When Does Paper Make Sense for Clinical Outcome Assessment Data Collection?

The seminal paper that served as the essential basis for what we now term electronic clinical outcome assessment (eCOA) was published in 2002.¹ While the adoption of electronic methods to authentically capture a trail participants health status continues to increase, there remains a persistent reversion to paper methods in some scenarios. When viewed through the lens of ALCOA++ principles, it’s exceptionally difficult to defend paper usage for COA data, especially where primary or secondary endpoints are involved. Some anecdotal evidence suggests that the rate of paper COA adoption is actually increasing.

Separate discussions held amongst the members of the Critical Path Institute’s eCOA Consortium and Patient-Centered Evidence (PCE) Consortium (formerly the PRO Consortium) have endeavored to try and understand why this is happening. This resulted in the topic serving as the focus of the latest eCOA Forum held in person in Washington D.C. on April 20, 2026, which saw a multistakeholder panel of representatives from the eCOA Consortium, PCE Consortium, and FDA try and answer the titular question: when do paper COAs (pCOAs) make sense?

The discussion was wide-ranging, candid, and at times surprising in its honesty. What emerged was not a simple verdict, but a nuanced picture of an industry still navigating the gap between what it knows to be best practice and what operational realities often demand.

Is this really happening?

The opening question—whether paper-based COA data collection is genuinely increasing—produced an immediate and telling response: nobody could say for certain, because there is no single source of truth. Some voices pushed back on the premise—eCOA has been established as the standard within many organizations, and there is no hard data confirming a reversal of that trend. An analogy offered was that the early drive toward electronic collection was an "acute phase" of advocacy, a period of active persuasion. That phase may be over, not because any argument has been lost, but because the field has matured enough to allow more critical, case-by-case thinking. The alarming aspect of this, from my own perspective, is how the argument for superior data quality can be bypassed in that “critical thinking” phase.

That said, the panel explored how the perception of paper's rise may stem from regional expansion strategies, where certain geographies lack the infrastructure or regulatory familiarity to support eCOA, leading teams to default to paper for those sites. When that happens program-wide, it can look like a preference for paper when it is really a pragmatic compromise.

The real drivers of paper

Several interconnected drivers were identified, and none of them reflect a principled preference for paper over electronics

Predictably, cost was cited most frequently and most forcefully. The expense of implementing an eCOA system is quantifiable and visible in a way that paper's hidden costs—monitoring, query resolution, source data verification, delayed data entry, re-training—are not. When a single-site study with twenty participants receives an eCOA quote of several hundred thousand dollars, the decision to use paper is not irrational; it is simply responding to incomplete financial data. There was broad agreement that a robust framework for calculating the true total cost of paper-based collection, including the downstream consequences, would be a powerful tool for making the case to budget holders. This is something the C-Path team is currently exploring.

Trial complexity and size were closely related factors. Small patient populations, short study durations, and rare disease programs present a genuine challenge to the cost-benefit calculation. One participant described managing a thirty-day trial with twelve patients and acknowledged that the economics simply did not support eCOA, regardless of the data quality arguments. Startup timelines is also a significant pressure point. The early weeks of a trial are among the most—if not the most—stressful, and eCOA implementation takes time. Paper, for all its limitations, can be deployed quickly. When teams are under pressure to open countries and enroll participants, speed can override quality considerations.

The group candidly explored how eCOA provider capabilities and past experiences may significantly influence thinking, but the same applies for all electronic systems used in clinical trials. There have unquestionably been cases where eCOA providers have overpromised and underdelivered, and if this occurs more than once, sponsor teams can become reluctant to re-engage with electronic solutions entirely, defaulting to paper not out of preference but out of accumulated frustration. The complexity of certain COA measures is another practical obstacle. Certain highly complex assessment tools, particularly in neurology and cognitive conditions, have proved difficult or expensive to digitize. In some cases, eCOA providers have themselves suggested paper as the more practical option for particularly intricate instruments.

The cost of paper no-one calculates

One of the sharpest observations in the discussion concerned the asymmetry in how costs are perceived. The upfront costs of eCOA are explicit and auditable. The costs of paper are diffuse, delayed, and often invisible to those making the initial decision. So, what does paper actually cost? Panelists pointed to the burden of source data verification, monitoring visits to check ALCOA compliance, the risk of data entry errors when transcribing to an electronic data capture system, the inability to conduct real-time data monitoring, and the regulatory risk of submitting data of uncertain quality for a key endpoint. Several participants also highlighted that clinical teams have, in many cases, lost the institutional knowledge of how to run a paper-based study properly. Protocols are now written with electronic collection in mind, and the practical and operational requirements of paper—when and how patients complete diaries, how materials are designed, how data is retrieved and verified—are no longer well understood. The result can be paper studies that are neither well-designed nor well-executed.

Defending the value of COA itself

An unexpected thread in the discussion concerned newer entrants to the field. A generation of clinical researchers who have only ever known eCOA as the default are now, in some cases, questioning not just the method of collection but the value of patient-reported outcomes as a whole, flying in the face of clear regulatory guidance around patient-focused drug development.² Faced with the complexity of setting up electronic data collection, some are asking whether outcomes data should simply be entered directly into the electronic data capture system by site staff—effectively eliminating the patient voice from the data.

This, panelists agreed, represents a step backwards that goes well beyond the paper versus electronic debate.

So, when does pCOA make sense?

The panel did reach a degree of consensus on the circumstances where paper-based collection remains a reasonable choice. These include single-site or very small studies where eCOA costs cannot be justified; regional expansion into areas where electronic infrastructure is genuinely insufficient; early-phase exploratory work where rapid iteration is more valuable than structured data capture; and specific patient populations, such as certain pediatric groups or those with severe illness, where electronic completion presents genuine practical barriers.

Importantly, these were framed as exceptions requiring explicit justification, not defaults. The question, as one participant put it, should always be: if eCOA were free and instant, would we still choose paper? In almost every case, the honest answer is no.

What could we do differently?

Earlier engagement between eCOA providers and clinical teams—at a program level, not just study-by-study—is a must, and would allow issues to be anticipated and resolved before they create pressure to default to paper. More standardized, pre-built instrument libraries would reduce build times and costs for common assessments. Better education across clinical operations, regulatory affairs, and budget holders about the true cost of paper would shift the terms of the debate. And greater transparency from providers about their real capabilities and limitations would reduce the likelihood of expensive, trust-damaging failures. As one participant splendidly summarized, if the eCOA experience is consistently excellent, the argument for paper largely disappears.

The panel closed without resolving the debate, which was the honest outcome. We did align on the fact that paper has a role, and electronic collection is better in most circumstances. The gap between those two truths is where most of the real work remains to be done.

Scottie Kern, executive director, eCOA Consortium

References

1. A. Stone et al. Patient non-compliance with paper diaries. BMJ 2002;324:1193. https://doi.org/10.1136/bmj.324.7347.1193
2. FDA Patient-Focused Drug Development Guidance Series for Enhancing the Incorporation of the Patient’s Voice in Medical Product Development and Regulatory Decision Making. https://www.fda.gov/drugs/development-approval-process-drugs/fda-patient-focused-drug-development-guidance-series-enhancing-incorporation-patients-voice-medical