OBJECTIVES. To try to identify the optimal time at which to start assessing new and fast-evolving health technologies. To provide insight into factors influencing the timing of assessments and the choice of methods for assessing new and fast-changing technologies. HOW THE RESEARCH WAS CONDUCTED. A series of literature reviews were undertaken covering the general principles involved in the timing of health technology assessments (HTAs). Additionally, the reported assessments of laparoscopic cholecystectomy, chorionic villus sampling (CVS), teleradiology, teledermatology, genetic screening for predisposition to breast cancer, and gene therapy for cystic fibrosis were reviewed to try to identify the factors that influenced the timing of these assessments. Key individuals in each field were also interviewed. The selected technologies allowed comparison between those that were new and evolving and those that were relatively well-established. A bibliometric study of publication trends was also undertaken to see whether these trends would suggest points in the development of a technology that could be used as indicators that assessment should be started. RESEARCH FINDINGS. TIMING. The precise point at which assessment should start was not identified but the bibliometric study suggested that extending this approach might give useful results. For all health technologies, more regular reporting of outcomes and side-effects should be encouraged during the period after initial assessment and, where the technology is fast-changing, reassessment should take place from time to time. The precise intervals were not identified and the problem remains of deciding when a technology has changed enough to warrant reassessment. FACTORS INFLUENCING TIMING. Published reports of assessments did not generally specify the reasons for their timing, but a number of factors appear to have influenced the timing of those assessments, directly or indirectly. Product champions and opinion leaders pioneer the introduction of new technologies into clinical practice, and their reports may lead to the rapid diffusion of such technologies before they have been adequately evaluated, as was the case with laparoscopic cholecystectomy; this diffusion may limit the methods of evaluation that can then be used. It is therefore important to assess new health technologies before diffusion takes place. The extent to which regulatory control is imposed on the introduction of new health technologies can also influence the timing of assessments. Such controls might have helped to restrict the diffusion of laparoscopic cholecystectomy, making a large and widely generalisable randomised controlled trial (RCT) feasible. The source and availability of funding for studies may influence the nature and timing of trials. Many telemedicine evaluations were funded by commercial telecommunications organisations and were thus restricted in their timing (and biased towards the technological aspects of the applications) by the availability of funds. Media coverage undoubtedly has an influence although this influence is not always predictable; it may generate 'favourable' publicity about new health technologies, which can lead to immediate demands for the new technique, as was the case with laparosocpic cholecystectomy with its apparent benefits. Thus assessments should be made before media coverage exerts popular pressure on purchasers to adopt the technology and dissuades patients from participating in RCTs (because of fear they may be randomised to the standard treatment as occurred in a US trial of CVS). Innovators should also be cautious in the claims that they make to the media.(ABSTRACT TRUNCATED)