oksister.co Reading the research record behind a treatment

What a clinical trial's own registry entry tells you

Behind most treatments sits a registry record: a structured public document stating who was studied, what was measured, and whether results were ever posted. Reading one takes about ten minutes and answers a question no summary article can — whether people like you were in the study at all. This page walks the record field by field.

The record, field by field

These fields appear on registry entries in a standard structure. Read them in this order; the later ones only make sense once the earlier ones are settled.

Registration IDThe unique identifier for this study.The one value that lets you find the same study cited under different titles in different articles.
StatusRecruiting, active, completed, terminated or withdrawn.Terminated and withdrawn are informative: a study can stop for slow recruitment, funding, or a safety finding — and the record often says which.
PhaseThe stage of investigation.Sets the outer limit on what the study can show at all. See the next section.
EnrolmentHow many participants, and whether that figure is anticipated or actual.An actual figure far below the anticipated one is a question worth carrying into the results.
Eligibility criteriaWho could join, and who was excluded, including age range and sex.The single most useful field for deciding whether a finding applies to you.
Primary outcomeThe one measure the study was designed to answer, and when it was measured.Stated before the study runs, which is what makes it different from any finding reported afterwards.
SponsorWho funds and runs the study.Not a verdict on quality — but part of the record, and stated plainly.
Results postedWhether summary results were submitted to the registry.Some studies are required to post results within a defined period after completion. An old completed study with none is a gap.

What each phase can answer

Phase is not a measure of quality. It states which question the study was built to answer, and therefore which questions it cannot.

Phase 1smallest

Primarily safety and dosing, usually in a small group. Not designed to show whether a treatment works.

Phase 2larger

Early efficacy signals and further safety, in people with the condition. Often too small to detect uncommon effects.

Phase 3largest

Comparison against existing treatment or placebo in a larger group. This is where effectiveness claims are usually established.

Phase 4after approval

Study after a treatment is in general use, where rarer effects and long-term outcomes become visible.

The practical consequence: a headline describing a promising result from an early-phase study is describing a signal, not a conclusion. That is not a criticism of the study — it is what the phase was designed to produce.

Five things in a record worth pausing on

None of these makes a study wrong. Each changes how far its finding travels.

The eligibility criteria exclude people like the reader

Age caps, sex restrictions and excluded conditions all narrow who the result describes. A finding from a trial that excluded people over sixty-five says little about people over sixty-five.

The primary outcome is a surrogate

Some studies measure a marker that stands in for the outcome people care about. A change in a marker is a real finding and is not the same as a change in how long or how well someone lives.

Actual enrolment came in far below plan

A study designed for one number of participants and completed with far fewer may no longer be able to detect the difference it was built to detect.

Completed years ago with no results posted

Unposted results are a known problem in the research record, and studies with disappointing findings have historically been less likely to be published. The absence is itself information.

The record was updated after the study began

Registries keep a history of changes. A primary outcome altered after enrolment started is something a reader is entitled to notice.

What reading a record does and does not equip you to do

The point of this skill is better questions in a consultation, not a substitute for one.

Reasonable to take from it

  • Whether people of your age and sex were included at all.
  • What the study set out to measure, stated before it ran.
  • How large it was, and whether it finished as planned.
  • Whether results exist in the public record.
  • A specific question to raise: "was this studied in women my age?"

Not reasonable to take from it

  • A decision to start, stop or change any treatment.
  • A conclusion that a treatment does or does not work for you personally.
  • A judgement about a clinician's recommendation.
  • Anything about a treatment's suitability during pregnancy or with other medicines.

Sources behind the dates above

Every date below names the policy that produced it, which is what lets a reader verify it.

  • US Food and Drug Administration guidance, 1977, and its subsequent revision. Source for the exclusion of women of childbearing potential from early-phase studies and for the later reversal of that position.
  • NIH Revitalization Act of 1993. Source for the requirement to include women and minority participants in NIH-funded clinical research.
  • NIH policy on sex as a biological variable, effective 2016. Source for the requirement that sex be factored into the design and analysis of funded research.
  • Public clinical trial registries. Source for the record structure, the phase definitions and the results-posting requirement described above.

Policies and registry requirements are revised. Where this page gives a year, it names the instrument that established it precisely so the current position can be checked rather than assumed.