Printer Friendly

The need for appropriate critical care service provision at non-tertiary hospitals in South Africa.

To the Editor: Critical care is an expensive and labour-intensive undertaking, and it is well recognised that there is a global shortage of intensive care unit (ICU) beds. (1) Although there are no recent published data regarding the cost of ICU care, total costs were estimated to exceed R2 000 per patient per day in 1991. (2) Despite the cost, it is widely accepted that patients refused ICU admission have a significantly worse outcome. (3,4)

In the Cape Town metropole, home to at least 2.7 million people, (5) there are 9 public hospitals providing approximately 2 800 adult beds, 186 of which offer critical care services (personal communication, medical superintendents, 2004). Approximately 87 (3% of all beds) beds have mechanical ventilation facilities, generally accepted to represent an ICU service. (6) The remaining beds are located in high-care units (HCUs) offering specialised care, e.g. burn units, coronary care units, and/or a higher level of care and more intensive monitoring than available in general medical wards. While the number of critical care service beds required to serve a population of this size has not been determined locally, a recent survey of service provision at an urban secondary-level hospital in Cape Town has provided some useful data. (7) G F Jooste Hospital (GFJH), a 200-bed facility serving an estimated population of at least 1.2 million people, (7) attends to approximately 58 000 emergency unit visits per annum. Of these, almost 5 500 require admission for an acute medical problem--an average of 454 admissions per month. The survey (7) assessed severity of illness on admission using the Modified Early Warning Score (MEWS) (8) and demonstrated that at least 25% of medical admissions were sufficiently ill to merit admission to a critical care service (HCU or ICU). Given this service demand, we prospectively surveyed all medical patients admitted to the 8-bed HCU at GFJH in 2003, so as to define and quantify the critical care service provided by this secondary-level hospital.


All data were prospectively collected and analysed using Statistica version 6 (StatSoft, Inc. 2004). Categorical variables were compared using chi-square analysis and continuous variables were compared using a Student's t-test. A p-value of 0.05 or less was considered significant.


A total of 859 medical patients were admitted to the HCU over the 12-month survey period, comprising 16% of all medical admissions--approximately 90 medical admissions per month. The spectrum of medical emergencies necessitating HCU admission is shown in Table I. Collectively acute coronary syndromes (ACSs), diabetic emergencies, drug overdose/ poisoning and sepsis accounted for 76.6% of patients admitted. The median age of admissions was 45.3 years (range 13-86 years), with a median stay of 2 (0.5-14) days. The overall mean ([+ or -] standard deviation (SD)) APACHE score was 10 ([+ or -] 7), with an overall HCU mortality of 10.7% (in-hospital mortality 14.8%). Mechanical ventilation was required by 213 patients (24.8% of admissions). Mortality rates in ventilated patients are shown in Table II. The mortality rate for ventilated patients was 30.1% (in-hospital 38.5%). While this was significantly higher than the 4.3% for non-ventilated patients (in-hospital 7.0%) (p < 0.01), it compares favourably with local and international data. (9-11) Ventilated survivors had a significantly lower mean (95% confidence interval (CI)) APACHE II score than ventilated non-survivors, viz 15 (14 - 16) compared with 19 (17 - 21) (p < 0.0001). This was true of all survivors compared with non-survivors: 9 (9 - 10) versus 18 (16 - 20) (p < 0.001).


The survey results demonstrate a good patient outcome overall, given that this HCU provides limited ICU facilities (mechanical ventilation) to more than 200 patients per year despite inadequate staff-patient ratios (1:2) and limited training--only 2 registered nurses have ICU training. In addition, the 2-day duration of stay clearly indicates the demands placed on this service. There are, however, significant differences in the outcome of patients within the different diagnostic groups. These are listed in Table I. The excellent outcome in patients presenting with acute drug overdose/poisoning, including those requiring ventilation, and the ACS group, particularly patients with unstable angina pectoris (mortality rate 0%), differs greatly from the mortality figures observed in ventilated patients with a high APACHE II admission score and/or a primary diagnosis of diabetic ketoacidosis (DKA) or sepsis.

Based on the survey data, it is clear that non-tertiary hospitals with mechanical ventilation facilities are currently providing much-needed critical care in the Western Cape. The lack of ICU beds nationwide would further magnify the issues if extrapolated nationally. Withdrawing these services is likely to precipitate a crisis at tertiary hospitals, given the documented service demand. There are, however, patients who demonstrate a poor outcome under the current conditions of service delivery. It is therefore a matter of urgent priority to develop national guidelines that will assist non-ICU clinicians to address the following issues: (i) which patients with ACS are least likely to benefit from a HCU admission, i.e. limit the unnecessary use of scare resources; (ii) provision of non-tertiary critical care, including short-term ventilation (less than 48 hours), for admissions with an expected favourable outcome, e.g. drug overdose patients and/or those with admission APACHE II scores of less than 15; and (iii) early identification and transfer of critically ill patients who require tertiary care, e.g. ventilated patients with an admission APACHE II score of 15 or more, and/or a diagnosis of DKA or sepsis. For these criteria to be useful they will need to be based on simple, reliable clinical data, such as those suggested here, viz. primary diagnosis necessitating critical care, APACHE II score on admission and need for mechanical ventilation beyond 48 hours. The data from this audit have been provided to superintendents at both tertiary and non-tertiary hospitals as well as key critical care planning personnel at provincial level including the co-ordinating clinician for critical care services for the Western Cape. The detailed information provided should assist in formulating clinical guidelines for tertiary and non-tertiary critical care service provision. There is, however, also a national group of clinicians working on the formulation of national guidelines for the care of critically ill patients. The data provided by our survey may therefore serve to inform policy-making decisions at this level as well. Given inter-provincial differences in health care resource allocations, policy adjustments between provinces may be required. Despite these potential differences, clear guidelines regarding assessment and transfer of critically ill patients would be widely welcomed by the large number of clinicians currently managing critically ill patients outside the comparatively well-resourced settings of tertiary hospitals in South Africa.

(1.) Mathivha L. ICUs worldwide: An overview of critical care medicine in South Africa. Critical Care 2002; 6(1): 22-23.

(2.) Potgieter PD, Hammond JM, Pike R. Tertiary role of critical care in academic hospitals. S Afr Med J 1991; 80: 629.

(3.) Joynt GM, Gomersall CD, Tan P, Lee A, Cheng CA, Wong EL. Prospective evaluation of patients refused admission to an intensive care unit: triage, futility and outcome. Intensive Care Med 2001; 27: 1459-1465.

(4.) Simchen E, Sprung CL, Galai N, et al. Survival of critically ill patients hospitalized in and out of intensive care units under paucity of intensive care unit beds. Crit Care Med 2004; 32: 1654-1661.

(5.) Dorrington R. Projection of the Population of the Cape Metropolitan Area 1996 - 2031. Cape Town: School of Public Health, University of the Western Cape, 2000.

(6.) Nasraway SA, Cohen IL, Dennis RC, et al. Guidelines on admission and discharge for adult intermediate care units. American College of Critical Care Medicine of the Society of Critical Care Medicine. Crit Care Med 1998; 26: 607-610.

(7.) Burch VC, Benatar SR. Rational planning for health care based on observed needs. S Afr Med J 2006; 96: 796-802.

(8.) Subbe C, Kruger M, Rutherford P, Gemmell L. Validation of a Modified Early Warning Score in medical admissions. QJM 2001; 94: 521-526.

(9.) Van der Merwe E, Kidd M, Metzker S, Bolliger C, Irusen E. Validating the use of the APACHE II score in a tertiary South African ICU. South African Journal of Critical Care 2005; 21(1): 46-54.

(10.) Badenhorst CH. Multidisciplinary intensive care in Bloemfontein, 1981-1984. S Afr Med J 1986; 70: 674-678.

(11.) Harrison D, Brady A, Rowan K. Case mix, outcome and length of stay for admissions to adult, general critical care units in England, Wales, and Northern Ireland: the Intensive Care National Audit and Research Centre Case Mix Programme Database. Crit Care 2004; 8(2): R99-R111.

Respiratory Clinic, Groote Schuur Hospital, Cape Town

Richard van Zyl-Smit, MB ChB, MRCP, FCP (SA), Dip HIV Man (SA), MMed Cert Pulm (SA)

Paul Willcox, BSc Hons, MB ChB, FRCP (Lond)

Department of Medicine, G F Jooste Hospital, Cape Town

Vanessa Burch, MB ChB, MMed, FCP (SA)

Corresponding author: R van Zyl Smit (
Table I. Primary acute medical problem requiring admission to HCU

 Number of
Diagnostic admissions Median age
group (%) years (range)

Metabolic disorders 257 (29.9) 42 (13-79)
Acute coronary syndromes 247 (28.8) 57 (24-85)
Drug overdose/poisoning 92 (10.7) 24 (13-74)
Sepsis syndrome 91 (10.6) 37 (13-86)
Other CVS disorders 77 (9.0) 54 (16-86)
Respiratory disorders 77 (9.0) 46 (19-74)
Neurological disorders 18 (2.1) 36.5 (14-63)

 rate (%)
Diagnostic (HCU/ Examples of diagnoses
group in-hospital) included in this group

Metabolic disorders 11.3/15.2 Diabetic emergencies
 (DKA, hypoglycaemia) renal
Acute coronary syndromes 5.3/5.3 Myocardial infarction and
 unstable angina pectoris
Drug overdose/poisoning 3.3/3.3 Drug overdose and poisoning
Sepsis syndrome 31.9/41.8 Pneumonia, meningitis,
 septic shock
Other CVS disorders 13.0/26.0 Stroke, pulmonary oedema,
 cardiac arrhythmias,
Respiratory disorders 9.0/10.2 Non-infective, e.g. asthma,
 chronic obstructive
 pulmonary disease
Neurological disorders 5.5/22.2 Status epilepsy, myasthenia
 gravis, Guillain-Barre

CVS = cardiovascular; DKA = diabetic ketoacidosis.

Table II. Mortality outcome of patients most frequently requiring

 Ventilation required

 Mean Mortality
 Number Median age score in-hospital
 admitted (yrs) (range) (95% CI) (%)

A 39 24 (13-57) 14 (12-15) 2.6/2.6
B 38 47 (20-73) 14 (12-16) 15.7/18.4
C 43 42 (13-74) 18 (16-20) 55.8/67.4
D 18 48 (18-68) 23 (18-27) 61.1/72.2

 Ventilation not required

 Mean Mortality
 Number Median age score in-hospital
 admitted (yrs) (range) (95% CI) (%)

A 53 25 (13-74) 5 (4-6) 1.9/1.9
B 39 45 (19-74) 8 (6-10) 2.5/2.5
C 48 38 (13-86) 13 (11-16) 10.4/18.8
D 141 32 (13-72) 11 (10-11) 2.8/4.3

A = drug overdose/poisoning; B = non-infective respiratory disorder;
C = sepsis; D = diabetic ketoacidosis.
COPYRIGHT 2007 South African Medical Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Scientific Letter
Author:van Zyl-Smit, Richard; Burch, Vanessa; Willcox, Paul
Publication:South African Medical Journal
Geographic Code:6SOUT
Date:Apr 1, 2007
Previous Article:Antibiotic resistance in hospitals--are surveillance data of any value?
Next Article:Antimicrobial susceptibility profile of selected bacteraemic pathogens from private institutions in South Africa.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters