Delivering an Interim Analysis During a Pandemic from a Data Management perspective
When the World Health Organisation (WHO) declared the COVID-19 pandemic on 11th March 2020, countries the world over implemented restrictions to manage this accordingly. As well as challenges faced by people and communities around the globe, the pandemic presented unique challenges for the clinical trials community, with many existing studies being either temporarily paused and / or adapting their process.
Interim analysis provides several options and opportunities for sponsors, including modifications of the trial design and adaptive designs, monitoring the safety of the trials with DSMB’s, post marketing surveillance and long-term follow up and monitoring if the trial is going to go ahead as originally intended (i.e. sample size). The statistical output for an Interim Analysis may include (but is not limited to) tables, figures and listings, data listings, interim study reports, IDMC and DSMB requirements. The Interim Analysis may be blinded, partially blinded, unblinded, or will use a “dummy” randomisation schedule.
The roadmap to database lock for an interim analysis is as follows.
- Data Capture
Data is captured at site in the eCRF, via eConsent, ePRO, eDiaries or by wearables or other devices eCRF i.e. ePRO.
- Data Review
Data is checked across data items, modules, visits and subjects to ensure that all pertinent data are available, consistent and of good quality.
Medical text and medications are coded, further facilitating the reporting of study data.
- SAE Reconciliation
SAE Data recorded in the trial r database is compared and reconciled with data recorded in the Patient Safety Database.
- Third Party Data Handling
Third party data is securely transferred and reconciled with the data captured in study database, identifying and resolving any issues that arise.
- Quality Control
The data and deliverables are independently reviewed, ensuring study procedures have been appropriately followed.
When running an Interim Analysis during a global pandemic, there are several challenges that can be faced. For example, local restrictions and strict national lockdown measures can see a disruption in study activities due to limited, or indeed no, access to the trial site for staff and for patients. There can also be challenges surrounding communications with the trial site due to issues surrounding working from home.
Furthermore, challenges surrounding country-specific regulations can mean different actions being required at various sites. To work towards mitigating this, the FDA and EMA issued guidance - “A listing of all participants affected by the COVID-19 related study disruption by unique subject number identifier and by investigational site, and a description of how the individual’s participation was altered”
During the decision-making around continuing or halting a trial it is important for sponsors to assess the impact of the pandemic from a patient’s perspective – what stage the of treatment the patient is in, any change in drug supply as a result of national lockdowns and also how data from patients is collected – changes in visits from in person to phone, completion of diaries or questionnaires and the impact this can have on the data collected.
If on-site monitoring visits are no longer possible, sponsors must document any inability to access or delayed monitoring of a site. Risks due to a lack of Source Data Verification (SDV) must be assessed and other options such as remote or centralised monitoring should be considered.
One of the main challenges faced when delivering an interim analysis during a pandemic stem from missing data. Delayed visits and patients lost to follow up can have a negative effect on quality and it is important for sponsors to question how these quality issues might impact the integrity of the analysis.
So, when faced with these challenges, how do we make an action plan to mitigate these and bring a successful interim analysis?
From a data collection perspective, we can implement a method for tracking COVID deviations – adding in an additional eCRF page and query or protocol deviations. We can update documents to include eCRF completion guidelines and training, monitoring plans, QC processes and SOP deviations. Sites can specify the forms affected so it’s clear of the impact that COVID had on that specific visit. This means in the analysis, the impacts of COVID will be clearly outlined and can then be used to justify certain changes, if needed.
A data review meeting ahead of an analysis should be organised to assess the status and integrity of the data This meeting will work to prioritise what has to be done in order to proceed with an analysis including a medical input on safety. Protocol deviations can be used as a way to assess where COVID presents an issue and potential impact on integrity. This can include delayed visits, missed visits, telephone visits in lieu of in-person visits and missed assessments.
Process changes should be implemented to allow for changes as a result of the pandemic. As an alternative approach to data collection, ePRO can potentially be utilised, allowing subject the opportunity to self-report information in diary form. Another benefit of ePRO is that data can be collected immediately, removing any risk of disruption from COVID. ePRO is likely to be utilised in the future, regardless of restrictions or limitations from the pandemic. A local nurse/ healthcare practitioner can be included to support home visits, providing blood collection, assessments and medication to patients.
Visualizations can play a key role in delivering an interim analysis during a pandemic. When teams are relying on information provided in tabular formats, it can be challenging to identify outliers and usual activity. However, visualizations are extremely useful, saving time and effort without compromising on quality, especially when they are interactive, targeted to specific areas of interest and have built-in algorithms that prioritise or alert users to specific items of interest.
Also playing a key role in risk based and centralised monitoring is dashboards, which can be intuitive, informative and offer real-time understanding of clinical, audit and query data sources. Critical to quality data items and thresholds are identified by the study team at set up and programmed for presentation in a dashboard. The dashboard provides the team with a tool for real time monitoring of the study progress and facilitates early intervention and mitigation activities for areas of concern.
PHASTAR’s bespoke visualization tool, Phizual, is a suite of interactive visualisations and dashboards which sit across the clinical data, its meta-data and audit data trials. We have the expertise to custom-build visualisations and dashboards, working iteratively with study teams to maximise the value and generate meaningful insights from their data and subsequent analyses.
The standard approach to data collection and trial reporting was thrown into turmoil due to COVID. However, the need for flexibility in approach, whilst safeguarding the patient’s well-being and data integrity has led to an evaluation of existing technologies and a leap to develop and deploy new methods both in trial design and execution. Technology together with well thought through procedural strategies are making it possible for patients to take part in studies without the need to visit investigational sites. The tools are also available for investigators and clinical research teams to carry out their work remotely and have real-time visibility of the data. The saying ‘necessity is the mother of invention’ has certainly been relevant in this pandemic.