1 University of New South Wales, Liverpool Hospital, Sydney, Australia 2 Aberdeen, UK
EditorI would like to congratulate Dr Cuthbertson on the excellent summary of outreach teams, including a plea for evidence of its effectiveness before encouraging its introduction.1
When discussing outreach teams it is important to look at two dimensions. First, is there a problem in acute hospitals, which needs to be addressed: and secondly, what is the most effective way of dealing with that problem? There is little doubt that there is a large number of potentially preventable deaths, cardiorespiratory arrests and other serious complications in acute hospitals across many countries.2 There are many documented reasons for this including disempowered nursing staff who record deterioration without directly intervening; junior medical staff who have little formal training, either at an undergraduate or postgraduate level in the increasing complex areas of resuscitation and acute medicine; and senior medical staff who, even if they were experts in resuscitation, do not have much opportunity to maintain and increase their knowledge and skills in that area, and furthermore, are not continuously available for urgent bed-side consultation.
One solution to the serious problem of potentially avoidable deaths and complications is to train existing nursing and medical staff to increase their knowledge and skills. Another option being explored in North America is the concept of hospitalists, who are specialists trained specifically in acute hospital medicine, including care of the seriously ill in general wards.3 Single organ specialists and/or primary care physicians hand over care responsibility to these hospital specialists for the course of the patients entire hospital stay.
Another solution, which the editorial discusses are outreach teams. I presume in this context, the meaning is used in the broadest sense, whereby those with specialized skills in acute medicine establish and are responsible for hospital-wide systems, which care for the at-risk and potentially seriously ill before they either die or develop severe multiorgan failure.
There are potentially many different ways one could construct an outreach team. The common features would include simple and early ways of recognizing those at risk; responding in a timely way to at risk patients with staff skilled in all aspects of resuscitation and knowledge of the seriously ill; and a comprehensive implementation of the system across the hospital.
We have operated a Medical Emergency Team (MET) system at Liverpool Hospital since 1989.4 It uses simple criteria based on extremes in vital signs (e.g. hypotension, tachypnoea, decreased level of consciousness) with a rapid response by staff trained in acute medicine from the intensive care unit. A cluster randomized controlled trial of 23 hospitals throughout Australia is currently being conducted by the Australian and New Zealand Intensive Care Societys (ANZICS) Clinical Trials Group (CTG) together with the Simpson Centre for Health Services Research, in order to test the MET system. An implementation programme was conducted in 12 hospitals and is now being compared with 11 control hospitals over a 6-month period. The three major outcome indicators of the Medical Early Response Intervention and Therapy (MERIT) study are not for resuscitation (NFR) deaths, cardiorespiratory arrests and unanticipated (usually from general wards) admissions to the Intensive Care Unit (ICU). Antecedents to these events will be examined in both groups. Results hopefully will be available towards the end of 2003.
I inferred from the editorial that there would be a simple answer to the question of whether, for example, I would dismantle the MET system at Liverpool Hospital if the trial is negative? But I would have to say, probably not. Moreover, the decision to withdraw the MET system would almost certainly not be mine to make alone. Many of the nursing staff, junior medical staff, and even senior medical colleagues in my hospital are now so used to a real-time consultancy service provided by trained acute care clinicians from the intensive care unit that they probably would not accept the service being withdrawn. This is not surprising. Other systems for caring for seriously ill patients, mentioned in the editorial, such as ICUs and high dependency units also have very little, if any, data justifying their role, and certainly no randomized controlled trials. It would probably be just as difficult to remove them and suggest that home teams had to revert to caring for these patients on general wards, until we have good evidence that they work. This does not mean we should make no attempt to evaluate the effectiveness of outreach teams and that is why the MERIT study is being undertaken. However, the search for better ways of intervening early in order to correct ischaemia and hypoxia would not stop if the MERIT trial was negative. I do not believe anyone familiar with the principles of acute medicine would ever suggest late correction of life-threatening situations is better than early intervention.
Disappointingly, the editorial did not address the more complex research approaches necessary to evaluate new health services such as outreach teams. Measuring new and complex health service interventions is not as simple as measuring the effect of a new drug or procedure. In the case of the MERIT study, institutions rather than patients needed to be randomized for obvious reasons. Failure to reach statistical significance may be related to factors such as poor discrimination in the calling criteria; lack of skills and expertise in the attending team; lack of timeliness of response; inadequate educational strategies across the whole hospital; receptiveness of an institution to cultural change and failure to develop adequate sustainability strategies, as well as many other factors.
Conducting Health Services Research requires an interdisciplinary approach with expertise in areas such as epidemiology, social science, medical anthropology, linguistics, education, and statistics, all working closely with clinicians. A negative result in the early trials on the effectiveness of outreach teams, such as the MERIT study, would not necessarily mean that resuscitating patients and restoring oxygen delivery at the earliest possible time is not beneficial; it simply means we need to work on more effective ways to provide that service to our patients.
K. Hillman
Sydney, Australia
EditorThank you for the opportunity to reply to Professor Hillmans letter. I agree with his comments that there are two dimensions that require to be considered when analysing problems with acute hospital care. Outreach critical care is just one possible solution to these problems. I also agree that training nurses and doctors to increase knowledge and skills is vital to improve care. We do not need an evidence base to undertake this activity.
I am very encouraged to hear of the ANZICS Clinical Trials Group who are attempting to answer some of these questions through a cluster randomized trial design. With their excellent research record they stand a high chance of producing an important result. Publication of these results is keenly awaited. Professor Hillman lists approaches and possible confounders when undertaking health service research into new complex interventions. These points are important and should be considered when designing future studies in this area.
Finally, he comments that he would probably not dismantle his Medical Emergency Team even if the results of the ANZICS study were negative. This may further display that clinical rationale and anecdotal experiences have more effect on our practice than well-designed study evidence ever will.
B. H. Cuthbertson
Aberdeen, UK
References
1 Cuthbertson BH. Outreach critical carecash for no questions? Br J Anaesth 2003; 90: 46
2 Hillman K, Parr M, Flabouris A, Bishop G, Stewart A. Redefining in-hospital resuscitation: the concept of the medical emergency team. Resuscitation 2001; 48: 10510[CrossRef][ISI][Medline]
3 Wachter RM, Gold L. The Hospitalist Movement 5 years later. JAMA 2002; 287: 48794
4 Lee A, Bishop G, Hillman KM, Daffurn K. The Medical Emergency Team. Anaesth Intens Care 1995; 23: 1836[ISI][Medline]