Knowledge transfer: what drug information would specialist doctors need to support their clinical practice? Results of a survey and of three focus groups in Italy

The research has been supervised by an Advisory Committee and has been carried out by Doxa Institute, an opinion research company, Italian branch of Gallup International Association. It consisted of two parts: a quantitative survey and a qualitative analysis by means of three focus groups.

This research offers a number of important elements to enlighten the debate on doctors’ information needs and information seeking behavior, integrating the results of a quantitative survey with findings from three focus groups, and gathering participants’ suggestions on how to improve access to, perceived quality, relevance and usability of, scientific information. In particular, focus groups enriched findings from the quantitative survey, allowing for better discrimination of perceived reliability and usefulness of different information sources, for in-depth analysis of physicians’ information needs and for defining their suggestions to meet these needs. Even considering the specificity of investigating opinions of specialists in infectious diseases and HIV therapies in Italy, which is the main limit of this research, we believe that it could bring relevant contents to the debate on drug information needs in general as drugs are the most important “equipment” for all clinicians - in either secondary or primary care - and all of them share the need to keep pace with drug related research in spite of hurdles like time constraints, pressure from drug companies and also from patients and health managers.

The quantitative survey showed that PSR are considered as valuable information providers, mainly because they are always available to supply information with a flexible schedule and also “on demand”. We cannot say whether such views were swayed by other positive reinforcements offered by PSR in addition to their support and time flexibility. However, appreciation of “traditional” channels (PSR) and info materials (brochures) seemed to be partly contradicted by doctors’ positive attitude to autonomous search of information, and by their wish to have more digital channels available to build an enhanced information framework; anyway, the overall picture emerging from the survey suggests that respondents were quite satisfied with the information they receive. Focus groups allowed for in-depth analysis and integration of these initial findings, showing a more articulated picture and a deeper understanding of limits that currently available information may have. In particular, participants’ opinions about strength and weaknesses of different information sources, channels and materials clearly revealed two general features that an information framework should have in order to meet with doctors’ perceived needs: on one side, physicians would like to be able to get – from dedicated professionals like PSR and also autonomously – complete, comparative (vs alternative treatments) and practical information quickly, targeting specific needs they would have in specific clinical circumstances, so that it can be easily used in everyday clinical practice; on the other side they would need to discuss their doubts with expert colleagues and get support from their experience. Therefore information focusing on specific needs should be “quick and easy” to access through PSR, internet and digital supports but should be also, and especially, shareable among practitioners and mediated by expert opinion: in other words, an evolution would be needed towards a more integrated use of the web for quick and targeted answers to clinical questions, combining the power of bibliographic and clinical databases on one side and clinical experience and expert opinions on the other side.

These findings are not completely new. Looking at the available literature, even though scientific articles from a few years ago may not fully reflect the current situation with the evolution of information frameworks, there are two main elements that characterize the information needs of doctors and their seeking behavior. First, their demand for quick answers associated to time constraints; [12‐14] and second, their reliance on experience from expert colleagues, who can (quickly) provide hints overcoming the difficulty to apply information about “average” patients to their own clinical encounters [2, 12, 15].

It is worthy of note that participants to our research were primarily concerned with the object of the information they want (targeted to specific patient needs - often not covered in scientific literature, which is mainly centered on “average” patients) and on the information framework that should provide it quickly (hyper-connected, with both expert colleagues and databases). Limited concerns were expressed on more specific characteristics of information, like its validity, potential biases and actual “evidence-base”: the need for information materials based on a reproducible analysis of the available evidence (therefore, based on a systematic methodology) was not focused on, and the threats due to non-publication of negative studies (publication bias) were generically dealt with as need to have also negative and not just positive data available. Limited concerns were also expressed about the need for information independent from vested interests, although independence – together with authoritativeness – of researchers has been cited as desirable. In this regard, knowing who produces the information was generally regarded as a fundamental element to qualify it as potentially useful, regardless of the presence of CoI. For example, participants did not show concerns about possible CoI and hidden biases associated to sponsored studies: they were confident that the filter operated by peer-review in scientific journals helps select valid research and that their own evaluation, together with their own and their colleagues’ personal experience, could help identify scientifically valid information as well as its applicability. Scientific validity seems to be assured by the perceived authoritativeness of information sources rather than the availability of an explicit and transparent evaluation framework, which is not asked for. Journal names, or names of experts who share their opinion, or names of scientific societies seem more important than an evidence-based assessment of the contents, that information materials generally do not provide in a plain and intelligible way and which would require time and skills that doctors generally do not have [16, 17]. No concerns about CoI of scientific journals [18] had been expressed either.

If we consider qualitative limits in the current offer of information to physicians, lack of EBM training in medical schools and time constraints, relying on the perceived authoritativeness of experts, local opinion leaders, [19] journals or societies may be an expected outcome and may often be a necessary shortcut to overcome difficulties in assessing the available evidence and using it in everyday practice. Risk of slipping from evidence to “eminence” based medicine – as self-referential assessment of health care interventions has been facetiously defined [20] - is concrete. More efforts are needed to provide doctors with more informative, useful and, at the same time, scientifically valid materials (either on web or on paper), based on systematic analyses of the available evidence [21] and synthesized through easy-to-read formats. These should ideally show a clear description of the characteristics of study populations, discuss results using absolute risk differences, evaluate possible study biases and highlight what the information does add in defining the place in therapy of drugs comparing to what is known about the therapeutic alternatives [5, 22]. Of course, web materials would allow different and additional layers of information through the use of hypertexts, and enhanced search engines would help accelerate the search for tailored answers.

Such ideal framework for evidence-based information would not be at odds with experience and expert opinions mediating its translation in clinical practice: evidence itself, if deployed by a thorough evaluation of the clinical context where it would be applied, is perceived as an academic exercise and can be of little use. The need for integrating a transparent assessment of evidence with experience is what the evidence-based movement itself has always advocated as David Sackett, considered one of the fathers of EBM, clearly wrote in a landmark 1996 editorial: “The practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research” [7]. In keeping with this principle, clinical practice guidelines are (or should be) developed through a systematic assessment of the available evidence [23] made by multidisciplinary groups of experts [24]. Such integration of evidence and experience, taking the collaborative nature of medical work into consideration, [25] is quite different from that “eminence-based” medicine, based on self-referential attitudes that many opinion leaders often show: while providing practical answers and hints to their colleagues, the so called experts would need to support their opinions with a transparent assessment of the available evidence, and show that they do not have potential CoI that may drive their opinions. This principle has been advocated for the development of clinical guidelines too, asking for clear disclosure of CoI and exclusion of experts with relevant and non-resolvable CoI [24].

This research suggests that institutional subjects who want to provide information to physicians should answer to a big challenge: providing quick, tailored and patient-centered answers through a transparent framework of available, comparative evidence, assessed for its validity and referred to the clinical context where it would be used, mediated by trusted experts with no relevant CoI who could help put the evidence in context; and favoring the exchange of information between practitioners. A big, big challenge, whose complexity had been already highlighted 20 years ago in a landmark paper by Richard Smith, where he weighted the “evergreen” role of experience from colleagues: [2]“… Probably there will be no single tool but a family of tools, and I suspect that, whatever sophisticated tools may be developed, the major source of information will remain colleagues”. Transparency of information (in its selection and assessment) and of experts’ CoI will be essential elements to consider for anybody who wants to compete on this field. As for specific key features, the development of evidence-based but uncomplicated information tools is already in a quite advanced stage in some countries like for example UK, Germany, Canada and the US. However, the development of relevant policies on CoI seems to be a general problem [26, 27]. Not to forget that patient-centered answers will require more patient-centered research, [28, 29] as our focus groups clearly indicated: participation of patients’ representatives in multidisciplinary panels defining research priorities [30] is an upstream requisite, so that information is not built on insubstantial grounds.