Are e-cigarettes a gateway to smoking?

Sanford Research scientist's findings, plus other Sanford Health research notes

Arielle Selya in her lab

E-cigarettes have skyrocketed in popularity over the past decade, especially among young people. Because vaping is relatively new and there are so many devices on the market, it’s not yet clear whether the practice leads some people to eventually smoke more harmful cigarettes, said Sanford Health researcher Arielle Selya, who recently presented updates from her laboratory at the annual conference of the Society for Research on Nicotine and Tobacco.

Selya, an assistant scientist at Sanford Research in Sioux Falls, South Dakota, and director of its Data Exchange, showed how trends in use of electronic cigarettes impact conventional smoking among adolescents and how advanced statistical methods can advance research on their use. She discusses her work in this Q&A and how the results could help guide policymakers who are debating whether vaping should be as heavily regulated as traditional cigarettes.

Selya’s background:

  • Hometown: Doylestown, Pennsylvania
  • Family: Husband; two boys, ages 5 and 2
  • Hobbies: Hiking, cycling
  • Education: B.A., physics, Cornell University, Ithaca, New York; Ph.D., neuroscience, Rutgers University, Newark, New Jersey; graduate work in system dynamics, Worcester Polytechnic Institute in Massachusetts
  • Work history: Postdoctoral fellow and research associate at Wesleyan University in Middletown, Connecticut; masters of public health assistant professor in the Department of Population Health with the School of Medicine and Health Sciences at the University of North Dakota in Grand Forks
  • Research interests: Adolescent smoking development and nicotine dependence; system dynamics modeling; predictive analytics of health outcomes

What do e-cigarettes include?

It’s a broad term that basically gets away from combustable cigarettes, which are the conventional form. E-cigarettes are anything like vaping or Juuls, anything that electronically heats the elements. And that’s how you get the vapor. There’s many products that fall under the category of e-cigarettes.

The other unique thing is that they can contain flavors. With traditional cigarettes there’s the tobacco flavor and there’s menthol with cigarettes and other flavors are banned. Whereas with e-cigarettes, you can have many flavors, so you can have candy flavors, fruit flavors, you can have menthol and tobacco flavor as well. That’s one of the unique aspects of e-cigarettes.

How did you get interested in e-cigarette research?

My undergrad degree was in physics and my Ph.D. was in neuroscience. I can make sense of this in hindsight now. I didn’t realize consciously what I was doing at the time. But I’m a quantitative thinker and I am really fascinated by using mathematical models to understand the world.

And when I transitioned from physics to neuroscience it was because I really wanted to understand the brain. I was recruited by a faculty member who sent out an email that said … all of the easy problems in neuroscience have already been solved. Where we are now in neuroscience really needs computational talent to understand the relationships between neurons and the network properties of the brain. So that fascinated me because my second love was psychology.

I really felt a calling to use these machine-learning and big data methods (from graduate school) and work in something more public health-related where somebody could benefit from my research a little more directly.

I looked around at different post-doctoral positions and I found one that was absolutely perfect at Wesleyan University in Connecticut, and my mentor studied addiction. … It was a perfect fit for me because my post-doc adviser was also kind of a quantitative thinker and she was great at training me with new methods. And I got to work in a different field and apply some of the mathematical modeling that I picked up before to study addiction.

What are the recent trends?

E-cigarettes hit the market around 2010 and if you look at the percentage of adolescents that are using these things, it rapidly and drastically increased over the next couple of years. Recent estimates are about 20 percent are using e-cigarettes. And this has been accompanied by a decline in use of conventional cigarettes. We’re at an all-time low with conventional cigarettes, which looks like a good thing until you look at the bigger picture. And then it’s not so certain because of the accompanying rise in e-cigarettes.

Are they as addictive?

We don’t know. There’s no blood test you can do to measure nicotine dependence, so you have to ask smokers a set of questions that are shown to measure it. But we don’t know whether these same questions apply for e-cigarettes. You can measure use, but you can’t measure how addicted somebody is. Some people can get addicted before they’re even smoking on a daily basis. And that’s especially true for adolescents. Whereas others can smoke a pack a day and not experience any dependence symptoms, so there’s something else going on inside the brain psychologically that makes some people more prone to addiction.

Can they be used to help people quit traditional cigarettes?

That’s another area of debate, but I think the research is coming down on the side that, yes, they can be an effective quit aid. And again it’s a situation of what’s the alternative? Obviously, it’s better if people quit nicotine and tobacco altogether. But you have heavy smokers that often have psychological issues. People with depression often smoke. It makes it very difficult for them to quit. So if you have somebody that you know is never going to quit, then e-cigarettes is probably a good alternative for that person.

Can a person get the same fix?

If you smoke first, usually you can’t get the same fix. E-cigarettes come close and they’re far better than other nicotine products like nicotine replacement therapy such as the patch and the gum. That just doesn’t give you the same fix. But for kids that use e-cigarettes first and get used to that, they usually don’t have a problem being satisfied with that. But there is a bigger trend toward using multiple tobacco products, whereas back in the day you were either a pure cigarette smoker or chewed tobacco or smoked cigars.

What impact do e-cigarettes have on the body compared with traditional cigarettes?

We really don’t know what the effects are because the data are so recent, so we don’t have the long-term health outcomes. Almost everybody agrees that e-cigarettes are not completely safe and almost everybody agrees that conventional cigarettes are probably more dangerous.

This is where my research really kind of advances what’s already known, what really is the effect. When we look at the national survey trends, if you look at kids that started with e-cigarettes and then track their later substance use, many, many of them go on to smoke conventional cigarettes. So there’s a concern out there that e-cigarettes are attracting adolescents to tobacco products in general, and the specific concern is that they will become conventional smokers one day, which is more dangerous for their long-term health.

My research kind of goes against the mainstream public health community. I don’t think it’s a pure gateway effect in that way. If you go back to the fact that conventional cigarette use is decreasing while e-cigarette use is increasing, that sort of suggests that some of the kids out there are initiating with e-cigarettes instead of traditional cigarettes. So if e-cigarettes weren’t on the market, they probably would have gone on to smoke conventional cigarettes anyway.

There’s a certain percentage of kids, maybe 10 to 20 percent, that are going to use some tobacco product, no matter what. So with e-cigarettes on the market, they’re diverting their use to those instead of conventional cigarettes. And given that e-cigarettes are relatively safer, even though they’re not completely safe, it probably is a better scenario for somebody to use e-cigarettes throughout their life compared to conventional cigarettes.

How can math advance research on e-cigarette use?

One of my in-progress developments, and this is the one I presented at the nicotine and tobacco conference, is a method called propensity scores. When we look at the conventional cigarette rate decreasing, we can’t say for sure what those people would have done if there were not e-cigarettes on the market. Propensity scores help to get at that question.

The way they do that is by choosing comparison groups for the analysis by looking at all the risk factors. We know what the risk factors for smoking are. If you have a parent that smokes, you’re more likely to smoke. If you have a group of friends that smoke, you’re more likely to smoke. If you have mental health, depression or anxiety issues, you’re more likely to smoke. So there’s a whole host of risk factors that are risk factors not just for conventional cigarette use but e-cigarette use as well. There are some differences, but by and large, the kids who use cigarettes look like the kids who use e-cigarettes on all of these respects.

In a traditional analysis, you compare the kids that are smoking, who are depressed and anxious and have smokers all around them with the kids that don’t smoke but are better off in terms of mental health, are better off in terms of not being exposed to cigarettes from their family and friend group. Traditional analysis can account for this to come extent, but the two groups look so different on all of these risk factors that it’s really not a valid comparison, and this can lead to answers that are not entirely accurate.

What this advanced method, propensity scores, does is it selects a more appropriate comparison group. I’m taking the kids that look the same on all these risk factors. And once I have a more appropriate comparison group, I’m looking at the kids who use e-cigarettes vs. don’t use e-cigarettes then looking later at their conventional smoking to see if e-cigarettes introduce an additional risk for smoking, above and beyond all of the pre-existing risk factors. I’m evaluating this gateway idea, about whether e-cigarettes are introducing an additional risk. My own research suggests that there’s a minimal or no gateway effect of e-cigarettes. But the verdict’s not in yet because I need to follow up with better data.

What happens next with your research?

I have a couple of things lined up next. I’m going to replicate the propensity scores analysis with better data that does have longer-term followup, so then I’ll really be able to isolate kids that tried e-cigarette use first and then progressed or didn’t progress to conventional cigarettes. I’ll be able to account for many of those pre-existing risk factors and we’ll see if there’s an additional effect of e-cigarettes or if it’s purely due to things like personality and background and peer group.

And then I have another project lined up this summer with a promising intern coming in through Sanford Health’s SPUR program to look at the trends across different types of products and get at the question of are there more tobacco users now or is it this steady level of tobacco use. Is there a certain percentage of the population that’s always going to use tobacco? Or is there a definite increase since the appearance of e-cigarettes?

What are the possible benefits from your research?

No. 1 is policy implications. There’s a lot of debate right now at the various state levels and at the national level whether we should regulate e-cigarettes as heavily as we regulate conventional tobacco products. My research is critical to this question because if e-cigarettes are a harmful gateway, then, yes, you should regulate e-cigarettes. And this would include things like banning advertising directly to kids, banning flavors, introducing age purchasing restrictions, same as cigarettes.

But on the other hand, if e-cigarettes do not have a gateway effect and they actually represent a diversion from cigarettes and therefore are important for harm reduction in the long-term, then you wouldn’t want to regulate them.

There’s actually a very important study, not mine, where the author looked at states that implemented e-cigarette legislation vs. states that did not. And they found that it actually backfired. Yes, it cut down on e-cigarette use, but it diverted kids back to conventional cigarettes.

Research notes

Selya’s presentation was one of numerous recent developments involving Sanford Research, one of the key areas of innovation at Sanford Health. Some other highlights:

Clinical research

Pediatric endocrinologists Kurt Griffin and Luis Casas will respectively oversee Sioux Falls and Fargo involvement in the TrialNet (TN01) Pathway to Prevention study. The overall objective of this study is to perform baseline and repeat assessments over time of metabolic and immunologic status of individuals at risk for type 1 diabetes (T1D) to: (1) characterize their risk for developing T1D and identify subjects eligible for prevention trials, (2) describe the pathogenic evolution of T1D, and (3) increase the understanding of the pathogenic factors involved in the development of T1D.

Clinical investigator and genetic counselor Quinn Stein authored an analysis of recent hiring trends in the genetic counseling profession that was published in the Journal of Genetic Counseling.

Sanford Health recently enrolled its first patient on the Multicenter Trial of Stem Cell Therapy for Osteoarthritis (MILES) study that seeks to identify the most effective treatment for knee osteoarthritis, comparing corticosteroid injections with three different stem cell options. The study involves 480 participants at four institutions: Sanford Health, Emory University, Duke University and the Andrews Research Education Foundation.

Lora Black recently was invited to speak about Sanford Health’s integration of clinical care and clinical research at the Clinical Research as a Care Option (CRAACO) conference in Raleigh-Durham, North Carolina. She told how the health system grew the number of its patients taking part in clinical trials from 3% percent to 10% in eight years.

Clinical investigator and gastroenterologist Muslim Atiq authored an epidemiologic review of pancreatic cancer in South Dakota in the Journal of Gastrointestinal Cancer. Pancreatic  cancer cases in South Dakota from 2005-2014 were found to occur at a similar incidence compared to national trends with age at presentation, tumor location and biological behavior of the tumor serving as predictors of patient survival.

Clinical investigators Lindsay Hines, Russell Wilke and Eric Larson recently published a study in Pharmacogenomics demonstrating that physicians changed the prescription pattern (drug and/or dose) and monitoring strategy for some patients. The change is made after responding to alerts in the electronic medical record for patients that poorly metabolized selective serotonin reuptake inhibitors for long QT syndrome due to the CYP2C19 genotype.

Pediatrics and rare diseases

Kameswaran Surendran received a grant supported by the Sanford Health Foundation for new research on the genetic basis of kidney disease in children with Alagille Syndrome (ALGS). An Alagille group came to Sanford Health last year as part of the rare disease event, which prompted the grant.

Jill Weimer received funding from the ForeBatten Foundation to determine the therapeutic efficacy of small molecule compounds in the treatment for CLN3 – Batten disease.


Alexei Savinov co-authored a study published in Frontiers in Immunology that could help researchers’ understanding of the onset of type 1 diabetes. He’s investigating how reactive oxygen species (ROS) synthesis in dendritic cells influences pathogenesis of the disease.

Behavioral Sciences

Stephen Wonderlich was awarded a subcontract from the North Dakota Council on Abused Women’s Services to expand the clinician base and increase utilization of services provided within the Treatment Collaborative for Traumatized Youth program.

DenYelle Kenyon showed that an apprenticeship program merged with an indigenous approach provides an effective framework for cultivating undergraduate Native American scientists. Emily Griese co-authored this work published in Cultural Studies of Science Education.

Susan Hoover co-authored a retrospective study published in Rheumatology International showing that anti-rheumatic therapy is safe for asymptomatic Coccidioidomycosis patients and that routine anti-fungal treatment may not always be warranted. Some rheumatology clinics perform blood tests that identify asymptomatic patients with the infection, but the management of such patients lacking clinical symptoms is not clear.

Paola Vermeer published a comprehensive review in Cancer Research on how tumor recruitment of nerves influences cancer progression and provides therapeutic opportunities. The new research focuses on how neurotrophins, guidance molecules and exosomes (small membrane-bound vesicles) work independently and in combination to regulate tumor innervation.

Paola Vermeer recently published in Gynecologic Oncology results that may help cancer patients with densely innervated tumors who suffer with increased metastasis and decreased survival compared to those with less innervated tumors. She showed that innervation and neurite outgrowth in human-papillomavirus (HPV)-positive cervical cancer is promoted by tumor-derived exosomes. 

Environmental influences on health and disease

Peter Vitiello presented on biomedical research infrastructure and community partnerships to the Augustana University Academy for Seniors in Sioux Falls. The academy invites speakers to contribute insights on topics to lifelong learners.

Genetics and genomics

Randy Faustino gave an invited talk at Winona State University in Winona, Minnesota, on his research program investigating how network biology and the application of graph theory principles is used to identify novel biomarkers of cardiac development and disease.

Cellular therapies and stem cell biology

University of South Dakota M.D.-Ph.D. candidate Elle Anderson and postdoctoral fellow Jordan Sheets, both members of the Francis laboratory, recently received travel awards from the Center for Brain and Behavior Research at USD to present their research at upcoming scientific conferences. Anderson will present her work on cholesterol and endocytic trafficking at the Kern Lipid Conference in August, while Sheets will present his work on astrogliosis in Batten disease at the Society for Neuroscience annual meeting in October.

Enabling technologies

Indra Chandrasekar was recently invited to participate in the American Society for Biochemistry and Molecular Biology undergraduate networking event as a mentor to provide career advice during the Experimental Biology annual meeting in Orlando, Florida.

Peter Vitiello contributed to this story.

Posted In Faces of Sanford Health, Innovations, Research

Leave A Reply