Dr Andrew Keegan
Consultant Physician & Former NSW President, AMA
Bed occupancy rates have been proposed to reflect the ability of a hospital to provide safe efficient patient care. The Australian Medical Association and the Australasian College of Emergency Medicine have acknowledged that bed occupancy rates above 85% negatively impact on the safe and efficient operation of a hospital. In its Position Statement on "Acute Hospital Bed Capacity" (March 2005), the Irish Medical Organisation has also acknowledged an average bed occupancy of 85% as an "internationally recognised measure" that should not be exceeded. In 2005 the average hospital bed occupancy in the 30 OECD countries was 75%. Furthermore, the Department of Health in the United Kingdom (UK)1 has found that bed occupancy rates exceeding 85% in acute hospitals are associated with problems dealing with both emergency and elective admissions. That county has instituted a target bed occupancy of 82% as one of its hospitals' quality measures.
There seem to be two lines of evidence that lead to and support the value of bed occupancy as an operational quality measure and target. Namely, the risk of cross-infection between inpatients in crowded wards and timely admission to an appropriate ward of patients presenting to emergency departments (ED) or for booked surgery.
Most reports concerning inpatient cross-infection consider the spread of methicillin resistant staphylococcus aureus (MRSA) in the hospital setting. The percentage bed occupancy was found to have a significant positive correlation with MRSA rates in hospitals in Northern Ireland between 2001 and 20032. However, it should be noted that all but one of the 12 hospitals in that study had bed occupancy rates below 90%, with only 4 below 80%. The same study also found that bed turnover interval influenced MRSA infection rates to an even greater extent.
Borg3 also found a significant correlation between bed occupancy and MRSA infection rates. Moreover, the introduction of additional beds to a ward during periods of high demand directly contributed to extra cases of MRSA. An extra bed added to a four bed room may lead to a three fold increased risk of acquiring MRSA4, even when staff numbers are increased to accommodate the increased patient load.
The association between nosocomal infection and bed occupancy rate was also highlighted in another UK Department of Health report5 . That report revealed that hospitals with occupancy rates of more than 90% had a 10.3% greater incidence of MRSA infection than those with occupancies below 85%. Furthermore, the UK House of Commons Committee of Public Accounts has "repeatedly noted that high levels of bed occupancy are not consistent with good control of infections" 6 .
The association between appropriate emergency bed access and hospital occupancy has been studied in a number of Western countries. Cooke et al7 found that in 147 hospitals in England, a higher average bed occupancy rate was associated with a greater proportion of ED patients needing admission waiting more than 4 hours (p < 0.0001). Those authors suggested that reducing bed occupancy rates below a "threshold" should decrease excessive waiting times in Accident and Emergency (A&E).
A UK Department of Health report8 found that during 1997-98 average bed occupancy levels in general and acute bed varied between trusts from around 50% to around 99%. However, most trusts reported periods during which inpatient bed demand exceeded availability. At those times, emergency patients were more likely to be assessed and have treatment initiated in A&E before moving to an inpatient ward than in would usually be the case. Furthermore, those hospitals with higher average bed occupancy rates cancelled a significantly higher proportion of elective operations and had longer delays in the transfer of patients from emergency to inpatient beds. The report also acknowledged that; "hospitals with average occupancy levels above 85% can expect to have regular bed shortages and periodic bed crises".
Observations made within the NHS have been used by Bagust et al9 to develop a discrete-event stochastic simulation model to study the relationship between demand and available hospital bed capacity. This model suggests that there is a discernable risk of a hospital failing to provide sufficient beds, and thereby safe efficient care, when average bed occupancies exceed 85%. Moreover, when a bed crisis does occur the recovery period could be also be significant. For instance, a hospital with an average bed occupancy of 85% would be expected to "run out of beds for four days in a year" with an associated disruption of normal function of up to eight weeks. The authors concluded that, considering the nature of hospital system, "spare (bed) capacity is essential if an emergency admissions service is to operate efficiently and at a level of risk acceptable to patients".
Orendi6 has recently compared the circumstances in the UK with those in the Netherlands where the average hospital bed occupancy rate was 64%, as opposed to 84% in the UK (2005), with the same number of beds per head of population. As a consequence of this that author felt that, in the Netherlands, admission decisions were not affected by bed pressures. The lesser pressure on hospital beds may in part have been the result of the special level of care provided to nursing home patients, which avoids the need for them to engage acute hospital services.
Canadian data also show that hospital bed availability has a significant influence on ED length of stay for admitted patients10 (access block) and thus a delay in patients reaching an appropriate inpatient bed. This was most marked when "hospital occupancy exceeded a threshold of 90%", as also found by Sprivulis et al11. On the other hand bed occupancy did not affect consultation and admission rates. Furthermore, analysis of emergency presentations to an Australian hospital has shown that access block may increase a patient's overall hospital length of stay12, particularly when they present after hours. Additionally, patients admitted to hospital during periods of ED overcrowding associated with access block, may be subject to an increased in-hospital mortality11,13. Moreover, high hospital bed occupancies have been directly related to an increase in the mortality of patients presenting to EDs in Western Australia11 independent age, season, diagnosis or urgency.
Using data from the Sydney South West Area Health Service Annual Report for 2006/2007 the linear regression line calculated from Percentage Occupancy Rate (BO) and Emergency Admission Performance (EAP, of hospitals performing below 95%) data has a negative correlation coefficient of -0.72 (EAP = -0.66 x BO + 133). That line suggests that a bed occupancy rate of 85% would be expected to be associated with an Emergency Admission Performance of 77%. Although this assessment is clearly an oversimplification of the situation, it is of note that the result is consistent with the literature.
Overall there appears to be sufficient evidence to support the contention that bed occupancy rates provide a useful measure of a hospital's ability to provide high quality patient care and that 85% is a reasonable target.
1. Department of Health (UK). Shaping the future NHS: long term planning for hospitals and related services. Consultation ducoment on the findings of the national beds inquiry. London: Department of Health, 2000
2. JB Cunningham, WG Kernohan, R Sowney. Bed occupancy and turnover interval as determinant factors in MRSA infections in acute settings in Northern Ireland: 1 April 2001 to 31 March 2003. J Hosp Infect 2005;61:189-193
3. MA Borg. Bed Occupancy and overcrowding as determinant factors in the incidence of MRSA infections within general ward settings. J Hosp Infect 2003;54:316-318
4. CC Kibbler, A Quick, AM O'Neill. The effect of increased bed numbers on MRSA transmission in acute medical wards. J Hosp Infect 1998; 39:213-219.
5. Department of Health (UK), Hospital organisation, specialty mix and MRSA.Report no 9163. December 2007.
6. J Orendi. Health-care organisation, hospital-bed occupancy, and MRSA. The Lancet 2008; 371:1401-1402.
7. MW Cooke, S Wilson, J Halsall, A Roalfe. Total time in English accident and emergency departments is related to bed occupancy. Emerg Med J 2004;21:575-576
8. Report by the Comptroller and Auditor General; Inpatient Admissions and Bed management in NHS acute hospitals. NHS Executive, London UK, February 2000.
9. A Bagust, M Place, J Posnett. Dynamics of bed use in accommodating emergency admissions: stochastic simulation model. BMJ 1999;319(7203):155-158
10. AJ Forster, I Stiell, G Wells, AJ Lee, C Walraven. The effect of hospital occupancy on emergency department length of stay and patient disposition. Acad Emerg Med 2003;10:127-133.
11. PC Sprivulis, JA Da Silva, IG Jacobs, ARL Frazer, GA Jelinek. The association between hospital overcrowding and mortality among patients admitted via Western Australian emergency departments. MJA 2006;184:208-212.
12. DB Richardson (1).The access-block effect: relationship between delay to reaching an inpatient bed and inpatient length of stay. MJA 2002:177:492-495
13. D Richardson (2). Increase in patient mortality at 10 days associated with emergency department overcrowding. MJA 2006;184:213-216
Australian Doctors Fund