Mortality Rate After Nonelective Hospital Admission

PAPER Mortality Rate After Nonelective Hospital Admission

Rocco Ricciardi, MD, MPH; Patricia L. Roberts, MD; Thomas E. Read, MD; Nancy N. Baxter, MD, PhD; Peter W. Marcello, MD; David J. Schoetz, MD

Objective: We hypothesized that the mortality rate af- patients with nonelective hospital admissions during the ter nonelective hospital admission is higher during week- 5-year study period: 6 842 030 during weekends and ends than weekdays. 23 149 591 during weekdays. Inpatient mortality was reported in 185 856 patients (2.7%) admitted for nonelec- Design: Retrospective cohort analysis. tive indications during weekends and 540 639 (2.3%) during weekdays (P Ͻ .001). The regression revealed Setting: Patients admitted to hospitals in the Nation- significantly higher mortality during weekends for 15 of wide Inpatient Sample, a 20% sample of US community 26 (57.7%) major diagnostic categories. The weekend hospitals. effect remained, and mortality was noted to be 10.5% higher during weekends (odds ratio, 1.10; 95% confi- Patients: We identified all patients with a nonelective dence interval, 1.10-1.11) compared with weekdays af- hospital admission from January 1, 2003, through De- ter adjusting for all other variables with the imputed data cember 31, 2007, in the Nationwide Inpatient Sample. set. Next, we abstracted vital status at discharge and calcu- lated the Charlson comorbidity index score for all pa- Conclusions: These data demonstrate significantly worse tients. We then compared odds of inpatient mortality af- outcomes after nonelective admission during the week- ter nonelective hospital admission during the weekend end compared with weekdays. Although the underlying compared with weekdays, after adjusting for diagnosis, mechanism of this finding is unknown, it is likely that age, sex, race, income level, payer, comorbidity, and hos- factors such as differences in hospital staffing and ser- pital characteristics. vices offered during the weekend compared with weekdays are causal and mutable. Main Outcome Measure: Mortality rate.

Results: Discharge data were available for 29 991 621 Arch Surg. 2011;146(5):545-551

HE HEALTH CARE SYSTEM IN during weekends for several urgent medi- the United States is rap- cal and surgical diagnoses. However, many idly evolving to provide the of these studies6-14 have focused on a single highest-quality care at the diagnosis or set of diagnoses. We hypoth- most reasonable cost. Al- esized that differences in mortality rates thoughT Americans anticipate receiving based on day of admission are present across quality health care based on the most the spectrum of clinical diagnoses. Thus, sound scientific knowledge available,1 we undertook a study to evaluate mortal- many patients do not receive optimal ity rate as a function of admission day across care.2,3 Differences in outcome and qual- a wide range of medical and surgical diag- ity of care have been described for mea- noses for patients admitted to hospitals sures of health and life expectancy in re- within the United States. Author Affiliations: Lahey lation to race, ethnicity, sex, educational Clinic, Department of level, income, geographic location, dis- METHODS Colorectal Surgery, Tufts ability status, and sexual orientation.4 University Medical School, These observations have led us to more Burlington, Massachusetts critical analyses of outcomes to ensure eq- DATA SOURCE (Drs Ricciardi, Roberts, Read, uitable and reliable high-quality care. Marcello, and Schoetz); and We obtained all-payer discharge data from Janu- Department of Surgery and Because marked variability exists in ary 1, 2003, through December 31, 2007, via Li Ka Shing Knowledge health care outcomes among patients in the Nationwide Inpatient Sample (NIS) of the Institute, St. Michael’s Hospital, hospitals, the inpatient setting has been an Healthcare Cost and Utilization Project of the University of Toronto, Toronto, area of relatively intense study.5 Many stud- Agency for Healthcare Research and Quality. Ontario, Canada (Dr Baxter). ies have identified increased mortality rates The NIS—the largest source of all-payer hos-

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 545

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 pital discharge information in the United States—contains data Bureau and categorized as Northeast, Midwest, South, and from approximately 7 million to 8 million hospital stays per West.15 Hospital rurality was categorized as rural or urban based year in 1000 hospitals in 35 states.15 It approximates a 20% strati- on Core Based Statistical Area codes. Before 2004, all metro- fied sample of US community hospitals, including large uni- politan statistical areas were considered urban, and nonmet- versity hospitals and smaller regional facilities. The database ropolitan statistical areas were classified as rural.15 The teach- provides information regarding patient demographics, socio- ing status of the hospital was obtained from the American economic factors, admission profiles, hospital profiles, state Hospital Association Annual Survey of Hospitals. A hospital is codes, discharge diagnoses, procedure codes, total charges, and considered to be a teaching hospital if it has an American Medi- vital status at hospital discharge. Along with other hospital dis- cal Association–approved residency program, is a member of charge databases, the NIS has been used to review trends in sur- the Council of Teaching Hospitals, or has a ratio of full-time gical care and outcomes,16 volume outcome relationships,17 and equivalent interns and residents to beds of 0.25 or higher.15 disparities in care.18 A data use agreement is held by the Agency for Healthcare Research and Quality, and our study was con- VITAL STATUS sidered exempt by the Lahey Clinic Institutional Review Board. The data set permits identification of vital status at the time of PATIENT SELECTION discharge. The variable is coded as died during hospitalization AND PREDICTOR VARIABLE or did not die during hospitalization. Deaths that occurred after discharge are not identifiable from our data set.15 All patients discharged during the time frame sampled were included. We used the elective variable to exclude all patients with STATISTICAL ANALYSIS an admission for elective reasons and included only those patients with nonelective admission.15 Thus, patients with emer- Statistical analyses were performed using SAS statistical soft- gency and urgent indications for admission were included. ware, version 9.2 (SAS Institute Inc, Cary, North Carolina). We The data set permits identification of admission day as a used t tests to analyze continuous variables and ␹2 tests for cat- weekend or weekday. We recorded this variable as admitted egorical variables. Results were considered statistically signifi- during a weekend (ie, Saturday or Sunday) or a weekday (ie, cance at P Ͻ .05, and all statistical tests were 2-tailed. We in- Monday through Friday).15 cluded all covariates in our regression model. The analyses were conducted with and without missing variables. To confirm results, we performed imputation of missing data using the mul- COVARIATES tiple imputation procedure from SAS Institute Inc. Imputa- tion substitutes missing values with plausible values that Our analysis adjusted for the following covariates: age, sex, race, characterize the uncertainty regarding the missing data. This income level, payer, major diagnostic categories (subgroup- process results in valid statistical inferences that properly re- ings of diagnosis-related groups), and the Charlson comorbid- flect the uncertainty due to missing values, for example, con- ity index score. Age was included as a continuous variable. Sex fidence intervals with the correct probability coverage.25 The was entered as a dichotomous variable. Race was divided into imputed data set was then analyzed by using standard logistic white, black, Hispanic, Asian or Pacific Islander, Native Ameri- regression for the complete data. can, or other. Income level was categorized into quartiles per Last, to assess whether the effect of admission day differed estimated median household income of residents in the pa- 15 as a function of diagnosis, we tested for interactions between tient’s zip code. The median income quartiles are classified admission day and major diagnostic category. Because of the as follows: $0 to $38 999, $39 000 to $47 999, $48 000 to 15 significant interaction between these variables, we reanalyzed $62 999, and $63 000 or more. the effect of admission day on mortality rate for each indi- Payer was recorded as follows: Medicare, Medicaid, private vidual major diagnostic category with the same covariates in including health maintenance organization, self-pay, no charge, the larger analysis described herein. or other.15 Major diagnostic categories were used to adjust for diagnoses and reflect larger groupings of diagnostic-related groups made available in the provided data set and download- RESULTS able for review from the US Department of Health and Human 19 Services, Centers for Medicare and Medicaid Services. Major COHORT diagnostic categories have been used to evaluate hospitaliza- 20 21 22 tion risk, mortality risk, and other outcomes. We also evalu- From January 1, 2003, through December 31, 2007, a total ated comorbidity with the Deyo modification of the Charlson comorbidity index.23 Briefly, we ascertained the presence of 17 of 40 095 587 discharges were recorded at NIS hospi- comorbid conditions and then weighted them according to the tals. From this total, admission information was avail- original report. An elevated Charlson comorbidity index score able for 29 991 621 patients (74.8%) who were admit- has been demonstrated to correlate with higher mortality rate.24 ted for nonelective reasons, of whom 726 495 patients We also examined the effect of hospital characteristics on (2.4%) died. mortality rate as a function of admission day. Hospital bed size A total of 6 842 030 patients (22.8%) were admitted categories are based on the number of short-term acute care during the weekend, and 23 149 591 patients (77.2%) beds in a hospital and obtained from the American Hospital As- 15 were admitted during a weekday, providing day-of- sociation Annual Survey of Hospitals. Bed size was classified admission information for 29 991 621 patients. On av- as small, medium, or large, depending on the location of the erage, patients admitted during the weekend were older hospital and its teaching status. The category for ownership and control of the hospital was obtained from the American Hos- and proportionately more likely to be male (Table 1). pital Association Annual Survey of Hospitals and included cat- The NIS included a large number of white patients in the egories for government nonfederal (public), not-for-profit pri- sample; however, proportionately more Asians came in vate (voluntary), and investor-owned private (proprietary).15 during weekends compared with other NIS-designated The census region for the hospital is defined by the US Census racial groups. Income categories were more evenly dis-

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 546

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 Table 1. Demographics of Patients Admitted to the Hospital for Nonelective Indications on a Weekday Compared With the Weekenda

Admission, No. (%)

Demographic Weekend Weekday All Patients, No. Age, mean (SE), y 47.3 (0.01) 46.0 (0.01) 46.3 (0.01) Sex Male 3 010 288 (23.1) 10 015 902 (76.9) 13 026 190 Female 3 809 421 (22.6) 13 065 024 (77.4) 16 874 445 Missing 22 321 (24.5) 68 665 (75.5) 90 986 Race White 3 220 315 (22.5) 11 097 824 (77.5) 14 318 139 Black 770 629 (23.2) 2 550 673 (76.8) 3 321 302 Hispanic 739 973 (23.0) 2 482 078 (77.0) 3 222 051 Asian or Pacific Islander 139 084 (23.6) 451 503 (76.4) 590 587 Native American 28 909 (23.3) 95 010 (76.7) 123 919 Other 160 541 (22.6) 551 156 (77.4) 711 697 Missing 1 782 579 (23.1) 5 921 347 (76.9) 7 703 939 Annual income, $ 0-38 999 1 974 533 (22.9) 6 666 323 (77.1) 8 640 856 39 000-47 999 1 716 207 (22.8) 5 801 677 (77.2) 7 517 884 48 000-62 999 1 565 569 (22.8) 5 291 453 (77.2) 6 857 022 Ն63 000 1 411 489 (22.7) 4 808 723 (77.3) 6 220 212 Missing 174 232 (23.1) 581 415 (76.9) 775 647 Payer Medicare 2 642 010 (23.4) 8 637 881 (76.6) 11 279 891 Medicaid 1 377 211 (22.5) 4 754 107 (77.5) 6 131 318 Private 2 140 024 (21.9) 7 612 939 (78.1) 9 752 963 Self-pay 431 099 (25.3) 1 271 137 (74.7) 1 702 236 None (no reported payer) 38 027 (24.3) 118 438 (75.7) 156 465 Other 201 554 (22.0) 715 176 (78.0) 916 730 Missing 12 105 (23.3) 39 913 (19.1) 52 018 Charlson comorbidity index score 0 3 643 749 (22.4) 12 627 933 (77.6) 16 271 682 1 1 402 080 (23.5) 4 569 704 (76.5) 5 971 784 Ն2 1 796 201 (23.2) 5 951 954 (76.8) 7 748 155 Missing 0 0 0 Mean (SE) 1.08 (0.001) 1.07 (0.001) 1.07 (0.001) Hospital size Small 867 117 (22.6) 2 965 832 (77.4) 3 832 949 Medium 1 751 846 (23.0) 5 857 844 (77.0) 7 609 690 Large 4 217 349 (22.8) 14 309 014 (77.2) 18 526 363 Missing 5718 (25.3) 16 901 (74.7) 22 619 Hospital control Government 4 052 649 (22.8) 13 720 395 (77.2) 17 773 044 Public 485 156 (22.8) 1 643 991 (77.2) 2 129 147 Not-for-profit private 1 343 449 (23.2) 4 458 160 (76.8) 5 801 609 Investor-owned private 690 375 (22.2) 2 423 092 (77.8) 3 113 467 Other private 264 683 (3.9) 887 052 (3.8) 1 151 735 Missing 5718 (25.3) 16 901 (74.7) 22 619 Hospital region Northeast 1 345 943 (22.5) 4 634 264 (77.5) 5 980 207 Midwest 1 474 302 (22.9) 4 962 317 (77.1) 6 436 619 South 2 647 705 (22.5) 9 101 920 (77.5) 11 749 625 West 1 374 080 (23.6) 4 451 090 (76.4) 5 825 170 Missing 0 0 0 Hospital rurality Rural 888 514 (22.7) 3 018 446 (77.3) 3 906 960 Urban 5 947 798 (22.8) 20 114 244 (77.2) 26 062 042 Missing 5718 (25.3) 16 901 (74.7) 22 619 Hospital teaching status Nonteaching 3 803 611 (22.9) 12 810 765 (77.1) 16 614 376 Teaching 3 032 701 (22.7) 10 321 925 (77.3) 13 354 626 Missing 5718 (25.3) 16 901 (74.7) 22 619

a All data are presented as number (percentage) of patients unless otherwise indicated. Missing variables were as follows: day of admission, 31 (0.0001%); age, 34 429 (0.1%); PϽ.001 for all entries.

tributed, but patients living in the lowest-annual- care coverage from Medicare, Medicaid, or private in- income areas were proportionately most likely to seek surance. Of interest, those patients categorized as self- treatment during the weekend. Most patients had health pay were proportionately more likely to be admitted dur-

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 547

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 Table 2. MDCs of Patients Admitted to the Hospital for Nonelective Indications on a Weekday Compared With the Weekenda

Mortality, % Multivariate Analyses Univariate MDC Weekend WeekdayP Value OR (95% CI) P Value Pre-MDC 4.4 3.5 .09 0.90 (0.60-1.32) .60 Nervous system 5.0 4.5 Ͻ.001 1.14 (1.12-1.16) Ͻ.001 Eyes 0.2 0.2 .60 0.85 (0.47-1.53) .60 Otorhinolaryngology 0.5 0.6 .002 0.95 (0.83-1.07) .40 Respiratory 5.7 5.4 Ͻ.001 1.08 (1.07-1.10) Ͻ.001 Circulatory 3.0 2.6 Ͻ.001 1.14 (1.12-1.16) Ͻ.001 Digestive 2.1 2.0 Ͻ.001 1.09 (1.07-1.12) Ͻ.001 Hepatobiliary 3.5 3.4 .20 1.04 (1.01-1.08) .008 Musculoskeletal 1.3 1.3 .60 1.05 (1.01-1.09) .03 Skin or breast 0.7 0.8 Ͻ.001 0.97 (0.90-1.05) .50 Endocrine system 1.6 1.6 .10 1.04 (1.00-1.09) .05 Kidneys 2.6 2.6 Ͼ.99 1.01 (0.98-1.04) .60 Male reproductive systemb 1.8 1.7 .20 1.12 (0.94-1.32)a .20 Female reproductive systemb 1.3 0.9 Ͻ.001 1.32 (1.16-1.51)a Ͻ.001 Pregnancy and childbirth 0.02 0.01 .001 1.37 (1.08-1.75)a .01 Neonatal 0.5 0.4 Ͻ.001 1.25 (1.20-1.30) Ͻ.001 Immunologic 1.5 1.4 .04 1.19 (1.10-1.28) Ͻ.001 Myeloproliferative 12.1 7.8 Ͻ.001 1.50 (1.43-1.58) Ͻ.001 Infectious 12.6 11.9 Ͻ.001 1.07 (1.05-1.09) Ͻ.001 Mental health 0.08 0.09 .03 0.83 (0.68-1.00) .05 Alcohol or drug use 0.1 0.1 .80 0.91 (0.70-1.17) .50 Poison 1.4 1.3 .009 1.12 (1.05-1.19) Ͻ.001 Burns 3.7 3.5 .40 1.17 (0.97-1.41) .10 Other health factors 1.5 1.2 Ͻ.001 1.20 (1.08-1.34) Ͻ.001 Trauma 9.3 9.9 .003 0.95 (0.90-1.01) .10 HIV 7.2 7.0 .40 1.05 (0.98-1.12) .20

Abbreviations: CI, confidence interval; HIV, human immunodeficiency virus; MDC, major diagnostic category; OR, odds ratio. a Multivariable analysis adjusted for age, sex, race, annual income, payer, comorbidity, hospital bed size, hospital control, hospital region, hospital rurality, and hospital teaching status. Univariate results are reported after ␹2 comparisons of proportions who died across admission day. The multivariate results include ORs and 95% CIs for weekend compared with weekday admission for individual regression analyses of each MDC with adjustment for age, sex, race, annual income, payer, and Charlson comorbidity index score. b No adjustment for sex in the MDCs of reproductive system health or childbirth.

ing the weekend. The Charlson comorbidity index scores tation of missing variables resulted in no significant ranged from 0 to 20, and the mean (SE) Charlson co- change in the response variables with the complete data morbidity index score was 1.07 (0.001). Patients who set. The weekend effect remained, and mortality was came in during weekends had a higher comorbidity score noted to be 10.3% higher during weekends (odds ratio, than patients admitted during weekdays (Table 1). Also, 1.10; 95% confidence interval, 1.10-1.11) compared slight differences existed in the types of hospitals to which with weekdays after adjusting for all other variables with patients were admitted on weekends compared with week- the imputed data set. days (Table 1). Tests for interaction revealed that the effect of admission day on mortality rate was significantly altered by UNIVARIATE ANALYSIS major diagnostic category. We reanalyzed the effect of admission day on mortality rate, adjusting for age, sex, Inpatient mortality was documented in 185 856 pa- race, payer, annual income level, and comorbidity for tients (2.7%) admitted during a weekend compared with each major diagnostic category. The regression revealed 540 639 patients (2.3%) admitted during a weekday significantly higher mortality during weekends: 12 of 26 (16.2% higher; P Ͻ .001). For 17 of 26 major diagnostic major diagnostic categories (46.2%) by univariate analy- categories (65.4%), higher weekend mortality was docu- sis and 15 of 26 major diagnostic categories (57.7%) by mented (Table 2) by univariate analysis. multivariate analysis (Table 2). The highest odds ratios for weekend mortality were identified for myeloprolif- MULTIVARIATE ANALYSIS erative disorders (odds ratio, 1.50; 95% confidence interval, 1.43-1.58), pregnancy and childbirth (1.37; After adjusting for age, sex, race, payer, annual income 1.08-1.75), and female reproductive system procedures level, comorbidity, hospital bed size, hospital control, (1.32; 1.16-1.51). Mortality rate was not statistically dif- hospital region, hospital rurality, hospital teaching sta- ferent during weekends and weekdays for 10 major tus, and major diagnostic categories, we found that the diagnostic categories and it was lower during weekends odds of death with admission during a weekend were for 1 major diagnostic category (mental health disor- 10.5% higher than during a weekday (Table 3). Impu- ders) (Table 2).

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 548

Variable OR (95% CI) P Value Admission during weekend 1.10 (1.10-1.11) .001 Increasing age 1.04 (1.04-1.04) .001 Sex Female 1 [Reference] Male 1.19 (1.19-1.20) Ͻ.001 Race White 1 [Reference] Black 0.99 (0.98-1.00) .02 Hispanic 0.89 (0.88-0.90) Ͻ.001 Asian or Pacific Islander 1.06 (1.04-1.08) Ͻ.001 Native American 0.97 (0.93-1.01) .20 Other 1.09 (1.07-1.11) Ͻ.001 Annual income, $ 0-38 999 1 [Reference] 39 000-47 999 0.98 (0.97-0.98) Ͻ.001 48 000-62 999 0.94 (0.93-0.95) Ͻ.001 Ն63 000 0.92 (0.91-0.93) Ͻ.001 Payer Private 1 [Reference] Medicare 1.01 (1.00-1.02) .06 Medicaid 1.27 (1.25-1.28) Ͻ.001 Self-pay 1.45 (1.42-1.47) Ͻ.001 None (no reported payer) 0.83 (0.78-0.87) Ͻ.001 Other 1.37 (1.34-1.40) Ͻ.001 Increasing Charlson comorbidity index score 1.23 (1.23-1.23) Ͻ.001 Hospital size Small 1 [Reference] Medium 1.07 (1.06-1.09) Ͻ.001 Large 1.14 (1.13-1.15) Ͻ.001 Hospital control Government 1.10 (1.08-1.13) Ͻ.001 Public 1.03 (1.01-1.06) .005 Not-for-profit private 1.01 (0.98-1.03) .60 Investor-owned private 0.97 (0.95-0.99) .01 Other private 1 [Reference] Hospital region Northeast 1 [Reference] Midwest 0.86 (0.86-0.87) Ͻ.001 South 1.01 (1.01-1.02) .002 West 1.04 (1.03-1.05) Ͻ.001 Hospital rurality Rural 1 [Reference] Urban 1.07 (1.06-1.08) Ͻ.001 Hospital teaching status Nonteaching 1 [Reference] Teaching 1.10 (1.09-1.11) Ͻ.001 Major diagnostic category Nervous system 1 [Reference] Eyes 0.07 (0.06-0.09) Ͻ.001 Otorhinolaryngology 0.21 (0.20-0.23) Ͻ.001 Respiratory 1.20 (1.19-1.21) Ͻ.001 Circulatory 0.54 (0.53-0.54) Ͻ.001 Digestive 0.47 (0.46-0.48) Ͻ.001 Hepatobiliary 0.62 (0.61-0.63) Ͻ.001 Musculoskeletal 0.31 (0.30-0.32) Ͻ.001 Skin or breast 0.24 (0.23-0.24) Ͻ.001 Endocrine system 0.39 (0.38-0.40) Ͻ.001 Kidneys 0.55 (0.54-0.56) Ͻ.001 Male reproductive system 0.29 (0.27-0.31) Ͻ.001 Female reproductive system 0.39 (0.37-0.41) Ͻ.001 Pregnancy and childbirth 0.02 (0.02-0.03) Ͻ.001 Neonatal 1.34 (0.31-0.37) Ͻ.001 Immunologic 0.37 (0.36-0.38) Ͻ.001 Myeloproliferative 1.51 (1.48-1.55) Ͻ.001 Infectious 3.10 (3.06-3.13) Ͻ.001 Mental health 0.05 (0.04-0.05) Ͻ.001 Alcohol or drug use 0.07 (0.07-0.08) Ͻ.001 Poison 0.53 (0.51-0.55) Ͻ.001 Burns 1.85 (1.70-2.01) Ͻ.001 Other health factors 0.26 (0.25-0.27) Ͻ.001 Trauma 5.34 (5.18-5.51) Ͻ.001 HIV 1.07 (1.04-1.11) Ͻ.001 Pre–major diagnostic category 1.68 (1.42-1.98) Ͻ.001

Abbreviations: CI, confidence interval; HIV, human immunodeficiency virus; OR, odds ratio. a Age and Charlson comorbidity index score were entered as continuous variables. Sex, race, income, payer, and major diagnostic categories were entered as categorical variables with referent values identified.

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 549

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 COMMENT outcomes and subsequent reduction in weekend admission mortality rates. Another possible cause of increased weekend mortal- In our study, using national all-payer discharge data from ity rate is low staffing levels and reduced staffing experi- the United States, we examined mortality rate in more ence during weekends. Staff who work weekends tend to than 29 million patients admitted nonelectively during have less experience and are often responsible for more a 5-year period. We identified a significantly higher mor- patients than staff employed during weekdays.6,27 This sce- tality rate for patients admitted during the weekend com- nario is particularly true for junior physicians and resi- pared with weekdays. This mortality rate difference re- dent trainees; unfortunately, studies evaluating outcome mained despite adjustment for age, sex, race, payer, in relation to physician staffing are few. Far more data ex- associated medical comorbidities, and hospital charac- ist that evaluate the role of nurse staffing regarding out- teristics. In addition, we identified mortality rate differ- comes.28,29 Much of these data demonstrate worse out- ences across most major diagnostic categories. These data comes with fewer nurses or reduced nurse staffing hours. are particularly concerning because the reported differ- For example, in a study30 from 210 adult general hospi- ences in mortality rate are noted across several key areas tals in Pennsylvania, the authors found that each addi- of health care and throughout the nation. tional patient per nurse was associated with a 7% in- Differences in mortality rate based on day of admis- crease in the likelihood of dying within 30 days of sion have similarly been identified in smaller studies and admission. Given these data, some researchers and poli- for more limited sets of urgent care diagnoses. One of cymakers have recommended mandatory staffing level leg- the largest and most inclusive studies6 from Ontario, islation as a solution.31 However, at present, an analysis Canada, revealed significantly higher in-hospital mor- documenting an improved outcome with increased staff- tality rates for patients admitted during weekends for 23 ing levels is not available, and a determination of the role of 100 leading causes of death. The differences in mor- of nurse staffing regarding weekend mortality rate has not tality rate were most pronounced in patients with emer- been conducted. gent conditions, such as ruptured abdominal aortic an- Our study is large and population based but it may eurysm, acute epiglottitis, and pulmonary embolism. Of be limited by information and misclassification bias given interest, no differences in mortality rate were observed the administrative data used for analyses. However, our for patients with myocardial infarction, intracerebral hem- selected outcome (mortality rate), our covariates, and ad- orrhage, or acute hip fracture. Other authors7-14 have dem- mission day are unlikely to be improperly abstracted from onstrated excess weekend mortality rates in patients with the medical record. The fact that the differences in mor- myocardial infarction, stroke, pulmonary embolism, gas- tality rate were identified across multiple medical diag- trointestinal bleeding, cardiac arrest, and other indi- noses reduces the potential importance of diagnostic mis- vidual diagnoses. Our study demonstrated an excess week- classification. However, it is possible that patients admitted end mortality rate for nonelective admissions across many during the weekend have more comorbidities or poten- major diagnostic categories and at the national level. tially more severe illnesses than those admitted during a An explanation for the differences in mortality rate is weekday. We have adjusted for this possibility by evalu- not immediately evident from our data. Health care out- ating comorbidity, but an assessment of disease severity comes, such as morbidity and mortality rate, are depen- at presentation is not possible with the available data. dent on patient comorbidities, structural elements of care, In conclusion, our data reveal a significantly increased and processes of care. Although our study demon- mortality rate for patients admitted during weekends strated an attenuation of the mortality rate differences across demographic groups for medical and surgical di- with adjustment for patient comorbidity, mortality rates agnoses. The consistency of the data across multiple di- remained higher during weekends despite adjustment for agnostic-related groups, patient demographics, comor- the Charlson comorbidity index score. Thus, the admis- bidities, and hospital characteristics indicates that a sion day outcome differences implicate a common struc- central and common factor is most likely responsible for tural or process measure. This theory is substantiated by the unfavorable outcomes. Given the comparatively the lack of a significant difference in admission mortal- similar weekend outcomes for those patients with disor- ity rate for trauma or burn care. The evaluation and man- ders treated under the direction of standard algorithms, agement of trauma and burns incorporate structured al- such as trauma and burns, our data raise serious con- gorithms for care that likely reduced much of the cerns regarding the adequacy of health care during week- variability in care practice that may be appreciated with ends for patients with many other diagnoses. An analysis other conditions. In addition, many patients who re- of potential causative factors is needed to identify modi- quire care for trauma or burns present during the night fiable components of care. Quality improvement strate- and/or weekends; thus, clinical services for these condi- gies can then be developed and implemented to stan- tions have been refined to account for these presenta- dardize care across admission day. tion patterns. Although speculative, clinical trial evidence of an outcome benefit for advanced trauma life Accepted for Publication: November 16, 2010. support is unavailable; yet, evidence exists that trauma Correspondence: Rocco Ricciardi, MD, MPH, Lahey educational initiatives improve hospital staff knowl- Clinic, Department of Colorectal Surgery, Tufts Univer- edge of available emergency interventions.26 It is un- sity Medical School, 41 Mall Rd, Burlington, MA 01805 clear whether similar standardization of medical care in ([email protected]). other major diagnostic categories may lead to improved Author Contributions: Study concept and design: Ric-

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 550

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 ciardi. Acquisition of data: Ricciardi, Roberts, and Baxter. for upper gastrointestinal hemorrhage: a nationwide analysis. Clin Gastroen- Analysis and interpretation of data: Ricciardi, Roberts, Read, terol Hepatol. 2009;7(3):296-302. 13. Shaheen AAM, Kaplan GG, Myers RP. Weekend versus weekday admission and Baxter, Marcello, and Schoetz. Drafting of the manu- mortality from gastrointestinal hemorrhage caused by peptic ulcer disease. Clin script: Ricciardi, Roberts, Read, Baxter, Marcello, and Gastroenterol Hepatol. 2009;7(3):303-310. Schoetz. Critical revision of the manuscript for important 14. Peberdy MA, Ornato JP, Larkin GL, et al; National Registry of Cardiopulmonary intellectual content: Ricciardi, Roberts, Read, Baxter, Resuscitation Investigators. Survival from in-hospital cardiac arrest during nights Marcello, and Schoetz. Statistical analysis: Ricciardi and and weekends. JAMA. 2008;299(7):785-792. 15. Healthcare Cost and Utilization Project (HCUP). Overview of the Nationwide In- Baxter. Administrative, technical, and material support: Ric- patient Sample; July 2010. Agency for Healthcare Research and Quality Web site. ciardi, Roberts, Read, Baxter, Marcello, and Schoetz. Study www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed November 1, 2010. supervision: Ricciardi. 16. Dimick JB, Wainess RM, Cowan JA, Upchurch GR Jr, Knol JA, Colletti LM. Financial Disclosure: None reported. National trends in the use and outcomes of hepatic resection. J Am Coll Surg. 2004;199(1):31-38. Previous Presentation: Presented at the New England Sur- 17. Ricciardi R, Virnig BA, Ogilvie JW Jr, Dahlberg PS, Selker HP, Baxter NN. Volume- gical Society Meeting; October 29, 2010; Saratoga Springs, outcome relationship for coronary artery bypass grafting in an era of decreasing New York. volume. Arch Surg. 2008;143(4):338-344. 18. Ricciardi R, Selker HP, Baxter NN, Marcello PW, Roberts PL, Virnig BA. Dispar- ate use of minimally invasive surgery in benign surgical conditions. Surg Endosc. REFERENCES 2008;22(9):1977-1986. 19. US Department of Health and Human Services, Centers for Medicare and Medicaid 1. Committee on Quality of Health Care in America, Institute of Medicine. Crossing Services. Details for Fiscal Year 2008 Major Diagnostic Categories File. Centers for the Quality Chasm: A New Health System for the Twenty-First Century. Wash- Medicare and Medicaid Services Web site. http://www.cms.gov/acuteinpatientpps ington, DC: National Academy Press; 2001. /ffd/ItemDetail.asp?ItemID=CMS1201739. Accessed October 1, 2010. 2. Soumerai SB, McLaughlin TJ, Spiegelman D, Hertzmark E, Thibault G, Goldman 20. Muennig P, Jia H, Khan K, Pallin DJ. Ascertaining variation in hospitalization risk L. Adverse outcomes of underuse of ␤-blockers in elderly survivors of acute myo- among immigrants using small area analysis. Prev Med. 2006;43(2):145-149. cardial infarction. JAMA. 1997;277(2):115-121. 21. Durairaj L, Will JG, Torner JC, Doebbeling BN. Prognostic factors for mortality 3. Meehan TP, Fine MJ, Krumholz HM, et al. Quality of care, process, and out- following interhospital transfers to the medical intensive care unit of a tertiary comes in elderly patients with pneumonia. JAMA. 1997;278(23):2080-2084. referral center. Crit Care Med. 2003;31(7):1981-1986. 4. US Department of Health and Human Services. Executive Summary: Goal 2: Elimi- 22. Brameld K, Holman D, Moorin R. Possession of health insurance in Australia— nate Health Disparities; Healthy People 2010. Healthy People Web site. http: how does it affect hospital use and outcomes? J Health Serv Res Policy. 2006; //www.healthypeople.gov/2010/data/midcourse/html/execsummary/goal2 11(2):94-100. .htm. Accessed March 10, 2011. 23. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with 5. Wennberg JE, Freeman JL, Shelton RM, Bubolz TA. Hospital use and mortality ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. among Medicare beneficiaries in Boston and New Haven. N Engl J Med. 1989; 24. Pompei P, Charlson ME, Ales K, MacKenzie CR, Norton M. Relating patient char- 321(17):1168-1173. acteristics at the time of admission to outcomes of hospitalization. J Clin Epidemiol. 6. Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on week- 1991;44(10):1063-1069. ends as compared with weekdays [published correction appears in N Engl J Med. 25. The Missing Imputation Procedure. SAS Institute Inc. SAS Institute Inc Web site. 2001;345(9):663-668]. N Engl J Med. 2001;345(9):663-668. http://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer 7. Clarke, Wills MS, Bowman R-A, et al. Exploratory study of the ‘weekend effect’ .htm#mitoc.htm. Accessed April 11, 2011. for acute medical admissions to public hospitals in Queensland, Australia. RV J 26. Jayaraman S, Sethi D. Advanced trauma life support training for hospital staff. Intern Med. 2010;40(1):777-783. Cochrane Database Syst Rev. 2009;(2):CD004173. 8. Crowley RW, Yeoh HK, Stukenborg GJ, Medel R, Kassell NF, Dumont AS. Influ- 27. McKee M, Black N. Does the current use of junior doctors in the United King- ence of weekend hospital admission on short-term mortality after intracerebral dom affect the quality of medical care? Soc Sci Med. 1992;34(5):549-558. hemorrhage. Stroke. 2009;40(7):2387-2392. 28. Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K. Nurse-staffing lev- 9. Kostis WJ, Demissie K, Marcella SW, Shao Y-H, Wilson AC, Moreyra AE; Myo- els and the quality of care in hospitals. N Engl J Med. 2002;346(22):1715- cardial Infarction Data Acquisition System (MIDAS 10) Study Group. Weekend 1722. versus weekday admission and mortality from myocardial infarction. N Engl J 29. Estabrooks CA, Midodzi WK, Cummings GG, Ricker KL, Giovannetti P. The im- Med. 2007;356(11):1099-1109. pact of hospital nursing characteristics on 30-day mortality. Nurs Res. 2005; 10. Aujesky D, Jime´nez D, Mor MK, Geng M, Fine MJ, Ibrahim SA. Weekend versus 54(2):74-84. weekday admission and mortality after acute pulmonary embolism. Circulation. 30. Aiken LH, Clarke SP, Sloane DM, Sochalski J, Silber JH. Hospital nurse staffing 2009;119(7):962-968. and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002; 11. Saposnik G, Baibergenova A, Bayer N, Hachinski V. Weekends: a dangerous time 288(16):1987-1993. for having a stroke? Stroke. 2007;38(4):1211-1215. 31. Becker DJ. Do hospitals provide lower quality care on weekends? Health Serv 12. Ananthakrishnan AN, McGinley EL, Saeian K. Outcomes of weekend admissions Res. 2007;42(4):1589-1612.

ARCH SURG/ VOL 146 (NO. 5), MAY 2011 WWW.ARCHSURG.COM 551