Evaluation of Machine-Learning Algorithms for Predicting Opioid Overdose Risk Among Medicare Beneficiaries With Opioid Prescriptions
Open accessThe DOI matches its Crossref record; 62 of the 62 references it cites were found in a registry or an index.
Abstract
Importance: Current approaches to identifying individuals at high risk for opioid overdose target many patients who are not truly at high risk. Objective: To develop and validate a machine-learning algorithm to predict opioid overdose risk among Medicare beneficiaries with at least 1 opioid prescription. Design, Setting, and Participants: A prognostic study was conducted between September 1, 2017, and December 31, 2018. Participants (n = 560 057) included fee-for-service Medicare beneficiaries without cancer who filled 1 or more opioid prescriptions from January 1, 2011, to December 31, 2015. Beneficiaries were randomly and equally divided into training, testing, and validation samples. Exposures: Potential predictors (n = 268), including sociodemographics, health status, patterns of opioid use, and practitioner-level and regional-level factors, were measured in 3-month windows, starting 3 months before initiating opioids until loss of follow-up or the end of observation. Main Outcomes and Measures: Opioid overdose episodes from inpatient and emergency department claims were identified. Multivariate logistic regression (MLR), least absolute shrinkage and selection operator-type regression (LASSO), random forest (RF), gradient boosting machine (GBM), and deep neural network (DNN) were applied to predict overdose risk in the subsequent 3 months after initiation of treatment with prescription opioids. Prediction performance was assessed using the C statistic and other metrics (eg, sensitivity, specificity, and number needed to evaluate [NNE] to identify one overdose). The Youden index was used to identify the optimized threshold of predicted score that balanced sensitivity and specificity. Results: Beneficiaries in the training (n = 186 686), testing (n = 186 685), and validation (n = 186 686) samples had similar characteristics (mean [SD] age of 68.0 [14.5] years, and approximately 63% were female, 82% were white, 35% had disabilities, 41% were dual eligible, and 0.60% had at least 1 overdose episode). In the validation sample, the DNN (C statistic = 0.91; 95% CI, 0.88-0.93) and GBM (C statistic = 0.90; 95% CI, 0.87-0.94) algorithms outperformed the LASSO (C statistic = 0.84; 95% CI, 0.80-0.89), RF (C statistic = 0.80; 95% CI, 0.75-0.84), and MLR (C statistic = 0.75; 95% CI, 0.69-0.80) methods for predicting opioid overdose. At the optimized sensitivity and specificity, DNN had a sensitivity of 92.3%, specificity of 75.7%, NNE of 542, positive predictive value of 0.18%, and negative predictive value of 99.9%. The DNN classified patients into low-risk (76.2% [142 180] of the cohort), medium-risk (18.6% [34 579] of the cohort), and high-risk (5.2% [9747] of the cohort) subgroups, with only 1 in 10 000 in the low-risk subgroup having an overdose episode. More than 90% of overdose episodes occurred in the high-risk and medium-risk subgroups, although positive predictive values were low, given the rare overdose outcome. Conclusions and Relevance: Machine-learning algorithms appear to perform well for risk prediction and stratification of opioid overdose, especially in identifying low-risk subgroups that have minimal risk of overdose.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The Elements of Statistical Learning
Hastie, T., Tibshirani, R., & Friedman, J. (2009) · Springer Series in Statistics · doi:10.1007/978-0-387-84858-7 · Cited by 25,970
DeLong, E. R., DeLong, D. M., & Clarke-Pearson, D. L. (1988) · Biometrics · doi:10.2307/2531595 · Cited by 22,649
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
John Lu, Z. Q. (2010) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2010.00646_6.x · Cited by 12,390
Comorbidity measures for use with administrative data.
Elixhauser, A., Steiner, C., Harris, D. R., & Coffey, R. M. (1998) · Medical Care · doi:10.1097/00005650-199801000-00004 · Cited by 9,921
Saito, T., & Rehmsmeier, M. (2015) · PLOS ONE · doi:10.1371/journal.pone.0118432 · Cited by 5,309
STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies
Bossuyt, P. M., Reitsma, J. B., Bruns, D. E., Gatsonis, C. A., Glasziou, P. P., Irwig, L., Lijmer, J. G., Moher, D., Rennie, D., de Vet, H. C. W., et al. (2015) · BMJ · doi:10.1136/bmj.h5527 · Cited by 3,660
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. M. (2015) · Annals of Internal Medicine · doi:10.7326/m14-0697 · Cited by 3,267
Estimation of the Youden Index and its Associated Cutoff Point
Fluss, R., Faraggi, D., & Reiser, B. (2005) · Biometrical Journal · doi:10.1002/bimj.200410135 · Cited by 2,608
Increases in drug and opioid-involved overdose deaths – United States, 2010-2015
Rudd, R. A., Seth, P., David, F., & Scholl, L. (2016) · MMWR. Morbidity and Mortality Weekly Report · doi:10.15585/mmwr.mm655051e1 · Cited by 2,194
Association Between Opioid Prescribing Patterns and Opioid Overdose-Related Deaths
Bohnert, A. S. B. (2011) · JAMA · doi:10.1001/jama.2011.370 · Cited by 1,509
Opioid Prescriptions for Chronic Pain and Overdose: A Cohort Study
Dunn, K. M., Saunders, K. W., Rutter, C. M., Banta-Green, C. J., Merrill, J. O., Sullivan, M. D., Weisner, C. M., Silverberg, M. J., Campbell, C. I., Psaty, B. M., et al. (2010) · Annals of Internal Medicine · doi:10.7326/0003-4819-152-2-201001190-00006 · Cited by 1,255
The economic burden of prescription opioid overdose, abuse and dependence in the United States, 2013
Florence, C. S., Zhou, C., Luo, F., & Xu, L. (2016) · Medical Care · doi:10.1097/mlr.0000000000000625 · Cited by 1,152
Webster, L. R., & Webster, R. M. (2005) · Pain Medicine · doi:10.1111/j.1526-4637.2005.00072.x · Cited by 1,096
Overdose deaths involving opioids, cocaine, and psychostimulants—United States, 2015–2016.
Seth, P., Scholl, L., Rudd, R. A., & Bacon, S. (2018) · MMWR. Morbidity and Mortality Weekly Report · doi:10.15585/mmwr.mm6712a1 · Cited by 821
Patterns of abuse among unintentional pharmaceutical overdose fatalities
Hall, A. J. (2008) · JAMA · doi:10.1001/jama.2008.802 · Cited by 815
Characteristics of Opioid Prescriptions in 2009
Volkow, N. D. (2011) · JAMA · doi:10.1001/jama.2011.401 · Cited by 815
Predictors of opioid misuse in patients with chronic pain: a prospective cohort study
Ives, T. J., Chelminski, P. R., Hammett-Stabler, C. A., Malone, R. M., Perhac, J. S., Potisek, N. M., Shilliday, B. B., DeWalt, D. A., & Pignone, M. P. (2006) · BMC Health Services Research · doi:10.1186/1472-6963-6-46 · Cited by 519
Systematic Review of Comorbidity Indices for Administrative Data
Sharabiani, M. T. A., Aylin, P., & Bottle, A. (2012) · Medical Care · doi:10.1097/mlr.0b013e31825f64d0 · Cited by 493
Becker, W. C., Sullivan, L. E., Tetrault, J. M., Desai, R. A., & Fiellin, D. A. (2008) · Drug and Alcohol Dependence · doi:10.1016/j.drugalcdep.2007.09.018 · Cited by 399
Data Mining and Statistics for Decision Making
Tufféry, S. (2011) · Wiley eBooks · doi:10.1002/9780470979174 · Cited by 384
An analysis of the root causes for opioid-related overdose deaths in the United States
Webster, L. R., Cochella, S., Dasgupta, N., Fakata, K. L., Fine, P. G., Fishman, S. M., Grey, T., Johnson, E. M., Lee, L. K., Passik, S. D., et al. (2011) · Pain Medicine · doi:10.1111/j.1526-4637.2011.01134.x · Cited by 295
Sullivan, M. D., Edlund, M. J., Fan, M.-Y., DeVries, A., Braden, J. B., & Martin, B. C. (2010) · Pain · doi:10.1016/j.pain.2010.05.020 · Cited by 292
Edlund, M. J., Martin, B. C., Devries, A., Fan, M.-Y., Brennan Braden, J., & Sullivan, M. D. (2010) · The Clinical Journal of Pain · doi:10.1097/ajp.0b013e3181b99f35 · Cited by 262
Unick, G. J., Rosenblum, D., Mars, S., & Ciccarone, D. (2013) · PLoS ONE · doi:10.1371/journal.pone.0054496 · Cited by 255
High-Risk Use by Patients Prescribed Opioids for Pain and Its Role in Overdose Deaths
Baumblatt, J. A. G., Wiedeman, C., Dunn, J. R., Schaffner, W., Paulozzi, L. J., & Jones, T. F. (2014) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2013.12711 · Cited by 253
Jann, M., Kennedy, W. K., & Lopez, G. (2014) · Journal of Pharmacy Practice · doi:10.1177/0897190013515001 · Cited by 213
Frequency of uterine contractions and the risk of spontaneous preterm delivery
Iams, J. D., Newman, R. B., Thom, E. A., Goldenberg, R. L., Mueller-Heubach, E., Moawad, A., Sibai, B. M., Caritis, S. N., Miodovnik, M., Paul, R. H., et al. (2002) · New England Journal of Medicine · doi:10.1056/nejmoa002868 · Cited by 205
Thottakkara, P., Ozrazgat-Baslanti, T., Hupf, B. B., Rashidi, P., Pardalos, P., Momcilovic, P., & Bihorac, A. (2016) · PLOS ONE · doi:10.1371/journal.pone.0155705 · Cited by 196
Oliva, E. M., Bowe, T., Tavakoli, S., Martins, S., Lewis, E. T., Paik, M., Wiechers, I., Henderson, P., Harvey, M., Avoundjian, T., et al. (2017) · Psychological Services · doi:10.1037/ser0000099 · Cited by 194
Data Mining and Statistics for Decision Making
Tuffry, S. (2011) · Medical Entomology and Zoology · Cited by 187
Analytic models to identify patients at risk for prescription opioid abuse.
White, A. G., Birnbaum, H. G., Schiller, M., Tang, J., & Katz, N. P. (2009) · PubMed · Cited by 171
Opioid utilization and opioid‐related adverse events in nonsurgical patients in US hospitals
Herzig, S. J., Rothberg, M. B., Cheung, M., Ngo, L. H., & Marcantonio, E. R. (2014) · Journal of Hospital Medicine · doi:10.1002/jhm.2102 · Cited by 167
Hsich, E., Gorodeski, E. Z., Blackstone, E. H., Ishwaran, H., & Lauer, M. S. (2011) · Circulation: Cardiovascular Quality and Outcomes · doi:10.1161/circoutcomes.110.939371 · Cited by 167
Doctor and Pharmacy Shopping for Controlled Substances
Peirce, G. L., Smith, M. J., Abate, M. A., & Halverson, J. (2012) · Medical Care · doi:10.1097/mlr.0b013e31824ebd81 · Cited by 155
Why the C-statistic is not informative to evaluate early warning scores and what metrics to use
Romero-Brufau, S., Huddleston, J. M., Escobar, G. J., & Liebow, M. (2015) · Critical Care · doi:10.1186/s13054-015-0999-1 · Cited by 144
A Model to Identify Patients at Risk for Prescription Opioid Abuse, Dependence, and Misuse
Rice, J. B., White, A. G., Birnbaum, H. G., Schiller, M., Brown, D. A., & Roland, C. L. (2012) · Pain Medicine · doi:10.1111/j.1526-4637.2012.01450.x · Cited by 143
Zedler, B., Xie, L., Wang, L., Joyce, A., Vick, C., Brigham, J., Kariburyo, F., Baser, O., & Murrelle, L. (2015) · Pain Medicine · doi:10.1111/pme.12777 · Cited by 138
Yang, Z., Wilsey, B., Bohm, M., Weyrich, M., Roy, K., Ritley, D., Jones, C., & Melnikow, J. (2015) · The Journal of Pain · doi:10.1016/j.jpain.2015.01.475 · Cited by 126
Zedler, B. K., Saunders, W. B., Joyce, A. R., Vick, C. C., & Murrelle, E. L. (2018) · Pain Medicine · doi:10.1093/pm/pnx009 · Cited by 119
Overdose Deaths Involving Prescription Opioids Among Medicaid Enrollees--Washington, 2004-2007
(2009) · PsycEXTRA Dataset · doi:10.1037/e552162010-002 · Cited by 117
Larochelle, M. R., Zhang, F., Ross‐Degnan, D., & Wharam, J. F. (2015) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2015.0914 · Cited by 113
Glanz, J. M., Narwaney, K. J., Mueller, S. R., Gardner, E. M., Calcaterra, S. L., Xu, S., Breslin, K. D., & Binswanger, I. A. (2018) · Journal of General Internal Medicine · doi:10.1007/s11606-017-4288-3 · Cited by 110
(2006) · Biometrical Journal · doi:10.1002/(issn)1521-4036 · Cited by 110
Logan, J. E., Liu, Y., Paulozzi, L. J., Zhang, K., & Jones, C. (2013) · Medical Care · doi:10.1097/mlr.0b013e318293c2c0 · Cited by 104
Prescription Practices involving Opioid Analgesics among Americans with Medicaid, 2010
Mack, K. A., Zhang, K., Paulozzi, L., & Jones, C. (2015) · Journal of Health Care for the Poor and Underserved · doi:10.1353/hpu.2015.0009 · Cited by 100
An Analysis of Heavy Utilizers of Opioids for Chronic Noncancer Pain in the TROUP Study
Edlund, M. J., Martin, B. C., Fan, M.-Y., Braden, J. B., Devries, A., & Sullivan, M. D. (2010) · Journal of Pain and Symptom Management · doi:10.1016/j.jpainsymman.2010.01.012 · Cited by 95
Potential misuse and inappropriate prescription practices involving opioid analgesics.
Liu, Y., Logan, J. E., Paulozzi, L. J., Zhang, K., & Jones, C. M. (2013) · PubMed · Cited by 91
Assessing Opioid Shopping Behaviour
Cepeda, M. S., Fife, D., Chow, W. S., Mastrogiovanni, G., & Henderson, S. C. (2012) · Drug Safety · doi:10.2165/11596600-000000000-00000 · Cited by 87
A decision support system to facilitate management of patients with acute gastrointestinal bleeding
Chu, A., Ahn, H., Halwan, B., Kalmin, B., Artifon, E. L. A., Barkun, A. N., Lagoudakis, M. G., & Kumar, A. (2007) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2007.10.003 · Cited by 86
Automated Prediction of Risk for Problem Opioid Use in a Primary Care Setting
Hylan, T. R., Korff, M. V., Saunders, K. W., Masters, E. T., Palmer, R. E., Carrell, D., Cronkite, D., Mardekian, J., & Gross, D. E. (2015) · Journal of Pain · doi:10.1016/j.jpain.2015.01.011 · Cited by 82
Patterns and Quality of Buprenorphine Opioid Agonist Treatment in a Large Medicaid Program
Gordon, A. J., Lo-Ciganic, W.-H., Cochran, G., Gellad, W. F., Cathers, T., Kelley, D., & Donohue, J. M. (2015) · Journal of Addiction Medicine · doi:10.1097/adm.0000000000000164 · Cited by 76
Self-citationAn Examination of Claims-based Predictors of Overdose from a Large Medicaid Program
Cochran, G., Gordon, A. J., Lo-Ciganic, W.-H., Gellad, W. F., Frazier, W., Lobo, C., Chang, C.-C. H., Zheng, P., & Donohue, J. M. (2017) · Medical Care · doi:10.1097/mlr.0000000000000676 · Cited by 70
Self-citationPatterns of Potential Opioid Misuse and Subsequent Adverse Outcomes in Medicare, 2008 to 2012
Carey, C., Jena, A. B., & Barnett, M. L. (2018) · Annals of Internal Medicine · doi:10.7326/m17-3065 · Cited by 70
Potentially Inappropriate Opioid Prescribing, Overdose, and Mortality in Massachusetts, 2011–2015
Rose, A. J., Bernson, D., Chui, K. K. H., Land, T. G., Walley, A. Y., Larochelle, M. R., Stein, B. D., & Stopka, T. J. (2018) · Journal of General Internal Medicine · doi:10.1007/s11606-018-4532-5 · Cited by 66
Gorodeski, E. Z., Ishwaran, H., Kogalur, U. B., Blackstone, E. H., Hsich, E., Zhang, Z.-M., Vitolins, M. Z., Manson, J. E., Curb, J. D., Martin, L. W., et al. (2011) · Circulation: Cardiovascular Quality and Outcomes · doi:10.1161/circoutcomes.110.959023 · Cited by 59
Amalakuhan, B., Kiljanek, L., Parvathaneni, A., Hester, M., Cheriyath, P., & Fischman, D. (2012) · Journal of Community Hospital Internal Medicine Perspectives · doi:10.3402/jchimp.v2i1.9915 · Cited by 59
Opioid Poisonings in Washington State Medicaid: Trends, Dosing, and Guidelines
Fulton‐Kehoe, D., Sullivan, M. D., Turner, J. A., Garg, R. K., Bauer, A. M., Wickizer, T. M., & Franklin, G. M. (2015) · Medical Care · doi:10.1097/mlr.0000000000000384 · Cited by 54
Chen, G., Kim, S., Taylor, J. M. G., Wang, Z., Lee, O., Ramnath, N., Reddy, R. M., Lin, J., Chang, A. C., Orringer, M. B., et al. (2011) · Journal of Thoracic Oncology · doi:10.1097/jto.0b013e31822918bd · Cited by 33
Predicting mortality over different time horizons: which data elements are needed?
Goldstein, B. A., Pencina, M. J., Montez‐Rath, M. E., & Winkelmayer, W. C. (2016) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocw057 · Cited by 32
Lange, A. V., Lasser, K. E., Xuan, Z., Khalid, L., Beers, D., Heymann, O. D., Shanahan, C. W., Crosson, J., & Liebschutz, J. M. (2015) · Pain · doi:10.1097/01.j.pain.0000460314.73358.ff · Cited by 28
Why 99% may not be as good as you think it is: limitations of screening for rare diseases.
Lutgendorf, M. A., & Stoll, K. A. (2016) · PubMed · doi:10.3109/14767058.2015.1039977 · Cited by 20
Chirikov, V., Shaya, F. T., Onukwugha, E., Mullins, C. D., dosReis, S., & Howell, C. D. (2015) · Medical Care · doi:10.1097/mlr.0000000000000405 · Cited by 19
Cited by
Lynam, A. L., Dennis, J. M., Owen, K. R., Oram, R. A., Jones, A. G., Shields, B. M., & Ferrat, L. A. (2020) · Diagnostic and Prognostic Research · doi:10.1186/s41512-020-00075-2 · Cited by 142
Predictors of attrition in a longitudinal population-based study of aging
Jacobsen, E., Ran, X., Liu, A., Chang, C.-C. H., & Ganguli, M. (2021) · International Psychogeriatrics · doi:10.1017/s1041610220000447 · Cited by 85
Afshar, M., Sharma, B., Dligach, D., Oguss, M., Brown, R., Chhabra, N., Thompson, H. M., Markossian, T., Joyce, C., Churpek, M. M., et al. (2022) · The Lancet Digital Health · doi:10.1016/s2589-7500(22)00041-3 · Cited by 55
“25,26 Our approach incorporated EHR notes and their interpretability as features in our deep learning models with approaches in eXplainable artificial intelligence, and showed that the features for identifying substance misuse were relevant terms and discovery was not limited to just explicit terms…”
Lo-Ciganic, W.-H., Donohue, J. M., Hulsey, E. G., Barnes, S., Li, Y., Kuza, C. C., Yang, Q., Buchanich, J., Huang, J. L., Mair, C., et al. (2021) · PLOS ONE · doi:10.1371/journal.pone.0248360 · Cited by 32
“Non-fatal overdose was defined as an opioid overdose code as the primary diagnosis, or other drug overdose or substance use disorders as the primary diagnosis (S2 Table) with opioid overdose as non-primary diagnosis, as defined previously.14 Fatal opioid overdose was measured using medical examiner’s autopsy data and…”
BackgroundMethodInfluentialSelf-citationDavis, M. A., Lim, N., Jordan, J., Yee, J., Gichoya, J. W., & Lee, R. (2023) · American Journal of Roentgenology · doi:10.2214/ajr.22.28802 · Cited by 19
Metadata
- Authors
- Lo-Ciganic, W.-H., Huang, J. L., Zhang, H. H., Weiss, J. C., Wu, Y., Kwoh, C. K., Donohue, J. M., Cochran, G., Gordon, A. J., Malone, D. C., Kuza, C. C., & Gellad, W. F.
- Venue
- JAMA Network Open
- Publisher
- American Medical Association (AMA)
- Published
- 2019-03-22
- Volume
- 2
- Issue
- 3
- Pages
- e190968
- ISSN
- 2574-3805 (ISSN-L)
- OpenAlex
- W2924338030
- Cited by
- 229
- References
- 62
- FWCI
- 24.54
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.