Hostname: page-component-74656f9884-8r8nv Total loading time: 0 Render date: 2026-11-11T01:20:41.274Z Has data issue: false hasContentIssue false
  • English
  • Français

Why so GLUMM? Detecting depression clusters through graphing lifestyle-environs using machine-learning methods (GLUMM)

Published online by Cambridge University Press:  23 March 2020

J.F. Dipnall ,
J.A. Pasco ,
M. Berk ,
L.J. Williams ,
S. Dodd ,
F.N. Jacka  and
D. Meyer
J.F. Dipnall*
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia Department of statistics, data science and epidemiology, Swinburne university of technology, Swinburne, Australia
J.A. Pasco
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia Melbourne clinical school-western campus, the university of Melbourne, Saint-Albans, VIC, Australia Department of epidemiology and preventive medicine, Monash university, Melbourne, VIC, Australia University hospital of Geelong, Geelong, VIC, Australia
M. Berk
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia University hospital of Geelong, Geelong, VIC, Australia Department of psychiatry, the university of Melbourne, Parkville, VIC, Australia Florey institute of neuroscience and mental health, Parkville, VIC, Australia Orygen, the National centre of excellence in youth mental health, Parkville, VIC, Australia
L.J. Williams
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia
S. Dodd
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia University hospital of Geelong, Geelong, VIC, Australia Department of psychiatry, the university of Melbourne, Parkville, VIC, Australia
F.N. Jacka
Affiliation:
Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia Department of psychiatry, the university of Melbourne, Parkville, VIC, Australia Centre for adolescent health, Murdoch children's research institute, Melbourne, Australia Black Dog institute, Sydney, Australia
D. Meyer
Affiliation:
Department of statistics, data science and epidemiology, Swinburne university of technology, Swinburne, Australia
*
⁎Corresponding author. Impact strategic research centre, school of medicine, Deakin university, PO Box 281, Geelong, Victoria 3220, Australia. E-mail address:jdipnall@deakin.edu.au (J.F. Dipnall).
Get access

Abstract

Background

Key lifestyle-environ risk factors are operative for depression, but it is unclear how risk factors cluster. Machine-learning (ML) algorithms exist that learn, extract, identify and map underlying patterns to identify groupings of depressed individuals without constraints. The aim of this research was to use a large epidemiological study to identify and characterise depression clusters through “Graphing lifestyle-environs using machine-learning methods” (GLUMM).

Methods

Two ML algorithms were implemented: unsupervised Self-organised mapping (SOM) to create GLUMM clusters and a supervised boosted regression algorithm to describe clusters. Ninety-six “lifestyle-environ” variables were used from the National health and nutrition examination study (2009–2010). Multivariate logistic regression validated clusters and controlled for possible sociodemographic confounders.

Results

The SOM identified two GLUMM cluster solutions. These solutions contained one dominant depressed cluster (GLUMM5-1, GLUMM7-1). Equal proportions of members in each cluster rated as highly depressed (17%). Alcohol consumption and demographics validated clusters. Boosted regression identified GLUMM5-1 as more informative than GLUMM7-1. Members were more likely to: have problems sleeping; unhealthy eating; ≤ 2 years in their home; an old home; perceive themselves underweight; exposed to work fumes; experienced sex at ≤ 14 years; not perform moderate recreational activities. A positive relationship between GLUMM5-1 (OR: 7.50, P < 0.001) and GLUMM7-1 (OR: 7.88, P < 0.001) with depression was found, with significant interactions with those married/living with partner (P = 0.001).

Conclusion

Using ML based GLUMM to form ordered depressive clusters from multitudinous lifestyle-environ variables enabled a deeper exploration of the heterogeneous data to uncover better understandings into relationships between the complex mental health factors.

Keywords

DepressionPsychiatryMachine learningBoosted regressionClusterLifestyle

Information

Type
Original article
Information
European Psychiatry , Volume 39 , January 2017 , pp. 40 - 50
Copyright
Copyright © Elsevier Masson SAS 2017

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

Footnotes

1

These authors contributed equally to this work.

Abbreviations: DIPIT, Data integration protocol in ten-steps, GLUMM, Graphing lifestyle-environs using machine-learning methods, GLUMM5-1, GLUMM solution 5 cluster 1, GLUMM5-2, GLUMM solution 5 cluster 2, GLUMM7-1, GLUMM solution 7 cluster 1, GLUMM7-3, GLUMM solution 7 cluster 3, GLUMM7-4, GLUMM solution 7 cluster 4, ML, Machine-learning, MART, Multiple additive regression trees, NCHS, National center for health statistics, NHANES, National health and nutrition examination survey, PHQ-9, Patient health questonnaire-9, SOMs, Self-organizing maps

References

Passos, ICMwangi, BKapczinski, FBig data analytics and machine learning: 2015 and beyond. Lancet Psychiatry 2016; 3: 13–15.CrossRefGoogle ScholarPubMed
Monteith, SGlenn, TGeddes, JBauer, MBig data are coming to psychiatry: a general introduction. Int J Bipolar Disord 2015; 3: 1–11.CrossRefGoogle ScholarPubMed
Samuel, ALSome studies in machine learning using the game of checkers. IBM J Res Develop 1959; 3: 210–229.CrossRefGoogle Scholar
Belson, WAMatching and prediction on the principle of biological classification. Appl Stat 1959;65–75.CrossRefGoogle Scholar
Witten, IHFrank, EHall, MAData mining: practical machine learning tools and techniques: practical machine learning tools and techniques. Morgan Kaufmann; 2011.Google Scholar
Kohenen, TSelf-organizing maps. Lecture notes in information sciences, 30. Springer; 1997.Google Scholar
Wehrens, RBuydens, LMSelf-and super-organizing maps in R: the Kohonen package. J Stat Softw 2007; 21: 1–19.CrossRefGoogle Scholar
Kohonen, TSelf-organized formation of topologically correct feature maps. Biol Cybern 1982; 43: 59–69.CrossRefGoogle Scholar
Mitchell, TMMachine learning Burr Ridge, IL: McGraw Hill; 1997. p. 45.Google Scholar
Shmueli, GPatel, NRBruce, PCData mining for business intelligence: concepts, techniques and applications in Microsoft Office Excel with XLMiner. Wiley; 2010.Google Scholar
Niculescu, ALevey, DPhalen, PLe-Niculescu, HDainton, HJain, Net al.Understanding and predicting suicidality using a combined genomic and clinical risk assessment approach. Mol Psychiatry 2015; 20: 1266–1285.CrossRefGoogle ScholarPubMed
Kessler, RCWarner, CHIvany, CPetukhova, MVRose, SBromet, EJet al.Predicting suicides after psychiatric hospitalization in US Army soldiers: the Army study to assess risk and resilience in service members (Army STARRS). JAMA Psychiatry 2015; 72: 49–57.CrossRefGoogle Scholar
Castro, VMRoberson, AMMcCoy, THWiste, ACagan, ASmoller, JWet al.Stratifying risk for renal insufficiency among lithium-treated patients: an electronic health record study. Neuropsychopharmacology 2015Google ScholarPubMed
Hahn, TKircher, TStraube, BWittchen, HUKonrad, CStröhle, Aet al.Predicting treatment response to cognitive behavioral therapy in panic disorder with agoraphobia by integrating local neural information. JAMA Psychiatry 2015; 72: 68–74.CrossRefGoogle ScholarPubMed
Chekroud, AMZotti, RJShehzad, ZGueorguieva, RJohnson, MKTrivedi, MHet al.Cross-trial prediction of treatment outcome in depression: a machine learning approach. Lancet Psychiatry 2016CrossRefGoogle ScholarPubMed
Monden, RWardenaar, KJStegeman, AConradi, HJde Jonge, PSimultaneous decomposition of depression heterogeneity on the person-, symptom-and time-level: the use of three-mode principal component analysis. Plos One 2015;10:e0132765.CrossRefGoogle ScholarPubMed
Widiger, TAClark, LAToward DSM-V and the classification of psychopathology. Psychol Bull 2000;126:946.CrossRefGoogle ScholarPubMed
Berk, MSarris, JCoulson, CJacka, FLifestyle management of unipolar depression. Acta Psychiatr Scand 2013; 127: 38–54.CrossRefGoogle Scholar
Hayley, ACSkogen, JCSivertsen, BWold, BBerk, MPasco, JAet al.Symptoms of depression and difficulty initiating sleep from early adolescence to early adulthood: a longitudinal study. Sleep 2014; 38: 1599–1606.CrossRefGoogle Scholar
Everitt, BHothorn, TCluster analysis. An introduction to applied multivariate analysis with R. Springer; 2011. p. 163–200.Google Scholar
Jain, AKDubes, RC Algorithms for clustering. Cap IV-cluster validity; 1988; 143–222.Google Scholar
Blashfield, RKMorey, LCThe classification of depression through cluster analysis. Compr Psychiatry 1979; 20: 516–527.CrossRefGoogle ScholarPubMed
Pilowsky, ILevine, SBoulton, DMThe classification of depression by numerical taxonomy. Br J Psychiatry 1969; 115: 937–945.CrossRefGoogle ScholarPubMed
Paykel, EClassification of depressed patients: a cluster analysis derived grouping. Br J Psychiatry 1971; 118: 275–288.CrossRefGoogle ScholarPubMed
Vesanto, JAlhoniemi, EClustering of the self-organizing map. Neural Networks IEEE Trans 2000; 11: 586–600.CrossRefGoogle ScholarPubMed
Van Hulle, MMSelf-organizing maps. Handbook of natural computing. Springer; 2012. p. 585–622.CrossRefGoogle Scholar
Waller, NGKaiser, HAIllian, JBManry, MA comparison of the classification capabilities of the 1-dimensional kohonen neural network with two partitioning and three hierarchical cluster analysis algorithms. Psychometrika 1998; 63: 5–22.CrossRefGoogle Scholar
Centers for disease control and prevention national center for health statistics. National health and nutrition examination survey: analytic guidelines, 1999–2010. U.S. department of health and human services; 2013.Google Scholar
Dipnall, JFBerk, MJacka, FNWilliams, LJDodd, SPasco, JAData integration protocol in ten-steps (DIPIT): a new standard for medical researchers. Methods 2014CrossRefGoogle ScholarPubMed
Kroenke, KSpitzer, RLThe PHQ-9: a new depression diagnostic and severity measure. Psychiatr Ann 2002; 32: 509–515.CrossRefGoogle Scholar
Kroenke, KSpitzer, RLWilliams, JBThe PHQ-9. J Gen Intern Med 2001; 16: 606–613.CrossRefGoogle ScholarPubMed
Grant, BFStinson, FSDawson, DAChou, SPDufour, MCCompton, Wet al.Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders: results from the National epidemiologic survey on alcohol and related conditions. Arch Gen Psychiatry 2004;61:807.CrossRefGoogle ScholarPubMed
Grant, BFHarford, TCComorbidity between DSM-IV alcohol use disorders and major depression: results of a national survey. Drug Alcohol Depend 1995; 39: 197–206.CrossRefGoogle ScholarPubMed
(CDC), CfDCaP, National center for health statistics (NCHS). National health and nutrition examination survey questionnaire. Hyattsville, MD: U.S. department of health and human services, centers for disease control and prevention; 2009-2010.Google Scholar
Kohonen, TSelf-organizing maps-springer series in information sciences, 30. Berlin: Springer Verlag; 1995.Google Scholar
Gabor, ALeach, RDowla, FAutomated seizure detection using a self-organizing neural network. Electroencephalogr Clin Neurophysiol 1996; 99: 257–266.CrossRefGoogle ScholarPubMed
Magdolen, JRappelsberger, PDorffner, GFlexer, AWinterer, GEvaluating multi-layer perceptrons and self-organising feature maps as a tool for identifying psychiatric disorders in EEG. Psychiatr Res Neuroimag 1997; 68: 171–172.CrossRefGoogle Scholar
Arnrich, BSetz, CLa Marca, RTröster, GEhlert, USelf organizing maps for affective state detection. Machine Learn Assist Technol 2010;45.Google Scholar
Köhn, HFHubert, LJHierarchical cluster analysis. Wiley StatsRef: statistics reference online; 2006.Google Scholar
Freund, YSchapire, REA decision-theoretic generalization of on-line learning and an application to boosting. J Comp Syst Sci 1997; 55: 119–139.CrossRefGoogle Scholar
Hastie, TTibshirani, RFriedman, JFranklin, JThe elements of statistical learning: data mining, inference and prediction. Math Intel 2005; 27: 83–85.Google Scholar
Dipnall, JFPasco, JABerk, MWilliams, LJDodd, SJacka, FNet al.Fusing data mining, machine learning and traditional statistics to detect biomarkers associated with depression. Plos One 2016;11:e0148195.CrossRefGoogle ScholarPubMed
Friedman, JHastie, TTibshirani, RThe elements of statistical learning. Springer series in statistics Berlin: Springer; 2001Google Scholar
Schonlau, MBoosted regression (boosting): an introductory tutorial and a Stata plugin. Stata J 2005;5:330.CrossRefGoogle Scholar
Friedman, JHMeulman, JJMultiple additive regression trees with application in epidemiology. Stat Med 2003; 22: 1365–1381.CrossRefGoogle ScholarPubMed
Van Voorhees, BWPaunesku, DGollan, JKuwabara, SReinecke, MBasu, APredicting future risk of depressive episode in adolescents: the Chicago Adolescent depression risk assessment (CADRA). Ann Fam Med 2008; 6: 503–511.CrossRefGoogle Scholar
Friedman, JHHall, POn bagging and nonlinear estimation. J Stat Plan Infer 2007; 137: 669–683.CrossRefGoogle Scholar
Friedman, JHastie, TTibshirani, RAdditive logistic regression: a statistical view of boosting (with discussion and a rejoinder by the authors). Ann Stat 2000; 28: 337–407.CrossRefGoogle Scholar
Rao, JNScott, AJThe analysis of categorical data from complex sample surveys: Chi2 tests for goodness of fit and independence in two-way tables. J Am Stat Assoc 1981; 76: 221–230.CrossRefGoogle Scholar
Rao, JNScott, AJOn Chi2 tests for multiway contingency tables with cell proportions estimated from survey data. Ann Stat 1984; 46–60.CrossRefGoogle Scholar
Sribney, WMTwo-way contingency tables for survey or clustered data. Stata Tech Bull 1999;8.Google Scholar
Archer, KJLemeshow, SGoodness-of-fit test for a logistic regression model fitted using survey sample data. Stata J 2006; 6: 97–105.CrossRefGoogle Scholar
Adrien, JNeurobiological bases for the relation between sleep and depression. Sleep Med Rev 2002; 6: 341–351.CrossRefGoogle ScholarPubMed
Jacka, FNPasco, JAMykletun, AWilliams, LJHodge, AMO’Reilly, SLet al.Association of Western and traditional diets with depression and anxiety in women. Am J Psychiatry 2010; 167: 305–311.CrossRefGoogle ScholarPubMed
Lai, JSHiles, SBisquera, AHure, AJMcEvoy, MAttia, JA systematic review and meta-analysis of dietary patterns and depression in community-dwelling adults. Am J Clin Nutr 2014; 99: 181–197.CrossRefGoogle ScholarPubMed
Dipnall, JFPasco, JAMeyer, DBerk, MWilliams, LJDodd, Set al.The association between dietary patterns, diabetes and depression. J Affect Disord 2015; 174: 215–224.CrossRefGoogle ScholarPubMed
Chief medical officers of England S, Wales, and Northern Ireland. Start active, stay active: a report on physical activity from the four home countries’ chief medical officers. Department of health; 2011.Google Scholar
Jacka, FNPJWilliams, LJLeslie, ERet al.Lower levels of physical activity in childhood associated with adult depression. J Sci Med Sport 2011; 14: 222–226.CrossRefGoogle ScholarPubMed
Pasco, JAWLJacka, FNet al.Habitual physical activity and the risk for depressive and anxiety disorders among older men and women. Int Psychogeriatr 2010; 24: 1–7.Google Scholar
Levesque, SSurace, MJMcDonald, JBlock, MLAir pollution & the brain: subchronic diesel exhaust exposure causes neuroinflammation and elevates early markers of neurodegenerative disease. J Neuroinflamm 2011;8:105.CrossRefGoogle ScholarPubMed
Weuve, JInvited commentary: how exposure to air pollution may shape dementia risk, and what epidemiology can say about it. Am J Epidemiol 2014; 180: 367–371.CrossRefGoogle Scholar
Nel, AAir pollution-related illness: effects of particles. Science 2005; 308: 804–806.CrossRefGoogle ScholarPubMed
Howren, MBLamkin, DMSuls, JAssociations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med. 2009; 71: 171–186.CrossRefGoogle ScholarPubMed
Pasco, JAWLJacka, FNet al.Tobacco smoking as a risk factor for major depressive disorder: population-based study. Br J Psychiatry 2008; 193: 322–326.CrossRefGoogle ScholarPubMed
Weissman, MMPaykel, ESMoving and depression in women. Society 1972; 9: 24–28.CrossRefGoogle Scholar
Grello, CMWelsh, DPHarper, MSDickson, JWDating and sexual relationship trajectories and adolescent functioning. Adolesc Fam Health 2003; 3: 103–112.Google Scholar
Davila, JStroud, CBStarr, LRMiller, MRYoneda, AHershenberg, RRomantic and sexual activities, parent–adolescent stress, and depressive symptoms among early adolescent girls. J Adolesc 2009; 32: 909–924.CrossRefGoogle ScholarPubMed
Joyner, KUdry, JRYou don’t bring me anything but down: Adolescent romance and depression. J Health Soc Behav 2000; 369–391.CrossRefGoogle ScholarPubMed
Bifulco, ABrown, GWAdler, ZEarly sexual abuse and clinical depression in adult life. Br J Psychiatry 1991; 159: 115–122.CrossRefGoogle ScholarPubMed
Submit a response

Comments

No Comments have been published for this article.