The importance of clinical design in ClinRO eCOA migration
An expert discussion
- Written with – Jan Malecki, Ph.D. – Associate Director eCOA Clinical Science at Clario, a part of Thermo Fisher Scientific
Clinician-reported outcome assessments (ClinROs) are central to many clinical trials, particularly in neuroscience, psychiatry and rare disease. Yet moving a ClinRO from paper to a screen does not automatically make it easier to administer or more consistent.
Complex scoring rules, branching logic and differences in rater experience can introduce variability into endpoint data. We spoke with Jan Malecki Ph.D., Associate Director eCOA Clinical Science, Clario a part of Thermo Fisher Scientific, about what sponsors should consider when implementing eClinROs and how clinical expertise can be embedded directly into assessment design.
What is a ClinRO, and how does it differ from a patient-reported outcome?
Jan: A clinician-reported outcome, or ClinRO, is an assessment completed by a trained healthcare professional or clinical rater, based on observation, interviews, clinical judgment or a combination of these.
Unlike a patient-reported outcome (PRO), where the participant reports directly, a ClinRO requires the rater to interpret clinical information before assigning a score. Because this introduces greater potential for variation between raters, a well-designed eClinRO should preserve the instrument’s clinical meaning while supporting consistent, standardized administration across sites and visits.
Why does ClinRO migration require more than simply digitizing the paper assessment?
Jan: A direct electronic reproduction may capture the same responses, but it can preserve many of the same opportunities for error.
A paper assessment may require the rater to remember detailed instructions, apply basal or ceiling rules, refer to previous answers, perform calculations or compare the current rating with baseline information. Some instruments also use normative data, age-based calculations, diagnostic algorithms or several domain scores.
Simply putting those items on a screen still leaves the rater to manage that complexity.
A clinically designed eClinRO can automate appropriate navigation and scoring, provide guidance when it is needed and flag potentially inconsistent responses before submission. That is the difference between simply digitizing an assessment and using technology to support standardized administration.
Do not assume that putting a ClinRO on a screen makes its administration standardized. A clinically designed eClinRO can automate appropriate navigation and scoring, provide guidance when it is needed and flag potentially inconsistent responses before submission. The electronic implementation should reflect the clinical intent of the instrument, the way raters actually work and the points where errors are most likely to occur. Bringing clinical science, technology and real-world assessment experience together can support more consistent administration and help protect endpoint reliability.
What role do Clario’s clinical scientists play?
Jan: Our clinical scientists build the instrument, configure and test the workflow, and refine the implementation as it develops.
Historical review of query data can also show where raters repeatedly encounter problems. If a response pattern frequently leads to a query, for example, the electronic form can prompt the rater to confirm that the responses reflect the intended rating.
Where appropriate, we work with scale authors, licensors and key opinion leaders to clarify the intended administration. The objective is to translate clinical intent into a practical electronic workflow without changing the construct being measured.
Is there evidence that this approach can reduce errors?
Jan: Across several CNS assessments, we show fewer scoring discrepancies or administration failures with eCOA than with paper.
For the Montgomery–Åsberg Depression Rating Scale, rater discrepancy declined from 97% with paper administration to 30% with electronic administration. Reviews with two discrepancies declined from 90% to 13% – an observed relative reduction of up to 86%.
Lower discrepancy rates were also observed across four assessments used in Alzheimer’s disease trials: the CDR, ADAS-Cog, ACDS-ADL-MCI and MMSE.
For the CAPS-5 in PTSD studies, poor-quality or failing administrations declined from 29% with paper to 12% with Clario’s electronic adaptation.
These findings should be considered alongside the methodology and limitations of each analysis. However, the consistent direction of the results suggests that thoughtful electronic design can help reduce predictable administration and scoring problems.
Can you give an example of how clinical design prevented an implementation problem?
Jan: I can give a case study example.
The challenges:
The TETRAS assessment required high-quality video capture so a third-party reviewer could centrally evaluate tremor severity. However, site teams historically struggled with the workflow, which requires raters and videographers to coordinate closely or the video will not always capture the movements or angles needed for reliable review. If the tremor is not clearly visible, the assessment becomes difficult or impossible to score. Additionally, training materials contain critical administration guidance, but raters do not always accurately recall or apply those details during a live participant visit.
The ClinRO design:
eCOA Clinical scientists redesigned the form to support the assessment and reliable video capture in real time. They included videographer instructions directly in the form, including example images to show what needed to be captured. They also incorporated verbatim participant instructions from the manual, helping raters administer the scale consistently with validated practices. In collaboration with KOLs from the TRG group, the team clarified instructions that were known to be difficult to apply in clinical settings. Logic was also added to improve accuracy for a commonly mis-scored item with nuanced scoring criteria.
The impact:
This design reduced the risk of unusable videos, mis-scored items, and delayed central review. It also decreased site burden by making the workflow clearer, less burdensome and helping raters and videographers coordinate more effectively during administration.
How does Clario’s assessment library support ClinRO quality?
Jan: Clario’s library includes more than 1,000 assessments, but its value is not simply its size.
A strong library preserves the work that has gone into the implementation: the clinical workflow, scoring logic, guidance, testing, translations and version control. When a sponsor uses an existing controlled assessment, it can be configured more efficiently than rebuilding it for every study.
It also supports consistency across a development program. Sponsors can reuse the same implementation from early development through later-phase trials, helping reduce unnecessary variation between studies.
What should sponsors ask an eCOA provider about ClinRO design?
Jan: Sponsors should ask who makes the clinical design decisions and whether those experts remain involved during the build.
They should also ask how the provider handles complex branching, basal and ceiling rules, scoring algorithms, historical error patterns, rater guidance and version control.
Most importantly, ask for evidence. A provider should be able to explain not only which features are available, but how those features address known assessment risks and what effect they have had on discrepancies, administration quality or operational burden.
Written with
Jan Malecki, Ph.D.
Associate Director, eCOA Clinical Science