OBJECTIVES: Value-based assessment will be introduced into National Institute for Health and Care Excellence (NICE) appraisals where the willingness-to-pay above £20,000 will depend upon the wider societal impact (absolute shortfall of quality-adjusted life years [QALYs]) and burden of illness (proportional QALY shortfall). The objective of this research was to re-evaluate past appraisals and determine whether there was a relationship between the absolute and proportional QALY shortfalls and the NICE decision. METHODS: Final incremental cost-effectiveness ratios (ICERs) were identified from NICE single technology appraisals published between January 2013 and May 2014. The age that treatment commenced was taken from manufacturer submissions and combined with the average life expectancy in the UK to calculate the discounted QALYs accrued by a healthy person (X). Discounted QALYs accrued by current treatment were extracted from manufacturer submissions (Y). Consequently, absolute (X-Y) and proportional (absolute/X) QALY shortfalls were calculated for each condition. Logistic regression (LR) was performed on the data. Appraisals were excluded if the manufacturer submission was missing. RESULTS: Of the appraisals assessed, 23 appraisals met the selection criteria; 13 were recommended and 10 not recommended by NICE. The LR confirmed that the probability of an intervention being recommended was reduced significantly if the ICER was >£30,000 (p<0.05). Taking into account the ICER, LR analysis demonstrated that neither the proportional nor absolute QALY shortfall had a significant effect on recommendation by NICE (p>0.05). CONCLUSIONS: These results suggest that in recent appraisals, proportional and absolute QALY shortfalls have had no significant effect on the NICE committee’s decisions, and that the introduction of value-based assessment may therefore cause a substantial change in the future outcomes of HTA processes. It is recognised that further appraisals before 2013 should be evaluated to confirm these results, as the sample size was small.