The Society of Clinical Trials (SCT) conference was held virtually on 17th - 20th May 2021. SCT runs a trials-related multidisciplinary program of plenary sessions, topical sessions and contributed presentations. The theme for the 42nd Annual Meeting was "Advancing Rigorous and Ethical Trials in the Pandemic Era".
Principles of ASPIRE: Ethics, Equipoise and Conduct by Professor David Beard
Professor David Beard (Professor of Musculoskeletal and surgical sciences, Rosetrees Royal College Surgeons [England] Director of SITU [NDORMS]) presented at the conference alongside Professor Marion Campbell (University of Aberdeen) and Dr Sian Cousins (University of Bristol) with Dr Dean Fergusson (University of Ottawa) chairing the session.
In this session, they presented their findings from the new ASPIRE (Applying Surgical Placebo In Randomised Evaluations) guidance on the use of placebo surgical in randomised trials and the related checklist developed by an international group of methodologists, trialists, ethicists and surgeons. Additionally, related work which provides an up-to-date review of current practice and the limitations in it was also presented.
David's presentation addressed the key principles of the ASPIRE guidance on how to design a surgical trial with a placebo control. It addressed issues such as fidelity to the real procedure, mitigation of risk and related ethical and conduct issues. How the new guidance has come about was also covered.
Members of the SITU team attended SCT as a training opportunity and Yara Neves Silva (Clinical Trial Coordinator) and Akiko Greshon (Data Support Clerk) have kindly shared their learnings from the virtual conference.
Yara's thoughts on the conference
On the first day, I wasn't very lucky choosing the lectures to attend, as almost all of them had connection problems ranging from lags, bad resolution, or poor audio quality. Thankfully, the videos were available for 90 days after the event ended.
Overall I found the SCT virtual experience very useful in that I can review the videos later, checking details and important information that can be easily missed during the real-time presentation.
One of the most interesting sessions that I attended was the Ethics and Future of Human Challenge Trials. It is well known that controlled studies of infectious diseases in humans (CHIs) can be as ethical as any phase 1 trials in healthy volunteers. But what caught my attention in this presentation was the discussion around some ethical problems of CHIs trials. The first of them is whether a person infected with a disease that can be transmitted to others can have a right to withdrawal from a study at any time. Another issue is whether it is acceptable to expose their partners and/or third parties to an infectious disease outside of the study without consent to be exposed to that risk or full information of what the risks are. Further, the protection of the vulnerable population excluding them from research studies would limit safe and valuable information for this particular group. Finally, another unresolved issue is related to the upper limit risk of CHIs that can be acceptable.
Aki's thoughts on the conference
Contributed Session 3 - Data Management I
FROM CLINICAL TRIAL TO OBSERVATIONAL STUDY: LONG-TERM RETENTION IN THE TODAY AND TODAY2 STUDIES
“Supportive relationship with study staff was the great contributor leads to success of high retention rate!”
This session was more than data management and rather philosophy to me. The vital key point to succeed the high retention rate is to maintain participant engagement. Although the nature and process of SITU trials may be slightly different from the presented trial, I thought it is worth comparing two and suggesting some points that may help our retention rate. I’ve put my thoughts together...
TODAY trial (from SCT session)
- Description of Trial - Treatment options for Type 2 Diabetes in Adolescents and Youths
- Participants - 10-17
- Duration - 15 years
- Challenge - Approaches to participant engagement needed to be age-appropriate.
- Contributed study team members - Approaches to participant engagement needed to be age-appropriate.
Retention strategies and approaches
- Communication - Regularly communicated with participants – not only about study-related but also normal conversation and check-ups. These participants were struggling with lives in general so stable presence of study staff was very important
- Dissemination - Study updates and results were distributed via newsletters/brochures/letters
- Incentives - Provide incentives for attendance such as monetary honorarium, retention gifts, participation appreciation certificate
- Study visits - Reimbursement for travel cost, re-schedule and be flexible (keep re-arranging if necessary), offer remote phone visit if allowed
How could this relate to SITU?
- Challenge and contributed study team members - The challenge we face is the retention of PROMS. Study team often needs an independent ‘phone-call’ team, who can be the key to a high retention rate.
- Communication and Dissemination - Only the questionnaire is sent according to time point. Can we regularly send study updates/results to all participants – something similar to a site newsletter?
- Incentives - ACL SNNAP voucher seems to be effective (young age group); NINJA encouraged small children with stickers and small toys for their clinical assessment attendance; NINJA encouraged site staff to be motivated by giving them a certificate. A participant certificate for completing questionnaires won't cost anything and can be done easily.
- Study visits - Visits are normally arranged by study sites. Any budget for participant's reimbursement for travel cost (car park, bus, taxi)?
Contributed Session 7 - Study Co-ordination II
MAKING A DISTINCTION BETWEEN DATA CLEANING AND CENTRAL MONITORING BY SHARON LOVE
“Can you describe the difference between data management and data monitoring? Do we need both documentation? Can we combine these two activities?” I thought about this session from SCT conference when Heidi’s Data Management Standard Operating Procedure (SOP) was presented at the SITU session a few weeks ago.
Sometimes the definition and purpose for SOP documents can be rather ambiguous. I have a better understanding in data management and data monitoring after watching this session at the SCT conference and reading the paper: "Making a Distinction between Data Cleaning and Central Monitoring" by Sharon Love. Read the paper here.
This session and paper can give us a good understanding of what should be described and determined in the data management plan and data monitoring plan as well as what actions we need to take in those two separate areas.
The information below summarises the difference between these two roles:
- Data Cleaning: Data cleaning addresses problems with data such as incomplete, invalid or inconsistent data. When data is entered, most databases have an automated way of checking the data and flagging any issues. On a regular basis or before the Data Monitoring Committee (DMC) meetings, a member of the trial team will run a check on the participant data and query any out of place or missing values with the hospital sites. Before any interim or final analysis, these processes will be repeated. These are all data cleaning activities. The main action is to send out data clarification requests.
- Central monitoring: This is a way to centrally identify any issues with trial conduct such as inadequate processes or procedures not being followed through a lack of clarity in the protocol or active fraud. We look through centrally held data by site, to discover any strange patterns or features in the site's data entry or unacceptable data activity. This may result in data queries or may lead to dedicated communication with sites or an on-site monitoring visit. Central monitoring results are an indicator of the quality of a trial and sow due diligence. Any issue found during the central monitoring process should be followed up with the site directly and may result in actions such as training or making an on-site visit. Central monitoring only needs to be repeated periodically, the period depending on the trials' needs, requirements and assessment of risk. Sometimes central monitoring is done across sites, comparing data between the sites to show differences. In some instances, this may be done across trials run from the same organisation.. Central monitoring can include review of trial management data too, such as protocol deivations.
Below are a few snapshots from the presentation
Ways in which we could take this authors advice
“Central monitoring may be split into many tasks which are completed across time in a rolling pattern. Here is an example for a plan.”
Week 1 - Serious adverse events
Week 2 - Protocol Deviations
Week 3 - Case report form return rate
Week 5 - Serious adverse events
As a SITU data management staff (now thinking about NEON!), I would probably suggest something like this:
- Week 1 - CRF overdue rate
- Week 2 - Participant questionnaire return rate
- Week 3 - Missing data - find out the frequent missing category and result by site
- Week 4 - Site responding rate
Or rather than split into different weeks, it could possibly be performed by producing a summary report once a month. The following was quoted from their paper published in SAGE journals.
“Without a clear understanding of data cleaning and central monitoring, the trial team and site staff may spend time and effort inappropriately or wastefully. If these activities are not separated, they can each occur at the wrong time: data cleaning too rarely and central monitoring too frequently. Data cleaning needs to happen often. It is easier to clarify, correct or locate previously missing or out-of-range datapoints when the query is asked close in time to when the data were collected. Data cleaning needs to be done often so that the data are as good quality as possible for central monitoring to be effective. Central monitoring is most effective on cleaned data, otherwise teams will focus on individual data errors rather than required process changes, or an incorrect process may be missed due to poor quality data. Repeat central monitoring needs to happen periodically.”
I totally agree with this comment. The Data Monitoring Plan should include the idea of having trial specific instructions (TSIs) and making a data entry cribsheet. The activities should be simple and on a regular basis.
I would say data cleaning/management is more like looking after your house! We just need to keep up the cleaning and tidying.