Clinical studies of innovative medical devices: what level of evidence for hospital-based health technology assessment?

• Published in: Journal of evaluation in clinical practice 2013-08, Vol.19 (4), p.697-702
• Main Authors: Boudard, Aurélie, Martelli, Nicolas, Prognon, Patrice, Pineau, Judith
• Format: Article
• Language: English
• Subjects: clinical trials; Clinical Trials as Topic; Clinical Trials as Topic - standards; Cost-Benefit Analysis; Equipment and Supplies; evaluation; Evidence-Based Practice; Evidence-Based Practice - standards; Hospital Administration; hospital-based health technology assessment; Humans; innovation; level of evidence; Life Sciences; medical device; Santé publique et épidémiologie; Technology Assessment, Biomedical; Technology Assessment, Biomedical - methods; Technology Assessment, Biomedical - standards
• ISSN: 1356-1294; 1365-2753
• DOI: 10.1111/jep.12024

Rationaleaims and objectives Like health technology assessment (HTA) agencies, hospitals are faced with requests for innovative and costly medical devices. However, local decision makers are frequently confronted with a lack of high‐quality clinical data when assessing the effectiveness of innovative medical devices. The aim of this study was to quantify the level of evidence available for innovative medical devices in the context of hospital‐based HTA. Methods We searched the Medline, Embase and Cochrane Library databases for articles, letters and reports relating to 32 innovative medical devices requested at our hospital between January 2008 and March 2012. All clinical studies retrieved were screened and classified according to the Sackett 5‐point level‐of‐evidence scale. Results We screened and classified 217 studies: 215 clinical trials and 2 cost‐effectiveness studies. Only 47 of the 215 clinical studies (22%) provided high‐level clinical evidence (levels 1–2); 33 (15%) were randomized controlled trials (RCTs). More than half of the 215 studies (52.1%) included fewer than 30 patients. Only 14 of the 47 high‐quality studies reported the amount of missing data. For implantable medical devices, 84 (71.8%) studies specified the follow‐up period and the mean follow‐up period was 18.9 months. Finally, methodological quality did not increase with the risk level of the medical device. Conclusions Our findings confirm that only a few studies of innovative medical devices provide high‐level clinical evidence. Nevertheless, RCT may be the ‘gold standard’ for drugs, but it is not always appropriate for medical devices. Changes to the European regulation of medical devices, with the requirement for a demonstration of clinical efficacy and safety before release onto the European market, have raised expectations.