Recent research has suggested that e-cigarette use is linked to bone damage, erectile dysfunction, smoking relapse, depression and heightened stroke risk. None of these associations is supported by good data, and it appears we may have to add a new malady to this list. A new study published March 2 in the American Journal of Preventative Medicine has supposedly linked vaping to an increased risk of prediabetes. The paper has all the hallmarks typical of research in the vaping-more-harmful-than-we-thought genre. And, as usual, the media has done nothing more than uncritically amplify its conclusions.
The researchers analyzed data from 600,046 individuals who reported their experience with vaping and smoking between 2016 and 2018 to the Behavioral Risk Factor Surveillance System (BRFSS), a CDC-sponsored project of telephone surveys that monitor health-related risk behaviors, chronic conditions, and use of preventive services across the US.
The prevalence of prediabetes was nine percent among current e-cigarette users and 13.4 percent among current smokers; the prevalence of prediabetes among sole e-cigarette users (those who had never smoked a cigarette) was 5.9 percent and 10.2 percent for dual users of combustible tobacco and electronic cigarettes. The study's conclusion:
"In this representative sample of U.S. adults, E-cigarette use was associated with greater odds of prediabetes."
As always, there's much more to the story. The study was based on incomplete data and its conclusions were restricted by serious limitations. For now, there's little reason to think vaping ups the risk of prediabetes.
Shaky exposure data
The first problem, and I must sound like a broken record for continuously pointing this out, is that the researchers failed to collect meaningful exposure data:
"Respondents were first asked: Have you ever used an e-cigarette or other electronic 'vaping' product, even just one time, in your entire life? Those who responded no were categorized as never E-cigarette users.
Those who answered yes were categorized as ever E-cigarette users and were then asked: Do you now use e-cigarettes or other electronic 'vaping' products every day, some days, or not at all? Ever E-cigarette users who answered every day or some days were categorized as current E-cigarette users, and those who answered not at all were categorized as former E-cigarette users."
As we noted in response to a recent study that linked vaping to bone damage, answers to these sorts of questions do not provide an adequate assessment of e-cigarette use. Substitute any other behavior for vaping, and you can see why. Classifying someone as an “ever chocolate user” because they reported eating a Snickers bar once in their life tells you absolutely nothing about their dietary habits. Asking them if they eat chocolate “every day or some days” is equally unhelpful. In this case, you need to know with more specificity how much someone vapes and what kind of device and e-liquid they use.
Serious limitations
It gets even worse when we survey the study's limitations:
"Limitations of this study include self-report of tobacco use and lack of medical confirmation of prediabetes and other diet information [my emphasis]. BRFSS is a cross-sectional survey, so a causal relationship between E-cigarette use and prediabetes cannot be inferred. Another potential limitation is recall bias. In addition, the authors do not have information on gaining weight for the former E-cigarette users, which might lead to misinformation bias."
Combined with the poor exposure data described above, these limitations are enough to cripple the study. The authors acknowledged that the correlation they uncovered was based entirely on self-reported data. Moreover, all they found was that a small percentage of participants had prediabetes and vaped during the study period; there was no way to know if vaping caused these individuals to become prediabetic, or even if their vaping habit preceded their prediabetes.
What about a mechanism?
This crucial detail was also buried deep in the paper: “The mechanisms behind the causality between E-cigarette use and prediabetes are not fully understood.” Stripped of its academic language, this sentence really means, “we don't know how vaping causes prediabetes.”
The authors cited a study showing that smokers registered higher blood glucose levels than non-smokers following an oral glucose tolerance test and suggested that nicotine may be the culprit in the present research. But it could just be that smokers have poorer glucose control due to some other variable; research has found that smokers tend to eat fewer fruits and vegetables and exercise less than non-smokers. Conversely, some studies have also shown that smoking correlates with weight loss, which tends to reduce the risk of diabetes.
All that to say, the researchers were correct; we have much to learn about the metabolic effects of vaping. Unfortunately, their study hasn't improved our understanding one bit.
Cameron English
Director of Bioscience
Cameron English is a writer, editor and co-host of the Science Facts and Fallacies Podcast. Before joining ACSH, he was managing editor at the Genetic Literacy Project.
