Abstract
The study of von Peter et al. (1) reported in this issue demonstrates once again how unwise it would be for physicians to ignore the rising tide of interest and use of complementary and alternative therapies amongst our patients. On this occasion, a high usage of complementary and alternative medicine (CAM) has been demonstrated in a population attending a specialist headache clinic, but the observation has been made amongst general neurological patients (2) and famously, in a wider, more generally representative population (3).
The authors discuss a number of reasons for physicians to take note of these findings, including the well-documented fact that most CAM use is not revealed to doctors (with likely attendant risks of treatment interactions). They mention that physicians should remain open to potentially beneficial treatments and that those that are deemed safe and effective should be incorporated into daily practice. The current lack of clarity on which therapies are effective means that they have to end by calling for more research of high quality. They suggest establishing new study designs which can combine ‘the specific demands of CAM with scientifically based, randomised controlled trials’ (RCTs). In other words, they feel a need for more evidence of efficacy before being able to recommend any forms of CAM as reasonable therapeutic options for headache sufferers and for full reimbursement by healthcare providers.
What kind of evidence though? Eighty-five percent of this study sample of 73 patients attending a head and neck pain clinic had used at least one of the 50 CAM therapies listed and 60% perceived the use to have been beneficial. Most of them had not responded to prior conventional therapy and by virtue of attending a specialist headache clinic, are likely to have been amongst the more difficult group to help with their headache symptomatology. One response to this kind of data is to wring the hands and bemoan the apparent willingness of ‘unscientific’ patients to respond to treatments for which there is no ‘evidence’. Here, the placebo effect, nonspecific, situational factors and regression to the mean are generally cited.
Another possible reaction is to wonder what conventional medicine might learn from CAM therapies. Quite apart from obtaining better understanding of CAM therapies and their practitioners, perhaps there are some CAM therapies which physicians might usefully and satisfyingly obtain training in themselves to improve their service to sufferers. If the bringing together of high quality conventional and complementary skills can happen within individual physicians, then I believe that the goals of the ‘integrative medicine’ to which the authors refer will truly begin to be met.
Many conventional interventions in the headache field are not well supported by results from RCTs and yet physicians continue to use them, satisfied with their real-world effectiveness (4). von Peter and colleagues acknowledge difficulties in the design of studies in CAM, at least those concerned with establishing treatment efficacy above placebo, but this is not the only form of evidence. Indeed, I would take from the current study a preliminary demonstration of some evidence that CAM approaches in general can be helpful in the management of headache. It has been well argued that much more use could be made of more formal outcome studies to decide on the usefulness or otherwise of specific CAM interventions – effectiveness research (5).
One example from the headache literature concerns homeopathy, a particularly implausible therapy, which some have suggested should not even be considered for investigation (6). There have been a number of well-conducted RCTs in chronic headache which failed to show efficacy over placebo (7, 8), but a recent observational study compared SF-36 scores before and 4–6 months after homeopathic treatment in 53 patients with longstanding migraine or tension-type headache (9). Highly significant improvements in all dimensions of health were found, with the largest improvements being in the ‘bodily pain’ and ‘vitality’ parameters. Here is a clear mismatch between failure to demonstrate an abstract ‘efficacy’ and demonstration of a concrete ‘effectiveness’, arguably of far more relevance to sufferers.
No one could argue with the goal of expanding the range of effective therapies generally available for headache sufferers. von Peter and colleagues have shown just how many potential therapies are already being used, waiting for assessment. If the only assessment tool allowed is to be the RCT though, then our patients are in for a long wait.