This study explores the factors (principally evidential) that predict guidance recommendations by this NICE committee. There are three main types of recommendations: Standard/normal arrangements (can be done without restriction in the NHS); Special arrangements (can be done under certain conditions); and Research only.

The following data were extracted from all published pieces of Interventional Procedure Guidance (IPGs) produced by this committee: year, IPG number, recommendations, evidence base (numbers and types of included studies, numbers of included patients etc.). All data were extracted independently by two researchers, and any disagreements clarified by consensus. Data were tabulated and descriptive statistics produced. Regression analyses will be performed using these data to identify any statistically significant predictors of recommendations.

IPG recommendations (n = 496); year range: 2003-2018. Proportion of IPGs by each recommendation: 50% Standard; 38% Special; 11% Research Only; 2% Do Not Do. Proportion of IPGs with highest level evidence (i.e. systematic review and/or RCT) by recommendation type: Standard = 64% (152/239); Special = 43% (77/180); Research Only = 48% (26/54); Do Not Do = 75% (6/8). Mean numbers of patients by recommendation type: Standard = 7,838; Special = 3,935; Research Only = 2,423. There is also a clear trend over time: Standard recommendations decrease for all IPGs from 63% in 2003-2009 to 40% in 2014-2018; and the evidence threshold for Standard recommendations increases over time from 56% based on systematic reviews and/or RCTs in 2003-2009 to 85% in 2014-2018; mean numbers of patients per Standard recommendation also increase from 2,002 to 6,098 over this period.

Higher levels evidence and numbers of patients increase the likelihood of the most positive recommendation. However, this evidence might still lack sufficient quality or certainty to answer a policy question. The evidence threshold to achieve a Standard recommendation has also increased markedly over time. As with other NICE committees, factors other than cost and perceived hierarchies of evidence clearly act as drivers of decisions.