|
|
| PMID |
|
|
| TITLE |
|
| Can We Improve the Prediction of Early Onset Mania and Hypomania in the Community? |
|
| ABSTRACT |
|
| BACKGROUND |
|
| There is broad agreement that the onset of bipolar disorders (BD) can be predicted by using combined estimates of familial, genetic and clinical risk. However, there is a lack of consensus about the operationalisation of different risk attributes (e.g., symptoms vs. sub-threshold syndromes; disorder-specific polygenic risk scores [PRS] vs. multiple-disorder PRS dimensions) and their utility for predicting bipolar type 1 (BD-I) and 2 (BD-II). Likewise, analyses often fail to consider the optimal model for predicting outcomes where true cases will be in the minority. |
| METHODOLOGY |
|
| There is broad agreement that the onset of bipolar disorders (BD) can be predicted by using combined estimates of familial, genetic and clinical risk. However, there is a lack of consensus about the operationalisation of different risk attributes (e.g., symptoms vs. sub-threshold syndromes; disorder-specific polygenic risk scores [PRS] vs. multiple-disorder PRS dimensions) and their utility for predicting bipolar type 1 (BD-I) and 2 (BD-II). Likewise, analyses often fail to consider the optimal model for predicting outcomes where true cases will be in the minority. Proof of concept study employing an ensemble machine learning approach (Boosting) to develop models for classifying BD cases vs. non-cases using different combinations of risk attributes extracted from a database from a prospective longitudinal follow-up of twin and non-twin siblings in the peak age range for onset of major mental disorders. |
| RESULTS |
|
| There is broad agreement that the onset of bipolar disorders (BD) can be predicted by using combined estimates of familial, genetic and clinical risk. However, there is a lack of consensus about the operationalisation of different risk attributes (e.g., symptoms vs. sub-threshold syndromes; disorder-specific polygenic risk scores [PRS] vs. multiple-disorder PRS dimensions) and their utility for predicting bipolar type 1 (BD-I) and 2 (BD-II). Likewise, analyses often fail to consider the optimal model for predicting outcomes where true cases will be in the minority. Proof of concept study employing an ensemble machine learning approach (Boosting) to develop models for classifying BD cases vs. non-cases using different combinations of risk attributes extracted from a database from a prospective longitudinal follow-up of twin and non-twin siblings in the peak age range for onset of major mental disorders. Of 1473 participants (mean age 26.3; female = 866), 104 developed BD-I (n = 30) or BD-II (n = 74). The best performing Boosting classification had an overall area under the receiver operating curve (AUROC) of 85.1% (95% Confidence Intervals: 80%, 88%); correctly identifying 86.7% BD cases. Variables with greatest relative influence were, in rank order: depressive symptoms, psychotic symptoms, a BD-Schizophrenia PRS dimension, hypomanic symptoms, and family history of BD. The model accurately classified 89% of manic cases but only 68% of hypomanic cases. |
| CONCLUSIONS |
|
| There is broad agreement that the onset of bipolar disorders (BD) can be predicted by using combined estimates of familial, genetic and clinical risk. However, there is a lack of consensus about the operationalisation of different risk attributes (e.g., symptoms vs. sub-threshold syndromes; disorder-specific polygenic risk scores [PRS] vs. multiple-disorder PRS dimensions) and their utility for predicting bipolar type 1 (BD-I) and 2 (BD-II). Likewise, analyses often fail to consider the optimal model for predicting outcomes where true cases will be in the minority. Proof of concept study employing an ensemble machine learning approach (Boosting) to develop models for classifying BD cases vs. non-cases using different combinations of risk attributes extracted from a database from a prospective longitudinal follow-up of twin and non-twin siblings in the peak age range for onset of major mental disorders. Of 1473 participants (mean age 26.3; female = 866), 104 developed BD-I (n = 30) or BD-II (n = 74). The best performing Boosting classification had an overall area under the receiver operating curve (AUROC) of 85.1% (95% Confidence Intervals: 80%, 88%); correctly identifying 86.7% BD cases. Variables with greatest relative influence were, in rank order: depressive symptoms, psychotic symptoms, a BD-Schizophrenia PRS dimension, hypomanic symptoms, and family history of BD. The model accurately classified 89% of manic cases but only 68% of hypomanic cases. Improving the accurate prediction of BD onset would benefit from greater consensus regarding the operationalisation of known risk attributes and selection of analytic models that consider sample imbalances. |
| © 2026 The Author(s). Bipolar Disorders published by John Wiley & Sons Ltd. |
|
| DATE PUBLISHED |
|
|
| HISTORY |
|
| PUBSTATUS |
PUBSTATUSDATE |
| revised |
2026/05/25 |
| received |
2025/11/03 |
| accepted |
2026/06/08 |
| medline |
2026/06/27 07:10 |
| pubmed |
2026/06/26 06:38 |
| entrez |
2026/06/26 03:02 |
| pmc-release |
2026/06/26 |
|
| AUTHORS |
|
| NAME |
COLLECTIVENAME |
LASTNAME |
FORENAME |
INITIALS |
AFFILIATION |
AFFILIATIONINFO |
| Scott J |
|
Scott |
Jan |
J |
|
Institute of Neuroscience, Newcastle University, Newcastle, UK. |
| Crouse JJ |
|
Crouse |
Jacob J |
JJ |
|
Brain and Mind Centre, The University of Sydney, Sydney, Australia. |
| Medland SE |
|
Medland |
Sarah E |
SE |
|
School of Psychology and Counselling, Queensland University of Technology, Brisbane, Queensland, Australia. |
| Mitchell BL |
|
Mitchell |
Brittany L |
BL |
|
Institute of Molecular Bioscience, The University of Queensland, Brisbane, Australia. |
| Gillespie NA |
|
Gillespie |
Nathan A |
NA |
|
Virginia Institute for Psychiatric and Behavioral Genetics, Virginia Commonwealth University, Richmond, Virginia, USA. |
| Martin NG |
|
Martin |
Nicholas G |
NG |
|
Brain and Mental Health Program, QIMR Berghofer Institute of Medical Research, Brisbane, Australia. |
| Hickie IB |
|
Hickie |
Ian B |
IB |
|
Brain and Mind Centre, The University of Sydney, Sydney, Australia. |
|
| INVESTIGATORS |
|
|
| JOURNAL |
|
| VOLUME: 28 |
| ISSUE: 5 |
| TITLE: Bipolar disorders |
| ISOABBREVIATION: Bipolar Disord |
| YEAR: 2026 |
| MONTH: Aug |
| DAY: |
| MEDLINEDATE: |
| SEASON: |
| CITEDMEDIUM: Internet |
| ISSN: 1399-5618 |
| ISSNTYPE: Electronic |
|
| MEDLINE JOURNAL |
|
| MEDLINETA: Bipolar Disord |
| COUNTRY: Denmark |
| ISSNLINKING: 1398-5647 |
| NLMUNIQUEID: 100883596 |
|
| PUBLICATION TYPE |
|
| PUBLICATIONTYPE TEXT |
| Journal Article |
| Research Support, Non-U.S. Gov't |
|
| COMMENTS AND CORRECTIONS |
|
|
| GRANTS |
|
| GRANTID |
AGENCY |
COUNTRY |
| 1031119 |
National Health and Medical Research Council |
|
| 1049911 |
National Health and Medical Research Council |
|
| APP10499110 |
National Health and Medical Research Council |
|
|
| GENERAL NOTE |
|
|
| KEYWORDS |
|
| KEYWORD |
| bipolar disorder |
| boosting |
| family history |
| onset |
| polygenic scores |
| prediction |
| youth |
|
| MESH HEADINGS |
|
| DESCRIPTORNAME |
QUALIFIERNAME |
| Humans |
|
| Bipolar Disorder |
classification |
| Mania |
genetics |
| Female |
genetics |
| Male |
genetics |
| Age of Onset |
genetics |
| Genetic Risk Score |
genetics |
| Boosting Machine Learning Algorithms |
genetics |
| Predictive Learning Models |
genetics |
| Prediction Algorithms |
genetics |
| Classification Algorithms |
genetics |
| Adolescent |
genetics |
|
| SUPPLEMENTARY MESH |
|
|
| GENE SYMBOLS |
|
|
| CHEMICALS |
|
|
| OTHER ID's |
|
|
|
