QALY (Quality Adjusted Life Years)

1. Description

QALY (Quality Adjusted Life Years) is the most used health index, where the time spent in a particular health state is multiplied by the Quality of Life (QoL) score in that state. The total QALY is simply the sum of all QALYs in all health states.[1][2] QALY combines both benefit criteria and risk criteria, and so can be directly used in benefit-risk assessment for the comparison of different options. QALY outcomes in a benefit-risk assessment can also be separated into health improvement with "QALY gain" and health adverse impact with "QALY loss".

2. Evaluation

2.1 Principle
  • QALYs are widely used in cost-effectiveness analysis of drug therapy.
  • QALYs provide a 'common currency' for both benefit and risk is crucial in evaluating the benefit-risk balance.
  • Standard statistical analysis can be applied to estimate the benefit-risk balance, as well as the difference of these balances between different treatments.

2.2 Features
  • Benefit and risk are integrated, and so is the time dimension.
  • QALY accommodates multiple criteria and multiple options.
  • In principle, a sensitivity analysis can be performed on the derivation of QoL, since weights or utilities of different criteria appear only in the derivation.
  • Incorporating multiple sources of QALYs data requires meta-analysis given that they are derived in the same way in all sources.

2.3 Visualisation
  • Incremental QALYs have been presented here

2.4 Assessability and accessibility
  • In principle, stakeholders' preference in this approach is reflected in QALYs.
  • The QoL usually is derived using standardised tools such as EQ-5D, before QALYs are calculated.
  • At population level, QALYs are considered as the average preference.
  • At population level, QALYs may be 'risk-neutral'.

3. References

[1] Weinstein MC, Torrance G, McGuire A. QALYs: the basics. Value Health 2009 Mar;12 Suppl 1:S5-S9.
[2] Pliskin JS, Shepard DS, Milton CW. Utility Functions for Life Years and Health Status. Operations Research 1980 Jan 1;28(1):206-24.