METHODS

A computerised search of the literature was undertaken using Medline and BIDS. Articles were searched for the text words “observation”, “assessment”, or “short stay” combined with “ward” or “unit” and “emerg$” ($ being a wild card). The medical sub heading (MESH) “observation” was also searched. The related topics facility on PubMed was also used.

Internet searches were carried out using the same text words with a medical filter (www.medisearch.co.uk). The Department of Health and National Institute of Health databases were also searched.

Review of article titles and abstracts identified were undertaken, and all potentially relevant articles were reviewed for inclusion. The bibliography of each relevant article was searched for related citations and these were also reviewed.

Because of the paucity of articles in the field, all those with original data were critically appraised for inclusion in this review. The methods used conform to those described for qualitative systematic reviews.¹⁶

RESULTS

The current problems associated with medical admissions have been recognised to be communication, appropriateness of referral, finding beds, waiting by patients, and organisation of care.¹⁷ It is recognised that 70% of medical admissions can be admitted through an admissions ward (6% go directly to a critical care ward and the remainder are elective or admissions from clinic).¹⁸ About one third can be discharged within 24 hours.¹⁸ This is not in itself an indication of inappropriate admission^10,19–21 as the patient’s condition may have required short-term care or observation to minimise the risk of complications or greater morbidity and to rule out certain serious conditions. However, it has been suggested that 9.5% of short-term admissions could be prevented, most by a senior specialist opinion in the A&E department.²² Another study showed that increased imaging would prevent 16% of admissions to a medical unit.¹⁰ Over half the emergency admissions will come via 999 ambulance.¹⁸ To allocate those arriving by 999 ambulance to the A&E and those referred by the general practitioner to an assessment ward is therefore illogical and results in duplication of services. Most delays in discharge are attributable to organisational problems.²¹ Although these wards may reduce admissions, one study suggested that they may increase the inappropriate care of those discharged home.²³

Little work has been undertaken on the role of the medical admissions unit. As there is variability in the nomenclature and roles of admission wards, assessment units, etc, it is difficult to combine results to produce findings that are generalisable. One study evaluated a package of changes including the formation of a medical admissions ward and a consultant dedicated to the unit each week. It concluded that medical patients were no longer on other wards, more patients saw a specialist physician earlier, medical staff had less concerns of “losing” patients on outlying wards, and patient and staff satisfaction rose significantly.²⁴

The functioning of assessment/admission wards has been described by various authors.^7,25–27 Goodacre²⁷ recently described the role of short stay observation units in UK A&E departments. Some 59% of A&E departments had a short stay ward. By comparison, a review of Australian emergency departments found that 50% had an observation ward and the number of beds present was 1:5254 emergency department (ED) attendances. The operational characteristics of Australian units were similarly variable to those in the UK and America, but most were under the control of emergency physicians and had a time limitation on stay in the ward. Some concerns were expressed that the presence of an observation ward may decrease decision making in the ED.²⁸

Brillman⁶ reviewed 10 descriptive studies and concluded that 2%–6% of patients seen in the ED may be suitable for observation unit care. She discovered a wide range in the number of beds, from 10%–40% of bed capacity of the parent emergency department. An earlier American study showed that 27% of EDs had observation wards and that most of these were located within the department. A further 16% were developing such units. Users perceived them to be safe and cost effective.² Therefore, observation units seem to be an increasingly important part of hospital care because they can act as an alternative to admission or discharge and for financial reasons, more patients are being managed on an outpatient basis.

One of the key findings from the literature was that patient satisfaction increased with the presence of an observation ward.^29–31 In a prospective, randomised, controlled trial with a sample of 163 patients presenting to the ED with acute asthma exacerbation, Rydman et al²⁹ assigned patients to an asthma observation unit based in the ED, or to customary inpatient care. On four summary ratings of patient satisfaction, the observation unit patients scored higher with fewer total problems and fewer problems regarding care received, communication, emotional support, physical comfort, and special needs, than did the inpatient group. Patients cared for in an observation unit also had fewer investigations and spent time in a more comfortable environment before admission.³² As an extension of ED evaluation, an observation unit has also been shown to reduce the workload in the ED, thus giving staff better flexibility and improving the flow of patients.⁶ Patients may also benefit from the increased time available for monitoring their clinical condition after ED treatment, and as a direct result of this, referral to a specialist team (for example, cardiology) can be made earlier.²⁴

The pressure on hospital staff can be reduced by the provision of observation services because more time can be spent making important decisions regarding diagnosis.^6,26 As a consequence of this, senior doctors can prevent inappropriate admissions.^36–38 Hadden et al³⁶ also found that observation in a general ward seems to increase the length of stay in hospital when compared with observation unit stay. Patients also had to wait longer in a ward before being seen by a senior doctor. One of the key causes of delayed discharge is the wait for the results of important investigations.³⁷ It is necessary therefore that 24 hour access to diagnostic facilities is available in these wards.^6,38 Again, time may be saved if the rapid turnaround of results is focused in one area, permitting earlier discharge. Moreover, the unit may act as a hospital safety net for patient evaluation, again reducing pressure and stress on ED staff.³⁹

Although there seem to be many advantages associated with the provision of assessment/admission wards, poor management and inadequate operational policies can have severe effects on the running of the ward and can cause additional problems in other hospital departments.⁴⁰ Certainly patient admission and discharge decisions can be complicated by poorly defined procedures. Rainer et al⁴¹ commented that patients remaining in the ward in excess of 72 hours can cause a significant increase in workload and hence patients need to be transferred out within 24 hours of arrival.³² For this reason, some A&E specialists have been averse to accepting responsibility for an observation ward, claiming that a unit may become a “dumping area” for patients who ought to have been either admitted or discharged.^6,41 It is also important that definitions are adhered to regarding admission criteria—that is, whether a patient is admitted under A&E or another specialist group.⁴¹ The potential for increased stress levels among staff and on resources is also clear. The possible drain in staff from the ED needs to be countered by increased staff levels over other wards and the variation in admission numbers each day means that bed occupancy has to be low to ensure adequate capacity for busy days.³² Another problem with short stay wards is that they could potentially become centres of social care above that of medical care. In areas where assault and alcohol related violence are prevalent, a short stay centre for dealing with patients may be of benefit, yet this should not be at the expense of adequate community arrangements.⁴¹

A number of studies have evaluated the use of observation wards for specific patient groups. Each study found specialist observation units to be advantageous in terms of cutting costs, preventing unnecessary hospital admissions, and reducing overall treatment times. Nevertheless, a recent study by Goodacre, who reviewed the literature regarding chest pain observation units in the UK, suggested that there was insufficient evidence as to whether an observation unit will improve results if clinical practice is already good.²⁷ Furthermore, the same study questions the financial benefits of an observation ward, although cost savings do seem to be made in the United States healthcare system. Table 1 shows a summary of the conclusions drawn from specialist patient group studies.

One of the key criteria for assessing the efficacy of observation wards is the reduction in hospital admissions. To date, there have been a number of trials to determine whether such reductions actually occur. One randomised trial, which used a small patient sample size of 222, assessed the treatment of asthma patients and compared hospital inpatient care with an ED observation unit.⁴² However, no difference was found in admission rates between the two groups. A before and after study investigating the introduction of a paediatric observation ward noted a reduction in admissions of 31% and the frequency of under 24 hour admissions decreased from 17% to 10%.²³ Brillman⁴³ described the effect of the opening of an observation ward for adults and children. They found no change in admission rate, but a greater number of patients were sent from the ED to the observation ward and there was a decrease in direct discharge from the ED.

A number of authors have compared the cost of observation ward treatment and inpatient care. From these studies, it has been widely reported that observation units produce savings although, as these analyses are not cost effectiveness studies, they must be interpreted with caution. Table 2 summarises a number of cost studies that have been performed with different patient types.

No clear studies have been performed that compare the different organisational structures running an observation unit but it seems that strong management is essential, especially to ensure that transfers to other wards are not delayed.⁴⁴ One study has shown that the time from arrival to the decision to admit was 47 minutes quicker (158 compared with 205 minutes) when the patient was seen by an emergency medicine consultant compared with a general physician.⁴⁵ In another study, at least 91% of observation wards had ward rounds by senior staff, with over half undertaking at least one ward round per day.²⁷ Certainly delegation of tasks and a predetermined line of command will also facilitate decision making processes and assist in the smooth operating of the unit. Staff should be specifically allocated to assessment/admission wards and it is essential that experienced nurses are present to perform traditional doctors’ roles.^6,46

The number of beds required in a paediatric observation unit was assessed by Beattie and Moir.⁵¹ They concluded that one observation unit bed for every 5000 paediatric attendances was appropriate. However, in comparison, an American study suggested that about 75 per 5000 visits are suitable for observation unit care.⁴⁸ Some other American studies have proposed that there is a much lower need for paediatric observation units.^49,50 This variation is probably related to variation in practice and definition of observation units. The random daily variation in demand would mean that a single bed would be inadequate 25% of the time and empty 37% of the time. In terms of observation unit location no clear directives exist, though it has been suggested that a position within, or in very close proximity to the A&E ward is very important.^6,51

CONCLUSIONS

In the UK there has been a proliferation of observation and assesment wards in the past few years. The main driving force has been the avoidance of long waits in A&E. This review looks at whether there is an evidence base for establishing such units. There is no evidence regarding whether the initial assessment is best undertaken in A&E or an assessment ward, local factors will probably determine the best configuration but as functions are similar it is unlikely that geographical location is significant but that the intervention offered is more important. These units also serve as areas where observation can be undertaken. They may result in benefit because the person is admitted to an area where the usual practice is to observe and then discharge and hence they are not mixed with cases where a longer stay is usual. This may therefore maintain a culture of rapid discharge for this group of patients. This, combined with rapid access to senior staff and diagnostics, is probably the key to “observation medicine”. Clearly there are a number of advantages over admission to a general ward, yet certain criteria need to be met if the ward is to be of benefit to patients, staff and the hospital as a whole.

Assessment/admission wards comprise a wide variety of organisational and functional units. This makes comparison of research literature difficult. Similarly the articles rarely state whether observation beds were additional to or in place of existing inpatient complement. To aid future research and quality assessment, it will be necessary to define the terms observation ward, assessment ward and admission ward. We propose the following definitions:

Assessment unit/ward

An area where emergency patients are assessed and initial management undertaken by inpatient hospital teams. The patient is only in this area while early assessment is made, for example, up to 12 hours and is then moved to another ward.

Observation ward

An area where patients can be observed or have early investigation/ management within the A&E department. Patients are admitted to this area with an expectation of discharge within 24 hours.

Admission ward

A ward to which people are admitted after clinical assessment for their continuing management.

They all have a commonality in only accepting emergency cases. The use of a “patient plan” delineating the reasons for observation, line of treatment, and an end point for patient disposition is imperative. Therefore, the patient’s care in the unit should have a focused goal depending on the conditions presented. For example, some patients may require evaluation of specific complaints, while some may need short-term treatment for an emergency condition and so particular problems can be investigated during an observation period.⁶ If a physician cannot reasonably set a goal of care for the patient that can be achieved in the agreed time frame, then the patient should not receive an observation unit bed place.

A maximum time limit of 24 hours should also be adhered to as this is important in terms of care costs and to reduce the presence of inappropriate patients who ought to have been discharged or admitted to hospital.⁴¹ In accordance with this, it is important that the severity of the patient’s illness is limited and balanced by the intensity of services available.⁷

The assessment/admission ward should be in a well defined area. Ideally this is within the emergency department or directly adjacent to it. An American study has noted that 93% of these wards were located within the department,² though an analysis of this sort has not been performed in the UK. Colocation is logical as they share many requirements, for example, proximity to diagnostic facilities, and transfer between the units will commonly occur.

Some wards are managed by the A&E department and others by inpatient specialist teams. If an assessment/admission ward is to be successful, the managerial structure is of primary importance and a designated clinical lead should be appointed.³⁸ They all seem to have advantages over traditional admission to a general ward. These advantages are, however, dependent on the over-arching structures. Access to regular senior consultations, good diagnostic facilities, and access to external agencies for discharge planning seem to be vital components of a successful assessment/admission ward. They are probably cost effective, although no robust cost effectiveness studies have been undertaken in the UK. The other keys to their success are firstly the clear definition of admission and care criteria, and secondly the ability to maintain the flow of patients through the ward.

Admission and assessment wards have many advantages providing they are correctly managed with appropriate staffing and facilities.

The ideal assessment ward

The evidence suggests that the ideal ward should be time limited (24 hours) and be staffed by senior personnel. Numbers of staff required has not been studied in the literature. There is some evidence that turnaround time is quicker when they are managed by emergency medicine specialists. These units must have access to imaging and laboratory facilities (the evidence is not conclusive on exactly which tests). Strong mangement and policing of operational policies is vital. In particular, procedures for transfer out after 24 hours must be adhered to. Those who will evidently need longer term admission should not use these wards but go directly to a main hospital ward. The required size cannot be predicted from the literature. Other characteristics were not adequately studied to draw conclusions.

Diagnostic groups shown to benefit from a short stay ward include asthmatic patients, diagnostic chest pain, high risk but non-evident trauma, pyelonephritis, deliberate self harm cases, head injuries; they have also been shown to be of use for the elderly population and for children.