Abstract
Definitions and demarcations are difficult in medicine, especially when it comes to poorly defined concepts such as traditional, complementary, alternative, and integrative medicine and their relation to (also poorly defined) conventional medicine. First, one is faced with the problem of what constitutes these practices in the first place. World Health Organization (WHO) defines traditional medicine as “the sum total of the knowledge, skill, and practices based on the theories, beliefs, and experiences indigenous to different cultures, whether explicable or not, used in the maintenance of health as well as in the prevention, diagnosis, improvement or treatment of physical and mental illness.” 1
The National Center for Complementary and Integrative Health (NCCIH) defines complementary medicine as the use of a non-mainstream approach together with conventional medicine and alternative medicine as the use of a nonmainstream approach instead of conventional medicine. According to NCCIH, integrative medicine brings conventional and complementary approaches together in a coordinated way. 2 Interestingly, conventional medicine is not defined in either source.
Since, apart from the WHO definition of traditional medicine, all definitions are primarily negative in nature (“non-mainstream”), Ng et al. have recently presented an operational definition of complementary, alternative, and integrative medicine (CAIM) based on systematic methods, which includes 604 different CAIM therapies. 3 This is an enormous step from mere demarcation to real content. Nevertheless, the question remains when a procedure actually is classified as CAIM and when as conventional medicine. A recurrent suggestion of the distinction between CAIM and conventional medicine is that the latter, unlike the former, is based on scientific criteria and evidence. 4
This corresponds to the self-perception of academic medicine as an evidence-based natural science—I am not excluding myself at all here. But our self-perception often plays tricks on us and, for example, often leads to the logical impossibility of everyone perceiving themselves as above average. 5 As scientists, however, we fortunately have some rather objective methods of challenging our collective self-perceptions.
And indeed, research to date has left little to the myth of a purely evidence-based (conventional) medicine: for example, a review of U.S. cardiology treatment guidelines found that in 2009, only 11% of treatment recommendations had level A evidence, that is, were based on at least one large randomized trial. 6 Thus, nearly 9 in 10 treatment recommendations were not really based on strong evidence but on low-quality evidence or expert opinion.
This situation, however, had changed in another review, almost 10 years later. Then, only 8.5% of the treatment recommendations were based on level A evidence. Thus, cardiology, at least in the United States, is moving further and further away from true evidence-based medicine. 7 But this of cause does not keep cardiology guidelines from advising against the use of CAIM procedures in, for example, hypertension, because the data were still too uncertain here. 8
More comprehensively, a recent meta-analysis looked at the evidence base of medicine (as a whole): 2428 randomly selected Cochrane Reviews published between January 2008 and March 2021, representing 35% of all Cochrane Reviews published during this period, were analyzed in terms of their findings and the quality of those findings. 9 Why Cochrane Reviews? Because they can be considered the gold standard of evidence-based medicine and hardly any other source provides better and more reliable data on the efficacy and safety of a wide variety of interventions for a wide variety of diseases. 10 The analysis found that across all conditions, only 5.6% of 1567 eligible interventions had high-quality evidence supporting their benefits. More than 9 out of 10 interventions thus failed to show positive effects of high methodological quality.
One thing I personally appreciate about Cochrane is that conventional and CAIM approaches are basically judged the same and according to identical criteria. Accordingly, the current analysis includes figures on all forms of medicine. And all of them perform poorly: slightly more than half of the reviews examined drugs, of which 7.8% found high-quality evidence for positive effects. Drugs thus still represented the “best” class of interventions, which is not surprising since the current evidence assessment methods are optimized for pharmacologic interventions.
Accordingly, for behavioral, psychological, exercise, or dietary interventions, high-quality evidence for positive effects was found in only 4.1% of the reviews, and for surgical interventions, in only 4.0%. For procedures classified by the authors of the current analysis as “alternative,” not a single review found high-quality evidence for positive effects. However, only 46 of the included Cochrane Reviews examined “alternative” methods, so it is better to speak of less than 2% high-quality evidence for positive effects.
I am not surprised that CAIM scores the “worst” here. This is not necessarily due to a lack of positive effects, but mainly due to the poor study quality: only 4.3% of the Cochrane Reviews on CAIM had high levels of evidence, compared with proportionally about three times as many pharmacologic Cochrane Reviews. Thus, the quality of CAIM studies is fundamentally too poor to make reliable statements about efficacy or inefficacy. And when I look at what is sometimes published in our field, I am painfully aware of this lack of high-quality studies. Pharmacologic research is much further along, the quality of the studies is much better, but the majority of the preparations are simply not effective.
Accordingly, I would still be cautious with CAIM definitions, which primarily classify it by distinguishing it from an ideal-type scientific evidence-based “conventional medicine,” as this simply does not exist in our notoriously imperfect world.
Ultimately, we are all in the same boat and should try to offer patients the best possible health promotion, prevention or therapy. Regardless of whether it bears the label conventional medicine or CAIM, since this often says frighteningly little about their evidence base.