Mira Analytics
Site Review

One site, in study context.

Site Review turns every site into a review-ready dossier: whether its scales agree with what the primary endpoint implies, whether the questionnaire battery is complete, whether visits happen in window, and how enrolment is tracking — each of them ranked against every other site in the study. Ready before the monitoring visit, not after it.

Request a demo

The dossier

Pick a site, get the scorecard.

One view answers the questions a monitoring visit is built around: is this site scoring like the others, is the assessment battery complete, are visits happening on schedule, and is enrolment healthy. Every number carries its study-wide comparison.

Site
Site 07

Enrolling since Mar 2025
19 participants · 112 visits

Scoring agreement
3.4
Critical

Study avg 1.6 pts

Cross-scale
18%
Watch

Study avg 6% discordant

Adherence
91%
Watch

Study avg 96% performed

Visit timing
6
Watch

Visits outside protocol window

Ranked 1st of 24 sites for scoring disagreement · 4th for cross-scale discordance

Concordance

Do the scales agree with each other?

The scales in a CNS battery overlap by design, so they should move together. Every secondary scale is checked against what the primary endpoint implies, visit by visit — and the primary is checked against Mira’s own model scores. Where they diverge, the visit is marked and the scale responsible is named.

The same stored signal drives every view in the platform, so the site dossier, the study-wide module and the participant record can never disagree with each other.

Scatter of two scales with discordant visits highlighted, beside a bar chart ranking sites by percentage of discordant visits.

Adherence

Is the battery actually being completed?

Every site against every protocol visit, as a share of the assessments that should have been performed. Missing questionnaires stop being a line in a listing and become a pattern you can see — a site that drifts, or a visit the whole study struggles with.

V1
V2
V3
V4
V5
V6
V7
V8
Overall
Site 03
100
100
97
89
86
96
98
100
96
Site 07
100
96
88
79
74
85
91
95
91
Site 11
100
100
100
98
97
99
100
100
99
Site 12
98
90
78
76
88
95
96
99
90
Site 19
100
99
96
92
87
89
95
—
94
Site 21
100
100
98
97
100
96
—
—
98
≥ 95% performed 85–95% below 85% Click any cell to drop to the participants behind it

Protocol & enrolment

Is the site on track and in window?

Participant disposition and enrolment over time, so recruitment and retention outliers, withdrawals and screen failures stand out immediately. Visit-interval deviation surfaces the assessments that happened outside their protocol window — per site, in one view.

Cumulative enrolment over time beside a per-site stacked bar of screen-failed, withdrawn and completed participants.

In context

A number only means something next to the others.

Nothing in Site Review is judged on its own. Every site is weighed against every other site and against the study as a whole, then ranked — so a site that looks acceptable in isolation still surfaces as the weakest in the study.

That is what makes it early enough to act on: feedback while the site is still enrolling, not a finding at database lock.

Sites ranked by discordant visits

Share of visits where the scales disagree

Site 0722%
Site 0319%
Site 1918%
Site 1117%
Site 058%
Site 126%
Site 215%
Site 144%

Study average 6% · 24 sites

See your sites ranked.

We run Mira against your blinded data and hand you the dossier for the site you are most worried about.

Request a demo