How to Use Screening Statistics to Better Understand Your Study

When you launch a study on Alethios, your focus naturally shifts from study planning to enrollment: how many participants are joining, how quickly they're moving through onboarding, and whether your cohort is filling as expected. But there's another signal that's available from the moment your first participant applies — and it's one that many researchers underuse.

Screening statistics, now available in the Screen tab within Live Mode on your study, give you a real-time view of how participants are moving through your eligibility criteria. You can see how many applicants were disqualified, at which questions they screened out, and the full distribution of answers across your screening survey.

It sounds simple. The implications are not.

The First 48 Hours: Is Your Criteria Working?

The first two days after a study goes live are among the most informative — not just for recruitment velocity, but for protocol design. Screening statistics give you an early diagnostic on whether your eligibility criteria are calibrated appropriately.

If a large proportion of applicants are being excluded, and the data shows they're clustering around one or two specific questions, that's a signal worth examining before you're deep into enrollment. Some teams have discovered, after reviewing their screening distributions, that a criterion they'd treated as a hard exclusion was either too broadly defined, or that the underlying answers warranted a closer look before automatic disqualification.

This is the moment to catch those issues. Adjusting a screening criterion after you've enrolled half your cohort is painful and may compromise your study's integrity. Catching it in the first 48 hours is just good design.

Over Time: Understanding Who You've Actually Enrolled

As your study progresses, your screening data becomes a longitudinal record of your participant population — not just who made it in, but who tried to and why they didn't.

This is the first real window into how representative your trial population is. The distribution of answers to your screening questions reveals the characteristics of the people who raised their hands to participate: their age ranges, the conditions they report, the criteria they didn't meet. That profile tells you something important about the generalizability of your findings — and about gaps between your intended population and the one you actually reached.

In disease-state research, this lens can surface something more uncomfortable: structural problems in how those conditions get diagnosed and reported in the real world.

Consider endometriosis as a current example. If you're running a study that requires a confirmed endometriosis diagnosis as an inclusion criterion, your screening data may show a pattern that reflects a well-documented reality: it takes seven to nine years after first symptoms for women to receive a diagnosis, in part because of limited diagnostic tools, but also because of a lack of awareness, dismissal by doctors, and symptoms that often resemble other illnesses. (Advisory). As many as six out of every ten cases of endometriosis remain undiagnosed (American Medical Association).

A high drop-off rate at a diagnosis screening question isn't just a recruitment problem. It's a reflection of a diagnostic gap in the healthcare system. Average delays range from 6 to 11 years, often despite disabling and ongoing symptoms, with symptoms being treated for years without a definitive diagnosis ever being made. (PSNet) Your screening data will show you exactly how that gap manifests in your applicant pool.

That kind of visibility matters. It can inform how you design future studies, how you communicate about your population's characteristics in publications, and how you think about the representativeness of the evidence you're generating.

How to Access Screening Statistics

Screening statistics are available in the Screen tab when you're in Live Mode on your study. From there, you can view:

• The number of participants disqualified at the first screening question that disqualified them
• The full distribution of answers across all screening questions
• How disqualification rates change over time as your study matures

There's no additional setup required. The data is collected automatically as applicants move through your screening survey, and updates in real time.

Using This Data Well

A few principles worth keeping in mind as you work with screening statistics:

• Don't over-interpret early snapshots. The first applicants to find your study are often not representative of the broader pool you'll ultimately reach. Patterns stabilize as volume increases.
• Review distributions, not just counts. Knowing that 40% of applicants were disqualified tells you less than knowing where they screened out and what they reported.
• Use it as a protocol signal, not just a reporting tool. If your screening data is telling you something unexpected, that's often useful information — about your study design, your recruitment channels, or the population you're trying to reach.

Screening data is the first lens you have into who your study is actually reaching. Used well, it's one of the most underrated tools in decentralized research design.

‍