Get a Quote

Free Tools › QUADAS-3 Tool

QUADAS-3 Tool for Diagnostic Test Accuracy Reviews

Record QUADAS-3 risk of bias and applicability judgments for each accuracy estimate, get the overall judgment from the QUADAS-3 rules, and download traffic light and summary plots.

FreeNo account neededYour data stays in your browser

Domains and how to use them
  1. Risk of bias: participantsSingle-gate rather than case-control design, prospective enrollment, consecutive or random sampling, and a sample representative of the intended-use population.
  2. Risk of bias: index testConducted and interpreted as recommended, without knowledge of the reference standard, with the information available in practice, and with a standard or pre-specified threshold.
  3. Risk of bias: target conditionA reference standard that identifies the condition correctly, applied to all participants in the same way, not incorporating the index test, interpreted blind, and at an appropriate interval.
  4. Risk of bias: analysisAll participants analyzed, missing data handled appropriately, a unit of analysis that matches the ideal trial, and accuracy estimates calculated appropriately.
  5. Applicability: participantsWhether the included participants match those in the ideal test accuracy trial for the synthesis question.
  6. Applicability: index testWhether the index test, its conduct, setting or interpretation match the ideal test accuracy trial.
  7. Applicability: target conditionWhether the target condition, as defined by the reference standard, matches the ideal test accuracy trial.

Answer the signaling questions in the official tool, then record your judgments here: QUADAS-3 tool and guidance (University of Bristol).

Risk of biasApplicability concerns
Study, estimatePAITTCANPAITTCRisk of biasApplicability
LowLow
HighLow
HighHigh
Insufficient informationInsufficient information

Your entries are saved in this browser only. Nothing is uploaded.

QUADAS-3 overall rule (phase 6), applied separately to risk of bias and applicability: high if any domain is high; low if all domains are low; otherwise insufficient information. Record your rationale alongside.

Summary plot

Traffic light plot

Risk of biasApplicability concernsStudy, estimatePAITTCANPAITTCRisk of biasApplicabilityExample study 1, threshold A+++++++++Example study 1, threshold B+×+++++×+Example study 2×+?+×++××Example study 3+?+?+?+??PA: participants · IT: index test · TC: target condition · AN: analysis+Low×High?Insufficient information

Summary plot

Risk of bias: participants75%25%Risk of bias: index test50%25%25%Risk of bias: target condition75%25%Risk of bias: analysis75%25%Applicability: participants75%25%Applicability: index test75%25%Applicability: target condition100%Overall risk of bias25%50%25%Overall applicability50%25%25%0%25%50%75%100%LowHighInsufficient information
Get new free research tools firstOccasional emails when we add tools and guides. No spam.

We only use your email to send these updates, never share it, and you can unsubscribe by replying to any email or writing to info@timelyscholar.com. Privacy Policy

How to Use the QUADAS-3 Tool

  1. 1

    Define questions and the ideal trial

    Phases 1 and 2, once per review, ideally in your protocol.

  2. 2

    Add one row per estimate

    Phase 4: name each row by study and estimate, such as a threshold or subgroup.

  3. 3

    Record judgments and download

    Phases 5 and 6: domain and overall judgments, plots as PNG or SVG, and a CSV.

What is QUADAS-3?

QUADAS-3 is the 2026 update of the QUADAS tool for assessing the risk of bias and applicability of diagnostic test accuracy studies in systematic reviews. Its developers recommend it as the current version.

It has six phases: state the synthesis question, define the ideal test accuracy trial, draw a flow diagram, identify the accuracy estimates to assess, assess risk of bias and applicability, and make an overall judgment. Risk of bias is assessed in four domains (participants, index test, target condition and analysis), and the first three are also assessed for applicability.

Each domain is judged low, high or insufficient information, guided by signaling questions answered yes, probably yes, probably no, no or no information. Unlike QUADAS-2, judgments are made for each accuracy estimate rather than for the whole study, so one study can contribute several rows, for example one per threshold.

Full guide: QUADAS-2 and QUADAS-3 explained: domains, signaling questions and examples

How it is calculated

Sources

QUADAS-3 Tool: FAQ

What is the difference between QUADAS-2 and QUADAS-3?

QUADAS-3 assesses each accuracy estimate rather than each study, starts by defining the ideal test accuracy trial, has six phases, replaces flow and timing with an analysis domain, uses 'insufficient information' where QUADAS-2 used 'unclear', and adds a formal overall judgment.

Why does QUADAS-3 assess estimates instead of studies?

A study can report several sensitivity and specificity pairs, for different thresholds, subgroups, reference standards or units of analysis, and their risk of bias can differ. After the first estimate, only domains that differ between estimates need to be reassessed.

When should I use 'insufficient information'?

Only when the study does not report enough to make a judgment. If a signaling question is answered no or probably no, decide whether the problem is likely to have affected the estimate and rate the domain low or high.

Can I use QUADAS-C with QUADAS-3?

Yes. For comparative accuracy studies the developers recommend using QUADAS-C alongside QUADAS-3, with the adaptations described in the QUADAS-3 explanation and elaboration paper.

Can I import my QUADAS-3 judgments from Excel?

Yes. Save the sheet as CSV with columns in this order: study and estimate, the four risk-of-bias domains, the three applicability domains, then optionally the two overall judgments and a weight. Low, High, II or Insufficient information are recognized.

Related Free Tools

QUADAS-2 ToolSensitivity & Specificity CalculatorSROC Curve GeneratorAll free tools →
Get a Free Quote