The increasing use and acceptance of alternative statistical approaches to indirect comparison in the NICE HTA submission process

Pooley N, Marjenberg Z, Embleton N, Langham S. The increasing use and acceptance of alternative statistical approaches to indirect comparison in the National Institute of Health and Care Excellence (NICE) health technology assessment (HTA) submission process. Value Health 2019:22(Suppl 3):S814

Abstract

Objectives: Indirect comparisons can be central to establishing efficacy estimates used in the HTA process. While traditional network meta-analysis (NMA) is considered to be the gold standard when performing indirect comparisons, it is not always possible. Instead, alternative statistical approaches such as matchingadjusted treatment comparison (MAIC) and simulated treatment comparison (STC) have been used. We reviewed technology appraisals (TAs) from NICE to assess how the use and acceptance of different approaches to indirect comparison have changed over time.

Methods: TAs that included a MAIC or STC were identified on the NICE website using the following search terms: matched adjusted, matching adjusted, MAIC, simulated treatment, STC, adjusted indirect. The statistical methods, clinical evidence, assessments and recommendations were extracted and summarised.

Results: Thirty-six TAs were identified on the NICE website: 21 included a MAIC (published 2014–2019), ten an adjusted indirect comparison (published 2006–2019), and four a STC (published 2015–2018). Using the MAIC approach has increased over time: 17/21 were published from 2017 onwards. MAICs have predominantly been used in oncology HTAs (19/21), often where only phase II single-arm data are available. There was a mixed response from NICE regarding the use of these alternative indirect comparisons. In some cases, NICE reported a preference for naive comparisons. In others, NICE requested such an approach be used where no other comparable data are available. NICE decisions on these submissions often included a market access agreement necessitating the collection of additional data.

Conclusions: There has been a large increase in the use and acceptance of alternative statistical approaches to indirect comparison, most notably MAICs. The use of MAICs is becoming central in the preparation of NICE submissions where traditional NMA is not possible, particularly in the area of oncology.