Why English? | Reporting PRISMA-DTA Diagnostic Accuracy Reviews Through Index Tests, Target Conditions, Thresholds and Synthesis

PRISMA-DTA diagnostic accuracy review reporting helps readers understand systematic reviews of tests instead of accepting one summary sensitivity or specificity. People searching for a PRISMA-DTA checklist, diagnostic test accuracy systematic review or index-test meta-analysis need a visible chain from participants and clinical role to target condition, reference standard, threshold, extracted data, bias assessment, synthesis and limitations.

The official PRISMA-DTA page, checked on 5 November 2026, states that the extension was published in 2018 and links the statement, explanation and elaboration, and a separate abstracts checklist. The original statement describes a 27-item checklist: eight items unchanged from the earlier PRISMA framework, 17 modified, two added and two omitted. The BMJ explanation stresses that PRISMA-DTA improves reporting and is not a comprehensive guide to conducting a diagnostic-accuracy review.

English does the connecting work. It keeps an index test attached to its version, threshold and intended clinical role; it keeps a target condition attached to the reference standard used to establish it; and it keeps a summary estimate attached to the participant spectrum, study design, unit of assessment, heterogeneity and uncertainty. Clear reporting also respects differences among health systems and languages instead of assuming one pathway transfers everywhere.

This article has a different owner from the existing STARD primary-study reporting guide, QUADAS-3 appraisal guide and QUADAS-C comparative appraisal guide. PRISMA-DTA owns transparent reporting of the systematic review itself.

---

Identify the Report as a DTA Systematic Review

Make diagnostic test accuracy and any meta-analysis visible in the title and abstract.

Diagnostic question and report trail

Start from the intended clinical role: replacement, triage, add-on, monitoring or another use, because the same accuracy may mean different things in different pathways. Under identify the report as a dta systematic review, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A test used for triage may reasonably prioritise sensitivity, while a confirmatory test may prioritise specificity; the report should not call one threshold universally best. For identify the report as a dta systematic review, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

PRISMA-DTA is a 27-item reporting guideline published in 2018; the official page also links explanation, elaboration and a separate abstracts checklist. This boundary is decisive for identify the report as a dta systematic review. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

State the Clinical Role of the Index Test

Explain whether the test is intended for triage, replacement, add-on, confirmation, monitoring or another role.

Diagnostic question and report trail

Keep the participant spectrum and setting visible; prevalence, severity, prior testing and referral patterns can change observed performance and applicability. Under state the clinical role of the index test, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

Two studies use the same test name but different software versions and positivity thresholds; pooling them without explanation can obscure clinically important differences. For state the clinical role of the index test, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

The guideline improves reporting of diagnostic-accuracy reviews and is not a manual for conducting the review or a certificate of methodological quality. This boundary is decisive for state the clinical role of the index test. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Define Participants, Index Tests and Target Conditions

Write an explicit review question with the people, tests and conditions being assessed.

Diagnostic question and report trail

Name the index-test version, threshold and interpretation rule rather than treating a product family or test label as one stable method. Under define participants, index tests and target conditions, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A reference standard incorporates the index test result, creating a path for agreement to be inflated; transparent reporting allows the bias to be appraised. For define participants, index tests and target conditions, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Accuracy does not establish clinical utility, patient benefit, cost-effectiveness or suitability for one individual. This boundary is decisive for define participants, index tests and target conditions. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Set Eligibility Criteria Around the Diagnostic Pathway

Include setting, prior testing, test role, thresholds, reference standards, design and report characteristics.

Diagnostic question and report trail

Describe the target condition and reference standard together so readers can judge how disease or status was established. Under set eligibility criteria around the diagnostic pathway, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A per-lesion analysis includes several lesions from one patient; the review must explain statistical dependence and avoid presenting lesion counts as patient counts. For set eligibility criteria around the diagnostic pathway, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Risk of bias and applicability are related but distinct: a well-conducted study may still address the wrong setting or participant group. This boundary is decisive for set eligibility criteria around the diagnostic pathway. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Report Protocol, Registration and Amendments

Name the prospective record or protocol and date substantive changes to methods.

Diagnostic question and report trail

Preserve the unit of assessment—per patient, lesion, sample, image or episode—through extraction and synthesis. Under report protocol, registration and amendments, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A review reports a summary sensitivity of 0.90, but confidence intervals and between-study variation show that performance is less stable than the headline implies. For report protocol, registration and amendments, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Threshold choice changes the balance between false positives and false negatives, so summary claims must keep the intended role visible. This boundary is decisive for report protocol, registration and amendments. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

List Information Sources and Search Dates

Report databases, registries, websites and the date each source was last searched.

Diagnostic question and report trail

Report sensitivity and specificity with uncertainty and thresholds; one number cannot describe the full trade-off between missed cases and false positives. Under list information sources and search dates, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A comparative review includes paired and unpaired designs; the synthesis keeps design type visible and avoids treating indirect contrasts as equivalent to within-study comparisons. For list information sources and search dates, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Publication-bias methods for diagnostic accuracy have limitations and should be reported without overstating what a statistical test can prove. This boundary is decisive for list information sources and search dates. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Publish a Reproducible Search Strategy

Provide complete syntax and restrictions while recognising that diagnostic-study filters may miss relevant records.

Diagnostic question and report trail

Show how study-level estimates entered the synthesis and how heterogeneity was explored rather than presenting only a summary point. Under publish a reproducible search strategy, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A flow diagram shows records and reports, while a characteristics table links multiple reports from the same diagnostic study to one study identity. For publish a reproducible search strategy, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Incomplete reporting creates uncertainty about the review; it does not automatically prove that the underlying study was conducted badly. This boundary is decisive for publish a reproducible search strategy. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Describe Study Selection and Report Linkage

Explain screening, eligibility decisions and how multiple reports from one study were connected.

Diagnostic question and report trail

Connect every accuracy conclusion to risk of bias, applicability and limitations in the evidence base. Under describe study selection and report linkage, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A school science analogy compares two thermometers against the same reference under the same conditions, then explains why medical decisions require more than this simplified model. For describe study selection and report linkage, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

A review should respect diagnostic evidence from different languages and health systems and report restrictions that may narrow transferability. This boundary is decisive for describe study selection and report linkage. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Define Data Extraction and Classifications

State how index tests, target conditions, reference standards, settings and designs were classified.

Diagnostic question and report trail

Start from the intended clinical role: replacement, triage, add-on, monitoring or another use, because the same accuracy may mean different things in different pathways. Under define data extraction and classifications, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A test used for triage may reasonably prioritise sensitivity, while a confirmatory test may prioritise specificity; the report should not call one threshold universally best. For define data extraction and classifications, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

PRISMA-DTA is a 27-item reporting guideline published in 2018; the official page also links explanation, elaboration and a separate abstracts checklist. This boundary is decisive for define data extraction and classifications. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Preserve Index-Test Version and Threshold

Record device, assay, software, reader rules and positivity thresholds used in each study.

Diagnostic question and report trail

Keep the participant spectrum and setting visible; prevalence, severity, prior testing and referral patterns can change observed performance and applicability. Under preserve index-test version and threshold, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

Two studies use the same test name but different software versions and positivity thresholds; pooling them without explanation can obscure clinically important differences. For preserve index-test version and threshold, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

The guideline improves reporting of diagnostic-accuracy reviews and is not a manual for conducting the review or a certificate of methodological quality. This boundary is decisive for preserve index-test version and threshold. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Define the Target Condition and Reference Standard

Keep the condition of interest separate from the method used to determine whether it was present.

Diagnostic question and report trail

Name the index-test version, threshold and interpretation rule rather than treating a product family or test label as one stable method. Under define the target condition and reference standard, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A reference standard incorporates the index test result, creating a path for agreement to be inflated; transparent reporting allows the bias to be appraised. For define the target condition and reference standard, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Accuracy does not establish clinical utility, patient benefit, cost-effectiveness or suitability for one individual. This boundary is decisive for define the target condition and reference standard. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

State the Unit of Assessment

Report whether estimates are per patient, lesion, image, specimen or episode and handle dependence accordingly.

Diagnostic question and report trail

Describe the target condition and reference standard together so readers can judge how disease or status was established. Under state the unit of assessment, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A per-lesion analysis includes several lesions from one patient; the review must explain statistical dependence and avoid presenting lesion counts as patient counts. For state the unit of assessment, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Risk of bias and applicability are related but distinct: a well-conducted study may still address the wrong setting or participant group. This boundary is decisive for state the unit of assessment. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Assess Risk of Bias and Applicability

Describe the appraisal approach and keep internal validity separate from match to the review question.

Diagnostic question and report trail

Preserve the unit of assessment—per patient, lesion, sample, image or episode—through extraction and synthesis. Under assess risk of bias and applicability, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A review reports a summary sensitivity of 0.90, but confidence intervals and between-study variation show that performance is less stable than the headline implies. For assess risk of bias and applicability, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Threshold choice changes the balance between false positives and false negatives, so summary claims must keep the intended role visible. This boundary is decisive for assess risk of bias and applicability. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Name the Principal Accuracy Measures

State sensitivity, specificity or other measures and explain how each connects to the clinical question.

Diagnostic question and report trail

Report sensitivity and specificity with uncertainty and thresholds; one number cannot describe the full trade-off between missed cases and false positives. Under name the principal accuracy measures, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A comparative review includes paired and unpaired designs; the synthesis keeps design type visible and avoids treating indirect contrasts as equivalent to within-study comparisons. For name the principal accuracy measures, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Publication-bias methods for diagnostic accuracy have limitations and should be reported without overstating what a statistical test can prove. This boundary is decisive for name the principal accuracy measures. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Report Two-by-Two Data When Available

Make true positives, false positives, false negatives and true negatives traceable to each study and threshold.

Diagnostic question and report trail

Show how study-level estimates entered the synthesis and how heterogeneity was explored rather than presenting only a summary point. Under report two-by-two data when available, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A flow diagram shows records and reports, while a characteristics table links multiple reports from the same diagnostic study to one study identity. For report two-by-two data when available, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Incomplete reporting creates uncertainty about the review; it does not automatically prove that the underlying study was conducted badly. This boundary is decisive for report two-by-two data when available. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Handle Thresholds as Clinical Decisions

Do not hide different positivity thresholds inside one convenient average.

Diagnostic question and report trail

Connect every accuracy conclusion to risk of bias, applicability and limitations in the evidence base. Under handle thresholds as clinical decisions, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A school science analogy compares two thermometers against the same reference under the same conditions, then explains why medical decisions require more than this simplified model. For handle thresholds as clinical decisions, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

A review should respect diagnostic evidence from different languages and health systems and report restrictions that may narrow transferability. This boundary is decisive for handle thresholds as clinical decisions. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Plan the Accuracy Synthesis

Describe statistical models, transformations, continuity corrections, random effects and software as applicable.

Diagnostic question and report trail

Start from the intended clinical role: replacement, triage, add-on, monitoring or another use, because the same accuracy may mean different things in different pathways. Under plan the accuracy synthesis, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A test used for triage may reasonably prioritise sensitivity, while a confirmatory test may prioritise specificity; the report should not call one threshold universally best. For plan the accuracy synthesis, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

PRISMA-DTA is a 27-item reporting guideline published in 2018; the official page also links explanation, elaboration and a separate abstracts checklist. This boundary is decisive for plan the accuracy synthesis. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Investigate Heterogeneity

Examine participant spectrum, setting, threshold, test execution, reference standard and design as possible sources of variation.

Diagnostic question and report trail

Keep the participant spectrum and setting visible; prevalence, severity, prior testing and referral patterns can change observed performance and applicability. Under investigate heterogeneity, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

Two studies use the same test name but different software versions and positivity thresholds; pooling them without explanation can obscure clinically important differences. For investigate heterogeneity, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

The guideline improves reporting of diagnostic-accuracy reviews and is not a manual for conducting the review or a certificate of methodological quality. This boundary is decisive for investigate heterogeneity. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Report Comparative Accuracy Transparently

Distinguish within-study comparisons from indirect contrasts and explain design-specific analysis.

Diagnostic question and report trail

Name the index-test version, threshold and interpretation rule rather than treating a product family or test label as one stable method. Under report comparative accuracy transparently, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A reference standard incorporates the index test result, creating a path for agreement to be inflated; transparent reporting allows the bias to be appraised. For report comparative accuracy transparently, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Accuracy does not establish clinical utility, patient benefit, cost-effectiveness or suitability for one individual. This boundary is decisive for report comparative accuracy transparently. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Address Publication Bias Cautiously

Report the method and its limits without treating asymmetry or a test result as proof of missing studies.

Diagnostic question and report trail

Describe the target condition and reference standard together so readers can judge how disease or status was established. Under address publication bias cautiously, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A per-lesion analysis includes several lesions from one patient; the review must explain statistical dependence and avoid presenting lesion counts as patient counts. For address publication bias cautiously, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Risk of bias and applicability are related but distinct: a well-conducted study may still address the wrong setting or participant group. This boundary is decisive for address publication bias cautiously. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Present Study Flow and Characteristics

Reconcile records, reports and studies and show the diagnostic pathway features of included evidence.

Diagnostic question and report trail

Preserve the unit of assessment—per patient, lesion, sample, image or episode—through extraction and synthesis. Under present study flow and characteristics, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A review reports a summary sensitivity of 0.90, but confidence intervals and between-study variation show that performance is less stable than the headline implies. For present study flow and characteristics, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Threshold choice changes the balance between false positives and false negatives, so summary claims must keep the intended role visible. This boundary is decisive for present study flow and characteristics. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Present Individual Study Accuracy

Display study estimates, thresholds and uncertainty before the summary result.

Diagnostic question and report trail

Report sensitivity and specificity with uncertainty and thresholds; one number cannot describe the full trade-off between missed cases and false positives. Under present individual study accuracy, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A comparative review includes paired and unpaired designs; the synthesis keeps design type visible and avoids treating indirect contrasts as equivalent to within-study comparisons. For present individual study accuracy, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Publication-bias methods for diagnostic accuracy have limitations and should be reported without overstating what a statistical test can prove. This boundary is decisive for present individual study accuracy. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Present Synthesis Results With Uncertainty

Report study and participant counts, summary estimates, confidence regions and heterogeneity.

Diagnostic question and report trail

Show how study-level estimates entered the synthesis and how heterogeneity was explored rather than presenting only a summary point. Under present synthesis results with uncertainty, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A flow diagram shows records and reports, while a characteristics table links multiple reports from the same diagnostic study to one study identity. For present synthesis results with uncertainty, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

Incomplete reporting creates uncertainty about the review; it does not automatically prove that the underlying study was conducted badly. This boundary is decisive for present synthesis results with uncertainty. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Interpret Accuracy Without Promising Patient Benefit

Connect results to intended use, bias, applicability and evidence limits while keeping clinical utility separate.

Diagnostic question and report trail

Connect every accuracy conclusion to risk of bias, applicability and limitations in the evidence base. Under interpret accuracy without promising patient benefit, preserve participants, index test, target condition, reference standard, clinical role and unit of analysis wherever they matter. A diagnostic number becomes interpretable only when readers know what counted as a positive index result and how target status was established.

Accuracy-review example

A school science analogy compares two thermometers against the same reference under the same conditions, then explains why medical decisions require more than this simplified model. For interpret accuracy without promising patient benefit, read the example in two layers: first, what the PRISMA-DTA report discloses; second, what a reader may reasonably infer about the review. Reporting completeness supports appraisal and replication, but it does not certify the accuracy of the index test or the absence of bias in included studies.

Interpretation boundary

A review should respect diagnostic evidence from different languages and health systems and report restrictions that may narrow transferability. This boundary is decisive for interpret accuracy without promising patient benefit. Finish with a bounded sentence naming the supported accuracy claim, the main source of uncertainty and the clinical or methodological question that remains. Do not turn sensitivity, specificity or a summary curve into a promise of patient benefit, admission, employment or personal outcomes.

---

Frequently Asked Questions

What is PRISMA-DTA?

It is the PRISMA reporting extension for systematic reviews and meta-analyses of diagnostic test accuracy studies.

How many checklist items are there?

The PRISMA-DTA statement contains 27 items.

When was PRISMA-DTA published?

The official PRISMA page identifies the DTA extension as published in 2018.

Does PRISMA-DTA tell me how to conduct the review?

No. It guides reporting and should be used with appropriate review-methods guidance.

What is an index test?

It is the test whose accuracy is being evaluated.

What is a target condition?

It is the disease, condition or state that the index test is intended to detect or classify.

Why do thresholds matter?

Changing the positivity threshold changes the balance between sensitivity and specificity.

How is PRISMA-DTA different from STARD?

PRISMA-DTA reports systematic reviews; STARD reports individual diagnostic-accuracy studies.

How is PRISMA-DTA different from QUADAS-3?

PRISMA-DTA is reporting guidance for the review; QUADAS-3 supports appraisal of diagnostic-accuracy results included in a review.

Does high accuracy prove patient benefit?

No. Clinical utility, downstream management, harms, costs and patient outcomes require additional evidence.

---

Useful Next Reading

• Read the official PRISMA-DTA page and checklist.
• Read the PRISMA-DTA explanation and elaboration.
• Connect review reporting to QUADAS-3 appraisal.
• Compare systematic-review reporting with STARD primary-study reporting.

Closing thought: PRISMA-DTA turns a headline accuracy number into a transparent review story in which patients, tests, thresholds, target conditions, studies and uncertainty remain connected.