High-Risk Phenotypes of Early Psychiatric Readmission in Bipolar Disorder With Comorbid Medical Illness.
bipolar disorder
comorbidity
decision tree
medical illness
readmission
Journal
Psychosomatics
ISSN: 1545-7206
Titre abrégé: Psychosomatics
Pays: England
ID NLM: 0376506
Informations de publication
Date de publication:
Historique:
received:
10
04
2019
revised:
13
05
2019
accepted:
13
05
2019
pubmed:
8
7
2019
medline:
22
8
2020
entrez:
8
7
2019
Statut:
ppublish
Résumé
Individuals with co-existing serious mental illness and non-psychiatric medical illness are at high risk of acute care utilization. Mining of electronic health record data can help identify and categorize predictors of psychiatric hospital readmission in this population. This study aimed to identify modifiable predictors of psychiatric readmission among individuals with comorbid bipolar disorder and medical illness. This goal was accomplished by applying objective variable selection via machine learning techniques. This was a retrospective analysis of electronic health record data derived from 77,296 episodes of care from 2006 to 2016 within the University of California Health Care System. Data included 1,250 episodes of care involving patients with bipolar disorder and serious comorbid medical illnesses (defined by transfer between medicine and psychiatry services or concomitant primary medical and psychiatric admission diagnoses). Machine learning (classification trees) was used to identify potential predictors of 30-day psychiatric readmission across hospital encounters. Predictors included demographics, medical and psychiatric diagnoses, medication regimen, and disposition. The algorithm was internally validated using 10-fold cross-validation. The model predicted 30-day readmission with high accuracy (98% unbalanced model, 88% balanced model). Modifiable predictors of readmission were length of stay, transfers between medical and psychiatric services, discharge disposition to home, and all-cause acute health service utilization in the year before the index hospitalization. Among bipolar disorder patients with comorbid medical conditions, characteristics of the index hospitalization (e.g., duration, transfer, and disposition) emerged as more predictive than static properties of the patient (e.g., sociodemographic factors and psychiatric comorbidity burden). Findings identified phenotypes of patients at high risk for rehospitalization and suggest potential ways of modifying the risk of early readmission.
Sections du résumé
BACKGROUND
Individuals with co-existing serious mental illness and non-psychiatric medical illness are at high risk of acute care utilization. Mining of electronic health record data can help identify and categorize predictors of psychiatric hospital readmission in this population.
OBJECTIVE
This study aimed to identify modifiable predictors of psychiatric readmission among individuals with comorbid bipolar disorder and medical illness. This goal was accomplished by applying objective variable selection via machine learning techniques.
METHOD
This was a retrospective analysis of electronic health record data derived from 77,296 episodes of care from 2006 to 2016 within the University of California Health Care System. Data included 1,250 episodes of care involving patients with bipolar disorder and serious comorbid medical illnesses (defined by transfer between medicine and psychiatry services or concomitant primary medical and psychiatric admission diagnoses). Machine learning (classification trees) was used to identify potential predictors of 30-day psychiatric readmission across hospital encounters. Predictors included demographics, medical and psychiatric diagnoses, medication regimen, and disposition. The algorithm was internally validated using 10-fold cross-validation.
RESULTS
The model predicted 30-day readmission with high accuracy (98% unbalanced model, 88% balanced model). Modifiable predictors of readmission were length of stay, transfers between medical and psychiatric services, discharge disposition to home, and all-cause acute health service utilization in the year before the index hospitalization.
CONCLUSION
Among bipolar disorder patients with comorbid medical conditions, characteristics of the index hospitalization (e.g., duration, transfer, and disposition) emerged as more predictive than static properties of the patient (e.g., sociodemographic factors and psychiatric comorbidity burden). Findings identified phenotypes of patients at high risk for rehospitalization and suggest potential ways of modifying the risk of early readmission.
Identifiants
pubmed: 31279490
pii: S0033-3182(19)30087-8
doi: 10.1016/j.psym.2019.05.002
pmc: PMC7071814
mid: NIHMS1533621
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
563-573Subventions
Organisme : NCATS NIH HHS
ID : UL1 TR001881
Pays : United States
Informations de copyright
Copyright © 2019 Academy of Consultation-Liaison Psychiatry. Published by Elsevier Inc. All rights reserved.
Références
Open Med. 2011;5(2):e104-11
pubmed: 21915234
PeerJ. 2017 Mar 29;5:e3137
pubmed: 28367375
Psychiatr Serv. 2012 Jan;63(1):87-90
pubmed: 22227766
Gen Hosp Psychiatry. 2016 Jul-Aug;41:6-12
pubmed: 27302718
BMJ. 1999 May 22;318(7195):1387-90
pubmed: 10334748
J Affect Disord. 2012 Dec 1;141(1):1-10
pubmed: 22497876
Early Interv Psychiatry. 2018 Aug;12(4):605-612
pubmed: 27587229
Psychiatr Serv. 2015 Feb 1;66(2):134-40
pubmed: 25642610
BMJ Open. 2016 Jun 27;6(6):e011060
pubmed: 27354072
Psychosomatics. 2014 Mar-Apr;55(2):134-43
pubmed: 24367898
Psychiatr Serv. 2002 Jan;53(1):76-9
pubmed: 11773653
Med Care. 2009 Jun;47(6):626-33
pubmed: 19433995
Psychiatr Serv. 2012 Sep 1;63(9):889-95
pubmed: 22751995
JAMA Psychiatry. 2013 Sep;70(9):931-9
pubmed: 23863861
World Psychiatry. 2014 Jun;13(2):153-60
pubmed: 24890068
World Psychiatry. 2017 Jun;16(2):163-180
pubmed: 28498599
Mil Med. 2010 Oct;175(10):732-41
pubmed: 20968262
JAMA Intern Med. 2016 Apr;176(4):496-502
pubmed: 26954698
J Emerg Nurs. 2014 May;40(3):e55-61
pubmed: 23628422
BMC Psychiatry. 2017 Jan 3;17(1):2
pubmed: 28049441
Acta Psychiatr Scand. 2000 May;101(5):367-73
pubmed: 10823296
Ann Pharmacother. 2009 Dec;43(12):1939-47
pubmed: 19934394
Br J Psychiatry. 2013 Mar;202(3):187-94
pubmed: 23457182
Compr Psychiatry. 2014 Apr;55(3):426-30
pubmed: 24405773
Psychiatry Clin Neurosci. 2014 Jun;68(6):418-24
pubmed: 24506520
Can J Psychiatry. 1996 Feb;41(1):49-51
pubmed: 8919424
Schizophr Res. 2016 Jan;170(1):184-90
pubmed: 26678982
J Am Acad Child Adolesc Psychiatry. 2004 Apr;43(4):440-51
pubmed: 15187804
Open Med. 2012 Jul 19;6(3):e80-90
pubmed: 23696773
PLoS One. 2011;6(5):e19590
pubmed: 21611123
Am J Epidemiol. 2013 Aug 15;178(4):645-51
pubmed: 23648805
Ann Clin Psychiatry. 2011 Feb;23(1):40-7
pubmed: 21318195
Psychiatr Serv. 2004 May;55(5):560-5
pubmed: 15128965
BMC Psychiatry. 2016 Dec 16;16(1):449
pubmed: 27986079
PLoS One. 2014 Oct 31;9(10):e111882
pubmed: 25360781
J Psychiatr Res. 2015 Feb;61:205-13
pubmed: 25537450
J Behav Health Serv Res. 2007 Apr;34(2):137-50
pubmed: 17437186
Psychiatr Serv. 2000 Jul;51(7):885-9
pubmed: 10875952
Biol Psychiatry. 2003 Aug 1;54(3):216-26
pubmed: 12893098