As I think about my academic career, my story is similar to many: I never thought I would end up doing what I ended up doing.
I graduated from Stanford with a Bachelor’s degree in psychology. I was also 3 courses short of a double degree in chemistry, but by the time I was supposed to finish up that degree, I was fully invested in a career in social psychology. I then entered the doctoral program in social psychology at the University of Michigan, where I worked with Hazel Markus on how one’s self-perceptions influence the perception of others, and with Richard Nisbett on a series of research studies on the impact of statistical training on thinking about everyday life (yes, people can be trained to think about the world in ways that avoid some of the well-documented errors and biases in reasoning).
My first faculty position was at Northwestern. After three years there, I went to Princeton’s psychology department as a visiting professor, where I began a 15-year collaboration with John and Loretta Jemmott in creating, implementing, and evaluating behavioral interventions to reduce risky sexual behavior among minority youth in West Philadelphia and the State of New Jersey. This collaboration led to landmark studies that were published in AJPH, JAMA, and JAMA Pediatrics (at the time known as Archives of Pediatric and Adolescent Medicine). This collaboration continued after I took a faculty position in the Department of Psychology at the University of Waterloo, and after John Jemmott moved from Princeton to the Annenberg School of Communication at the University of Pennsylvania.
My background in research methodology dovetailed nicely with John Jemmott’s own expertise in methods. We designed the behavioral intervention studies to be as rigorous as possible. Indeed, the studies that we conducted are widely known for their rigorous design, using design principles in experimental social psychology to isolate the specific causal elements of the safer sex intervention. For example, every 4-6 hour intervention that was created by our team to promote safer sex was matched with an identically structured “control” intervention to promote healthy behaviors, which used the same materials, and included activities for the 11-13-year old participants that had the same structure as the safer sex intervention. Thus, the only difference between the safer sex intervention and the control health intervention was the content. Further, all of the adult facilitators that were involved in delivering the intervention were trained on BOTH interventions equally, and on the day of the intervention, they were randomly assigned to deliver the safer sex intervention or the health control intervention. Thus, not only were the youth participants randomized to the interventions, but the facilitators themselves were randomized to deliver those interventions. This double randomization served to equate both the characteristics of the participants and those of the facilitators. And after the intervention, the youth participants rated the intervention and their experiences, including how much liked the intervention, how much they learned from it, how much they would recommend the program to their friends, and how much they liked their facilitator. Over the 5 trials that the team conducted, there were never any differences between the two interventions on any of these measures. We found that the safer sex intervention did indeed lead to significant reductions in risky sex at followup (with over 95% retention) at 3-months, 6-months, 12-months and 24-months after the intervention day. In another trial, involving a clinic for adolescent females, the same result was obtained showing the effectiveness of the 6-hour intervention over a long term, but in that trial with the added validity from serotesting: those in the safer sex intervention were significantly less likely to have contracted a sexually transmitted infection.
In early 1999, I was told by a close colleague in the School of Public Health at Waterloo about the upcoming introduction of graphic health warnings on tobacco packaging in Canada—which would be the first country to have implemented this policy. I looked into the research literature on graphic warnings, and was astonished to find that out of the 40+ studies on health warnings, there was only one study that had ever looked at the impact of this public health intervention at the population level. I also knew that the existing health behavior surveillance systems were inadequate to properly evaluate this policy or any other population-level policy, not only in tobacco control but in any domain.
There was an urgent need to create a rigorous evaluation system since the graphic warnings were being introduced at the end of 1999. So I designed a cohort study to evaluate the impact of the new warnings. It was this study that led to a much larger concept: to create a global evidence system to rigorously evaluate the impact of the global treaty on tobacco control that was being negotiated at the WHO—the Framework Convention on Tobacco Control (FCTC).
I recognized the importance of doing something to fill the evidence gap, and so I dove headfirst into becoming a tobacco researcher.
This was the beginning of the International Tobacco Control Policy Evaluation (ITC) Project, which started off as a set of parallel longitudinal cohort studies of nationally representative samples of smokers in four countries: Canada, United States, United Kingdom, and Australia. Despite the fact that up to this point, I had published only one article on tobacco, I was fortunate to be able to enlist the generous collaboration of world-leading researchers on tobacco use, including Gary Giovino, Mike Cummings, and Andy Hyland at Roswell Park, and Frank Chaloupka at University of Illinois at Chicago, Gerard Hastings and Ann McNeill in the UK, and Ron Borland at the Cancer Council Victoria in Australia—collaborations that have lasted for nearly 20 years. With the guidance and leadership of Mary Thompson, a world-leading statistician of estimating equations and sampling design in surveys (when you are using GEE models, you will never know that some of its mathematical foundation comes from Mary’s theoretical work), we created a longitudinal cohort study that had key design elements and structures that would allow for rigorous evaluation of FCTC policies that we anticipated would be implemented throughout the world. In the formative stages of the ITC Project, my graduate students—David Hammond and Tara Elton-Marshall—worked tirelessly to help design and operationalize the creation of the ITC Project’s Four Country Survey. The first wave of this landmark survey was conducted in October 2002.
From the very beginning, I aspired to bring the rigorous methodology and design of the ITC Project to low- and middle-income countries (LMICs), where the vast majority of the devastation of the global tobacco epidemic will be experienced in the 21st century. Accordingly, over the years, I sought funding from key funders—NIH, the Canadian Institutes of Health Research, and foundations and health ministries in key LMICs. In 2005, the ITC Southeast Asia Survey—parallel nationally representative cohort surveys conducted in Thailand and Malaysia—was launched. Soon after, we formed a partnership with the China CDC’s National Tobacco Control Office to create the ITC China Project, an extensive ITC cohort survey in cities (e.g., Beijing, Shanghai, Guangzhou) and rural areas of China, which to date has been conducted 5 times.
We formed a partnership with Dr. Prakash Gupta and Dr. Mangesh Pednekar to conduct an ITC cohort survey study of 8000 tobacco users (cigarette smokers, bidi smokers, and smokeless tobacco users) and non-users in representative samples of four states in India. Other LMICs joined the ITC Project, including Bangladesh, Brazil, Uruguay, Mexico, and then three Sub-Saharan African countries: Mauritius, Zambia, and Kenya. Other high-income countries also joined, including Ireland (where the ITC Project evaluated Ireland’s landmark comprehensive smoke-free law), France, New Zealand, and the Netherlands. In 2016, the ITC Project formed a partnership with leading EU tobacco control researches in the creation of the ITC Six European Country Survey—a two-wave cohort survey of representative samples of smokers in Germany, Greece, Hungary, Poland, Romania, and Spain—as part of the large EUREST-PLUS Research program, funded by the European Commission, to evaluate and understand the impact of the EU’s Tobacco Products Directive, the first EU-wide regulatory package on tobacco products in 15 years.
One unique feature of the ITC Project is its ability to adapt to changing features of the tobacco control policy landscape. Unlike virtually every surveillance survey, which only rarely adds/subtracts or changes its measures over decades, the ITC Project surveys have constantly evolved from wave-to-wave in response to the policy landscape. For example, over the past 8 years, when plain packaging evolved from an interesting policy concept to its first implementation in Australia in December 2012 to its adoption by a dozen countries today, the ITC Project added new questions to be able to conduct pre-post evaluations of plain packaging in Australia, England, New Zealand, France, Canada, and The Netherlands. We have made similar changes to the ITC surveys in different countries to position ourselves to evaluate other policies/regulations (e.g., bans on smoking in cars with children, flavor bans, including menthol bans in Canada, Tobacco 21, bans on smoking in multi-unit housing).
For over a decade, our flexibility and adaptability has extended to the changing features of the nicotine product landscape. We have been tracking the use of e-cigarettes/vaping products and other alternative nicotine delivery products in all of our ITC countries, including the new heated tobacco products (HTPs), of which the global leading brand, IQOS, has just received from FDA authorization to make limited modified risk claims under the MRTP route. We have increased the number of questions devoted to the use of these products, reasons for use, perceptions of these products, including perceptions of the relative harmfulness of vaping products and HTPs compared to cigarettes.
For nearly 8 years, the ITC Project has also added an important focus on alternative nicotine delivery products, such as e-cigarettes and These alternative nicotine products are the focus of the ITC Project’s current Program Project Grant (P01) from NCI, which is conducting 4 integrated projects designed to examine the impact and interplay of policies on both cigarettes and these alternative nicotine products across four countries—US, Canada, England, and Australia.
The need for industry-independent evidence on the many aspects of these new products—including toxicology, emissions testing, and exposure to toxicants, as measured through biomarker tests—has led our ITC Project team to expand our scope both in the diversity of research methods utilized but also the diversity of tobacco product environments and regulatory and policy approaches taken across a widening set of countries. Most recently, the ITC Project has joined with leading researchers and stakeholders in Japan and the Republic of Korea to establish ongoing ITC cohort surveys in both countries—the epicenter of the introduction of HTPs. These ITC Projects are providing leading-edge evidence on the possible impact of HTPs in other countries, including the US and Canada, where IQOS and other HTPs are just being introduced.
Today, the ITC Project is a consortium of over 150 researchers across 29 countries, covering over half of the world’s population and over two-thirds of the world’s tobacco users. The ITC Project is the world’s largest tobacco research program, and has published over 550 articles and 90 national and policy reports and has presented over 1,200 papers at scientific meetings since 2002.
I’m proud of the diversity of our research team. Across our vast research team, over 10 scientific disciplines are represented, including psychology, sociology, statistics, economics, public health, public policy, toxicology, chemistry, medicine, demography. I and several members of the ITC Project team are founding members of the Scientific Leadership Team that created the US FDA’s Population Assessment of Tobacco and Health (PATH) Project. The Conceptual Model of the PATH Study is the ITC Conceptual Model, since I wrote that section of the proposal. And the ITC Four Country Survey was a key source of survey content for the PATH Survey.
Along the way, we have recognized the need for using our evidence to promote stronger, evidence-based action against the global tobacco epidemic. We have produced over 90 national reports, policy reports, briefings, government submissions, and letters to national leaders (e.g., President of Kenya) across over 20 countries, which present ITC evidence in to these non-research stakeholders.
In addition, we have been at the forefront of supporting governments whose tobacco control policies have been challenged through litigation, including playing a key role in the two major challenges brought by the tobacco industry via trade treaties. I was a central expert for the Australian government in their defense at the World Trade Organization of their plain packaging law, and for the Uruguay government in their defense of larger health warnings at a challenge brought by Philip Morris International via a bilateral investment treaty between Switzerland and Uruguay. In both cases, the governments scored resounding victories against these challenges. In the case of Uruguay, PMI’s claim that Uruguay’s larger warnings had no impact was completely destroyed by the longitudinal data from the ITC Uruguay cohort survey demonstrating that on each of the 6 key indicators of policy impact, the larger warnings led to significantly greater impact.
Most recently, as the COVID-19 pandemic took root globally, I stopped the fieldwork on the ITC Four Country Smoking and Vaping Survey at the beginning, and designed a set of questions on COVID and smoking/vaping, as well as interesting potential explanatory variables to understand the impact of COVID on nicotine product use (as well as some key general protective behaviors). We then restarted the survey in early April, and are now in the process of data cleaning. We are also inserting these questions in 5 other ITC countries this year: The Netherlands, Japan, Korea, New Zealand, and Mexico. The result will be an extensive international dataset on the impact of the pandemic.
Our amazing ITC Project team at the University of Waterloo has made it all possible. We currently have 27 full-time research staff that work on the many aspects of the ITC Project, including the management of every step of the development of a new ITC country—sampling design, research protocol, collaboration with in-country research partners and in many cases, collaborations across multiple government, academic, and advocacy partners within each country, extensive training protocols for the fieldwork (the ITC Project has one of the most complex and intricate research protocols, which allows for considerable variation across countries, depending on the environment and capacity and cultural practices, but always staying within the bounds of the broad ITC Project parameters, allowing for cross-country comparisons), translation of surveys (our ITC Managing Director has a Ph.D. in translation), extensive training for the fieldwork staff, data cleaning, data weighting, and analysis.
To date, we have conducted nearly 160 survey waves across our 29 countries, invoving over 160,000 unique individuals with over 350,000 observations. With over 400 variables on each observation, including derived variables, we now have a global dataset of over 140 million data points. We also have a unique SQL-based survey question database that includes detailed information of nearly tens of thousands of questions from the ITC Project, all linked by an extensive indexing system which allows us at an instant to create “harmony reports”, that list the questions on a particular topic and sub-topic asked in every country, at every wave, and of every user group (e.g, smoker, vaper, dual user for the smoking and vaping surveys) throughout our history.
We have worked extensively to increase the research capacity of our ITC countries, especially in LMICs. I am most proud of our record in research productivity in LMICs: 135 of the 560 ITC articles that we have published to date have reported data from LMICs, and an LMIC researcher has been first author on 77 articles (57%). We have also given extensive workshops in many of our ITC countries on tobacco control research and on the analysis of ITC data for our LMIC colleagues.
My career story is that of so many others in academia, and in other professional fields: I never thought I would do what I ended up doing, but along the way, I have been able to adapt and evolve in my research because of my strong belief in the power of research methods to produce evidence that can be helpful to those who are making decisions that affect our health and well-being. Being open to collaborating with others who have greater expertise in areas outside of my own and to find ways of collaborating with smart people who have different perspectives and views because that collaboration can produce great work is always at the front of my mind. It is always challenging, but always rewarding.