Saying quiet has no effect in medicine (obviously)

Superstitions are sort of fun. They represent a lack of scientific understanding and critical thinking that I obviously hate, but they also offer us interesting and relatively harm free insights into the many failings of human cognition. I say “quiet” a lot. I like the word. Although it occasionally upsets a colleague, I have previously discussed why I am right and they are wrong. However, that post was criticized for lacking evidence (which is strange on a blog dedicated to evidence based medicine). Honestly, I think that post was enough to dispel the silly “quiet” myth. When you have perfect logic, you don’t necessarily need evidence. It is impossible for the words I say in the emergency department to affect the health of people in the community and therefore the workload of the emergency department. However, there was a fun new study published in the BMJ Christmas edition, so I decided to tackle the evidence against this superstition…

Paper #1: Saying quiet has no effect in medicine

Brookfield CR, Phillips PPJ, Shorten RJ. Q fever-the superstition of avoiding the word “quiet” as a coping mechanism: randomised controlled non-inferiority trial. BMJ (Clinical research ed.). 2019; 367:l6446. PMID: 31852676 [free full text]

This is a single centre, prospective, randomized, non-inferiority trial looking at the impacts of saying the word “quiet” on workload in a medical microbiology department in a teaching hospital in the United Kingdom. For 61 consecutive days, this team was randomized to either start the day by saying “today will be a quiet day” or to refrain from saying the word “quiet”. They then tracked the total workload on each day, which can generally be measured objectively in a microbiology lab. Before the trial started, they determined that adding 10 extra tasks per person over the course of a shift would make the day noticeably busier, but anything less than that would not be noticeable, so they set their non-inferiority margin as 30 total tasks for the day (10 for each of the 3 people working). (They do a much better job explaining their rationale for a non-inferiority margin than most trials, where the number seems arbitrary and often ridiculous.) During their 61 day trial, there were 2 full moons and a summer solstice, but no Friday the 13ths. The mean number of clinical events was 139 on control days and 145 on days when they said “quiet”, which was non-inferior according to their definition. By chance, more control days were scheduled on weekends, which are less busy in the lab. When they adjust for this, any difference between the two groups completely disappears. Of course, they did not control for other confounding factors, such as the “presence of black cats, cracked mirrors, or lone magpies.” This is a single centre study, and didn’t take place in the emergency department, so external validity might be limited, but there are 3 ED based studies that confirm these results, which are discussed below.

Bottom line: It shouldn’t need to be said, but of course saying “quiet” has no impact on medical work loads.

Paper #2: Saying quiet still has no effect in medicine, despite these results

Lamb J, Howard A, Marciniak J, Shenolikar A. Does the word ‘quiet’ really make things busier? Bulletin. 2017; 99(4):133-136. [full text]

As I write the bottom line for paper 1, I can already hear people clamoring. “But wasn’t there a paper that proved saying quiet was bad??!!” In this RCT, they either said “have a quiet night” or “have a good night” based on a coin flip before orthopedic house officer night shifts. They then looked at the number of admissions on those shifts. There were 18 shifts where they said “quiet” and 24 control shifts, but unlike the last study they don’t give us any information about potential confounders, such as the day of the week. There were more admissions on the “quiet” nights – 3.1 vs 1.7, p=0.04. Although this difference is statistically significant, there are a few red flags. Allocation concealment using a coin flip is far from ideal. It is possible that some house officers simply refused to work the “quiet” shifts, and so demanded a repeat coin flip. This is also a convenience sample, rather than consecutive patients. Furthermore, there was no sample size calculation or justification for why they decided to look at 42 shifts. Thus, the just barely statistically significant p value of 0.04 is highly suspicious of “p-hacking”, even if entirely unintentional by the researchers. It is also a little concerning that the journal that published this study isn’t indexed on PubMed. Perhaps most importantly, the study isn’t blinded, and the decision to admit is pretty subjective. It is entirely possible that we are seeing a self-fulfilling prophecy, in which house officers expected sicker patients on nights when they heard the word “quiet” and therefore ended up admitting more borderline cases.

Bottom line: Although this study is statistically significant, there is no mechanism through which a word can impact a community’s need for orthopedic admissions. This study is a good reminder that a “positive” study does not necessarily indicate a true effect. Bias, P hacking, and chance alone can produce apparently positive results when no real effect exists.

Paper #3: An abstract that states saying quiet has no effect in medicine

Johnson G. The Q**** Study – basic randomised evaluation of attendance at a children’s emergency department. Emergency Medicine Journal 2010;27:A11.

This third paper is only available in abstract form. The study took place in a pediatric emergency department, and at the beginning of each shift they opened an envelope that contained the word “quiet”, “busy”, or was blank. (I like the control with “busy”, which is lacking in the other studies). This word was said aloud, and then displayed in the department throughout the shift. Of course, the number of patients arriving on each shift was identical, no matter what was said.

Bottom line: The full study doesn’t seem to be published, but the results are clearly correct. Saying a word can’t impact ED workload. 

Paper #4: Two more studies verifying that saying quiet has no effect in medicine

Kuriyama A, Umakoshi N, Fujinaga J, et al. Impact of Attending Physicians’ Comments on Residents’ Workloads in the Emergency Department: Results from Two J(^o^)PAN Randomized Controlled Trials. PloS one. 2016; 11(12):e0167480. PMID: 27936189 [free full text]

This report actually covers 2 RCTs, both using the same methods, but looking at patients in two different areas of the emergency department – ambulatory patients and transferred patients. At the beginning of every shift, the residents working in the assigned area were randomized to receive either a “peppy” message from their attending, such as “hope you have a quiet day”, or no message. At the end of their shift, they filled out questionnaires about the number of patients seen, the perceived business, stressfulness, difficulty of the shift, mealtime duration, and fatigue at the end of the shift. They admit that they did not consider or adjust their data for the superstition level of the residents, or whether the residents were considered “black clouds”. They only randomized on weekdays, and when certain attending staff were working. The person analyzing the questionnaires was blinded to the shift allocation. 25 residents participated, and there were about 170 shifts randomized in both trials. Of course, there were no real differences in any of the outcomes. In J(^o^)PAN-2, there was statistically more patients seen in the “quiet” group, but it was 0.5 patients over a shift, there were no adjustments for the many statistical comparisons being made, and the residents rated the busyness and difficulty of their shifts identically despite that extra half patient.

Bottom line: Once again, saying quiet has no impact workload. 

Summary

It really shouldn’t need to be studied. Saying the word “quiet” cannot possibly affect how busy an emergency department is.

There is a difference between saying quiet randomly and actually describing a quiet department. When the department is actually quiet, which is pretty rare in emergency medicine, we expect a regression to the mean. The following hours are likely to be busier as things average out. Thus, although these studies clearly tell us that the simple act of saying quiet has no impact on workloads in medicine, emergency departments might actually get busier after someone points out how quiet it has been. However, the extra patients have nothing to do with the word “quiet”. They would have come anyway.

Similarly, words can’t make our shifts easier. There was one paper where they tried to jinx residents by saying “you will have a great call day”, and the “jinxed” group actually had fewer admissions, more sleep, and easier on call days. (Ahn 2002) Although they didn’t actually use the word “quiet”, the concept is the same. However, much like the paper by Lamb and colleagues above, the effect seen is clearly not real. The words didn’t make the shifts better – and difference had to have been from bias in the research design or chance alone. 

If you were wondering about other superstitions, many have also been studied, and they are all silly. The occurrence of a full moon has no impact on ED patient volume, ambulance runs, or admissions. (Thompson 1996; Zargar 2004; Kamat 2014) The full moon also doesn’t affect ICU mortality, seizures, dog bites, agitation, or renal colic. (Cohen-Mansfield 1989; Chapman 2000; Benbadis 2004; Molaee 2011; Nadeem 2014) A full moon might actually be protective against intracranial aneurysm rupture, except that this is almost certainly just the result of too many studies examining a phenomenon with no effect (which could teach us something about other areas of the medical literature). (Banfield 2017)

Lunar phase and star sign don’t impact outcomes after spinal surgery. (Joswig 2016) Moon phases, zodiac signs, and Friday the 13th don’t influence surgical blood loss or frequency of emergencies. (Schuld 2011) Lunar phases don’t correlate with acute MI or cardiac arrest. (Eisenburger 2003) Lunar cycles don’t influence the rate of childbirth or complications. (Arliss 2005) In a superstition entirely unknown to me, the consumption of bao buns has no impact on inpatient admissions, mortality, or sleep duration while on call. (Tan 2008)

Of course, there will be the occasional outlying study, but because there is no prior plausibility, they just serve as good examples of the fact that “positive” studies frequently occur by chance alone, and that researcher bias can easily shape research results. (Onozuka 2018)

Bottom line

Although it is sort of fun to test these superstitions, it is probably time for us to stop wasting time and money on voodoo and focus on real medical issues. 

Other FOAMed

I said quiet

The Q Word Superstition

References

Ahn A, Nallamothu BK, Saint S. “We’re jinxed”–are residents’ fears of being jinxed during an on-call day founded? The American journal of medicine. 2002; 112(6):504. [pubmed]

Arliss JM, Kaplan EN, Galvin SL. The effect of the lunar cycle on frequency of births and birth complications. American journal of obstetrics and gynecology. 2005; 192(5):1462-4. [pubmed]

Banfield JC, Abdolell M, Shankar JS. Secular pattern of aneurismal rupture with the lunar cycle and season. Interventional neuroradiology : journal of peritherapeutic neuroradiology, surgical procedures and related neurosciences. 2017; 23(1):60-63. [pubmed]

Benbadis SR, Chang S, Hunter J, Wang W. The influence of the full moon on seizure frequency: myth or reality? Epilepsy & behavior : E&B. 2004; 5(4):596-7. [pubmed]

Brookfield CR, Phillips PPJ, Shorten RJ. Q fever-the superstition of avoiding the word “quiet” as a coping mechanism: randomised controlled non-inferiority trial. BMJ (Clinical research ed.). 2019; 367:l6446. PMID: 31852676

Chapman S, Morrell S. Barking mad? another lunatic hypothesis bites the dust. BMJ (Clinical research ed.). 2000; 321(7276):1561-3. [pubmed]

Cohen-Mansfield J, Marx MS, Werner P. Full moon: does it influence agitated nursing home residents? Journal of clinical psychology. 1989; 45(4):611-4. [pubmed]

Eisenburger P, Schreiber W, Vergeiner G, et al. Lunar phases are not related to the occurrence of acute myocardial infarction and sudden cardiac death. Resuscitation. 2003; 56(2):187-9. [pubmed]

Johnson G. The Q**** Study – basic randomised evaluation of attendance at a children’s emergency department. Emergency Medicine Journal 2010;27:A11.

Joswig H, Stienen MN, Hock C, Hildebrandt G, Surbeck W. The influence of lunar phases and zodiac sign ‘Leo’ on perioperative complications and outcome in elective spine surgery. Acta neurochirurgica. 2016; 158(6):1095-101. [pubmed]

Kamat S, Maniaci V, Linares MY, Lozano JM. Pediatric psychiatric emergency department visits during a full moon. Pediatric emergency care. 2014; 30(12):875-8. [pubmed]

Kuriyama A, Umakoshi N, Fujinaga J, et al. Impact of Attending Physicians’ Comments on Residents’ Workloads in the Emergency Department: Results from Two J(^o^)PAN Randomized Controlled Trials. PloS one. 2016; 11(12):e0167480. PMID: 27936189 [free full text]

Lamb J, Howard A, Marciniak J, Shenolikar A. Does the word ‘quiet’ really make things busier? Bulletin. 2017; 99(4):133-136. [full text]

Molaee Govarchin Ghalae H, Zare S, Choopanloo M, Rahimian R. The lunar cycle: effects of full moon on renal colic. Urology journal. 2011; 8(2):137-40. [pubmed]

Nadeem R, Nadeem A, Madbouly EM, Molnar J, Morrison JL. Effect of the full moon on mortality among patients admitted to the intensive care unit. JPMA. The Journal of the Pakistan Medical Association. 2014; 64(2):129-33. [pubmed]

Onozuka D, Nishimura K, Hagihara A. Full moon and traffic accident-related emergency ambulance transport: A nationwide case-crossover study. The Science of the total environment. 2018; 644:801-805. [pubmed]

Schuld J, Slotta JE, Schuld S, Kollmar O, Schilling MK, Richter S. Popular belief meets surgical reality: impact of lunar phases, Friday the 13th and zodiac signs on emergency operations and intraoperative blood loss. World journal of surgery. 2011; 35(9):1945-9. [pubmed]

Tan MH, Lee Z, Ng B, et al. The Tao of bao: a randomised controlled trial examining the effect of steamed bun consumption on night-call inpatient course and mortality. Annals of the Academy of Medicine, Singapore. 2008; 37(3):255-3. [pubmed]

Thompson DA, Adams SL. The full moon and ED patient volumes: unearthing a myth. The American journal of emergency medicine. 1996; 14(2):161-4. [pubmed]

Zargar M, Khaji A, Kaviani A, Karbakhsh M, Yunesian M, Abdollahi M. The full moon and admission to emergency rooms. Indian journal of medical sciences. 2004; 58(5):191-5. [pubmed]

Image by Vitabello from Pixabay