Apr 15, 2020

Are you more likely to die in an Ontario hospital on a weekend?

Rendek v. Dufresne, 2006 ABQB 663 (CanLII)

Are you more likely to die in an Ontario hospital on a weekend?

The medical literature identifies a ‘weekend effect’ in mortality data (worldwide). This is prima facie negligence: if the hospital has one implicit standard for weekdays (as proven by a lower weekday mortality rate) and a more accepting standard for weekends (higher mortality rate) – the only cause is that the hospital and the doctors have so arranged their own comfort that patients die on weekends at a higher rate.

What does the data say?

Faust Feldman and Chawla look at mortality at the Beth Israel Deaconess Medical Center in Boston MA (2001-2012). They break 40,000 admissions into various cohorts including single hospital admission, age, emergency, ICU stay. Most importantly, they controlled for time of day. Critical incidents seem to occur in off-hours when the regular (i.e. highly experienced) staff are not generally present. Having created identical weekday and weekend subgroups. The mortality coefficient as between weekday treatment and weekend treatment, should be 1:1. In other words there should be no higher mortality on a weekend ( a number >1). Here are the actual results:

  • With all controls applied to make both groups identical in composition, you are more likely to die on a weekend operating table (2.03). The 95% confidence interval is 1.52 and 2.71. (p.5 of 9).
  • Interestingly, trauma-mortality is effectively identical between weekend and weekday. This hospital has so evolved that its trauma output is identical regardless of day of the week.
  • Ventilators are all the rage in the last two months. Everybody wants a ventilator until they get ventilator-acquired pneumonia at twice the rate on weekends (4.16) as on weekdays (2.81)(p.6 of 9)

What does the law do with this weekend effect?

Various cases identify the fact that medical care is, in that case, deficient on a weekend. Some cases fail to prove the absence of specific procedures earlier (i.e. on the weekend) would have had an appreciable effect (LaViolette v. Memiche, [2008] N.B.J. No. 182)

Lahey v Craig is an excellent example where the judge points out that the head nurse of outpatient emergency never worked nights and weekends, leaving the ‘charge’ nurse to rotate among other nurses who were not paid for their supervisory responsibilities. This self-serving conduct was a factor in that judge’s finding of liability (as was the apparently intentional failure to properly record events in the patient’s chart).

In Rendek at 144 Burrows J. criticizes a hospital expert opinion that the weekend standard of care is lower than on a weekday. This was a very sneaky way for the negligent doctor in this case to pretend that she could not have ordered weekend tests even if she wanted to.

This is a small sample of the ‘weekend effect’ in Canadian negligence law.

economics comment: regulatory capture

The manner in which medical service has evolved (within a state-regulated environment) permits the development of a situation where hospitals are not run like McDonalds. McDonalds has evolved to give you the same hamburger anywhere in North America at 4am on a weekend as a 12 noon on a weekday. Medical service has evolved to put inexperienced partial staff in hospitals on weekends. This produces the inevitable mortality result.

take-away: lawyers are evidently aware of the weekend effect. That it is allowed to continue is an artifact of public medicine and is overwhelming evidence of regulatory capture. Hospitals, being in effect, state-agencies (who do not answer to the patient because the patient doesn’t pay) allow doctor comfort and doctor convenience to rule output. Like every other demonstration of Canadian public sector negligence, nothing will be done about this.


see also

Dennis Buchanan commented
The assertion that the "only cause" could be differences in the quality of care...seems an oversimplification.
Dennis Buchanan commented
The assertion that the "only cause" could be differences in the quality of care...seems an oversimplification.
Edward Conway commented
It would be a 'simplification' if the studies did not control the identical characteristics of the weekday and weekend surgical cohorts. If patients are identical in risk profile, and the law of large numbers is in operation, significantly elevated weekend mortality establishes that surgery is different on weekends - why - we all know why - hospital staffs are dramatically different on weekends. That is the glaring and generally admitted fact across North American hospitals. Why - because staff have generally pressed for a 'normal' work week regardless of what this does to patient output. This is not the fault of staff - it is the hospital who permits it. It is no different than any factory which produces on a 24 hour cycle. There is no excuse for your vehicle falling apart at twice the rate of a 'weekday' car because it was produced on the 'weekend' shift at the Ford factory.