Integrating human services and criminal justice data with claims data to predict risk of opioid overdose among Medicaid beneficiaries: A machine-learning approach.
Adolescent
Adult
Aged
Analgesics, Opioid
/ adverse effects
Child
Criminal Law
/ statistics & numerical data
Female
Humans
Machine Learning
Male
Medicaid
/ statistics & numerical data
Middle Aged
Opiate Overdose
/ epidemiology
Predictive Value of Tests
Retrospective Studies
Risk Assessment
/ methods
Risk Factors
Social Work
/ statistics & numerical data
United States
Young Adult
Journal
PloS one
ISSN: 1932-6203
Titre abrégé: PLoS One
Pays: United States
ID NLM: 101285081
Informations de publication
Date de publication:
2021
2021
Historique:
received:
23
09
2020
accepted:
24
02
2021
entrez:
18
3
2021
pubmed:
19
3
2021
medline:
14
10
2021
Statut:
epublish
Résumé
Health system data incompletely capture the social risk factors for drug overdose. This study aimed to improve the accuracy of a machine-learning algorithm to predict opioid overdose risk by integrating human services and criminal justice data with health claims data to capture the social determinants of overdose risk. This prognostic study included Medicaid beneficiaries (n = 237,259) in Allegheny County, Pennsylvania enrolled between 2015 and 2018, randomly divided into training, testing, and validation samples. We measured 290 potential predictors (239 derived from Medicaid claims data) in 30-day periods, beginning with the first observed Medicaid enrollment date during the study period. Using a gradient boosting machine, we predicted a composite outcome (i.e., fatal or nonfatal opioid overdose constructed using medical examiner and claims data) in the subsequent month. We compared prediction performance between a Medicaid claims only model to one integrating human services and criminal justice data with Medicaid claims (i.e., integrated model) using several metrics (e.g., C-statistic, number needed to evaluate [NNE] to identify one overdose). Beneficiaries were stratified into risk-score decile subgroups. The samples (training = 79,087, testing = 79,086, validation = 79,086) had similar characteristics (age = 38±18 years, female = 56%, white = 48%, having at least one overdose = 1.7% during study period). Using the validation sample, the integrated model slightly improved on the Medicaid claims only model (C-statistic = 0.885; 95%CI = 0.877-0.892 vs. C-statistic = 0.871; 95%CI = 0.863-0.878), with small corresponding improvements in the NNE and positive predictive value. Nine of the top 30 most important predictors in the integrated model were human services and criminal justice variables. Using the integrated model, approximately 70% of individuals with overdoses were members of the top risk decile (overdose rates in the subsequent month = 47/10,000 beneficiaries). Few individuals in the bottom 9 deciles had overdose episodes (0-12/10,000). Machine-learning algorithms integrating claims and social service and criminal justice data modestly improved opioid overdose prediction among Medicaid beneficiaries for a large U.S. county heavily affected by the opioid crisis.
Identifiants
pubmed: 33735222
doi: 10.1371/journal.pone.0248360
pii: PONE-D-20-30028
pmc: PMC7971495
doi:
Substances chimiques
Analgesics, Opioid
0
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Validation Study
Langues
eng
Sous-ensembles de citation
IM
Pagination
e0248360Subventions
Organisme : NIDA NIH HHS
ID : R01 DA044985
Pays : United States
Déclaration de conflit d'intérêts
The authors have declared that no competing interests exist.
Références
Public Health Rep. 2018 Nov;133(6):658-666
pubmed: 30300555
Am J Med. 2017 Mar;130(3):e113
pubmed: 28215952
JAMA. 2019 Feb 12;321(6):609-611
pubmed: 30747958
Med Care. 2016 Oct;54(10):901-6
pubmed: 27623005
Ann Intern Med. 2015 Jan 6;162(1):W1-73
pubmed: 25560730
Ann Intern Med. 2010 Jan 19;152(2):85-92
pubmed: 20083827
Pain Med. 2015 Aug;16(8):1566-79
pubmed: 26077738
Pain Med. 2016 Dec;17(12):2268-2279
pubmed: 28025361
Am J Manag Care. 2009 Dec;15(12):897-906
pubmed: 20001171
Pain. 2010 Aug;150(2):332-339
pubmed: 20554392
J Am Med Inform Assoc. 2017 Nov 1;24(6):1204-1210
pubmed: 29016967
Pharmacoepidemiol Drug Saf. 2019 Jan;28(1):62-69
pubmed: 29687539
Drug Alcohol Depend. 2008 Apr 1;94(1-3):38-47
pubmed: 18063321
Gend Med. 2010 Oct;7(5):402-13
pubmed: 21056867
J Pain. 2015 May;16(5):445-53
pubmed: 25681095
Am J Prev Med. 2019 Dec;57(6):e211-e217
pubmed: 31753274
J Gen Intern Med. 2018 Oct;33(10):1646-1653
pubmed: 29380216
BMC Health Serv Res. 2006 Apr 04;6:46
pubmed: 16595013
Pain. 2013 Nov;154(11):2287-2296
pubmed: 23792283
Clin J Pain. 2010 Jan;26(1):1-8
pubmed: 20026946
Pain Med. 2005 Nov-Dec;6(6):432-42
pubmed: 16336480
Pain Med. 2018 Jan 1;19(1):68-78
pubmed: 28340046
J Community Hosp Intern Med Perspect. 2012 Apr 30;2(1):
pubmed: 23882354
JAMA Intern Med. 2014 May;174(5):796-801
pubmed: 24589873
MMWR Morb Mortal Wkly Rep. 2018 Jan 04;67(5152):1419-1427
pubmed: 30605448
Psychol Serv. 2017 Feb;14(1):34-49
pubmed: 28134555
NCHS Data Brief. 2015 Mar;(190):1-8
pubmed: 25932890
JAMA. 2008 Dec 10;300(22):2613-20
pubmed: 19066381
PLoS One. 2020 Jan 17;15(1):e0227966
pubmed: 31951640
Subst Abus. 2015;36(2):192-202
pubmed: 25671499
Crit Care. 2015 Aug 13;19:285
pubmed: 26268570
Med Care. 2017 Mar;55(3):291-298
pubmed: 27984346
Int J Drug Policy. 2019 Jun;68:37-45
pubmed: 30981166
JAMA. 2011 Apr 6;305(13):1315-21
pubmed: 21467284
Drug Alcohol Depend. 2015 Aug 1;153:346-9
pubmed: 26077605
JAMA Netw Open. 2019 Mar 1;2(3):e190968
pubmed: 30901048
J Pain. 2015 Apr;16(4):380-7
pubmed: 25640294
JAMA Netw Open. 2019 Jun 5;2(6):e196373
pubmed: 31251376
Drug Alcohol Depend. 2020 May 23;213:107997
pubmed: 32534407
JAMA. 2019 Oct 10;:1-3
pubmed: 31600370
Pharmacoepidemiol Drug Saf. 2017 May;26(5):509-517
pubmed: 28074520
Med Care. 2012 Jun;50(6):494-500
pubmed: 22410408
MMWR Morb Mortal Wkly Rep. 2009 Oct 30;58(42):1171-5
pubmed: 19875978
Med Care. 2017 Dec;55(12):e104-e112
pubmed: 29135773
Drug Saf. 2012 Apr 1;35(4):325-34
pubmed: 22339505
Am J Public Health. 2018 Feb;108(2):182-186
pubmed: 29267060
Ann Intern Med. 2018 Jun 19;168(12):837-845
pubmed: 29800019
PLoS One. 2015 Mar 04;10(3):e0118432
pubmed: 25738806
Pain Med. 2011 Jun;12 Suppl 2:S26-35
pubmed: 21668754
PLoS One. 2016 May 27;11(5):e0155705
pubmed: 27232332
Pain Med. 2012 Sep;13(9):1162-73
pubmed: 22845054
J Gen Intern Med. 2018 Sep;33(9):1512-1519
pubmed: 29948815
Public Health Rep. 2018 Nov/Dec;133(1_suppl):24S-34S
pubmed: 30426871
JAMA. 2011 Apr 6;305(13):1299-301
pubmed: 21467282
AMIA Annu Symp Proc. 2018 Dec 05;2018:527-534
pubmed: 30815093
BMJ. 2015 Oct 28;351:h5527
pubmed: 26511519