Major depressive disorder subtypes to predict long-term course

Hanna M. van Loo; Tianxi Cai; Michael J. Gruber; Junlong Li; Peter de  Jonge; Maria Petukhova; Sherri Rose; Nancy A. Sampson; Robert A. Schoevers; Klaas J. Wardenaar; Marsha A. Wilcox; Ali Obaid Al-Hamzawi; Laura Helena Andrade; Evelyn J. Bromet; Brendan Bunting; John Fayyad; Silvia E. Florescu; Oye Gureje; Chiyi Hu; Yueqin Huang; Daphna Levinson; Maria Elena Medina-Mora; Yoshibumi Nakane; Jose Posada-Villa; Kate M. Scott; Miguel Xavier; Zahari Zarkov; Ronald C. Kessler

doi:10.1002/da.22233

Major depressive disorder subtypes to predict long-term course

Hanna M. van Loo, Tianxi Cai, Michael J. Gruber, Junlong Li, Peter de Jonge, Maria Petukhova, Sherri Rose, Nancy A. Sampson, Robert A. Schoevers, Klaas J. Wardenaar, Marsha A. Wilcox, Ali Obaid Al-Hamzawi, Laura Helena Andrade, Evelyn J. Bromet, Brendan Bunting, John Fayyad, Silvia E. Florescu, Oye Gureje, Chiyi Hu, Yueqin HuangDaphna Levinson, Maria Elena Medina-Mora, Yoshibumi Nakane, Jose Posada-Villa, Kate M. Scott, Miguel Xavier, Zahari Zarkov, Ronald C. Kessler^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

56 Citations (Scopus)

Abstract

BACKGROUND: Variation in the course of major depressive disorder (MDD) is not strongly predicted by existing subtype distinctions. A new subtyping approach is considered here.

METHODS: Two data mining techniques, ensemble recursive partitioning and Lasso generalized linear models (GLMs), followed by k-means cluster analysis are used to search for subtypes based on index episode symptoms predicting subsequent MDD course in the World Mental Health (WMH) surveys. The WMH surveys are community surveys in 16 countries. Lifetime DSM-IV MDD was reported by 8,261 respondents. Retrospectively reported outcomes included measures of persistence (number of years with an episode, number of years with an episode lasting most of the year) and severity (hospitalization for MDD, disability due to MDD).

RESULTS: Recursive partitioning found significant clusters defined by the conjunctions of early onset, suicidality, and anxiety (irritability, panic, nervousness-worry-anxiety) during the index episode. GLMs found additional associations involving a number of individual symptoms. Predicted values of the four outcomes were strongly correlated. Cluster analysis of these predicted values found three clusters having consistently high, intermediate, or low predicted scores across all outcomes. The high-risk cluster (30.0% of respondents) accounted for 52.9-69.7% of high persistence and severity, and it was most strongly predicted by index episode severe dysphoria, suicidality, anxiety, and early onset. A total symptom count, in comparison, was not a significant predictor.

CONCLUSIONS: Despite being based on retrospective reports, results suggest that useful MDD subtyping distinctions can be made using data mining methods. Further studies are needed to test and expand these results with prospective data.

Original language	English
Pages (from-to)	765-777
Number of pages	13
Journal	Depression and Anxiety
Volume	31
Issue number	9
DOIs	https://doi.org/10.1002/da.22233
Publication status	Published - Sept-2014

Keywords

epidemiology
depression
anxiety
anxiety disorders
suicide
self-harm
panic attacks
RECURSIVE PARTITIONING ANALYSIS
WORLD-HEALTH-ORGANIZATION
COMMON MENTAL-DISORDERS
LARGE-SAMPLE
GENERAL-POPULATION
ANXIETY DISORDERS
RISK
COMORBIDITY
INPATIENTS
REMISSION

Access to Document

10.1002/da.22233

Handle.net

Persistent link

Embargoed Document

Major depressive disorder subtypes to predict long-term course
Final publisher's version, 469 KB
Request copy

Cite this

van Loo, H. M., Cai, T., Gruber, M. J., Li, J., de Jonge, P., Petukhova, M., Rose, S., Sampson, N. A., Schoevers, R. A., Wardenaar, K. J., Wilcox, M. A., Al-Hamzawi, A. O., Andrade, L. H., Bromet, E. J., Bunting, B., Fayyad, J., Florescu, S. E., Gureje, O., Hu, C., ... Kessler, R. C. (2014). Major depressive disorder subtypes to predict long-term course. Depression and Anxiety, 31(9), 765-777. https://doi.org/10.1002/da.22233

@article{cff27cece2f14f9e89e35c451e5ba670,

title = "Major depressive disorder subtypes to predict long-term course",

abstract = "BACKGROUND: Variation in the course of major depressive disorder (MDD) is not strongly predicted by existing subtype distinctions. A new subtyping approach is considered here.METHODS: Two data mining techniques, ensemble recursive partitioning and Lasso generalized linear models (GLMs), followed by k-means cluster analysis are used to search for subtypes based on index episode symptoms predicting subsequent MDD course in the World Mental Health (WMH) surveys. The WMH surveys are community surveys in 16 countries. Lifetime DSM-IV MDD was reported by 8,261 respondents. Retrospectively reported outcomes included measures of persistence (number of years with an episode, number of years with an episode lasting most of the year) and severity (hospitalization for MDD, disability due to MDD).RESULTS: Recursive partitioning found significant clusters defined by the conjunctions of early onset, suicidality, and anxiety (irritability, panic, nervousness-worry-anxiety) during the index episode. GLMs found additional associations involving a number of individual symptoms. Predicted values of the four outcomes were strongly correlated. Cluster analysis of these predicted values found three clusters having consistently high, intermediate, or low predicted scores across all outcomes. The high-risk cluster (30.0% of respondents) accounted for 52.9-69.7% of high persistence and severity, and it was most strongly predicted by index episode severe dysphoria, suicidality, anxiety, and early onset. A total symptom count, in comparison, was not a significant predictor.CONCLUSIONS: Despite being based on retrospective reports, results suggest that useful MDD subtyping distinctions can be made using data mining methods. Further studies are needed to test and expand these results with prospective data.",

keywords = "epidemiology, depression, anxiety, anxiety disorders, suicide, self-harm, panic attacks, RECURSIVE PARTITIONING ANALYSIS, WORLD-HEALTH-ORGANIZATION, COMMON MENTAL-DISORDERS, LARGE-SAMPLE, GENERAL-POPULATION, ANXIETY DISORDERS, RISK, COMORBIDITY, INPATIENTS, REMISSION",

author = "{van Loo}, {Hanna M.} and Tianxi Cai and Gruber, {Michael J.} and Junlong Li and {de Jonge}, Peter and Maria Petukhova and Sherri Rose and Sampson, {Nancy A.} and Schoevers, {Robert A.} and Wardenaar, {Klaas J.} and Wilcox, {Marsha A.} and Al-Hamzawi, {Ali Obaid} and Andrade, {Laura Helena} and Bromet, {Evelyn J.} and Brendan Bunting and John Fayyad and Florescu, {Silvia E.} and Oye Gureje and Chiyi Hu and Yueqin Huang and Daphna Levinson and Medina-Mora, {Maria Elena} and Yoshibumi Nakane and Jose Posada-Villa and Scott, {Kate M.} and Miguel Xavier and Zahari Zarkov and Kessler, {Ronald C.}",

note = "{\textcopyright} 2014 Wiley Periodicals, Inc.",

year = "2014",

month = sep,

doi = "10.1002/da.22233",

language = "English",

volume = "31",

pages = "765--777",

journal = "Depression and Anxiety",

issn = "1091-4269",

publisher = "Wiley",

number = "9",

}

van Loo, HM, Cai, T, Gruber, MJ, Li, J, de Jonge, P, Petukhova, M, Rose, S, Sampson, NA, Schoevers, RA , Wardenaar, KJ, Wilcox, MA, Al-Hamzawi, AO, Andrade, LH, Bromet, EJ, Bunting, B, Fayyad, J, Florescu, SE, Gureje, O, Hu, C, Huang, Y, Levinson, D, Medina-Mora, ME, Nakane, Y, Posada-Villa, J, Scott, KM, Xavier, M, Zarkov, Z & Kessler, RC 2014, 'Major depressive disorder subtypes to predict long-term course', Depression and Anxiety, vol. 31, no. 9, pp. 765-777. https://doi.org/10.1002/da.22233

TY - JOUR

T1 - Major depressive disorder subtypes to predict long-term course

AU - van Loo, Hanna M.

AU - Cai, Tianxi

AU - Gruber, Michael J.

AU - Li, Junlong

AU - de Jonge, Peter

AU - Petukhova, Maria

AU - Rose, Sherri

AU - Sampson, Nancy A.

AU - Schoevers, Robert A.

AU - Wardenaar, Klaas J.

AU - Wilcox, Marsha A.

AU - Al-Hamzawi, Ali Obaid

AU - Andrade, Laura Helena

AU - Bromet, Evelyn J.

AU - Bunting, Brendan

AU - Fayyad, John

AU - Florescu, Silvia E.

AU - Gureje, Oye

AU - Hu, Chiyi

AU - Huang, Yueqin

AU - Levinson, Daphna

AU - Medina-Mora, Maria Elena

AU - Nakane, Yoshibumi

AU - Posada-Villa, Jose

AU - Scott, Kate M.

AU - Xavier, Miguel

AU - Zarkov, Zahari

AU - Kessler, Ronald C.

PY - 2014/9

Y1 - 2014/9

N2 - BACKGROUND: Variation in the course of major depressive disorder (MDD) is not strongly predicted by existing subtype distinctions. A new subtyping approach is considered here.METHODS: Two data mining techniques, ensemble recursive partitioning and Lasso generalized linear models (GLMs), followed by k-means cluster analysis are used to search for subtypes based on index episode symptoms predicting subsequent MDD course in the World Mental Health (WMH) surveys. The WMH surveys are community surveys in 16 countries. Lifetime DSM-IV MDD was reported by 8,261 respondents. Retrospectively reported outcomes included measures of persistence (number of years with an episode, number of years with an episode lasting most of the year) and severity (hospitalization for MDD, disability due to MDD).RESULTS: Recursive partitioning found significant clusters defined by the conjunctions of early onset, suicidality, and anxiety (irritability, panic, nervousness-worry-anxiety) during the index episode. GLMs found additional associations involving a number of individual symptoms. Predicted values of the four outcomes were strongly correlated. Cluster analysis of these predicted values found three clusters having consistently high, intermediate, or low predicted scores across all outcomes. The high-risk cluster (30.0% of respondents) accounted for 52.9-69.7% of high persistence and severity, and it was most strongly predicted by index episode severe dysphoria, suicidality, anxiety, and early onset. A total symptom count, in comparison, was not a significant predictor.CONCLUSIONS: Despite being based on retrospective reports, results suggest that useful MDD subtyping distinctions can be made using data mining methods. Further studies are needed to test and expand these results with prospective data.

AB - BACKGROUND: Variation in the course of major depressive disorder (MDD) is not strongly predicted by existing subtype distinctions. A new subtyping approach is considered here.METHODS: Two data mining techniques, ensemble recursive partitioning and Lasso generalized linear models (GLMs), followed by k-means cluster analysis are used to search for subtypes based on index episode symptoms predicting subsequent MDD course in the World Mental Health (WMH) surveys. The WMH surveys are community surveys in 16 countries. Lifetime DSM-IV MDD was reported by 8,261 respondents. Retrospectively reported outcomes included measures of persistence (number of years with an episode, number of years with an episode lasting most of the year) and severity (hospitalization for MDD, disability due to MDD).RESULTS: Recursive partitioning found significant clusters defined by the conjunctions of early onset, suicidality, and anxiety (irritability, panic, nervousness-worry-anxiety) during the index episode. GLMs found additional associations involving a number of individual symptoms. Predicted values of the four outcomes were strongly correlated. Cluster analysis of these predicted values found three clusters having consistently high, intermediate, or low predicted scores across all outcomes. The high-risk cluster (30.0% of respondents) accounted for 52.9-69.7% of high persistence and severity, and it was most strongly predicted by index episode severe dysphoria, suicidality, anxiety, and early onset. A total symptom count, in comparison, was not a significant predictor.CONCLUSIONS: Despite being based on retrospective reports, results suggest that useful MDD subtyping distinctions can be made using data mining methods. Further studies are needed to test and expand these results with prospective data.

KW - epidemiology

KW - depression

KW - anxiety

KW - anxiety disorders

KW - suicide

KW - self-harm

KW - panic attacks

KW - RECURSIVE PARTITIONING ANALYSIS

KW - WORLD-HEALTH-ORGANIZATION

KW - COMMON MENTAL-DISORDERS

KW - LARGE-SAMPLE

KW - GENERAL-POPULATION

KW - ANXIETY DISORDERS

KW - RISK

KW - COMORBIDITY

KW - INPATIENTS

KW - REMISSION

U2 - 10.1002/da.22233

DO - 10.1002/da.22233

M3 - Article

C2 - 24425049

SN - 1091-4269

VL - 31

SP - 765

EP - 777

JO - Depression and Anxiety

JF - Depression and Anxiety

IS - 9

ER -

Major depressive disorder subtypes to predict long-term course

Abstract

Keywords

Access to Document

Handle.net

Embargoed Document

Fingerprint

Cite this