Journal of General Internal Medicine

, Volume 33, Issue 6, pp 898–905 | Cite as

Prediction of Future Chronic Opioid Use Among Hospitalized Patients

  • S. L. Calcaterra
  • S. Scarbro
  • M. L. Hull
  • A. D. Forber
  • I. A. Binswanger
  • K. L. Colborn
Original Research



Opioids are commonly prescribed in the hospital; yet, little is known about which patients will progress to chronic opioid therapy (COT) following discharge. We defined COT as receipt of ≥ 90-day supply of opioids with < 30-day gap in supply over a 180-day period or receipt of ≥ 10 opioid prescriptions over 1 year. Predictive tools to identify hospitalized patients at risk for future chronic opioid use could have clinical utility to improve pain management strategies and patient education during hospitalization and discharge.


The objective of this study was to identify a parsimonious statistical model for predicting future COT among hospitalized patients not on COT before hospitalization.


Retrospective analysis electronic health record (EHR) data from 2008 to 2014 using logistic regression.


Hospitalized patients at an urban, safety net hospital.

Main Measurements

Independent variables included medical and mental health diagnoses, substance and tobacco use disorder, chronic or acute pain, surgical intervention during hospitalization, past year receipt of opioid or non-opioid analgesics or benzodiazepines, opioid receipt at hospital discharge, milligrams of morphine equivalents prescribed per hospital day, and others.

Key Results

Model prediction performance was estimated using area under the receiver operator curve, accuracy, sensitivity, and specificity. A model with 13 covariates was chosen using stepwise logistic regression on a randomly down-sampled subset of the data. Sensitivity and specificity were optimized using the Youden’s index. This model predicted correctly COT in 79% of the patients and no COT correctly in 78% of the patients.


Our model accessed EHR data to predict 79% of the future COT among hospitalized patients. Application of such a predictive model within the EHR could identify patients at high risk for future chronic opioid use to allow clinicians to provide early patient education about pain management strategies and, when able, to wean opioids prior to discharge while incorporating alternative therapies for pain into discharge planning.


hospital medicine statistical modeling prediction rules 


Funding Information

Funders: The authors would like to acknowledge the University of Colorado, Department of Medicine, Division of General Internal Medicine Small Grants Program for their generous grant which funded this project. Dr. Binswanger was supported by the National Institute On Drug Abuse of the National Institutes of Health under Award Numbers R34DA035952 and R01DA042059. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Prior Presentations: This work was presented at the National Society of General Internal Medicine Conference on April 21, 2017.

Compliance with Ethical Standards

This study was approved by the Colorado Multiple Institutional Review Board and adhered to the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) statement on reporting predictive models.21

Conflict of Interest

The authors declare that they do not have a conflict of interest.

Supplementary material

11606_2018_4335_MOESM1_ESM.docx (15 kb)
Appendix 1 (DOCX 14 kb)
11606_2018_4335_MOESM2_ESM.docx (13 kb)
Appendix 2 (DOCX 13 kb)


  1. 1.
    Center for Behavioral Health Statistics and Quality, Results from the 2015 National Survey on Drug Use and Health 2016, Substance Abuse and Mental Health Services Administration: Rockville, MD. Available from: 12/11/2017. Accessed 12/11/2017.
  2. 2.
    Han B., et al., Prescription opioid use, misuse, and use disorders in US adults: 2015 National Survey on Drug Use and Health. Ann Intern Med, 2017. 167(5): p. 293–301.CrossRefPubMedGoogle Scholar
  3. 3.
    Wasan A.D., J. Wootton, and R. N. Jamison, Dealing with difficult patients in your pain practice. Reg Anesth Pain Med, 2005. 30(2): p. 184–92.CrossRefPubMedGoogle Scholar
  4. 4.
    Edlund M. J., et al., Risk factors for clinically recognized opioid abuse and dependence among veterans using opioids for chronic non-cancer pain. Pain, 2007. 129(3): p. 355–62.CrossRefPubMedGoogle Scholar
  5. 5.
    Sullivan M. D., et al., Association between mental health disorders, problem drug use, and regular prescription opioid use. Arch Intern Med, 2006. 166(19): p. 2087–93.CrossRefPubMedGoogle Scholar
  6. 6.
    Butler S. F., et al., Development and validation of the current opioid misuse measure. Pain, 2007. 130(1–2): p. 144–56.CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Akbik H., et al., Validation and clinical application of the Screener and Opioid Assessment for Patients with Pain (SOAPP). J Pain Symptom Manage, 2006. 32(3): p. 287–93.CrossRefPubMedGoogle Scholar
  8. 8.
    Webster L.R. and R.M. Webster, Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the Opioid Risk Tool. Pain Med, 2005. 6(6): p. 432–42.CrossRefPubMedGoogle Scholar
  9. 9.
    Compton P., J. Darakjian, and K. Miotto, Screening for addiction in patients with chronic pain and “problematic” substance use: evaluation of a pilot assessment tool. J Pain Symptom Manag, 1998. 16(6): p. 355–63.CrossRefGoogle Scholar
  10. 10.
    Herzig S.J., et al., Opioid utilization and opioid-related adverse events in nonsurgical patients in US hospitals. J Hosp Med, 2014. 9(2): p. 73–81.CrossRefPubMedGoogle Scholar
  11. 11.
    Calcaterra S.L., et al., Opioid prescribing at hospital discharge contributes to chronic opioid use. J Gen Intern Med, 2016. 31(5): p. 478–85.CrossRefPubMedGoogle Scholar
  12. 12.
    Sun E.C., et al., Incidence of and risk factors for chronic opioid use among opioid-naive patients in the postoperative period. JAMA Intern Med, 2016. 176(9): p. 1286–93.CrossRefPubMedGoogle Scholar
  13. 13.
    Bateman B.T., et al., Persistent opioid use following cesarean delivery: patterns and predictors among opioid-naive women. Am J Obstet Gynecol, 2016. 215(3): p. 353.e1–353.e18.CrossRefGoogle Scholar
  14. 14.
    Thielke S.M., et al., A prospective study of predictors of long-term opioid use among patients with chronic noncancer pain. Clin J Pain, 2017. 33(3): p. 198–204.PubMedGoogle Scholar
  15. 15.
    Iwashyna T.J. and V. Liu, What’s so different about big data?. A primer for clinicians trained to think epidemiologically. Ann Am Thorac Soc, 2014. 11(7): p. 1130–5.CrossRefPubMedPubMedCentralGoogle Scholar
  16. 16.
    Chekroud A. M., et al., Cross-trial prediction of treatment outcome in depression: a machine learning approach. Lancet Psychiatry, 2016. 3(3): p. 243–50.CrossRefPubMedGoogle Scholar
  17. 17.
    Churpek M.M., et al., Multicenter comparison of machine learning methods and conventional regression for predicting clinical deterioration on the wards. Crit Care Med, 2016. 44(2): p. 368–74.CrossRefPubMedPubMedCentralGoogle Scholar
  18. 18.
    Taylor R.A., et al., Prediction of in-hospital mortality in emergency department patients with sepsis: a local big data-driven, machine learning approach. Acad Emerg Med, 2016. 23(3): p. 269–78.CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Gabow P., S. Eisert, and R. Wright, Denver Health: a model for the integration of a public hospital and community health centers. Ann Intern Med, 2003. 138(2): p. 143–9.CrossRefPubMedGoogle Scholar
  20. 20.
    Gabow P. A. and P. S. Mehler, A broad and structured approach to improving patient safety and quality: lessons from Denver Health. Health Aff (Millwood), 2011. 30(4): p. 612–8.CrossRefGoogle Scholar
  21. 21.
    Collins G. S., et al., Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (tripod): the tripod statement. Ann Intern Med, 2015. 162(1): p. 55–63.CrossRefPubMedGoogle Scholar
  22. 22.
    Paice J. A., et al., Management of chronic pain in survivors of adult cancers: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol, 2016. 34(27): p. 3325–3345.CrossRefPubMedGoogle Scholar
  23. 23.
    Chow R., et al., Needs assessment of primary care physicians in the management of chronic pain in cancer survivors. Support Care Cancer, 2017. 25(11): p. 3505–3514.CrossRefPubMedGoogle Scholar
  24. 24.
    Vanderlip E. R., et al., National study of discontinuation of long-term opioid therapy among veterans. Pain, 2014. 155(12): p. 2673–2679.CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Von Korff M., et al., De facto long-term opioid therapy for noncancer pain. Clin J Pain, 2008. 24(6): p. 521–7.CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    Amit Y. and D. Geman, Shape quantization and recognition with randomized trees. Neural Comput, 1997. 9(7): p. 1545–1588.CrossRefGoogle Scholar
  27. 27.
    CDC, Classification of Diseases, Functioning, and Disability. 2013. Available from: Accessed 12/11/2017.
  28. 28.
    Charlson M. E., et al., A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis, 1987. 40(5): p. 373–83.CrossRefPubMedGoogle Scholar
  29. 29.
    Colorado, S.O., Colorado Amendment 20. Legalization of Medicinal Marijuana. 2000. Available from: Accessed 12/11/2017.
  30. 30.
    Cleveland W. S. and S. J. Devlin, Locally weighted regression: an approach to regression analysis by local fitting. J Am Stat Assoc, 1988. 83(403): p. 596–610.CrossRefGoogle Scholar
  31. 31.
    Von Korff M., et al., Defacto long-term opioid therapy for non-cancer pain. Clin J Pain, 2008. 24(6): p. 521–527.CrossRefPubMedPubMedCentralGoogle Scholar
  32. 32.
    Herzig S. J., et al., Opioid utilization and opioid-related adverse events in non-surgical patients in US hospitals. J Hosp Med, 2014. 9(2): p. 73–81.CrossRefPubMedGoogle Scholar
  33. 33.
    Lin W.-J. and J. J. Chen, Class-imbalanced classifiers for high-dimensional data. Brief Bioinform, 2013. 14(1): p. 13–26.CrossRefPubMedGoogle Scholar
  34. 34.
    Ruopp M. D., et al., Youden Index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom J, 2008. 50(3): p. 419–30.CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Foundation, T.R. What is R? [cited 2017 11/3/2017]; Available from: Accessed 12/11/2017.
  36. 36.
    Butler S. F., et al., Validation of the revised Screener and Opioid Assessment for Patients with Pain (SOAPP-R). J Pain, 2008. 9(4): p. 360–72.CrossRefPubMedPubMedCentralGoogle Scholar
  37. 37.
    Raebel M. A., et al., Chronic opioid use emerging after bariatric surgery. Pharmacoepidemiol Drug Saf, 2014. 23(12): p. 1247–57.CrossRefPubMedGoogle Scholar
  38. 38.
    Inacio M. C., et al., Risk factors for persistent and new chronic opioid use in patients undergoing total hip arthroplasty: a retrospective cohort study. BMJ Open, 2016. 6(4): p. e010664.Google Scholar
  39. 39.
    Gudin J. A., et al., Risks, management, and monitoring of combination opioid, benzodiazepines, and/or alcohol use. Postgrad Med, 2013. 125(4): p. 115–130.CrossRefPubMedPubMedCentralGoogle Scholar
  40. 40.
    Guccione A. A., et al., The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health, 1994. 84(3): p. 351–8.CrossRefPubMedPubMedCentralGoogle Scholar
  41. 41.
    Schmaling K. B. and K. L. Betterton, Neurocognitive complaints and functional status among patients with chronic fatigue syndrome and fibromyalgia. Qual Life Res, 2016. 25(5): p. 1257–63.CrossRefPubMedGoogle Scholar
  42. 42.
    Murray C. J., et al., The state of US health, 1990-2010: burden of diseases, injuries, and risk factors. JAMA, 2013. 310(6): p. 591–608.CrossRefPubMedGoogle Scholar
  43. 43.
    Casanova R., et al., Prediction of incident diabetes in the Jackson Heart Study using high-dimensional machine learning. PLoS One, 2016. 11(10): p. e0163942.Google Scholar
  44. 44.
    Pearce C. B., et al., Machine learning can improve prediction of severity in acute pancreatitis using admission values of APACHE II score and C-reactive protein. Pancreatology, 2006. 6(1–2): p. 123–31.CrossRefPubMedGoogle Scholar
  45. 45.
    Shameer K., et al., Predictive modeling of hospital readmission rates using electronic medical record-wide machine learning: a case-study using Mount Sinai Heart Failure Cohort. Pac Symp Biocomput, 2016. 22: p. 276–287.PubMedCentralGoogle Scholar
  46. 46.
    Alsolamy S., et al., Diagnostic accuracy of a screening electronic alert tool for severe sepsis and septic shock in the emergency department. BMC Med Inform Decis Mak, 2014. 14: p. 105.CrossRefPubMedPubMedCentralGoogle Scholar
  47. 47.
    Kucher N., et al., Electronic alerts to prevent venous thromboembolism among hospitalized patients. N Engl J Med, 2005. 352(10): p. 969–977.CrossRefPubMedGoogle Scholar
  48. 48.
    Ng C. K., et al., Clinical and economic impact of an antibiotics stewardship programme in a regional hospital in Hong Kong. Qual Saf Health Care, 2008. 17(5): p. 387–92.CrossRefPubMedGoogle Scholar
  49. 49.
    McNeil V., M. Cruickshank, and M. Duguid, Safer use of antimicrobials in hospitals: the value of antimicrobial usage data. Med J Aust, 2010. 193(8 Suppl): p. S114–7.PubMedGoogle Scholar
  50. 50.
    MacDougall C. and R. E. Polk, Antimicrobial stewardship programs in health care systems. Clin Microbiol Rev, 2005. 18(4): p. 638–56.CrossRefPubMedPubMedCentralGoogle Scholar
  51. 51.
    Fishman N., Antimicrobial stewardship. Am J Med, 2006. 119(6 Suppl 1): p. S53–61; discussion S62-70.CrossRefPubMedGoogle Scholar

Copyright information

© Society of General Internal Medicine 2018

Authors and Affiliations

  • S. L. Calcaterra
    • 1
    • 2
  • S. Scarbro
    • 3
    • 4
  • M. L. Hull
    • 1
  • A. D. Forber
    • 5
  • I. A. Binswanger
    • 2
    • 6
  • K. L. Colborn
    • 5
  1. 1.Hospital MedicineDenver Health Medical CenterDenverUSA
  2. 2.Division of General Internal Medicine, Department of MedicineUniversity of Colorado School of MedicineAuroraUSA
  3. 3.University of Colorado Adult and Child Consortium for Health Outcomes Research and Delivery ScienceAuroraUSA
  4. 4.Rocky Mountain Prevention Research Center, Colorado School of Public Health, University of Colorado DenverAuroraUSA
  5. 5.Department of Biostatistics and InformaticsUniversity of Colorado Anschutz Medical CampusAuroraUSA
  6. 6.Institute for Health Research, Kaiser Permanente ColoradoDenverUSA

Personalised recommendations