Is it helpful to talk about the Christmas Blues when research strongly suggests that depression rates are lower and most people feel better at this time of year?
Recent research suggests the pundits who write compelling accounts about seasonal sadness that appear in the media around this time every year are getting it all wrong.
This article started off innocently enough. It is December, and like many psychologically minded blog authors who are looking for a topic in line with the “content calendar”, I began dutifully gathering my facts about the “Christmas Blues”. Just as carols and advent candles, or stockings and santa claus are traditions of families who celebrate this occasion, it seems that articles on stress and symptoms of depression have become a tradition of therapists at this time of year throughout North America.
As with most of my articles, I began by reviewing what other writers have said about the darker side of the holiday season. My first impression was that most of the newer material merely rehashed information gleaned from older sources, and these in turn did the same. Alternately, many sites rely on archived articles rather than going to the bother of penning fresh pieces. In virtually all of the cases, the evidence supporting the Christmas Blues phenomenon is quickly summarized so that the reader can get to the meat and potatoes part about coping. The popular online version of the magazine Psychology Today has an only slightly stale article by Ray Williams titled “How to Deal with Christmas” (1) that offers a fairly standard opening salvo:
According to the National Institute of Health, Christmas is the time of year that many people experience sustained “blues” and even depression. Hospitals and police forces report the high incidences of suicide and attempted suicide. Psychiatrists, psychologists and other mental health professionals report a significant increase in patients complaining about depression. One North American survey reported that 45% of respondents dreaded the festive season.
This is all pretty typical of the claims that are thrown about by most writers who have chosen to illuminate their readers on this problem. It all sounds pretty straightforward, not to mention alarming, until you actually check the facts. Take for example the reference to the National Institute of Health, which most people would assume is pretty airtight. A quick perusal of NIH information (2) on the topic reveals that they are just one more link in the long chain of agencies repeating seasonal truisms, but failing to backup their claims. What do we actually know about hospital and police forces reporting high incidences of suicide, or the skyrocketing increase in complaints about depression from patients? Follow me, and I will take you on a quick journey through the data. However, just as was the case when Fanny Bright sat down beside her beau for the jingle bells sleigh ride behind a lean and lank horse, “misfortune seemed his lot, we got into a drifted bank, and then we got upsot”
To begin our sleigh ride in search of facts, I began with the most prevalent studies on this topic – those that looked at hospital records of suicide and attempted suicide. It is a very sad truth that suicide, suicide attempts, and suicidal ideation (thinking about suicide, but not acting on those thoughts), are very common symptoms of severe major depressive disorder. Because it is relatively easy to obtain hospital or clinic records of these events, and because the act of killing oneself or attempting to can be documented fairly reliably, researchers really like to use suicide as a marker of illness prevalence. Unlike many other measures of depression, such as a simple questionnaire, the act of committing suicide or attempting suicide does not rely on subjective opinion (i.e., am I a little bit depressed, somewhat depressed, very depressed or really, really depressed?). Furthermore, at least in the case of actual suicides, there is no need to distinguish between mild, moderate and severe. It is a very indisputable and tragic fact that dead is dead, period.
Like the snowdrift in jingle bells, the results of this initial foray into the data brought me to a sudden standstill. Sure, I could find lots of data on suicide rates, and even information looking at its temporal sequence and calendar trends, but nowhere could I find any data suggesting that suicide rates peak during the Christmas season. Nada, Nothing. In fact, not only is there no evidence of a peak in the incidence of suicide either on or before Christmas, but it appears that suicide rates actually decline during this time of the year – significantly. A large UK study carried out in the city of Edinburgh looking at almost 20 years of records and more than 20,000 deaths cited a 20% decrease for the month of December (3). A US study looking at 6 years of data for close to 200,000 deaths by suicide noted up to 15% fewer suicides before Christmas (4). In another large study of more than 30,000 deaths, the overall reduction was 6% compared to the yearly average, and as much as 30% less on Christmas eve (5). For all you fact checkers out there, the truth is that people commit suicide a lot less in December, and the trend strengthens as you approach Christmas day (6). Then, as though the population has been holding its emotional breath in anticipation of the big event, suicide rates show a slight increase after the holiday period, in early January.
In almost all studies, suicide rates peak in the spring, a completely different season about as far removed from Christmas as you could imagine.
Okay, so perhaps suicide isn’t the way to go on this. Though it is a startlingly common outcome for people who are severely depressed, most people who have mood disorders do not kill themselves, and many do not even think about it. Furthermore, lots of people who get to this stage of desperation have problems other than depression, like terrible anxiety, or psychosis, or substance use problems. Perhaps suicide is simply not a good way to measure how prevalent the Christmas Blues are.
The next most common way to measure the prevalence of depression in a population is to look at its treatment, and the most common treatment in North America for clinical depression is antidepressant medication. If depression is more common during the Christmas season, we should see a corresponding rise in prescriptions for this sort of medication at this time of year. Like suicide, getting a prescription is a behavior that is quite easy to record, and it does not rely on someone’s subjective judgement when they are looking over the data. Either a person fills a prescription or they don’t. And once again, this is a behavior that assumes a certain level of severity for the condition, though in the case of choosing to take a medication we are talking about moderate to severe levels of sadness, as opposed to the extreme levels of hopelessness that can drive a person to the desperate act of suicide.
A pretty fascinating example of a study that uses antidepressant medication as a measure of mood across larger groups of people can be found in a time-series analysis conducted by Hartig et al just a few years ago (7). They looked at the population of Sweden over the course of more than 12 years starting in 1993, logging a month by month account of the dispensation of antidepressant medication by pharmacists. As an aside, this study was uniquely suited to this type of investigation in that Sweden has a pretty sophisticated public health system that at the time provided everyone in the country with free antidepressant medication supplied from just one drug manufacturer, Apotek. This meant that most people in Sweden used this medication to treat their mood symptoms because otherwise they would have to pay out of pocket.
Hartig et al used a type of trend analysis for their research, which not only looked at antidepressant dispensation month to month, but also things like the number of men and women not working due to vacation. Surely if depression were more common due to the Christmas season, the Swedish sample would be able to demonstrate how strong the effect was. In fact, like the evidence concerning suicide rates, their findings were the exact opposite of what you might expect if the Christmas Blues was a significant problem in society:
Results of the analyses showed that a decline in antidepressant dispensing was associated with each additional vacationing worker, and that the decline increased as the number of vacationing workers increased. Year after year a large decline in medication dispensing was evident for a roughly two week period in July, when most Swedes are on holiday, and a lesser but still notable decline was evident for other periods where more people were off work.
Once again, we appear to be striking out on on the idea that Christmas time brings with it more depression. Not only that, suicide and the prescription of antidepressants are both related to many other troubling symptoms, like those that go with anxiety problems. Based on the available evidence, it’s beginning to appear as though people not only don’t feel unusually depressed over the holidays, but depressed and anxious people may actually feel better than usual.
But we should not give up so easily. There is still the more hard nosed approach of looking at the chemistry underlying negative emotions. Perhaps if we looked for signs that the chemistry behind depression (and anxiety) was more active during the Christmas season we would have the evidence we need for our hypothesis. Coming from this physiological perspective, Molendijk et al (8) looked at a compound called Brain Derived Neurotrophic Factor (BDNF), which is known to interact with serotonin (the feel good chemical associated with the positive effect of the most common antidepressant medications) and be closely tied to neuronal functioning and plasticity – the brain’s ability to work efficiently and adapt to change. BDNF is thought to be one of the mechanisms by which antidepressants work their magic, hence one would expect quantities available in the brain to vary with mood states, and possibly with the seasons. In fact, this proved to be true. In Molendijk et al’s large scale study involving close to 3,000 people, serum BDNF concentrations were lower from January to May and higher from June to December. But according to these results, the highest point for BDNF concentrations is in September, and the lowest is in March.
If low BDNF levels predict depression, then Molendijk’s results suggest that depression should be more prevalent at the start of spring, not the beginning of winter, when BDNF levels (and presumably serotonin levels) are at their lowest.
By now we are completely mired in a snowdrift as we hunt for our quarry, and there are only a couple of ways left to dig ourselves out. The next logical thing to consider is the literature on seasonal variations in mood. Most people have heard about the Winter Blues, and lots of research has looked at the idea that depression could be related to the amount of sunshine we receive. According to this hypothesis, it makes sense that Christmas, which coincides with the longest night of the year, would be associated with higher rates of depression. There are plenty of studies from the 80’s and 90’s looking at this question. Most were conducted on small samples and lacked good control groups. Nevertheless, some trends seemed to support the Christmas Blues theory and the even larger idea that there was some sort of “seasonal affective disorder” at work here. Yet in the past several years it turns out that most of the positive results supporting this idea could not be replicated. In fact, the case for seasonal mood swings has lost much of its momentum, and Seasonal Affective Disorder has been removed from the latest version of the diagnostic manual, DSM5, to be replaced by a seasonal qualifier when someone is diagnosed with a mood disorder.
We are left with one final, obvious way to test our hypothesis about Christmas causing the blues – we can simply ask people how they are feeling. While you may be wondering why I’ve waited so long to get to what may seem to be the best way to test our assumptions, may I remind you that in the field of psychology it is the general consensus that asking people things like this is the weakest form of evidence. People are subjective, interviewers are biased, and there is often no reliable way to introduce a control group. Still, this is the method that has the most “face validity”, meaning that it just seems obvious that if someone says something then it must be true. Using data from an annual phone survey of health, M.K. Traffanstedt and her colleagues took responses from more than 34,000 people on the PHQ-8, a common screen for signs of clinical depression, and compared them to people’s geographic location, latitude, the day of the year, and the amount of sunlight they were receiving at the time of the call (9).
There was no evidence that symptoms were associated with any of these measures. Furthermore, when the researchers looked at the close to 2,000 participants whose scores indicated they were clinically depressed, there was still no relationship observed between mood symptoms and the time of year.
At this point we have considered evidence based on suicide rates, the dispensation of antidepressant medication, the level of chemicals associated with mood symptoms, and the association between people’s self reported symptoms of depression and the time of year. The results of all these studies provide no support whatsoever for the notion that more people are susceptible to the blues during the Christmas season. In fact, the results are very consistent in suggesting the opposite to be the case. Christmas appears to be a time of year when people are less likely to harm themselves, require less medication to treat mood symptoms, and have more “positive” chemistry in their brain. While there is evidence that mood symptoms vary according to the time of year, this may be a consequence of things like work demands, or the cumulative effect of a long winter, or past bad experiences. But the Christmas season does not appear to have any bearing on the population trends in major depressive disorder or most other forms of mental illness.
Does this mean that the Christmas Blues do not exist? Not at all. However, the evidence does suggest that being depressed at Christmas is probably different than being depressed because of Christmas. Like most days of the year, there are many cues during the Christmas holiday season that may remind people of past difficulties or losses. More importantly, we know that rumination about these sort of negative things is bad for you (10) (11) (12). If we focus more on the ruminations, and less on what is happening around us, it is certainly possible, and even very probable that sadness, if not outright depression will settle in. The cure for this sort of problem is not to change Christmas or its traditions, but to change how people approach their own negative thoughts and sad memories. We know that things like keeping active, getting involved socially, and engaging the mind in tasks such as planning and solving problems rather than worrying about them tends to protect us from negative emotions. It is behaviors like this that Christmas encourage, along with many of the other holidays and celebrations of various faiths and groups, because they help us feel better. Or as we learn in Dicken’s Christmas Carol,
“Men’s courses will foreshadow certain ends, to which, if persevered in, they must lead,” said Scrooge. “But if the courses be departed from, the ends will change.”
(Dickens, A Christmas Carol)
And if you can’t shake off the blues at this time of year? This can be a sign that depression is the problem, and this is what needs to be treated. Perhaps you need to make some changes in your life to get unstuck. Many people find it helpful to learn more about depression and carry out some self care. Unlike Ebenezer in Dickens’ A Christmas Carol, we should not wait until the symptoms get out of control, leaving us feeling isolated and unhappy, nor can we count on divine intervention to help set the record straight just in the nick of time. If your symptoms are more serious, both counseling and medication have been shown to be effective treatments. These require professional advice and competent guidance, and anyone can get started on this path to feeling better by talking to a healthcare provider. It can be hard to take the first step, but it is well worth the effort.
Williams, R. (2009). Why People Get Depressed at Christmas. Psychology Today.
NIH News in Health, January 2013 Beat the winter blues
Masterton, G. (1991). Monthly and seasonal variation in parasuicide: a sex difference. Brit J Psychiatry. 199(158), 155-7.
Phillips, D.P., & Wills, J.S. (1987). A drop in suicides around major national holidays. Suicide Life Threat Behav. 17, 1-12.
Cullum, S.J. et al (1993). Deliberate self-harm and public holidays: is there a link? Crisis. 14, 39-42.
Jessen, G. & Jensen, B.F. (1999). Attempted suicide and major holidays in Europe: findings from the WHO/EURO Multicentre study on parasuicide. Acta Psychiatr Scand. (99), 412-18.
Hartig, T. et al, (2013). Vacation, collective restoration, and mental health in a population. Societ Ment Health. 3(3), 221-36.
Molendijk, M.L. et al (2012). Serum BDNF concentrations show strong seasonal variation and correlations with the amount of ambient sunlight. PLoS ONE 7(11) DOI: 10.1371/journal.pone.004046
Traffanstedt, M.K., Mehta, S., & LoBello, S.G. (2016) Major Depression with seasonal variation: is it a valid construct? Clin Psychological Sci. DOI: 10.1177/2167702615615867
Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review, 30, 217– 237. doi:10.1016/j.cpr.2009.11.004.
Moyal, N., Henik, A., & Anholt, G. (2013). Cognitive strategies to regulate emotions - current evidence and future directions. Front Psychol. DOI: 10.3389/fpsyg.2013.01019
Cohen, N., Mor, N., & Henik, A. (2015). Linking Executive Control and Emotional response: a training procedure to reduce rumination. Clin Psychological Sci. 3(1), 15-25. DOI: 10.1177/2167702614530114
Leave a Reply