Letters to the Editor
Seasonal Affective Disorder: The Latitude Hypothesis Revisited
Axelsson and others’ recent article concerning the prevalence of seasonal
affective disorder (SAD) in people of Icelandic decent makes for interesting
reading (1). The authors report that people of wholly Icelandic descent
living in Winnipeg, Manitoba, show significantly lower rates of seasonality
than do residents of non-Icelandic decent (as measured by the Seasonal
Pattern Assessment Questionnaire [SPAQ] ). This finding complements
those of an earlier study conducted in the Interlake District of Manitoba
(3). The authors conclude that genetic factors play an important role in
the etiology of SAD, and they express some doubt regarding the veracity
of the “latitude hypothesis,” which posits a higher prevalence of SAD in
more northern latitudes, owing to the shorter winter day.
An earlier review of epidemiological studies in this area did not find
a significant overall correlation between prevalence of SAD and latitude
(r [n = 13] = 0.07, P = 0.42) (4). However, the authors did observe meaningful
correlations between SAD and latitude when the studies were split into
those conducted in North America (r [n = 7] = 0.90, P = 0.003) and those
conducted in Europe (r [n = 6] = 0.70, P = 0.06). While this review provided
a useful contribution, it had some limitations: it included several studies
not published in peer-reviewed journals, and it omitted published work
conducted in areas outside North America and Europe.
To further inform this issue, we performed a slightly different quantitative
synthesis of the literature. We included all peer-reviewed studies published
before October 2000 that examined the prevalence of SAD using the SPAQ
(n = 22), irrespective of the country in which they were conducted. We
analyzed the studies in 2 separate categories: those conducted in general
population samples (which are more representative of overall prevalence
at any given latitude) and those performed in specific subpopulations.
Where studies had been performed over several latitudes, we calculated
mean prevalence rates and latitudes.
The results of this analysis showed a significant correlation between SAD
and latitude in the general population studies (r [n = 12] = 0.66, P =
0.019) and an insignificant correlation in the specific subgroup studies
(r [n = 10] = 0.34, P = 0.34).
Our results suggest that the prevalence of SAD in the general population
does increase with higher latitude, consistent with the latitude hypothesis.
However, although latitude does reflect the daily photoperiod in a given
location, it is likely to be only a crude measure of other variables, such
as climatic conditions. These studies are also limited by the fact that
the SPAQ does not provide clinical diagnoses. We agree with Axelsson and
others’ view (1) that SAD probably has a complex etiology and pathophysiology
(5) influenced by several variables, such as environment, genetics, sociocultural
context, and psychosocial factors. Nevertheless, given the study results,
we do not wish to potentially “throw the baby out with the bathwater” and
discontinue studying the relation between SAD and latitude.
1. Axelsson J, Stefansson JG, Magnusson A, Sigvaldason H, Karlsson MM.
Seasonal affective disorders: relevance of Icelandic and Icelandic-Canadian
evidence to etiologic hypotheses. Can J Psychiatry 2002;47:153–8.
2. Rosenthal NE, Bradt GH, Wehr TA. Seasonal Pattern Assessment Questionnaire.
Bethesda (MD): National Institute of Mental Health; 1987.
3. Magnusson A, Axelsson J. The prevalence of seasonal affective disorder
is low among descendants of Icelandic emigrants in Canada. Arch Gen Psychiatry
4. Mersch PP, Middendorp HM, Bouhuys AL, Beersma DG, van den Hoofdakker
RH. Seasonal affective disorder and latitude: a review of the literature.
J Affect Disord 1999;53:35–48.
5. Lam RW, Levitan RD. Pathophysiology of seasonal affective disorder:
a review. J Psychiatr Neurosci 2000;25:469–80.
Erin E Michalak, PhD
Raymond W Lam, MD, FRCPC
Vancouver, British Columbia