Hospital admission on weekends for patients who have surgery and 30-day mortality in Ontario, Canada: A matched cohort study

Abstract

Background

Healthcare interventions on weekends have been associated with increased mortality and adverse clinical outcomes, but these findings are inconsistent. We hypothesized that patients admitted to hospital on weekends who have surgery have an increased risk of death compared with patients who are admitted and have surgery on weekdays.

Methods and findings

This matched cohort study included 318,202 adult patients from Ontario health administrative and demographic databases, admitted to acute care hospitals from 1 January 2005 to 31 December 2015. A total of 159,101 patients who were admitted on weekends and underwent noncardiac surgery were classified by day of surgery (weekend versus weekday) and matched 1:1 to patients who both were admitted and had surgery on a weekday (Tuesday to Thursday); matching was based on age (in years), anesthesia basic unit value for the surgical procedure, median neighborhood household income quintile, resource utilization band (a ranking system of overall morbidity), rurality of home location, year of admission, and urgency of admission. Of weekend admissions, 16.2% (25,872) were elective and 53.9% (85,744) had surgery on the weekend of admission. The primary outcome was all-cause mortality within 30 days of the date of hospital admission. The 30-day all-cause mortality for patients admitted on weekends who had noncardiac surgery was 2.6% (4,211/159,101) versus 2.5% (3,901/159,101) for those who were admitted and had surgery on weekdays (adjusted odds ratio [OR] 1.05; 95% CI 1.00 to 1.11; P = 0.03). However, there was significant heterogeneity in the increased odds of death according to the urgency of admission and when surgery was performed (weekend versus weekday). For urgent admissions on weekends ( n = 133,229), there was no significant increase in odds of mortality when surgery was performed on the weekend (adjusted OR 1.02; 95% CI 0.95 to 1.09; P = 0.7) or on a subsequent weekday (adjusted OR 1.05; 95% CI 0.98 to 1.12; P = 0.2) compared to urgent admissions on weekdays. Elective admissions on weekends ( n = 25,782) had increased risk of death both when surgery was performed on the weekend (adjusted OR 3.30; 95% CI 1.98 to 5.49; P < 0.001) and when surgery was performed on a subsequent weekday (adjusted OR 2.70; 95% CI 1.81 to 4.03; P < 0.001). The main limitations of this study were the lack of data regarding reason for admission and cause of increased time interval from admission to surgery for some cases, the small number of deaths in some subgroups (i.e., elective surgery), and the possibility of residual unmeasured confounding from increased illness severity for weekend admissions.

Conclusions

When patients have surgery during their hospitalization, admission on weekends in Ontario, Canada, was associated with a small but significant proportional increase in 30-day all-cause mortality, but there was significant heterogeneity in outcomes depending on the urgency of admission and when surgery was performed. An increased risk of death was found only for elective admissions on weekends; whether this is a function of patient-level factors or represents a true weekend effect needs to be further elucidated. These findings have potential implications for resource allocation in hospitals and the redistribution of elective surgery to weekends.

Abstract

In their cohort study, James Daniel O'Leary find that data stratification leads to fuller insight into the 'weekend effect' of increased mortality following hospital admission and surgery.

Author summary

Why was this study done?

• There is ongoing debate about whether a weekend effect found in most observational research studies reflects actual differences in healthcare outcomes on the weekend or is a data artifact from using administrative datasets.

• The possibility of a weekend effect continues to inform the allocation of hospital resources, healthcare policies, and clinical practices.

What did the researchers do and find?

• We used provincial administrative healthcare and demographic databases in Ontario, Canada, from 2005 to 2015 to assemble a cohort of 159,101 adults who were admitted on weekends and had surgery performed during their hospital stay on the same weekend or a subsequent weekday, and who were matched directly to adults with weekday admissions and surgery.

• We found an increase in the crude mortality rate for all weekend admissions, but there was significant heterogeneity in adjusted outcomes according to the urgency of weekend admission and when surgery was performed (weekend versus weekday).

• For urgent admissions on weekends, there was no significant increase in the adjusted odds of death. However, elective admissions on weekends were associated with clinically important increases in odds of death compared with weekday admissions.

What do these findings mean?

• The proportional increase in the adjusted mortality rate for weekend admissions was lower than that found in most previous observational studies using administrative data.

• However, stratification of these findings in the current study suggests that only elective admissions on weekends are associated with an increase in odds of death.

• Whether the increased risk of death found for elective admissions on weekends is a factor of increased illness severity in this population, requiring preoperative optimization, or a true weekend effect needs to be elucidated in further studies.