This study investigates the association between long-term opioid use among family members and persistent opioid use among opioid-naive adolescents and young adults undergoing surgical and dental procedures.
Key Points
Question
Is long-term opioid use among family members associated with persistent opioid use among opioid-naive adolescents and young adults undergoing surgical and dental procedures?
Findings
In this cohort study of 346 251 opioid-naive individuals aged 13 to 21 years undergoing common surgical and dental procedures with an initial opioid prescription fill, persistent opioid use occurred in 453 patients (4.1%) with long-term opioid use in family members compared with 5940 patients (2.4%) without long-term opioid use in family members.
Meaning
The findings suggest that long-term opioid use among family members is associated with persistent opioid use among opioid-naive adolescents and young adults undergoing surgery and should be screened for in the preoperative period.
Abstract
Importance
Prior studies have found a substantial risk of persistent opioid use among adolescents and young adults undergoing surgical and dental procedures. It is unknown whether family-level factors, such as long-term opioid use in family members, is associated with persistent opioid use.
Objective
To determine whether long-term opioid use in family members is associated with persistent opioid use among opioid-naive adolescents and young adults undergoing surgical and dental procedures.
Design, Setting, and Participants
This retrospective cohort study used data from a commercial insurance claims database for January 1, 2010, to June 30, 2016, to study 346 251 opioid-naive patients aged 13 to 21 years who underwent 1 of 11 surgical and dental procedures and who were dependents on a family insurance plan.
Exposures
Long-term opioid use in family members, defined as having 1 or more family members who (1) filled opioid prescriptions totaling a 120 days’ supply or more during the 12 months before the procedure date or (2) filled 3 or more opioid prescriptions in the 90 days before the procedure date.
Main Outcomes and Measures
The main outcome measure was persistent opioid use, defined as 1 or more postoperative prescription opioid fills between 91 and 180 days among patients with an initial opioid prescription fill. Generalized estimating equations with robust SEs clustered at the family level were used to model persistent opioid use as a function of long-term opioid use among family members, controlling for procedure, total morphine milligram equivalents of the initial fill, and patient and family characteristics.
Results
A total of 346 251 patients (mean [SD] age, 17.0 [2.3] years; 175 541 [50.7%] female) were studied. Among these patients, 257 085 (74.3%) had an initial opioid fill. Among patients with an initial opioid fill, 11 016 (4.3%) had long-term opioid use in a family member. Persistent opioid use occurred in 453 patients (4.1%) with long-term opioid use in a family member compared with 5940 patients (2.4%) without long-term opioid use in a family member (adjusted odds ratio, 1.54; 95% CI, 1.39-1.71).
Conclusion and Relevance
The findings suggest that long-term opioid use among family members is associated with persistent opioid use among opioid-naive adolescents and young adults undergoing surgical and dental procedures. Physicians should screen young patients for long-term opioid use in their families and implement heightened efforts to prevent opioid dependence among patients with this important risk factor.
Introduction
Opioid-related overdose deaths in the United States continue to increase, including among adolescents and young adults.1,2 Between 2015 and 2016, prescription opioid–related overdose deaths increased by 30% among individuals aged 15 to 24 years, the highest of any age group.2 Surgical and dental procedures are among the most frequent indications for opioid prescriptions to adolescents and young adults.3,4 Opioids prescribed after procedures have been associated with development of persistent opioid use (ie, opioid use beyond the postoperative recovery period) in 5% of adolescents and young adults,5,6 mirroring trends in adults.7,8,9,10 It is crucial to identify which patients are at elevated risk to implement prevention measures.
Prior studies5,11 have identified several patient-level factors associated with persistent opioid use, including prior mental health conditions and substance use disorders. Family-level factors, such as long-term opioid use in family members, may also play an important role but, to our knowledge, have not previously been explored in this context. It is possible that long-term opioid use in family members may indicate the presence of shared genetic polymorphisms associated with chronic pain, opioid metabolism phenotypes, or neurochemical signaling pathways associated with opioid dependence.12,13,14,15 In addition, families may share common environmental exposures16,17 or attitudes toward pain management and opioid use that increase the risk of persistent opioid use when adolescents and young adults are exposed to opioids.18 It is also possible that an opioid prescription filled by an adolescent or young adult may have been diverted for use by another family member. By linking individuals within families by insurance plan within a large administrative database, important early work demonstrated a 0.7% higher 1-year risk of prescription opioid use initiation among individuals in households exposed to opioids compared with individuals in households exposed to nonsteroidal anti-inflammatory drugs.19
Using a similar approach, we investigated the association between long-term opioid use among family members and persistent opioid use among opioid-naive adolescents and young adults undergoing surgical and dental procedures. We hypothesized that long-term opioid use among family members would be associated with an increased risk of persistent opioid use as well as increased risk of opioid fills at the time of surgery and during the postoperative recovery periods.
Methods
Data Sources
Data sources were the Truven Health MarketScan Commercial Claims and Encounters Database and the MarketScan Dental Database for January 1, 2009, to December 31, 2016. The MarketScan Commercial Claims and Encounters Database contains linked and deidentified inpatient, outpatient, and pharmacy insurance claims for nonelderly enrollees in all 50 states who receive coverage from 1 of approximately 150 large employers. Annual sample sizes range from 43 million to 55 million beneficiaries.20 Groups of beneficiaries sharing a common insurance plan are assigned a unique encrypted family identifier, and individuals within each insurance plan are identified by relationship to the employee (primary insurance holder, spouse, or dependent).19 The MarketScan Dental Database includes dental insurance claims for 8.8 million beneficiaries through 2016, among whom approximately 81% link to the MarketScan Commercial Claims and Encounters Database.21 For this study, beneficiaries in the MarketScan Dental Database were only included if they could be linked to the MarketScan Commercial Claims and Encounters Database. Because of the use of deidentified claims, the University of Michigan Institutional Review Board exempted this study from human subject review. This study reporting is adherent with Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.22
Study Population
The study population consisted of adolescents and young adults aged 13 to 21 years at the time of surgery who had 1 of 11 common surgical or dental procedures between January 1, 2010, and June 30, 2016. Procedures included surgical tooth extraction, tonsillectomy and/or adenoidectomy, appendectomy, cholecystectomy, pectus repair, umbilical or epigastric hernia repair, inguinal hernia repair, orchiopexy, arthroscopic knee surgery, arthroscopic shoulder surgery, and upper extremity fracture with surgical fixation. Dental procedures were identified by Codes on Dental Procedures and Nomenclature, and surgical procedures were identified by Current Procedural Terminology codes (eTable 1 in the Supplement). The procedure date was defined as the first instance of any of these codes during the study period.
Patients were excluded if they (1) did not have continuous insurance enrollment and pharmacy benefit coverage in the 12 months before to 6 months after the procedure date; (2) were not opioid naive (defined as having no opioid fills between the 365 days and 7 days before the procedure date); (3) had an inpatient admission immediately after the procedure that lasted more than 30 days; (4) had any additional anesthetic code during the 6 months after surgery (eTable 1 in the Supplement); (5) did not have 1 or more family members on the same insurance plan who also had continuous enrollment and pharmacy benefit coverage in the 12 months before to 6 months after the procedure date; (6) were the primary insurance holder or spouse to the primary insurance holder (ie, were not dependents on the insurance plan); or (7) had any code for a complex chronic condition (including malignant neoplasm) on any claim in the 12 months before the procedure date (to include only medically noncomplex patients)23 (Figure 1).
Figure 1. Data Set Construction.
Opioid Prescription Fills
Opioid prescription fills were identified by therapeutic drug class and generic name in the MarketScan pharmacy file, including (1) all analgesic or antipyretic opiate agonists, (2) buprenorphine and buprenorphine hydrochloride in the all analgesic or antipyretic opiate partial agonist class, and (3) all medications with the generic name tramadol that were not a topical formulation. Opioid prescription fills with any of the following medications in the generic name were excluded as cough or cold preparations: brompheniramine, chlorpheniramine, pseudoephedrine, phenylephrine, and guaifenesin.
Study Variables
The primary outcome was persistent opioid use, defined as the occurrence of 1 or more opioid prescription fills 91 to 180 days after the procedure date among patients with 1 or more initial opioid prescription fills (see below).5,7,8,9,10 This time frame was based on the International Association for the Study of Pain’s definition of long-term postsurgical pain as pain that persisted beyond 3 months after surgery.24 The secondary outcomes were (1) initial opioid fill, defined as an opioid prescription filled in the perioperative window (eg, between 7 days before and 3 days after the procedure date), and (2) postoperative opioid refill, defined as the occurrence of 1 or more opioid prescription fills 4 to 90 days after the procedure date among patients with an initial fill.9,11
The primary exposure variable was long-term opioid use among any family member, defined as having 1 or more family members who (1) filled opioid prescriptions totaling 120 days’ supply or more during the 12 months before the procedure date or (2) filled 3 or more opioid prescriptions in the 90 days before the procedure date.25,26 This level of opioid use was associated with an increased risk of opioid overdose mortality.26
Covariates included patient age, sex, geographic region, procedure type, and patient and family comorbidities. Comorbidities included (1) mental health diagnoses, including mood; disruptive; suicide, self-harm, or miscellaneous; and nonopioid substance use disorders defined by Clinical Classifications Software (Agency for Healthcare Research and Quality); (2) chronic pain diagnoses5,8; (3) opioid use disorder, including opioid-related overdose; and (4) uncomplicated opioid dependence (to reflect nomenclature established by the Diagnostic and Statistical Manual of Mental Disorders [Fifth Edition]). Comorbidities were defined as the occurrence of 1 or more claims with a relevant International Classification of Diseases, Ninth Revision or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnosis code during the 12 months before the procedure date (eTable 2 in the Supplement). For analyses of persistent opioid use and postoperative opioid refill, covariates also included total morphine milligram equivalents of opioid(s) prescriptions filled in the perioperative window, which was calculated using standard published conversions.27 Winsorization was performed for total morphine milligram equivalent outliers greater than 3 SDs above the mean.
Statistical Analysis
To assess unadjusted associations between the exposure and the primary and secondary outcomes, χ2 analyses were used. To assess adjusted associations, generalized estimating equations with robust SEs clustered at the family level were used to model all outcomes as a function of having 1 or more family members with long-term opioid use, controlling for all covariates. To convert adjusted odds ratios (aORs) into more interpretable estimates of risk, marginal effects were calculated, setting each covariate to their observed value in the data. In analyses of initial opioid prescription fills, we included all eligible patients. In analyses of postoperative opioid refill and persistent opioid use, we studied a subset of patients who had an initial opioid prescription fill.
In a secondary analysis, regressions were repeated except that the exposure variable was replaced with an indicator for the number of parents with long-term opioid use (1 vs none, 2 vs none) and an indicator for the number of other dependents with long-term opioid use (1 vs none). Only 2 families had 2 dependents with long-term opioid use; these families were excluded for this analysis. Parents were defined as the primary insurance holder or the spouse of the primary insurance holder. This analysis was conducted to evaluate whether there was an association of the number and role of family members with long-term opioid use with all outcomes. In an additional subgroup analysis, family members with long-term opioid use were categorized as malignant or benign based on the occurrence of any diagnostic code that indicated malignant tumors in the 12 months before the procedure (eTable 2 in the Supplement).
Statistical analyses were performed using 2-sided tests; P < .05 was considered to be statistically significant. All analyses were conducted using SAS statistical software, version 9.4 (SAS Institute Inc) and Stata SE statistical software, version 15.1 (StataCorp).
Results
Sample Characteristics
A total of 346 251 patients (mean [SD] age, 17.0 [2.3] years; 175 541 [50.7%] female) met the sample inclusion and exclusion criteria. Overall, 13 929 patients (4.0%) had 1 more family members with long-term opioid use; 13 560 (97.4%) lived in families in which 1 or more parents had long-term opioid use, and 437 (3.1%) lived in families in which 1 or more dependents had long-term opioid use. Among these families, 68 (0.5%) had 1 or more parents and 1 or more other dependents with long-term opioid use.
Compared with patients without long-term opioid use among any family members, patients with long-term opioid use in family members were more likely to have prior diagnoses of chronic pain, mental health, and opioid use disorders. Type of procedure and geographic region also differed between these groups of patients (eTable 3 in the Supplement).
Initial Opioid Fill
Overall, 257 085 patients (74.3% of the sample) had an initial opioid fill (Table 1). Among patients with a family member with long-term opioid use, 11 016 (79.1%) had an initial fill, compared with 246 069 (74.1%) who did not have a family member with long-term opioid use (P < .001 for unadjusted comparison; aOR, 1.23; 95% CI, 1.18-1.29; marginal effect, 5.0 percentage points; 95% CI, 4.4–5.7 percentage points) (Table 2 and Figure 2).
Table 1. Patient and Family Characteristics by Long-term Opioid Use in Family Members Among Patients With an Initial Opioid Filla.
| Characteristic | Long-term Opioid Use in Family | No Long-term Opioid Use in Family | P Value |
|---|---|---|---|
| Patients | 11 016 (4.3) | 246 069 (95.7) | NA |
| Female sex | 5641 (51.2) | 126 018 (51.2) | .99 |
| Age, y | |||
| 13-15 | 2467 (22.4) | 52 698 (21.4) | .01 |
| 16-18 | 5308 (48.2) | 121 932 (49.6) | |
| 19-21 | 3241 (29.4) | 71 439 (29.0) | |
| Geographic region | |||
| Northeast | 1437 (13.0) | 41 199 (16.7) | <.001 |
| Midwest | 3169 (28.8) | 70 548 (28.7) | |
| South | 4239 (38.5) | 88 321 (35.9) | |
| West | 2131 (19.3) | 45 263 (18.4) | |
| Unknown | 40 (0.4) | 738 (0.3) | |
| Procedure typeb | |||
| Surgical tooth extraction | 6275 (57.0) | 151 066 (61.4) | <.001 |
| Tonsillectomy and/or adenoidectomy | 1513 (13.7) | 25 581 (10.4) | |
| Appendectomy | 712 (6.5) | 13 357 (5.4) | |
| Cholecystectomy | 380 (3.4) | 4822 (2.0) | |
| Pectus repair | 28 (0.3) | 486 (0.2) | |
| Umbilical or epigastric hernia repair | 30 (0.3) | 590 (0.2) | |
| Inguinal hernia repair | 164 (1.5) | 3233 (1.3) | |
| Orchiopexy | 40 (0.4) | 795 (0.3) | |
| Knee arthroscopy | 1069 (9.7) | 28 446 (11.6) | |
| Shoulder arthroscopy | 389 (3.5) | 8514 (3.5) | |
| Upper extremity fracture with operative fixation | 416 (3.8) | 9179 (3.7) | |
| Total MME during the perioperative window, mean (SD) | 269.6 (648.8) | 248.4 (630.2) | .001 |
| Patient comorbidities | |||
| Chronic pain diagnosis | 1091 (9.9) | 21 646 (8.8) | <.001 |
| Mood disorder | 1570 (14.3) | 25 762 (10.5) | <.001 |
| Disruptive disorder | 926 (8.4) | 15 364 (6.2) | <.001 |
| Suicide/self-harm or miscellaneous disorder | 183 (1.7) | 3167 (1.3) | .001 |
| Nonopioid substance use disorder | 224 (2) | 3172 (1.3) | <.001 |
| Opioid use disorder or overdose | 39 (0.4) | 261 (0.1) | <.001 |
| Uncomplicated opioid dependence | 31 (0.3) | 181 (0.1) | <.001 |
| Family comorbidities | |||
| Chronic pain diagnosis | 10 390 (94.3) | 155 280 (63.1) | <.001 |
| Mood disorder | 5090 (46.2) | 60 645 (24.7) | <.001 |
| Disruptive disorder | 1363 (12.4) | 20 415 (8.3) | <.001 |
| Suicide/self-harm or miscellaneous disorder | 938 (8.5) | 8550 (3.5) | <.001 |
| Nonopioid substance use disorder | 897 (8.1) | 6172 (2.5) | <.001 |
| Opioid use disorder or overdose | 101 (0.9) | 270 (0.1) | <.001 |
| Uncomplicated opioid dependence | 537 (4.9) | 871 (0.4) | <.001 |
| No. of parents with long-term opioid use | |||
| 0 | 295 (2.7) | NA | NA |
| 1 | 10 080 (91.5) | NA | |
| 2 | 641 (5.8) | NA | |
| No. of other dependents with long-term opioid use | |||
| 0 | 10 668 (96.8) | NA | NA |
| 1 | 347 (3.2) | NA | |
| 2 | 1 (0.001) | NA | |
| Postoperative opioid fill rates | |||
| Postoperative refill (days 4-90) | 2515 (22.8) | 28 508 (11.6) | <.001 |
| Persistent use (days 91-180) | 453 (4.1) | 5940 (2.4) | <.001 |
Abbreviations: MME, morphine milligram equivalents; NA, not applicable.
Data are presented as number (percentage) of patients unless otherwise indicated.
Does not add up to 100% owing to rounding error.
Table 2. Coefficients From Multivariable Regressions.
| Characteristic | Initial Opioid Filla | Postoperative Opioid Refillb | Persistent Opioid Use (Primary Outcome)b | |||
|---|---|---|---|---|---|---|
| aOR (95% CI) | P Value | aOR (95% CI) | P Value | aOR (95% CI) | P Value | |
| Long-term opioid use in family | 1.23 (1.18-1.29) | <.001 | 2.14 (2.03-2.25) | <.001 | 1.54 (1.39-1.71) | <.001 |
| Female (reference: male) | 1.12 (1.11-1.14) | <.001 | 1.14 (1.11-1.17) | <.001 | 1.30 (1.24-1.37) | <.001 |
| Age, y (reference: 13-15 y) | ||||||
| 16-18 | 3.46 (3.39-3.52) | <.001 | 1.46 (1.41-1.52) | <.001 | 1.57 (1.46-1.69) | <.001 |
| 18-21 | 3.34 (3.27-3.42) | <.001 | 1.98 (1.91-2.05) | <.001 | 1.66 (1.54-1.79) | <.001 |
| Geographic region (reference: Northeast) | ||||||
| Midwest | 1.00 (0.98-1.03) | .87 | 1.55 (1.48-1.61) | <.001 | 1.31 (1.21-1.43) | <.001 |
| South | 1.21 (1.18-1.24) | <.001 | 1.49 (1.43-1.55) | <.001 | 1.57 (1.45-1.70) | <.001 |
| West | 0.86 (0.84-0.88) | <.001 | 1.36 (1.30-1.43) | <.001 | 1.32 (1.20-1.45) | <.001 |
| Unknown | 0.95 (0.81-1.11) | .49 | 1.43 (1.17-1.75) | <.001 | 1.94 (1.36-2.77) | <.001 |
| Procedure (reference: surgical tooth extraction) | ||||||
| Tonsillectomy and/or adenoidectomy | 3.98 (3.82-4.13) | <.001 | 7.87 (7.61-8.14) | <.001 | 2.01 (1.86-2.17) | <.001 |
| Appendectomy | 1.33 (1.28-1.37) | <.001 | 1.00 (0.93-1.07) | .96 | 2.47 (2.25-2.72) | <.001 |
| Cholecystectomy | 1.39 (1.31-1.49) | <.001 | 1.26 (1.14-1.38) | <.001 | 2.66 (2.34-3.02) | <.001 |
| Pectus repair | 3.33 (2.64-4.20) | <.001 | 6.22 (5.11-7.58) | <.001 | 1.76 (1.03-3.02) | .04 |
| Umbilical or epigastric hernia repair | 1.48 (1.25-1.75) | <.001 | 0.86 (0.62-1.21) | .39 | 2.32 (1.52-3.52) | <.001 |
| Inguinal hernia repair | 2.18 (2.00-2.38) | <.001 | 1.37 (1.21-1.54) | <.001 | 1.96 (1.60-2.41) | <.001 |
| Orchiopexy | 2.70 (2.30-3.16) | <.001 | 1.08 (0.81-1.45) | .59 | 1.88 (1.20-2.94) | .01 |
| Knee arthroscopy | 3.09 (2.98-3.20) | <.001 | 3.23 (3.11-3.36) | <.001 | 2.09 (1.94-2.26) | <.001 |
| Shoulder arthroscopy | 2.89 (2.70-3.08) | <.001 | 3.52 (3.31-3.73) | <.001 | 2.43 (2.16-2.74) | <.001 |
| Upper extremity fracture with operative fixation | 2.70 (2.56-2.85) | <.001 | 2.39 (2.24-2.54) | <.001 | 1.99 (1.75-2.26) | <.001 |
| Total MME during the perioperative window | NA | NA | 1.00 (1.00-1.00) | .23 | 1.00 (1.00-1.00) | .27 |
| Patient comorbidities | ||||||
| Chronic pain diagnosis | 1.22 (1.18-1.26) | <.001 | 1.33 (1.28-1.39) | <.001 | 1.21 (1.11-1.31) | <.001 |
| Mood disorder | 1.21 (1.18-1.25) | <.001 | 1.28 (1.23-1.33) | <.001 | 1.26 (1.17-1.36) | <.001 |
| Disruptive disorder | 1.09 (1.05-1.12) | <.001 | 1.12 (1.06-1.17) | <.001 | 1.15 (1.04-1.27) | .01 |
| Suicide/self-harm or miscellaneous disorder | 0.97 (0.89-1.05) | .43 | 1.03 (0.93-1.14) | .60 | 1.18 (0.98-1.42) | .09 |
| Nonopioid substance use disorder | 0.93 (0.86-1.01) | .07 | 1.43 (1.30-1.58) | <.001 | 1.33 (1.11-1.60) | <.001 |
| Opioid use disorder or overdose | 0.73 (0.46-1.14) | .16 | 1.11 (0.59-2.08) | .75 | 1.17 (0.42-3.25) | .77 |
| Uncomplicated opioid dependence | 0.56 (0.34-0.93) | .03 | 1.49 (0.73-3.03) | .28 | 1.58 (0.50-4.99) | .44 |
| Family comorbidities | ||||||
| Chronic pain diagnosis | 1.10 (1.09-1.12) | <.001 | 1.04 (1.01-1.07) | <.001 | 1.14 (1.08-1.20) | <.001 |
| Mood disorder | 1.08 (1.06-1.10) | <.001 | 1.12 (1.09-1.16) | <.001 | 1.03 (0.97-1.10) | .29 |
| Disruptive disorder | 1.03 (1.00-1.06) | .08 | 1.07 (1.02-1.12) | <.001 | 1.02 (0.93-1.11) | .70 |
| Suicide/self-harm or miscellaneous disorder | 1.01 (0.96-1.05) | .82 | 1.04 (0.97-1.11) | .24 | 0.99 (0.87-1.12) | .84 |
| Nonopioid substance use disorder | 1.02 (0.97-1.08) | .39 | 1.14 (1.05-1.22) | <.001 | 1.14 (0.98-1.31) | .08 |
| Opioid use disorder or overdose | 0.79 (0.64-0.99) | .04 | 0.98 (0.72-1.32) | .88 | 0.56 (0.26-1.20) | .14 |
| Uncomplicated opioid dependence | 1.03 (0.92-1.17) | .59 | 1.07 (0.91-1.26) | .41 | 1.04 (0.76-1.43) | .78 |
Abbreviations: aOR, adjusted odds ratio; MME, morphine milligram equivalents.
Multivariable regression includes all eligible patients (N = 346 251).
Multivariable regression includes only eligible patients with an initial opioid fill (n = 257 085).
Figure 2. Adjusted Initial Fill, Postoperative Refill, and Persistent Use by Exposure to Long-term Opioid Use in Family Members.
Multivariable regression models were used to model initial opioid fill, postoperative opioid refill, and persistent opioid use as a function of long-term opioid use in family members.
Postoperative Opioid Refill
Among the 257 085 patients who had an initial opioid prescription fill, 11 016 (4.3%) had a family member with long-term opioid use (Table 1). Among patients with an initial fill and exposure to long-term opioid use in family members, 2515 (22.8%) had an additional postoperative refill. In comparison, among patients who had an initial fill but did not have a family member with long-term opioid use, 28 508 (11.6%) had an additional postoperative opioid refill (P < .001 for unadjusted comparison; aOR, 2.14; 95% CI, 2.03–2.25; marginal effect, 11.2 percentage points. 95% CI, 10.5-12.0 percentage points) (Table 2 and Figure 2).
Persistent Opioid Use
Among the patients who had an initial opioid prescription fill and a family member with long-term opioid use, 453 (4.1%) had persistent opioid use compared with 5940 patients (2.4%) who had an initial fill but did not have a family member with long-term opioid use (P < .001 for unadjusted comparison; aOR, 1.54; 95% CI, 1.39–1.71; marginal effect, 1.7 percentage points; 95% CI, 1.3-2.1 percentage points) (Table 2 and Figure 2).
Secondary Analysis
When regressions were repeated to model long-term opioid use by number of parents and number of other dependents, long-term opioid use among 1 parent and 2 parents was associated with an increase in persistent opioid use compared with no parent (1 parent: aOR, 1.53; 95% CI, 1.37-1.70; 2 parents: aOR, 1.56; 95% CI, 1.07-2.28) (eTable 4 in the Supplement). Having another dependent with long-term opioid use was not significantly associated with an increase in persistent opioid use, although the direction of the association was positive (aOR, 1.50; 95% CI, 0.90-2.51). Similar patterns were seen for the association between parental long-term opioid use and both initial opioid fills and postoperative opioid refills. Having another dependent with long-term opioid use was associated with increased adjusted odds of postoperative refill only (aOR, 1.60; 95% CI, 1.20-2.14). With no family members who were long-term opioid users as the reference, the marginal effects were 4.3 percentage points (95% CI, 1.6-7.0 percentage points) for initial opioid fill, 19.1 percentage points (95% CI, 15.7-22.5 percentage points) for postoperative opioid refill, and 2.1 percentage points (95% CI, 0.01-3.7 percentage points) for persistent opioid use for patients with 2 family members with long-term opioid use and 5.0 percentage points (95% CI, 4.3-5.7 percentage points) for initial opioid fill, 10.8 percentage points (95% CI, 10.0-11.5 percentage points) for postoperative opioid refill, and 1.7 percentage points (95% CI, 1.3-2.0 percentage points) for persistent opioid use for patients with 1 family member with long-term opioid use (Figure 3A). In patients with other dependents with long-term opioid use, the marginal effects were 5.6 percentage points (95% CI, 2.0-9.2 percentage points) for initial opioid use, 10.9 percentage points (95% CI, 6.6-15.2 percentage points) for postoperative opioid refill, and 2.1 percentage points (95% CI, 0.001-4.3 percentage points) for persistent opioid use (Figure 3B).
Figure 3. Secondary Analysis Defining Exposure as Number and Role of Family Members With Long-term Opioid Use.
Multivariable regression models were used to model initial opioid fill, postoperative opioid refill, and persistent opioid use as a function of the number of members with chronic opioid use and the family member role. Parent was defined as the primary insurance holder or spouse of the primary insurance holder on the common insurance plan, and other dependent was defined as any additional dependent on the common insurance plan. Only 1 family with an initial opioid fill had more than 1 dependent with chronic opioid use, precluding this patient from analysis. Error bars indicate 95% CI.
Subgroup Analysis for Malignant Tumor–Related Long-term Opioid Use in Family Members
Among 11 016 patients with long-term opioid use in family members, 549 patients (4.9%) had a family member with a malignant tumor. In this subgroup, the risks of initial fill, postoperative refill, and prolonged use were similar compared with families without malignant tumors but were nonsignificant with the exception of postoperative refill (aOR, 1.88; 95% CI, 1.50-2.35) (eTable 5 in the Supplement).
Discussion
In this study of more than 300 000 adolescents and young adults undergoing common surgical and dental procedures across all geographic regions, having a family member with long-term opioid use was associated with a higher probability of persistent opioid use. Moreover, long-term opioid use in family members was associated with a substantial increase in initial opioid prescription fills and postoperative opioid refills, even after adjustment for patient and family member comorbidities. To date, clinical and policy-based interventions have primarily focused on individual risk factors for opioid dependence, misuse, and overdose (eg, prescription drug monitoring programs) but have not placed the individual within the context of the family. Our results suggest that additional screening and heightened prevention efforts may be needed when there is long-term opioid use among the families of young surgical patients.
Our findings are consistent with prior work16,17,28 that found that a family history of substance use is a potent risk factor for opioid misuse in relatives and that permissive family norms and exposure to substance use among parents and siblings are associated with increased risk of substance use disorders in adolescents. The association of long-term opioid use in family members with persistent opioid use is likely mediated through several pathways. For example, family members play an important role in exposures, social norms, and genetic predispositions that might place young patients at an elevated risk for complications, such as new persistent opioid use.12,13,14,15,18,29 Family members are also vital sources of information about substance use risk in children and adolescents.30 A family-oriented approach to pediatric care improves quality and outcomes across care settings and may show similar benefit in the treatment and prevention of opioid use in adolescents and young adults.31
To date, studies5,11 have primarily focused on identifying individual risk factors for opioid dependence after acute care prescribing but have not examined family-level factors. This study suggests that preoperative risk stratification for adolescents and young adults should include individual and family factors. However, challenges remain to implementation of routine screening for long-term opioid use among family members. Current screening tools, such as the Current Opioid Misuse Measure, Opioid Risk Tool, and Screening, Brief Intervention, and Referral to Treatment, are not validated in youths and may not use terms that youths understand.30,32,33 In addition, current risk assessment tools are designed to detect a family history of substance abuse28,34 rather than a family history of long-term opioid use. Finally, risk assessments assume that family history was solicited and that sensitive information was disclosed, but conversations about opioid use among family members may be uncomfortable because of the stigma of opioid use in the current era.35
To overcome these challenges, coordinated efforts by practitioners, policymakers, and researchers are critically needed. Research is needed to develop screening tools among younger populations that capture both individual and environmental risk factors. Policymakers must support evidence-based implementation of screening and facilitate access to specialized care when opioid misuse or dependence is identified. Lastly, practitioners should facilitate conversations about opioid use among family members by alleviating stigma and judgment through careful use of language.35 Like other patients at high risk for opioid-related harm, patients who screen positive for long-term opioid use in family members should receive greater anticipatory guidance regarding addiction and diversion. Similarly, primary care practitioners seeing patients in the postoperative period should also screen for signs of opioid misuse and dependence.
Limitations
This study has several limitations. First, we were unable to explain the causal mechanism between long-term opioid use in family members and persistent opioid use; however, we assessed for exposure to long-term opioid use as an indicator for clinical screening. Second, our findings were limited to adolescents and young adults undergoing specific surgical and dental procedures with private employer-based insurance. These findings likely generalize to other acute care opioid prescribing to this age group, but they may not generalize to publicly insured or uninsured patients. Our claims database did not capture opioid prescription fills paid for out-of-pocket, reimbursed by secondary insurance plans, or filled by family members not enrolled on the patient’s insurance plan. We also relied on diagnosis codes to identify comorbidities, but coding is not always reliable and may be subject to misclassification. It is also possible that results could be biased by confounders not captured in the database, such as race and socioeconomic status. In addition, we defined opioid-naive patients based on the occurrence of a prior opioid claim, but some patients classified as opioid naive through this approach may have taken opioids from other sources (eg, nonmedical use). We are also unable to determine whether opioids prescribed to patients were misused by family members. Third, our definition of persistent opioid use was based on the occurrence of an opioid prescription fill between 91 and 180 days after the procedure. Opioid prescription fills may not have been related to the initial procedure. However, family history of long-term opioid use was associated with postoperative opioid fills between 4 and 90 days after the procedure, supporting the notion that persistent opioid use occurred among patients who were transitioning to long-term opioid use.
Conclusions
Among adolescents and young adults undergoing common surgical and dental procedures, long-term opioid use among family members appears to be associated with persistent opioid use. Our findings suggest that, during the preoperative evaluation, practitioners should screen young patients undergoing procedures for long-term opioid use in their families and implement heightened efforts to prevent opioid dependence among patients with this important risk factor.
eTable 1. Inclusion and Exclusion Categories With Current Procedure Terminology (CPT) or Current Dental Terminology (CDT) Codes
eTable 2. Mental Health, Opioid Use, and Chronic Pain Disorders Among Patients and Family Members
eTable 3. Patient and Family Exposure Characteristics by Presence of Chronic Opioid Use Among Any Family Member (N = 346,251)
eTable 4. Secondary Analysis Examining Number and Role of Family Members With Chronic Opioid Use
eTable 5. Adjusted Odds Ratios for Association of Malignant and Benign Chronic Opioid Use Among Family Members
References
- 1.Gomes T, Tadrous M, Mamdani MM, Paterson JM, Juurlink DN. The burden of opioid-related mortality in the United States. JAMA Netw Open. 2018;1(2):. doi: 10.1001/jamanetworkopen.2018.0217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Seth P, Scholl L, Rudd RA, Bacon S. Overdose deaths involving opioids, cocaine, and psychostimulants—United States, 2015-2016. MMWR Morb Mortal Wkly Rep. 2018;67(12):349-. doi: 10.15585/mmwr.mm6712a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Groenewald CB, Rabbitts JA, Gebert JT, Palermo TM. Trends in opioid prescriptions among children and adolescents in the United States: a nationally representative study from 1996 to 2012. Pain. 2016;157(5):1021-1027. doi: 10.1097/j.pain.0000000000000475 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Volkow ND, McLellan TA, Cotto JH, Karithanom M, Weiss SR. Characteristics of opioid prescriptions in 2009. JAMA. 2011;305(13):1299-1301. doi: 10.1001/jama.2011.401 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Harbaugh CM, Lee JS, Hu HM, et al. Persistent opioid use among pediatric patients after surgery. Pediatrics. 2018;141(1):e20172439. doi: 10.1542/peds.2017-2439 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bennett KG, Harbaugh CM, Hu HM, et al. Persistent opioid use among children, adolescents, and young adults after common cleft operations. J Craniofac Surg. 2018;29(7):1697-1701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Alam A, Gomes T, Zheng H, Mamdani MM, Juurlink DN, Bell CM. Long-term analgesic use after low-risk surgery: a retrospective cohort study. Arch Intern Med. 2012;172(5):425-430. doi: 10.1001/archinternmed.2011.1827 [DOI] [PubMed] [Google Scholar]
- 8.Brummett CM, Waljee JF, Goesling J, et al. New persistent opioid use after minor and major surgical procedures in US adults. JAMA Surg. 2017;152(6):e170504. doi: 10.1001/jamasurg.2017.0504 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Clarke H, Soneji N, Ko DT, Yun L, Wijeysundera DN. Rates and risk factors for prolonged opioid use after major surgery: population based cohort study. BMJ. 2014;348:g1251. doi: 10.1136/bmj.g1251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sun EC, Darnall BD, Baker LC, Mackey S. Incidence of and risk factors for chronic opioid use among opioid-naive patients in the postoperative period. JAMA Intern Med. 2016;176(9):1286-1293. doi: 10.1001/jamainternmed.2016.3298 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Harbaugh CM, Nalliah RP, Hu HM, Englesbe MJ, Waljee JF, Brummett CM. Persistent opioid use after wisdom tooth extraction. JAMA. 2018;320(5):504-506. doi: 10.1001/jama.2018.9023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Elmallah RK, Ramkumar PN, Khlopas A, et al. Postoperative pain and analgesia: is there a genetic basis to the opioid crisis? Surg Technol Int. 2018;32:306-314. [PubMed] [Google Scholar]
- 13.Mistry CJ, Bawor M, Desai D, Marsh DC, Samaan Z. Genetics of opioid dependence: a review of the genetic contribution to opioid dependence. Curr Psychiatry Rev. 2014;10(2):156-167. doi: 10.2174/1573400510666140320000928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Angst MS, Phillips NG, Drover DR, et al. Pain sensitivity and opioid analgesia: a pharmacogenomic twin study. Pain. 2012;153(7):1397-1409. doi: 10.1016/j.pain.2012.02.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nielsen CS, Knudsen GP, Steingrímsdóttir ÓA. Twin studies of pain. Clin Genet. 2012;82(4):331-340. doi: 10.1111/j.1399-0004.2012.01938.x [DOI] [PubMed] [Google Scholar]
- 16.Pestka EL, Craner J, Evans M, et al. Impact of family history of substance abuse on admission opioid dose, depressive symptoms, and pain catastrophizing in patients with chronic pain. Pain Manag Nurs. 2018;19(2):115-124. doi: 10.1016/j.pmn.2017.09.007 [DOI] [PubMed] [Google Scholar]
- 17.Merikangas KR, Stolar M, Stevens DE, et al. Familial transmission of substance use disorders. Arch Gen Psychiatry. 1998;55(11):973-979. doi: 10.1001/archpsyc.55.11.973 [DOI] [PubMed] [Google Scholar]
- 18.Voepel-Lewis T, Zikmund-Fisher BJ, Smith EL, Zyzanski S, Tait AR. Parents’ preferences strongly influence their decisions to withhold prescribed opioids when faced with analgesic trade-off dilemmas for children: a prospective observational study. Int J Nurs Stud. 2015;52(8):1343-1353. doi: 10.1016/j.ijnurstu.2015.05.003 [DOI] [PubMed] [Google Scholar]
- 19.Seamans MJ, Carey TS, Westreich DJ, et al. Association of household opioid availability and prescription opioid initiation among household members. JAMA Intern Med. 2018;178(1):102-109. doi: 10.1001/jamainternmed.2017.7280 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hansen L. The Truven Health MarketScan Databases for life sciences researchers. http://content.truvenhealth.com/rs/699-YLV-293/images/2017%20MarketScan%20Databases%20for%20Life%20Sciences%20Researchers%20WP.pdf?aliId=4361930. Published 2017. Accessed September 18, 2018.
- 21.Nasseh K, Vujicic M, Glick M. The relationship between periodontal interventions and healthcare costs and utilization: evidence from an integrated dental, medical, and pharmacy commercial claims database. Health Econ. 2017;26(4):519-527. doi: 10.1002/hec.3316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Vandenbroucke JP, von Elm E, Altman DG, et al. ; STROBE Initiative . Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007;4(10):e297. doi: 10.1371/journal.pmed.0040297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Feudtner C, Feinstein JA, Zhong W, Hall M, Dai D. Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation. BMC Pediatr. 2014;14:199. doi: 10.1186/1471-2431-14-199 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.International Association for the Study of Pain IASP Task Force for the Classification of Chronic Pain in ICD-11 prepares new criteria on postsurgical and posttraumatic pain. https://www.iasp-pain.org/PublicationsNews/NewsDetail.aspx?ItemNumber=4611. Accessed March 18, 2017.
- 25.Lee JS, Hu HM, Edelman AL, et al. New persistent opioid use among patients with cancer after curative-intent surgery. J Clin Oncol. 2017;35(36):4042-4049. doi: 10.1200/JCO.2017.74.1363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bohnert AS, Ilgen MA, Trafton JA, et al. Trends and regional variation in opioid overdose mortality among Veterans Health Administration patients, fiscal year 2001 to 2009. Clin J Pain. 2014;30(7):605-612. [DOI] [PubMed] [Google Scholar]
- 27.Centers for Disease Control, US Department of Health and Human Services Calculating total daily dose of opioids for safer dosage. https://www.cdc.gov/drugoverdose/pdf/calculating_total_daily_dose-a.pdf. Accessed September 18, 2018.
- 28.Meier MH, Hall W, Caspi A, et al. Which adolescents develop persistent substance dependence in adulthood? using population-representative longitudinal data to inform universal risk assessment. Psychol Med. 2016;46(4):877-889. doi: 10.1017/S0033291715002482 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cambron C, Kosterman R, Catalano RF, Guttmannova K, Hawkins JD. Neighborhood, family, and peer factors associated with early adolescent smoking and alcohol use. J Youth Adolesc. 2018;47(2):369-382. doi: 10.1007/s10964-017-0728-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ozechowski TJ, Becker SJ, Hogue A. SBIRT-A: adapting SBIRT to maximize developmental fit for adolescents in primary care. J Subst Abuse Treat. 2016;62:28-37. doi: 10.1016/j.jsat.2015.10.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Park M, Giap TT, Lee M, Jeong H, Jeong M, Go Y. Patient- and family-centered care interventions for improving the quality of health care: a review of systematic reviews. Int J Nurs Stud. 2018;87:69-83. doi: 10.1016/j.ijnurstu.2018.07.006 [DOI] [PubMed] [Google Scholar]
- 32.Palamar JJ, Shearston JA, Cleland CM. Discordant reporting of nonmedical opioid use in a nationally representative sample of US high school seniors. Am J Drug Alcohol Abuse. 2016;42(5):530-538. doi: 10.1080/00952990.2016.1178269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Morris C, Green KE, Chimuma LL. A Rasch analysis of the Current Opioid Misuse Measure for patients with chronic pain. J Opioid Manag. 2018;14(1):9-14. doi: 10.5055/jom.2018.0424 [DOI] [PubMed] [Google Scholar]
- 34.Webster LR, Webster RM. Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the opioid risk tool. Pain Med. 2005;6(6):432-442. doi: 10.1111/j.1526-4637.2005.00072.x [DOI] [PubMed] [Google Scholar]
- 35.Botticelli MP, Koh HK. Changing the language of addiction. JAMA. 2016;316(13):1361-1362. doi: 10.1001/jama.2016.11874 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Inclusion and Exclusion Categories With Current Procedure Terminology (CPT) or Current Dental Terminology (CDT) Codes
eTable 2. Mental Health, Opioid Use, and Chronic Pain Disorders Among Patients and Family Members
eTable 3. Patient and Family Exposure Characteristics by Presence of Chronic Opioid Use Among Any Family Member (N = 346,251)
eTable 4. Secondary Analysis Examining Number and Role of Family Members With Chronic Opioid Use
eTable 5. Adjusted Odds Ratios for Association of Malignant and Benign Chronic Opioid Use Among Family Members
