts from all over the world. It is not possible anymore to do this type of work alone, like I did in Mauritius 20 years ago…’

Old hand

Can you give an example of your daily work?

‘At the moment, I am working on the submission of a paper for the Lancet about nonfatal consequences of disease. This article is about incidence, prevalence and severity of all consequences of diseases. This morning I worked on malaria, dementia, air pollution, and zinc deficiency. Because I am one of the old hands here, I work in a very broad area. That is very important, because 25 years ago we asked folks at the WHO with responsibility for a specific disease, “How many people die from the disease you are working on?” Adding up all these estimates, we got 2.5 times the amount of global deaths that would have been possible! There are many reasons for overestimating the numbers of deaths from your “favourite” disease, for example financial reasons, but professional interests may have a strong influence as well. That obviously leads to exaggeration, although not always intentional. At our institute, we claim to be less influenced by bias, since we don’t have a particular stake in any particular disease. We treat all diseases in the same way, and my main role is to be involved in all the different areas that people work on.’

Where to put your money

‘We create visualization tools for our data that are very user-friendly (check them out at http://www.healthdata.org/gbd/data-visualizations). Our main target group consists of policy makers, but also the general public, researchers and funders. The origin of this project lies in the Gates Foundation. Bill Gates invested large amounts of money in healthcare and he wanted to know the impact of his investments. He realized early on that when you only focus on specific problems, for example TB, malaria or HIV, you can get a very distorted picture because you ignore the rest. He funded the major part of our global project and will continue to do so the coming ten years. That is a tremendous endorsement. We built a great reputation as a neutral institute that applies the best possible scientific methods and statistics to produce estimates of the global burden of disease.

Interestingly, this type of work used to be done for a long time by the World Health Organization (WHO). That led to problems, not only because some people had financial interests, but also certain countries would disagree on certain statistics for their country. The WHO culture is to have full consultation with all countries, which makes projects like this very slow and difficult.’

From doctoring patients to doctoring global health numbers

‘This title of my congress* lecture was meant as a joke. Of course, our goal is to improve the care that health workers give to patients. When you look at the bigger problems and trends, you can influence health policy and find the right balance between preventive measures and care for individual medical problems. Also, how are populations changing, and how do we plan healthcare accordingly? For example, the ageing of populations and changes in disease pattern require the flexibility to shift focus to new areas beside those that have been successful in the past.’

Message for the clinician

Do you have a message for tropical doctors in the field?

‘With the visualization tools, you can display the health profile of any country in the world, and it is very interesting to see how this relates to what you see in your clinical practice. Are there diseases you rarely see? For example, mental disorders often present differently in African countries than in western countries. Fact is that wherever we measure, there is a substantial burden of mental health problems. Should you look differently at certain complaints? Could chronic pain be a clue of underlying depression or other mental disorders?

Another message is the importance of data collection. One of the main things that can be of enormous help is the correct registration of death certificates. I remember that when I was doing clinical work, every time I was asked to fill out such a certificate, I felt I had more urgent things to do, with a full ward of living patients. However, good registration of deaths and underlying causes is extremely valuable. The same holds for registration of usual work activities and even information about correct measurements of children’s weight and length – you name it. Take this seriously. It might not have an immediate benefit, but it is an important long-term goal.’

The French paradox

One of the challenges of our work is to compare data from different countries. A famous example is the ‘French paradox’. French people misbehave in many ways. They smoke a lot, drink too much and eat fatty food, yet they have very little ischemic heart disease. But when you look at the data, France marks a lot of deaths as cardiac arrests or atherosclerosis, rather vague diagnoses, most of which most likely represent ischemic heart disease. When you correct for this, half of the paradox disappears. Still, a part of the paradox remains, so there might still be some secret in the petit rouge…

* Theo Vos is keynote speaker at the symposium From Tropical Medicine to Global Health, June 9, 2017, with his presentation ‘From doctoring patients to doctoring global health numbers’.

[Image: Example of data visualisation of Global Health at the Institute for Health Metrics and Evaluation. https://vizhub.healthdata.org/gbd-compare]