Diagnostic accuracy study
How accurately does a test identify a target condition in its intended setting?
When to use this project
Evaluate an index test against a defined reference standard in people representing the intended clinical pathway. Specify the target condition, test role and decision threshold.
How it differs
Accuracy concerns classification against a reference standard. A prediction model combines inputs to estimate an individual probability; an accuracy study alone does not establish that using a test improves patient outcomes.
Decisions before you start
- Define index test, reference standard and target condition, including the consequences of an imperfect reference.
- Prespecify thresholds and reader blinding. Record test order and the time between tests.
- Plan sensitivity and specificity precision using expected numbers with and without the condition. Address missing, indeterminate and partially verified results.
How the flows work together
Each flow below links to its diagram, steps and tables. Flow order organizes the project; it does not imply that all work is strictly sequential. Data assessment, analysis planning and protocol refinement may inform one another.
1. Diagnostic study overview
Overview · Clinical question through accuracy interpretation
Use this overview to state the purpose and link to detailed decisions. Keep a versioned protocol and decision register, and add evidence references for the rationale.
2. Participant and specimen selection
Selection · Track eligibility sampling and specimens
Sampling should represent the intended clinical pathway. Keep participant and specimen identifiers separate, and show the reasons that paired results are unavailable. Excluding hard-to-test patients can change the apparent performance.
3. Index and reference testing
Conduct · Specify thresholds blinding and timing
The index test is the test under evaluation; the reference standard determines target-condition status. Neither is automatically error-free. Specify thresholds before examining accuracy and distinguish masking test readers from concealing randomized allocation.
4. Accuracy and precision plan
Statistics · Specify paired measures and uncertainty
Sensitivity uses reference-positive participants; specificity uses reference-negative participants. Positive and negative predictive values also depend on the clinical population. Preserve indeterminate and incompletely verified results rather than silently deleting them.
5. Bias and applicability assessment
Statistics · Assess verification and spectrum effects
Review selection, test interpretation, reference classification and participant timing together. An apparently precise estimate can remain biased or inapplicable to the intended setting. Record the version and scope of any appraisal tool used.
Equations and statistical interpretation
These explanations describe notation and assumptions. Flow does not calculate these values.
Sensitivity and specificity
TP and FN are index-test positive and negative results among reference-positive participants; TN and FP are negative and positive results among reference-negative participants.
These ratios assume a specified threshold and observed reference classification. Report intervals and missing or indeterminate results separately; selective verification can bias both estimates.
Positive predictive value
π is the target-condition prevalence in the population where the test is used.
This expression assumes sensitivity and specificity apply to that population. A predictive value from an enriched accuracy sample may not transport to routine care.
Published example
Trace the index tests, template mapping biopsy reference and paired observations. Distinguish diagnostic performance from the effect of a subsequent care pathway.
This publication illustrates a study design, not an endorsement of every method or a reproduction of its results. Follow its protocol, supplement and update history when studying the example.
Hypothetical scenario
A hospital wants to evaluate a rapid infection test before introducing it in its emergency department.
- Diagnostic study overview: State whether the test will rule out infection, rule it in or replace an existing procedure.
- Participant and specimen selection: Recruit the intended patient spectrum and track paired specimens without silently excluding difficult cases.
- Index and reference testing: Specify the rapid test, reference procedure, blinded interpretation and handling of indeterminate results.
- Accuracy and precision plan: Build the paired result table and estimate accuracy with intervals at the prespecified threshold.
- Bias and applicability assessment: Examine incomplete verification, spectrum differences and whether findings apply to the proposed pathway.
Deliver a populated protocol, decision log, data specification, and planned outputs. Enter numerical findings only after analysis and verification. Keep decisions made before seeing results distinguishable from later amendments.
Sources
Source check: 2026-09-17. Record the version of each reporting or appraisal tool used in your actual protocol.
Compare all Diagnostic and prediction studies project types · Open editor in a new tab