Understanding test accuracy research: a test consequence graphic
Abstract Background Presenting results of diagnostic test accuracy research so that it is accessible to users is challenging. Commonly used accuracy measures (e.g. sensitivity and specificity) are poorly understood by health professionals and the public. Evidence suggests that presenting probabiliti...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-02-01
|
| Series: | Diagnostic and Prognostic Research |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s41512-017-0023-0 |
| _version_ | 1828842712671977472 |
|---|---|
| author | Penny Whiting Clare Davenport |
| author_facet | Penny Whiting Clare Davenport |
| author_sort | Penny Whiting |
| collection | DOAJ |
| description | Abstract Background Presenting results of diagnostic test accuracy research so that it is accessible to users is challenging. Commonly used accuracy measures (e.g. sensitivity and specificity) are poorly understood by health professionals and the public. Evidence suggests that presenting probabilities as natural frequencies rather than percentages facilitates understanding. We present a test consequence graphic to display results based on natural frequencies and test consequences. Methods The graphic was developed as part of a project to develop guidance for writing plain language summaries for Cochrane diagnostic test accuracy (DTA) reviews. Using a mixed methods approach (focus groups, user testing, web-based surveys, public engagement and piloting), the graphic emerged as a clear preference out of a range of methods for presenting probabilistic information (text only, numbers embedded in text, annotated graphic) across patient representatives, media representatives and health professionals. The structure of the graphic was refined during the research process. Results The test consequence graphic displays the results of diagnostic test accuracy study or review as natural frequencies based on a hypothetical cohort of 1000 patients receiving the test. Conclusions The test consequence graphic provides a tool to help researchers communicate the results of diagnostic research in a simple, easy to access format and encourage meaningful application of research findings to practice. Key to this is linking estimates of test accuracy to potential downstream consequences of testing. |
| first_indexed | 2024-12-12T20:30:10Z |
| format | Article |
| id | doaj.art-84c96a34bbba4b9fbc84ec6567523ee6 |
| institution | Directory Open Access Journal |
| issn | 2397-7523 |
| language | English |
| last_indexed | 2024-12-12T20:30:10Z |
| publishDate | 2018-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Diagnostic and Prognostic Research |
| spelling | doaj.art-84c96a34bbba4b9fbc84ec6567523ee62022-12-22T00:13:02ZengBMCDiagnostic and Prognostic Research2397-75232018-02-01211510.1186/s41512-017-0023-0Understanding test accuracy research: a test consequence graphicPenny Whiting0Clare Davenport1NIHR CHLARC West, University Hospitals Bristol NHS Foundation TrustInstitute of Applied Health Research, University of BirminghamAbstract Background Presenting results of diagnostic test accuracy research so that it is accessible to users is challenging. Commonly used accuracy measures (e.g. sensitivity and specificity) are poorly understood by health professionals and the public. Evidence suggests that presenting probabilities as natural frequencies rather than percentages facilitates understanding. We present a test consequence graphic to display results based on natural frequencies and test consequences. Methods The graphic was developed as part of a project to develop guidance for writing plain language summaries for Cochrane diagnostic test accuracy (DTA) reviews. Using a mixed methods approach (focus groups, user testing, web-based surveys, public engagement and piloting), the graphic emerged as a clear preference out of a range of methods for presenting probabilistic information (text only, numbers embedded in text, annotated graphic) across patient representatives, media representatives and health professionals. The structure of the graphic was refined during the research process. Results The test consequence graphic displays the results of diagnostic test accuracy study or review as natural frequencies based on a hypothetical cohort of 1000 patients receiving the test. Conclusions The test consequence graphic provides a tool to help researchers communicate the results of diagnostic research in a simple, easy to access format and encourage meaningful application of research findings to practice. Key to this is linking estimates of test accuracy to potential downstream consequences of testing.http://link.springer.com/article/10.1186/s41512-017-0023-0DiagnosisNatural frequenciesGraphical display |
| spellingShingle | Penny Whiting Clare Davenport Understanding test accuracy research: a test consequence graphic Diagnostic and Prognostic Research Diagnosis Natural frequencies Graphical display |
| title | Understanding test accuracy research: a test consequence graphic |
| title_full | Understanding test accuracy research: a test consequence graphic |
| title_fullStr | Understanding test accuracy research: a test consequence graphic |
| title_full_unstemmed | Understanding test accuracy research: a test consequence graphic |
| title_short | Understanding test accuracy research: a test consequence graphic |
| title_sort | understanding test accuracy research a test consequence graphic |
| topic | Diagnosis Natural frequencies Graphical display |
| url | http://link.springer.com/article/10.1186/s41512-017-0023-0 |
| work_keys_str_mv | AT pennywhiting understandingtestaccuracyresearchatestconsequencegraphic AT claredavenport understandingtestaccuracyresearchatestconsequencegraphic |