Patient engagement in clinical research has a measurement problem, and it is not what most people assume. The front end of the recruitment journey is measured carefully; impressions, page views, questionnaire starts, are tracked, reported, and optimized. The back end is measured too; enrollments, randomizations, retention rates, and drop-off. But between those two measurement zones sits a gap that very few programs are equipped to see clearly, let alone act on. That gap belongs to the patients who engaged, who raised their hand, who tried to qualify, and couldn't, or wouldn't, for reasons that were never connected in a way that drove any change. The gap in reporting is not only what happens in the middle, but also the inability to connect those two zones in a meaningful way when patient recruitment activity is spread across multiple vendors, platforms, and data sources with no shared infrastructure to harmonize the picture.
In this post we will discuss what that gap in clinical trial recruitment contains, why it matters strategically, and how disqualification analysis (DQA) — understood in its full scope across three distinct signal sources — gives clinical research teams the intelligence to close it.
Disqualification analysis is not yet a standard term in clinical trial recruitment. It names something the industry has not had a consistent framework for and building that framework is what makes it actionable.
At its most useful, disqualification analysis is the synthesis of multiple signal streams, each capturing a different stage of the patient journey, and answering a different question about where and why patients fall out of the recruitment funnel. Carebox structures DQA across three distinct sources and understanding all three is what transforms it from an operational metric into a strategic intelligence tool.
The first signal source is where Carebox's approach diverges most sharply from how the rest of the industry operates. Most recruitment programs that analyze eligibility failures are limited to patients who completed a trial-specific pre-screener for that study. They can tell you why patients didn't qualify for the trial in front of them. What they cannot do is analyze eligibility patterns from all patients who completed condition-based questionnaires across an entire network, regardless of which specific trial they encountered first.
This is the layer that the Carebox Connect Network Intelligence report captures. Across the Carebox Connect Network, patients complete condition-based questionnaires through trusted sources including advocacy organization websites and the public Carebox Connect site. When there is no match, the system records which specific criterion caused the failure, primary diagnosis, metastatic status, age, treatment history, ECOG performance score, and so on, and that record becomes part of a dataset that spans the entire network.
The question this signal answers is: across all the patients who tried to qualify, who couldn't, and why?
Once a patient clears the questionnaire and is referred to a study site, a second disqualification layer begins — one that questionnaire data alone cannot see. The site may determine that the patient is still not a fit, based on chart review, additional eligibility screening, or an in-person assessment that reveals something the pre-screener did not capture. But screen failures are only one part of the story. Patients may never be contacted after referral, may not schedule a screening visit, or may withdraw before reaching that point. Each of those outcomes tells a different story about where the process broke down.
Every site captures some version of this data in isolation. What most programs lack is the ability to normalize it. When site outcome data is collected in a structured, consistent format across trials and across different site solution vendors, patterns that would otherwise be invisible become legible. A site that is consistently failing to contact referred patients is a different problem than a site with elevated screen failures on a specific eligibility criterion. A screen failure pattern that repeats across multiple sites is a different problem than one concentrated at a single location. Carebox's referral management platform tracks site outcomes structurally — contact status, screening visit scheduling, screen failure, enrollment — in a way that makes those distinctions visible and comparable across the program.
The question this signal answers is: of the patients who moved forward, what happened at the site, and what does that pattern reveal?
“Measurement without action is just documentation."
- Kyle Smith, VP of Sales at Carebox
The third signal source is the most frequently overlooked, and in some ways the most revealing. It captures the reasons patients do not move forward that have nothing to do with eligibility criteria. Those reasons matter as much as eligibility mismatch for understanding where a recruitment program is losing people.
A patient may have matched on every inclusion and exclusion criterion and still be unable to participate because the study site is three hours away, because their physician has advised against trial participation, because they are already enrolled in another study, or because the visit schedule is incompatible with their work or caregiving responsibilities. These barriers can be surfaced through multiple mechanisms including, navigation conversations, patient-facing tools, and other touchpoints that capture patient-reported context beyond what a questionnaire records.
These are not the same problem as eligibility mismatch, and they do not cancel out the eligibility data. A patient can have both an I/E disqualification reason and a logistical barrier. A patient who matched completely may still not proceed for reasons that are entirely non-clinical. This is why DQA is most usefully understood as "disqualification plus barriers" — not just eligibility mismatch. Signal 3 captures the reasons patients, including those who qualified, do not complete the referral process, and it surfaces those reasons in a way that structured eligibility data never will.
The question this signal answers is: beyond eligibility mismatch, what else is stopping patients from completing the referral process?
Considered separately, each of these three signal sources is useful. Considered together, they create a picture of the recruitment funnel that most programs currently cannot see.
Signal 1 tells you who tried to qualify and couldn't, and which eligibility criteria are functioning as the most significant recruitment barriers across the patient population. Signal 2 tells you what happened to the patients who did qualify and were referred — whether they made it to a screening visit, and whether they enrolled or failed at the site level. Signal 3 tells you what the patients were really dealing with: the logistical, social, and clinical realities that structured data does not capture.
Together, those three signals span three categories of disqualification: eligibility-based disqualification from questionnaire data, site-reported eligibility and non-eligibility barriers, and patient-reported barriers beyond eligibility. Understanding DQA broadly as the intersection of all three is what makes it a tool for strategic action rather than a retrospective report. “Measurement without action”, as Kyle Smith, Vice President of Sales at Carebox, noted at the Measuring Patient Engagement Summit in May 2026, “is just documentation”.
The value of a complete view of disqualification is easy to understand. When a program can see exactly which eligibility criteria are driving failures, what happened to referred patients at the site level, and what barriers prevented qualified patients from completing the process, it becomes possible to act on that intelligence in specific, targeted ways.
Those action pathways — eliminating dead ends in the patient experience, building continuous engagement beyond disqualification, feeding structured intelligence back to study teams, and calibrating upstream outreach for measurable ROI — are the subject of the next post in this series. The foundation starts here: with the patients who engaged and didn't make it through. Their data touches every stage of the recruitment journey, and programs that learn to read it find that it changes what questions they ask, what they optimize for, and ultimately how many patients make it through.