General practitioners' knowledge and practice of complementary/alternative medicine and its relationship with life-styles: a population-based survey in Italy

Complementary and alternative medicine (CAM) includes a variety of diagnostic and therapeutic practices whose underlying theory or explanatory mechanisms do not conform to current medical thinking [1]. In its various forms, CAM is enjoying a growing popularity among the public [2, 3]. Dissatisfaction with mainstream modern medicine particularly with regards to patient-physician relationship, concerns about the side effects of chemical drugs, and personal beliefs favouring a more holistic orientation to health care are often quoted as possible explanations [3‐5]. Estimates of CAM use in Western countries range from about one-third to half of the general population [6, 7]. In Italy the proportion has almost doubled during the last decade [8], although it still remains far below the estimates reported in many European countries and the US. The analysis of data collected in 1999–2000 among the general population by the Italian National Institute for Statistics showed that in Tuscany 13.6% of adults had made use of CAM in the previous year [9]; yet the local Government is providing cost reimbursement for some CAM treatments under the National Health System for certain select conditions.

The use of CAM remains controversial [6]. Most of the debate focuses on its safety and effectiveness. Proper scientific evidence is lacking for most forms of CAM, partly because little methodologically rigorous evaluation has been carried out [6, 10]. The controversy also concerns the costs and reimbursement system [7]. So far, in most countries people have been paying out of pocket for these therapies, and providers' fees and national total expenditures appear to be considerable [6, 7]. There is a trend towards an increasing insurance coverage to include CAM treatments, in some cases covered by public money [7]. These issues are relevant for general practitioners (GPs), whose "gatekeeping" role with respect to health care expenditure is critically important. Faced with the increasing demand for CAM by their patients, GPs have to be prepared to discuss its uses and limitations, as well as its possible adverse effects [11]. They have to watch for signs of non-compliance with prescribed conventional treatments that may be associated with the use of CAM, as this is often not explicitly reported by patients [12]. Sometimes GPs are known to practise CAM themselves, particularly in certain countries [4, 13].

Several studies have described GPs' CAM views [14] and practices [15]; however, most of them have been hampered by methodological problems such as small sample size, convenience sampling, or unsatisfactory response rates [16].

This paper reports the results of a large population-based survey carried out in Tuscany with the aim of exploring GPs' knowledge of CAM and patterns of recommendation and practice, as well as the relationship between these patterns and the personal and professional characteristics of the physicians.

Overall, 1484 completed questionnaires were collected, corresponding to a response rate of 82% (83% for men and 80% for women); non-respondents were 13.0% and explicit refusals 4.9% (figure 1). Participants were mainly middle-aged with an overall median of 50 years, and a 25th and 75th centile of 47 and 53 years respectively (data not shown). Women were younger than men (mean age 48.2 and 52.1 years respectively, p < 0.001). Additional personal and professional characteristics of participants are shown in Table 1.

This paper reports the findings of a large population-based survey on the knowledge, recommendation and practice of CAM among GPs in Tuscany. Unlike previous studies, the random sampling strategy and the high response rate ensure that the sample is representative and that the results can be generalised. Tuscany is one central region of the country where the populations' use of CAM, which in Italy decreases gradually from the North to the South of the country [18], is similar to the national average [9]. Should the same geographical gradient apply to physicians, our findings may be considered indicative of GPs' level of knowledge and behaviours regarding CAM throughout the entire country.

In the present study over half of the respondents reported to recommend CAM to their patients while a far smaller fraction (13%) acted as CAM practitioners themselves. Published figures on the practice of CAM by primary care physicians vary greatly across countries, ranging from 8% [19] and 13% in Israel [20]; 16% [21] and 20% in the UK [22, 23]; 16% in Canada [24]; 20% [25] and 38% in Australia [26]; 30% in New Zealand [27]; 47% in the Netherlands [28], and up to 95% in Germany [29]. The popularity of different types of CAM also appear to vary geographically, with homeopathy being practised by 40% of Dutch GPs [28] and herbal medicine being favoured in Germany [29]. Year of data collection, different methodological choices and sampling strategies can certainly explain part of the variation [16]. The frequency of CAM provision by GPs may also be related to the extent to which it is demanded by patients and also to different policies of financial coverage across countries.

To our knowledge, the relationship between personal life-styles and attitudes towards CAM among physicians has not been previously explored. In this study factors such as following a vegetarian or macrobiotic diet and practising physical activity appear to be associated with the professional use of CAM. These findings support the theory that a holistic view of health and health care can be one of the possible determinants of GPs' interest in this type of medicine [5, 30].

The conditions for which physicians apply CAM are similar to those for which consumers themselves use these therapies, the most common being pain syndromes, psychological conditions, and chronic illnesses in general. Indeed these conditions, often dealt with in primary care, may still lack fully satisfactory treatments through conventional medicine. This "effectiveness gap" may partially explain the appeal of CAM to primary care physicians [31]. Yet, lack of conventional effective treatment cannot in itself justify the adoption of another treatment for which evidence of effectiveness is lacking and for which there is no control over the training and experience of the practitioners applying it. Under this aspect, our study has shown some worrisome findings. Among almost two hundred Tuscan family doctors currently practising CAM, more than one third reported that no specific certificated training was completed or in progress. This was the case for approximately 40% of GPs practising acupuncture and homeopathy and 82% GPs practising phytotherapy. Admittedly, the absolute number of CAM practitioners without specific training is small; yet the issue of professional competence in CAM is an important concern that has been highlighted also by other authors [4, 7, 32]. In Italy CAM training is mostly offered by private schools or associations and there is little control over the quality of the education provided. Even in countries where CAM is more widely included in medical curricula, education and certification of CAM practitioners is still an existing problem [12, 32]. The rapidly spreading public enthusiasm for CAM, coupled with the increasing acceptance of a consumerist and market-driven approach to health care, may push some physicians to respond to the patients' requests without having appropriate training. In addition, the perception of CAM as generally safe in terms of side effects, although incorrect [33], may contribute to its uncontrolled use.

Nevertheless, our study reveals that as many as 42% of Tuscan GPs never recommend CAM, principally for the insufficient evidence of its effectiveness. The lack of proper randomized controlled trials in the field of CAM and of substantial evidence of its effectiveness is a serious concern [12, 13, 27, 34].

CAM knowledge among GPs is not as widespread as the public demand seems to require, and the scarce evidence of CAM effectiveness hinders its professional use among a considerable number of GPs. Some CAM practitioners maintain that the assumptions underlying CAM treatment and procedures make it unsuitable for conventional scientific assessment. Yet, many of the methodological problems encountered are common to other therapeutic research [35]. Methodologically sound CAM research is essential to guide physicians' behaviour, to safeguard patients' safety, and to assist policy-makers in developing regulations for usage.