Trauma care involves a disproportionately small amount of research activity relative to the burden of the disease to society1, so that few injury treatments have a solid evidence base in randomized clinical trials. The lack of clinical research into trauma care is often blamed on unpopularity among grant-awarding bodies. There are, however, ethical, practical and personnel-related barriers that inhibit the ability to deliver clinical studies.
With regard to ethical barriers, many injured patients are unable to give informed consent in an emergency. Even if patients have ‘capacity’, they may not be able to give fully informed consent as they do not have sufficient time to weigh up the alternatives, ask questions and reach a decision, raising the issue of whether trauma care research is ethical2.
The Declaration of Helsinki recognizes the need for special measures to allow emergency care research, with an exception to the overarching rule that all patients must give informed consent before entry into a clinical trial3. ‘In such circumstances the physician should seek informed consent from the legally authorized representative. If no such representative is available and if the research cannot be delayed, the study may proceed without informed consent…’ In this quote the term ‘legally authorized representative’ does not mean a lawyer—simply a person who is recognized in law to represent the patient, such as a relative or friend.
The simple concept underlying the statement, however, is not easily translated into a regulation that covers all of the different situations that might arise. A hierarchy of consent exists on the basis that the people most likely to understand a patient's wishes are those who know the patient best. Thus the hierarchy is the patient, a relative or friend, an independent clinician, a clinician caring for the patient and, finally, a research ethics committee (waiver of consent).
The severely injured patient may be unable to give consent owing to a combination of impairment by the injury (for example brain injury), response to the injury (pain and endogenous catecholamine responses) and the effects of treatment (for example opiate analgesia). The patient may be in a highly aroused cognitive state, associated with a decreased ability to process information. In addition, there is likely to be insufficient time for the patient to decide on trial entry as many treatments have to be given urgently. It may be possible to ask the patient to sign a consent form, but this is different from gaining informed consent.
Relatives or friends (‘personal legal representatives’ in the Helsinki Declaration) may also not be in a good position to give informed consent. They have to absorb a great deal of clinical information about the injuries, investigations required, treatment needed and prognosis, all in a short space of time. They are often in a state of ‘information overload’ with insufficient time to understand what is happening, ask questions, consider alternatives and reach a decision.
Some systems require an independent person to give ‘consent’ if neither the patient nor a relative or friend is able to do this. It is difficult to see how an independent clinician, with no prior knowledge of the patient, can make any sort of judgement about what the patient's wishes might be. The independent clinician can only say that, if most eligible patients have given consent to take part in a particular study, it is probable that the patient would want to participate. In effect this is the same as a ‘waiver of consent’. Finding an ‘independent’ person at very short notice can also be difficult as there is some degree of interdependence between almost all staff within a hospital.
A ‘waiver of consent’ requires the research ethics committee to agree that individual patient consent will not be obtained. A prerequisite is often significant public or patient involvement involving focus groups, lay involvement in the research team, or public awareness through meetings, advertising or the media.
Many countries have systems in place that allow research in patients who cannot give consent. Trauma care researchers need to acquire the skills and resources to find their way through these complex administrative systems in order to address these ethical issues.
In terms of practical barriers, severe injury is relatively infrequent and happens at unpredictable intervals, often at night. The infrastructure needed to recruit patients to trials is often greater, therefore, than that required in studies on elective patients attending a specific clinic. The cost of always having a researcher present makes trauma trials expensive, although without this research infrastructure patient recruitment becomes unlikely. Initial identification of potential subjects requires a hospital information system flexible enough to alert research staff. Researchers need to be available rapidly enough to perform the study tasks, whether this involves doctors or research nurses.
Trial design may not allow the use of optimal endpoints. A survival endpoint with a clinically significant absolute risk reduction of 1 per cent from an overall 20 per cent mortality rate in severe injury requires about 20 000 patients to be randomized. Much of trauma research still follows an explanatory model, whereas the practical nature of the specialty requires a more pragmatic approach. In an ideal setting, every trauma trial would have a clinical endpoint, a patient-reported endpoint and an economic evaluation.
A lot of trauma research has weak methodology with too much reliance on observational studies. A good example is use of fresh frozen plasma (FFP) in traumatic haemorrhage and the recent change in practice to a 1:1 FFP:blood ratio. This move has been based solely on retrospective cohort studies, potentially affected by survivor bias. Did the patients survive because they got more FFP, or did they get more FFP because they survived?
With regard to personnel-related barriers, a lack of personnel restricts the amount of research into trauma care. In the UK, there are relatively few academic clinicians involved in trauma care. Of the 985 Fellows of the Academy of Medical Sciences, only three mention the word ‘trauma’ in their areas of interest, and only an additional nine use the word ‘injury’ (excluding injury at the cellular level).
With few exceptions, trauma appears rarely in the top-ranking medical journals and consequently fails to attract large public funding and industry support to compete at the top levels within universities. The highest-ranked journals in trauma care have impact factors well below those of the top journals in many other specialties. Across the world, relatively few trainees in trauma care wish to compete for the best academic training schemes because the specialty has an intense and practical clinical commitment that often involves working antisocial hours. This leaves little time and energy for clinical research.
Research funders do, however, seem increasingly to be interested in trauma care, but there is a need to develop centres with the people and academic leaders to deliver the clinical studies that are required. In an ideal world the networks necessary to deliver trauma care research would be linked to the trauma systems that deliver clinical care. In both the UK and Germany, this process is becoming government policy rather than being left to the enthusiasms of individual practitioners.
Few clinicians will lead trauma research projects, but a culture in which every trauma practitioner is motivated to take part in clinical research must be created. This must involve an environment where trauma practitioners want to contribute to research projects. The two successful Clinical Randomisation of an Antifibrinolytic in Significant Haemorrhage (CRASH) trials (which randomized 10 000 and 20 000 patients) showed the power of the trauma community when inspired to produce the evidence to improve care4, 5. Although training academic leaders for the future is essential, they alone cannot deliver what is needed. Few will be research leaders, but all must make a contribution.
The key to funding and delivery of high-quality trauma care research in the future will be collaboration in large research networks that share research expertise, involving many ‘non-academic’ practitioners. Creating a mindset in which clinical research is an integral part of clinical care remains equally important.