From the Head of School...
It's a pretty typical weekend in July for me—a slower start to the morning, a longer workout, and a quieter house that provides time to catch up on email, review news feeds, and check in with Canterbury's social media channels. In the process, I crossed paths on Twitter with Dr. Tracy A. Dennis-Tiwary's article, Taking Away the Phones Won't Solve Our Teenagers' Problems. Instantly intrigued, I read it start-to-finish (and then re-Tweeted it, of course!).
There are three aspects of this article I found most interesting and compelling. The first is that it addresses questions we are all asking as parents and educators: what is the impact of technology, smartphones, and social media on our children and students? How worried should we be? How can we effectively limit and monitor use? From my own discomfort with being separated from my iPhone, to weighing the pros and cons of my sons' love for Fortnite, to being mystified by the amount of time a teenager can spending watching YouTube videos or Netflix episodes, I spend a fair amount of energy wishing for clarity and strategy.
The second is that Dr. Dennis-Tiwary addresses the very real and concerning rise in anxiety we are seeing among our adolescent population. While I am no expert, I believe that one factor in this statistical increase is that more families are more comfortable acknowledging and seeking help for their children's mental health concerns. While this hopeful guess would lead to an increase in diagnoses, it certainly would not account for what has been described as an epidemic of teenage anxiety and suicide. Rather, the article's reference to "uncertain economic lives," "uncertain truths," and "uncertain independence" (boarding school is an excellent avenue toward more certain independence, by the way) builds a credible description of the ingredients fueling this epidemic. Mix in the constant comparing/self-assessment our teenagers endure when immersed in the world of posted pictures and events, and there is no question that they are navigating a world rife with opportunity to question one's strengths, value, and self-worth.
The third is that the article is written within a public health framework, a framework I grew to love during graduate school. For years I taught a junior/senior elective entitled Epidemiology, and the following quotation creates an opportunity for a quick Epi 101 lesson:
"...there simply does not yet exist a prospective longitudinal study showing that, all things being equal, teenagers who use smartphones more often or in certain ways are more likely than their fellows to subsequently develop mental illness."
The holy grail of public health research is the ability to prove a causative relationship. This is harder to do than you might think, and causal links which we now take for granted often started as anecdotes. John Snow, known as the "Father of Epidemiology," linked the devastating illness of cholera to a water pump handle by observing the connection between those who were ill to their use of that particular pump handle. In 1854, he certainly could not prove (nor see!) the bacterium itself that was transmitted by unsanitary water to wreak havoc on the digestive tracts of London residents. Fast forward to the 20th century, and what is now a proven, causative association between cigarettes and lung cancer began with anecdotal "evidence" showing a correlation between frequent smoking and newly-diagnosed cases of lung cancer. For these anecdotes to shift to proven causation, the gold standard of research studies — prospective, longitudinal — follows study participants for years to measure the effect of a particular factor on a particular outcome.
Meanwhile, when as association is observed — for instance, an increase in use of smartphones and an increase in mental illness — Epidemiology teaches us that there are 5 possible explanations for that association (longitudinal studies reduce, if not eliminate, the likelihood of explanations #1-4):
- Chance. What we are observing has taken place purely by coincidence, by accident, by random error.
- Bias. What we are observing can be explained by the impact of some bias; for instance, Recall Bias occurs when those with an illness (e.g., breast cancer) are more likely to remember a specific exposure (e.g., garden pesticides). The Placebo Effect is also an example.
- Confounding. What we are observing can be explained by a third factor. For example, in studies which explore possible connections between coffee and heart disease, researchers must "control" for confounders like caffeine and sugar to ensure that they are isolating the impact solely of coffee.
- Reverse Time Order. What we are observing might be backwards. In other words, does an increase in time spent on smartphones lead to mental illness . . . or does mental illness lead to an increase in time spent on smartphones?
- Cause. What we are observing can be explained by the effect of one factor ("exposure") on the outcome/disease. For example, exposure to UV radiation causes melanoma (though not always, and not for everyone...).
While this concludes the Epi 101 lesson (for now, at least), the central questions of the article remain: how can we understand, explain, and address the rise in mental health concerns among our children and students? How can we untangle the threads of technology use, self-esteem, mental illness, and "normal," healthy development? Research will provide clarity over time, but while we wait, I encourage all of us to see these interwoven issues through the lens of day-to-day public health as one path toward greater understanding and new solutions:
"Yes, we should devote resources to making smartphones less addictive, but we should devote even more resources to addressing the public health crisis of anxiety that is causing teenagers so much suffering and driving them to seek relief in the ultimate escape machines."