Advertisement

A Tool to Assess Risk of De Novo Opioid Abuse or Dependence

      Abstract

      Background

      Determining risk factors for opioid abuse or dependence will help clinicians practice informed prescribing and may help mitigate opioid abuse or dependence. The purpose of this study is to identify variables predicting opioid abuse or dependence.

      Methods

      A retrospective cohort study using de-identified integrated pharmacy and medical claims was performed between October 2009 and September 2013. Patients with at least 1 opioid prescription claim during the index period (index claim) were identified. We ascertained risk factors using data from 12 months before the index claim (pre-period) and captured abuse or dependency diagnosis using data from 12 months after the index claim (postperiod). We included continuously eligible (pre- and postperiod) commercially insured patients aged 18 years or older. We excluded patients with cancer, residence in a long-term care facility, or a previous diagnosis of opioid abuse or dependence (identified by International Classification of Diseases 9th revision code or buprenorphine/naloxone claim in the pre-period). The outcome was a diagnosis of opioid abuse (International Classification of Diseases 9th revision code 304.0x) or dependence (305.5).

      Results

      The final sample consisted of 694,851 patients. Opioid abuse or dependence was observed in 2067 patients (0.3%). Several factors predicted opioid abuse or dependence: younger age (per decade [older] odds ratio [OR], 0.68); being a chronic opioid user (OR, 4.39); history of mental illness (OR, 3.45); nonopioid substance abuse (OR, 2.82); alcohol abuse (OR, 2.37); high morphine equivalent dose per day user (OR, 1.98); tobacco use (OR, 1.80); obtaining opioids from multiple prescribers (OR, 1.71); residing in the South (OR, 1.65), West (OR, 1.49), or Midwest (OR, 1.24); using multiple pharmacies (OR, 1.59); male gender (OR, 1.43); and increased 30-day adjusted opioid prescriptions (OR, 1.05).

      Conclusions

      Readily available demographic, clinical, behavioral, pharmacy, and geographic information can be used to predict the likelihood of opioid abuse or dependence.

      Keywords

      Clinical Significance
      • Readily available variables can help quantify the risk of developing opioid abuse.
      • Chronic opioid use and history of mental illness are the strongest predictors of abuse.
      The United States has seen a dramatic increase in opioid prescriptions in the past decade with a concomitant increase in abuse of opioid medications.
      • Dart R.C.
      • Surrat H.L.
      • Cicero T.J.
      • et al.
      Trends in opioid analgesic abuse and mortality in the United States.
      There has been a tripling in the rate of opioid-related overdose deaths from 2000 to 2014, with more than 28,000 deaths in 2014.
      • Rudd R.
      • Aleshire N.
      • Zibbell J.E.
      • Gladden M.
      Increases in drug and opioid overdose deaths — United States, 2000-2014.
      This epidemic creates a dilemma for prescribers who seek to provide adequate pain relief while minimizing risks of abuse and dependence. Abuse is defined as the intentional self-administration of a medication for a nonmedical reason,
      • Katz N.P.
      • Adams E.H.
      • Chilcoat H.
      • et al.
      Challenges in the development of prescription opioid abuse-deterrent formulations.
      whereas dependence is a maladaptive pattern of substance use.
      American Psychiatric Association
      Diagnostic and Statistical Manual of Mental Disorders.
      • Savage S.R.
      • Joranson D.E.
      • Covington E.C.
      • Schnoll S.H.
      • Heit H.A.
      • Gilson A.M.
      Definitions related to the medical use of opioids: evolution towards universal agreement.
      Guidelines exist for using opioids in noncancer pain,
      • Chou R.
      • Fanciullo G.J.
      • Fine P.G.
      • et al.
      Clinical guidelines for the use of chronic opioid therapy in chronic noncancer pain.
      but prescribers face challenging situations when prescribing opioids and need tools to aid their decisions. Prescription drug monitoring programs can help reveal aberrant behavior. Forty-nine states have enacted these programs; however, monitoring alone does not prevent abuse.
      • Paulozzi L.J.
      • Jones C.M.
      • Mack K.A.
      • Rudd R.A.
      Vital signs: overdoses of prescription opioid pain relievers United States, 1999-2008.
      • Katz N.
      • Panas L.
      • Kim M.
      • et al.
      Usefulness of prescription monitoring programs for surveillance–analysis of Schedule II opioid prescription data in Massachusetts, 1996-2006.
      • Wang J.
      • Christo P.J.
      The influence of prescription monitoring programs on chronic pain management.
      Centers for Medicare and Medicaid Services
      The role of a prescription drug monitoring program in reducing prescription drug diversion, abuse, and misuse.
      Currently, there are limited tools that help predict which patients may develop opioid abuse or dependence. The Opioid Risk Tool identifies at-risk patients on the basis of medical, family, and social history.
      • Webster L.R.
      • Webster R.M.
      Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the opioid risk tool.
      However, the Opioid Risk Tool does not combine patient and prescription drug monitoring program information to assess risk. Clinicians need to know how risk factors ascertained at the time of prescribing opioids predict subsequent abuse or dependence.
      The objective of this study is to identify demographic characteristics, clinical and behavioral factors obtained from prescription drug monitoring programs, and pharmacy and geographic factors that quantify the risk of developing opioid abuse or dependence. These factors are immediately available to a prescriber by patient interview and by accessing a prescription drug monitoring program and could help assess the risk of prescribing opioids. Once at-risk patients are identified, additional screening tests could be used by the prescriber
      • McNeely J.
      • Strauss S.M.
      • Saitz R.
      • et al.
      A brief patient self-administered substance use screening tool for primary care: two-site validation study of the Substance Use Brief Screen (SUBS).
      • Bowman S.
      • Eiserman J.
      • Beletsky L.
      • Stancliff S.
      • Bruce R.D.
      Reducing the health consequences of opioid addiction in primary care.
      and treatment of abuse and dependence could be pursued.

      Materials and Methods

      We used de-identified (in accordance with Health Insurance Portability and Accountability Act requirements) pharmacy and medical claims data from a pharmacy benefit manager (Express Scripts) from October 1, 2009, to September 30, 2013. These data include health insurance claims (inpatient/outpatient medical and outpatient pharmacy) and enrollment data from large employers and health plans across the United States. This study included patients aged 18 years or older as of the index opioid claim date.
      International Classification of Diseases, Ninth Revision (ICD–9) codes were used to identify medical diagnoses. First Data Bank “Smart Key” classifications were used to identify opioids on the basis of pharmacy claims.

      First Databank. Available at: http://www.fdbhealth.com/. Accessed April 24, 2015.

      Smart Key Specific Therapeutic Class designations (4-digit codes describing therapeutic drug classes) and Generic Code Numbers (5-digit numbers that group equivalent products based on active ingredients) were used to classify pharmacy claims (Appendix 1, available online). Dosage strengths for Specific Therapeutic Class were used in calculating daily morphine equivalent dosing and to classify immediate- vs extended-release opioids.
      Exclusion criteria included patients with a cancer diagnosis (Appendix 2, available online), with claims for chemotherapy or antiemetics (Appendix 3, available online), in residence in long-term care facilities (residence code of 03 from the National Council of Prescription Drug Programs 384-4x classification), in convalescence after chemotherapy (ICD-9 V66.2), or in hospice/palliative/end-of-life care (ICD-9 V66.7). Patients with a prior opioid dependency diagnosis (within 365 days before the index claim) or who were taking buprenorphine/naloxone (typically used to treat opioid dependence) also were excluded (Appendix 4, available online).
      To predict the likelihood of opioid abuse or dependency, we conducted a retrospective claims analysis. Derivation and validation models were developed. For the derivation model (Figure), we identified patients on the basis of 1 or more claims for opioids in the index period (October 1, 2011, to September 30, 2012); the index claim was a randomly selected opioid claim. For the validation model, we identified patients on the basis of 1 or more claims for opioids in the index period (October 1, 2010, to September 30, 2011), again randomly selected. For both models, we ascertained risk factors using data from 12 months before the index claim (pre-period) and captured abuse or dependency diagnosis by ICD-9 code using data from 12 months postindex claim (postperiod). All patients were continuously eligible during pre- and postperiods.
      Figure thumbnail gr1
      FigureStudy timeline for the derivation model.
      The primary outcome measure was an ICD-9 diagnosis of nondependent opioid abuse (304.0x) or dependence (305.5x) in the postperiod. Patient characteristics based on pharmacy and medical claims were included as independent variables, including demographic, clinical, behavioral, pharmacy claims, and geographic factors. All factors were measured before the index date.
      Variables included age
      • Cepeda M.S.
      • Fife D.
      • Chow W.
      • Mastrogiovanni G.
      • Henderson S.C.
      Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
      (calculated at the index claim) and the chronic use of opioids
      • Von Korff M.
      • Saunders K.
      • Thomas Ray G.
      • et al.
      De facto long-term opioid therapy for noncancer pain.
      (defined as claims for >90 days of opioids in the 6 months before and including the index date). Clinical variables of history of mental illness,
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      • Skala K.
      • Reichl L.
      • Ilias W.
      • et al.
      Can we predict addiction to opioid analgesics? A possible tool to estimate the risk of opioid addiction in patients with pain.
      nonopioid substance abuse,
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      and nondependent alcohol abuse
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      • Skala K.
      • Reichl L.
      • Ilias W.
      • et al.
      Can we predict addiction to opioid analgesics? A possible tool to estimate the risk of opioid addiction in patients with pain.
      • McCabe S.E.
      • Cranford J.A.
      • Boyd C.J.
      The relationship between past-year drinking behaviors and nonmedical use of prescription drugs: prevalence of co-occurrence in a national sample.
      were identified by ICD-9 codes (Appendix 5, available online). We identified high morphine equivalent dose users (≥120 mg morphine equivalent dosing daily)
      • Manchikanti L.
      • Helm II, S.
      • Fellows B.
      • et al.
      Opioid epidemic in the United States.
      • Manchikanti L.
      • Abdi S.
      • Atluri S.
      • et al.
      American Society of Interventional Pain Physicians (ASIPP) guidelines for responsible opioid prescribing in chronic non-cancer pain: Part 2 –guidance.
      • Sehgal N.
      • Manchikanti L.
      • Smith H.S.
      Prescription opioid abuse in chronic pain: a review of opioid abuse predictors and strategies to curb opioid abuse.
      by using pharmacy claims. The other clinical variable, tobacco use disorder,
      • Skala K.
      • Reichl L.
      • Ilias W.
      • et al.
      Can we predict addiction to opioid analgesics? A possible tool to estimate the risk of opioid addiction in patients with pain.
      • Cheatle M.D.
      • O'Brien C.P.
      • Mathai K.
      • Hansen M.
      • Grasso M.
      • Yi P.
      Aberrant behaviors in a primary care-based cohort of patients with chronic pain identified as misusing prescription opioids.
      also was identified by ICD-9 code.
      Prescriber shopping was hypothesized to be a risk factor,
      • Cepeda M.S.
      • Fife D.
      • Chow W.
      • Mastrogiovanni G.
      • Henderson S.C.
      Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
      • Parente S.T.
      • Kim S.S.
      • Finch M.D.
      • et al.
      Identifying controlled substance patterns of utilization requiring evaluation using administrative claims data.
      • Hall A.J.
      • Logan J.E.
      • Toblin R.L.
      • et al.
      Patterns of abuse among unintentional pharmaceutical overdose fatalities.
      and patients were identified as prescriber shoppers if they received opioid prescriptions from ≥2 prescribers
      • Cepeda M.S.
      • Fife D.
      • Chow W.
      • Mastrogiovanni G.
      • Henderson S.C.
      Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
      within 60 days before and inclusive of the index date. Geographic region is associated with opioid abuse or dependence.
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      Patients were classified into geographic regions (Northeast, South, West, or Midwest as defined by US Census Bureau

      US Census Bureau. Geographic terms and concepts: census divisions and census regions. Available at: https://www.census.gov/geo/reference/gtc/gtc_census_divreg.html. Accessed February 9, 2015.

      ) according to their index claim state of residence.
      Pharmacy shopping also was considered as a risk factor,
      • Cepeda M.S.
      • Fife D.
      • Chow W.
      • Mastrogiovanni G.
      • Henderson S.C.
      Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
      • Parente S.T.
      • Kim S.S.
      • Finch M.D.
      • et al.
      Identifying controlled substance patterns of utilization requiring evaluation using administrative claims data.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Tang J.
      • Katz M.P.
      Analytic models to identify patients at risk for prescription opioid abuse.
      and patients were considered pharmacy shoppers if they filled opioid prescriptions at 3 or more pharmacies
      • Cepeda M.S.
      • Fife D.
      • Chow W.
      • Mastrogiovanni G.
      • Henderson S.C.
      Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
      • Parente S.T.
      • Kim S.S.
      • Finch M.D.
      • et al.
      Identifying controlled substance patterns of utilization requiring evaluation using administrative claims data.
      within 60 days before and inclusive of the index date. Prior research indicates that men are more likely to be opioid abusers than women.
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      • Roland C.L.
      • Joshi A.V.
      • Mardekian J.
      • Walden S.C.
      • Harnett J.
      Prevalence and cost of diagnosed opioid abuse in a privately insured population in the United States.
      We hypothesized that the same relationship would be observed for opioid abuse or dependence.
      Pharmacy claims were used to determine the number of opioid prescriptions in the pre-period.
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      To capture prior use of opioids, the number of 30-day adjusted opioid prescriptions in the pre-period was used. Day's supply of all opioid prescriptions was divided by 30.4 days/month to convert to months and address the day's supply differential between dispensing channels. Pharmacy claims also identified long-term use of immediate-release opioids. For patients who were taking opioids for at least 6 months of the pre-period, we computed a ratio of immediate release to total opioids being taken. Patients were considered chronic immediate-release users if this ratio exceeded 0.5.
      We computed a binary distance variable of less than and greater than 50 miles from patient to index opioid prescriber based on centroids of the respective ZIP codes. We hypothesized that potential opioid abusers or dependents would travel farther to receive an opioid prescription.
      From a population of approximately 1.4 million patients with at least 1 opioid claim during the index period for the derivation model, 694,851 patients constituted the final analytic sample (Table 1). Datasets were created and statistical analyses were conducted using SAS version 9.3 and SAS Enterprise Guide version 5.1 (SAS Institute Inc, Cary, NC). Descriptive statistics included comparison of bivariate differences in risk factors between opioid abusers or dependents and nonabusers or nondependents using analysis of variance for continuous variables and chi-square tests for categoric variables. All comparisons were 2-tailed. Variance inflation factor was analyzed to ascertain multicollinearity among independent variables.
      Table 1Sample Selection Methodology and Description of Sample Size
      Study Selection CriteriaDerivation ModelValidation Model
      NN
      Total patients with opioid prescription claims during index period
      Derivation model: index period Q4 2011 to Q3 2012; validation model: index period Q4 2010 to Q3 2011.
      1,428,1371,453,996
      No cancer diagnosis or medication1,348,7931,376,236
      Not in long-term care facilities1,345,9081,376,210
      Not in hospice care facilities1,345,7201,376,052
      No diagnosis for prior drug dependency1,339,4181,370,631
      Continuously eligible during pre- and postperiod751,937689,519
      Aged ≥18 y as of the index date696,922636,620
      No missing values for key covariates694,851634,588
      Derivation model: index period Q4 2011 to Q3 2012; validation model: index period Q4 2010 to Q3 2011.
      Multivariate logistic regression analyses were performed to predict the likelihood of opioid abuse or dependence. To address potential bias in the estimated coefficient and to test the robustness of the findings, 2 sensitivity analyses were conducted (Appendix 6, available online).
      Validation was conducted to assess performance of the predictive model in an independent sample. The validation model design was identical to the derivation model with the exception of index period. The index period for the validation model was 1 year earlier, from October 1, 2010, to September 30, 2011, which resulted in a cohort of 634,588 patients.
      This research was exempt from institutional review board approval on the basis of the Code of Federal Regulation, §46.101b, from the US Department of Health and Human Resources,

      Basic Health and Human Services policy for protection of human research subject. 45 C.F.R. § 46 (2009).

      and exempted from the Washington University Institutional Review Board.

      Results

      The derivation cohort included 694,851 patients, of whom 2067 (0.3%) were opioid abusers/dependents. They were significantly younger (Table 2). There were more chronic opioid users (55.8% vs 10.4%) in the group that developed abuse or dependence.
      Table 2Baseline Characteristics
      All data were significantly different at P <.05 between opioid dependents and nondependents, except for cells marked with a note indicating otherwise.
      MeasureDerivation ModelValidation Model
      DependentNondependentDependentNondependent
      No.2067692,7841580633,008
      Age, mean (SD)44.1 (15.5)48.8 (15.6)43.6 (15.1)49.1 (15.6)
      Chronic users, N (%)1154 (55.8)72,072 (10.4)869 (55.0)62,597 (9.9)
      Mental illness, N (%)1076 (52.1)103,398 (14.9)781 (49.4)86,546 (13.7)
      Nonopioid substance abuse, N (%)84 (4.1)1696 (0.2)59 (3.7)1280 (0.2)
      Nondependent alcohol abuse, N (%)82 (4.0)3191 (0.5)55 (3.5)2623 (0.4)
      Daily MED ≥120 mg/d, N (%)398 (19.3)13,075 (1.9)301 (19.1)15,663 (2.5)
      Tobacco use disorder, N (%)401 (19.4)30,584 (4.4)290 (18.4)23,663 (3.7)
      Prescriber shoppers, N (%)735 (35.6)80,354 (11.6)575 (36.4)72,373 (11.4)
      Region, N (%)
       Northeast386 (18.7)218,291 (31.5)313 (19.8)206,560 (32.6)
       South772 (37.4)198,229 (28.6)489 (31.0)172,457 (27.2)
       West409 (19.8)111,168 (16.1)368 (23.3)104,301 (16.5)
       Midwest500 (24.2)165,096 (23.8)410 (26.0)149,690 (23.7)
      Pharmacy shoppers, N (%)141 (6.8)3855 (0.6)125 (7.9)3406 (0.5)
      Percent male, N (%)994 (48.1)298,126 (43.0)780 (49.4)271,038 (42.8)
      Prior opioid 30-d adjusted prescriptions, mean (SD)9.3 (9.3)1.8 (4.1)9.1 (9.2)1.7 (4.0)
      Chronic immediate-release users, N (%)665 (32.2)46,839 (6.8)536 (33.9)40,069 (6.3)
      Distance from patient to prescriber, N (%)
       ≤50 miles1747 (84.5)600,035 (86.6)1343 (85.0)
      Not significantly different between opioid dependents and nondependents at P <.05.
      539,518 (85.2)
      Not significantly different between opioid dependents and nondependents at P <.05.
       >50 miles320 (15.5)92,749 (13.4)237 (15.0)
      Not significantly different between opioid dependents and nondependents at P <.05.
      93,490 (14.8)
      Not significantly different between opioid dependents and nondependents at P <.05.
      MED = morphine equivalent dose; SD = standard deviation.
      All data were significantly different at P <.05 between opioid dependents and nondependents, except for cells marked with a note indicating otherwise.
      Not significantly different between opioid dependents and nondependents at P <.05.
      Clinical factors significantly varied between the 2 groups of patients. Opioid abusers/dependents had a higher proportion of mental illness (52.1% vs 14.9%) and nonopioid substance abuse (4.1% vs 0.2%), and nondependent alcohol abuse (4.0% vs 0.5%) compared with nonabusers/nondependents. Furthermore, opioid abuse/dependence was associated with high morphine equivalent dose users (19.3% vs 1.9%) and tobacco use disorder (19.4% vs 4.4%).
      Opioid abuser/dependents were more likely to be prescriber shoppers (35.6% vs 11.6%). There were also significant regional differences among the 2 groups, with the South, West, and Midwest having a higher percentage of abusers or dependents compared with the Northeast.
      Pharmacy shopping differed between the 2 groups. There was a higher percentage of pharmacy shoppers (6.8% vs 0.6%) in the opioid abuse/dependence group than in the nonopioid abusers/dependence group. There was a higher proportion of men in the abuse/dependence group than in the nonabuser/nondependent group. Patients who developed abuse or dependency averaged higher numbers of 30-day adjusted opioid prescriptions in the pre-period (9.3 vs 1.8) and more chronic immediate-release users (32.2% vs 6.8%).
      The derivation and validation data set found similar effects for all variables, except that a long (>50 miles) distance between patient and index opioid prescriber was significantly more common among opioid abusers/dependents in the derivation dataset but not in the validation data.
      As indicated by a variance inflation factor of less than 10 for all variables, independent variables in the model did not have a high level of collinearity. Thus, all variables were retained in the model. The c-statistic was 0.852 for the derivation model and 0.847 for the validation model, indicating that the 2 models (Table 3) successfully discriminate between opioid abusers/dependents and nonabusers or nondependent patients.
      Table 3Multivariate Adjusted Odds Ratio for Opioid Dependency Models
      ReferenceDerivation Model
      The c-statistics were 0.852 for the derivation model and 0.847 for the validation model.
      Validation Model
      The c-statistics were 0.852 for the derivation model and 0.847 for the validation model.
      OR
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      95% CIOR
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      95% CI
      Age (per decade of life)NA0.680.65-0.700.650.63-0.68
      Chronic usersAbsent4.393.71-5.194.293.53-5.22
      Mental illnessAbsent3.453.13-3.793.373.02-3.76
      Nonopioid substance abuseAbsent2.822.18-3.642.872.11-3.89
      Nondependent alcohol abuseAbsent2.371.84-3.052.101.55-2.85
      Daily MED ≥120 mg/dAbsent1.981.68-2.341.931.61-2.32
      Tobacco use disorderAbsent1.801.60-2.042.091.81-2.40
      Prescriber shoppersAbsent1.711.55-1.891.741.55-1.95
      South regionNortheast1.651.45-1.871.441.25-1.67
      West regionNortheast1.491.29-1.721.701.46-1.99
      Midwest regionNortheast1.241.08-1.421.311.13-1.53
      Pharmacy shoppersAbsent1.591.31-1.921.981.61-2.43
      MaleFemale1.431.31-1.571.521.37-1.68
      Prior opioid 30-d adjusted prescriptionsNA1.051.04-1.061.041.03-1.05
      Chronic immediate-release userAbsent1.07
      Not significant at P <.05.
      0.93-1.221.271.09-1.48
      Distance from patient to prescriber≤50 miles1.12
      Not significant at P <.05.
      0.99-1.270.95
      Not significant at P <.05.
      0.83-1.10
      CI = confidence interval; MED = morphine equivalent dose; NA = not applicable (for continuous variables); OR = odds ratio.
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      Not significant at P <.05.
      The c-statistics were 0.852 for the derivation model and 0.847 for the validation model.
      Younger age (odds ratio [OR], 0.68 per decade older; 95% confidence interval [CI], 0.65-0.70) significantly predicated opioid abuse or dependence. Chronic use of opioids (OR, 4.39; 95% CI, 3.71-5.19) and history of mental illness (OR, 3.45; 95% CI, 3.13-3.79) were strong predictors of developing opioid abuse/dependence. Histories of other substance abuse (OR, 2.82; 95%, CI, 2.18-3.64) and alcohol abuse (OR, 2.37; 95% CI, 1.84-3.05), and doses of opioids ≥120 mg or morphine equivalents per day (OR, 1.98; 95% CI, 1.68-2.34) elevated the risk of developing opioid abuse or dependence. Tobacco use (OR, 1.80; 95% CI, 1.60-2.04); prescriber shoppers (OR, 1.71; 95% CI, 1.55-1.89); and residing in the South (OR, 1.65; 95% CI, 1.45-1.87), West (OR, 1.49; 95% CI, 1.29-1.72), and Midwest (OR, 1.24; 95% CI, 1.08-1.42) compared with the Northeast also were significant predictors of developing opioid abuse/dependence.
      Finally, pharmacy shoppers (OR, 1.59; 95% CI, 1.31-1.92), male gender (OR, 1.43; 95% CI, 1.31-1.57), and each additional 30-day adjusted opioid prescription (OR, 1.05; 95% CI, 1.04-1.06) were predictive of developing opioid abuse/dependence.
      Distance from patient to prescriber was not statistically significant in either model. Being a chronic immediate-release user was insignificant in the derivation model.
      With only 1 exception (chronic immediate-release opioid), the predictors of opioid abuse or dependence that were significant in the derivation model also were significant in the validation model. Additional sensitivity analyses (Appendix 6, available online) corroborated the associations in the derivation and validation models.

      Discussion

      This study identified 12 patient characteristics that predict increased risk of de novo abuse or dependence in opioid users. The strongest predictors were chronic use, mental illness, nonopioid substance use, alcohol abuse, high morphine equivalent dose per day, younger age, and male gender. These effects were in the direction as hypothesized. In this study, the relationships between the distance from patient to prescriber and being a chronic immediate-release user to the odds of developing opioid abuse or dependence were not consistently significant. All identified risk factors are available through patient history or a prescription drug monitoring program. Thus, our study provides useful risk factors for prescribers to be able to determine a patient's risk of developing opioid abuse or dependence in the next 12 months. These factors can help prescribers weigh the risks and benefits of prescribing opioids.
      Our findings are consistent with prior research. Dufour et al
      • Dufour R.
      • Markekian J.
      • Pasquale M.K.
      • Schaaf D.
      • Andrews G.
      • Patel N.
      Understanding predictors of opioid abuse: predictive model development and validation.
      developed a predictive model using data from 1 commercial insurer with 3500 cases of opioid abuse or dependence. They also found a lower risk with advanced age and high risks among men.
      Our results also are consistent with those of Edlund et al,
      • Edlund M.J.
      • Martin B.C.
      • Fan M.Y.
      • Devries A.
      • Braden J.B.
      • Sullivan M.D.
      Risks for opioid abuse and dependence among recipients of chronic opioid therapy: results from the TROUP study.
      who evaluated 46,000 patients in Arkansas. They reported that opioid abuse or dependence was associated with mental health disorders, prior opioid abuse, younger age (18-30 years), prior nonopioid substance abuse, and higher morphine equivalent dose per day. White et al
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Tang J.
      • Katz M.P.
      Analytic models to identify patients at risk for prescription opioid abuse.
      developed an abuse prediction model based on 116,382 patients in Maine who used opioids. One of their main findings was that ≥4 opioid prescriptions (OR, 7.34) and early refills (OR, 3.39) predicted abuse. Dose escalation also was a significant risk factor (OR, 1.88). We did not assess early refills or dose escalation because they cannot be assessed at the time of first prescription. Compared with these seminal studies, our study is larger and more representative of the US population.
      Rice et al
      • Rice J.B.
      • White A.G.
      • Birnbaum H.G.
      • Schiller M.
      • Brown D.A.
      • Roland C.L.
      A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
      also studied a large, representative dataset. Their findings of nonopioid drug abuse (OR, 9.89) and a history of mental illness (OR, 2.45) increasing the risk for opioid abuse support our findings. Our study differentiates from prior studies in that we quantified how readily available demographic, clinical, behavioral, pharmacy, and geographic information predict opioid abuse or dependence. Prescribers will be able to use these variables in real-time to make a more accurate risk assessment of developing opioid abuse or dependence.

      Study Limitations

      First, the model is not implementable in states without a prescription drug monitoring program, but 49 states have a program in place or pending. Second, our study did not include Medicare, Medicaid, or Veterans Administration patients and awaits validation in these populations. However, most of the total US population is covered by private insurance.
      Kaiser Family Foundation
      Health insurance coverage of the total population.
      Third, because we used 1 year of ICD-9 codes after the index claim for identifying opioid abuse or dependence, we likely failed to capture some episodes of abuse or dependence. Future studies could include longer follow-up. Finally, the relationship between receiving ≥50% of the total dose in the immediate-release form and being diagnosed with opioid abuse or dependence was significant in the validation model, but not in the derivation model.

      Conclusions

      In light of the opioid abuse epidemic, the findings of this study warrant updating tools that estimate the risk for abuse or dependence. We recommend incorporating factors found in a prescription drug monitoring program into a patient's risk analysis. We found that risk factors for a patient being diagnosed with opioid abuse or dependence are younger age; being a chronic opioid user; histories of mental illness, nonopioid substance abuse, and alcohol abuse; being a high morphine equivalent dose user; a history of tobacco use; using multiple prescribers; residing in the South, West, or Midwest; using multiple pharmacies; male gender; and an increasing number of opioid prescriptions. Our study quantifies risk factors that are available to prescribers who are considering prescribing opioids. These insights highlight the importance of using readily available demographic, clinical, pharmacy, and geographic information to estimate the risk for opioid abuse or dependence.

      Acknowledgments

      H. M. Dinesh, MS, Genpact, contributed to data collection. From Express Scripts, Craig Reno, BS, MBA, provided clinical expertise on the analysis, Ria Westergaard, PharmD, provided clinical expertise on the analysis. In addition to the authors, Ruth Martinez, RPh, contributed to writing and editing the manuscript.

      Appendices

      Appendix 1Specific Therapeutic Class and Associated Description for Opioid Claims
      STCSTC Description
      0268Analgesics, narcotics
      6122Narcotic antitussive first-generation antihistamine-decongestant combination
      7740Analgesics narcotic, anesthetic adjunct agents
      8483Narcotic antitussive-anticholinergic combination
      8485Narcotic antitussive first-generation antihistamine
      8502Narcotic antitussive-decongestant combinations
      8514Narcotic antitussive-decongestant-expectorant combination
      8518Narcotic antitussive-expectorant combination
      8769Analgesic narcotic agonist NSAID combination
      B902Analgesic narcotics-dietary supplement combination
      B947Narcotic nonsalicylate analgesic-barbiturate-xanthine combination
      B955Narcotic salicylate analgesics-barbiturate-xanthine combination
      B974Narcotic analgesic-nonsalicylate analgesic combination
      C423Narcotic antitussive-decongestant-analgesic-expectorant combination
      C431Narcotic antitussive first-generation antihistamine-analgesic, nonsalicylate combination
      C618Narcotic salicylate analgesic combination
      NSAID = nonsteroidal anti-inflammatory drug; STC = Specific Therapeutic Class.

      Appendix 2. International Classification of Diseases, 9th Revision Cancer Diagnosis Codes for Patient Exclusion Criteria

      ICD-9 codes 140-165, 170-176, 179-209 excluding benign neoplasms under 209.x (209.4*, 209.5*, and 209.6*) were used to identify patients with cancer.
      *Denotes all combination codes under the ICD-9.
      ICD-9 = International Classification if Diseases, 9th Revision.
      Appendix 3Generic Code Numbers Associated with Cancer Drug Markers for Patient Exclusion Criteria
      STC DescriptionGeneric Code Numbers Used
      Antineoplastic – selective retinoid receptor agonists (retinoid X receptor)92373
      Antibiotic antineoplastics29203, 34241, 34242, 34247, 34248, 35080, 38581, 38590, 38591, 38592, 38593, 38594, 38600, 38601, 38602, 38610, 38613, 38622, 38623, 38630, 47340, 47343, 94175, 96679, 97242, 97271, 97272, 97277, 97278, 97282, 99510, 99835
      Anti-CD20 (B lymphocyte) monoclonal antibody27827, 30137
      Antiemetic/antivertigo agents16007, 16008, 17256, 17258, 20011, 20228, 23756, 29247, 33531, 60548, 99260, 99267, 99335, 99862
      Antileprotics19321, 28301, 95392, 98220
      Antineoplast humanized VEGF inhibitor recomb monoclonal antibody21427
      Antineoplast, histone deacetylase inhibitors28397, 97345
      Antineoplast – alkylating agents6939, 7182, 7196, 9217, 12014, 14401, 17724, 24699, 24701, 34221, 34310, 38232, 38340, 38350, 38351, 38352, 38353, 38357, 38360, 38361, 38370, 38380, 38390, 38410, 38420, 38422, 38431, 38432, 38433, 38440, 38450, 38451, 38910, 38911, 38912, 38920, 48862, 60901, 72722, 72730, 72731, 72732, 72733, 72734, 92893, 92903, 92913, 92933, 97957, 98310, 98311, 98709, 98710, 98813
      Antineoplast – antidrogenic agents450, 22642, 22645, 25740, 29886, 33183
      Antineoplast – antimetabolites880, 10290, 12473, 19901, 21179, 21473, 21485, 21501, 21503, 22663, 23432, 23439, 24037, 25932, 27027, 27365, 27663, 27664, 30776, 30777, 30778, 31611, 31612, 32981, 34230, 34231, 38490, 38500, 38520, 38530, 38531, 38532, 38540, 38541, 38542, 38543, 93472, 97455, 97456, 97457, 97458, 97825, 99268
      Antineoplast – Aromatase inhibitors17300
      Antineoplast – epothilones and analogs98998, 98999
      Antineoplast – halichondrin B analogs29249
      Antineoplast – Hedgehog pathway inhibitor31307
      Antineoplast – Janus kinase inhibitors30892, 30893, 30894, 30895, 30896
      Antineoplast – mTOR kinase inhibitors20784, 20844, 28783, 31396, 34589, 34590, 34592, 98597
      Antineoplast – topoisomerase I inhibitors14254, 14256, 22661, 29519, 97955, 97956, 99056, 99790
      Antineoplast – VEGF A and B isoforms, and PLGF inhibitors32988, 32989
      Antineoplast – vinca alkaloids38560, 38572, 38580, 38820, 38970, 97327, 97630
      Antineoplast antibody/radioactive-drug complexes20159, 20160
      Antineoplast epidermal growth factor receptor blocker monoclonal antibody13632, 13638, 13639, 15979, 15983, 28471, 32343
      Antineoplast immunomodulator agents26314, 26315, 27276, 27277, 29809, 29811, 29812, 31911, 34147, 34148, 34149, 34150, 34743
      Antineoplast LHRH and GNRH agonist, pituitary suppressant13133, 15338, 15344, 16945, 16946, 17377, 18155, 19219, 21004, 23768, 24301, 28506, 28507, 29894, 30083, 84590, 84591, 84592, 84593, 84594, 84596, 84597, 84598, 84601, 84602, 99763, 99764
      Antineoplast systemic enzyme inhibitors13369, 19586, 19656, 19907, 19908, 23793, 23794, 23795, 26263, 26452, 26453, 26454, 27257, 27258, 27259, 27829, 28737, 29405, 29406, 29817, 29818, 30332, 30457, 30458, 31294, 31295, 32722, 33199, 33202, 33363, 33873, 33874, 33903, 33904, 33905, 34723, 34724, 34726, 34727, 98140, 99070, 99867
      Antineoplast antibody/antibody-drug complexes14171, 18373, 18374, 20158, 21050, 24507, 30404, 34234, 34235
      Antineoplast – miscellaneous7480, 7481, 7544, 7550, 7552, 7560, 14103, 24094, 24231, 28663, 28762, 29066, 29591, 29662, 29663, 29664, 30918, 33734, 38710, 38730, 38731, 38732, 38740, 38750, 39000, 39150, 39152, 39153, 39154, 47410, 48480, 48481, 48590, 85410, 85602, 85602, 93610
      Chemotherapy rescue/antidote agents1330, 27236, 31194, 36901, 38950, 38953, 38955, 87552, 87553, 87554, 87555, 87556, 87557, 87558, 87559, 87562, 87563, 89655
      CXCR4 chemokine receptor antagonist16124
      Cytotoxic T-lymphocyte antigen recombinant monoclonal antibody29688, 29689
      Immunomodulators26405, 46471, 46472, 47511, 47512, 47513, 47520, 47521, 47522, 47523, 47524, 47525, 47526, 47527, 47528, 47529, 47530, 47600, 47601, 47602, 47603, 47604, 47605, 47661, 47662, 47663, 48891, 48931, 48941, 49031, 90823, 90833
      Keratinocyte growth factor23928
      Leukocyte (WBC) stimulants13206, 13308, 13309, 15666, 26001, 26220, 26221, 26222
      LHRH (GNRH) agonist analog pituitary suppressants23768, 80254, 84350
      Selective estrogen receptor modulators17307, 17308, 38720, 38721, 50377
      Steroid antineoplastics38640, 38661, 38700
      Tissue protective treatment of chemotherapy extravasation30562
      Topical antineoplastic and premalignant lesion agents89921
      GNRH = gonadotropin-releasing hormone; LHRH = luteinizing hormone; MTOR = mammalian target of rapamycin; PLGF = placental growth factor; STC = specific therapeutic class; VEGF = vascular endothelial growth factor; WBC = white blood cell.
      Appendix 4Generic Code Numbers Associated with Suboxone for Patient Exclusion Criteria
      STC DescriptionGeneric Code Numbers Used
      Narcotic withdrawal therapy agents18973, 18974, 28958, 28959, 33741, 33744, 34904, 34905, 36677, 36678, 36679
      STC = specific therapeutic class.
      Appendix 5International Classification of Diseases, Ninth Revision Codes Associated with Independent Variables
      DescriptionICD-9 Code
      Nonopioid substance abuse304.1-304.9 and 305.2-305.9, excluding 305.5x
      Tobacco use disorder305.1
      Nondependent alcohol abuse303.9 and 305.0x
      Mental illness290-302 and 306-316
      ICD-9 = International Classification of Diseases, Ninth Revision.

      Appendix 6. Sensitivity Analyses to Test the Robustness of the Findings

      The first sensitivity analysis was “Firth's bias-adjusted estimation,” which maximizes a penalized likelihood function and provides finite parameter estimates. Second was “oversampling” to increase the target rate by 10 times. This helps address the bias resulting from the margin of sampling error being related to the outcome sample size.
      Tabled 1Sensitivity Analysis Using Firth Bias-Adjusted Estimation Method and Oversampling Method
      ReferenceFirth's Bias-Adjusted Estimation Model
      The c-statistics for the Firth's bias-adjusted estimation model was 0.853, and oversampling method model was 0.876.
      Oversampling Method Model
      The c-statistics for the Firth's bias-adjusted estimation model was 0.853, and oversampling method model was 0.876.
      OR
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      95% CIOR
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      95% CI
      AgeNA0.960.96-0.970.960.96-0.96
      Chronic usersAbsent4.393.71-5.193.983.28-4.83
      History of mental illnessAbsent3.453.13-3.793.633.28-4.03
      History of nonopioid substance abuseAbsent2.832.18-3.633.932.84-5.44
      History of nondependent alcohol abuseAbsent2.381.84-3.042.641.93-3.60
      Daily MED ≥120 mg/dAbsent1.981.67-2.341.831.50-2.24
      History of tobacco use disorderAbsent1.801.59-2.042.041.77-2.34
      Prescriber shoppersAbsent1.711.55-1.891.711.53-1.91
      Region
       SouthNortheast1.651.45-1.871.791.56-2.05
       WestNortheast1.491.29-1.721.521.30-1.78
       MidwestNortheast1.241.08-1.421.291.12-1.50
      Pharmacy shoppersNo1.591.31-1.922.031.58-2.61
      MaleFemale1.431.31-1.571.521.37-1.68
      Prior opioid 30-d adjusted prescriptionsNA1.051.04-1.061.061.05-1.07
      Chronic immediate-release usersAbsent1.06
      Not significant at P <.05.
      0.93-1.221.03
      Not significant at P <.05.
      0.88-1.21
      Distance from patient to prescriber≤50 miles1.12
      Not significant at P <.05.
      0.99-1.271.12
      Not significant at P <.05.
      0.98-1.28
      CI = confidence interval; MED = morphine equivalent dosing; NA = not applicable (for continuous variables); OR = odds ratio.
      All data were significant at P <.05 unless marked with a note indicating otherwise.
      Not significant at P <.05.
      The c-statistics for the Firth's bias-adjusted estimation model was 0.853, and oversampling method model was 0.876.

      References

        • Dart R.C.
        • Surrat H.L.
        • Cicero T.J.
        • et al.
        Trends in opioid analgesic abuse and mortality in the United States.
        N Engl J Med. 2015; 372: 241-248
        • Rudd R.
        • Aleshire N.
        • Zibbell J.E.
        • Gladden M.
        Increases in drug and opioid overdose deaths — United States, 2000-2014.
        MMWR Morb Mortal Wkly Rep. 2016; 64: 1378-1382
        • Katz N.P.
        • Adams E.H.
        • Chilcoat H.
        • et al.
        Challenges in the development of prescription opioid abuse-deterrent formulations.
        Clin J Pain. 2007; 23: 648-660
        • American Psychiatric Association
        Diagnostic and Statistical Manual of Mental Disorders.
        4th ed. American Psychiatric Association, Washington, DC1994
        • Savage S.R.
        • Joranson D.E.
        • Covington E.C.
        • Schnoll S.H.
        • Heit H.A.
        • Gilson A.M.
        Definitions related to the medical use of opioids: evolution towards universal agreement.
        J Pain Symptom Manage. 2003; 26: 655-667
        • Chou R.
        • Fanciullo G.J.
        • Fine P.G.
        • et al.
        Clinical guidelines for the use of chronic opioid therapy in chronic noncancer pain.
        J Pain. 2009; 10: 113-130
        • Paulozzi L.J.
        • Jones C.M.
        • Mack K.A.
        • Rudd R.A.
        Vital signs: overdoses of prescription opioid pain relievers United States, 1999-2008.
        MMWR Morb Mortal Wkly Rep. 2011; 60: 1487-1492
        • Katz N.
        • Panas L.
        • Kim M.
        • et al.
        Usefulness of prescription monitoring programs for surveillance–analysis of Schedule II opioid prescription data in Massachusetts, 1996-2006.
        Pharmacoepidemiol Drug Saf. 2010; 19: 115-123
        • Wang J.
        • Christo P.J.
        The influence of prescription monitoring programs on chronic pain management.
        Pain Physician. 2009; 12: 507-515
        • Centers for Medicare and Medicaid Services
        The role of a prescription drug monitoring program in reducing prescription drug diversion, abuse, and misuse.
        June 2014 (Available at:) (Accessed May 4, 2015)
        • Webster L.R.
        • Webster R.M.
        Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the opioid risk tool.
        Pain Med. 2005; 6: 432-442
        • McNeely J.
        • Strauss S.M.
        • Saitz R.
        • et al.
        A brief patient self-administered substance use screening tool for primary care: two-site validation study of the Substance Use Brief Screen (SUBS).
        Am J Med. 2015; 128: 784-784.e19
        • Bowman S.
        • Eiserman J.
        • Beletsky L.
        • Stancliff S.
        • Bruce R.D.
        Reducing the health consequences of opioid addiction in primary care.
        Am J Med. 2013; 126: 565-571
      1. First Databank. Available at: http://www.fdbhealth.com/. Accessed April 24, 2015.

        • Cepeda M.S.
        • Fife D.
        • Chow W.
        • Mastrogiovanni G.
        • Henderson S.C.
        Opioid shopping behavior: how often, how soon, which drugs, and what payment method.
        J Clin Pharmacol. 2013; 53: 112-117
        • Von Korff M.
        • Saunders K.
        • Thomas Ray G.
        • et al.
        De facto long-term opioid therapy for noncancer pain.
        Clin J Pain. 2008; 24: 521-527
        • Rice J.B.
        • White A.G.
        • Birnbaum H.G.
        • Schiller M.
        • Brown D.A.
        • Roland C.L.
        A model to identify patients at risk for prescription opioid abuse, dependence, and misuse.
        Pain Med. 2012; 13: 1162-1173
        • Skala K.
        • Reichl L.
        • Ilias W.
        • et al.
        Can we predict addiction to opioid analgesics? A possible tool to estimate the risk of opioid addiction in patients with pain.
        Pain Physician. 2013; 16: 593-601
        • McCabe S.E.
        • Cranford J.A.
        • Boyd C.J.
        The relationship between past-year drinking behaviors and nonmedical use of prescription drugs: prevalence of co-occurrence in a national sample.
        Drug Alcohol Depend. 2006; 84: 281-288
        • Manchikanti L.
        • Helm II, S.
        • Fellows B.
        • et al.
        Opioid epidemic in the United States.
        Pain Physician. 2012; 15: ES9-ES38
        • Manchikanti L.
        • Abdi S.
        • Atluri S.
        • et al.
        American Society of Interventional Pain Physicians (ASIPP) guidelines for responsible opioid prescribing in chronic non-cancer pain: Part 2 –guidance.
        Pain Physician. 2012; 15: S67-S116
        • Sehgal N.
        • Manchikanti L.
        • Smith H.S.
        Prescription opioid abuse in chronic pain: a review of opioid abuse predictors and strategies to curb opioid abuse.
        Pain Physician. 2012; 15: ES67-ES92
        • Cheatle M.D.
        • O'Brien C.P.
        • Mathai K.
        • Hansen M.
        • Grasso M.
        • Yi P.
        Aberrant behaviors in a primary care-based cohort of patients with chronic pain identified as misusing prescription opioids.
        J Opioid Manag. 2013; 9: 315-324
        • Parente S.T.
        • Kim S.S.
        • Finch M.D.
        • et al.
        Identifying controlled substance patterns of utilization requiring evaluation using administrative claims data.
        Am J Manag Care. 2004; 10: 783-790
        • Hall A.J.
        • Logan J.E.
        • Toblin R.L.
        • et al.
        Patterns of abuse among unintentional pharmaceutical overdose fatalities.
        JAMA. 2008; 300: 2613-2620
      2. US Census Bureau. Geographic terms and concepts: census divisions and census regions. Available at: https://www.census.gov/geo/reference/gtc/gtc_census_divreg.html. Accessed February 9, 2015.

        • White A.G.
        • Birnbaum H.G.
        • Schiller M.
        • Tang J.
        • Katz M.P.
        Analytic models to identify patients at risk for prescription opioid abuse.
        Am J Manag Care. 2009; 15: 897-906
        • Roland C.L.
        • Joshi A.V.
        • Mardekian J.
        • Walden S.C.
        • Harnett J.
        Prevalence and cost of diagnosed opioid abuse in a privately insured population in the United States.
        J Opioid Manag. 2013; 9: 161-175
      3. Basic Health and Human Services policy for protection of human research subject. 45 C.F.R. § 46 (2009).

        • Dufour R.
        • Markekian J.
        • Pasquale M.K.
        • Schaaf D.
        • Andrews G.
        • Patel N.
        Understanding predictors of opioid abuse: predictive model development and validation.
        Am J Pharm Benefits. 2014; 6: 208-216
        • Edlund M.J.
        • Martin B.C.
        • Fan M.Y.
        • Devries A.
        • Braden J.B.
        • Sullivan M.D.
        Risks for opioid abuse and dependence among recipients of chronic opioid therapy: results from the TROUP study.
        Drug Alcohol Depend. 2010; 112: 90-98
        • Kaiser Family Foundation
        Health insurance coverage of the total population.
        2015 (Available at:) (Accessed March 24, 2015)