When a new drug or a new presentation of a drug is marketed, physicians need to be able to determine whether it is likely to be more useful for the treatment of their patients than the alternatives already available, and whether it is likely to modify their treatment practices. It is difficult for physicians to form their own opinions about a new drug. The pharmaceutical industry, through drug advertising, has a predominant and not always objective influence. Analyses of the official information about a new manufactured product would require considerable effort. This information is contained in different types of documents: (i) the summary of product characteristics (SPC), which forms the basis of its monograph, (ii) the SPCs of other drugs currently used for the same indication, (iii) the evaluation report for the new drug published by the national drug agencies.
Monographs have a standardized structure that varies little between countries. They include sections on composition, route of administration, indications, contraindications, adverse reactions, treatment regimens, etc. However, the nature of the novelty of the new drug is not specified clearly in the monograph, and the monograph alone provides no comparison with existing drugs for the same indications. Drug monographs provide sufficient information for the safe prescription of a given drug, but too little information to allow a physician to develop a therapeutic strategy.
When evaluating a new manufactured product with respect to existing treatments, physicians must first identify the set of drugs used for a given indication, and then study the monographs for each drug in the set, section by section, comparing the properties of each drug with those of the new drug. This process is time-consuming and unrealistic.
The national and international drug agencies (EMA , ANSM , HAS , FDA ) publish evaluation reports for each new manufactured product. These reports usually contain a comparison of the new drug with other drugs with same indications, recommendations for approval, and an overview of the clinical program, efficacy data, safety findings, recommended doses, and information for use in specific populations. These reports often conclude with a nation-specific drug-novelty index (e.g. in France, the Actual Benefit of the drug), which classifies the innovation as major, important or minor.
Very few physicians read these evaluation reports because they tend to be long and time-consuming to read. They also tend to focus on comparisons of the efficacy of the new manufactured product with other drugs for a given indication; aspects relating to safety or ease of use are described in less detail and may even be completely ignored. Improvements in drug labelling have been proposed, to provide patients and physicians with better information, including the results of studies comparing effectiveness . Other authors  have highlighted problems due to a lack of information concerning the value of the innovation brought by the new drug after entry into the market. The effectiveness of drug fact boxes for communicating information about the benefits and adverse effects of drugs has been evaluated experimentally. The results obtained were encouraging, but this approach essentially targets patients and not physicians . A comprehensive, but easy to read description of the various aspects of pharmaceutical innovation is required. There is therefore a need to develop new tools enabling physicians and other health professionals to comprehend the main characteristics and clinical impact of a new manufactured product both rapidly and easily. This requires the selection and structuring of the elementary pieces of information for each pharmaceutical innovation to be provided to the physician.
The objective of this study was to identify these pieces of information and to build and validate a conceptual model including the essential aspects of a new drug to be presented to physicians.
We describe the methods used for the selection and modeling of information about new drugs and provide conceptual maps defining the final model.