3a Research in palliative care

The origins of palliative care research in the modern UK Hospice movement date from the founding of St Christopher’s Hospice in 1967. Dame Cicely Saunders advocated scientific observation and systematic research as an essential component of the specialty.

To ensure that patients are managed in the most appropriate way a solid body of knowledge must be developed. However, this can only be done on the basis of good research, which some would say is an absolute moral imperative.

Many treatments widely used in the palliative care setting have never been proven to be effective and their use is based on anecdotal evidence and doctor preference. For instance, nebulized morphine was in vogue for many years within the palliative care setting but it has since been shown to be only as effective as normal saline in helping with breathlessness. However, it would be unethical and impractical to generate doubt in nearly all palliative medicine treatments purely on the basis that they had not been submitted to quantitative trials.

Some of the evidence cited to support the use of medication in palliative care comes from populations of patients who do not traditionally fall within the remit of palliative care. This is particularly relevant for drugs used for neuropathic pain, where the research has largely been focused on patients with post-herpetic neuralgia or diabetic neuropathy and not those with complex mixed pains seen in palliative care.

The overarching difficulty in palliative care research is the balance between the needs of the individual patient–who, due to the advanced nature of their disease, may not have a ‘second chance’ of another treatment–with those of future patients for whom their treatment should be improved and based on research evidence.

The Declaration of Helsinki was drawn up by the World Medical Association in 1964 in response to the need for a code of ethics on human experimentation. This is particularly pertinent in the field of palliative care where the core practice is looking after the dying and the clear need for guidance for the physician caught in the conflict between patients’ own best interests and the necessity to advance knowledge for society as a whole.

Some people feel that palliative care research in dying patients is always inappropriate, an affront to dignity and an expression of profound disrespect for the emotional and physical state of those people who are terminally ill. Others feel that precious time, which is limited by disease and growing physical incapacity, should not be wasted or taken from patients or their families by conducting ‘research’, particularly when patients may be emotionally vulnerable and may feel easily coerced into studies in order to maintain the level of care that they need from staff.

Research has shown, however, that patients are not always adverse to participating even when it is clear that such research will have no immediate benefit for them. They may not share the concerns of ethicists about the difficulties and hazards of research with the terminally ill.

Patient-identified reasons for being involved in research:

Thus the methodological, ethical and practical difficulties encountered in conducting palliative care studies need to be looked at clearly, and strategies devised which are both sensitive to the needs of this particular group of patients as well as to the need of similar patients in the future to have access to improved care.

Evidence based practice is the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients

Sackett et al., 1996

Evidence-based practice is graded according to a system which attributes high evidence to the ‘gold standard’ research method of the randomized controlled trial (RCT) and lower evidence to studies of a more descriptive nature (Table 3a.1).

RCTs have not been used a great deal within the field of palliative care. There are different possible explanations for this, but the nature of palliative care itself may sometimes fit more readily with research methodologies such as qualitative and descriptive studies. Although tools are available to investigate issues such as quality of life and emotional distress as part of quantitative trials, qualitative research may better assess patient experience (Table 3a.2).

Qualitative research techniques, which are often more suited to palliative care, incorporate the subjective experience that cannot be measured so easily within a mathematical framework. Evidence from qualitative research studies have not been bestowed with the weight of evidence attributed to quantitative research, although the techniques can be as rigorous and are gradually becoming more accepted. Specific aims (generic goals) and objectives (specific ends or outcomes) and precision and clarity are important whether or not the methodology, data collection and analysis are qualitative or quantitative.

Qualitative research takes account of ways in which the research subject makes sense of his/her individual experience. Ideas and concepts develop as the research progresses, which may then be redirected back to further inform the research findings. Words are used as opposed to numerical data. The method is inductive, to discover new knowledge and to ‘ground’ it into the subjective experience. Although hotly challenged by enthusiasts of quantitative methodology, qualitative research–which often uses the imaginative expression of language–may have the power to disrupt existing assumptions and to challenge what has been considered as reliable, factual material.

Qualitative and quantitative research methods can be combined to bring a different perspective and to enhance knowledge in a more holistic way.

A range of techniques, guided by set principles, exist. Techniques include:

Palliative care covers a very wide range of patients with different morbidities. Such patients may include patients with slow growing metastastic cancer who may have many months or even years to live to others with end-stage heart failure and only a few days to live.

For clinical research to be clinically applicable it is important that the research is carried out in relevant patient groups rather than extrapolated from studies, which, although superficially similar, may include patients with widely varying characteristics.

Further, many patients with palliative care needs will have significant co-morbidities in addition to their primary illness, which can make defining a uniform palliative study population very difficult.

A significant proportion of patients with palliative care needs will be unable to adequately report their symptoms or complete questionnaires, either because they are too ill, too fatigued or have a cognitive impairment. This raises issues of the validity of consent to participate in research. Will consent for today cover consent in a week’s time when the patient’s condition deteriorates and they are no longer able to communicate clearly? Setting appropriate eligibility criteria is crucial and trial design needs to take account of these issues at the outset.

Setting appropriate endpoints in palliative care studies can be very difficult. If these endpoints are not specific enough then outcomes will be hard to evaluate. If the endpoints are too specific then the trial will be at risk of irrelevance to the complexity of the clinical situation encountered by patients at the end of life. It is very difficult to isolate a single variable and monitor its changes over time, particularly in the palliative care population which is commonly frail and elderly. Furthermore, patients are often receiving multiple interventions for several co-morbid conditions and facing the emotional and spiritual demands of confronting mortality. A more complex approach is necessary to take account of these different factors and to view interventions in the context of the patient’s overall disease journey.

Recruiting patients to trials is difficult for many reasons. They are often ‘protected’ by their families and also by clinical staff, who see them as being vulnerable and in need of protection against unnecessary burdens. There may be only limited opportunities to approach such patients to discuss trial involvement since rapidly changing clinical and emotional situations may make recruitment inappropriate. The recruitment to trials is therefore often much slower than anticipated.

Once patients are enrolled into studies, sample attrition rates up to 60% have been recorded due to rapidly changing physical and emotional conditions, or even death, during the course of the study.

Compliance can become a particular issue as the disease progresses in terms, for instance, of completing questionnaires. In designing trials which extend into the last weeks of life this needs to be anticipated and other simpler methods of evaluation built into the study from the outset.

There is often much confusion on the differences between research, audit or service evaluation (Table 3a.3) and in particular whether ethical approval is required.

For example, questions are often raised if palliative care patients are interviewed as part of a service evaluation and whether ethical approval is required.

Many clinicians will often approach an ethics committee for clarification.

Palliative care patients are considered a vulnerable group and include those with dementia, learning difficulties and children. This makes it difficult to allocate such patients to any form of care that could be deemed in any way less optimal than another.

Thus, randomized trials, which are understood by many to offer the best opportunity to minimize bias, should only involve patients with palliative care needs if there is a high degree of demonstrable equipoise between the interventions being studied. No possibility of disadvantage or morbidity associated with trial participation can be tolerated.

The European Trials Directive, introduced in 2004, brought with it increased protection for vulnerable patient groups. It has brought increased scrutiny from ethical and sponsorship committees.

With the implementation of the directive the length of time required to bring a trial through the process has increased considerably, adversely affecting cost and the practical administrative process. This difficulty has been addressed, in part, with the setting up of two National Palliative Care support collaboratives, SuPaC and COMPASS, and through organizations like the National Cancer Research Institute and other Supportive and Palliative Care initiatives to promote appropriately funded and supported multicentre studies.

Funding for palliative care research is limited, especially in comparison with cancer research. The necessary infrastructure to conduct palliative care research and to gather the teams of people together who have the necessary skills and knowledge is therefore difficult.

As previously discussed, matters of competence and consent are particularly important issues in relation to running studies in patients with palliative care needs, especially when their physical and emotional condition is deteriorating. Researchers are currently devising methods of consenting patients to enable recruitment into a specific study in the future should the patient become incompetent in the meantime.

In the UK, government funding for hospice services is still less than 50% of the total running costs. Given a choice between a research programme and clinical services, most hospice directors will be obliged to maintain the latter. There are however, many examples of small unfunded studies taking place in hospices. However, larger more comprehensive and time-consuming studies need funding either from industry, e.g. pharmaceutical, or other sources such as research charities.

The need for sponsorship and trial insurance is a major issue for palliative care units in the UK, two-thirds of which are charities without access to NHS research governance.

Most of those involved in palliative care research are heavily involved in clinical work without specific time dedicated to research. The need to collaborate with other research teams, with logistical and administrative support, therefore becomes very important.

The difficulties in conducting palliative care research are therefore many, but this must not reduce its importance nor its potential to make a real difference to the lives of the growing number of patients who will need palliative care in the years to come. Collaboration, finance, new methodologies, new research tools and ways of studying clinically complex patients will make future palliative care research more robust.

The models of research inherited from oncology may not all be appropriate, and as the specialty becomes more involved with patients who do not have cancer, new and distinctive ways of conducting clinical trials amongst the palliative care population are needed. Studies must be of the highest standard and yet practical with the promise to produce significant outcomes. Examples, include N of 1 studies, which are trials in which patients are repeatedly treated with the same treatment in order to compare their usefulness on an individual basis.