Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Oct 1.
Published in final edited form as: Am J Obstet Gynecol. 2020 Mar 23;223(4):566.e1–566.e13. doi: 10.1016/j.ajog.2020.03.020

New persistent opioid use after acute opioid prescribing in pregnancy: a nationwide analysis

Alex F Peahl 1,2,3, Daniel M Morgan 1,2, Vanessa K Dalton 1,2,4, Kara Zivin 1,2,5, Yen-Ling Lai 6, Hsou Mei Hu 6, Elizabeth Langen 1, Lisa Kane Low 1,2,7, Chad M Brummett 6,8, Jennifer F Waljee 4,6,9, Melissa E Bauer 7
PMCID: PMC7508788  NIHMSID: NIHMS1578864  PMID: 32217114

Abstract

Objective:

To evaluate the association between opioid prescribing during pregnancy and new persistent opioid use in the year following delivery.

Study Design:

This nationwide retrospective cohort study included patients ages 12–55 in Optum’s de-identified Clinformatics® Data Mart Database, undergoing vaginal birth (VB) or cesarean birth (CB) from 2008–2016, with continuous enrollment from 2-years before birth to 1-year post-discharge. Women were included if they were opioid-naive in pregnancy (i.e. did not fill an opioid prescription 2-years to 9-months before delivery) and did not undergo a procedure within the year after discharge. The exposure was filling an opioid prescription in pregnancy.

The primary outcome was new persistent opioid use (NPOU), defined as a pharmacy claim for ≥1 opioid prescription between 4–90 days post-discharge and ≥1 prescription between 91–365 days post-discharge. Clinical and demographic covariates were included. Analyses included descriptive statistics and multivariable logistic regression, adjusting for clinical and demographic covariates.

Results:

Of 158,425 childbirths identified, 101,013 (63.8%) were VB and 57,412 (36.2%) were CB. Among all patients, 6.0% (9,429) filled an opioid prescription during pregnancy. The factors associated with filling an opioid in pregnancy were having a non-delivery procedure in pregnancy [adjusted OR (aOR) 9.60, 95% CI (8.81–10.47)] and having an emergency room visit during pregnancy [aOR 2.48, 95% CI (2.37–2.59)]. Of women who received an opioid in pregnancy, 4% (379) developed NPOU. The factors most associated with NPOU were receiving an opioid prescription during pregnancy [aOR 3.45 95% CI (3.04–3.92)] and filling a peripartum opioid prescription (1-week prior to 3-days post-discharge) [aOR 2.28, 95% CI (2.02–2.57)]. Though having a procedure during pregnancy was associated with increased receipt of an opioid prescription, it was also associated with reduced NPOU [aOR 0.72, 95% CI 0.52–0.99)].

Conclusion:

Women who receive an opioid prescription during pregnancy are more likely to experience new persistent opioid use. Maternity care providers must balance pain management in pregnancy with potential risks of opioids.

Keywords: pregnancy, vaginal delivery, cesarean delivery, opioid, pain management

Introduction:

As maternity care providers have recognized the role of peripartum opioid prescribing in the opioid epidemic, reducing opioid prescribing following delivery has become a national priority.1–4 This increased focus is warranted, as 1 in 75 women in the United States who fill an opioid prescription in the peripartum period will continue filling prescriptions up to one year postpartum.5 In fact, exposure to postpartum opioids has been linked to new persistent use after delivery, independent of the type of birth (vaginal vs. cesarean), suggesting the risk is inherent to the opioid prescription.5

Less is known, however, about the long-term consequences of acute opioid prescribing during pregnancy, outside of the delivery episode. It is possible that opioid prescribing for pain in pregnancy could have a similar effect as postpartum opioid prescribing. Women may experience pain from physiologic changes of pregnancy (e.g. round ligament pain), recurrence of chronic conditions (e.g. back pain), or pathologic pain related to new conditions (e.g. appendicitis).6–9 Pain management in pregnancy must balance risks to both the pregnant women and fetus. As opioid-alternatives like non-steroidal anti-inflammatory medications (NSAIDS), may cause fetal complications including miscarriage and renal dysfunction,10,11 pregnant patients may be more likely to receive opioids because of potential differences in safety profile during pregnancy. However, there is a growing body of evidence that acute opioid exposure can lead to long-term use.12,13

In order to understand the effects of opioid prescribing during maternity care, we assessed rates of opioid prescribing during pregnancy and their impact on long-term opioid use following delivery in opioid-naïve patients.

Methods:

In this retrospective cohort study, we analyzed claims from a national single private-payer including medical and prescription drug coverage, aggregated in Optum’s de-identified Clinformatics® Data Mart Database. We included women ages 12–55 who underwent vaginal or cesarean birth from 2008–2016, with continuous enrollment from 2-years before birth to 1-year after discharge. We selected patients with 2-years of continuous enrollment prior to birth in order to identify women who were opioid-naïve in the year prior to pregnancy. Similarly, we required an additional year of enrollment postpartum to assess opioid use in the year following delivery. Women were excluded if they 1) filled an opioid prescription from 2-years to 9-months before delivery admission (i.e. were not opioid-naïve in the year before pregnancy); 2) underwent a procedure as identified by anesthesia codes within 1-year of discharge, or 3) had an index hospitalization length of stay (defined as time from cesarean delivery to discharge) of >7 days, as these deliveries were not thought to represent routine care. If patients had multiple deliveries in the specified time period, only the first birth was included.

The main exposure in this analysis was filling an opioid prescription in pregnancy. The primary outcome was new persistent opioid use (NPOU) after vaginal or cesarean delivery, defined as at least two pharmacy claims after discharge: ≥1 opioid prescription between 4–90 days post-discharge and ≥1 prescription between 91–365 days post-discharge.5,14 This definition of NPOU is consistent with prior work, and represents a time when patients are expected to have recovered from their initial delivery episode.

To assess the factors associated with filling an opioid prescription during pregnancy and developing NPOU, we included demographic and clinical characteristics as covariates in adjusted models. Demographic variables included age, race, education, region, and year of delivery. Clinical characteristics included pregnancy comorbidities as measured by the Bateman comorbidity index15, delivery type (vaginal vs. cesarean), and delivery specific complications/additional procedures defined by billing codes for both vaginal delivery (e.g. prolonged labor, bilateral tubal ligation) and cesarean delivery (e.g. unscheduled delivery and repeat cesarean). Characteristics of opioid exposure during pregnancy included receipt of an opioid prescription during pregnancy and oral morphine equivalents prescribed. We also report if patients received a peripartum opioid prescription (defined as 1-week prior to 3-days after discharge), as this has previously been independently associated with NPOU.5 Health care utilization measures including emergency department visits, inpatient hospitalizations, and procedures in pregnancy (as defined by non-delivery anesthesia codes) were also included as dichotomous variables. To determine the diagnoses associated with opioid prescriptions, we reported the diagnosis codes from encounters 3-days prior to prescription fills, with a frequency of ≥50 across the included population.

Analyses included descriptive statistics for cohort characteristics, rates of opioid prescription fills in pregnancy, diagnoses associated with opioid prescriptions, and rates of NPOU. Multivariable logistic regression models were used to separately estimate the adjusted odds ratio (aOR) for filling an opioid prescription in pregnancy and developing NPOU. We also conducted a sensitivity analysis using two additional definitions of NPOU from the literature: 1) Brummett et al defined NPOU in patients undergoing major and minor procedures as ≥2 opioid prescriptions filled after the perioperative period, ≥1 prescription before 90 days, and ≥1 prescription between 90–180 days post-discharge; and 2) Bateman et al defined NPOU as ≥1 opioid prescription filled in at least 4 of 12 months after the perioperative period to 365 days after discharge.16,17 All statistical analyses were performed using SAS version 9.4. Statistical significance was set at P<0.05 with 2-sided tests. This study was deemed exempt by the University of Michigan Institutional Review Board.

Results:

We identified 988,036 women who underwent vaginal or cesarean delivery from 2008–2016, of whom, 158,425 met inclusion criteria [101,013 (63.8%) vaginal deliveries; 57,412 (36.2%) cesarean deliveries]. [Figure 1] Descriptive data for the cohort is presented in Table 1. Among all deliveries, 6.0% (9,429) of women filled an opioid in pregnancy [vaginal delivery, 5.6% (5,619); cesarean delivery, 6.6% (3,810)]. Of patients who filled an opioid prescription during pregnancy, the median OMEs per patient was 108 (IQR 81–146), and 16% filled ≥2 prescriptions during pregnancy (range 1–22 prescriptions in pregnancy). Only 1.8% (2,871) of patients in the cohort had a procedure with anesthesia during pregnancy.

Figure 1:

Figure 1:

Open in a new tab

Flowchart of Patient Inclusions and Exclusions

Table 1:

Characteristics of patients undergoing vaginal or cesarean delivery, with and without new persistent opioid use

Vaginal Delivery (n=101,013) Persistent Opioid Use(n=724) No Persistent Opioid Use (n=100,28 9) P* Value Cesarean Delivery (n=57,412) Persiste nt Opioid Use (n=882) No Persistent Opioid Use (n=56,530) P* Value Vaginal and Cesarean Delivery (n=158,425) Persistent Opioid Use (n=1,606) No persistent Opioid Use (n=156,819) P* Value
n % n % n % n % n % n % n % n % n %
Age <0.001 <0.001 <0.001
 <20 3802 3.8 58 8.0 3744 3.7 1131 2.0 40 4.5 1091 1.9 4933 3.1 98 6.1 4835 3.1
 20–29 34423 34.1 312 43.1 34111 34.0 14916 26.0 276 31.3 14640 25.9 49339 31.1 588 36.6 48751 31.1
 30–39 58153 57.6 324 44.8 57829 57.7 36454 63.5 504 57.1 35950 63.6 94607 59.7 828 51.6 93779 59.8
 >=40 4635 4.6 30 4.1 4605 4.6 4911 8.6 62 7.0 4849 8.6 9546 6.0 92 5.7 9454 6.0
Race <0.001 <0.001 <0.001
 White 50032 49.5 75 51.8 49657 49.5 26641 46.4 70 53.3 26171 46.3 76673 48.4 845 52.6 75828 48.4
 Black 6561 6.5 74 10.2 6487 6.5 4524 7.9 91 10.3 4433 7.8 11085 7.0 165 10.3 10920 7.0
 Hispanic 10361 10.3 66 9.1 10295 10.3 6466 11.3 96 10.9 6370 11.3 16827 10.6 162 10.1 16665 10.6
 Asian 7621 7.5 24 3.3 7597 7.6 4794 8.4 18 2.0 4776 8.5 12415 7.8 42 2.6 12373 7.9
 Unknown 26438 26.2 185 25.6 26253 26.2 14987 26.1 207 23.5 14780 26.2 41425 26.2 392 24.4 41033 26.2
Region <0.001 <0.001 <0.001
 Northeast 11499 11.4 33 4.6 11466 11.4 7201 12.5 49 5.6 7152 12.7 18700 11.8 82 5.1 18618 11.9
 Midwest 27182 26.9 162 22.4 27020 26.9 12732 22.2 186 21.1 12546 22.2 39914 25.2 348 21.7 39566 25.2
 South 40142 39.7 362 50.0 39780 39.7 26095 45.5 70 53.3 25625 45.3 66237 41.8 832 51.8 65405 41.7
 West 21876 21.7 165 22.8 21711 21.7 11238 19.6 176 20.0 11062 19.6 33114 20.9 341 21.2 32773 20.9
 Unknown 314 0.3 2 0.3 312 0.3 146 0.3 1 0.1 145 0.3 460 0.3 3 0.2 457 0.3
Education Level <0.001 <0.001 <0.001
 Less than 12th Grade 555 0.6 6 0.8 549 0.6 325 0.6 7 0.8 318 0.6 880 0.6 13 0.8 867 0.6
 HighSchool 18815 18.6 196 27.1 18619 18.6 11350 19.8 240 27.2 11110 19.7 30165 19.0 436 27.2 29729 19.0
Diploma
 Less than Bachelor Degree 53682 53.1 401 55.4 53281 53.1 29826 52.0 467 53.0 29359 51.9 83508 52.7 868 54.1 82640 52.7
 Bachelor Degree Plus 27105 26.8 116 16.0 26989 26.9 15364 26.8 161 18.3 15203 26.9 42469 26.8 277 17.3 42192 26.9
 Unknown 856 0.9 5 0.7 851 0.9 547 1.0 7 0.8 540 1.0 1403 0.9 12 0.8 1391 0.9
Household Income Range <0.001 <0.001 <0.001
 <$40K 9192 9.1 97 13.4 9095 9.1 9192 9.1 97 13.4 9095 9.1 9192 9.1 97 13.4 9095 9.1 9192 9.1 97
 $40K-$49K 4716 4.7 41 5.7 4675 4.7 4716 4.7 41 5.7 4675 4.7 4716 4.7 41 5.7 4675 4.7 4716 4.7 41
 $50K−$59K 5040 5.0 52 7.2 4988 5.0 5040 5.0 52 7.2 4988 5.0 5040 5.0 52 7.2 4988 5.0 5040 5.0 52
 $60K− 7932 7.9 55 7.6 7877 7.9 7932 7.9 55 7.6 7877 7.9 7932 7.9 55 7.6 7877 7.9 7932 7.9 55
$74K 3 7
 $75K-$99K 12102 12.0 79 10.9 12023 12.0 6813 11.9 106 12.0 6707 11.9 18915 11.9 185 11.5 18730 11.9
 $100K+ 39219 38.8 218 30.1 39001 38.9 21822 38.0 264 29.9 21558 38.1 61041 38.5 482 30.0 60559 38.6
 Unknown 22812 22.6 182 25.1 22630 22.6 13098 22.8 208 23.6 12890 22.8 35910 22.7 390 24.3 35520 22.7
Delivery Year <0.001 <0.001 <0.001
 2008 14326 14.2 102 14.1 14224 14.2 7662 13.4 147 16.7 7515 13.3 21988 13.9 249 15.5 21739 13.9
 2009 14883 14.7 133 18.4 14750 14.7 8567 14.9 168 19.1 8399 14.9 23450 14.8 301 18.7 23149 14.8
 2010 13054 12.9 123 17.0 12931 12.9 7827 13.6 146 16.6 7681 13.6 20881 13.2 269 16.8 20612 13.1
 2011 11760 11.6 90 12.4 11670 11.6 6907 12.0 110 12.5 6797 12.0 18667 11.8 200 12.5 18467 11.8
 2012 10931 10.8 72 9.9 10859 10.8 6380 11.1 106 12.0 6274 11.1 17311 10.9 178 11.1 17133 10.9
 2013 9608 9.5 66 9.1 9542 9.5 5441 9.5 70 7.9 5371 9.5 15049 9.5 136 8.5 14913 9.5
 2014 9250 9.2 52 7.2 9198 9.2 4927 8.6 51 5.8 4876 8.6 14177 9.0 103 6.4 14074 9.0
 2015 8989 8.9 52 7.2 8937 8.9 4967 8.7 49 5.6 4918 8.7 13956 8.8 101 6.3 13855 8.8
 2016 8212 8.1 34 4.7 8178 8.2 4734 8.3 35 4.0 4699 8.3 12946 8.2 69 4.3 12877 8.2
Tobacco use history 5695 5.6 91 12.6 5604 5.6 <0.001 3704 6.5 120 13.6 3584 6.3 <0.001 9399 5.9 211 13.1 9188 5.9 <0.001
Mental Health Disorder
 Adjustment 3514 3.5 44 6.1 3470 3.5 <0.001 2119 3.7 45 5.1 2074 3.7 0.025 5633 3.6 89 5.5 5544 3.5 <0.001
 Anxiety 7895 7.8 121 16.7 7774 7.8 <0.001 4657 8.1 141 16.0 4516 8.0 <0.001 12552 7.9 262 16.3 12290 7.8 <0.001
 Mood 6369 6.3 116 16.0 6253 6.2 <0.001 3772 6.6 121 13.7 3651 6.5 <0.001 10141 6.4 237 14.8 9904 6.3 <0.001
 Suicide or self-harm 282 0.3 8 1.1 274 0.3 <0.001 87 0.2 1 0.1 86 0.2 0.769 369 0.2 9 0.6 360 0.2 0.006
 Personality 176 0.2 5 0.7 171 0.2 <0.001 71 0.1 1 0.1 70 0.1 0.930 247 0.2 6 0.4 241 0.2 0.026
 Disruptive 1588 1.6 34 4.7 1554 1.6 <0.001 828 1.4 25 2.8 803 1.4 <0.001 2416 1.5 59 3.7 2357 1.5 <0.001
 Psychosis 156 0.2 5 0.7 151 0.2 <0.001 74 0.1 3 0.3 71 0.1 0.078 230 0.2 8 0.5 222 0.1 <0.001
 Alcohol or substance abuse 1874 1.9 45 6.2 1829 1.8 <0.001 1145 2.0 48 5.4 1097 1.9 <0.001 3019 1.9 93 5.8 2926 1.9 <0.001
 Other mental disorder 2664 2.6 47 6.5 2617 2.6 <0.001 1551 2.7 52 5.9 1499 2.7 <0.001 4215 2.7 99 6.2 4116 2.6 <0.001
Pain Disorders
 Arthritis 27381 27.1 276 38.1 27105 27.0 <0.001 16476 28.7 309 35.0 16167 28.6 <0.001 43857 27.7 585 36.4 43272 27.6 <0.001
 Back 20065 19.9 216 29.8 19849 19.8 <0.001 11587 20.2 248 28.1 11339 20.1 <0.001 31652 20.0 464 28.9 31188 19.9 <0.001
 Neck 9062 9.0 90 12.4 8972 9.0 0.001 5409 9.4 112 12.7 5297 9.4 <0.001 14471 9.1 202 12.6 14269 9.1 <0.001
 Other Pain 23257 23.0 249 34.4 23008 22.9 <0.001 14358 25.0 315 35.7 14043 24.8 <0.001 37615 23.7 564 35.1 37051 23.6 <0.001
Bateman’s Comorbidity Index 0.062 0.265 <0.001
 0 57407 56.8 372 51.4 57035 56.9 22197 38.7 326 37.0 21871 38.7 79604 50.3 698 43.5 78906 50.3
 1 23043 22.8 177 24.5 22866 22.8 14715 25.6 214 24.3 14501 25.7 37758 23.8 391 24.4 37367 23.8
 2 12302 12.2 103 14.2 12199 12.2 9997 17.4 167 18.9 9830 17.4 22299 14.1 270 16.8 22029 14.1
 3 4595 4.6 38 5.3 4557 4.5 5121 8.9 79 9.0 5042 8.9 9716 6.1 117 7.3 9599 6.1
 4 1911 1.9 19 2.6 1892 1.9 2521 4.4 39 4.4 2482 4.4 4432 2.8 58 3.6 4374 2.8
 >=5 1755 1.7 15 2.1 1740 1.7 2861 5.0 57 6.5 2804 5.0 4616 2.9 72 4.5 4544 2.9
Care Utilization in Pregnancy
AnyEmergency room visit 22203 22.0 247 34.1 21956 21.9 <0.001 13829 24.1 302 34.2 13527 23.9 <0.001 36032 22.7 549 34.2 35483 22.6 <0.001
Any non-delivery 3456 3.4 54 7.5 3402 3.4 <0.001 2605 4.5 59 6.7 2546 4.5 0.002 6061 3.8 113 7.0 5948 3.8 <0.001
Inpatient hospitalization
Opioid filled during pregnancy 5619 5.6 179 24.7 5440 5.4 <0.001 3810 6.6 200 22.7 3610 6.4 <0.001 9429 6.0 379 23.6 9050 5.8 <0.001
Quartile of the first Opioid prescription (mg of OME equivalent) <0.001 <0.001 <0.001
 <75 96567 95.6 577 79.7 95990 95.7 54358 94.7 711 80.6 53647 94.9 150925 95.3 1288 80.2 149637 95.4
 75–108 1656 1.6 51 7.0 1605 1.6 1097 1.9 62 7.0 1035 1.8 2753 1.7 113 7.0 2640 1.7
 108–150 804 0.8 26 3.6 778 0.8 579 1.0 26 3.0 553 1.0 1383 0.9 52 3.2 1331 0.9
 >=150 1986 2.0 70 9.7 1916 1.9 1378 2.4 83 9.4 1295 2.3 3364 2.1 153 9.5 3211 2.1
AnyAnesthesiaprocedure 1443 1.4 15 2.1 1428 1.4 0.143 1428 2.5 29 3.3 1399 2.5 0.124 2871 1.8 44 2.7 2827 1.8 0.005
Delivery Characteristics
Hospital Length of Stay for delivery 0.376 0.094 <0.00>1
 <=3 days 97310 96.3 693 95.7 96617 96.3 33831 58.9 544 61.7 33287 58.9 131141 82.8 1237 77.0 129904 82.8
 4–7 days 3703 3.7 31 4.3 3672 3.7 23581 41.1 338 38.3 23243 41.1 27284 17.2 369 23.0 26915 17.2
Opioid filled duringperipartumperiod 25943 25.7 382 52.8 25561 25.5 <0.001 42899 74.7 753 85.4 42146 74.6 <0.001 68842 43.5 1135 70.7 67707 43.2 <0.001
Open in a new tab

Patients were more likely to receive an opioid prescription during pregnancy if they received non-delivery anesthesia [aOR=9.60, 95% CI (8.81–10.47)], presented for an emergency room visit [aOR=2.48, 95% CI (2.37–2.59)] or underwent an inpatient hospitalization [aOR=1.92, 95% CI (1.78–2.08)]. Women who used tobacco [aOR=1.48, 95% CI (1.37–1.60)], had comorbid psychiatric diagnoses such as mood disorders [aOR=1.23, 95% CI (1.13–1.33)], and had comorbid pain disorders such as back pain [aOR=1.57, 95% CI (1.48–1.65)], were also more likely to receive an opioid prescription than women without these characteristics. [Table 2] The most common diagnoses associated with receiving an opioid in pregnancy were abdominal pain [4,278(14.5%)], urinary tract infections (UTI) [1,997(6.8%)], and back pain [1,470 (6.3%)]. Diagnoses associated with opioid prescriptions were unspecified for 10,071 (34.2%).

Table 2.

Logistic regression model for filling an opioid prescription during pregnancy

Characteristic Vaginal Delivery Cohort (n=101,013) Cesarean Delivery Cohort (n=57,412) VD & CD Combined Cohort (n=158,425)
Adjusted OR 95% CI P Value Adjusted OR 95% CI P Value Adjusted OR 95% CI P Value
Age
 20–29 (reference) 1 NA NA 1 NA NA 1 NA NA
 <20 0.88 (0.77- 0.081 0.88 (0.69- 0.281 0.88 (0.78- 0.035
 30–39 0.93 (0.87- 0.023 0.93 (0.86- 0.081 0.94 (0.89- 0.008
 >=40 0.92 (0.79- 0.280 0.97 (0.84- 0.683 0.96 (0.86- 0.403
Race
 White (reference) 1 NA NA 1 NA NA 1 NA NA
 Black 1.06 (0.96- 0.260 0.99 (0.88- 0.900 1.04 (0.96- 0.389
 Hispanic 0.77 (0.69- <.001 0.79 (0.70- <.001 0.78 (0.72- <.001
 Asian 0.84 (0.74- 0.008 0.81 (0.70- 0.005 0.83 (0.76- <.001
 Unknown 0.93 (0.87- 0.044 0.92 (0.84- 0.046 0.93 (0.88- 0.006
Region
 Northeast (reference) 1 NA NA 1 NA NA 1 NA NA
 Midwest 1.93 (1.70- <.001 1.81 (1.57- <.001 1.87 (1.70- <.001
 South 2.29 (2.04- <.001 2.15 (1.89- <.001 2.23 (2.05- <.001
 West 2.27 (2.01- <.001 2.07 (1.79- <.001 2.18 (1.98- <.001
 Unknown 3.50 (2.23- <.001 1.27 (0.51- 0.606 2.66 (1.78- <.001
Education Level
 Less than Bachelor Degree 1 NA NA 1 NA NA 1 NA NA
 Less than 12th Grade 0.96 (0.65- 0.821 1.11 (0.70- 0.650 1.02 (0.76- 0.890
 High School Diploma 1.07 (0.99- 0.077 1.09 (0.99- 0.071 1.08 (1.02- 0.010
 Bachelor Degree Plus 0.84 (0.78- <.001 0.95 (0.87- 0.206 0.88 (0.83- <.001
 Unknown 0.93 (0.68- 0.628 0.88 (0.60- 0.508 0.91 (0.72- 0.449
Delivery Year
  2008 1 NA NA 1 NA NA 1 NA NA
  2009 0.99 (0.89- 0.796 1.01 (0.89- 0.856 1.00 (0.92- 0.949
  2010 1.03 (0.93- 0.600 1.06 (0.93- 0.372 1.04 (0.96- 0.295
  2011 0.98 (0.88- 0.741 0.94 (0.82- 0.322 0.97 (0.89- 0.392
  2012 0.83 (0.74- 0.001 0.91 (0.80- 0.168 0.86 (0.79- <.001
  2013 0.83 (0.74- 0.001 0.78 (0.67- 0.001 0.81 (0.74- <.001
  2014 0.60 (0.53- <.001 0.64 (0.55- <.001 0.62 (0.56- <.001
  2015 0.55 (0.48- <.001 0.59 (0.50- <.001 0.57 (0.51- <.001
  2016 0.53 (0.46- <.001 0.56 (0.47- <.001 0.54 (0.49- <.001
Tobacco Use 1.48 (1.34- <.001 1.46 (1.30- <.001 1.48 (1.37- <.001
Psychiatric diagnoses
 Adjustment disorder 1.06 (0.92- 0.455 1.03 (0.87- 0.763 1.04 (0.94- 0.443
 Anxiety disorder 1.04 (0.94- 0.483 1.18 (1.05- 0.005 1.10 (1.02- 0.020
 Mood disorder 1.27 (1.14- <.001 1.17 (1.03- 0.015 1.23 (1.13- <.001
 Substance use disorder 1.16 (0.98- 0.088 1.29 (1.06- 0.013 1.21 (1.06- 0.004
Pain Condition
 Arthritis 1.14 (1.07- <.001 1.13 (1.04- 0.003 1.13 (1.08- <.001
 Back 1.57 (1.46- <.001 1.56 (1.43- <.001 1.57 (1.48- <.001
 Neck 0.95 (0.86- 0.288 0.90 (0.80- 0.073 0.93 (0.87- 0.054
 Other pain 1.44 (1.35- <.001 1.47 (1.37- <.001 1.45 (1.38- <.001
Bateman’s Comorbidity Index 1.02 (1.00- 0.045 1.02 (1.00- 0.077 1.03 (1.01- 0.002
Emergency room visit during 2.59 (2.44- <.001 2.31 (2.15- <.001 2.48 (2.37- <.001
Inpatient during pregnancy 2.07 (1.87- <.001 1.73 (1.54- <.001 1.92 (1.78- <.001
Any Anesthesia procedure during 10.96 (9.72- <.001 8.23 (7.27- <.001 9.60 (8.81- <.001
Open in a new tab

Of patients who received an opioid in pregnancy, 4.0% (379) developed new persistent opioid use. Of those undergoing vaginal birth who received an opioid during pregnancy, 3.2% (179) had postpartum NPOU. Among women undergoing cesarean delivery who received an opioid during pregnancy, 5.2% (200) had postpartum NPOU. Rates of NPOU after vaginal delivery and cesarean delivery in women who did not receive an opioid prescription in pregnancy were 0.6% (545) and 1.3% (682) respectively.

We identified several risk factors associated with new persistent opioid use. [Figure 2] Women were more likely to develop NPOU if they received an opioid during pregnancy [aOR 3.45 95% CI (3.04–3.92)] or filled a peripartum opioid prescription [aOR 2.28, 95% CI (2.02–2.57)]. While having an emergency department visit during pregnancy was associated with NPOU [aOR 1.19, 95% CI (1.06–1.33), inpatient hospitalization was not [aOR 1.15, 95% CI (0.94–1.42)]. Though undergoing a procedure in pregnancy was the factor most strongly associated with filling an opioid prescription, it was associated with a decreased rate of NPOU after delivery [aOR 0.72, 95% CI 0.52–0.99]. Patients who had lower education, history of substance abuse, and cesarean delivery were also more likely to have NPOU.

Figure 2:

Figure 2:

Open in a new tab

Plot of adjusted odds ratios of characteristics associated with new persistent opioid use after delivery

NPOU after delivery was higher in patients who filled an opioid prescription during pregnancy across all definitions of NPOU included in the sensitivity analysis. Even using the most conservative definition of new persistent opioid use16, rates were higher in those with opioid use during pregnancy compared to those without (vaginal and cesarean delivery combined: 1.2% (115) vs. 0.1% (176); vaginal delivery 1.2% (70) vs. 0.1% (96); cesarean delivery 1.2% (45) vs. 0.1% (80)) [p<0.01 for all comparisons].

Structure Discussion/Comment:

Principal Findings:

In our study, receipt of an opioid prescription during pregnancy was independently associated with new persistent opioid use in the year following delivery. Undergoing a procedure with anesthesia was most strongly associated with receiving an opioid prescription during pregnancy, but with lower rates of NPOU. In contrast, emergency department visits were associated with both receipt of an opioid prescription in pregnancy and NPOU. We also confirmed factors previously associated with NPOU, including prescription factors (e.g. filling a peripartum prescription) and individual factors (e.g. younger age, lower education, race, tobacco use, and psychiatric and pain diagnoses).5,14,17 Going forward, future efforts targeting opioid reduction throughout pregnancy, in addition to the peripartum period, are critical to reduce the long-term harms of opioid exposure.18

Results:

Previous estimates suggest 14–22% of patients fill opioid prescriptions during pregnancy.19,20 We only included women in our study who were opioid-naive in the year before pregnancy and had 3 years of continuous enrollment to isolate the independent effect of opioid prescribing during pregnancy. These criteria explain the lower rates of opioid prescribing seen in our population when compared to previous groups. Thus, our findings likely underestimate national rates of NPOU, as prior work has demonstrated higher rates of prescribing for patients who have Medicaid or are not opioid-naive prior to pregnancy.19–21 Thus, almost 1 in 5 women in the general population are exposed to opioids during pregnancy, a particularly vulnerable period often associated with stress and anxiety that may further potentiate the risk of prolonged use.18

Clinical Implications:

There are many potential consequences of opioid prescribing in pregnancy for both women and their children. For women who are opioid-naive, opioid prescribing can be associated with NPOU and greater risk for opioid harms in the future.5,16 Additionally, some data suggest a possible increased risk of neural tube defects, congenital heart defects, and gastroschisis in fetuses exposed to opioids during the first trimester.22,23 Exposure to opioids near the time of birth may also contribute to neonatal abstinence syndrome (NAS), NICU admissions, and potential neurodevelopmental effects.24,25,26. While adequate pain management must be a priority, approaches should also balance the potential maternal and fetal risks and benefits of treatment.

Research Implications:

While there are now clear recommendations for post-delivery prescribing for pregnant women, current recommendations for pain management in pregnancy provide little guidance on appropriate management of acute pain.3,6 There are several reasons for this omission: 1) efforts to reduce post-operative prescribing in other surgical specialties have paved the way for post-delivery recommendations but have not been as well developed for prenatal pain management; 2) there are potential concerns about the safety of opioid-alternatives (such as NSAIDs and acetaminophen) in pregnancy. There are data linking use of NSAIDs with complications in the third trimester, including premature closure of the ductus arteriosus and oligohydramnios from impaired renal function.27–29 In contrast, the validity of studies that associate NSAIDs with miscarriage and acetaminophen with increased rates of ADHD, is limited by recall bias and inadequate adjustment for confounding.24,30,31 Still, in the absence of robust safety data, providers may resort to using opioids for acute pain, though these medications are not without harm. Acute pain management guidelines will require a clearer understanding of the actual risks of non-opioid medications, and the potential benefits of non-pharmacologic treatments for acute pain such as deep breathing, visualization, massage, and physical therapy in the context of pregnancy.32–34

In our study, the most common diagnoses associated with opioid fills in pregnancy were unspecified, or attributable to visits for non-specific abdominal pain, urinary tract symptoms, and back pain. It is difficult to assess which of these indications is “appropriate” for opioid prescribing, particularly because we cannot see alternative pain strategies utilized by patients: opioid-sparing medications are frequently prescribed over-the-counter and non-pharmacologic regimens leave no footprint in claims data. Interestingly, undergoing a procedure in pregnancy was associated with a reduced risk of NPOU: this may be because surgical management provided definitive treatment of patients’ pain (e.g. performing a cholecystectomy for a patient with cholelithiasis or the surgical management of nephrolithiasis), or the pain related to non-surgical diagnoses conferred unique risks of NPOU. More granular understanding of the diagnoses associated with opioid prescribing will be crucial for setting best practices for pain management.

Strengths and Limitations:

Our study has several important limitations. First, the generalizability of our study is limited by our requirement for 3 years of continuous insurance enrollment; however, this was necessary to clearly define the impact of opioid prescribing in pregnancy, and likely underestimates rates of NPOU. Additionally, we assessed pharmacy claims for women with commercial insurance, limiting the generalizability of our findings. As half of all births in the United States are covered by private insurance, these results are applicable to a wide group. Second, the number of eligible births in our cohort decreased between 2008–2016. While decreased fertility rates may account for a portion of this reduction, it does not fully explain the 40% decline. It is possible that fewer women in 2016 were opioid-naïve entering pregnancy, or continuous enrollment for our prespecified period is less common in 2016 compared to prior years. Future analyses will investigate these findings in broader cohorts. Finally, using pharmacy claims allows us to identify only prescriptions filled. We cannot identify if prescriptions were written but not filled, prescriptions were filled but not all opioid pills consumed, or if opioids were consumed that were not prescribed. Future work including direct patient inquiry will be helpful for elucidating these differences.

Still, we believe these limitations are outweighed by our study’s strengths. These include the novel evaluation of pregnancy opioid exposure as a risk factor for NPOU, the large sample size, and the longitudinal assessment of risk across pregnancy up to one year postpartum. This work fills an important knowledge gap about the long-term effects of opioid prescribing during pregnancy, and represents the best currently available data.

Conclusions:

Many women receive opioid prescriptions during pregnancy, of whom, 1 in 25 develop new persistent opioid use after childbirth. Future work is needed to clarify the safety of opioid alternatives, define prescribing guidelines in pregnancy, and implement opioid-sparing regimens not just at the time of delivery, but also throughout pregnancy.

AJOG at a Glance:

Why was this study conducted?

  1. Between 14–22% of patients fill an opioid prescription in pregnancy, however, little is known about the consequences of this acute prescribing in pregnancy. We evaluated rates of NPOU after delivery in opioid-naïve women who received an opioid prescription during pregnancy, and the factors most associated with NPOU.

  2. What are the key findings? In opioid-naïve patients who received an acute opioid prescription during pregnancy, 4% developed new persistent opioid use in the year after delivery. Filling an opioid in pregnancy and postpartum were associated with NPOU, while having surgery in pregnancy was associated with lower NPOU.

  3. What does this study add to what is already known? These data suggest opioid prescribing during pregnancy poses a risk of postpartum persistent opioid use. Maternity providers will need to consider how to best balance adequate pain management in pregnancy with the risk of long-term opioid harms.

Acknowledgments

Financial support:

Drs. Brummett and Waljee received funding for research from the National Institute on Drug Abuse (R01DA042859).

This work is supported by a grant from the Michigan Department of Health and Human Services, Blue Cross Blue Shield of Michigan, and S AMSHA.

Dr. Dalton received grant funding from the Agency on Health Care Quality and Research, American Association of Obstetricians and Gynecologists Foundation, the Laura and John Arnold Foundation, National Institute for Reproductive Health, Blue Cross Blue Shield Foundation and the National Institutes for Health. She is also a contributing editor for the Medical Letter and an author for Up-to-Date. She is also a consultant for Bind.

Footnotes

Disclosures: Dr. Brummett is a consultant for Heron Therapeutics (San Diego, CA).

Condensation: Of opioid-naïve women who fill an opioid prescription during pregnancy, 1 in 25 will develop new persistent opioid use in the year after delivery.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References:

  • 1.Wilson RD, Caughey AB, Wood SL, et al. Guidelines for Antenatal and Preoperative care in Cesarean Delivery: Enhanced Recovery After Surgery Society Recommendations (Part 1). American journal of obstetrics and gynecology. 2018;219(6):523.e521– 523.e515. [DOI] [PubMed] [Google Scholar]
  • 2.Caughey AB, Wood SL, Macones GA, et al. Guidelines for intraoperative care in cesarean delivery: Enhanced Recovery After Surgery Society Recommendations (Part 2). American journal of obstetrics and gynecology 2018;219(6):533–544. [DOI] [PubMed] [Google Scholar]
  • 3.Macones GA, Caughey AB, Wood SL, et al. Guidelines for postoperative care in cesarean delivery: Enhanced Recovery After Surgery (ERAS) Society recommendations (part 3). American journal of obstetrics and gynecology 2019;221(3):247.e241–247.e249. [DOI] [PubMed] [Google Scholar]
  • 4.ACOG Committee Opinion on Postpartum Pain Management. In: Gynecologists ACoOa, ed. Vol 7422018:e35–e43. [Google Scholar]
  • 5.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 Network Open. 2019;2(7):e197863-e197863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lawrence HC. ACOG Statement on Opioid Use During Pregnancy. 2016. [Google Scholar]
  • 7.Kristiansson P, Svardsudd K, von Schoultz B. Back pain during pregnancy: a prospective study. Spine. 1996;21(6):702–709. [DOI] [PubMed] [Google Scholar]
  • 8.Gartland D, Brown S, Donath S, Perlen S. Women’s health in early pregnancy: findings from an Australian nulliparous cohort study. The Australian & New Zealand journal of obstetrics & gynaecology. 2010;50(5):413–418. [DOI] [PubMed] [Google Scholar]
  • 9.Lutterodt MC, Kahler P, Kragstrup J, Nicolaisdottir DR, Siersma V, Ertmann RK. Examining to what extent pregnancy-related physical symptoms worry women in the first trimester of pregnancy: a cross-sectional study in general practice. BJGP open. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Li D-K, Ferber JR, Odouli R, Quesenberry C. Use of nonsteroidal antiinflammatory drugs during pregnancy and the risk of miscarriage. American journal of obstetrics and gynecology. 2018;219(3):275.e271–275.e278. [DOI] [PubMed] [Google Scholar]
  • 11.Schreuder MF, Bueters RR, Huigen MC, Russel FG, Masereeuw R, van den Heuvel LP. Effect of drugs on renal development. Clinical journal of the American Society of Nephrology: CJASN. 2011;6(1):212–217. [DOI] [PubMed] [Google Scholar]
  • 12.Friedman BW, Ochoa LA, Naeem F, et al. Opioid Use During the Six Months After an Emergency Department Visit for Acute Pain: A Prospective Cohort Study. Ann Emerg Med. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Delgado MK, Huang Y, Meisel Z, et al. National Variation in Opioid Prescribing and Risk of Prolonged Use for Opioid-Naive Patients Treated in the Emergency Department for Ankle Sprains. Ann Emerg Med. 2018;72(4):389–400.e381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.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] [PMC free article] [PubMed] [Google Scholar]
  • 15.Bateman BT, Mhyre JM, Hernandez-Diaz S, et al. Development of a Comorbidity Index for Use in Obstetric Patients. Obstetrics & Gynecology. 2013;122(5):957–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bateman BT, Franklin JM, Bykov K, et al. Persistent opioid use following cesarean delivery: patterns and predictors among opioid-naive women. American journal of obstetrics and gynecology. 2016;215(3):353.e351–353.e318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Brummett CM, Waljee JF, Goesling J, et al. New Persistent Opioid Use After Minor and Major Surgical Procedures in US Adults. JAMA surgery. 2017;152(6):e170504-e170504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Fairbrother N, Young AH, Zhang A, Janssen P, Antony MM. The prevalence and incidence of perinatal anxiety disorders among women experiencing a medically complicated pregnancy. Arch Womens Ment Health. 2017;20(2):311–319. [DOI] [PubMed] [Google Scholar]
  • 19.Desai RJ, Hernandez-Diaz S, Bateman BT, Huybrechts KF. Increase in Prescription Opioid Use During Pregnancy Among Medicaid-Enrolled Women. Obstetrics & Gynecology. 2014;123(5):997–1002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Bateman BT, Hernandez-Diaz S, Rathmell JP, et al. Patterns of opioid utilization in pregnancy in a large cohort of commercial insurance beneficiaries in the United States. Anesthesiology. 2014;120(5):1216–1224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Osmundson SS, Wiese AD, Min JY, et al. Delivery type, opioid prescribing, and the risk of persistent opioid use after delivery. American journal of obstetrics and gynecology. 2019;220(4):405–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yazdy MM, Mitchell AA, Tinker SC, Parker SE, Werler MM. Periconceptional Use of Opioids and the Risk of Neural Tube Defects. Obstetrics & Gynecology. 2013;122(4):838–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Broussard CS, Rasmussen SA, Reefhuis J, et al. Maternal treatment with opioid analgesics and risk for birth defects. American journal of obstetrics and gynecology 2011;204(4) :314.e311–314.e311. [DOI] [PubMed] [Google Scholar]
  • 24.Azuine RE, Ji Y, Chang HY, et al. Prenatal Risk Factors and Perinatal and Postnatal Outcomes Associated With Maternal Opioid Exposure in an Urban, Low-Income, Multiethnic US Population. JAMA Netw Open. 2019;2(6):e196405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ko JY, Patrick SW, Tong VT, Patel R, Lind JN, Barfield WD. Incidence of Neonatal Abstinence Syndrome - 28 States, 1999–2013. MMWR Morb Mortal Wkly Rep. 2016;65(31):799–802. [DOI] [PubMed] [Google Scholar]
  • 26.Tolia VN, Patrick SW, Bennett MM, et al. Increasing incidence of the neonatal abstinence syndrome in U.S. neonatal ICUs. N Engl J Med. 2015;372(22):2118–2126. [DOI] [PubMed] [Google Scholar]
  • 27.Moise KJ Jr. Effect of advancing gestational age on the frequency of fetal ductal constriction in association with maternal indomethacin use. Am J Obstet Gynecol. 1993;168(5):1350–1353. [DOI] [PubMed] [Google Scholar]
  • 28.Kirshon B, Moise KJ Jr., Mari G, Willis R. Long-term indomethacin therapy decreases fetal urine output and results in oligohydramnios. Am JPerinatol. 1991;8(2):86–88. [DOI] [PubMed] [Google Scholar]
  • 29.Karadeniz C, Ozdemir R, Kurtulmus S, Doksoz O, Yozgat Y, Mese T. Diclofenac-induced intrauterine ductal closure. FetalDiagn Ther. 2013;34(2):133–134. [DOI] [PubMed] [Google Scholar]
  • 30.Bai D, Yip BHK, Windham GC, et al. Association of Genetic and Environmental Factors With Autism in a 5-Country Cohort. JAMA psychiatry 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Ji Y, Azuine RE, Zhang Y, et al. Association of Cord Plasma Biomarkers of In Utero Acetaminophen Exposure With Risk of Attention-Deficit/Hyperactivity Disorder and Autism Spectrum Disorder in Childhood. JAMA psychiatry 2019:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Unalmis Erdogan S, Yanikkerem E, Goker A. Effects of low back massage on perceived birth pain and satisfaction. Complementary therapies in clinical practice. 2017;28:169–175. [DOI] [PubMed] [Google Scholar]
  • 33.Yuksel H, Cayir Y, Kosan Z, Tastan K. Effectiveness of breathing exercises during the second stage of labor on labor pain and duration: a randomized controlled trial. Journal of integrative medicine. 2017;15(6):456–461. [DOI] [PubMed] [Google Scholar]
  • 34.Pregnancy and low back pain: physical therapy can reduce back and pelvic pain during and after pregnancy. The Journal of orthopaedic and sports physical therapy. 2014;44(7):474. [DOI] [PubMed] [Google Scholar]

RESOURCES