Basic↦Statistic Methodology↦Study Design↦Interim Analysis
What is it? Why is it important?
An Interim Analysis (IA) is a planned evaluation of the data collected in an ongoing study before its completion.
The goal of an IA is to assess the safety, efficacy, and/or futility of a study. The aim is to provide grounds for the decision to modify, continue, or terminate a study.
IA results are reported to a Data Safety Monitoring Board (DSMB), also referred to as an Independent Data Monitoring Committee (IDMC).
Upon an IA assessment, the DSMB recommends to the SP-INV to either:
- Continue the study as planned
- Continue the study with modifications
- Put enrolment (participant recruitment) on hold pending further DSMB recommendations
- Terminate the study early
A study may terminate early due to the study`s:
- Futility: a high likelihood that the intervention is ineffective or has only minimal benefit
- Safety: unacceptable safety risks to participants (e.g. an unacceptable risk-benefit ratio)
- Efficacy: A significant beneficial treatment effect with study success being declared early
More
An IA is recommended when:
- The effect of the treatment is unknown and may be:
- Particularly beneficial (efficacy)
- Very limited or null (futility)
- Deleterious (safety)
- The sample size estimation is based on uncertain pre-study assumptions, which can be re-evaluated using study data.
IAs implemented during study conduct must be approved by the Ethics Committee (EC), and if applicable Swissmedic (risk-category B and C studies). Special circumstances may dictate that an IA be performed that was not planned at study start. In that case, the protocol must be amended and approved prior to performing the IA.
What do I need to do?
As a SP-INV:
- Describe planned IA(s) in the study protocol, include the:
- Rational (e.g. for safety-, efficacy reasons)
- Number of planned IAs during study conduct
- Timing of IA(s) (e.g. after 25% and/or 50% of participants did complete the study)
- Statistical method (e.g. safety assessment based on AE/SAE frequency between intervention and control group)
- Stopping criteria: define study termination criteria (e.g. SAEs/SAE-Medical Device in the intervention group as compared to the control group exceeds a pre-determined threshold)
- Appoint members of the DSMB (i.e. describe procedures in a DSMB charter)
- Appoint a study independent statistician who performs the IA and reports the results to the DSMB (e.g. ideally blinded to treatment allocation)
A statistician is blinded to the study when he/she does not know the study`s treatment allocation (i.e. which study participants are in the treatment group and which are in the control group).
More
Know that conducting multiple tests increases the likelihood of observing statistically significant results by chance, thereby inflating the Type I risk.
Depending on the type and number of interim analyses, the sample size calculation may thus need to be adapted to maintain the overall statistical power and control the Type I error rate.
Where can I get help?
Your local CTU↧ can support you with experienced staff regarding this topic
Basel, Departement Klinische Forschung, CTU, dkf.unibas.ch
Lugano, Clinical Trials Unit, CTU-EOC, www.ctueoc.ch
Bern, Clinical Trials Unit, CTU, www.ctu.unibe.ch
Geneva, Clinical Research Center, CRC, crc.hug.ch
Lausanne, Clinical Research Center, CRC, www.chuv.ch
St. Gallen, Clinical Trials Unit, CTU, www.kssg.ch
Zürich, Clinical Trials Center, CTC, www.usz.ch
References
ICH GCP E6(R2) – see in particular guidelines
- 5.4.1 Interim report
- 6.9.1 Statistics, interim analysis
ICH Topic E9 statistical Principles for Clinical Trials – see in particular
- 4 Trial conduct considerations
- 4.5 Interim analysis and early stopping
ISO 14155:2020 Medical device (access liable to costs) – see in particular sections
- 6.2 Risk management
- Art.7 Statistical design and analysis