Handling an uncertain control group event risk in non-inferiority trials: non-inferiority frontiers and the power-stabilising transformation

Quartagno, Matteo; Walker, A. Sarah; Babiker, Abdel; Turner, Rebecca; Parmar, Mahesh; Copas, Andrew; White, Ian R.

doi:10.1186/s13063-020-4070-4

Cited by 21 publications

(29 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If it is considered, then a correction for type I error rate inflation must be used. Some advice for simulation-based correction methods are given in Quartagno et al 22 The issues we describe are particularly important for non-inferiority trials since changing the analysis scale requires redefining the non-inferiority margin. In superiority trials, the neutral comparison values (zero for the RD and one for the RR or OR) do not change when switching the analysis scale such that no large differences between scales are expected.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Choosing and changing the analysis scale in non-inferiority trials with a binary outcome

Quartagno

Böhringer

et al. 2021

Clinical Trials

Self Cite

View full text Add to dashboard Cite

Background The size of the margin strongly influences the required sample size in non-inferiority and equivalence trials. What is sometimes ignored, however, is that for trials with binary outcomes, the scale of the margin – risk difference, risk ratio or odds ratio – also has a large impact on power and thus on sample size requirement. When considering several scales at the design stage of a trial, these sample size consequences should be taken into account. Sometimes, changing the scale may be needed at a later stage of a trial, for example, when the event proportion in the control arm turns out different from expected. Also after completion of a trial, a switch to another scale is sometimes made, for example, when using a regression model in a secondary analysis or when combining study results in a meta-analysis that requires unifying scales. The exact consequences of such switches are currently unknown. Methods and Results This article first outlines sample size consequences for different choices of analysis scale at the design stage of a trial. We add a new result on sample size requirement comparing the risk difference scale with the risk ratio scale. Then, we study two different approaches to changing the analysis scale after the trial has commenced: (1) mapping the original non-inferiority margin using the event proportion in the control arm that was anticipated at the design stage or (2) mapping the original non-inferiority margin using the observed event proportion in the control arm. We use simulations to illustrate consequences on type I and type II error rates. Methods are illustrated on the INES trial, a non-inferiority trial that compared single birth rates in subfertile couples after different fertility treatments. Our results demonstrate large differences in required sample size when choosing between risk difference, risk ratio and odds ratio scales at the design stage of non-inferiority trials. In some cases, the sample size requirement is twice as large on one scale compared with another. Changing the scale after commencing the trial using anticipated proportions mainly impacts type II error rate, whereas switching using observed proportions is not advised due to not maintaining type I error rate. Differences were more pronounced with larger margins. Conclusions Trialists should be aware that the analysis scale can have large impact on type I and type II error rates in non-inferiority trials.

show abstract

Section: Discussionmentioning

confidence: 99%

“…If it is considered, then a correction for type I error rate inflation must be used. Some advice for simulation-based correction methods are given in Quartagno et al 22 …”

Section: Discussionmentioning

confidence: 99%

Choosing and changing the analysis scale in non-inferiority trials with a binary outcome

Quartagno

Böhringer

et al. 2021

Clinical Trials

Self Cite

View full text Add to dashboard Cite

show abstract

“…23 The non-inferiority margin can be reviewed if the event rate in the control arm differs from the original assumption within the non-inferiority frontier framework approach. 24 …”

Section: Methods and Analysismentioning

confidence: 99%

Randomised controlled trial to evaluate the effectiveness of using the RD-1-based C-Tb skin test as a replacement for blood-based interferon-γ release assay for detection of, and initiation of preventive treatment for, tuberculosis infection: RID-TB:Dx study protocol

et al. 2021

View full text Add to dashboard Cite

IntroductionThe predictive utility for incident tuberculosis (TB) of the purified protein derivative tuberculin skin test and region of difference 1 (RD1)-based interferon-gamma release assays (IGRA) is comparable; and either is recommended to test for latent TB infection (LTBI). Despite associated high costs of IGRA, sites participating in LTBI screening in many high-income settings pragmatically favour IGRA due to its higher specificity and simpler logistics. A new RD1-based skin test, C-Tb, could offer an acceptable and as accurate, cheaper alternative to IGRA. Evaluating the impact of C-Tb on process and patient-related outcomes would provide important information to help guide its use in LTBI testing strategies.Methods and analysisThis is a pragmatic multicentre, open-label, non-inferiority, randomised controlled trial. The trial will assess the initiation of LTBI treatment following a positive result of the randomised test as the primary outcome. Participants will be randomised to receive the C-Tb test (intervention) or IGRA (usual care, control) for initiation of treatment. We will enrol 1530 participants in England aged≥16 years who are eligible for LTBI testing and treatment according to UK guidance. In the C-Tb arm, skin induration will be assessed 2–3 days after intradermal injection and measured in millimetres of induration. Results of IGRA will be obtained in line with standard practice. Behavioural studies will explore people’s experiences, perspectives and preferences of LTBI testing and treatment. Economic analysis will estimate cost-effectiveness of changes to the diagnostic algorithm for LTBI. The protocol was developed with Patient and Public Involvement (PPI), which will continue throughout the trial.Ethics and disseminationEthics approval has been obtained from The NHS Health Research Authority (269485). We will share results of the trial in peer-reviewed journals and conferences.Trial registration numberEudraCT 2019-002592-34; ISRCTN17936038.

show abstract

“…Limitations to the design of noninferiority trials are set by the arbitrary definition of the noninferiority margin, 15,16 by uncertainty in estimates of benefit for the previously approved treatment regimen, 17 and by a design that focuses on a single therapeutic outcome. Factors such as toxicity and cost are relegated to secondary end points.…”

Section: Re-evaluation Of Failed Noninferiority Trialsmentioning

confidence: 99%

Near-Equivalence: Generating Evidence to Support Alternative Cost-Effective Treatments

Tannock

Ratain

Goldstein

et al. 2021

JCO

View full text Add to dashboard Cite

Handling an uncertain control group event risk in non-inferiority trials: non-inferiority frontiers and the power-stabilising transformation

Cited by 21 publications

References 22 publications

Choosing and changing the analysis scale in non-inferiority trials with a binary outcome

Choosing and changing the analysis scale in non-inferiority trials with a binary outcome

Randomised controlled trial to evaluate the effectiveness of using the RD-1-based C-Tb skin test as a replacement for blood-based interferon-γ release assay for detection of, and initiation of preventive treatment for, tuberculosis infection: RID-TB:Dx study protocol

Near-Equivalence: Generating Evidence to Support Alternative Cost-Effective Treatments

Contact Info

Product

Resources

About