Does the time of admission to ITU affect mortality?
*Magdalena A. Wujtewicz, Aleksandra Suszyńska-Mosiewicz, Wioletta Sawicka, Arkadiusz Piankowski, Anna Dylczyk-Sommer, Radosław Owczuk, Maria Wujtewicz
Department of Anaesthesiology and Intensive Therapy, Medical University of Gdansk
- Table 1. Mortality in patients admitted to ITU during daytime and at night
- Table 2. Characteristics of groups of patients admitted at daytime and at night
- Table 3. Mortality among patients admitted to ITU on weekdays and at weekends
- Table 4. Characteristics of groups of patients admitted on weekdays and at weekends
Background. Among many factors that may affect mortality among ITU patients, the time of admission has been reported to play some, but ill-defined role. In the retrospective study, we analysed the time of admission, severity of the underlying disease, clinical status on admission and mortality among adult patients treated in a single ITU over a six-year period.
Methods. We compared the mortality of patients who were admitted during daytime (07:00 to 18:59) and at night (19:00 to 06:59). We also compared those admitted on weekdays (Monday 07:00 to Friday 18:59) to those admitted during weekends (Friday 19:00 to Monday 06:59). The patient’s condition was assessed using the APACHE II scale. Brain dead organ donors and re-admissions were excluded from the analysis.
Results. The prospective study involved the data of 1789 patients. Mortality was higher in patients who were admitted during the night and during weekends, when compared to daytime and weekdays, respectively. Mortality was also higher in patients admitted directly from the operating theatre after emergency surgery, but only during nights and weekends.
The following independent factors in ITU mortality have been identified: length of ITU stay (OR 1.015; 95% CI 1.005-1.024), admission from a hospital ward (OR 1.39; 95% CI 1.04-1.86) and APACHE II score (OR 1.177; 95% CI 1.156-1.198).
Conclusion. Time of admission has not been identified as a single independent factor in ITU mortality, but admissions at night and during weekends were associated with higher mortality, probably because of emergency conditions.
Under ideal conditions, the identical 24-hour standard of treatment in the intensive therapy unit (ITU) should be provided. The therapeutic management implemented within the first hours after ITU admission may be crucial for treatment outcomes , yet it is not always the case. To analyse the hospital mortality, many authors considered various factors associated with admissions to ITU, including the season [2, 3], month [4, 5], weekends  or time of admission [7, 8]. However, the differences in study methods, as well as in organisation and specificity of ITUs in individual countries do not allow referring directly their study results to the Polish conditions.
The aim of the present study was to analyse the mortality in the intensive therapy unit according to one factor, i.e. the time of ITU admission.
The retrospective analysis, approved by the Bioethics Committee, Medical University of Gdańsk, involved electronic and paper-based medical records of patients treated in the ITU in the years 2001-2006. According to the APACHE II score, the following patients were excluded: those below the age of 16 years, after cardiac surgeries, with burns, admitted directly from another ITU, re-admitted to ITU and treated for less than 8 h. Moreover, brain dead organ donors were not included.
Two times of admissions were distinguished - day and night, the former from 7 a.m. to 6.59 p.m. and the latter between 7 p.m. and 6.59 a.m. Moreover, we compared admissions on weekdays (Monday 7 a.m. through Friday 6.59 p.m.) and at weekends (Friday 7 p.m. through Monday 6.59 a.m.).
The number of specialists working in ITU was the same during 24 h; the number of nurses per patient changed depending on the number of patients treated.
The relations between ITU mortality and ITU admission during daytime vs night and during weekdays vs weekends were analysed.
Patients were divided into groups according to the time of ITU admission – group D (daytime), group N (at night), group Wd (during weekdays) and group We (during weekends). Groups D vs N and Wd vs We were compared.
The results were statistically analysed. Normal distribution of continuous variables was verified using the Shapiro-Wilk`s W test. Depending on the type of data, the Mann-Whitney U or χ2 test were used for intergroup comparisons. Death risk factors were identified with logistic regression and the quasi-Newton test was applied for estimation. P<0.05 was considered significant.
The total number of patients treated in ITU was 1947; only first ITU admissions were analysed. The medical records of 26 patients were excluded, as the APACHE II scale could not be used. According to the APACHE II score, the following patients were excluded: 6 <16 years of age, 52 – transferred directly from another ITU, 75 – treated shorter than 8 h and 7 – admitted for procedures to determine brain death. Some patients belonged to more than one category; therefore, the number of exclusions and inclusions did not equal 1947.
One hundred patients required re-admissions, including four admitted three times and one – four times. In this group, 35 died in ITU. Finally, the study group consisted of 1789 patients.
During the first ITU stay, 404 patients died, which constitutes 22.6% of the study population. The further 202 (11.2%) died before discharge from hospital (including 35 during the next ITU stay). In total, 606 patients died in hospital, which consisted 33.7% of those admitted to ITU during the study period. Amongst 1385 patients discharged from ITU, 14.6% died.
Day-night admissions: 1205 patients were admitted during the day and 584 at night. The groups differed in ITU and hospital mortality (which included the death in ITU and death after ITU discharge). A significantly higher mortality was observed in patients admitted at night, yet no differences were demonstrated in the number of deaths in other wards after ITU discharge (Table 1).
Moreover, there were no differences according to gender, duration of ITU treatment and places from which the patients were admitted to ITU (operating room/ward). At nights, patients after emergency procedures were significantly more often admitted whereas during daytime – those after scheduled procedures. The groups differed in age ( significantly higher age in the day group), APACHE II score, predicted death rate (PDR) and predicted death rate adjusted (PDRA) – significantly higher values were found in the night group (Table 2).
Weekday-weekend admissions: during weekdays, 1372 patients were admitted to ITU whereas during weekends – 417. The groups differed in ITU mortality, mortality after ITU discharge, and hospital mortality. A significantly higher mortality was found in patients admitted during weekends (Table 3). No gender-related differences were demonstrated. Significant differences between the groups were observed in relation to the place from which the patients were admitted to ITU (operating room/ward) – during weekdays, patients were more frequently admitted from the operating room. Amongst the patients admitted from the operating room during weekends, more patients were after emergency procedures. The groups differed significantly for age (higher age in the weekday group), duration of ITU treatment (longer in weekend patients), APACHE II score, PDR and PDRA (higher values in weekend patients) (Table 4).
Based on multi-variable logistic regression, the following independent risk factors of ITU death were identified: duration of stay (OR 1.015; 95% CI 1.005-1.024), admission from another hospital ward (OR 1.39; 95% CI 1.04-1.86) and APACHE II score (OR 1.177; 95% CI 1.156-1.198).
Higher ITU mortality was observed in patients admitted at nights and during weekends, compared to days and weekdays. The general hospital mortality was found similar. Multi-variable logistic regression, which refers to the total study population (without division into groups) did not reveal correlations between ITU death and the time of ITU admission.
As expected, the groups of patients with higher mortality had higher APACHE II score, which was an independent ITU death risk factor, higher predictable death risk and predictable death risk adjusted (for the cause of ITU admission).
Kuijsten and co-workers  demonstrated that the hospital death risk was higher in patients during working hours and weekends. Likewise, in our study, mortality amongst patients admitted during weekends was higher than in those admitted during weekdays; however, the weekend admission was not an independent risk factor. Additionally, according to the authors mentioned, the ITU admissions on Friday were associated with higher risk of death, yet their definition of daytime was different from ours; the main criterion defining day hours was the availability of intensive care specialists , which improved the outcomes of ITU treatment. In our study, the specialist in anesthesiology and intensive therapy was available in the unit over the period of 24 h.
The predictable risk of death according to APACHE II score, type of admission and patient’s age (but not the fact of admission after day working hours) are considered independent factors of hospital mortality . According to our findings, also the time of admission was not an independent risk factor.
Furthermore, some reports demonstrated that mortality in patients admitted during weekdays and weekends was the same, which is contrary to our findings. Admissions after working hours (4 p.m.-8 a.m.) were associated with lower mortality. Logistic regression showed, however, that weekend admissions were associated with higher mortality; once SAPS II and TISS-28 scores were considered, admissions after working hours were not found to be associated with lower mortality . Our results also confirm the reports on higher mortality amongst patients admitted during weekends ; however, according to some other studies , this pattern concerns patients treated due to selected diseases being the cause of ITU admission. The literature review reveals higher death risk of patients admitted to ITU during weekends; no such a relation was observed for admissions at nights .
The analysis of admissions of over 56 000 patients to 102 British ITUs did not disclose the relations between the time of admission or admission on a specific weekday and hospital mortality of patients treated in ITU (29.4%) . In our study, mortality amongst patients discharged from ITU was 11.2% and was the same , comparable  or higher  than that in other reports. It seems, however, that differences in the course of treatment of patients discharged from ITUs in different centres hinder explicit comparisons of results.
Various aspects of work organisation and medical management are likely to affect mortality in ITU patients. For instance, in some countries, residents start their work in teaching hospitals in July (the ‘July phenomenon’ – USA), hence, supposedly higher mortality and longer stay of patients admitted to ITU in this month. However, the studies undertaken did not confirm this relation [4, 5].
The comparison of our findings with those reported in literature is extremely difficult. On the one hand, the Polish reports are lacking; on the other hand, studies from other countries regard units of various profiles, working in varied organizational forms of health care system, which was stressed by almost all authors. Moreover, the limitation of our study is its retrospective character; the data of some, small group of patients were not complete due to lack of medical records or lack of access to complete records.
1. In the total population of patients treated in the intensive therapy unit, the time of admission was not found to be an independent risk factor of death in the unit. The factors increasing the death risk include duration of ITU stay, admission from another hospital ward and APACHE II score.
2. Admissions during nights and weekends are associated with higher mortality compared to admissions during days and weekdays.
3. The groups characterized by higher ITU mortality had higher APACHE II scores and higher predictable risk of death.
1. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M, Early Goal-Directed Therapy Collaborative Group: Early goal directed therapy in the treatment of severe sepsis and sepsis shock. N Engl J Med 2001; 345: 1368-1377.
2. Reinkainen M, Uusaro A, Ruokonen E, Niskanen M: Excess mortality in winter in Finnish intensive care. Acta Anaesthesiol Scand 2006; 50:706-711.
3. Harrison DA, Lertsithichai P, Brady AR, Carpenter JR, Rowan K: Winter excess mortality in intensive care in the UK: an analysis of outcome adjusted for patients case mix and unit workload. Intensive Care Med 2004; 30: 1900-1907.
4. Finkielman JD, Morales IJ, Peters SG, Keegan MT, Ensminger SA, Lymp JF, Afessa B: Mortality lenght of stay of patients admitted to the intensive care unit in July. Crit Care Med 2004; 32: 1161-1165.
5. Barry WA, Rosenthal GE: The effect of July admission on intensive care mortality and lenght of stay in teaching hospitals. J Gen Intern Med 2003; 18: 639-645.
6. Ensminger SA, Morales IJ, Peters SG, Keegan MT, Finkelman JD, Lymp JF, Afessa B: The hospital mortality of patients admitted to the ICU on weekends. Chest 2004; 126: 1292-1298.
7. 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; 30: 895-901.
8. Kuijsten HAJM, Brinkman S, Meynaar IA, Spronk PE, van der Spoel JI, Bosman RJ, de Keizer NF, Abu-Hanna A, de Lange D: Hospital mortality is associated with ICU admission time. Intensive Care Med 2010; 36: 1765-1771.
9. Knaus WA, Draper EA, Wagner DP, Zimmerman JE: Apache II: a severity of disease classification system. Crit Care Med 1985; 13: 818-829.
10. Blunt MC, Burchett KR: Out-of-hours consultant cover and case-mix-adjusted mortality in intensive care. The Lancet 2000; 356: 735-736.
11. 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; 13: R84.
12. Uusaro A, Kari A, Ruokonen E: The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Med 2003; 29: 2144-2148.
13. Bell CM, Redelmeier DA: Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med 2001; 345: 663-668.
14. Cavallazzi R, Marik PE, Hirani A, Pachinburavan M, Vasu TS, Leiby BE: Association between time of admission to the ICU and mortality: a systematic review and metaanalysis. Chest 2010; 138: 68-75.
15. Campbell AJ, Cook JA, Adey G, Cuthbertson BH: Predicting death and readmission after intensive care discharge. Br J Anaesth 2008; 100: 656-662.
16. Azoulay E, Adrie C, de Lassence A, Pochard F, Moreau D, Thiery G, Cheval C, Moine P, Garrouste-Orgeas M, Alberti C, Cohen Y, Timsit JF: Determinants of postintensive care unit mortality: a prospective multicenter study. Crit Care Med 2003; 31: 428-432.
17. Fernández R, Serrano JM, Umaran I, Abizanda R, Carrillo A, López-Pueyo MJ, Rascado P, Balerdi B, Suberviola B, Hernanadez G, Sabadell Score Study Group: Ward mortality after ICU discharge: a multicenter validation of the Sabadell Score. Intensive Care Med 2010; 36:1196-1201.
*Magdalena A. Wujtewicz
Klinika Anestezjologii i Intensywnej Terapii UCK GUMed
ul. Dębinki 7, 80-211 Gdańsk
tel.: +48 58 349 24 06, fax: 58 346 11 82
received: 21.05.2011 r.
accepted: 18.07.2011 r.