PT  - JOURNAL ARTICLE
AU  - Osipenko, Leeza
AU  - Ul-Hasan, Saba Ajwat
AU  - Winberg, Debra
AU  - Prudyus, Kseniia
AU  - Kousta, Marina
AU  - Rizoglou, Artemis
AU  - Rustignoli, Isabella
AU  - van der Maas, Laurens
TI  - Assessment of quality of data submitted for NICE technology appraisals over two decades
AID  - 10.1136/bmjopen-2023-074341
DP  - 2024 Jan 01
TA  - BMJ Open
PG  - e074341
VI  - 14
IP  - 2
4099  - http://bmjopen.bmj.com/content/14/2/e074341.short
4100  - http://bmjopen.bmj.com/content/14/2/e074341.full
SO  - BMJ Open2024 Jan 01; 14
AB  - Background The National Institute for Health and Care Excellence (NICE) pioneered the Health Technology Assessment (HTA) processes and methodologies. Technology appraisals (TAs) focus on pharmaceutical products and clinical and economic data, which are presented by the product manufacturers to the NICE appraisal committee for decision-making. Uncertainty in data reduces the chance of a positive outcome from the HTA process or requires a higher discount.Objective To investigate the quality of clinical data (comparator, quality of life (QoL), randomised controlled trials (RCTs) and overall quality of evidence) submitted by the manufacturers to NICE.Design This retrospective evaluation analysed active TAs published between 2000 and 2019 (up to TA600).Methods For all TAs, we extracted data from the Assessment Group and Evidence Review Group reports and Final Appraisal Determinations on (1) the quality of submitted RCTs and (2) the overall quality of evidence submitted for decision-making. For single TAs, we also extracted data and its critique on QoL and comparators. Each category was scored for quality and analysed using descriptive statistics.Results 409 TAs were analysed (multiple technology appraisals (MTA)=104, single technology appraisal (STA)=305). In two-thirds of TAs, the overall quality of evidence was either poor (n=224, 55%) or unacceptable (n=41, 10%). In 39% (n=119) of the STAs, the quality of comparative evidence was considered poor, and in 17% (n=51) unacceptable. In 44% (n=135) of STAs, the quality of QoL data was considered poor, 15% (n=47) unacceptable, 33% (n=102) acceptable and 7% (n=21) as good. Over 20 years of longitudinal analysis did not show improvements in the quality of evidence submitted to NICE.Conclusion We found that the primary components of clinical evidence influencing NICE’s decision-making framework were of poor quality. It is essential to continue to generate robust clinical data for premarket and postmarket introduction of medicines into clinical practice to ensure they deliver benefits to patients.Data are available in a public, open access repository. All data used for this research are publicly available and can be rechecked by third parties. Data that we extracted for analysis can be accessed through a public repository zenodo.org.