0
Original Research: CRITICAL CARE MEDICINE |

Association Between Time of Admission to the ICU and Mortality: A Systematic Review and Metaanalysis FREE TO VIEW

Rodrigo Cavallazzi, MD; Paul E. Marik, MD, FCCP; Amyn Hirani, MD; Monvasi Pachinburavan, MD; Tajender S. Vasu, MD; Benjamin E. Leiby, PhD
Author and Funding Information

From the Division of Pulmonary and Critical Care Medicine (Drs Cavallazzi, Hirani, Pachinburavan, and Vasu) and the Division of Biostatistics, Jefferson Medical College (Dr Leiby), Thomas Jefferson University, Philadelphia, PA; and the Division of Pulmonary and Critical Care Medicine (Dr Marik), Eastern Virginia Medical School, Norfolk, VA.

Correspondence to: Paul E. Marik, MD, FCCP, 825 Fairfax Ave, Ste 410, Norfolk, VA 23507; e-mail: marikpe@evms.edu


Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/site/misc/reprints.xhtml).


© 2010 American College of Chest Physicians


Chest. 2010;138(1):68-75. doi:10.1378/chest.09-3018
Text Size: A A A
Published online

Background:  The organizational and staffing structure of an ICU influences the outcome of critically ill and injured patients. A change in the ICU staffing structure frequently occurs at nighttime and on weekends (off-hours). We postulated that patients who are admitted to an ICU during off hours may be at an increased risk of death.

Methods:  We performed a systematic review of the literature to assess whether admission to an ICU during off-hours is associated with an increased mortality. We selected studies that evaluated the association between time of admission to the ICU and mortality, with adjustment for severity of disease. We excluded studies that included pediatric and non-ICU patients. Study characteristics extracted included date of publication, study design, country where study was done, study population, time factor (weekend or night shift), severity adjustment tool, and outcome.

Results:  Ten cohort studies met our inclusion criteria; eight of these studies evaluated nighttime admissions, whereas six studies evaluated weekend admissions. The pooled analysis demonstrated that nighttime admission was not associated with an increased mortality (odds ratio [OR], 1.0 [95% CI, 0.87-1.17]; P = .956); however, patients admitted over the weekend had a significant increase in the adjusted risk of death (OR, 1.08 [95% CI, 1.04-1.13]; P < .001). Significant heterogeneity was found in the studies that evaluated nighttime admissions.

Conclusions:  Whereas patients admitted to an ICU over the weekend appear to be at an increased risk of death, nighttime admissions were not associated with an increased mortality. The lower level of staffing and intensity of care provided by many hospitals over the weekend may account for this finding. The heterogeneity noted between studies evaluating nighttime admissions likely reflects the diverse organizational structure of the hospitals and ICUs where these studies were carried out.

Figures in this Article

Patients in the ICU often have several organ-system abnormalities and physiologic disarrangements that require highly complex and timely diagnostic tests and therapies. The appropriate timeliness of care and availability of complex tests and therapy are even more imperative in the first hours after admission to an ICU, when patients are more likely to be unstable and require vigorous resuscitative measures. Although admission of patients who are unstable to ICUs occurs 24 h a day, not all units maintain the same level of staffing during off-hours, that is, during nighttime, weekends, and holidays. This raises concern in view of evidence showing that the organizational structure of an ICU influences the outcome of patients who are critically ill. A number of studies have demonstrated that units with high-intensity physician staffing (mandatory intensivist consultation or closed ICU) are associated with reduced mortality when compared with units with low-intensity physician staffing (no intensivist or elective intensivist consultation).1 However, a recent, large retrospective study found higher odds of death in patients who are critically ill and fully treated by critical care physicians as compared with those who are not.2 The absence of daily rounds by an ICU physician is associated with a threefold increase in mortality,3 and increased nurse staffing is associated with lower hospital mortality and adverse patient events.4

Patients who are admitted to ICUs during off-hours may face worse outcomes. This hypothesis has been evaluated by several cohort studies. These studies, however, have not shown uniform results. Whereas some showed an increased risk of death for patients admitted during off-hours,5,6 others showed a surprising protective effect.7,8 If patients admitted during off-hours do indeed have a worse outcome, this would have important implications for intensivists and health-care administrators involved in ICU staffing as well as health-care policy makers and insurance providers. We, therefore, performed a systematic review to evaluate the available evidence on the association between off-hour admissions (nighttime and weekends) and mortality for patients in ICUs.

Search Strategy

Two investigators independently searched the Medline and Embase databases through September 2009 using a combination of terms that included “critical care,” “time,” and “admission,” with no restriction to date or language. We also reviewed the bibliographies of all selected articles for other potential relevant articles.

Study Selection

Although the definition of “off-hours” differed among the studies, we use the term “off-hours” to refer to either nighttime or weekend admissions. Thus, studies were eligible to be included in this review if they evaluated the association between being admitted to the ICU during either nighttime or weekends (or both) and mortality, with adjustment for severity of disease. We excluded studies in which the setting was not an ICU (eg, hospital admissions or ED admissions) and ones that evaluated a pediatric population. We classified studies according to the time factor evaluated: nighttime, weekend, or both. When nighttime was the time factor evaluated, the reference group was composed of patients admitted during daytime. When weekend was the time factor evaluated, the reference group was composed of patients admitted during weekdays and their mortality was compared with the group of patients admitted during the weekend.

Reporting Quality

To evaluate the reporting quality of the articles included in this review, we used variables as suggested by the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.9 We assessed whether there was a description of the following items: objectives, setting, eligibility criteria, variables, number of participants, patient characteristics, and outcome data.

Data Extraction

Two investigators (R. C., A. H.) independently abstracted the data using a standardized form. Disagreements were resolved by discussion between all the reviewers. Study characteristics extracted included author name, date of publication, study design (eg, cohort vs case control), country where the study was carried out, study population (eg, adult vs pediatric), time factor evaluated (eg, weekend or nighttime), severity adjustment tool, and outcome. We also abstracted reporting quality items as outlined above. Numerical data extracted from each individual study included total number of patients in the study, number of patients in each group, mortality in each group, and the adjusted relative risk estimate (odds ratio [OR]) for mortality with corresponding 95% CI.

Statistical Analysis

We performed the data analysis using Stata 10 (StataCorp LP; College Station, TX) software. We combined individual study data (adjusted OR and 95% CI), performed the metaanalysis by means of a DerSimonian and Laird random effects model,10 and report the pooled OR for mortality with corresponding 95% CI. For the purpose of this metaanalysis, the numerator of the OR was the odds of the group with the time factor in consideration (either nighttime or weekend admission). If a study reported the OR and CI with the odds of the group with the time factor in consideration in the denominator, we took the inverse (1 divided by the estimate) of both the OR and the CI to make the combination of data consistent. We performed subgroup analysis with stratification according to the presence of an intensivist on-site during off-hours. We performed a χ2 test to evaluate the presence of heterogeneity among the studies. We considered a P value < .1 as indicative of heterogeneity. For all other analyses, we considered a P value < .05 statistically significant. We tested the inter-rater agreement of the reporting quality with κ statistic.

We explored the presence of publication bias by performing a linear regression of the standardized effect estimates against their precision.11 If the intercept deviates significantly from zero, there is evidence of bias.

The Medline and Embase searches retrieved 796 and 173 citations, respectively. After screening the titles and abstracts from these citations and the bibliographic references from relevant articles, we selected 21 studies for full review.5-8,12-26 Although our initial search was not restricted to any specific language, only English-language articles were included in the final analysis.22 Although Arabi et al20 evaluated weekend or nighttime admissions, their study reports standardized mortality ratios and was therefore excluded. Arslankoylu et al24 did not provide a risk estimate; this study was also excluded. Three studies were excluded because they included pediatric populations.7,18,21 Four studies were excluded because the setting was not an ICU.25-28 A further study was excluded because the time factor in consideration was the “ICU-rounding” period.23 A study, included in the nighttime admissions metaanalysis, was not included in the weekend admissions metaanalysis because it compared mortality by individual day of admission using Wednesday as the reference.16

Characteristics of the Included Studies

Figure 1 shows details of study identification and inclusion and exclusion criteria. Ten studies met the inclusion criteria for this metaanalysis (Table 1).5,6,8,12-17,19 Eight studies reported in-hospital mortality,5-8,12-14,16,17 and two reported ICU mortality.15,19 The number of ICUs in each study varied markedly, with three single-ICU studies8,14,19 and one study with 102 ICUs.16 Four studies were performed in North America,5,6,8,12 four in Europe,13,15-17 and two in Asia.14,19 The time factor evaluated included weekends in six studies,5,6,12,14,15,19 night shifts in eight studies,6,8,13-17,19 and weekends plus night shifts in two studies.13,14

Figure Jump LinkFigure 1. Flow chart with the number of studies evaluated at each stage of the systematic review. OR = odds ratio.Grahic Jump Location
Table Graphic Jump Location
Table 1 —Characteristics of the Included Studies

APACHE = Acute Physiology and Chronic Health Evaluation; NS = night shift; PR-GCS = postresuscitation Glasgow Coma Scale; SAPS = Simplified Acute Physiology Score; TISS = Therapeutic Intervention Scoring System; UK = United Kingdom; US = United States.

a 

Adjustment used: age score, APACHE II, severe medical history, surgical status, and primary reason for admission.

Studies showed variation in the organizational structure during off-hours. When we evaluated studies for the presence of intensivists during nighttime, we were able to establish that in two studies intensivists were present on-site at nighttime,13,19 in five studies they were not present,6,8,14,15,17 and this information was not available in one study.16 During weekends, an intensivist was present 24 h per day in one study,19 part of the day in four studies,6,12,14,15 and this information was not available in one study.5 In four studies, either resident physicians or fellows were present during off-hours despite the absence of an on-site intensivist at all times.6,8,12,14 In the study by Luyt et al,13 which included 23 closed ICUs in a metropolitan area, a board-certified or in-training intensivist (with > 52 on-site night shifts of experience) was always available on-site during off-hours. For the purpose of this review, we classified the study of Luyt et al13 as having an intensivist present during off-hours.

Reporting Quality

Two reviewers independently assessed the reporting quality of the studies and had an inter-rater agreement of 95% (κ, 0.69). The proportion of items clearly described by the studies was the following: patient characteristics, 80%; organizational structure, 80%; outcome definition, 100%; variables, 90%; number of participants, 100%; eligibility criteria, 100%; study setting, 100%; and study objectives, 100%.

Nighttime Admission and Mortality

Eight studies with a total of 135,220 patients evaluated daytime/nighttime admissions (Table 2).6,8,13-17,19 The number of admissions and percentage of deaths were 73,676 and 21.4%, respectively, for daytime admissions and 61,544 and 20.8%, respectively, for nighttime admissions. There was no difference in the adjusted odds of death between patients admitted during nighttime as compared with those admitted during daytime (OR, 1.0 [95% CI, 0.87-1.17]; P = .956) (Fig 2). There was significant heterogeneity among the studies (P < .001).

Table Graphic Jump Location
Table 2 —Number of Patients and Death Percentage per Group, Odds Ratio With 95% CI, and Organizational Features of Studies that Evaluated Nighttime Admissions

NA = information not available; OR = odds ratio.

a 

Fellow was present.

b 

The reference group schedule was from 8:00 am to 5:59 pm; there was a third group, not included in our analysis, with a schedule from 6:00 pm to 11:59 pm.

c 

Resident physician was present.

d 

During off-hours, intensivists were on duty, but they were not routinely present.

Figure Jump LinkFigure 2. Adjusted effect of nighttime admission to the ICU on mortality. Weight is the relative contribution of each study to the overall OR (random effects model with 95% CI). See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

We performed a subgroup analysis in which we stratified studies according to the presence of an intensivist physician on-site during off-hours, with the following OR for death in patients admitted during off-hours according to study stratification: no intensivist6,8,14,15,17 (OR, 1.05 [95% CI, 0.79-1.41]; P = .73), heterogeneity P < .001; intensivist13,19 (OR, 0.93 [95% CI, 0.87-0.99]; P = .02), heterogeneity P = .558; and information not available16 (OR, 1.02 [95% CI, 0.96-1.09]; P = .44).

Weekend Admission and Mortality

Six studies with a total of 180,600 patients evaluated weekday/weekend admissions (Table 3).5,6,12,14,15,19 The number of patients admitted to the ICU and the percentage of deaths were 133,150 and 11.1%, respectively, for weekday admissions and 47,450 and 15.6%, respectively, for weekend admissions. The adjusted odds of death were significantly higher in patients admitted over the weekend as compared with patients admitted during weekdays (OR, 1.08 [95% CI, 1.04-1.13]; P < .001) (Fig 3). There was no significant heterogeneity between studies (P = .585).

Table Graphic Jump Location
Table 3 —Number of Patients and Death Percentage per Group, OR with 95% CI, and Organizational Features of Studies that Evaluated Weekend Admissions

See Table 2 for expansion of abbreviations.

a 

One ICU out of 18 had intensivist coverage 7 d per week.

b 

Fellow.

c 

Resident physician.

Figure Jump LinkFigure 3. Adjusted effect of weekend admission to the ICU on mortality. Weight is the relative contribution of each study to the overall OR (random effects model with 95% CI). See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

A subgroup analysis in which we stratified studies according to the presence of an intensivist physician on-site over the weekend found the following OR for death: no intensivist6,12,14,15 (OR, 1.07 [95% CI, 0.99-1.15]; P = .081), heterogeneity P = .319; intensivist19 (OR, 1.03; [95% CI, 0.61-1.73]; P = .916); and information not available5 (OR, 1.09; [95%, CI, 1.04-1.15]; P = .001).

We performed a sensitivity analysis by excluding the study of Barnett et al,5 which contributed to 64% of the pooled OR. After excluding this study, the risk of death remained higher in patients admitted over the weekend (OR, 1.07 [95% CI, 1.0-1.14]; P = .05). There was no significant heterogeneity between studies (P = .473).

Publication Bias Analysis

Linear regression of the standard normal deviate against precision showed that the intercepts did not significantly deviate from zero (P = .96 for nighttime studies; P = .48 for weekend studies). Thus, there was no evidence of publication bias according to Egger test.

The main finding of this metaanalysis is that patients admitted to an ICU over the weekend have an 8% higher risk of death, after adjusting for severity of illness, as compared with patients admitted during weekdays. Accounting for sampling variability, this increased risk could be as low as 4% or as high as 13%. This finding should, however, be viewed with caution because the metaanalysis was dominated by one study.5 Furthermore, the control group in that study included admissions only during midweek (Tuesday, Wednesday, and Thursday). It is unclear whether the increased risk would remain had the authors used admissions throughout all weekdays as a control group. Nighttime admissions were not associated with a change in the risk of mortality; however, the pooled analysis of individual study results showed significant heterogeneity.

A change in the organizational/staffing structure of an ICU and hospital over the weekend likely explains the increased mortality noted. It is likely that a number of factors in combination, including a decreased physician-to-patient ratio, unavailability of board-certified intensivists, physician fatigue, and difficulty in obtaining complex diagnostic tests or therapies, account for the higher risk of death over the weekend.12,14,15

What can be done to mitigate the increased mortality for ICU admissions over the weekends? While answering this question requires an understanding of the factors underlying the increased risk, it is reasonable to presume that efforts to make the organizational structure of ICUs (and hospitals) homogeneous during weekdays and weekends will likely have a favorable impact on patient outcomes. Such efforts may include not only maintaining the same number of health-care professionals and ensuring the same level of staffing seniority throughout the week (including weekends) in ICUs, but also guaranteeing the appropriate timeliness of tests and therapies during weekends. It seems improbable that ICUs will be able to set up such important organizational changes without adequate leadership. Worryingly, Kahn et al29 surveyed 72 ICUs in 2003 and found that more than one-half of them had no identifiable ICU director in charge.

The Leapfrog Group ICU standard suggests that intensivists should be present in the ICU during daytime hours, but they are not required to be present during off-hours provided that pages are returned at least 95% of the time within 5 min and arrangements are made for a qualified health-care professional to reach the patient within 5 min.30 A subgroup analysis of studies that evaluated nighttime admission showed that the absence of an intensivist during off-hours is not significantly associated with a worse outcome. However, our analysis was based on pooled results of only five studies. Studies have shown conflicting results with respect to the impact of intensivists on the outcome of patients who are critically ill. Whereas a recent retrospective study showed higher odds of hospital mortality in patients who are critically ill and treated by critical care physicians compared with those who were not,2 a large body of literature shows that higher intensivist staffing is associated with improved outcome.1 In fact, two observational, single-ICU studies showed that the implementation of 24-h intensivist coverage was associated with a significantly improved standardized mortality ratio.31,32 A third study found a decrease in ICU complications and adjusted length of stay after setting 24-h intensivist coverage in the ICU, but there was no significant change in mortality.33

It is not entirely clear why weekend admissions were associated with an increased risk of dying whereas this association was not present for nighttime admissions. The greater availability of clinicians and support staff during the week as well as the shorter “downtime” could account for this finding. The studies included in this review are based on the premise that patients admitted during off-hours are unstable and more susceptible to the deleterious effects of substandard care in the early period after admission to the ICU; however, those admitted during a weekday remain critically ill and are as likely to receive substandard care during off-hours. Thus, it is possible that the deleterious consequences of being admitted during off-hours may have been underestimated by the current studies because of a dilutional effect in the analysis.5,17

The association between off-hour admission and mortality has been evaluated in settings other than the ICU. Bell and Redelmeier25 compared the risk of in-hospital mortality of Canadian patients admitted from the ED on weekends with those admitted during the week. Patients with serious medical conditions, such as abdominal aortic aneurysm, acute epiglottitis, and pulmonary embolism, had a higher risk of death if they were admitted on a weekend. Other authors have noted an increased risk of death for weekend admissions in patients with intracerebral hemorrhage,34 but not in patients with burns.27 Similarly, trauma patients are more likely to die if they are admitted to the ICU between midnight and 6:00 am.35 Earlier studies36,37 showed an increase in neonatal mortality for infants born on the weekend, but a more recent study did not find a statistically significant increase in the risk of death after adjusting for birth weight.38

Further studies are necessary to determine to what extent a change in the organizational structure of an ICU during off-hours may affect outcome. This may be feasible if future studies evaluating the outcome of patients admitted to ICUs during off-hours provide a more detailed description of the organizational changes during off-hours. For instance, it is important to determine whether the presence of housestaff (residents or fellows) in teaching hospitals is sufficient to ensure appropriate care during off-hours. In community hospitals, it is equally important to establish whether outcomes are substantially different if nonintensivist physicians, nurse practitioners, or physician assistants provide on-site care during off-hours. Further studies should also establish the magnitude of organizational changes in hospitals and ICUs during off-hours. As an example, while it is known that increased nursing staffing is associated with better outcomes,4 it is important to properly establish how nursing staffing changes occur in relation to off-hours.

Our systematic review has a number of limitations. Most importantly, different definitions were used for both nighttime and weekend hours, and the description of the organizational change was not well described in all studies. In addition, different severity adjustment tools were used, and it is possible that residual confounding may have occurred in some of these analyses. For example, Ensminger et al12 did not show an increase in the adjusted mortality of patients admitted over the weekend; however, had the authors not adjusted their analysis for the intensity of treatment and the ICU admission source, they would have demonstrated an increased mortality in those admitted over the weekend despite adjustment for severity of disease.

In conclusion, our systematic review suggests that patients admitted to the ICU over the weekend have a higher risk of dying than those admitted on weekdays. Nighttime admission to the ICU did not appear to be associated with an increased mortality; however, there was significant heterogeneity between studies, which likely reflects the diverse organizational structure in the ICUs where these studies were carried out. Additional studies are required to determine the need for the presence of on-site intensivists during off-hours.

Author contributions:Dr Cavallazzi: contributed to the literature search, data extraction, statistics, and writing of the manuscript.

Dr Marik: contributed to the study concept and design, and writing and final editing of the manuscript.

Dr Hirani: contributed to the literature search, data extraction, and review of the manuscript.

Dr Pachinburavan: contributed to the data review and review of the manuscript.

Dr Vasu: contributed to the data review and review of the manuscript.

Dr Leiby: contributed to the statistics and review of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Pronovost PJ, Angus DC, Dorman T, Robinson KA, Dremsizov TT, Young TL. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA. 2002;28817:2151-2162. [CrossRef] [PubMed]
 
Levy MM, Rapoport J, Lemeshow S, Chalfin DB, Phillips G, Danis M. Association between critical care physician management and patient mortality in the intensive care unit. Ann Intern Med. 2008;14811:801-809. [PubMed]
 
Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA. 1999;28114:1310-1317. [CrossRef] [PubMed]
 
Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care. 2007;4512:1195-1204. [CrossRef] [PubMed]
 
Barnett MJ, Kaboli PJ, Sirio CA, Rosenthal GE. Day of the week of intensive care admission and patient outcomes: a multisite regional evaluation. Med Care. 2002;406:530-539. [CrossRef] [PubMed]
 
Laupland KB, Shahpori R, Kirkpatrick AW, Stelfox HT. Hospital mortality among adults admitted to and discharged from intensive care on weekends and evenings. J Crit Care. 2008;233:317-324. [CrossRef] [PubMed]
 
Numa A, Williams G, Awad J, Duffy B. After-hours admissions are not associated with increased risk-adjusted mortality in pediatric intensive care. Intensive Care Med. 2008;341:148-151. [CrossRef] [PubMed]
 
Morales IJ, Peters SG, Afessa B. Hospital mortality rate and length of stay in patients admitted at night to the intensive care unit. Crit Care Med. 2003;313:858-863. [CrossRef] [PubMed]
 
von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. STROBE Initiative STROBE Initiative The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;1478:573-577. [PubMed]
 
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;73:177-188. [CrossRef] [PubMed]
 
Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;3157109:629-634. [CrossRef] [PubMed]
 
Ensminger SA, Morales IJ, Peters SG, et al. The hospital mortality of patients admitted to the ICU on weekends. Chest. 2004;1264:1292-1298. [CrossRef] [PubMed]
 
Luyt CE, Combes A, Aegerter P, et al. Mortality among patients admitted to intensive care units during weekday day shifts compared with “off” hours. Crit Care Med. 2007;351:3-11. [CrossRef] [PubMed]
 
Sheu CC, Tsai JR, Hung JY, et al. Admission time and outcomes of patients in a medical intensive care unit. Kaohsiung J Med Sci. 2007;238:395-404. [CrossRef] [PubMed]
 
Uusaro A, Kari A, Ruokonen E. The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Med. 2003;2912:2144-2148. [CrossRef] [PubMed]
 
Wunsch H, Mapstone J, Brady T, Hanks R, Rowan K. Hospital mortality associated with day and time of admission to intensive care units. Intensive Care Med. 2004;305:895-901. [CrossRef] [PubMed]
 
Meynaar IA, van der Spoel JI, Rommes JH, van Spreuwel-Verheijen M, Bosman RJ, Spronk PE. Off hour admission to an intensivist-led ICU is not associated with increased mortality. Crit Care. 2009;133:R84. [CrossRef] [PubMed]
 
Arias Y, Taylor DS, Marcin JP. Association between evening admissions and higher mortality rates in the pediatric intensive care unit. Pediatrics. 2004;1136:e530-e534. [CrossRef] [PubMed]
 
Lee KK, Ng I, Ang BT. Outcome of severe head injured patients admitted to intensive care during weekday shifts compared to nights and weekends. Ann Acad Med Singapore. 2008;375:390-396. [PubMed]
 
Arabi Y, Alshimemeri A, Taher S. Weekend and weeknight admissions have the same outcome of weekday admissions to an intensive care unit with onsite intensivist coverage. Crit Care Med. 2006;343:605-611. [CrossRef] [PubMed]
 
Hixson ED, Davis S, Morris S, Harrison AM. Do weekends or evenings matter in a pediatric intensive care unit? Pediatr Crit Care Med. 2005;65:523-530. [CrossRef] [PubMed]
 
Weng L, Meng YL, Lu MS, Du B. Association between hospital mortality and day of week of intensive care unit admission [in Chinese]. Zhonghua Yi Xue Za Zhi. 2006;8626:1841-1844. [PubMed]
 
Afessa B, Gajic O, Morales IJ, Keegan MT, Peters SG, Hubmayr RD. Association between ICU admission during morning rounds and mortality. Chest. 2009;1366:1489-1495. [CrossRef] [PubMed]
 
Arslankoylu AE, Bayrakci B, Oymak Y. Admission time and mortality rates. Indian J Pediatr. 2008;757:691-694. [CrossRef] [PubMed]
 
Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med. 2001;3459:663-668. [CrossRef] [PubMed]
 
Schmulewitz L, Proudfoot A, Bell D. The impact of weekends on outcome for emergency patients. Clin Med. 2005;56:621-625. [PubMed]
 
Taira BR, Meng H, Goodman MS, Singer AJ. Does “off-hours” admission affect burn patient outcome? Burns. 2009;358:1092-1096. [CrossRef] [PubMed]
 
Carmody IC, Romero J, Velmahos GC. Day for night: should we staff a trauma center like a nightclub? Am Surg. 2002;6812:1048-1051. [PubMed]
 
Kahn JM, Matthews FA, Angus DC, Barnato AE, Rubenfeld GD. Barriers to implementing the Leapfrog Group recommendations for intensivist physician staffing: a survey of intensive care unit directors. J Crit Care. 2007;222:97-103. [CrossRef] [PubMed]
 
The Leapfrog GroupThe Leapfrog Group ICU physician staffing. http://www.leapfroggroup.org/media/file/FactSheet_IPS.pdf.
 
Blunt MC, Burchett KR. Out-of-hours consultant cover and case-mix-adjusted mortality in intensive care. Lancet. 2000;3569231:735-736. [CrossRef] [PubMed]
 
Goh AY, Lum LC, Abdel-Latif ME. Impact of 24 hour critical care physician staffing on case-mix adjusted mortality in paediatric intensive care. Lancet. 2001;3579254:445-446. [CrossRef] [PubMed]
 
Gajic O, Afessa B, Hanson AC, et al. Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Crit Care Med. 2008;361:36-44. [CrossRef] [PubMed]
 
Crowley RW, Yeoh HK, Stukenborg GJ, Medel R, Kassell NF, Dumont AS. Influence of weekend hospital admission on short-term mortality after intracerebral hemorrhage. Stroke. 2009;407:2387-2392. [CrossRef] [PubMed]
 
Vaziri K, Roland JC, Robinson L, Fakhry SM. Optimizing physician staffing and resource allocation: sine-wave variation in hourly trauma admission volume. J Trauma. 2007;623:610-614. [CrossRef] [PubMed]
 
Mathers CD. Births and perinatal deaths in Australia: variations by day of week. J Epidemiol Community Health. 1983;371:57-62. [CrossRef] [PubMed]
 
MacFarlane A. Variations in number of births and perinatal mortality by day of week in England and Wales. BMJ. 1978;26153:1670-1673. [CrossRef] [PubMed]
 
Gould JB, Qin C, Marks AR, Chavez G. Neonatal mortality in weekend vs weekday births. JAMA. 2003;28922:2958-2962. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. Flow chart with the number of studies evaluated at each stage of the systematic review. OR = odds ratio.Grahic Jump Location
Figure Jump LinkFigure 2. Adjusted effect of nighttime admission to the ICU on mortality. Weight is the relative contribution of each study to the overall OR (random effects model with 95% CI). See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 3. Adjusted effect of weekend admission to the ICU on mortality. Weight is the relative contribution of each study to the overall OR (random effects model with 95% CI). See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Characteristics of the Included Studies

APACHE = Acute Physiology and Chronic Health Evaluation; NS = night shift; PR-GCS = postresuscitation Glasgow Coma Scale; SAPS = Simplified Acute Physiology Score; TISS = Therapeutic Intervention Scoring System; UK = United Kingdom; US = United States.

a 

Adjustment used: age score, APACHE II, severe medical history, surgical status, and primary reason for admission.

Table Graphic Jump Location
Table 2 —Number of Patients and Death Percentage per Group, Odds Ratio With 95% CI, and Organizational Features of Studies that Evaluated Nighttime Admissions

NA = information not available; OR = odds ratio.

a 

Fellow was present.

b 

The reference group schedule was from 8:00 am to 5:59 pm; there was a third group, not included in our analysis, with a schedule from 6:00 pm to 11:59 pm.

c 

Resident physician was present.

d 

During off-hours, intensivists were on duty, but they were not routinely present.

Table Graphic Jump Location
Table 3 —Number of Patients and Death Percentage per Group, OR with 95% CI, and Organizational Features of Studies that Evaluated Weekend Admissions

See Table 2 for expansion of abbreviations.

a 

One ICU out of 18 had intensivist coverage 7 d per week.

b 

Fellow.

c 

Resident physician.

References

Pronovost PJ, Angus DC, Dorman T, Robinson KA, Dremsizov TT, Young TL. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA. 2002;28817:2151-2162. [CrossRef] [PubMed]
 
Levy MM, Rapoport J, Lemeshow S, Chalfin DB, Phillips G, Danis M. Association between critical care physician management and patient mortality in the intensive care unit. Ann Intern Med. 2008;14811:801-809. [PubMed]
 
Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA. 1999;28114:1310-1317. [CrossRef] [PubMed]
 
Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care. 2007;4512:1195-1204. [CrossRef] [PubMed]
 
Barnett MJ, Kaboli PJ, Sirio CA, Rosenthal GE. Day of the week of intensive care admission and patient outcomes: a multisite regional evaluation. Med Care. 2002;406:530-539. [CrossRef] [PubMed]
 
Laupland KB, Shahpori R, Kirkpatrick AW, Stelfox HT. Hospital mortality among adults admitted to and discharged from intensive care on weekends and evenings. J Crit Care. 2008;233:317-324. [CrossRef] [PubMed]
 
Numa A, Williams G, Awad J, Duffy B. After-hours admissions are not associated with increased risk-adjusted mortality in pediatric intensive care. Intensive Care Med. 2008;341:148-151. [CrossRef] [PubMed]
 
Morales IJ, Peters SG, Afessa B. Hospital mortality rate and length of stay in patients admitted at night to the intensive care unit. Crit Care Med. 2003;313:858-863. [CrossRef] [PubMed]
 
von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. STROBE Initiative STROBE Initiative The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;1478:573-577. [PubMed]
 
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;73:177-188. [CrossRef] [PubMed]
 
Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;3157109:629-634. [CrossRef] [PubMed]
 
Ensminger SA, Morales IJ, Peters SG, et al. The hospital mortality of patients admitted to the ICU on weekends. Chest. 2004;1264:1292-1298. [CrossRef] [PubMed]
 
Luyt CE, Combes A, Aegerter P, et al. Mortality among patients admitted to intensive care units during weekday day shifts compared with “off” hours. Crit Care Med. 2007;351:3-11. [CrossRef] [PubMed]
 
Sheu CC, Tsai JR, Hung JY, et al. Admission time and outcomes of patients in a medical intensive care unit. Kaohsiung J Med Sci. 2007;238:395-404. [CrossRef] [PubMed]
 
Uusaro A, Kari A, Ruokonen E. The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Med. 2003;2912:2144-2148. [CrossRef] [PubMed]
 
Wunsch H, Mapstone J, Brady T, Hanks R, Rowan K. Hospital mortality associated with day and time of admission to intensive care units. Intensive Care Med. 2004;305:895-901. [CrossRef] [PubMed]
 
Meynaar IA, van der Spoel JI, Rommes JH, van Spreuwel-Verheijen M, Bosman RJ, Spronk PE. Off hour admission to an intensivist-led ICU is not associated with increased mortality. Crit Care. 2009;133:R84. [CrossRef] [PubMed]
 
Arias Y, Taylor DS, Marcin JP. Association between evening admissions and higher mortality rates in the pediatric intensive care unit. Pediatrics. 2004;1136:e530-e534. [CrossRef] [PubMed]
 
Lee KK, Ng I, Ang BT. Outcome of severe head injured patients admitted to intensive care during weekday shifts compared to nights and weekends. Ann Acad Med Singapore. 2008;375:390-396. [PubMed]
 
Arabi Y, Alshimemeri A, Taher S. Weekend and weeknight admissions have the same outcome of weekday admissions to an intensive care unit with onsite intensivist coverage. Crit Care Med. 2006;343:605-611. [CrossRef] [PubMed]
 
Hixson ED, Davis S, Morris S, Harrison AM. Do weekends or evenings matter in a pediatric intensive care unit? Pediatr Crit Care Med. 2005;65:523-530. [CrossRef] [PubMed]
 
Weng L, Meng YL, Lu MS, Du B. Association between hospital mortality and day of week of intensive care unit admission [in Chinese]. Zhonghua Yi Xue Za Zhi. 2006;8626:1841-1844. [PubMed]
 
Afessa B, Gajic O, Morales IJ, Keegan MT, Peters SG, Hubmayr RD. Association between ICU admission during morning rounds and mortality. Chest. 2009;1366:1489-1495. [CrossRef] [PubMed]
 
Arslankoylu AE, Bayrakci B, Oymak Y. Admission time and mortality rates. Indian J Pediatr. 2008;757:691-694. [CrossRef] [PubMed]
 
Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med. 2001;3459:663-668. [CrossRef] [PubMed]
 
Schmulewitz L, Proudfoot A, Bell D. The impact of weekends on outcome for emergency patients. Clin Med. 2005;56:621-625. [PubMed]
 
Taira BR, Meng H, Goodman MS, Singer AJ. Does “off-hours” admission affect burn patient outcome? Burns. 2009;358:1092-1096. [CrossRef] [PubMed]
 
Carmody IC, Romero J, Velmahos GC. Day for night: should we staff a trauma center like a nightclub? Am Surg. 2002;6812:1048-1051. [PubMed]
 
Kahn JM, Matthews FA, Angus DC, Barnato AE, Rubenfeld GD. Barriers to implementing the Leapfrog Group recommendations for intensivist physician staffing: a survey of intensive care unit directors. J Crit Care. 2007;222:97-103. [CrossRef] [PubMed]
 
The Leapfrog GroupThe Leapfrog Group ICU physician staffing. http://www.leapfroggroup.org/media/file/FactSheet_IPS.pdf.
 
Blunt MC, Burchett KR. Out-of-hours consultant cover and case-mix-adjusted mortality in intensive care. Lancet. 2000;3569231:735-736. [CrossRef] [PubMed]
 
Goh AY, Lum LC, Abdel-Latif ME. Impact of 24 hour critical care physician staffing on case-mix adjusted mortality in paediatric intensive care. Lancet. 2001;3579254:445-446. [CrossRef] [PubMed]
 
Gajic O, Afessa B, Hanson AC, et al. Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Crit Care Med. 2008;361:36-44. [CrossRef] [PubMed]
 
Crowley RW, Yeoh HK, Stukenborg GJ, Medel R, Kassell NF, Dumont AS. Influence of weekend hospital admission on short-term mortality after intracerebral hemorrhage. Stroke. 2009;407:2387-2392. [CrossRef] [PubMed]
 
Vaziri K, Roland JC, Robinson L, Fakhry SM. Optimizing physician staffing and resource allocation: sine-wave variation in hourly trauma admission volume. J Trauma. 2007;623:610-614. [CrossRef] [PubMed]
 
Mathers CD. Births and perinatal deaths in Australia: variations by day of week. J Epidemiol Community Health. 1983;371:57-62. [CrossRef] [PubMed]
 
MacFarlane A. Variations in number of births and perinatal mortality by day of week in England and Wales. BMJ. 1978;26153:1670-1673. [CrossRef] [PubMed]
 
Gould JB, Qin C, Marks AR, Chavez G. Neonatal mortality in weekend vs weekday births. JAMA. 2003;28922:2958-2962. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543