Association of weekend admission on immediate postoperative outcomes following non-elective cardiac operations

Abstract

Background

The “weekend effect” refers to inferior outcomes among patients admitted during weekends. Although documented across multiple medical specialties, studies evaluating weekend admission for cardiac surgery have yielded conflicting results. We assessed immediate postoperative outcomes following weekend cardiac admissions using a contemporary national database.

Methods

Adults (≥18 years) undergoing non-elective major cardiac operations, including coronary artery bypass grafting, valvular, and aortic procedures, were identified from the 2016–2022 National Inpatient Sample. Patients were stratified by weekend versus weekday admission, defined as operative intervention within a day of admission. The primary outcome was in-hospital mortality. Secondary outcomes included length of stay (LOS), hospitalization costs, and non-home discharge. Multivariable regression models adjusted for demographics, comorbidity burden, operative type, hospital volume, and year.

Results

Among an estimated 99,865 patients, 14,565 were admitted over the weekend. Compared to others, weekend patients were younger, more commonly female, and were more frequently insured by Medicaid (all p < 0.001). Following risk adjustment, weekend admission was associated with higher odds of in-hospital mortality (AOR 1.37, 95% CI 1.18–1.60), increased hospitalization costs (β +$5,831), and prolonged LOS (β +0.86 days, all p < 0.001). Mortality risk was attenuated at higher-volume centers.

Conclusion

Weekend admission for non-elective major cardiac surgery is associated with worse clinical outcomes and increased resource utilization nationwide, highlighting a persistent weekend care disparity which warrants further investigation into underlying drivers.

Graphical abstract

Introduction

The “weekend effect”, which describes the inferior patient outcomes observed on weekends compared to weekdays, represents a widely described yet incompletely understood phenomenon in healthcare. Despite incremental advances in perioperative care delivery and protocolized management, weekend admission has been associated with increased mortality and complication rates across multiple surgical populations. In fact, adverse outcomes are reported to range from 5% to over 20%, depending on diagnosis and acuity [1], [2], [3]. Smith et al. reported weekend admissions account for up to 25% of all acute hospitalizations, demonstrating the broad clinical relevance of this phenomenon [4]. Notably, mechanisms underlying the weekend effect remain obscure, as differences in patient severity, staffing patterns, and access to specialized resources may contribute to observed outcome disparities [5].

Prior studies have corroborated the detrimental effects of weekend admission on perioperative processes and recovery trajectories. Liu and colleagues demonstrated reduced adherence to enhanced recovery measures, such as early mobilization and invasive catheter removal, among patients undergoing surgery near or during the weekend, particularly at hospitals with lower nurse-to-bed ratios [6]. Additionally, several studies have suggested weekend admission to be associated with delayed diagnostics, reduced continuity of care, and limited multidisciplinary engagement, which may elevate postoperative risk independent of operative complexity [7], [8], [9]. Nevertheless, evidence specific to cardiac operations remains conflicting, with some registry-based analyses reporting increased mortality and others demonstrating no independent weekend effect after comprehensive risk adjustment [10], [11]. This heterogeneity may be a result of patients presenting with greater case severity. Furthermore, weekend staffing models involve reduced availability of dedicated cardiac anesthesia, perfusion, and intensive care coverage which may delay time-sensitive perioperative decision-making. Additionally, reduced access to interventional and diagnostic resources as well as advanced imaging may further constrain timely escalation of care.

Despite relevance to policy and clinical care, outcomes associated with weekend admission following major cardiac operations remain insufficiently characterized at the national level. The present study employed a contemporary, nationally representative cohort to evaluate the association between weekend admission and postoperative outcomes following major cardiac surgery. We hypothesized weekend admission would be independently associated with increased mortality, postoperative morbidity, and resource utilization.

Methods

This was a retrospective cohort study utilizing the 2016–2022 National Inpatient Sample (NIS). Maintained by the Agency for Healthcare Research and Quality, the NIS is the largest publicly available all-payer inpatient database in the United States, capturing approximately 20% of all hospitalizations annually. Using validated discharge and survey-weighting methodology, the NIS provides accurate national estimates for approximately 97% of all inpatient hospitalizations nationwide [12]. Owing to the deidentified nature of the NIS, this study was exempt from review by the Institutional Review Board (IRB#17-001112).

All adult patients (≥18 years) undergoing non-elective major cardiac operations within the first procedure day were identified using International Classification of Diseases, Tenth Revision (ICD-10) procedure codes. Only non-elective admissions were considered in order to optimize comparison between groups. Major cardiac operations were classified hierarchically into three mutually exclusive categories: proximal aortic (any patient with a proximal aortic procedure, regardless of concomitant coronary or valvular surgery), valvular (any patient with a valve procedure who did not undergo a concomitant proximal aortic procedure, including isolated valve repair/replacement and combined CABG-valve procedures), and isolated CABG (patients undergoing CABG without a concomitant valve or proximal aortic procedure). The study population was stratified based on admission timing into weekend (Weekend) versus weekday (Weekday) cohorts. Weekend admission was defined as hospital admission on a Saturday or Sunday that resulted in immediate operative intervention within a day of admission as defined by the “Procedure Day” variable, in accordance with Healthcare Cost and Utilization Project (HCUP) reporting guidelines [13]. Under this definition, a patient admitted on a weekend day but not operated on until the following Monday was classified as Weekday, since operative intervention occurred beyond one hospital day of admission; conversely, a patient admitted on a Friday but taken to the operating room on Saturday was classified as Weekend, as the qualifying operative day itself fell on a weekend day. Records with missing key demographic or outcome data were excluded from analysis (1.1%, Fig. 1). Furthermore, those requiring heart transplantation, ventricular assist device placement, or with a history of endocarditis, were not considered (0.32%).

Fig. 1.

Consolidated standards of reporting trails diagram of estimates.

Patient-level characteristics included age, sex, race, insurance coverage, and comorbidity burden. The burden of chronic conditions was quantified using the Elixhauser Comorbidity Index, a validated composite of 30 comorbidities. Hospital-level variables included annual facility cardiac operation volume and year of admission. Annual institutional cardiac volume was calculated as the sum of all interventional included for analysis, performed at each center. Hospitalization costs were calculated via the application of center-specific cost-to-charge ratios to overall charges followed by inflation adjustment to the 2022 Personal Health Care Price Index [14].

The primary outcome was in-hospital mortality, defined as postoperative, in-hospital death during the index admission following operative intervention. Secondary outcomes included postoperative complications, length of stay (LOS), hospitalization costs, and non-home discharge, defined as discharge to a skilled nursing facility, rehabilitation center, or other acute care facility.

Categorical variables are reported as proportions (%) and continuous variables as medians with interquartile range (IQR). Bivariate comparisons were performed using Pearson χ² test for categorical and Mann-Whitney U test for continuous variables. A doubly robust risk adjustment framework was employed to ensure appropriate comparisons. Entropy balancing was first used to reach covariate-balanced groups between Weekend and Weekday cohorts (Supplemental Table S1) [15]. Multivariable linear and logistic regression models were then developed to evaluate associations between weekend admission and outcomes of interest. Several variables used in regression models were automatically selected by elastic net regularization and included patient age, Elixhauser Comorbidity Index, race, insurance coverage, operation type, presence of comorbidities, annual facility cardiac operation volume, and year of admission, and history of cardiac operations as well as redo sternotomy. To better assess patient severity, additional factors such as cardiogenic shock and history of myocardial infarction were also integrated into regression models. It is important to note that while elastic net regularization was used to guide model selection, clinically significant variables were retained for analysis. Model performance was assessed using receiver operating characteristic curves, calibration, and determination coefficients (R²) as necessary. Linear regression model outputs are reported as beta coefficients (β), and logistic estimates are detailed as adjusted odds ratios (AOR), both with 95% confidence intervals (CI). Statistical significance was defined as an α of 0.05. Linear regression estimates were used to provide more directly interpretable estimates of absolute risk difference for a clinical and administrative audience. This approach is double robust in which entropy balancing first achieves covariate balance between exposure groups, such that the subsequent regression step principally adjusts for residual imbalance rather than models the full conditional distribution of the outcome. Moreover, the Central Limit Theorem ensures that the sampling distribution of the regression coefficients approximates normality, such that skew in the outcome variable itself does not compromise the validity of the coefficient estimates or confidence intervals.

The relationship between annual cardiac surgery volume and risk-adjusted mortality was modeled separately for Weekend and Weekday cohorts using restricted cubic splines, with predicted margins plotted against volume.

Rates of weekend admission were further analyzed by stratifying by U.S. Census divisions (New England, Middle Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain and Pacific), in accordance with Healthcare Cost and Utilization Project reporting standards [13]. All analyses were performed using Stata 18.1 (StataCorp, College Station, TX) and incorporated NIS survey weights to generate national estimates.

Results

Patient characteristics

Of an estimated 99,865 adult hospitalizations for major cardiac surgery during the study period, 14,565 (14.6%) comprised Weekend, with baseline characteristics summarized in Table 1. Those within Weekend were younger (63 [54–72] years vs 65 [57–72], p < 0.001), more commonly female (30.1 vs 28.0%, p < 0.001), and more frequently of Black race (14.3 vs 10.2%, p < 0.001). Compared to others, Weekend more commonly was of the lowest income quartile (27.6 vs 27.0%, p < 0.001) and more often insured by Medicaid (12.7 vs 9.7%, p < 0.001). Furthermore, Weekend more commonly underwent CABG (47.1 vs 55.7%, p < 0.001). Those within the Weekend group more often presented with hypertension (67.1 vs 73.5%, p < 0.001), liver disease (8.2 vs 5.2%, p < 0.001), and peripheral vascular disease (51.3 vs 32.0%, p < 0.001). Across US Census Divisions, the South Atlantic and Pacific demonstrated the greatest proportion of Weekend operation (Fig. 2). However, the Pacific had the highest mean hospitalization costs (Fig. 3, Supplemental Table S2).

Table 1.

Patient, and hospital characteristics of patients receiving an operation during the Weekday or Weekend. IQR: Interquartile Range, CABG: Coronary Artery Bypass Grafting.

	Weekday (n = 85,300)	Weekend (n = 14,565)	p-Value
Age (years [IQR])	65 [57–72]	63 [54–72]	<0.001
Female (%)	28.0	30.1	<0.001
Elixhauser Comorbidity Index (median [IQR])	4 [3–5]	5 [4–6]	<0.001
Annual Cardiac Surgery Volume (median [IQR])	66 [37–103]	70 [38–108]	<0.001
Operation type (%)			<0.001
CABG	55.7	47.1
Valvular	20.0	7.9
Proximal Aortic	25.0	45.1
Income quartile (%)			0.25
>75%	21.2	21.1
51–75%	25.8	24.0
26–50%	26.4	27.2
0–25%	27.0	27.6
Insurance coverage (%)			<0.001
Private	33.7	34.6
Medicare	49.1	43.6
Medicaid	9.7	12.7
Other payer	7.7	9.1
Race (%)			<0.001
White	75.0	70.0
Black	10.2	14.3
Hispanic	8.2	8.1
Asian/Pacific Islander	3.3	3.8
Other	3.4	3.9
Comorbidities (%)
Congestive heart failure	31.9	30.7	<0.001
Diabetes	28.6	22.0	<0.001
Hypertension	73.5	67.1	<0.001
Liver disease	5.2	8.2	<0.001
Peripheral vascular disease	32.0	51.3	<0.001
Hospital teaching status (%)			0.01
Non-Metropolitan	85.1	87.5
Metropolitan non-teaching	12.8	11.0
Metropolitan teaching	2.1	1.5
Hospital division (%)			<0.001
New England	3.6	3.4
Middle Atlantic	13.1	10.9
East North Central	11.0	12.6
West North Central	6.8	7.6
South Atlantic	25.2	23.0
East South Central	7.8	8.4
West South Central	11.6	10.3
Mountain	9.1	7.6
Pacific	11.8	16.2

Fig. 2.

Proportion of weekend admission stratified by US Census Divisions. The Pacific Division had the greatest proportion of Weekend Admission.

Fig. 3.

Mean hospitalization costs stratified by US Census Divisions. Hospitalization costs were the greatest in the Pacific Division.

Unadjusted outcomes

Unadjusted rates of mortality were significantly higher among patients receiving an operation over the weekend (11.1 vs 6.6%, p < 0.001, Table 2). Additionally, those undergoing operative intervention over the weekend more frequently experienced cardiac (53.0 vs 37.3%, p < 0.001), infectious (7.4 vs 4.5%, p < 0.001), and renal complications (35.6 vs 25.7%, p < 0.001, Fig. 4). Moreover, individuals receiving weekend management had lower unadjusted costs ($51,921 [37,901-77,814] vs $67,012 [46,213-98,916], p < 0.001), and longer length of stay (9 days [5–13] vs 7 [5–10], p < 0.001).

Table 2.

Unadjusted and risk-adjusted outcomes of patients undergoing operative intervention during the Weekday or Weekend. Estimates from dichotomous outcomes are reported as % or AORs and continuous outcomes as median with IQR or β-coefficients. IQR, Interquartile Range; AOR, Adjusted Odds Ratio; CI, Confidence Interval.

	Unadjusted			Adjusted (AOR/β)
	Weekday	Weekend	p-Value	Estimate	95% CI
Clinical outcomes
In-hospital mortality	6.6	11.1	<0.001	1.37	1.18–1.60
Individual complications
Blood transfusion	23.3	27.8	<0.001	1.13	1.03–1.26
Cardiac	37.3	53.0	<0.001	2.33	2.12–2.56
Infectious	4.5	7.4	<0.001	1.40	1.17–1.67
Respiratory	18.2	23.8	<0.001	1.22	1.10–1.35
Renal	25.7	35.6	<0.001	1.25	1.13–1.37
Thromboembolic	1.2	1.9	0.01	1.43	0.99–2.06
Resource utilization
Non-home discharge	24.6	30.9	<0.001	1.22	1.09–1.36
Duration of stay (days) [IQR]	7 [5–10]	9 [5–13]	<0.001	+0.86	+0.45–1.30
Costs ($1000) [IQR]	67.0 [46.2–98.9]	51.9 [37.9–77.8]	<0.001	+5.83	3.21–8.44

Fig. 4.

Association of Weekend Operation with outcomes following non-elective cardiac surgery. Following entropy balancing and risk-adjustment, Weekend remained significantly associated with greater odds of mortality and postoperative complications. Reference: Weekday. *Statistical Significance, p < 0.05. Error bars represent 95% confidence intervals.

Risk-adjusted outcomes

Adjusted outcomes are summarized in Table 2. Following comprehensive risk adjustment, weekend intervention was independently associated with significantly greater odds of in-hospital mortality (AOR 1.37, 95% CI 1.18–1.60, LROC = 0.76). Receipt of care during the weekend was also linked with greater odds of cardiac (AOR 2.33, 95% CI 2.12–2.56, LROC = 0.72), infectious (AOR 1.40, 95% CI 1.17–1.67, LROC = 0.74), and renal (AOR 1.25, 95%CI 1.13–1.37, LROC = 0.74) complications. Furthermore, a weekend operation was also associated with increased LOS (β +0.86 days, 95% CI 0.45–1.30, R² = 0.1327), higher hospitalization costs (β + $5831, 95% CI $3216–$8444, R² = 0.1272), and a greater likelihood of non-home discharge (AOR 1.22, 95% CI 1.09–1.36, LROC = 0.73). Although, weekend operative intervention remained associated with a greater risk of mortality, this effect was attenuated at higher-volume canters (Fig. 5).

Fig. 5.

Risk-adjusted relationship between annual cardiac surgery volume and risk-adjusted mortality. As volume increases, Weekend operation remains associated with greater risk of mortality. This model was creating using a restricted cubic spline. To formally test whether the relationship between volume and mortality differed by weekend status, a separate model including volume spline terms interacted with weekend status was compared against a reduced model omitting the interaction terms using a likelihood ratio test. *Statistical Significance, p < 0.0001, LRχ² = 74.27.

In a subgroup analysis, weekend operative intervention remained linked with greater odds of in-hospital mortality across procedure subtypes (CABG: AOR 1.71, 95% CI 1.31–2.20, LROC = 0.72; Valvular: AOR 5.12, 95% CI 3.51–7.62, LROC = 0.70; Proximal Aortic: AOR 1.04, 95% CI 1.02–1.25, LROC = 0.73). Additionally, across all operations, female sex was linked with increased odds of in-hospital mortality following weekend intervention (AOR 1.36, 95% CI 1.20–1.53, LROC = 0.75) as well as Black race (AOR 1.29, 95% 1.08–1.55, LROC = 0.76).

Discussion

In the present study, we utilized a nationally representative dataset to examine the association between weekend admission and immediate postoperative outcomes following non-elective major cardiac operations. Despite ongoing efforts to standardize perioperative care, weekend admission was associated with a substantial increase in in-hospital mortality, postoperative complications, and resource utilization. Specifically, weekend admission conferred increased odds of prolonged hospitalization, higher hospitalization costs, and greater likelihood of non-home discharge compared with weekday admission. Furthermore, we identified several patient, hospital, and regional- factors modifying the relationship between weekend admission and outcomes, including institutional surgical volume and geographic variation. The magnitude of these disparities and their resource implications emphasize the importance of optimizing weekend cardiac surgical care.

Over the study period, we noted weekend admission for major cardiac operations to be associated with substantially worse immediate postoperative outcomes and increased resource utilization compared with weekday admission. Prior national work has similarly shown weekend surgery confers significantly elevated perioperative risk, with Glance and colleagues reporting more than a two-fold increase in mortality and higher rates of major complications among weekend elective and urgent operations after risk adjustment [16]. These findings have been replicated across diverse surgical populations, suggesting the weekend effect extends beyond specific procedure types [4], [11]. Additionally, weekend admissions have been linked with higher rates of adverse events and delayed recovery trajectories even after adjusting for patient acuity [9]. Previous literature has demonstrated weekend outcome disparities to reflect systemic differences in care delivery rather than patient-level factors alone, with reducing staffing levels, limited access to subspeciality support, and delays in postoperative escalation of care which have been proposed as potential contributors to inferior weekend outcomes [2], [8]. Recent analyses further suggest these disparities persist across diverse international health systems, indicating temporal variation in care remains a ubiquitous and unresolved challenge [17]. Nevertheless, our findings demonstrate weekend admission itself remains an independent marker of increased perioperative risk and inefficiency following major cardiac operations, warranting focused efforts to standardize perioperative resources and staffing across the week. Given that our dataset does not capture staffing levels, resource availability, or perioperative processes directly, future work is needed to evaluate this directly as our work aims to serve as a foundational assessment of the data provided.

Notably, the adverse association between weekend admission and in-hospital mortality was significantly attenuated at higher-volume cardiac centers, suggesting that institutional capacity modifies the “weekend risk”. This observation is consistent with the established volume-outcome relationship in cardiac surgery, where high-volume centers achieve superior perioperative survival and lower resource utilization through enhanced infrastructure and care coordination [18], [19], [20]. Prior work by Schwann and colleagues revealed progressively lower operative mortality with increasing institutional case volume among CABG patients, showing that even modest volume differences translate into meaningful reductions in adverse outcomes at the population level [21]. Gonzalez et al. reported patients treated at low-volume hospitals had two-fold greater odds of mortality after developing a postoperative complication despite comparable complication incidence, highlighting rescue efficiency as a critical determinant of survival [22]. Hospital-level infrastructure, including nurse-to-patient ratios, intensivist coverage, and electronic medical record integration, has been shown to mitigate weekend-associated delays in escalation of care [23]. Furthermore, increasing center case volume has been independently associated with lower failure-to-rescue rates, with each quartile increase in volume conferring an approximate 13% reduction in rescue failure [24]. In aggregate, these findings suggest high-volume centers possess resilient perioperative systems, remaining effective across temporal variation. Our results reinforce the premise that institutional resources and organizational readiness can buffer weekend care disparities, supporting targeted regionalization and systems-based quality improvement as strategies to mitigate weekend mortality risk following major cardiac operations. While our admirative dataset does not permit direct evaluation of staffing or organizational mechanism underlying such interventions, the results are presented are reflective of broader literature, this providing a direction for future work.

Understanding geographic variation in weekend admission outcomes is critical for contextualizing disparities in access, cost structures, and weekend care delivery. In our analysis, weekend admission was associated with significantly worse outcomes across all U.S. Census regions, however, the magnitude of these disparities varied by region. Prior work has demonstrated substantial regional heterogeneity in cardiac surgical outcomes, with Quin and colleagues reporting risk-adjusted CABG mortality ranging from 1.6% in the Northeast to over 2.2% in the South, accounting for more than 15% of observed mortality variation nationally [25]. Indeed, it is plausible that regional differences in referral patterns, access to high-volume centers, and workforce distribution amplify weekend-associated risk [2], [26]. Notably, we observed weekend admissions to be more frequent in the South yet associated with lower incremental hospitalization costs compared with the Northeast, suggesting divergent regional cost structures and practice patterns rather than uniform inefficiency. Similar findings have been reported in contemporary analyses demonstrating hospital region to be a dominant predictor of cardiac surgical expenditures independent of patient risk [8], [27]. Moreover, regional differences in weekend staffing models and perioperative infrastructure may further compound these disparities [28]. In this context, our findings highlight the need for region-specific quality initiatives and policy strategies to mitigate weekend care disparities while accounting for underlying structural differences in healthcare delivery.

The present study should be interpreted in the context of several limitations inherent to its retrospective design and reliance on administrative data. Identification of weekend admission, operative timing, comorbidities, and outcomes was based on ICD-10 diagnosis and procedure codes, which are subject to variation in coding practices across institutions and may result in misclassification. Although comprehensive risk adjustment was performed, the NIS lacks granular clinical detail, precluding assessment of physiologic severity, operative urgency, intraoperative variables, staffing levels, or postoperative management strategies, possibly influence outcomes. Additionally, causality cannot be inferred, and unmeasured confounding from patient selection, referral patterns, or institutional practices may persist. The NIS is unable to distinguish the pathway by which patients arrive at the operating hospital, such as an outpatient-directed urgent referral versus weekend self-presentation which is not captured by the nature of this database. Multicollinearity amount candidate covariates is partially mitigated by elastic net regularization. Given the administrative nature of this database, the possibility of endogeneity and residual confounding from unmeasured factors cannot be excluded despite double robust adjustment framework. The NIS does not capture surgeon-level factors, longitudinal outcomes beyond the index hospitalization, or interfacility transfers occurring after discharge. Furthermore, regional analyses may also reflect underlying differences in healthcare infrastructure not fully accounted for by available covariates. Future work is needed to determine whether this association reflects deficiencies in weekend care delivery, unmeasured differences in patient acuity at presentation, or both within the entire cohort as well as across various cardiac operation subtypes. Despite these limitations, this study leveraged a large, nationally representative cohort and robust multivariable methods to characterize the association between weekend admission and immediate postoperative outcomes following major cardiac surgery.

In conclusion, weekend admission for major cardiac operations was associated with substantially worse immediate postoperative outcomes and increased resource utilization compared with weekday admission. Our findings highlight the elevated risks of in-hospital mortality, prolonged hospitalization, non-home discharge, and higher costs associated with weekend surgical care. Advances in perioperative management and overall improvements in cardiac surgical outcomes over recent decades may have narrowed disparities in some settings, yet the persisting weekend effect underscores ongoing vulnerabilities in care delivery. Notably, attenuation of risk at higher-volume centers emphasizes the importance of institutional capacity in mitigating adverse outcomes. Multidisciplinary, system-level efforts aimed at optimizing staffing, perioperative resources, and regional care coordination may be critical to reducing weekend-associated inequities and improving outcomes in cardiac surgery. While several findings require extensive support from data sources which provide staffing measures and other perioperative details, the present study aims to function as a contemporary nationally representative analysis to assist the direction of more comprehensive future work.

CRediT authorship contribution statement

Konmal Ali: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Conceptualization. Syed Shaheer Ali: Writing – review & editing, Validation, Methodology, Conceptualization. Troy Coaston: Writing – review & editing, Visualization, Validation. Melissa Justo: Writing – review & editing, Methodology, Investigation. Jaden Rahmani: Writing – review & editing, Investigation. Sidney Hinks: Writing – review & editing, Investigation. Giselle Porter: Writing – review & editing, Investigation. Owen Lai: Writing – review & editing, Investigation. Peyman Benharash: Writing – review & editing, Resources, Investigation.

Ethical approval statement

This study did not require ethical approval as no human subjects were directly involved. Data utilized in this study were derived from the National Inpatient Sample (NIS), which is publicly available upon purchase through the Healthcare Cost and Utilization Project (HCUP).

Funding sources statement

The authors declare that no external funding was received for this work.

Declaration of competing interest

The authors declare that no external funding was received for this work. Dr. Peyman Benharash discloses being a proctor for Atricure. However, this relationship is unrelated to the present study.

Appendix A. Supplementary data

Supplementary Table S1

Covariate balance and weight diagnostics before and after entropy balancing.

Standardized mean differences (SMD) were calculated as (mean_Weekend − mean_Weekday) / pooled SD, using unweighted (before) and entropy-balancing-weighted (after) group moments. The threshold of |SMD| < 0.10 is shown for reference.

Supplementary Table S2

Regional distribution of weekend admission and hospitalization costs by US Census Division.

Values are survey-weighted estimates with 95% confidence intervals. “Share of weekend cohort” denotes the percentage of all weekend admissions nationally originating from each division (i.e., the composition of the Weekend group), corresponding to the values reported in Table 1 and depicted geographically in Fig. 2. Mean hospitalization cost by division corresponds to Fig. 3.