• Critical care; outreach, effectiveness


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Summary Re-admissions have been cited as a measure of critical care quality and outreach teams have recently been introduced to improve critical care delivery. The aim of this study was to examine whether the number, causes and sequence of re-admissions to critical care altered as a result of the introduction of an outreach team. Re-admissions between April 2000 and November 2001 were examined. The reasons for re-admission were classified as (i) same pathology or disease process; (ii) new, but related, pathology; (iii) new and unrelated pathology; (iv) exacerbation of other comorbidities. During the two-year period, a total of 2546 patients were admitted to critical care of which 100 were re-admitted (49 before outreach and 51 after outreach). The reasons for re-admission did not vary before or after the introduction of the outreach team (same pathology 15 vs. 15; new, but related, pathology 17 vs. 23; new, but unrelated, 14 vs. 9; exacerbation of comorbidity 3 vs. 4, respectively, Chi-squared = 2.07, df = 3, p = 0.56). There was also no difference between the duration of stay on the general ward in between the critical care unit admissions before (median 2.93 [interquartile range 1.32–6.05] days) or after (median 2.25 [interquartile range 1.06–6.32] days) the introduction of an outreach team. As we could not detect any change in patterns of re-admissions as a result of the introduction of an outreach team, we would suggest that although outreach is an important development for critical care, its performance should be measured by other parameters.

In 2000, the government released £142.5 million for modernising critical care [1] following the publication of a comprehensive review of critical care services [2]. The review outlined proposals that included the development of critical care delivery groups, outreach teams and the formation of critical care networks. These networks are the mechanism by which improvements in critical care delivery across several trusts is achieved. To monitor improvements, the local East Anglian Network has included the number of critical care re-admissions as a quality marker; these data are to be collected monthly in each of the four network trusts in addition to the National Global Aims and Perfomance Measures. Re-admissions have previously been cited as a measure of critical care quality [3] and on a larger scale, the American Society of Critical Care Medicine now ranks re-admission to the intensive care unit (ICU) within 48 h of discharge as the top indicator of critical care quality [4].

The main roles of an outreach team are to avert admission to critical care and enable discharge by sharing critical care skills with general ward staff [5]. The principle behind outreach is to extend critical care services beyond the usual physical limit of an ICU and act as a service and educational partnership between critical care and the general wards. Although this principle seems intuitive, finding objective and reliable parameters to measure the impact of outreach services may be difficult. Up to 2002, only one of seven studies examining the impact of outreach on in-hospital mortality or cardiac arrests demonstrated improved outcome following the introduction of outreach services [6–12].

Outreach teams support the general ward staff by following up patients recently discharged from the critical care unit. As a result, alterations in the number and patterns of re-admissions to the critical care unit might be expected following the introduction of an outreach team. Indeed, substandard care following ICU discharge is a topic in the Royal College of Anaesthetists' publication describing a range of audit projects [13].

Therefore, the aims of this study were to examine whether the number, causes and duration of stay of patients re-admitted to critical care altered following the introduction of an outreach team and whether re-admission is a reliable marker of ICU quality.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

At the time of the study the Norfolk & Norwich was a 1000-bed university NHS trust serving a population of 500 000. All types of services were provided within the trust, except for paediatric intensive care, cardiac surgery and the care of major burns. The critical care complex comprised 14 beds (8 intensive and 6 high-dependency care) that were used flexibly as the workload demanded. The study was registered with the trust's research and development directorate but local ethics committee approval was not sought as the study involved secondary analysis of routinely collected data.

At a local level, the most tangible outcome of the Department of Health's release of funds was the development of an outreach team that became operational in February 2001. Therefore, to assess the impact of the outreach team, the number, causes and sequences of re-admissions to the critical care complex between April 2000 and November 2001 were examined (i.e. 10 months either side of the introduction of the outreach team). This period was chosen on the basis that it would include 1300 patients before and after introduction of an outreach team and detect a 50% reduction in re-admission rate with a power of 0.8 and an alpha error of 0.05. A re-admission was defined as a subsequent re-admission to either the ICU or high-dependency unit (HDU) prior to the patient's death or discharge from hospital. Re-admissions were identified by inspection of the critical care complex's patient demographics database and subsequently confirmed by inspection of the medical records.

The reasons for re-admission were classified as follows:

  • 1
    Same pathology or disease process precipitated the second admission (e.g. another exacerbation of acute pancreatitis).
  • 2
    A new, but related, pathology (e.g. enterocutaneous fistula following bowel resection, atrial fibrillation following an oesophagogastrectomy).
  • 3
    A new and unrelated pathology (e.g. pulmonary embolism, cardiovascular accident, chest infection following abdominal surgery).
  • 4
    Exacerbation of other comorbidities (e.g. chronic renal failure or chronic obstructive pulmonary disease).

Both investigators independently classified the reasons following review of the medical records and any differences were resolved by discussion and further review of the records. The duration of the original critical care complex admission, the interval on the general ward prior to re-admission and the duration of the second admission were also recorded to see how many re-admissions fell within the American Society of Critical Care Medicine's limit (48 h).

Any differences in the distributions of the reasons for admission and lengths of stay were examined using a Chi-squared test. A p-value of < 0.05 was considered significant. Data are expressed as mean (SD) for normally distributed data and median and interquartile ranges (IQR) for non-normally distributed data.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Between April 2000 and November 2001, 2546 patients were admitted to both the ICU and HDU of which 100 were re-admitted (49 before and 51 after outreach, a re-admission rate of 4%; 95% confidence interval 4.76–3.24%). The demographic details of all patients are given in Table 1, whereas those of the 100 patients re-admitted are shown in Table 2. All medical records were traced (100% retrieval rate). The most common reason for re-admission to the critical care complex both before and after the introduction of outreach was the development of a new but related pathology (n = 17 and 23, respectively) (Table 3) and the three most common reasons were respiratory dysfunction (n = 29), a leak from the gastrointestinal tract (n = 18) and re-admission of a high-risk patient previously managed within the critical care complex (n = 18) (Table 4). However, there was no difference in the numbers and distribution of the causes of re-admission after the introduction of the outreach team (Chi-squared = 2.07, df = 3, p = 0.56). The patients' ultimate outcomes are shown in Table 3. The trend towards increased ‘within critical care unit mortality’ following the institution of outreach was not significantly different [6 (12.2)% vs. 10 (19.6)%] deaths, Chi-squared = 1.18, df = 1, p = 0.28). Table 5 illustrates the lengths of stay of patients on the critical care complex and the general ward. There was no difference in the distribution of the lengths of stay of the first admission, the duration of general ward care prior to re-admission to the critical care complex and the duration of the subsequent critical care admission before or after the instigation of the outreach services. Twenty patients (41%) remained on the general ward for less than 48 h prior to the outreach team's introduction, whereas 23 (45%) were managed on the general ward for less than 48 h after outreach. The length of stay of all critically ill patients was not affected by introduction of an outreach team (Table 1).

Table 1.  Demographic details of all 2546 patients studied.
 Pre outreachPost outreach
  1. LOS: length of stay, SD: standard deviation, IQR: interquartile range.

Patients; n500791530825
Elective; n (%)71 (14)450 (57)87 (16)326 (60)
Emergency; n (%)429 (86)341 (43)443 (84)499 (40)
Survivors; n (%)352 (70)764 (97)432 (82)805 (98)
Female; n (%)191 (38)324 (41)213 (40)337 (41)
Male; n (%)309 (62)467 (59)317 (60)488 (59)
Mean age; years (SD)58.4 (21.6)63.9 (18.7)57.8 (20.0)63.3 (18.1)
Median LOS; days (IQR) 1.26 (0.69–3.59)1.02 (0.80–1.99)1.21 (0.68–3.12)0.98 (0.78–1.90)
Table 2.  Semographic details of the 100 patients re-admitted to the critical care complex.
 Pre outreachPost outreach
Patients; n4951
Elective; n (%)18 (37)24 (47)
Emergency; n (%)31 (63)27 (53)
Female; n (%)13 (26)20 (39)
Male; n (%)36 (74)31 (61)
Mean age; years (SD)62.0 (15.2)62.3 (15.8)
Table 3.  Distribution of the reasons for patient re-admission to the critical care complex and mortality.
Category of re-admissionDeaths before outreach; n (total = 49)Deaths after outreach (total = 51)
1 (same pathology)15 (2)15 (0)
2 (new but related pathology)17 (2)23 (7)
3 (new but unrelated)14 (1)9 (3)
4 (exacerbation of comorbidity)3 (1)4 (0)
Table 4.  Re-admission reasons and their relationship to first admission diagnosis.
Reasons for re-admissionTotal (n)Same pathology (n)New, but related (n)New but unrelated (n)Exacerbation of comorbidity (n)
Impaired respiratory function (e.g. infection, muscle weakness, exacerbation of chronic obstructive airways disease, pulmonary embolism, aspiration)2910811 
Post operative re-admission in high-risk patient previously admitted to critical care (e.g. internal fixation of cervical fracture, acute confusional state, re-fashioning ileostomy, division of adhesions, urinary leak after cystectomy, acetabular reconstruction, second carotid endarterectomy, debridement of infected tissues, intra-abdominal collection)1813311
Tachyarrhythmia (e.g. atrial fibrillation, ventricular tachycardia)81322
Leak from gastrointestinal tract (e.g. perforated duodenal ulcer, anastomotic leak)181161 
Severe ileus, bowel obstruction413  
Left ventricular failure81 52
Post operative haemorrhage (e.g. intra-abdominal, airway)9 621
Renal impairment2 11 
Hypovolaemia ans hypotension21  1
Table 5.  The distribution of the lengths of stay of re-admitted patients.
 Before outreachAfter outreach
  1. IQR: interquartile range

First critical care admission; days (median, IQR)1.68 (0.96–3.18)1.80 (0.96–4.03)
General ward admission; days (median, IQR)2.93 (1.32–6.05)2.25 (1.06–6.32)
Subsequent critical care admission; days (median, IQR)2.68 (0.94–5.79)2.02 (0.91–6.32)


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Our readmission rate was below the national figures reported by the Intensive Care National Audit and Research Centre (ICNARC) for re-admission rates in England, Wales and Northern Ireland (6.3%). Contrary to a previous report [14] showing that patients re-admitted to the ICU had a dismal prognosis with a reported mortality rate up to seven times that of patients recovering after only one admission, the mortality rate of our re-admitted patients varied between 11.7 and 19.6%. This is lower than the overall ICU mortality rate for the entire study period (23.8%) but higher than the combined mortality rate for the ICU and HDU (8.6%).

The most common reason precipitating re-admission was the development of a new, but related, pathology (40% of patients re-admitted). In only 15% of re-admissions was the second admission precipitated by an exacerbation of the original pathology. This is encouraging from a critical care perspective as it confirms that the patients' discharge was appropriate, coinciding with resolution of the initial pathology. The similar number of re-admissions before and after outreach suggests that the new, but related, pathology that prompted the second admission may have developed irrespective of whether the patient remained on critical care. This agrees with earlier research which confirmed that although over 50% of patients receive substandard care prior to intensive care admission only 8.5% of emergency admissions were independently assessed as being probably or definitely avoidable [15].

One of the main aims of outreach is to warn of patient deterioration and allow timely application of critical care. Our study was not designed to explore whether the patients re-admitted were identified earlier by the outreach team and so re-admitted to the critical care complex before they had deteriorated markedly from a physiological perspective. During the study period, the trust's outreach team operated only during normal working hours and so this may have been one reason for the apparent lack of impact. Either the outreach team did in fact have minimal impact on re-admission rates or such rates are the wrong parameter with which to try to assess an outreach team's performance.

The effect of outreach critical care has been formally examined by numerous previous authors [5–10] and reviewed by two authors [16,17]. Unfortunately, none of these studies or reviews could identify a clearly significant improvement in outcome following the introduction of outreach services. Only Buist et al. [12] were able to show a significant impact in reducing number of cardiac arrests and improving mortality following these cardiac arrests. However, this study has been criticised as there may have been other explanations for the improvement in outcome; for example, the outreach team may simply have increased ‘Do Not Resuscitate’ orders, so avoiding pointless resuscitations [18]. Therefore, selecting a valid and reliable performance indicator for outreach services is likely to be difficult using in-hospital mortality rate, cardiac arrest and re-admission rates.

Outreach is meant to speed up the identification of sick patients on the ward and so precipitate timely medical intervention which may avert critical care admission. However, if the reasons for admission and the numbers of re-admission have not changed, the impact of the outreach team on critical care services seems limited. Indeed earlier work has suggested that outreach services may actually increase the mean length of stay on ICU [18]; however, we could not detect any change in length of stay in the 1291 patients admitted before and the 1355 patients admitted after the introduction of outreach services.

The widespread introduction of outreach services was in response to a political directive [2]. Although the benefits may seem intuitive, our study adds to the growing body of evidence which suggests that the assumed benefits are difficult to quantify scientifically. Although the organisation of outreach services had undergone trials in various hospitals, no national model has been established. Until such a model is developed, it may be difficult to identify worthwhile performance indicators. Earlier work has shown that the introduction of an outreach team may have benefits in terms of reducing the number of unexpected ICU admissions because the critical care services are already aware of the patient's deterioration on the general ward [8,10]. Outreach teams have also been shown to increase the number of ‘Do Not Resuscitate’ orders, so sparing seriously ill patients the insult of ultimately futile intensive care [19]. These may be better indicators of both improvements in the delivery of care and the outreach team's performance.

In conclusion, we found that outreach services appear not to affect re-admission in terms of reasons or numbers. The effectiveness of outreach services should be measured but we would suggest that re-admission rate is not useful for this purpose.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  • 1
    Department of Health. Health Circular 2000/017. London: Department of Health, 2000.
  • 2
    Department of Health. Comprehensive critical care: a review of adult critical care services 2000.
  • 3
    Angus DC. Grappling with intensive care unit quality – does the readmission rate tell us anything? Critical Care Medicine 1998; 26: 177980.
  • 4
    SCCM Quality Indicators Committee. Candidate Critical Care Indicators. Anaheim, CA: Society of Critical Care Medicine, 1995.
  • 5
    Intensive Care Society. Guidelines for the Introduction of Outreach Services. Intensive Care Society, London, 2002.
  • 6
    Lee A, Bishop G, Hillman KM, Daffurn K. The medical emergency team. Anaesthesia and Intensive Care 1995; 23: 1836.
  • 7
    Hourihan F, Bishop G, Hillman KM, Daffurn K, Lee A. The medical emergency team: a new strategy to identify and intervene in high risk patients. Clinical Intensive Care 1996; 7: 2423.
  • 8
    Daly FFS, Sidney K, Fatovich DM. The medical emergency team (MET). A model for the district general hospital. Australian and New Zealand Journal of Medicine 1998; 28: 7958.
  • 9
    Goldhill DR, Worthington L, Mulcahy A, Tarling M, Sumner A. The patient at risk team: identifying and managing seriously ill ward patients. Anaesthesia 1999; 54: 85360.
  • 10
    Bristow PJ, Hillman KM, Chey T et al. Rates of in-hospital arrests, deaths and intensive care admissions: the effect of a medical emergency team. Medical Journal of Australia 2000; 173: 23640.
  • 11
    Salamonson Y, Kariyawasam A, Van Heere B, O'Connor C. The evolutionary process of medical emergency team (MET) implementation: reduction in unanticipated ICU transfers. Resuscitation 2001; 49: 13541.
  • 12
    Buist MD, Moore GE, Bernard SA, Waxman BP, Anderson JN, Nguyen TV. Effects of a medical emergency team on reduction of incidence of and mortality from unexpected cardiac arrests in hospital: preliminary study. British Medical Journal 2002; 324: 16.
  • 13
    Lack JA, White LA, Thoms GM, Rollin A-M. Raising the Standard. London: Royal College of Anaesthetists, 2000: Section 8 Number 9.
  • 14
    Chen L, Martin CM, Keenan SP, Sibbald WJ. Patients readmitted to the intensive care unit during the same hospitalization. clinical features and outcomes. Critical Care Medicine 1998; 26: 183441.
  • 15
    McQuillan P, Pilkington S, Allan A et al. Confidential inquiry into quality of care before admission to intensive care. British Medical Journal 1998; 316: 18538.
  • 16
    Hillman K, Parr M, Flabouris Bishop G, Stewart A. Refining in-hospital resuscitation. the concept of the medical emergency team. Resuscitation 2001; 48: 10510.
  • 17
    McArthur-Rouse F. Critical care outreach services and early warning scoring systems: a review of the literature. Journal of Advanced Nursing 2001; 36: 696704.
  • 18
    Smith GB, Nolan J. Results may have been due to education of ward staff. British Medical Journal 2002; 324: 1215.
  • 19
    Parr MJA, Hadfield JH, Flabouris Bishop G, Hillman K. The medical emergency team, 12 month analysis of reasons for activation, immediate outcome and not-for-resuscitation orders. Resuscitation 2001; 50: 3944.