Feb

20

Copy/paste from http://groups.yahoo.com/group/TWSAPI/ (30/01/2008):

"I spent 35 years tending to the the Emergency Rooms as an orthopaedic surgeon. I can assure you that if you draw a graph of visits per day it produces a perfect sine curve. The apogee coincides with the full moon and the perigee with the new moon. The variation is about 25%. Large ERs plan their staffing levels on the phases of the moon. This has been going on for decades. The visits are not only mental (lunacy) but also physical. It is clear that the phase of the moon effects mankind. Why not its trading? …"

In a past life in the health care industry, we noted a few studies establishing correlations between moon phase and ER visits. Often times, these studies purported to zero in on specific conditions or causes of injury, like assault or mental disorder, which might be supposed to support the popular notion of 'crazy behavior' at the full moon. However, I recall most of the evidence of same being anecdotal and no statistically significant correlation between the variables over broad samples or studies. That is not to say that this phenomenon could not have been or has not been experienced by a particular hospital ER or at a local or regional level from time to time.

Another such popular supposed relationship is between incidence of crime, particularly a variety of assaults, during the full moon phase. Similar studies with similar results have been conducted and the anecdotal evidence is also there, but overall, the relationship seems to be no stronger than does it in health care.

One could take this a step further in the quest of causality by hypothesizing that the full moon caused a spike in mental disorder, which caused a spike in assaults, which in turn caused a spike in ER visits, perhaps at least in part a reasonable argument, certainly a romantic one, but I for one have no interest in taking that leap. We here have been cautioned to beware spurious correlation in counting. Let's do so with the full moon and the ER.

That said, the facts seem to indicate that on a broad level, ER visits are pretty flat, rarely spiking at all any more. Certainly there are observable peaks, but weekdays are as busy or busier than weekends, the peaks often being Monday and Friday.

Again, this is not to say an individual hospital or groups of hospitals may not see a spike in the pattern from time to time, or that any given study may not find such evidence. For instance, we might try studying Flint, Michigan hospital ER's the night of GM's payday that happens to fall on a full moon. Guaranteed spike in visits for broken noses and knuckles, beer bottle concussions and alcohol poisoning. Sorry, though, no man-in-the-moon causality.

But, back to the point. I would suggest that, if they ever existed, any pattern exhibiting a spike in ER visits for ICD9 codes in the 800-900's correlating with the moon phase, or with any other variable, would have been subsumed in the past few decades by the propensity of many of us to use the ER as primary care provider. This is not to mention the number of visits by bored or lonely people with no companionship and nowhere to go, and there are plenty of those visits to go around, particularly in large urban centers. Ask any ER doc. Instead of hours of boredom separated by moments of terror, it's quite the opposite.

There are over 200MM visits made to hospitals in this country each year, over 100MM of them to the ER. About 300,000 ER visits per day, every day. As an aside, only 15% or so of all hospital visits result in a hospitalization, the remaining 35% or so are for outpatient services. Going out on a limb, albeit a pretty thick one, I'm more or less certain that backing up and viewing a daily plot of these 300K ER visits from across the room, the curve would look pretty flat.

In other words, the universe of ER visits has expanded pretty dramatically. Samples within it may exhibit variations, spikes even, but they're simply going to be less evident when the universe is being viewed as a whole. As I last saw it, the moon seems to impact the earthly hospital universe equally, not just samples of it. Frankly, I don't know of any studies of the ER universe, then or now, so I'll take the evidence of moon phase/ER visit correlation from samples like I take studies of coffee. One week it's good for you, the next it's not, but I'm still drinkin' it.

Would you like cream with those stitches, sir?


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search