Focus on Alternative and Complementary Therapies
www.pharmpress.com/fact
Focus Alternat Complement Ther©2005 Pharmaceutical Press
Focus Altern Complement Ther 2004; 9: 87–90
The discussion around whole systems research (WSR) is not new, but began in earnest at the First International Workshop on Research Methods for the Investigation of CAM Whole Systems in Vancouver, Canada, in October 2002. At this workshop whole systems were described as ‘approaches to health care in which practitioners apply bodies of knowledge and associated practices in order to maximise the patients’ capacity to achieve mental and physical balance and restore their own health, using individualised, non-reductionist approaches to diagnosis and treatment. In whole systems the practitioner–patient relationship plays an important role and continues to evolve over time.’1 WSR effectively brings together many of the themes that define the uniqueness of CAM while at the same time creating obstacles to its evaluation by applying conventional clinical research methodology. Recognising the integrity of whole systems may allow us to develop a more appropriate, but equally rigorous, scientific understanding of a wide variety of traditional systems of medicine (from Shamanism through European herbal medicine to traditional Chinese medicine). WSR is, to some extent, rooted in the frustration of scientifically rigorous and committed CAM practitioners and researchers, with conventional clinical research methods. These methods often appear to fail to understand and assess the important subtleties that exist for both patient and practitioner within the whole process of CAM; WSR is an attempt to capture and evaluate that process in its unique context.
During the first workshop, we defined WSR as research that:
We believe that WSR methodology is equally necessary and applicable to conventional medicine as it is to CAM.
The Second International WSR workshop was held in order to capture the developments that had occurred over the previous year.1 It was advertised as a pre-conference satellite meeting to the 10th Annual Symposium on Complementary Health Care in London in November 2003. Thirty individuals participated in a combination of plenary and small group discussions. Thirty more individuals expressed an interest in attending the workshop but accommodation for the workshop required numbers to be limited.
The meeting focused on three main areas:
Patient–practitioner interactions are at the heart of all clinical medicine. There is considerable information available to suggest that patients seek, and indeed find, something quite unique in CAM that is simply not provided in their conventional care.2 However, the amount of time spent with the patient does not appear to be the most significant factor in explaining the benefits of the interaction between the patient and the practitioner. Astin et al. suggest that the key motivation for patients seeking CAM is a process of self-empowerment and communality of belief that they may feel when consulting a practitioner.2 Vincent and Furnham direct us to a similar process in which the egalitarian nature of the homoeopathic consultation and the shared belief system between patient and practitioner empower the patient through the whole process of receiving a CAM therapy.3 These issues are obviously complex and our small group discussions resulted in a recommendation that we create a model that could serve as a conceptual framework upon which we could map the issues that are important within this therapeutic process and interaction.
The group believed that a future model should incorporate increased recognition of the importance of measuring the practitioner’s role and perspective. Model development should also be informed by a review of the literature within, but also outside, CAM, drawing substantially from the experience of general practice and almost certainly using a Delphi process to define consensus among CAM researchers. The Delphi process is a method of obtaining judgements from a panel of experts; these experts are questioned individually, and a summary of the judgements is circulated to the entire panel. The experts are questioned again, with further iterations introduced as needed, until there is some consensus. While the Delphi process does not provide definitive answers, it does provide a starting point. Some of the issues that we think are unique to CAM might include the intent of the practitioner, the transformational nature of the consultation, the use of individualised therapies, the development of a therapeutic partnership, the use of a vitalistic philosophy that underpins almost all CAM systems, and the recognition that small therapeutic changes may have substantial effects. These issues are more distinct in CAM than in conventional medicine because of the differences in underlying philosophy. Finally, the group concluded that, in its quest for evidence, conventional medicine has not (sufficiently) recognised and dealt with individualised approaches, nor has it recognised the vitalistic philosophy that appears intuitively important to many CAM practitioners and to the patients who seek their advice.
The debate about outcomes in CAM continues to be controversial. Long has pleaded for the development of unique and sensitive CAM outcomes but others argue that we have a plethora of outcomes available to us and we are simply not applying them appropriately.4 Some of this conflict originates from the fact that many rigorous studies of CAM interventions appear to produce equivocal or negative outcomes when evaluated in the context of a conventional RCT. For instance, the Southampton research group’s study on homoeopathic immunotherapy for asthma was a large, well-powered and rigorous clinical trial that failed to demonstrate a difference between verum and placebo in the context of the patient’s asthma. However, it did demonstrate a possible homoeopathic effect with unexplained but significant improvements and aggravations in the homoeopathically treated group. This might indicate the need to look at individual rather than group outcomes in CAM studies.5
The search for patient-centred outcomes is clearly relevant in the context of individualised care. Such outcomes are relevant to patients both in terms of their initial choices about which therapies they would like to receive and their subsequent evaluation of the therapy’s impact on their condition. The relevance and appropriateness of such outcomes are indeed complex issues as far as CAM is concerned. The way forward will depend, to some extent, on our ability to distinguish between the patients’ and practitioners’ perceptions of effectiveness and the scientific community’s understanding of efficacy.6
Our approach during the workshops was not to reinvent the wheel. The group felt that reviewing the literature relating to measuring the benefits of therapeutic relationships (e.g. psychotherapeutic outcomes) and seeking more background data about patient-reported benefits within the context of qualitative research would be valuable starting points. However, we also debated whether there are some particular outcomes that might eventually prove more relevant to CAM than to conventional medicine, one example being a measure that captures ‘intention’ on the part of practitioner and patient. The group also recognised the importance of distinguishing between outcomes that evaluate a process (empathy or empowerment as part of a homoeopathic consultation, for instance) and outcomes that evaluate a primary disease-based endpoint (e.g. lung function in the case of asthma).
In identifying appropriate outcome measures, we also need to take into account the many different patients’ and caregivers’ perspectives in relation to the illness so that we can develop a holistic package that provides a pragmatic context to outcome. This may mean that we need to use a number of primary and secondary outcomes in combination, but with a clearly defined, a priori, analysis plan. Many of these issues are closely related to patient–practitioner interaction and therefore, while not explicitly discussed in the workshop, model development in this field should also inform the debate about appropriate outcomes.
The group concluded that in the next workshop we would like to see more formal presentations relating to the practical lessons learned from using particular outcome measures in evaluative research and audit relating to CAM care.
In the plenary session, Mikel Aickin pointed out that the issue that we should address is not whether or not RCTs are appropriate in CAM, or how we can make them so, but rather what methods need to be entangled with the explosion of CAM research. One very important line of development appears to be integrating the qualitative and quantitative aspects of measurements. Event-stream analysis, an evolution from N-of-1 research, captures data as a stream of unanticipated as well as anticipated measurement points and types of measurements. This analysis has the advantage of capturing numbers as well as expressions/meanings. A second potential methodological direction would be to focus more on the individual patient and less on group-averaged results. The assumption is that there will be patient heterogeneity in response and that there is information about the healing approach in that heterogeneity.
In the small group a five-phase framework was discussed in order to better understand and evaluate WSR.7
Phase I: Understanding what is going on – descriptive; involves practitioners and patients, e.g. what are patients using CAM therapies for and what do practitioners believe they can do best? It is important to systemise and standardise observations in order to be able to pool data. Qualitative methods will also be very useful. Methods such as meta-ethnography or meta-analysis of quality of life and treatment experience studies might be useful.
Phase II: Assessing safety through questionnaires and recording of adverse events (if not safe, should discontinue use of therapy).
Phase III: Effectiveness of the whole system: uncontrolled trials, pragmatic trials, observational studies, combination of CAM and conventional medicine (integrative medicine should be considered in this phase, including conventional medicine and CAM). In this phase it is also important to collect process data.
Phase IV: Efficacy of components and their interaction (recognises the complexity of interventions).
Phase V: Understanding the underlying mechanisms such as neurophysiological processes, etc.
We felt that the most important immediate step for us to make in relation to these design issues would be to carefully document what processes different research groups are focusing on at the present time and to follow their development within the context of this framework.
Invitations to participants in the First International Workshop on Research Methods for the Investigation of CAM Whole Systems were based on their research expertise in a wide range of CAM fields. Defining whole systems and WSR, and the conceptualisation of this field were most important in the discussions. Participants in the second workshop were self-selected and included more practitioners. While conceptualisation of the field of WSR was still important, in this workshop there was a greater focus on practice-based examples and the development of models within specific areas, based on participants’ experience and interest. It is clear that the development of a WSR methodology will be an ongoing process requiring input from researchers and practitioners with a wide range of experience and expertise. This was recognised by the participants, who agreed about the need to continue this work and organise a third workshop next year. A planning committee (the authors of this paper) was established to organise such a meeting. It was suggested that the workshop format was useful for enhancing discussions in the early development of this field. This means that the number of people attending the next meeting should be relatively small (less than 50), but an attempt should be made to bring in different participants to ensure a wide range of input in this field.
The broad goals that emerged from the workshop were remarkably similar for each of the three themes. The need for multidisciplinary literature searches, for building conceptual frameworks and for a forum to share this information was endorsed by all. What type of forum would be appropriate, other than another workshop, was not addressed in detail, but could include a newsletter or website. In addition, the proposed Delphi study with workshop participants (and possibly others) has the potential to be a major instrument in moving the field of WSR forward. What happens beyond that will depend on the degree of interest in this field in the CAM community, on sharing ideas and suggestions with the planning committee, on funding and logistical support, and on collaboration within the CAM community.
We thank Rebecca Brundin-Mather for her invaluable assistance with the organisation of the workshop and the production of the workshop report, and Jackie Burnham for her assistance with the preparation of the first draft of this manuscript. We also thank the workshop participants for their thoughtful input and suggestions; without them we could not have written this report. Organisation of the workshop was supported by funding from the Cancer and Complementary Therapies Research Team of the Sociobehavioural Cancer Research Network, administered by the Centre for Behavioural and Program Evaluation of the National Research Cancer Institute of Canada. Full reports of the First and the Second International Workshops on Research Methods for the Investigation of CAM Whole Systems can be obtained from Marja Verhoef.
