How good is the test, across every study?

Pool sensitivity and specificity from every published evaluation of a test, model the trade-off between them, and say how the test would perform in your patients.

What it is

Sensitivity and specificity, pooled the right way.

Diagnostic studies report two linked numbers, sensitivity and specificity, and pooling each on its own gives the wrong answer. A diagnostic test accuracy meta-analysis fits a bivariate or HSROC model that keeps them together, draws the summary ROC curve, and turns it into post-test probabilities a clinician can use.

What we run

Bivariate and HSROC models, the Cochrane-recommended methods, fitted in R or Stata with model diagnostics.

QUADAS-2, risk of bias and applicability judged by two reviewers for every study.

Threshold and spectrum effects, meta-regression and subgroups for cut-offs, reference standards, and populations.

Clinical meaning, likelihood ratios, Fagan nomogram, Deeks' test, and PRISMA-DTA reporting.

Summary ROC plot: individual studies, the HSROC curve, and the bivariate summary point with confidence and prediction regions
The summary ROC: every study, the fitted curve, and the pooled point with its regions.
Paired forest plot of sensitivity and specificity for ten studies with the pooled bivariate estimate
The paired forest plot: sensitivity and specificity side by side, study by study.

How it works

The method, in plain terms.

01

Protocol

PROSPERO, PRISMA-DTA

02

Search

Databases, no DTA filters

03

Screen & extract

2 x 2 tables, two reviewers

04

QUADAS-2

Bias and applicability

05

Model

Bivariate, HSROC

06

Interpret

Likelihood ratios, GRADE

How you benefit

Why a test deserves its own meta-analysis.

The project, step by step

How a diagnostic accuracy meta-analysis runs with Cohira.

Start your diagnostic accuracy meta-analysis →

1

Scope the question

We define the index test, the reference standard, the target condition, and the setting, and check enough studies exist.

2

Protocol & PROSPERO

We write and register the protocol with the models and subgroups fixed in advance.

3

Search, screen, extract

We run the search, screen in duplicate, and extract the 2 x 2 tables with a full audit trail.

4

Model

We fit the bivariate and HSROC models, check thresholds and spectrum, and derive likelihood ratios.

5

Manuscript

We draft the PRISMA-DTA paper with you: summary ROC, paired forest plots, nomogram, and an honest discussion.

6

Submit & revise

We pick the target journal, submit, and answer reviewers with you through to a decision.

Published examples

Our diagnostic accuracy work, in the literature.

A peer-reviewed diagnostic test accuracy meta-analysis co-authored by our founder, in an indexed journal.

Abdom RadiolOur work

Standalone 29-MHz micro-ultrasound for clinically significant prostate cancer: a diagnostic test accuracy meta-analysis

Abdel Gawad AM, Aboelsaad AY, et al., with Elkasaby MH · 2025

Bivariate and HSROC models, Fagan nomogram, and meta-regression for threshold and spectrum effects.

View in journal →

These are peer-reviewed papers co-authored by our founder. A fuller list is available on request.

Pricing

One fee, from idea to publication.

Diagnostic test accuracy meta-analysis, idea to publication
Protocol, bivariate and HSROC analysis with reproducible code, manuscript, journal shortlist, submission, and reviewer responses.
$1,600per study

Per project, never per author. Journal fees and database access are billed at cost. Pairwise systematic review with meta-analysis: $1,500. Your week-by-week timeline is written into the free feasibility note before you pay. Booking more than one study? 10% off the second, 15% off from the third. Acceptance Guarantee & group pricing →

More study types

TriNetXNIS / real-world dataNRD readmissionsGlobal Burden of DiseaseSEER cancer registryUK BiobankAI & ML prediction modelsSystematic reviews & meta-analysesNetwork meta-analysesNarrative reviewsCross-sectional studiesYour clinical dataSingle stages

Start a conversation

Have a test worth pooling?

Tell us the test and the condition. We will tell you honestly whether the published studies can support a strong accuracy meta-analysis, and how we would build it.

Get a free feasibility note See all study types →