Abstract
Objective:
To evaluate outpatient opioid dispensing and the incidence of persistent opioid use after ectopic pregnancy.
Methods:
This cohort study used US employer-based claims data to identify opioid-naïve individuals experiencing ectopic pregnancy from November 1, 2008 to September 30, 2015. Treatment was categorized as surgical, medical (using methotrexate), or unknown. New opioid prescriptions were defined as prescriptions filled from one week before to one week after an ectopic pregnancy treatment. For those who filled a new opioid prescription, we calculated the incidence and risk factors for persistent opioid use. Persistent opioid use was defined as having filled at least one opioid prescription both from 8 to 90 days after treatment and from 91 to 365 days after treatment.
Results:
Of the 15,338 individuals who had an ectopic pregnancy and were opioid naïve, 7,047 (45.9%, 95% Confidence Interval (CI) 45.2%, 46.7%) filled an opioid prescription at the time of treatment, of whom 4.1% (95% CI 3.6%, 4.6%) developed persistent opioid use. The risk of persistent opioid use was lower among those who had surgical versus medical treatment (3.7% and 6.8%, respectively; Relative Risk (RR) = 0.54, 95% CI 0.38, 0.77). Variables most strongly associated with persistent opioid use were a history of benzodiazepine use (RR = 1.99; 95% CI 1.43, 2.78; adjusted RR (aRR) = 1.57, 95% CI 1.11, 2.22), antidepressant use (RR = 1.91, 95% CI = 1.45, 2.53; aRR = 1.53, 95% CI 1.08, 2.18), and a pre-existing pain disorder (RR = 1.58, 95% CI 1.26, 1.99; aRR = 1.47, 95% CI = 1.16, 1.85) in the year before treatment.
Conclusion:
New opioid use is common following an ectopic pregnancy, with approximately 4% of those with new opioid use developing persistent opioid use. New pain management guidelines need to be developed to prevent persistent opioid use after ectopic pregnancy.
Précis:
New opioid use is common among opioid-naïve individuals following an ectopic pregnancy, with about 4% continuing to use opioids persistently.
Introduction
Ectopic pregnancy occurs when an embryo implants outside the uterus, most commonly in the Fallopian tube.1 In the United States, approximately 126,000 individuals experience an ectopic pregnancy each year.2,3 Ectopic pregnancies are painful, and can be life threatening if left untreated.4 Nearly two-thirds of individuals experiencing ectopic pregnancies undergo surgical treatment, which may prolong or exacerbate pain.2 Pain management after an ectopic pregnancy may include opioids; however, patterns of opioid prescribing following treatment have not been described. Given that reducing opioid overprescribing and the risk of persistent opioid use from excess exposure to opioids is a major health policy issue,5,6 examining opioid prescription patterns and the incidence of persistent opioid use after an ectopic pregnancy would inform future guidelines for pain management in this setting.
Our primary aim was to examine the incidence of filled outpatient opioid prescriptions and persistent opioid use among opioid-naïve individuals after an ectopic pregnancy. Our secondary aim was to identify risk factors for persistent opioid use among opioid-naïve individuals after ectopic pregnancy.
Methods
We conducted a cohort study of individuals who experienced an ectopic pregnancy from January 1, 2008 to September 30, 2015, identified in the IBM® MarketScan® Research Databases (MarketScan). MarketScan is a large administrative claims database of individuals covered by employer-sponsored insurance in the United States, and includes information from more than 200 U.S. commercial health plans (> 60 million lives annually).7 Within the MarketScan databases is patient-level information of inpatient and outpatient healthcare encounters (with associated diagnoses and procedures), outpatient pharmacy dispensing, and healthcare plan enrollment.7 Information on inpatient dispensing of medications was not available.
Cohort
Using approaches described by Hoover et al. and Scholes et al.,2,8 ectopic pregnancies were identified by an International Classification of Diseases, ninth edition, (ICD-9) ectopic pregnancy diagnosis code followed within a 14 day period by any of the following: an ICD-9 or Current Procedural Terminology (CPT) surgical procedure code, a second ICD-9 ectopic pregnancy diagnosis code, or medical treatment with methotrexate therapy (identified using ICD-9, Healthcare Common Procedure Coding System (HCPCS), or national drug codes).2,8 Table e1 in Appendix A contains a full list of codes used to define ectopic pregnancies. Treatment type was classified as surgical, medical, or unknown (if no surgical or methotrexate codes were identified). Those who had surgical treatment after medical treatment were classified as having had surgical treatment. When treatment type was unknown, we defined treatment date as the date of their first ectopic pregnancy diagnosis.
If an individual had more than one ectopic pregnancy, we only included data from their first ectopic pregnancy. To allow adequate analyses of comorbidities and opioid use before ectopic pregnancy, we restricted our study cohort to include only individuals with continuous medical insurance and prescription drug coverage from 1 year before and to 1 year after treatment for the ectopic pregnancy. In keeping with prior studies on persistent opioid use,9,10 we restricted our cohort to those who were opioid-naïve, defined as not having filled an opioid prescription from 365 to 8 days before their ectopic pregnancy treatment (i.e. from one year to one week before treatment). Based on these criteria, our analytic cohort comprised 15,338 individuals (see eFigure 1 in the Appendix B).
Outcomes
Using the approach described by Peahl et al.,9 we defined new opioid use as filling at least one opioid prescription from one week before to one week after treatment for an ectopic pregnancy among those who were opioid-naïve. This approach accounts for individuals who received a pre-treatment prescription for pain control or who delayed the initial prescription fill until after treatment was initiated. Among those with new opioid use, we examined the following characteristics for their first opioid prescription: number of days’ supply (1–3, ≥4 days; cut-point based on the median number of days’ supply) and daily dose, measured in oral morphine milligram equivalents (MME) (0–30, 31–45, 46–74, ≥75; cut-points based on quartiles).11
Persistent opioid use among those with new opioid use was defined using an approach described by Peahl et al.,9 which required individuals to fill at least one opioid prescription from 8 to 90 days post-treatment and fill at least one additional opioid prescription from 91 to 365 days post-treatment.
Patient Factors
Based on literature review, we selected a number of sociodemographic and clinical factors as potential risk factors of persistent opioid use.10,12,13 Potential risk factors included: age at the time of diagnosis, year of diagnosis, medical conditions in the year before treatment associated with ectopic pregnancy or opioid use (pain disorders, pelvic inflammatory disease, mental disorder, substance use disorder, and smoking) and prescriptions for psychotropic medications (benzodiazepines and antidepressants) filled in the year before treatment. Diagnostic and treatment codes used to classify these clinical factors are presented in eTable 2 in Appendix C.
Statistical Analysis
Among opioid-naïve individuals experiencing ectopic pregnancy, we first calculated the frequency and risk of new and persistent opioid use, overall and by treatment type. Modified Poisson regression models were used to quantify the relative risk (RR) of new and persistent opioid use, comparing those whose ectopic pregnancies were treated surgically or had an unknown treatment type relative to those whose ectopic pregnancies were treated medically.
For individuals with new opioid use, we used univariate and multivariate modified Poisson regression models to identify potential risk factors of persistent opioid use.
Sensitivity Analysis
In our primary analysis, opioid-naïve individuals were defined as those who did not fill an opioid prescription from 365 to 8 days before their ectopic pregnancy treatment date. Time-periods for classifying opioid-naiveté before surgery or delivery can vary, with some studies using a period shorter than one year.14–16 Therefore, we conducted a sensitivity analysis in which opioid-naïve was applied to those who did not fill an opioid prescription from 6 months to 8 days before their ectopic pregnancy treatment.16
Analyses were performed using SAS, version 9.4 (Cary, N.C.). This study uses pre-existing, de-identified data, and was exempt from review by the Stanford University Institutional Review Board.
Results
We identified 21,615 individuals who had an ectopic pregnancy and had continuous insurance coverage from one year before to one year after their treatment. After excluding those who filled at least one opioid prescription during the baseline period before their ectopic pregnancy diagnosis, the final study cohort comprised of 15,338 opioid-naïve individuals. Of these, 7,047 (45.9%; 95% CI 45.2%, 46.7%) filled an opioid prescription from one week before to one week after their ectopic pregnancy diagnosis.
Characteristics of individuals with and without new opioid use are provided in Table 1. Compared to those who did not use opioids, individuals who filled a new opioid prescription were more likely to be younger, have a pre-existing substance use disorder, smoke, and have filled a psychotropic medication prescription in the baseline period (Table 1).
Table 1.
Characteristics of individuals with and without new opioid use associated with an ectopic pregnancy
| All (n = 15,338) | With new opioid use (n = 7,047) | Without new opioid use (n = 8,291) | |
|---|---|---|---|
| n (%) | n (%) | n (%) | |
| Age | |||
| 15–24 | 2,014 (13.1) | 981 (13.9) | 1,033 (12.5) |
| 25–29 | 3,602 (23.5) | 1,694 (24.0) | 1,908 (23.0) |
| 30–34 | 5,301 (34.6) | 2,419 (34.3) | 2,882 (34.8) |
| 35–39 | 3,395 (22.1) | 1,535 (21.8) | 1,860 (22.4) |
| 40–45 | 1,026 (6.7) | 418 (5.9) | 608 (7.3) |
| Year | |||
| 2008–2009 | 2,504 (16.3) | 1,149 (16.3) | 1,355 (16.3) |
| 2010–2011 | 4,988 (32.5) | 2,262 (32.1) | 2,726 (32.9) |
| 2012–2013 | 4,475 (29.2) | 2,046 (29.0) | 2,429 (29.3) |
| 2014–2015 | 3,371 (22.0) | 1,590 (22.6) | 1,781 (21.5) |
| Diagnoses in the year before end of ectopic pregnancy | |||
| Pain Disorders | 5,404 (35.2) | 2,463 (35.0) | 2,941 (35.5) |
| Pelvic Inflammatory Disease | 1,979 (12.9) | 891 (12.6) | 1,088 (13.1) |
| Mental Disorder | 2,210 (14.4) | 1,051 (14.9) | 1,159 (14.0) |
| Substance Use Disorder | 426 (2.8) | 250 (3.6) | 176 (2.1) |
| Smoking | 485 (3.2) | 293 (4.2) | 192 (2.3) |
| Psychotropic prescriptions filled in the year before end of ectopic pregnancy | |||
| Antidepressants | 1.611 (10.5) | 808 (11.5) | 803 (9.7) |
| Benzodiazepines | 1,015 (6.6) | 488 (6.9) | 527 (6.4) |
Rates of new opioid use among those who had surgery, medical treatment, and unknown treatment were 69.8% (95% CI 68.8, 70.9), 17.5% (95% CI 16.1, 18.9), and 27.0% (95% CI 25.8, 28.2), respectively. Compared to those receiving medical treatment, the relative risk of new opioid use was nearly 4-fold (RR = 3.99, 95% CI 3.68, 4.33) and 1.5-fold (RR = 1.54, 95% CI 1.41, 1.69) higher for individuals who had surgery or unknown treatment, respectively (Table 2). Among those with new opioid use, 287 (4.1%; 95% CI 3.6%, 4.6%) developed persistent opioid use. Based on treatment modality, rates of persistent opioid use were lowest after surgical treatment (3.7%; 95% CI 3.2%, 4.3%) and highest after medical treatment (6.8%; 95% CI 4.9%, 9.4%). Compared to those receiving medical treatment, the relative risk of persistent opioid use was reduced by 46% (RR = 0.54, 95% CI 0.38, 0.77) and 35% (RR = 0.65, 95% CI 0.44, 0.98) for individuals who had surgery or unknown treatment, respectively.
Table 2.
Frequency and risk of post-ectopic pregnancy opioid use and persistent opioid use among opioid-naïve individuals
| All Individuals who were Opioid-Naïve in the Year before Treatment | New Opioid Use among Opioid-Naïve Individuals | Persistent Opioid Use among Individuals with New Opioid Use | |||||||
|---|---|---|---|---|---|---|---|---|---|
| N | Risk† (95% CI) | Risk Difference† (95% CI) | Relative Risk (95% CI) | N | Risk‡ (95% CI) | Risk Difference‡ (95% CI) | Relative Risk (95% CI) | ||
| All | 15,338 | 7,047 | 45.9 (45.2, 46.7) | - | - | 287 | 4.1 (3.6, 4.6) | - | - |
| By Treatment Type | |||||||||
| Surgical | 7,426 | 5,183 | 69.8 (68.8, 70.9) | 52.3 (50.6, 54.1) | 3.99 (3.68, 4.33) | 192 | 3.7 (3.2, 4.3) | −3.1 (−5.4, −0.9) | 0.54 (0.38, 0.77) |
| Medical | 2,850 | 498 | 17.5 (16.1, 18.9) | 0.0 (Reference) | 1.00 (Reference) | 34 | 6.8 (4.9, 9.4) | 0.0 (Reference) | 1.00 (Reference) |
| Unknown* | 5,062 | 1,366 | 27.0 (25.8, 28.2) | 9.5 (7.7, 11.4) | 1.54 (1.41, 1.69) | 61 | 4.5 (3.5, 5.7) | −2.4 (−4.8, 0.1) | 0.65 (0.44, 0.98) |
Could have been managed expectantly or have received inpatient medical treatment
Per 100 individuals experiencing ectopic pregnancy who were opioid-naïve in the year before treatment
Per 100 individuals experiencing ectopic pregnancy who were opioid-naïve in the year before treatment and had new opioid use around treatment
Persistent Opioid Use
We first examined whether, among those with new opioid use, prescription opioid characteristics increased the risk of persistent opioid use. In univariate modified Poisson regression models, daily opioid dose and the number of days’ supply of opioids were not associated with persistent opioid use (Table 3).
Table 3.
Frequency and Relative Risk of persistent opioid use among individuals with new opioid use, by characteristics of first opioid dispensation (n = 7,047)
| Characteristics of First Opioid Dispensation | N | Persistent Opioid Use (Risk*) | Risk Difference* (95% CI) | Relative Risk (95% CI) |
|---|---|---|---|---|
| Daily Dose (MME) | ||||
| 0–30 | 1,838 | 87 (4.7) | (Reference) | 1.00 (Reference) |
| 31–45 | 2,054 | 82 (4.0) | −0.7 (−2.0, 0.6) | 0.84 (0.63, 1.13) |
| 46–74 | 1,468 | 60 (4.1) | −0.7 (−2.1, 0.8) | 0.86 (0.63, 1.19) |
| ≥75 | 1,687 | 58 (3.4) | −1.3 (−2.6, 0.0) | 0.73 (0.52, 1.01) |
| Days’ Supply | ||||
| 1–3 | 3,276 | 138 (4.2) | (Reference) | 1.00 (Reference) |
| ≥4 | 3,771 | 149 (4.0) | −0.3 (−1.2, 0.7) | 0.94 (0.75, 1.18) |
Per 100 individuals experiencing ectopic pregnancy with new opioid use
In the modified Poisson regression models, factors associated with increased risk of persistent opioid use among those with new opioid use were: pain disorders (RR = 1.58, 95% CI 1.26, 1.99; adjusted RR = 1.47, 95% CI = 1.16, 1.85), antidepressant prescriptions (RR = 1.91, 95% CI = 1.45, 2.53; adjusted RR = 1.53, 95% CI 1.08, 2.18) and benzodiazepine prescriptions (RR = 1.99; 95% CI 1.43, 2.78; adjusted RR = 1.57, 95% CI 1.11, 2.22) filled in the year before treatment (Table 4). Of the 287 individuals who continued to use opioids persistently, 161 (56.1%) had at least one of these three risk factors.
Table 4.
Frequency and risk of persistent opioid use among Individuals with new opioid use (n = 7,049)
| Individual Characteristic | N | Persistent Opioid Use (Risk*) | Risk Difference* (95% CI) | Relative Risk (95% CI) | |
|---|---|---|---|---|---|
| Unadjusted | Adjusted† | ||||
| Age (years) | |||||
| 15–24 | 981 | 47 (4.8) | 1.1 (−0.5, 2.6) | 1.29 (0.91, 1.82) | 1.31 (0.93, 1.86) |
| 25–29 | 1,694 | 62 (3.7) | 0.1 (−1.2, 1.1) | 0.98 (0.72, 1.35) | 0.98 (0.72, 1.35) |
| 30–34 | 2,419 | 90 (3.7) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| 35–39 | 1,535 | 72 (4.7) | 1.0 (−0.3, 2.3) | 1.26 (0.93, 1.71) | 1.24 (0.92, 1.68) |
| 40–45 | 418 | 16 (3.8) | 0.1 (−1.9, 2.1) | 1.03 (0.61, 1.73) | 0.97 (0.58, 1.64) |
| Year | |||||
| 2008 −2009 | 1,149 | 60 (5.2) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| 2010–2011 | 2,262 | 93 (4.1) | −1.1 (−2.6, 0.4) | 0.80 (0.57, 1.08) | 0.77 (0.56, 1.06) |
| 2012–2013 | 2,046 | 82 (4.0) | −1.2 (−2.8, 0.3) | 0.77 (0.55, 1.06) | 0.75 (0.54, 1.04) |
| 2014–2015 | 1,590 | 52 (3.3) | −2.0 (−3.5, −0.4) | 0.62 (0.44, 0.90) | 0.60 (0.42, 0.86) |
| Diagnoses in the year before end of ectopic pregnancy | |||||
| Pain Disorders | |||||
| No | 4,584 | 155 (3.4) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 2,463 | 132 (5.4) | 2.0 (1.0, 3.0) | 1.58 (1.26, 1.99) | 1.47 (1.16, 1.85) |
| Pelvic Inflammatory Disease | |||||
| No | 6,156 | 252 (4.1) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 891 | 35 (3.9) | −0.2 (−1.5, 1.2) | 0.96 (0.68, 1.36) | 0.89 (0.63, 1.27) |
| Mental Disorder | |||||
| No | 5,996 | 225 (3.8) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 1,051 | 62 (5.9) | 2.2 (0.6, 3.7) | 1.57 (1.20, 2.07) | 1.02 (0.72, 1.43) |
| Substance Use Disorder | |||||
| No | 6,797 | 268 (3.9) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 250 | 19 (7.6) | 3.7 (0.3, 7.0) | 1.93 (1.23, 3.02) | 1.58 (0.97, 2.57) |
| Smoking | |||||
| No | 6,754 | 268 (4.0) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 293 | 19 (6.5) | 2.5 (−0.3, 5.4) | 1.63 (1.04, 2.56) | 1.02 (0.63, 1.63) |
| Psychotropic prescriptions filled in the year before end of ectopic pregnancy | |||||
| Antidepressants | |||||
| No | 6,239 | 230 (3.7) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 808 | 57 (7.1) | 3.4 (1.5, 5.2) | 1.91 (1.45, 2.53) | 1.53 (1.08, 2.18) |
| Benzodiazepines | |||||
| No | 6,559 | 250 (3.8) | 0.0 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Yes | 488 | 37 (7.6) | 3.8 (1.4, 6.2) | 1.99 (1.43, 2.78) | 1.57 (1.11, 2.22) |
Per 100 individuals experiencing ectopic pregnancy with new opioid use
Adjusted model includes all variable in Table 1
Sensitivity Analysis
In the sensitivity analysis we examined whether rates of new and persistent opioid use differed when we defined opioid-naïve individuals as not receiving an opioid prescription from 6 months to 8 days before the ectopic pregnancy treatment date. Using this definition, the rates of new opioid use (47.1%; 95% CI 46.4%, 47.9%) and persistent opioid use among individuals with new opioid use (5.0%; 95% CI 4.5%, 5.4%) were similar to our main findings (Table 5). By treatment group, the rate of persistent opioid use remained highest for those receiving medical treatment (8.7%; 95% CI 6.8%, 11.3%).
Table 5.
Frequency of post-ectopic pregnancy opioid use and persistent opioid use among individuals who were opioid-naïve in the six months before treatment
| All Individuals who were Opioid-Naïve 6 months before Treatment | New Opioid Use among Individuals who were Opioid-Naïve for 6 months before Treatment | Persistent Opioid Use among Individuals with New Opioid Use | |||||||
|---|---|---|---|---|---|---|---|---|---|
| N | Risk† (95% CI) | Risk Difference† (95% CI) | Relative Risk (95% CI) | N | Risk‡ (95% CI) | Risk Difference‡ (95% CI) | Relative Risk (95% CI) | ||
| All | 17,751 | 8,363 | 47.1 (46.4, 47.9) | - | - | 414 | 5.0 (4.5, 5.4) | - | - |
| By Treatment Type | |||||||||
| Surgical | 8,628 | 6,143 | 71.2 (70.3, 72.2) | 52.5 (50.8, 54.1) | 3.80 (3.53, 4.08) | 270 | 4.4 (3.9, 4.9) | −4.3 (−6.6, −2.1) | 0.50 (0.38, 0.67) |
| Medical | 3,302 | 619 | 18.8 (17.5, 20.1) | 0.0 (Reference) | 1.00 (Reference) | 54 | 8.7 (6.8, 11.3) | 0.0 (Reference) | 1.00 (Reference) |
| Unknown* | 5,821 | 1,601 | 27.5 (26.4, 28.7) | 8.8 (7.0, 10.5) | 1.47 (1.35, 1.59) | 90 | 5.6 (4.6, 6.9) | −3.1 (−5.6, −0.6) | 0.64 (0.47, 0.89) |
Could have been managed expectantly or have received inpatient medical treatment
Per 100 individuals experiencing ectopic pregnancy who were opioid-naïve in the year before treatment
Per 100 individuals experiencing ectopic pregnancy who were opioid-naïve in the year before treatment and had new opioid use around treatment
Discussion
We found that, after an ectopic pregnancy, 46% of opioid-naïve individuals filled a new opioid prescription. Among those who filled a new opioid prescription, 4% developed persistent opioid use. New opioid use was more common after surgical treatment than after medical treatment (69.8% versus 17.5%). In contrast, persistent opioid use was more common among new opioid users whose ectopic pregnancy was treated medically than surgically (6.8% and 3.7%, respectively).
Recent obstetric studies have reported rates of opioid use and persistent opioid use after vaginal or cesarean delivery. In these studies, opioid prescription rates were high after vaginal delivery (27%) and cesarean delivery (76%),9 with rates of persistent opioid ranging from 0.6% to 1.7% after vaginal delivery and from 0.4% to 2.2% after caesarean delivery.9,17,18 Because opioids are commonly overprescribed after birth, concern has been raised that this degree of exposure increases the risk of persistent opioid use.18 Several approaches have been investigated to reduce opioid overprescribing, including shared decision making and individualized prescribing.18,19 However, less attention has been given to opioid prescription patterns and persistent opioid use after treatment for an ectopic pregnancy. Our findings indicate that outpatient opioids are commonly prescribed and that the risk of persistent opioid use is likely underappreciated in this population. Given these findings and current knowledge of the extent and potential consequences of opioid overprescribing after surgery,5 patients and physicians need to be made aware and educated about the potential risk of persistent opioid use related to opioid prescription fills after treatment for an ectopic pregnancy. Stakeholders involved in providing care to individuals following an ectopic pregnancy should employ multimodal analgesia with non-opioid medications such as acetaminophen to minimize outpatient opioid exposure and over prescription.5 Interdisciplinary care coordination for those who have characteristics associated with higher risk of persistent opioid use, such as preexisting pain disorders, is essential to balance optimal pain care and conservative opioid prescribing.20
Among opioid-naïve individuals whose ectopic pregnancy was treated medically, 17.5% filled an opioid prescription. Medical treatment was associated with a higher risk of persistent opioid use among new opioid users compared with surgical treatment or unknown treatment. These findings are consistent with those from a multicenter trial in which individuals with ectopic pregnancies who received medical treatment reported more pain than those treated surgically.21 However, it is unclear why medical treatment was associated with a higher risk of persistent opioid use than surgical treatment. One possible reason is that methotrexate cannot be co-prescribed with non-steroidal anti-inflammatory drugs (NSAIDs) because of a drug interaction potentiating methotrexate toxicity.22 Therefore, with a limited choice of non-opioids for analgesia, physicians may be more likely to prescribe opioids which increases the risk of persistent opioid use. For individuals receiving methotrexate treatment, further research is needed to identify a non-NSAID based multimodal analgesic regimen that allows effective analgesia whilst minimizing opioid exposure and overprescribing. Given that pain and opioid use trajectories after injury do not overlap, and postoperative pain often continues long after opioid cessation,20 further research is needed to determine whether reducing persistent opioid use corresponds to other favorable outcomes including pain reduction.
Approximately half of the individuals in our cohort were treated surgically, with 3.7% of those who filled an opioid prescription after surgery developing persistent opioid use. This rate is slightly lower than the range for persistent opioid use (4–6%) reported after other types of laparoscopic surgery (e.g. appendectomy, cholecystectomy).10 In 2016, post-operative Enhanced Recovery After Surgery (ERAS) guidelines were released for gynecologic surgery, which include guidance on reducing post-surgery opioid use.23 The American College of Obstetricians and Gynecologists included this guidance in their recommendations for ERAS implementation, along with stepwise, multimodal pain management strategies to minimize opioid use.24 Because the guidelines were published after the last year of our study period, follow-up studies are needed to determine whether clinicians caring for individuals with ectopic pregnancy have adopted these guidelines and whether rates of new and persistent opioid use have changed over time.
In our study, the strongest risk factors for persistent opioid use were pre-existing pain disorders and antidepressant and benzodiazepine use. Consistent with these findings, preoperative exposure to these psychotropic medications has been associated with persistent opioid use after cesarean delivery and non-obstetric surgery.17,25 Pre-existing pain and psychiatric disorders, including depression, have also been linked to persistent opioid use after surgery.25 Based on our findings, physicians should be aware of the potential for persistent opioid use when prescribing outpatient opioids to individuals with a pre-existing pain, mood or anxiety disorder after an ectopic pregnancy. Our findings that daily dose and days’ supply of opioids were not associated with persistent opioid use are consistent with previous findings from a population-wide study of persistent opioid use after cesarean delivery.17 However, we recommend conservative opioid prescription practices because patients typically do not dispose of leftover opioids after surgery.25 This has important public health ramifications as leftover opioids can be potentially lead to opioid misuse or diversion.5
The use of the MarketScan database is a significant strength of our study, as it comprises data for a large contemporary cohort of individuals with ectopic pregnancy, objective information of potential drug exposures, and the ability to account for a range of potential confounders through linkages to medical claims data. However, our analysis has several limitations. First, we relied on outpatient prescription data to identify opioid use, which could result in underestimation or overestimation of actual opioid consumption due to lack of data on the actual number of consumed or diverted opioid tablets. Second, we were unable to identify ectopic pregnancy treatment type for approximately one third of our population, classified as unknown treatment. These pregnancies may have been managed expectantly, or treated with oral methotrexate in an inpatient setting. Further research using data from electronic medical records is needed to determine how these ectopic pregnancies were treated. Third, we relied on inpatient and outpatient records to identify potential predictors of persistent opioid use, including pre-existing substance use disorders, mental health conditions, and smoking. These conditions are typically underestimated in claims data.26 Additionally, the dataset does not contain information on demographic and socioeconomic factors that may be associated with persistent opioid use. Fourth, our cohort included only those with employer-based medical insurance coverage, therefore our findings may not be generalizable to individuals without insurance or who have government-assisted insurance. Fifth, we did not examine other procedures or diagnoses in the year after the ectopic pregnancy that may have influenced the risk of persistent opioid use. Lastly, definitions of persistent opioid use are inconsistent between studies.27 Although we based our definition on that used in other obstetric research, we acknowledge that the incidence of persistent opioid use may vary according to the chosen definition.
Our findings indicate that opioid use is high among opioid-naïve individuals after an ectopic pregnancy, with approximately 4% developing persistent opioid use. Factors most strongly associated with persistent opioid use were history of a pain disorder, and pre-existing antidepressant or benzodiazepine use; more than 50% of those with persistent opioid use had at least one of these three risk factors. Multifaceted efforts such as evidence-based guidelines and physician education are urgently needed to reduce excessive opioid prescribing and promote safe and effective pain management practices after an ectopic pregnancy.
Supplementary Material
Disclosure and acknowledgements:
This study was funded through a Banting Postdoctoral Fellowship (Elizabeth Wall-Wieler), a Stanford Maternal and Child Health Research Institute Postdoctoral Award (Elizabeth Wall-Wieler), and the National Institutes on Drug Abuse of the National Institutes of Health (R01DA045027 - Jennifer M. Hah; T32 DA035165 - Chelsea L. Shover). Data access for this project was provided by the Stanford Center for Population Health Sciences Data Core. The PHS Data Core is supported by a National Institutes of Health National Center for Advancing Translational Science Clinical and Translational Science Award (UL1 TR001085) and internal Stanford funding. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Footnotes
Source of work: Data were sourced from ectopic pregnancies identified in the IBM® MarketScan® Databases from November 1, 2008 to September 30, 2015.
References
- 1.Lin EP, Bhatt S, Dogra VS. Diagnostic clues to ectopic pregnancy. Radiographics 2008;28(6):1661–1671 [DOI] [PubMed] [Google Scholar]
- 2.Hoover K, Tao G, Kent C. Trends in the Diagnosis and Treatment of Ectopic Pregnancy in the United States. Obstet Gynecol 2010;115(3):495–502. [DOI] [PubMed] [Google Scholar]
- 3.Curtin SC, Abma JC, Ventura SJ, Henshaw S. Pregnancy Rates for U.S. Women Continue to Drop. NCHS Data Brief 2013;(136):1–8. [PubMed] [Google Scholar]
- 4.Committee on Practice Bulletins - Gynecology, Barnhart K, Franasiak H. ACOG Practice Bulletin No. 193: Tubal Ectopic Pregnancy. Obstet Gynecol 2018;131(3):e91–e103. [DOI] [PubMed] [Google Scholar]
- 5.Neuman MD, Bateman BT, Wunsch H. Inappropriate opioid prescription after surgery. Lancet 2019;393(10180):1547–1557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bonnie RJ, Schumacher MA, Clark DJ, Kesselheim AS. Pain management and opioid regulation: Continuing public health challenges. Am J Public Health 2019;109(1):31–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.IBM Corporation. IBM MarketScan Research Databases. Somers, NY; 2018. [Google Scholar]
- 8.Scholes D, Yu O, Raebel MA, Trabert B, Holt VL. Improving automated case finding for ectopic pregnancy using a classification algorithm. Hum Reprod 2011;26(11):3163–3168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Peahl AF, Dalton VK, Montgomery JR, Lai Y-L, Hu HM, Waljee JF. Rates of New Persistent Opioid Use After Vaginal or Cesarean Birth Among US Women. JAMA Netw Open 2019;2(7):e197863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.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). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.National Center for Injury Prevention and Control. CDC Compliation of Benzodiazepines, Muscle Relaxants, Stimulatns, Zolpidem, and Opioid Analgesics with Orl Morphine Milligram Equivalent Conversion Factors, 2017 Version. Atlanta, GA; 2017. [Google Scholar]
- 12.Lin HW, Tu YY, Lin SY, Su WJ, Lin WL, Lin WZ, et al. Risk of ovarian cancer in women with pelvic inflammatory disease: A population-based study. Lancet Oncol 2011;12(9):900–904. [DOI] [PubMed] [Google Scholar]
- 13.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(February):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Swenson CW, Kamdar NS, Seiler K, Morgan DM, Lin P, As-Sanie S. Definition development and prevalence of new persistent opioid use following hysterectomy. Am J Obstet Gynecol 2018;219(5):486.e1–486.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Brat GA, Agniel D, Beam A, Yorkgitis B, Bicket M, Homer M, et al. Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: Retrospective cohort study. BMJ 2018;360:j5790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Burke LG, Zhou X, Boyle KL, et al. Trends in Opioid Use Disorder and overdose among opioid-naive individuals receiving an opioid prescription in Massachusetts from 2011–2014. Addiction 2020;115(3):493–504. [DOI] [PubMed] [Google Scholar]
- 17.Bateman B, Franklin J, Bykov K, Avorn J, Shrank WH, Brennan TA, et al. Persistent opioid use following Cesarean delivery: patterns and predictors among opioid naïve women. Am J Obstet Gynecol 2016;215(3):353.e1–353.e18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Osmundson S, Wiese A, Min J, Hawley R, Patrick S, Griffin M. Delivery type, opioid prescribing, and the risk of persistent opioid use after delivery. Am J Obstet Gynecol 2019;220(4):405–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Prabhu M, Garry EM, Hernandez-Diaz S, MacDonald SC, Huybrechts KF, Bateman BT. Frequency of opioid dispensing after vaginal delivery. Obstet Gynecol 2018;132(2):459–465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hah JM, Cramer E, Hilmoe H, Schmidt P, McCue R, Trafton J, et al. Factors Associated With Acute Pain Estimation, Postoperative Pain Resolution, Opioid Cessation, and Recovery: Secondary Analysis of a Randomized Clinical Trial. JAMA Netw Open 2019;2(3):e190168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nieuwkerk PT, Hajenius PJ, Ankum WM, Van Der Veen F, Wijker W, Bossuyt PMM. Systemic methotrexate therapy versus laparoscopic salpingostomy in patients with tubal pregnancy. Part I. Impact on patients’ health-related quality of life. Fertil Steril 1998;70(3):511–517. [DOI] [PubMed] [Google Scholar]
- 22.Svanström H, Lund M. Concomitant use of low - dose methotrexate and NSAIDs and the risk of serious adverse events among patients with rheumatoid arthritis. Pharmacoepidemiol Drug Saf 2018;(April):885–893. [DOI] [PubMed] [Google Scholar]
- 23.Nelson G, Altman AD, Nick A, Meyer LA, Ramirez PT, Achtari C, et al. Guidelines for postoperative care in gynecologic/oncology surgery: Enhanced Recovery after Surgery (ERAS®) Society recommendations - Part II. Gynecol Oncol 2016;140(2):323–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.American College of Obstetricians and Gynecologists. Perioperative pathways: enhanced recovery after surgery. ACOG Committee Opinion No. 750. American. Obstet Gynecol 2018;132(3):e120–e130. [DOI] [PubMed] [Google Scholar]
- 25.Hah JM, Bateman BT, Ratliff J, Curtin C, Sun E. Chronic Opioid Use After Surgery. Anesth Analg. 2017;125(5):1733–1740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kim H, Smith E, Stano C, Ganoczy D, Zivin K, Walters H, et al. Validation of key behaviourally based mental health diagnoses in administrative data: suicide attempt, alcohol abuse, illicit drug abuse and tobacco use. BMC Health Serv Res 2012;12(18). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jivraj NK, Raghavji F, Bethell J, Wijeysundera DN, Ladha KS, Bateman BT, et al. Persistent postoperative opioid use: A systematic literature search of definitions and population-based cohort study. Anesthesiology 2020;[Epub ahead of print]. doi: 10.1097/aln.0000000000003265 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.