Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods
Introduction: Meta-analyses of diagnostic/prognostic studies for calculating the pooled sensitivity and specificity require true positive (TP), true negative (TN), false positive (FP), and false negative (FN) counts. However, few studies report these values directly. This study aimed to consolidate...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Shahid Beheshti University of Medical Sciences
2025-06-01
|
| Series: | Archives of Academic Emergency Medicine |
| Subjects: | |
| Online Access: | https://journals.sbmu.ac.ir/aaem/index.php/AAEM/article/view/2678 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839642944828604416 |
|---|---|
| author | Reihanesadat Khatami Mohammadsadegh Faghihi Hannanesadat Khatami Mahmoud Yousefifard Seyedhesamoddin Khatami |
| author_facet | Reihanesadat Khatami Mohammadsadegh Faghihi Hannanesadat Khatami Mahmoud Yousefifard Seyedhesamoddin Khatami |
| author_sort | Reihanesadat Khatami |
| collection | DOAJ |
| description |
Introduction: Meta-analyses of diagnostic/prognostic studies for calculating the pooled sensitivity and specificity require true positive (TP), true negative (TN), false positive (FP), and false negative (FN) counts. However, few studies report these values directly. This study aimed to consolidate practical methods to reconstruct sensitivity and specificity from minimal data.
Methods: Our framework addresses three main situations: (1) algebraic rearrangements to compute specificity given partial metrics; (2) digitization of receiver operating characteristic (ROC) curves to obtain threshold-specific sensitivity and specificity; and (3) application of the binormal model when only AUC and prevalence are available. We tested these methods on a dataset related to mortality prediction in myocardial infarction (MI) using machine learning models, assessing how well they reconstructed sensitivity and specificity.
Results: Algebraic formulas and ROC digitization yielded reliable estimates when partial metrics or graphical curves were sufficiently detailed. However, the binormal model, which assumes equal variances, showed noticeable inaccuracies, especially for sensitivity. Linear regression analyses indicated that higher prevalence and higher AUC reduced estimation errors.
Conclusion: These methods offer practical alternatives for reconstructing diagnostic accuracy measures when data are incomplete. Relying solely on AUC-based estimations may introduce substantial bias, particularly in low-prevalence contexts. We recommend that primary studies report threshold-specific sensitivity and specificity to support more accurate meta-analytic estimations.
|
| format | Article |
| id | doaj-art-aa5b904e44e2496ca8d00aedc7fa6ce3 |
| institution | Matheson Library |
| issn | 2645-4904 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Shahid Beheshti University of Medical Sciences |
| record_format | Article |
| series | Archives of Academic Emergency Medicine |
| spelling | doaj-art-aa5b904e44e2496ca8d00aedc7fa6ce32025-07-02T05:19:39ZengShahid Beheshti University of Medical SciencesArchives of Academic Emergency Medicine2645-49042025-06-0113110.22037/aaemj.v13i1.2678Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical MethodsReihanesadat Khatami0Mohammadsadegh Faghihi1Hannanesadat Khatami2Mahmoud Yousefifard3Seyedhesamoddin Khatami4Technische Universität Berlin, Faculty of Electrical Engineering and Computer Science, Berlin, GermanyPhysiology Research Center, Iran University of Medical Sciences, Tehran, IranPhysiology Research Center, Iran University of Medical Sciences, Tehran, IranPhysiology Research Center, Iran University of Medical Sciences, Tehran, IranPhysiology Research Center, Iran University of Medical Sciences (IUMS), Tehran, Iran Introduction: Meta-analyses of diagnostic/prognostic studies for calculating the pooled sensitivity and specificity require true positive (TP), true negative (TN), false positive (FP), and false negative (FN) counts. However, few studies report these values directly. This study aimed to consolidate practical methods to reconstruct sensitivity and specificity from minimal data. Methods: Our framework addresses three main situations: (1) algebraic rearrangements to compute specificity given partial metrics; (2) digitization of receiver operating characteristic (ROC) curves to obtain threshold-specific sensitivity and specificity; and (3) application of the binormal model when only AUC and prevalence are available. We tested these methods on a dataset related to mortality prediction in myocardial infarction (MI) using machine learning models, assessing how well they reconstructed sensitivity and specificity. Results: Algebraic formulas and ROC digitization yielded reliable estimates when partial metrics or graphical curves were sufficiently detailed. However, the binormal model, which assumes equal variances, showed noticeable inaccuracies, especially for sensitivity. Linear regression analyses indicated that higher prevalence and higher AUC reduced estimation errors. Conclusion: These methods offer practical alternatives for reconstructing diagnostic accuracy measures when data are incomplete. Relying solely on AUC-based estimations may introduce substantial bias, particularly in low-prevalence contexts. We recommend that primary studies report threshold-specific sensitivity and specificity to support more accurate meta-analytic estimations. https://journals.sbmu.ac.ir/aaem/index.php/AAEM/article/view/2678Diagnostic AccuracyPrognostic studySensitivity and SpecificityMeta-AnalysisPredictive Value of Testsspecificity |
| spellingShingle | Reihanesadat Khatami Mohammadsadegh Faghihi Hannanesadat Khatami Mahmoud Yousefifard Seyedhesamoddin Khatami Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods Archives of Academic Emergency Medicine Diagnostic Accuracy Prognostic study Sensitivity and Specificity Meta-Analysis Predictive Value of Tests specificity |
| title | Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods |
| title_full | Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods |
| title_fullStr | Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods |
| title_full_unstemmed | Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods |
| title_short | Calculation of Sensitivity and Specificity from Partial Data for Meta-Analyses: Introducing Some Practical Methods |
| title_sort | calculation of sensitivity and specificity from partial data for meta analyses introducing some practical methods |
| topic | Diagnostic Accuracy Prognostic study Sensitivity and Specificity Meta-Analysis Predictive Value of Tests specificity |
| url | https://journals.sbmu.ac.ir/aaem/index.php/AAEM/article/view/2678 |
| work_keys_str_mv | AT reihanesadatkhatami calculationofsensitivityandspecificityfrompartialdataformetaanalysesintroducingsomepracticalmethods AT mohammadsadeghfaghihi calculationofsensitivityandspecificityfrompartialdataformetaanalysesintroducingsomepracticalmethods AT hannanesadatkhatami calculationofsensitivityandspecificityfrompartialdataformetaanalysesintroducingsomepracticalmethods AT mahmoudyousefifard calculationofsensitivityandspecificityfrompartialdataformetaanalysesintroducingsomepracticalmethods AT seyedhesamoddinkhatami calculationofsensitivityandspecificityfrompartialdataformetaanalysesintroducingsomepracticalmethods |