Assessing the validity of a self-reported clinical diagnosis of schizophrenia

Woolway, Grace E., Legge, Sophie E., Lynham, Amy J., Smart, Sophie E. ORCID: https://orcid.org/0000-0002-6709-5425, Hubbard, Leon, Daniel, Ellie R., Pardiñas, Antonio F. ORCID: https://orcid.org/0000-0001-6845-7590, Escott-Price, Valentina ORCID: https://orcid.org/0000-0003-1784-5483, O’Donovan, Michael C., Owen, Michael J. ORCID: https://orcid.org/0000-0003-4798-0862, Jones, Ian R. and Walters, James T. R. ORCID: https://orcid.org/0000-0002-6980-4053 2024. Assessing the validity of a self-reported clinical diagnosis of schizophrenia. Schizophrenia 10 (1) , 99. 10.1038/s41537-024-00526-5

	PDF - Published Version Download (3MB)
	PDF - Supplemental Material Download (702kB)

Abstract

The increasing availability of biobanks is changing the way individuals are identified for genomic research. This study assesses the validity of a self-reported clinical diagnosis of schizophrenia. The study included 1744 clinically-ascertained participants with schizophrenia or schizoaffective disorder depressed-type (SA-D) diagnosed by self-report and/or research interview and 1453 UK Biobank participants with self-reported and/or medical record diagnosis of schizophrenia or SA-D. Unaffected controls included a total of 501,837 participants. We assessed the positive predictive values (PPV) of self-reported clinical diagnoses against research interview and medical record diagnoses. Polygenic risk scores (PRS) and phenotypes relating to demographics, education and employment were compared across diagnostic groups. The variance explained (r2) in schizophrenia PRS for each diagnostic group was compared to samples in the Psychiatric Genomics Consortium (PGC). In the clinically-ascertained participants, the PPV of self-reported schizophrenia for a research diagnosis of schizophrenia was 0.70, which increased to 0.81 after expanding the research diagnosis to schizophrenia or SA-D. In UK Biobank, the PPV of self-reported schizophrenia for a medical record diagnosis was 0.74. Compared to participants who self-reported, participants with a clinically-ascertained research diagnosis were younger and more likely to have a high school qualification. Participants with a medical record diagnosis in UK Biobank were less likely to be employed or have a high school qualification than those who self-reported. Schizophrenia PRS did not differ between participants that had a diagnosis from self-report, research diagnosis or medical records. Polygenic liability r2, for all diagnosis definitions, fell within the distribution of PGC schizophrenia cohorts. Self-reported measures of schizophrenia are justified in genomic research to maximise sample size and reduce the burden of in-depth interviews on participants, although within sample validation of diagnoses is recommended.

Item Type:	Article
Date Type:	Published Online
Status:	Published
Schools:	Medicine
Additional Information:	License information from Publisher: LICENSE 1: URL: http://creativecommons.org/licenses/by/4.0/, Type: open-access
Publisher:	Nature Publishing Group UK
ISSN:	2754-6993
Funders:	MRC, NIH
Date of First Compliant Deposit:	1 November 2024
Date of Acceptance:	15 October 2024
Last Modified:	04 Dec 2024 09:45
URI:	https://orca.cardiff.ac.uk/id/eprint/173591

Actions (repository staff only)

Edit Item