2
|
Ferris LM, Weiner JP, Saloner B, Kharrazi H. Comparing person-level matching algorithms to identify risk across disparate datasets among patients with a controlled substance prescription: retrospective analysis. JAMIA Open 2022; 5:ooac020. [PMID: 35571361 PMCID: PMC9097759 DOI: 10.1093/jamiaopen/ooac020] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/23/2021] [Revised: 02/25/2022] [Accepted: 03/10/2022] [Indexed: 11/13/2022] Open
Abstract
Abstract
Background
The opioid epidemic in the United States has precipitated a need for public health agencies to better understand risk factors associated with fatal overdoses. Matching person-level information stored in public health, medical, and human services datasets can enhance the understanding of opioid overdose risk factors and interventions.
Objective
This study compares approximate match versus exact match algorithms to link disparate datasets together for identifying persons at risk from an applied perspective.
Methods
This study used statewide prescription drug monitoring program (PDMP), arrest, and mortality data matched at the person-level using an approximate match and 2 exact match algorithms. Impact of matching was assessed by analyzing 3 independent concepts: (1) the prevalence of key risk indicators used by PDMP programs in practice, (2) the prevalence of arrests and fatal opioid overdose, and (3) the performance of a multivariate logistic regression for fatal opioid overdose. The PDMP key risk indicators included (1) multiple provider episodes (MPE), or patients with prescriptions from multiple prescribers and dispensers, (2) high morphine milligram equivalents (MMEs), which represents an opioid’s potency relative to morphine, and (3) overlapping opioid and benzodiazepine prescriptions.
Results
Prevalence of PDMP-based risk indicators were higher in the approximate match population for MPEs (n = 4893/1 859 445 [0.26%]) and overlapping opioid/benzodiazepines (n = 57 888/1 859 445 [4.71%]), but the exact-basic match population had the highest prevalence of individuals with high MMEs (n = 664/1 910 741 [3.11%]). Prevalence of arrests and deaths were highest for the approximate match population compared with the exact match populations. Model performance was comparable across the 3 matching algorithms (exact-basic validation area under the receiver operating characteristic curve [AUC]: 0.854; approximate validation AUC: 0.847; exact + zip validation AUC: 0.826) but resulted in different cutoff points balancing sensitivity and specificity.
Conclusions
Our study illustrates the specific tradeoffs of different matching methods. Further research should be performed to compare matching algorithms and its impact on the prevalence of key risk indicators in an applied setting that can improve understanding of risk within a population.
Collapse
Affiliation(s)
- Lindsey M Ferris
- Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
- The Chesapeake Regional Information System for our Patients, Baltimore, Maryland, USA
| | - Jonathan P Weiner
- Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
- Johns Hopkins Center for Population Health Information Technology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
| | - Brendan Saloner
- Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
| | - Hadi Kharrazi
- Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
- Johns Hopkins Center for Population Health Information Technology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
| |
Collapse
|
3
|
Ripperger M, Lotspeich SC, Wilimitis D, Fry CE, Roberts A, Lenert M, Cherry C, Latham S, Robinson K, Chen Q, McPheeters ML, Tyndall B, Walsh CG. Ensemble learning to predict opioid-related overdose using statewide prescription drug monitoring program and hospital discharge data in the state of Tennessee. J Am Med Inform Assoc 2021; 29:22-32. [PMID: 34665246 PMCID: PMC8714265 DOI: 10.1093/jamia/ocab218] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/04/2021] [Revised: 09/03/2021] [Indexed: 12/11/2022] Open
Abstract
Objective To develop and validate algorithms for predicting 30-day fatal and nonfatal opioid-related overdose using statewide data sources including prescription drug monitoring program data, Hospital Discharge Data System data, and Tennessee (TN) vital records. Current overdose prevention efforts in TN rely on descriptive and retrospective analyses without prognostication. Materials and Methods Study data included 3 041 668 TN patients with 71 479 191 controlled substance prescriptions from 2012 to 2017. Statewide data and socioeconomic indicators were used to train, ensemble, and calibrate 10 nonparametric “weak learner” models. Validation was performed using area under the receiver operating curve (AUROC), area under the precision recall curve, risk concentration, and Spiegelhalter z-test statistic. Results Within 30 days, 2574 fatal overdoses occurred after 4912 prescriptions (0.0069%) and 8455 nonfatal overdoses occurred after 19 460 prescriptions (0.027%). Discrimination and calibration improved after ensembling (AUROC: 0.79–0.83; Spiegelhalter P value: 0–.12). Risk concentration captured 47–52% of cases in the top quantiles of predicted probabilities. Discussion Partitioning and ensembling enabled all study data to be used given computational limits and helped mediate case imbalance. Predicting risk at the prescription level can aggregate risk to the patient, provider, pharmacy, county, and regional levels. Implementing these models into Tennessee Department of Health systems might enable more granular risk quantification. Prospective validation with more recent data is needed. Conclusion Predicting opioid-related overdose risk at statewide scales remains difficult and models like these, which required a partnership between an academic institution and state health agency to develop, may complement traditional epidemiological methods of risk identification and inform public health decisions.
Collapse
Affiliation(s)
- Michael Ripperger
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Sarah C Lotspeich
- Department of Biostatistics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Drew Wilimitis
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Carrie E Fry
- Department of Health Policy, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Allison Roberts
- Office of Informatics and Analytics, Tennessee Department of Health, Nashville, Tennessee, USA
| | - Matthew Lenert
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Charlotte Cherry
- Office of Informatics and Analytics, Tennessee Department of Health, Nashville, Tennessee, USA
| | - Sanura Latham
- Office of Informatics and Analytics, Tennessee Department of Health, Nashville, Tennessee, USA
| | - Katelyn Robinson
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Qingxia Chen
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA.,Department of Biostatistics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Melissa L McPheeters
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA.,Department of Health Policy, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Ben Tyndall
- Office of Informatics and Analytics, Tennessee Department of Health, Nashville, Tennessee, USA
| | - Colin G Walsh
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA.,Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA.,Department of Psychiatry, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| |
Collapse
|
4
|
Moyo P, Zhao X, Thorpe CT, Thorpe JM, Sileanu FE, Cashy JP, Hale JA, Mor MK, Radomski TR, Donohue JM, Hausmann LRM, Hanlon JT, Good CB, Fine MJ, Gellad WF. Patterns of opioid prescriptions received prior to unintentional prescription opioid overdose death among Veterans. Res Social Adm Pharm 2018; 15:1007-1013. [PMID: 30385111 DOI: 10.1016/j.sapharm.2018.10.023] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2018] [Revised: 07/24/2018] [Accepted: 10/17/2018] [Indexed: 12/20/2022]
Abstract
BACKGROUND Few studies have assessed prescription opioid supply preceding death in individuals dying from unintentional prescription opioid overdoses, or described the characteristics of these individuals, particularly among Veterans. OBJECTIVES To describe the history of prescription opioid supply preceding prescription opioid overdose death among Veterans. METHODS In a national cohort of Veterans who filled ≥1 opioid prescriptions from the Veterans Health Administration (VA) or Medicare Part D during 2008-2013, we identified deaths from unintentional or undetermined-intent prescription opioid overdoses in 2012-2013. We captured opioid prescriptions using both linked VA and Part D data, and VA data only. RESULTS Among 1181 decedents, 643 (54.4%) had prescription opioid supply on the day of death, and 735 (62.2%) within 30 days based on linked data, compared to 40.1% and 46.7%, respectively, using VA data alone. Decedents with prescription opioid supply were significantly older and less likely to have alcohol or illicit drugs as co-occurring substances involved in the overdose. Using linked data, 241 (20.4%) decedents lacked prescription opioid supply within a year of death. CONCLUSIONS Many VA patients who die from prescription opioid overdose receive opioid prescriptions outside VA or not at all. It is important to supplement VA with non-VA data to more accurately measure prescription opioid exposure and improve opioid medication safety.
Collapse
Affiliation(s)
- Patience Moyo
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Health Services, Policy and Practice, Brown University School of Public Health, Providence, RI, USA
| | - Xinhua Zhao
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA
| | - Carolyn T Thorpe
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, USA
| | - Joshua M Thorpe
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, USA
| | - Florentina E Sileanu
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA
| | - John P Cashy
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA
| | - Jennifer A Hale
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA
| | - Maria K Mor
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Department of Biostatistics, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA, USA
| | - Thomas R Radomski
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Division of General Internal Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Julie M Donohue
- Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Health Policy & Management, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA, USA
| | - Leslie R M Hausmann
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Division of General Internal Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Joseph T Hanlon
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, USA; Division of Geriatric Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Chester B Good
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, USA; Division of Geriatric Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA; Center for Value Based Pharmaceutical Initiatives, UPMC Health Plan, Pittsburgh, PA, USA
| | - Michael J Fine
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Division of General Internal Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Walid F Gellad
- Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA; Center for Pharmaceutical Policy and Prescribing, University of Pittsburgh Health Policy Institute, Pittsburgh, PA, USA; Division of General Internal Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
| |
Collapse
|