Abstract
Background Weekend vs weekday hospital admission has been associated with poorer mortality rates for many conditions. Studies evaluating weekend admission for hip fractures have resulted in contradictory conclusions regarding outcomes.
Methods We conducted a retrospective analysis of all patients who underwent surgery for a fragility hip fracture at a quaternary level teaching hospital during a 6-year period. A total of 1,164 patients were included: 796 weekday admissions (Monday through Friday) vs 368 weekend admissions (Saturday and Sunday). Patients were subdivided based on surgeon experience level (473 consultants vs 690 nonconsultants). Statistical tests included chi-square tests and logistic regression. Demographic data included age, sex, prior hip fracture, fracture type, operation, and American Society of Anesthesiologists grade. The primary outcome was 1-year mortality. Secondary outcomes were acute mortality (<24 hours), subacute mortality (1 to 30 days), change in mobility from baseline at 1 year, preoperative delay (>48 hours), and surgical duration.
Results The weekend admission cohort had a higher 1-year mortality rate than the weekday admission cohort (30.4% vs 23.2%; P=0.029), while subacute mortality trended toward significance (P=0.083). No significant difference was seen in acute mortality (P=0.5). Hemiarthroplasty was associated with increased mortality at 12 months (P=0.012) compared to the other operative interventions. The median duration of surgery was lower in the weekend cohort vs the weekday cohort (1.15 hours [69 minutes] vs 1.23 hours [73.8 minutes]; P<0.001). Consultants performed surgeries 16.2 minutes faster than nonconsultants (P<0.001) and trended toward a lower 1-year mortality rate (22.1% vs 27.9%; P=0.058). No significant difference was seen in mobility change at 1 year in both the consultant vs nonconsultant analysis (P>0.9) and in the weekday vs weekend analysis (P>0.12).
Conclusion A significantly increased 1-year mortality rate and a shorter surgical duration were observed among patients admitted on the weekends.
INTRODUCTION
Globally, more than 1 million people sustain hip fractures yearly, with the worldwide prevalence expected to increase from 1.6 million in 2000 to between 4.5 and 6.3 million by 2050.1-3 This increasing prevalence, driven by an aging population, resulted in 50,900 hospitalizations between 2015 and 2016 in Australia alone.4 Even with advances in care, mortality rates remain high at approximately 30% within 1 year of fracture.5 Postoperative challenges include diminished mobility and ability to perform activities of daily living.6,7 The severe impact on patients underscores the potential for any improvements in operation efficacy to have a meaningful effect on patient outcomes.
Weekday vs weekend hospital admission has long been theorized to affect patient outcomes. This phenomenon has been termed the “weekend effect,” and multiple conditions have been studied for its impact.8-11 Notably, weekend vs weekday hospital admission has been associated with poorer mortality rates for aortic aneurysms, pulmonary embolisms, and stroke.12,13 The correlation between hip fractures and the weekend effect has been less clear because of conflicting evidence.
Multiple analyses have found that factors known to affect hip fracture outcomes—including age, sex, American Society of Anesthesiologists (ASA) grade, preoperative delay, preoperative mobility, and fracture type—are associated with increased mortality rates.14-18 Lower surgeon experience level has also been associated with poorer surgical outcomes.19 However, studies have reported conflicting evidence regarding an association between weekend admissions and mortality.14,17,20-24 Potential causes for a weekend effect on hip fracture surgical outcomes are unclear and require further investigation. One factor that has not been robustly studied is the variation in surgeon experience level between weekday and weekend admissions. Likewise, many databases only tracked acute mortality for 30 days. A systematic review by Downey et al found that only 1 hip fracture registry of 8 followed mortality up to 1 year after surgery.25 Thus, increased length of monitoring is also indicated to ensure poor outcomes are not missed.
This study examined the relationships between weekend admission, surgeon experience level, and outcomes up to the 1-year time point for patients who underwent operative intervention for hip fracture. We hypothesized that weekend admission would be associated with negative outcomes.
METHODS
Ethical approval was provided by the Queensland Metro South Human Research Ethics Committee as a negligible risk retrospective study. We then conducted a retrospective analysis of all patients who underwent surgery for a fragility hip fracture at a quaternary level teaching hospital between January 2011 and July 2018. Fragility hip fractures were defined as femoral neck (displaced and nondisplaced intracapsular), intertrochanteric, and subtrochanteric fractures caused by low-energy trauma, such as falls from a standing height. Data were collected from a deidentified hip fracture database maintained by the hospital orthopedics department under institutional review board approval. This database included information on all patients with fragility hip fractures, with 1-year follow-up collected from clinic evaluations and telephone communication with patients. Patients who did not undergo surgery or who were transferred from other facilities were excluded because they had no hospital admission time point. Patients who died after admission but prior to a planned surgery were included in the analysis of mortality relative to the hospital admission time point only.
Subjects were stratified by hospital admission time point—weekday admission (Monday through Friday) vs weekend admission (Saturday and Sunday)—and by surgeon experience level—consultant vs nonconsultant. Consultants are attending-level surgeons (senior medical officers and specialists). Nonconsultants are surgeons in training such as fellows, principal house officers, and registrars (residents).
The primary outcome was mortality up to 1 year after surgery. Secondary outcomes were acute mortality (<24 hours), subacute mortality (1 to 30 days), change in mobility from baseline at 1 year, preoperative delay (measured as a categorical variable ≤48 or >48 hours postadmission), and surgical duration. Baseline mobility was determined using the Parker Mobility Score (scored from 1 to 9) that assesses the ability to move around the house, leave the house, and go shopping.26 Higher scores represent better mobility. Demographic variables were age, sex, previous hip fracture, fracture type, and ASA grade. Categorical variables are presented as counts and percentages. Continuous variables are presented using medians and interquartile ranges. Differences between groups were determined using chi-square or Mann-Whitney tests. Statistical analysis was performed by Queensland Facility for Advanced Bioinformatics. The relationships between outcomes and surgeon experience level and between outcomes and hospital admission time point were assessed using chi-square or Mann-Whitney tests. Analysis of the relationship between each outcome and hospital admission time point was performed using logistic regression for binary categorical outcomes. The relationship between each outcome and surgeon experience level was also analyzed using logistic regression. Confounding factors were assessed for significant relationships with hospital admission time point, surgeon experience level, operation type, and outcomes of interest using logistic regression. Significance was determined by P<0.05.
RESULTS
A total of 1,164 patients were included in the analysis, with 796 in the weekday admission cohort vs 368 in the weekend admission cohort (Table 1). The population had a median age of 83 years (range, 58 to 102 years; P=0.5); 792 (68.0%) were females and 372 (32.0%) were males (P=0.5). One hundred twenty-one patients had had previous hip fractures, while 1,024 had first-time fractures (19 patients unknown) (P=0.2). No significant differences were found for fracture type, operation type, or ASA grade between the patients admitted on weekdays vs weekends.
Patient Demographics Overall and by Hospital Admission Time Point
Surgeon Experience Level
Surgeon experience level included 473 consultant level, 690 nonconsultant level, and 1 unknown (Table 2). Two significant relationships were found in the surgeon experience level analysis. Consultants were more likely to operate on displaced intracapsular fractures, while nonconsultants were more likely to operate on intertrochanteric and nondisplaced intracapsular fractures (P<0.001). Consultants performed 54.3% (220/405) of displaced intracapsular fracture surgeries despite only performing 40.7% (473/1,163) of total surgeries. Comparatively, nonconsultants handled the majority of intertrochanteric (63.3%; 105/166), nondisplaced intracapsular (68.8%; 352/512), and subtrochanteric (59.5%; 47/79) fractures.
Patient Demographics Overall and by Surgeon Experience Level
Consultants were more likely to perform hemiarthroplasties and total hip arthroplasties, while nonconsultants were more likely to perform cannulated screw, dynamic hip screw, and intramedullary nail surgeries (P<0.001). We found no relationship between hospital admission time point and surgeon experience level (P=0.2).
Outcomes Analysis by Hospital Admission Time Point
Weekend admission was associated with a 30.4% 1-year mortality rate compared to 23.2% for weekday admission (P=0.029) (Table 3). No significant difference was seen between the admission time point cohorts in acute mortality (<24 hours) (P=0.5), but the weekend cohort had a higher subacute mortality (1 to 30 days) rate than the weekday cohort (9.9% vs 6.7%, respectively), and the difference trended toward significance (P=0.083). The median duration of weekend surgeries was less than the median duration of weekday surgeries (1.15 hours [69 minutes] vs 1.23 hours [73.8 minutes]; P≤0.001). Changes in Parker Mobility Score from baseline to the 1-year time point were not significant between the weekday and weekend admission cohorts (P=0.12).
Outcomes by Hospital Admission Time Point
Outcomes Analysis by Surgeon Experience Level
Analysis of outcomes based on surgeon experience level (Table 4) showed that higher surgeon experience level was associated with decreased duration of surgery, with a median duration of 1.03 hours (61.8 minutes) for consultants vs 1.3 hours (78 minutes) for nonconsultants (P<0.001). Additionally, delayed surgeries (>48 hours from admission) were more frequently performed by consultants, with 20.9% of delayed cases performed by consultants vs 15.5% of delayed cases performed by nonconsultants (P=0.017). A favorable trend was noted regarding mortality at 1 year for consultant-led operations vs nonconsultant-led operations (22.1% vs 27.9%, respectively; P=0.058).
Outcomes by Surgeon Experience Level
Logistic Regression Analysis
Table 5 presents the logistic regression analyses of clinical outcomes by hospital admission time point and surgeon experience level. Mortality at 1 year was significant in the hospital admission time point analysis, and preoperative delay was significant in the surgeon experience level analysis.
Logistic Regression Analysis of Clinical Outcomes by Hospital Admission Time Point and Surgeon Experience Level
Although Table 4 showed a favorable trend for 1-year mortality for consultant-led vs nonconsultant-led operations, after adjusting for type of operation, consultant status no longer trended toward decreased mortality at 1 year (Table 6).
Logistic Regression Analysis of 1-Year Mortality by Surgeon Experience Level and Type of Operation
In the analysis of 1-year mortality by type of operation, hemiarthroplasty was associated with increased mortality at 1 year (P=0.012) (Table 6). However, no significant associations were seen between operation type and delay to surgery (>48 hours) (Table 7).
Logistic Regression Analysis of Preoperative Delay by Surgeon Experience Level and Type of Operation
DISCUSSION
Evidence is clear that a weekend effect exists for a wide variety of emergent surgical procedures, with studies showing an association between weekend admission and higher rates of morbidity and mortality.12,13,20,27-29 Controversy regarding the existence of a weekend effect for hip fractures is ongoing. Our retrospective study of 1,164 patients demonstrated that weekend admissions had a significant effect on patient mortality rates at 1 year (30.4% vs 23.2%; P=0.029), but no other significant negative outcomes were observed in the study. The weekend cohort had a higher subacute mortality (1 to 30 days) rate than the weekday cohort (9.9% vs 6.7%, respectively), although this difference only trended toward significance (P=0.083). We found no significant differences in mobility change from baseline (P=0.12) or preoperative delay (P=0.9) to help explain the mortality rate.
Weekend admission was associated with a significant, but minimal, difference in surgical duration compared to weekday admission: a <5-minute (0.08 hour) difference between median times (1.15 hours vs 1.23 hours, respectively; P<0.001). This difference is unlikely to justify the increased mortality rate, as shorter procedures are typically associated with lower complication rates (wound infection, revision, and death) in joint arthroplasty cases.30-32
Hemiarthroplasty was associated with increased 1-year mortality (P=0.012), but the fewer hemiarthroplasties performed on weekends than weekdays (29.9% vs 32.4%, respectively) do not explain the increased 1-year mortality rate in the weekend cohort. Furthermore, we found no significant associations between operation type and a >48-hour delay to surgery.
While one could theorize that patients admitted on the weekends have more significant preoperative risk factors than patients admitted on weekdays, we found no association between ASA grades (P=0.2) or prior hip fracture (P=0.2) and hospital admission time point.
Consultant surgeons performed significantly more of the cases delayed >48 hours (P=0.017). This trend could be partially explained by the fact that consultant-level surgeons were performing surgeries on higher risk patients who required more extensive preoperative optimization or on the more difficult cases. However, we found no difference in surgeon experience level between admission time point cohorts (P=0.2) and no relationship between hospital admission time point and preoperative delay (P=0.9).
While many potential explanations exist for differences in weekend outcomes, no clear cause has been established for hip fractures. In some cases, weekend admission may be associated with sicker patients or more severe injuries, leading to a higher rate of negative outcomes.33,34 Conventionally, degree of injury severity is unlikely to be the cause of the hip fracture weekend effect as hip fractures require urgent treatment and patients are unlikely to delay presentation. However, a 2020 study demonstrated that only 66.3% of patients presented within 2 days while some waited until weeks after injury,35 but whether this delay is related to a weekend effect is unclear. Another widely accepted hypothesis is that surgeon experience level plays a role in the weekend effect, with the assumption that upper-level surgeons are less likely to perform weekend procedures, thereby leading to poorer outcomes. However, we found no relationship between surgeon experience level and hospital admission time point.
Given the retrospective nature of our study, we were unable to account for potential differences in nursing staff experience and availability that have been shown to impact weekend operations and outcomes in other fields.36,37 Similarly, we were not able to account for the availability of services such as physiotherapy, occupational therapy, and geriatric specialists on weekends. Boutera et al suggested changes in discharge and follow-up after weekend admission as potential causes for an increased 2-month mortality.22 Asheim et al supported this explanation of possible postoperative factors impacting weekend admission.38 In their study, weekend discharge was associated with poorer long-term outcomes, including increased mortality.38
Our study has several limitations, including the retrospective nature, the narrow focus on a single institution, and the smaller sample size compared to other studies, making it difficult to account for differences between departmental protocols. In addition, follow-up data at 1 year were limited, likely because the quaternary hospital treats a predominantly rural population, and the patients can be difficult to contact. Future studies that look at longer term outcomes may highlight additional differences between the groups.
CONCLUSION
Among this fragility hip fracture cohort, weekend admission was associated with increased 1-year mortality. This study provides important evidence that a clinically significant weekend effect cannot be dismissed regarding hip fracture treatments. The need for further evaluation of possible causes for the weekend effect to improve future care is clear.
This article meets the Accreditation Council for Graduate Medical Education and the American Board of Medical Specialties Maintenance of Certification competencies for Patient Care, Medical Knowledge, and Practice-Based Learning and Improvement.
ACKNOWLEDGMENTS
Dr Cooke consults on behalf of Medacta International and DePuy Synthes. The other authors have no financial or proprietary interest in the subject matter of this article.
- ©2025 by the author(s); Creative Commons Attribution License (CC BY)
©2025 by the author(s); licensee Ochsner Journal, Ochsner Clinic Foundation, New Orleans, LA. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (creativecommons.org/licenses/by/4.0/legalcode) that permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
REFERENCES
In this issue
Jump to section
Cited By...
- No citing articles found.