Around March, some of us take a kick at the snow mounded on the curb and wonder if spring is finally going to drop by. The sun sets before we go home, and the cold coops us up except for runs to the grocery store. All of this amounts to something known informally as the winter blues, because those wintry days and dead trees can put us in a glum mood. But in the 1980s, research at the National Institutes of Mental Health led to recognition of a form of depression known as seasonal affective disorder (shortened, of course, to SAD). Seasonal affective disorder was categorized under major depression to signify depression with a yearly recurrence, a condition far more debilitating than your average “winter blues.” Mention of SAD in research and books peaked in the 1990s, and today SAD is considered a diagnosable (and insurable) disorder. Treatment ranges from psychotherapy to antidepressants to light therapy — large boxes filled with lightbulbs that look like tanning beds for your face.

However, a recent study questions the existence of seasonal depression entirely. Each year, the Centers for Disease Control conducts a large cross-sectional study of the US population. A group of researchers realized they could use the CDC results independently to investigate how much depression changes by season. The 2006 version of the CDC study included a set of questions typically used to screen for depression. By analyzing the answers gathered from 34,000 adults over the course of the year, the researchers might detect flareups of seasonal affective disorder. They might see wintertime surges in depression. “To be honest, we initially did not question the [SAD] diagnosis,” writes investigator Dr. Steven LoBello, the goal being “to determine the actual extent to which depression changes with the seasons.”

Other patterns might emerge. There could be an overall increase in reported depressive symptoms at northernmost latitudes if low light exposure is what brings about a cycle of the depression. After all, light seems to be key. The development of artificial light therapies for SAD patients has relied on links between increased light and increased mood. For those living with SAD, both season and latitude should vary in light levels and interact with depressive symptoms.

Megan Traffanstedt and Dr. LoBello, in collaboration with Dr. Sheila Mehta, searched the CDC’s survey results for links between high scores on the depression screen and particular seasons or latitudes. The researchers also looked to see if high latitudes combined with the winter season to raise the frequency of depressed answers more than high latitude or winter alone. Hours of sunlight at a given location and date are available from the U.S. Naval Observatory, so the researchers even tested for links between depression scores and hours of sunlight on the day a score was collected. If light is responsible for SAD, then looking at hours of sunlight should be a sensitive way to detect people with SAD among the general population, they thought.

Instead, the CDC survey revealed no evidence for seasonal affective disorder. The researchers were wary of overlooking SAD trends among the huge non-SAD population, so they reanalyzed answers from a subset of people who classified as depressed at the time of the survey. Still no sign of SAD. No seasonal or light-dependent increases appeared in the depression measures. We might wonder if something was wrong with the phone study, but other well-established trends appeared in the survey data, such as higher rates of depression for women and the unemployed. The fluctuation in depression from SAD was either nonexistent or undetectable.

As you might expect, other researchers have also asked how SAD plays out across the seasons and if it is more common in countries at different latitudes. The results are mixed, and these studies seem to face two problems. One problem lies with the questions being asked. Those questions typically used to screen for SAD, while very specific, are also incredibly leading. Imagine you are answering a battery of questions about how your mood, weight, appetite vary over the year, then filling out charts to compare your habits over the the seasons. You are asked to consider at which time of the year you gain or lose weight, sleep or eat less. Using this standard assessment for SAD, you might reconsider whether you have a yearly mood cycle after all. The CDC telephone survey did not ask the usual SAD questions but relied on eight questions commonly used to screen for major depression. These questions ask about topics not covered in standard SAD assessments — hopelessness, lack of pleasure in activities, trouble concentrating — and concern the past two weeks rather than the past fifty-two. It is much easier to recall how you felt in the recent past than to remember how you felt last October. Our memories can smudge and blot over time, especially if we expect a certain pattern.

This introduces another problem with SAD studies: expectations. Even hearing about the “winter blues” in popular culture could plant confirmation biases, encouraging potential patients (or researchers) to find evidence of SAD whether it exists or not. Our preconceptions are always hard at work, so survey data collected without any mention of seasonal mood disorders is a safer bet for avoiding biases. A person experiencing major depression episodes in the winter might need therapy for SAD or they might need therapy to cope with holiday-time stresses.

Other studies, particularly in Norway, have also called into question how seasonal depression is measured. If changes in sunlight or other qualities of winter can provoke seasonal depression, then why doesn’t the Norwegian winter, with severely shortened daylight, report higher rates? Perhaps Norwegian culture helps ward off any negative effects of winter, or perhaps SAD is not what we think it is.

A massive telephone survey of the US regarding depression is a valuable opportunity to track SAD in the population, but lack of evidence for SAD does not prove SAD isn’t real. We do know that light orchestrates effects on our health. Melatonin and other hormones certainly respond to light, and there is a part of the brain which takes inputs from the eyes not to see but to keep the circadian beat.  It is possible that seasonal depression is exceedingly rare and hard to detect at the population level — some evidence for that comes from a 1998 US survey of 8000 people. It is also possible that cases of SAD are mood disorders which do not share many symptoms with depression. Understandably, people who believe they have SAD remain doubtful, especially those who see improvement after light therapy. If light boxes or light visors can alleviate mood disturbances with no serious side effects, there will continue to be investigations into light therapies as well as other SAD-specific treatments. For now, though, we can’t quite reconcile seasonal affective disorder with the stability of the national mood.