Letters to the Editor
Emergency Medicine News welcomes letters to the editor about any subject related to emergency medicine. Please limit your letter to 250 words, and include your full name, credentials, and city and state of residence or practice.
Letters may be edited for content, length, and grammar. Submission of a letter constitutes the author's permission to publish on all media, including print, online, and social media, but does not guarantee publication. Letters express the views of the authors and do not necessarily reflect those of Emergency Medicine News and Wolters Kluwer.
Letters to the editor may be sent to [email protected].
I read with interest the article analyzing the association between inclement weather and psychiatric presentations to the ED. (EMN. 2023;4517; http://bit.ly/3RvApOL.) I particularly commend Dr. Tom Belanger for presenting his primary data to facilitate further scrutiny. However, I came to some different conclusions.
I agree that his data do support the colloquial belief that psychiatric presentations appear disproportionately high on evenings and weekends. However, without comparing these numbers with other (nonpsychiatric) ED presentations, as he did with similar data for presentations over the course of the year, this association is not entirely clear.
As for the finding that behavioral presentations to the ED increase with inclement weather (i.e., colder days), I came to a different conclusion. These data all came from EDs in Texas, and temperatures there should not be generalized to more temperate areas of the United States. The temperature data show that daily high temperatures were rarely below 40°F and never below 20°F. Conversely, quite a lot of days feature highs above 90°F.
While visually the trend does appear to hold up, I have to believe that much of this trend suggests that presentations do in fact increase when the weather is inclement, but in this region, that would more often be from excessively hot temperatures. Dr. Belanger says as much and even cites a study that similarly regarded presentations under extreme heat, but I think this is all still in support of the idea that when the weather gets rough, those at the limits of their coping skills seek emergency care.
I expect the common wisdom driving this study—that presentations rise when the temperature drops—comes from areas with greater morbidities from exposure to cold than heat, which likely represents a greater portion of the United States. I suspect this is really a U-shaped curve, that the further temperatures go from a balmy 70, the higher the presentation rate.
Moreover, data may look different when analyzing the daily low temperature, especially because his data also suggest presentations increase later in the day when the temperature drops. I'd suggest this study be recapitulated in the Northeast or the upper Midwest before generalizing any conclusions. Credit and kudos to Dr. Belanger for presenting his analysis of a phenomenon common to our field and all the work that such a study entails.
Eric Goldlust, MD, PhD
Santa Clara, CA
Dr. Belanger responds: This critique is excellent! To your first point, I reported the periodicity data (time of day and day of week) as a percentage of behavioral health-related visits of the total; I think this is a decent initial way to address your concern. To your second point, I would absolutely bet you are right! This is a great point. It may just not have been cold enough to see the effect I was looking for near the low temps. Incidentally, I did look just at the range of very low temperatures to see if a different trend emerged (as you got closer to freezing, would the curve bend upward again?), and I just don't think my n was high enough to visualize this with confidence.
I also looked at the daily lows in addition to the daily highs; this didn't seem to matter. And I looked to see if nighttime visits changed during the cold days (because I reasoned it was probably colder at night on the cold days), and that also didn't appear to change things. I also really like your idea of seeing if a model of distance from ideal temperature better explains reality than a linear model linked to absolute temperature (as I did here). That said, I would absolutely love to repeat this in a different environment (and with competing models). Let me know if you're interested.