Association of Opioid and Concurrent Benzodiazepine,Skeletal Muscle Relaxant,and Gabapentinoid Usage on Healthcare Expenditure and Resource Utilization: A Serial Cross-Sectional Study,2009 to 2019

Abstract

Background:

Healthcare providers may be utilizing central nervous system (CNS) depressants to reduce opioid use due to recent changes in public policy. Combination use of these agents with opioids increases the risk of respiratory depression and death. Healthcare expenditures by individuals using these drug combinations have not been previously quantified. We sought to characterize healthcare costs and expenditures associated with a population reporting concurrent CNS depressants and opioid use compared with nonopioid analgesics in the United States from 2009 to 2019.

Methods:

A serial cross-sectional design was used to compare the healthcare expenditures of adult Medical Expenditure Panel Survey respondents who were prescribed nonopioid analgesics, opioids only, opioids/benzodiazepines (BZD), opioids/BZD/skeletal muscle relaxants (SMR), or opioids/gabapentin (gaba) using pooled data from 2009 to 2019. Expenditure (cost and resource utilization) categories included inpatient, outpatient, office-based, and prescription medicine. Average marginal effects were used to compare survey-weighted annual costs and resource utilizations across the groups as compared to nonopioid analgesic respondents, adjusted for covariates.

Results:

A weighted total of 34 241 838 individuals were identified. Most were opioid-only respondents (46.5%), followed by nonopioid analgesic (43.4%), opioid/BZD (5.3%), opioid-gaba (3.5%), and opioid/BZD/SMR respondents (1.3%). In comparison to the study groups with nonopioid analgesics, opioid-gaba users had the highest significant incremental cost difference among the different pairings (+$11 684, P < .001). Opioid-gaba, opioid/BZD, and opioid/BZD/SMR respondents had significantly higher inpatient, emergency department, and prescription drug costs and use compared to nonopioid analgesic respondents. Opioid-only respondents had higher outpatient and office-based costs and visits compared to nonopioid analgesic respondents.

Conclusions:

As healthcare providers seek to utilize fewer opioids for pain management, attention must be paid to ensuring safe and effective use of concurrent CNS depressants to mitigate high healthcare costs and burden.

Keywords

opioid benzodiazepine skeletal muscle relaxant gabapentin resource utilization healthcare costs

Highlights

Of all concurrent opioid drug categories, opioid-gabapentin respondents had the highest total annual medical costs compared to nonopioid analgesic respondents.

Opioid-only respondents had a significant increase in outpatient use and a nonsignificant increase in other healthcare services over time.

Opioid/benzodiazepines (BZD) and opioid/BZD/skeletal muscle relaxants respondents had the highest emergency department use of the study groups compared to nonopioid analgesics, likely driven by complex patient presentations.

Introduction

In 2019, an estimated 50.2 million US adults reported suffering from chronic pain.¹ The use of prescription opioids as treatment for chronic pain has resulted in significant morbidity and mortality.¹ Recently, the coronavirus pandemic exacerbated the occurrence of opioid use disorder and associated mortality rates by aggravating social and economic stressors.² In addition, individuals with opioid use disorder faced increased challenges in accessing addiction treatment during this time.^2,3 In 2021, the number of drug overdose deaths involving a prescription opioid was almost 5 times as high as those reported in 1999.⁴ Between 2008 and 2017, it was estimated that the total annual health-related economic burden of patients using a prescription opioid was $524 billion.⁵ Although the release of the Centers for Disease Control and Prevention’s (CDC’s) opioid prescribing guidelines in 2016 has been associated with a reduction in opioid prescription rates over time, the deleterious effects of the opioid epidemic continue to persist.^6,7 While focus has been centered on the abuse of opioids alone, less has been devoted to understanding the potential harmful effects of opioids with co-occurring use of other central nervous system depressants, including benzodiazepines, gabapentinoids, and skeletal muscle relaxants (referred to as “concurrent opioid use”).

Use of these medications in combination with each other increases risk of overdose from opioid-induced respiratory depression, sedation, and impaired sensory-perceptual processing.⁸ Concurrent use of opioids with benzodiazepines, skeletal muscle relaxants, and/or gabapentinoids results in significantly higher magnitudes of mortality, drug-related hospitalization, and emergency department (ED) use for respiratory depression compared to use of these drug classes alone.^9-13 These risks are further underscored in the case of triple combinations, where patients concurrently utilize an opioid, a benzodiazepine, and a skeletal muscle relaxant, colloquially referred to as the “Holy Trinity.”¹⁴ In 2017, the CDC estimated that of all deaths related to prescription opioids, 5.7% had evidence of the triple combination in the 6-month period preceding death.¹⁵ Triple combination use has also been estimated to increase the risk of ED visits by almost 9 times the rate of nonopioid users, surpassing comparable ED rates for patients using concurrent opioids and benzodiazepines.¹⁶

It has been hypothesized that utilization of concurrent opioid and triple combination use has resulted from mounting regulatory pressure to reduce use of opioids as treatment for pain.^17-19 As clinical guidelines continue to advocate for a multimodal analgesic approach to pain, clinicians may find themselves gravitating toward the use of nonopioid analgesics, skeletal muscle relaxants, gabapentinoids, and/or benzodiazepines to mitigate the risk of death imparted by opioids.¹⁸ Moreover, as individuals with chronic pain often present with comorbid anxiety, use of benzodiazepines or gabapentinoids to treat both conditions may be highly desirable.^20-23 Despite the high risk of death demonstrated by the triple combination or concurrent opioid combinations, the CDC’s updated 2022 Clinical Guidelines for Pain Management have only warned against concurrent opioid use with either benzodiazepines or gabapentinoids, based on limited evidence.¹⁸ We sought to expand on this evidence base by investigating the differences in healthcare costs and resource utilizations among nonopioid analgesic, opioid-only, opioid and concurrent benzodiazepine, opioid and concurrent gabapentin (gaba), and opioid concurrent skeletal muscle relaxant and benzodiazepine adults using pooled data from 2009 to 2019 in a nationally representative cohort. We also sought to evaluate changes in these outcomes over time among the 5 groups.

Materials and Methods

Study Design

A serial cross-sectional analysis design using pooled data from the Medical Expenditure Panel Survey (MEPS) between 2009 and 2019 was used to evaluate differences in economic burden and resource utilization between 5 groups: nonopioid analgesic respondents, opioid-only respondents, concurrent opioid/benzodiazepine respondents (“double threat”), concurrent opioid/gabapentinoid respondents (opioid-gaba), and triple combination respondents, consisting of concurrent opioid, benzodiazepine, and skeletal muscle relaxant respondents (“triple threat”). MEPS is a nationally representative survey of the US civilian, noninstitutionalized population that utilizes complex survey design methods to yield nationally weighted estimates using clustering, stratification, and multistage and disproportional sampling curated by the Agency for Healthcare Research and Quality (AHRQ).^24,25 MEPS collects data for up to 5 rounds within a 24-month period, including prescription fills.^23,24 This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional study designs.²⁶

Study Sample

We pooled data from adult MEPS participants (18 years or older) between 2009 and 2019. The full-year consolidated MEPS Household files, Prescription Medicines files, and Medical Conditions files were used to identify and assess participants.^24,27,28 The Full-Year Consolidated files contain data on demographics, socioeconomic status, insurance status, and health-related data.²⁴ The Medical Conditions file contains data on self-reported diagnoses from MEPS participants.²⁸ The Prescription Medicines file contains data on self-reported prescription drug fills from participants in outpatient pharmacies.²⁷ Drugs administered in an inpatient setting or over-the-counter drugs are not captured in the Prescription Medicines file. Healthcare resource use and related costs reported by MEPS respondents are validated with each respondent’s medical provider and pharmacy.²⁶

Characterization of Study Groups

Respondents in the “non-opioid analgesic” group were defined as those who reported purchasing a prescription for a nonopioid analgesic (eTable1, Supplementary Materials) at any round in that specific year. Respondents in the “non-opioid analgesic” group did not report any opioids, benzodiazepines, skeletal muscle relaxants, or gabapentinoids throughout the study period. We excluded selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors (SNRIs), tricyclic antidepressants, antipsychotics, antiepileptics, and hydroxyzine as “non-opioid analgesics” because we could not confirm whether these medications were reported for pain control or for another indication (eg, major depressive disorder, general anxiety disorder, etc). “Opioid-only” respondents were defined as those who only reported purchase of one or more opioids at any round in that specific year. Opioid and benzodiazepine respondents (“double threat”) were characterized as those reporting prescription purchases of both an opioid and a benzodiazepine in the same round for that year. Opioid and benzodiazepine and skeletal muscle relaxant (“triple threat”) respondents were defined as those reporting prescription purchases of an opioid, a benzodiazepine, and a skeletal muscle relaxant in the same round for that year. Opioid-gaba respondents were defined as those reporting purchase of an opioid and a gabapentinoid (pregabalin or gaba) prescription in the same round for that year. These categories were mutually exclusive (eg, double-threat respondents did not report use of a gabapentinoid or a skeletal muscle relaxant during any other round for that year). Medications were flagged using Multum Lexicon Variables from Cerner Multum, Inc (Supplementary Materials) and were manually reviewed for appropriate inclusion into the study. We did not restrict our analyses to a single type of drug formulation (eg, oral, topical, injectable, etc) to best evaluate respondents who may have used analgesics for various types of pain (eg, neuropathic, visceral, somatic, etc). Methadone was excluded as an opioid from this analysis as we could not confirm whether reported methadone was for pain control or for opioid use disorder.^29
-31

Study Outcomes

The primary outcomes of interest were total yearly healthcare costs and yearly resource utilizations. Secondary outcomes of interest included prescription costs, outpatient costs, ED costs, inpatient costs, number of prescriptions filled, number of office-based visits, number of ED visits, number of outpatient visits, and number of inpatient hospitalizations. Costs are the sum of direct payments provided, including out-of-pocket payments and payments by private insurance, Medicare, Medicaid, or other sources and were adjusted for inflation using the Personal Consumption Expenditure Health price index as recommended by AHRQ to reflect 2021 USD.^24,32 Data on prescriptions filled included information on medication name, number of fills and refills for that medication, and round reported. Medical provider visits were captured using the number of office-based visits and number of outpatient visits to a medical provider. Data on ED visits included the count of all ED visits reported for that year. Inpatient hospitalizations were captured as the total number of hospital discharges.

Demographic and Other Variables

Participant demographic data used in our analysis included age category (18-24 years, 25-44 years, 45-64 years, and 65 years or older), sex (female/male), race (White/Black or African American/American Indian or Alaska Native/Asian or Native Hawaiian or Pacific Islander/Multiple), ethnicity (Hispanic/Non-Hispanic), marital status (married/widowed/divorced/separated/never), education (no diploma/General Education Development or high school/associate, other, <4 years college/BS degree/Master’s, doctorate, professional degree), geographic region (Northeast/Midwest/South/West), poverty status (poor or negative/near poor/low income/middle income/high income), insurance coverage (any private/public/uninsured), and Charlson Comorbidity Index (CCI) score. CCI was calculated using a previously published algorithm using International Classification of Diseases 10 (ICD-10) diagnosis codes to identify comorbidities from each year’s Medical Conditions file to account for disease burden.³³ For years 2009 to 2015, ICD-9 diagnosis codes were used to identify CCI conditions (Supplementary Materials). Poverty status category was determined by each participant’s total annual income of his or her family for each year of data.³⁴ Categories were defined according to federal poverty level (FPL): poor or negative (<100% of FPL), near poor (100% to <120% of FPL), low income (125% to <200% of FPL), middle income (200% to <400% of FPL), and high income (greater than or equal to 400% of FPL).³⁴

Statistical Analysis

Demographic data were descriptively analyzed between the 5 study groups using Analysis of Variance (ANOVA) and chi-squared tests for continuous and categorical variables, respectively. Means and standard deviations were derived for continuous data, and frequencies with proportions were summarized for categorical variables. Survey weights were applied to estimate the population standard errors and 95% confidence intervals (CIs) using STATA’s svy command.³⁵

For our primary aim, we individually compared opioid-only, double-threat, triple-threat, and opioid-gaba group healthcare costs and resource utilizations to the nonopioid analgesic group. We further compared the groups on total prescription drug costs, total outpatient costs, total ED costs, total inpatient hospitalization costs, total office-based costs, total number of prescriptions filled, total number of office-based visits, total number of ED visits, and total number of inpatient hospitalizations. Survey-weighted two-part models were used to evaluate cost-related outcomes due to a large number of respondents with $0 expenditure with STATA’s twopm package.^36
-38

We used logistic regression for the model’s first part to estimate the likelihood of a respondent having nonzero healthcare costs. Generalized linear models with gamma distribution were used for the second part of the models to estimate the relationship between healthcare cost and drug group contingent on whether the participant had nonzero healthcare costs in the first part, adjusting for covariates.^36
-38 All models were adjusted for covariates including age, year, sex, race/ethnicity, marital status, education attained, geographic region, income category, insurance status, and CCI score. All cost model results were reported using average marginal effects in USD with 95% CIs. Differences in resource utilization categories (eg, number of prescription medications, etc) between the study drug groups were evaluated using survey-weighted negative binomial regression models to account for overdispersion, adjusting for covariates listed previously. Results were reported as incidence rate ratios (IRRs) with 95% CIs and robust standard errors. All models utilized the nonopioid analgesic group as a reference group.

For our secondary aim, we evaluated whether the study outcomes (costs, resource utilizations) significantly changed over time for each study group compared to the nonopioid analgesic group. This was performed using a series of linear regression models with an interaction term between year and study group, adjusting for covariates. Findings were presented as average annual differences using regression model coefficients with standard errors and 95% CIs. Statistical significance was defined using a 2-tailed alpha value of <0.05. All analyses were performed using STATA version 17 (StataCorp, College Station, TX).

Results

In total, 370 738 unique unweighted MEPS respondents were pooled between 2009 and 2019 (Figure 1). Among these, 18 214 (4.9%) purchased only a nonopioid analgesic at least once, 16 796 (4.5%) purchased only an opioid at least once, 1742 (0.47%) purchased an opioid and a benzodiazepine in the same round, 439 (0.12%) purchased an opioid, a benzodiazepine, and a skeletal muscle relaxant in the same round, and 1230 (0.33%) purchased an opioid and a gabapentinoid in the same round (Table 1). A total of 370 738 unweighted MEPS respondents represented an annual weighted population of 34.2 million lives. Opioid respondents accounted for 172.1 billion dollars spent in total medical expenditure, followed by nonnarcotic opioid respondents ($103.2 billion), double-threat respondents ($29.4 billion), opioid-gaba respondents ($26.3 billion), and triple-threat respondents ($7.5 billion). Demographic characteristics varied across the study groups. Nonopioid analgesic respondents were predominately middle-aged (37.4%), female (56.9 %), white (75.2%), not Hispanic (81.4%), and married (53.0%) with higher education and income levels (high income: 37.8%). Opioid-only respondents were younger with a higher proportion of white individuals (82.0%), slightly higher income (high income: 40.1%), and a high rate of private insurance coverage (7.2%). Double-threat respondents were mostly male (62.6%), white (88.5%), and non-Hispanic (93.7%) with lower income (poor/negative: 18.4%) and more public insurance coverage (37.1%). Triple-threat respondents had similar characteristics to nonopioid analgesic respondents. Opioid-gaba respondents were predominantly over age 65 (46.6%) with a higher comorbidity burden (mean CCI: 3.2).

Figure 1.

Number of unique respondents per drug group per year, 2009 to 2019.

Table 1.

Cohort Demographics, Weighted.

Characteristic	Opioid only		Opioid + BZD		Opioid + BZD + SMR		Opioid + GABA		Nonopioid analgesic only		P value
Characteristic	N	%	N	%	N	%	N	%	N	%	P value
Unweighted sample size	16 796		1742		439		1230		18 214
Weighted sample size	15 916 017		1 802 395		437 755		1 186 319		14 899 352
Age
18 – 24 years	1 622 366	10.2	43 525	2.4	5430	1.2	16 284	1.4	1 069 958	7.2	<.001
25 – 44 years	4 852 035	30.5	387 835	21.5	97 346	22.2	143 216	12.1	3 636 846	24.4
45 – 64 years	5 596 927	35.2	800 983	44.4	230 563	52.7	473 786	39.9	5 567 689	37.4
65+ years	3 844 689	24.2	570 053	31.6	104 415	23.9	553 033	46.6	4 624 858	31.0
Sex
Male	7 120 351	44.7	1 128 334	62.6	137 467	31.4	498 572	42.0	6 425 311	43.1	<.001
Female	8 795 666	55.3	674 061	37.4	300 288	68.6	687 747	58.0	8 474 041	56.9	<.001
Race
White	13 045 941	82.0	1 594 253	88.5	386 468	88.3	989 566	83.4	11 208 145	75.2	<.001
Black	1 932 855	12.1	109 310	6.1	32 281	7.4	11 8651	10.0	2 346 182	15.7
American Indian/Alaska Native	165 887	1.0	26 367	1.5	5211	1.2	13 670	1.2	163 835	1.1
Asian/NativeHawaiian/Pacific Islander	411 031	2.6	22 108	1.2	2947	0.7	14 450	1.2	806 859	5.4
Multiple	354 961	2.2	49 643	2.8	10 848	2.5	49 981	4.2	366 249	2.5
Ethnicity
Hispanic	1 643 920	10.3	113 927	6.3	26 882	6.1	82 041	6.9	2 773 540	18.6	<.001
Not Hispanic	14 272 097	89.7	1 688 468	93.7	410 873	93.9	1 104 278	93.1	12 125 812	81.4	<.001
Marital status
Married	8 376 113	52.6	923 988	51.3	203 625	46.5	620 974	52.3	7 900 830	53.0	<.001
Widowed	1 312 721	8.2	212 405	11.8	62 254	14.2	198 666	16.7	1 489 461	10.0
Divorced	2 208 480	13.9	344 141	19.1	100 966	23.1	209 304	17.6	1 992 373	13.4
Separated	350 973	2.2	70 015	3.9	19 227	4.4	32 448	2.7	407 076	2.7
Never married	3 667 730	23.0	251 847	14.0	51 682	11.8	124 927	10.5	3 109 130	20.9
Not ascertained/don’t know	1 704 910	10.7	216 104	12.0	47 528	10.9	140 858	11.9	1 515 819	10.2	<.001
Education
No diploma	1 866 441	11.7	238 463	13.2	67 590	15.4	154 435	13.0	2 192 431	14.7
GED or HS	6 440 032	40.5	777 714	43.1	187 467	42.8	544 234	45.9	5 820 541	39.1
Associate, other, degree, <4 years college	2 146 580	13.5	258 060	14.3	83 271	19.0	140 397	11.8	1 731 211	11.6
BS degree	2 446 728	15.4	188 042	10.4	33 341	7.6	137 570	11.6	2 257 555	15.2
Master’s, doctorate, professional degree	1 311 325	8.2	124 013	6.9	18 557	4.2	71 825	6.1	1 381 795	9.3
Geographic region
Midwest	3 748 226	23.6	460 992	25.6	82 801	18.9	288 839	24.3	2 972 770	20.0	<.001
Northeast	2 460 310	15.5	280 001	15.5	59 081	13.5	127 351	10.7	2 754 604	18.5
South	6 195 133	38.9	699 753	38.8	209 599	47.9	520 248	43.9	5 555 493	37.3
West	3 512 348	22.1	361 650	20.1	86 274	19.7	249 881	21.1	3 616 484	24.3
Income*
Poor/negative	2 081 369	13.1	332 761	18.5	107 474	24.6	206 762	17.4	2 137 192	14.3	<.001
Near poor	691 838	4.3	127 592	7.1	33 158	7.6	82 861	7.0	754 974	5.1
Low income	2 194 505	13.8	289 396	16.1	78 375	17.9	240 676	20.3	2 163 427	14.5
Middle income	4 563 728	28.7	483 258	26.8	109 489	25.0	329 540	27.8	4 213 332	28.3
High income	6 384 578	40.1	569 389	31.6	109 259	25.0	326 480	27.5	5 630 427	37.8
Insurance coverage
Any private	10 924 926	68.6	1 006 052	55.8	220 497	50.4	635 859	53.6	9 294 882	62.4	<.001
Public	3 839 027	24.1	669 418	37.1	190 927	43.6	502 392	42.3	4 600 770	30.9
Uninsured	1 152 063	7.2	126 926	7.0	26 331	6.0	48 068	4.1	1 003 699	6.7
Charlson Comorbidity Index (mean, [SE])	1.7 (0.025)		2.5 (0.085)		2.3 (0.11)		3.2 (0.085)		2.0 (0.028)		<.001

Abbreviations: SE, standard error; BZD, benzodiazepine; SMR, skeletal muscle relaxant; GABA, gabapentinoid; GED, General Education Development; HS, high school; BS, Bachelor of Science; FPL, federal poverty level.

Poor/negative = <100% FPL; near poor = 100% to 124% FPL; low income = 125% to 199% FPL; middle income = 200% to 399% FPL; high income = 400% FPL or greater.

Healthcare Costs and Resource Utilization

Between 2009 and 2019, opioid-gaba respondents had the highest average annual total medical costs ($22 129) compared to opioid-only ($10 813), double-threat ($16 317), triple-threat ($17 084), and nonopioid analgesic only respondents ($6929). This trend was also observed in average inpatient hospitalization costs, outpatient costs, office-based costs, and prescription drug costs (Figure 2). Opioid-gaba respondents had the highest average number of inpatient hospitalizations, outpatient visits, office-based visits, and prescription drugs. Triple-threat respondents had the highest average ED costs ($5699) and average yearly number of ED visits (Figure 2). Adjusted for covariates, opioid-gaba respondents also had the highest incremental total annual medical costs (+$11 684, 95% CI: +$8633, +$13 571) of the study groups when compared to nonopioid analgesic respondents. Opioid-only respondents had the highest incremental effect on outpatient costs (+$950, 95% CI: +$856, +$1064; Table 2) and the lowest incremental effect in other cost categories compared to nonopioid analgesic respondents. Double-threat respondents had a higher incremental effect on total medical costs (+$8789, 95% CI: +$6893, +$5471), driven by inpatient costs and prescription drug costs compared to nonopioid analgesic respondents. Double-threat respondents also had higher incidence rates of inpatient hospitalizations (IRR: 2.69, 95% CI: 2.37, 3.04) and ED visits compared to nonopioid respondents. Triple-threat respondents had a similar result to double-threat respondents, with a high incremental effect on total medical costs compared to nonopioid analgesic respondents (+$9876, 95% CI: +$6527, +$12 240).

Figure 2.

Average yearly costs and resource use by drug group, 2009 to 2019. The top half of the figure illustrates unadjusted average yearly cost by cost category compared across nonnarcotic analgesic, opioid-only, opioid/BZD, opioid/BZD/SMR, and opioid-gaba respondents. The bottom half illustrates unadjusted average yearly resource use by resource use category compared across the 5 drug groups of interest.

Table 2.

Estimated Average Annual Differences in Healthcare Costs and Resource Utilization by Treatment Group, 2009 to 2019.

	Nonopioid analgesic	Opioid only	Opioid + BZD	Opioid + BZD + SMR	Opioid + GABA
Average annual difference, $ (95% CI)
Total medical costs	Reference	5228** (4463, 5471)	8789** (6893, 5471)	9876** (6527, 12 240)	11 684** (8633, 13 571)
Inpatient costs	Reference	1918** (1661, 2175)	2594** (1936, 3252)	2665** (1043, 4287)	3897** (2766, 5028)
outpatient costs	Reference	950** (836, 1064)	699** (467, 931)	518** (166, 870)	687** (385, 988)
Office-based visit costs	Reference	741** (625, 857)	1429** (1083, 1776)	1485** (912, 2059)	1849** (1145, 2553)
ED costs	Reference	253** (210, 296)	322** (228, 416)	397** (161, 633)	258** (130, 385)
Prescription drug costs	Reference	117 (−64, 293)	2560** (1754, 3211)	3270** (1833, 4509)	3334** (2416, 4051)
Resource utilization: incident rate ratios (95% CI)
Number of inpatient discharges	Reference	2.02** (1.87, 2.18)	2.69** (2.37, 3.04)	2.39** (1.79, 3.18)	2.93** (2.53, 3.38)
Number of outpatient visits	Reference	1.56** (1.43, 1.71)	1.80** (1.51, 2.16)	1.65** (1.31, 2.08)	2.02** (1.60, 2.55)
Number of office-based visits	Reference	1.24** (1.19, 1.30)	1.92** (1.75, 2.12)	2.12** (1.88, 2.38)	2.01** (1.84, 2.19)
Number of ED visits	Reference	1.68** (1.59, 1.77)	1.97** (1.79, 2.18)	1.97** (1.56, 2.50)	1.75** (1.54, 2.00)
Number of prescription medications	Reference	0.94**(0.91, 0.96)	1.87** (1.77, 1.98)	2.74** (2.48, 3.04)	1.92** (1.79, 2.06)

Average marginal effects were estimated using 2-part survey weighted models for each treatment group individually as compared to the nonopioid analgesic reference group.

Abbreviations: BZD, benzodiazepine; SMR, skeletal muscle relaxant; GABA, gabapentinoid; CI, confidence interval; ED, emergency department.

Statistical significance (P < .001).

Annual Trends in Healthcare Costs and Resource Utilization

Table 3 describes the yearly trends in costs and resource utilizations for the study drug groups compared to the nonopioid analgesic group. All groups had a significantly higher average annual increase in total medical costs compared to nonopioid analgesics, with the opioid-gaba group reporting the highest average annual increase of all pairs (+$1237, P < .001). Opioid-only respondents had a significant average annual increase in outpatient costs (+$98.84, P < .001), office-based costs (+$106.14, P < .001), and prescription drug costs (+$53.99, P = .03) compared to nonopioid analgesic respondents. Opioid-only respondents also had significantly higher average annual increases in outpatient visits (+0.042, P = .03) and office-based visits (+0.22, P < .001) compared to nonopioid analgesic respondents. Triple-threat respondents had significantly higher average annual increases in office-based costs compared to nonopioid respondents (+$189.36, P = .02). Double-threat respondents had significantly higher average annual increases in prescription costs (+$317.42, P < .001), number of inpatient hospitalizations (+0.030, P < .001), number of outpatient visits (+0.17, P = .04), and number of ED visits (+0.042, P = .02) compared to nonopioid analgesic respondents.

Table 3.

Average Annual Differences in Costs and Resource Utilizations Trends of Groups Compared to Nonopioid Analgesic Only Respondents, 2009 to 2019.

	Average annual difference in slopes (SE)	P value	95% confidence interval
Total medical costs
Opioid	$364.70 (87.70)*	<.001	($192.20, $537.00)
Opioid + BZD	$653.60 (298.70)*	.031	($66.10, $1341.00)
Opioid + BZD + SMR	$1173.60 (491.70)*	.022	($206.60, $2141.00)
Opioid + GABA	$1237.80 (371.80)*	<.001	($506.60, $1969.00)
Inpatient costs
Opioid	$106.86 (57.80)	.073	(−$6.80, $220.00)
Opioid + BZD	$109.73 (188.17)	.561	(−$260.30, $480.00)
Opioid + BZD + SMR	$480.57 (396.25)	.230	(−$298.80, $1260.00)
Opioid + GABA	$504.60 (264.88)	.061	(−$16.30, $1026.00)
Outpatient costs
Opioid	$98.84 (24.40)*	<.001	($50.90, $147.00)
Opioid + BZD	$98.09 (62.91)	.121	(−$25.60, $222.00)
Opioid + BZD + SMR	$61.38 (58.61)	.300	(−$53.90, $177.00)
Opioid + GABA	$157.52 (81.83)	.064	(−$3.40, $318.00)
Office-based costs
Opioid	$106.14 (28.20)*	<.001	($51.00, $161.60)
Opioid + BZD	$116.69 (77.33)	.133	(−$35.00, $268.80)
Opioid + BZD + SMR	$189.36 (82.41)*	.024	($27.00, $351.40)
Opioid + GABA	$125.19 (131.18)	.340	(−$133.00, $383.20)
Emergency department costs
Opioid	$14.83 (7.92)	.061	(−$0.74, $30.00)
Opioid + BZD	$1.21 (14.7)	.934	(−$27.78, $30.00)
Opioid + BZD + SMR	$18.22 (19.40)	.352	(−$19.92, $56.00)
Opioid + GABA	$9.80 (12.30)	.431	(−$14.31, $34.00)
Prescription medicine costs
Opioid	$53.99 (25.39)*	.030	($4.10, $104.00)
Opioid + BZD	$178.66 (113.00)	.113	(−$43.60, $401.00)
Opioid + BZD + SMR	$380.97 (293.38)	.190	(−$196.00, $958.00)
Opioid + GABA	$317.42 (104.13)*	<.001	($112.60, $522.00)
Number of inpatient discharges
Opioid	0.0024 (0.0024)	.310	(−0.0022, 0.0071)
Opioid + BZD	−0.0073 (0.0086)	.390	(−0.024, 0.0095)
Opioid + BZD + SMR	0.032 (0.019)	.102	(−0.006, 0.069)
Opioid + GABA	0.030 (0.0088)*	<.001	(0.0081, 0.043)
Number of outpatient visits
Opioid	0.042 (0.019)*	.030	(0.0058, 0.079)
Opioid + BZD	0.042 (0.066)	.531	(−0.089, 0.17)
Opioid + BZD + SMR	0.16 (0.091)	.112	(−0.034, 0.32)
Opioid + GABA	0.17 (0.079)*	.04	(0.011, 0.32)
Number of office-based visits
Opioid	0.22 (0.067)*	<.001	(0.092, 0.36)
Opioid + BZD	0.28 (0.24)	.264	(−0.21, 0.76)
Opioid + BZD + SMR	0.54 (0.37)	.142	(−0.18, 1.26)
Opioid + GABA	0.19 (0.22)	.403	(−0.26, 0.63)
Number of emergency department visits
Opioid	−0.0014 (0.0034)	.681	(−0.0082, 0.0053)
Opioid + BZD	−0.0052 (0.011)	.632	(−0.026, 0.016)
Opioid + BZD + SMR	0.042 (0.018)*	.023	(0.0057, 0.078)
Opioid + GABA	0.0083 (0.01)	.433	(−0.012, 0.029)
Number of prescription medicines
Opioid	0.065 (0.09)	.470	(−0.111, 0.242)
Opioid + BZD	−0.079 (0.35)	.824	(−0.77, 0.613)
Opioid + BZD + SMR	0.29 (0.80)	.713	(−1.28, 1.87)
Opioid + GABA	−0.65 (0.38)	.090	(−1.41, 0.10)

Abbreviations: SE, standard error; BZD, benzodiazepine; SMR, skeletal muscle relaxant; GABA, gabapentinoid.

Statistical significance at P < .05.

Discussion

In this analysis, we found that double- or triple-threat respondents demonstrated greater increases in the growth of inpatient hospitalizations, ED use, office-based visits, and total medical costs over time compared to nonopioid analgesic respondents. Participants in these two groups were older and had a higher comorbidity burden compared to nonopioid analgesic respondents, and tended to fall into lower socioeconomic strata with higher rates of public insurance. Our findings indicate that there is an association between double- and triple-threat status and healthcare expenditures; however, the underlying mechanism for this is unclear.^11,39,40 We believe this may be due to several potential hypotheses. First, double- or triple-threat respondents may have complicated clinical presentations that include the presence of multiple comorbidities, such as anxiety or depression in addition to chronic pain.⁴¹ Second, the association between these drug combinations with increased inpatient hospitalizations and ED use may suggest that these patients have a higher risk of experiencing compounded adverse events, such as respiratory depression. This is particularly concerning, as participants in these groups were, on average, older than the nonopioid analgesic group. Respiratory depression in these patients may lead to a higher risk of negative consequences, such as opioid-related overdose compared to opioid-only respondents.^10,42 Of the five study groups, respondents in the triple-threat group had the greatest increase in growth of inpatient hospitalizations, ED costs, and number of ED visits compared to nonopioid analgesic respondents.

Of the study drug group comparators, opioid-gaba respondents had the highest total medical costs in total and in growth over time when compared to nonopioid analgesic respondents. Opioid-gaba respondents also had greater inpatient hospitalizations and prescription drug costs and use compared to nonopioid analgesic respondents over time. These findings are similar to other studies that indicate opioid prescribing rates are greater among individuals who are also prescribed gabapentinoids.^43
-46 The steady increase in gaba use may reflect prescriber’s perception that gabapentinoids are a safer option for chronic pain management.⁴⁷

We hypothesize that the effect of gabapentinoids potentiating the effect of opioids may be responsible for the increase in growth of inpatient hospitalization costs and use over time for opioid-gaba respondents. While these higher inpatient expenditures have been noted and somewhat mitigated through policy changes by certain regulatory bodies, such as the Centers for Medicare and Medicaid Services and third-party payers, it may not be enough to stave off the rising trends noted here.^48
-50 Gaba and pregabalin are foundational in the use of multimodal analgesia to treat chronic pain.⁵¹ To continue encouraging safe use of these agents concurrently with opioids, investigators should focus on identifying factors associated with inpatient events and costs among opioid-gaba users.

We noted higher healthcare costs across various resource categories among double-threat, triple-threat and opioid-gaba respondents in total and over time. While we hypothesized that these results may potentially be due to a higher rate of severe adverse events commonly observed with these medication combinations (eg, respiratory depression, etc), we cannot rule out other potential causative factors. For example, the triple-threat and opioid-gaba groups were older than the other study groups and had a higher comorbidity burden. It is possible that healthcare costs were simply higher in these groups due to higher disease burden, which would require more intensive care. However, literature suggests that the negative effects of concurrent opioid-CNS depressant use persist even among cohorts matched on opioid use/nonuse. Hawkins and colleagues reported a higher rate of all-cause mortality and opioid overdose with concurrent opioid/BZD use among Veterans Affairs patients with posttraumatic stress disorder compared to matched groups of opioid-only or BZD-only patients.⁵² Weisberg et al reported similar results among a population of long-term prescription opioid and/or BZD users with HIV infection.⁵³ Finally, Sharma and colleagues found that concurrent use of opioids and BZDs resulted in a higher risk of hospitalization/ED visits and mortality over a matched group of opioid-only users in a Canadian sample.¹³ Although our analyses did not utilize matching, we posit that the overall increase in healthcare expenditure among triple-threat and opioid-gaba respondents noted here may be, in part, associated with a higher use of resources linked to opioid overdose and related events.

Compared to nonopioid analgesic respondents, the number of opioid-only respondents decreased over time. This downward trend supports other published national trends starting in 2012 for opioid prescriptions reported by the CDC and other sources.^54,55 We hypothesize the reason for this decline to be multifactorial. Primarily, US policies at both the federal and state level aimed at increasing scrutiny on opioid use may have played a strong role in our observed reduction of opioid-only respondents. For example, state prescription drug monitoring programs (PDMP), an increase in the amount and quality of provider education, and the release of clinical guidelines aimed at restricting the use of opioids for noncancer pain may have contributed to the decline noted here.^6,17,56
-61

There was an increase in the use of outpatient and office-based services and costs over time among opioid-only respondents compared to nonopioid analgesic respondents. Although the average annual increases in outpatient visit, office-based visit, and prescription drug costs among opioid-only respondents may seem modest, third-party payers tasked with covering healthcare expenses for thousands of such patients should remain mindful of these gradually increasing costs, which can have an impact on patients due to increased opportunity costs. Opioid-only respondents also had lower inpatient hospitalizations, ED use, and ED costs when compared to nonopioid analgesic respondents. We hypothesize that various policies aimed toward expanding behavioral health services and increasing access to medications used for opioid use disorder (MOUD) may have driven these observations. For example, Medicaid expansion after implementation of the Affordable Care Act has been associated with reductions in opioid-related hospital use and an increase in access to buprenorphine prescriptions for MOUD.^62
-64 Federal policies have also greatly expanded access to treatment for opioid use disorder, such as the Comprehensive Addiction and Recovery Act (CARA) of 2016, the 21st Century Cures Act, the Substance Use-Disorder Prevention that Promotes Opioid Recovery and Treatment (SUPPORT) for Patients and Communities Act of 2018, and the Consolidated Appropriations Act.^65
-68 While these are only hypotheses for the underlying reasons for the trends among opioid-only patients noted here, future research may want to focus on these trends with the advent of newer legislation, such as elimination of the X-waiver, which was previously a required training and registration process by physicians to prescribe buprenorphine.⁶⁹

Limitations

Our study had limitations of note. Some MEPS data components may be subject to recall bias, including reported demographic characteristics. We did not distinguish between chronic or acute opioid respondents. Therefore, our results cannot be extrapolated to one subgroup or the other. Nonopioid analgesic respondents may have used their medications for reasons other than for alleviating pain, such as low-dose aspirin for the use of cardiovascular disease prevention. Similarly, respondents reporting use of benzodiazepines or gabapentinoids may have been using their medication for psychiatric conditions (eg, anxiety), or alcohol withdrawal. Accordingly, we cannot assume that our results apply only to patients who were using these medications for pain. We could not ascertain whether opioid respondents also had substance use disorder, or whether these respondents were prescribed their medications from 1 physician versus several (eg, “doctor shopping”). We could not be certain whether MEPS respondents who reported purchase of prescription medications took those medications as directed or whether use of medications in concurrent categories (double threat, triple threat, opioid-gaba) overlapped in any 1 round. Our study groups were also significantly different from each other with regard to demographic characteristics, which may have affected our findings. For example, the opioid-gaba group was, on average, the oldest group, which may have played a role in their high use of inpatient hospitalizations. However, we did control for this bias to a certain extent by adjusting for age in our models. Because we could not be certain of outpatient or office-based physician specialty type, we also could not determine whether the increase in use of these services among opioid-only respondents was due to an increase in the use of behavioral health services related to substance use disorder or not. We could not ascertain the primary reason for each MEPS respondent’s reported healthcare use event (eg, inpatient hospitalization, outpatient visit, etc). Therefore, we could not ascertain whether resources utilized were primarily used for pain, or for another reason. While we utilized mutually exclusive categories to define our study groups, there may have also been some cases of chronic opioid respondents picking up a single (“one time”) concurrent CNS depressant (eg, a benzodiazepine). The effects of a singular concurrent prescription may be less impactful than the effects of a chronic, long-term concurrent prescription, which may have affected our results. Finally, we did not include other medications used for pain that may also have been used for other indications (eg, SSRIs, SNRIs, etc). This may have resulted in an underestimation of individuals in the “non-opioid analgesic” group.

Conclusions

Compared to nonopioid analgesic respondents, opioid-only respondents had a greater use of outpatient or office-based services. Double- or triple-threat respondents and opioid-gaba respondents demonstrated increased total medical costs and use of inpatient hospitalizations and ED services over time compared to nonopioid analgesic users. Policymakers and third-party payers should orient future legislation, regulation, and education on emphasizing and rewarding safe and continuous oversight from physicians and pharmacists for patients taking concurrent opioid combinations to stagnate the rising risk of negative clinical consequences.

Supplemental Material

sj-docx-1-saj-10.1177_29767342241247372 – Supplemental material for Association of Opioid and Concurrent Benzodiazepine, Skeletal Muscle Relaxant, and Gabapentinoid Usage on Healthcare Expenditure and Resource Utilization: A Serial Cross-Sectional Study, 2009 to 2019

Supplemental material, sj-docx-1-saj-10.1177_29767342241247372 for Association of Opioid and Concurrent Benzodiazepine, Skeletal Muscle Relaxant, and Gabapentinoid Usage on Healthcare Expenditure and Resource Utilization: A Serial Cross-Sectional Study, 2009 to 2019 by Aryana Sepassi, Meng Li, Kangho Suh, Britney Stottlemyer and Mark Bounthavong in Substance Use & Addiction Journal

Footnotes

Author Contributions

AS: Conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, software, validation, visualization, roles/writing—original draft, writing—review and editing.

ML: Conceptualization, methodology, roles/writing—review and editing, supervision.

KS: Conceptualization, methodology, roles/writing—review and editing, supervision.

BS: Conceptualization, methodology, roles/writing—review and editing.

MB: Conceptualization, methodology, roles/writing—review and editing.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Compliance,Ethical Standards,and Ethical Approval

The Ethics Committee of the University of California, Irvine, waived the need for ethics approval and the need to obtain consent for the collection, analysis, and publication of the retrospectively obtained and anonymized data for this noninterventional study.

ORCID iD

Aryana Sepassi

Supplemental Material

Supplementary material for this article is available online at the SAJ website .

References

Yong

Ananda-Rajah

Cameron

, et al. Cytomegalovirus reactivation is associated with increased risk of late-onset invasive fungal disease after allogeneic hematopoietic stem cell transplantation: a multicenter study in the current era of viral load monitoring. Biol Blood Marrow Transplant. 2017;23(11):1961-1967. doi:10.1016/j.bbmt.2017.07.025

Joint Economic Committee Democrats. 2022. The economic toll of the opioid crisis reached nearly $1.5 trillion in 2020. Accessed April 7, 2023. https://www.jec.senate.gov/public/_cache/files/67bced7f-4232-40ea-9263-f033d280c567/jec-cost-of-opioids-issue-brief.pdf

Haley

Saitz

The opioid epidemic during the COVID-19 pandemic. JAMA. 2020;324(16):1615-1617. doi:10.1001/jama.2020.18543

CDC Injury Center. Death rate maps and graphs. Drug overdose deaths. 2022. Accessed April 7, 2023. https://www.cdc.gov/drugoverdose/deaths/index.html

Bounthavong

Suh

Spoutz

Stottlemyer

Sepassi

Trends in healthcare expenditures and resource utilization among a nationally representative population with opioids in the United States: a serial cross-sectional study, 2008 to 2017. Subst Abuse Treat Prev Policy. 2021;16(1):80. doi:10.1186/s13011-021-00415-5

Bohnert

ASB

Guy

Jr Losby

JL.

Opioid prescribing in the United States before and after the Centers for Disease Control and Prevention’s 2016 Opioid Guideline. Ann Intern Med. 2018;169(6):367-375. doi:10.7326/M18-1243

Salvatore

Guy

Mikosz

CA.

Changes in opioid dispensing by medical specialties after the release of the 2016 CDC Guideline for Prescribing Opioids for Chronic Pain. Pain Med. 2022;23(11):1908-1914. doi:10.1093/pm/pnac068

Horsfall

Sprague

. The pharmacology and toxicology of the ‘Holy Trinity.’ Basic Clin Pharmacol Toxicol. 2017;120(2):115-119. doi:10.1111/bcpt.12655

Evoy

Sadrameli

Contreras

Covvey

Peckham

Morrison

MD.

Abuse and misuse of pregabalin and gabapentin: a systematic review update. Drugs. 2021;81(1):125-156. doi:10.1007/s40265-020-01432-7

10.

Hernandez

Brooks

Zhang

Exposure-response association between concurrent opioid and benzodiazepine use and risk of opioid-related overdose in Medicare Part D beneficiaries. JAMA Netw Open. 2018;1(2):e180919. doi:10.1001/jamanetworkopen.2018.0919

11.

Delcher

Wei

, et al. Risk of opioid overdose associated with concomitant use of opioids and skeletal muscle relaxants: a population-based cohort study. Clin Pharmacol Ther. 2020;108(1):81-89. doi:10.1002/cpt.1807

12.

Mooney

Zhu

Yoo

Wolitzky-Taylor

Hser

YI.

Association between benzodiazepine and opioid prescription and mortality among patients in a large healthcare system. Jo Addict Med. 2022;16(1):65-71. doi:10.1097/ADM.0000000000000828

13.

Sharma

Simpson

Samanani

Jess

Eurich

DT.

Concurrent use of opioids and benzodiazepines/Z-drugs in Alberta, Canada and the risk of hospitalisation and death: a case cross-over study. BMJ Open. 2020;10(11):e038692. doi:10.1136/bmjopen-2020-038692

14.

Forrester

MB.

Ingestions of hydrocodone, carisoprodol, and alprazolam in combination reported to Texas poison centers. J Addict Dis. 2011;30(2):110-115. doi:10.1080/10550887.2011.554778

15.

Scholl

Seth

Kariisa

Wilson

Baldwin

Drug and opioid-involved overdose deaths—United States, 2013-2017. MMWR Morb Mortal Wkly Rep. 2019;67:1419-1427.

16.

Watanabe

Yang

Association of combination opioid, benzodiazepine, and muscle relaxant usage with emergency department visits in a nationwide cohort in the United States. Int J Clin Pharm. 2021;43(2):358-364. doi:10.1007/s11096-020-01012-5

17.

Dowell

Haegerich

Chou

. CDC Guideline for Prescribing Opioids for Chronic Pain—United States, 2016. Morbidity and Mortality Weekly Report (MMWR). Accessed April 7, 2023. https://www.cdc.gov/mmwr/volumes/65/rr/rr6501e1.htm

18.

Dowell

Ragan

Jones

Baldwin

Chou

CDC Clinical Practice Guideline for Prescribing Opioids for Pain—United States, 2022. MMWR Recomm Rep. 2022;71:1-95. doi:10.15585/mmwr.rr7103a1

19.

US Food and Drug Administration. FDA Drug Safety Communication: FDA warns about serious risks and death when combining opioid pain or cough medicines with benzodiazepines; requires its strongest warning. 2019. Accessed April 7, 2023. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-warns-about-serious-risks-and-death-when-combining-opioid-pain-or

20.

Ahmed

Bachu

Kotapati

, et al. Use of gabapentin in the treatment of substance use and psychiatric disorders: a systematic review. Front Psychiatry. 2019;10:228. doi:10.3389/fpsyt.2019.00228

21.

Markota

Morgan

RJ.

Treatment of generalized anxiety disorder with gabapentin. Case Rep Psychiatry. 2017;2017:6045017. doi:10.1155/2017/6045017

22.

Slee

Nazareth

Bondaronek

Liu

Cheng

Freemantle

Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis. Lancet. 2019;393(10173):768-777. doi:10.1016/S0140-6736(18)31793-8

23.

Verduin

McKay

Brady

KT.

Gabapentin in comorbid anxiety and substance use. Am J Addict. 2007;16(2):142-143. doi:10.1080/10550490601186238

24.

Agency for Healthcare Research and Quality. MEPS HC-216 2019 Full Year Consolidated Data File Documentation. 2021. Accessed December 1, 2022. https://meps.ahrq.gov/data_stats/download_data/pufs/h216/h216doc.shtml

25.

Chowdhury

Machin

Gwet

. Sample designs of the Medical Expenditure Panel Survey Household Component, 1996-2006 and 2007-2016. Methodology Report #33. 2019. Accessed December 1, 2022. https://meps.ahrq.gov/data_files/publications/mr33/mr33.shtml

26.

von Elm

Altman

Egger

Pocock

Gøtzsche

Vandenbroucke

JP.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349. doi:10.1016/j.jclinepi.2007.11.008

27.

Agency for Healthcare Research and Quality. MEPS HC-213A Prescribed Medicines Documentation. 2021. Accessed April 7, 2023. https://meps.ahrq.gov/data_stats/download_data/pufs/h213a/h213adoc.shtml

28.

Agency for Healthcare Research and Quality. MEPS HC-214 2019 Medical Conditions Documentation. 2021. Accessed April 7, 2023. https://meps.ahrq.gov/data_stats/download_data/pufs/h214/h214doc.shtml

29.

Stokes

Lundberg

Hempstead

Berry

Baker

Preston

SH.

Obesity and incident prescription opioid use in the U.S., 2000-2015. Am J Prev Med. 2020;58(6):766-775. doi:10.1016/j.amepre.2019.12.018

30.

Bart

Maintenance medication for opiate addiction: the foundation of recovery. J Addict Dis. 2012;31(3):207-225. doi:10.1080/10550887.2012.694598

31.

Wakeman

Larochelle

Ameli

, et al. Comparative effectiveness of different treatment pathways for opioid use disorder. JAMA Network Open. 2020;3(2):e1920622. doi:10.1001/jamanetworkopen.2019.20622

32.

Agency for Healthcare Research and Quality. 2023. Using appropriate price indices for analyses of health care expenditures or income across multiple years. Accessed April 7, 2023. https://meps.ahrq.gov/about_meps/Price_Index.shtml#t1a1

33.

Glasheen

Cordier

Gumpina

Haugh

Davis

Renda

Charlson Comorbidity Index: ICD-9 update and ICD-10 translation. Am Health Drug Benefits. 2019;12(4):188-197.

34.

Agency for Healthcare Research and Quality. MEPS-HC Summary Data Tables Technical Notes. 2013. Accessed April 7, 2023. https://meps.ahrq.gov/survey_comp/hc_technical_notes.shtml

35.

Lumley

Analysis of complex survey samples. J Stat Softw. 2004;9:1-19. doi:10.18637/jss.v009.i08

36.

Belotti

Deb

Manning

Norton

EC.

Twopm: Two-Part Models. Stata J. 2015;15(1):3-20. doi:10.1177/1536867X1501500102

37.

Boulton

Williford

Analyzing skewed continuous outcomes with many zeros: a tutorial for social work and youth prevention science researchers. J Soc Soc Work Res. 2018;9(4):721-740. doi:10.1086/701235

38.

Neelon

O’Malley

Smith

VA.

Modeling zero-modified count and semicontinuous data in health services research part 2: case studies. Stat Med. 2016;35(27):5094-5112. doi:10.1002/sim.7063

39.

Khan

Bykov

Barnett

Glynn

Vine

Gagne

JJ.

Comparative risk of opioid overdose with concomitant use of prescription opioids and skeletal muscle relaxants. Neurology. 2022;99(13):e1432-e1442. doi:10.1212/WNL.0000000000200904

40.

Soprano

Hennessy

Bilker

Leonard

CE.

Assessment of physician prescribing of muscle relaxants in the United States, 2005-2016. JAMA Netw Open. 2020;3(6):e207664. doi:10.1001/jamanetworkopen.2020.7664

41.

Vadiei

Bhattacharjee

Concurrent opioid and benzodiazepine utilization patterns and predictors among community-dwelling adults in the United States. Psychiatr Serv. 2020;71(10):1011-1019. doi:10.1176/appi.ps.201900446

42.

Sun

Dixit

Humphreys

Darnall

Baker

Mackey

Association between concurrent use of prescription opioids and benzodiazepines and overdose: retrospective analysis. BMJ. 2017;356:j760. doi:10.1136/bmj.j760

43.

Johansen

ME.

Gabapentinoid use in the United States 2002 through 2015. JAMA Intern Med. 2018;178(2):292-294. doi:10.1001/jamainternmed.2017.7856

44.

IMS Institute for Healthcare Informatics. Medicines use and spending shifts: a review of the use of medicines in the US in 2014. 2015. Accessed April 7, 2023. https://www.iqvia.com/-/media/iqvia/pdfs/institute-reports/medicines-use-and-spending-shifts-in-the-us-in-2014.pdf

45.

Pauly

Delcher

Slavova

Lindahl

Talbert

Freeman

PR.

Trends in gabapentin prescribing in a commercially insured U.S. adult population, 2009-2016. J Manag Care Spec Pharm. 2020;26(3):246-252. doi:10.18553/jmcp.2020.26.3.246

46.

Serpell

; Neuropathic Pain Study Group. Gabapentin in neuropathic pain syndromes: a randomised, double-blind, placebo-controlled trial. Pain. 2002;99(3):557-566. doi:10.1016/S0304-3959(02)00255-5

47.

Goodman

Brett

AS.

Gabapentin and pregabalin for pain—is increased prescribing a cause for concern?

N Engl J Med. 2017;377(5):411-414. doi:10.1056/NEJMp1704633

48.

Kouzoukas

Lazio

. Advance Notice of Methodological Changes for Calendar Year (CY) 2019 for Medicare Advantage (MA) Capitation Rates, Part C and Part D Payment Policies and 2019 draft Call Letter. 2018. Centers for Medicare and Medicaid Services. Accessed April 7,2023. https://www.cms.gov/medicare/health-plans/medicareadvtgspecratestats/downloads/advance2019part2.pdf

49.

Margolis

Cao

Onukwugha

, et al. Healthcare utilization and cost effects of prior authorization for pregabalin in commercial health plans. Am J Manag Care. 2010;16(6):447-456.

50.

Peckham

Fairman

Sclar

DA.

Policies to mitigate nonmedical use of prescription medications: how should emerging evidence of gabapentin misuse be addressed?

Expert Opin Drug Saf. 2018;17(5):519-523. doi:10.1080/14740338.2017.1390081

51.

Schmidt

Ruchelli

Mackey

Carroll

IR.

Perioperative gabapentinoids: choice of agent, dose, timing, and effects on chronic postsurgical pain. Anesthesiology. 2013;119(5):1215-1221. doi:10.1097/ALN.0b013e3182a9a896

52.

Hawkins

Goldberg

Malte

Saxon

AJ.

New coprescription of opioids and benzodiazepines and mortality among Veterans Affairs patients with posttraumatic stress disorder. J Clin Psychiatry. 2019;80(4):18m12689. doi:10.4088/JCP.18m12689

53.

Weisberg

Gordon

Barry

, et al. Long-term prescription opioids and/or benzodiazepines and mortality among HIV-infected and uninfected patients. J Acquir Immune Defic Syndr. 2015;69(2):223-233. doi:10.1097/QAI.0000000000000591

54.

IQVIA Institute. Prescription opioid trends in the United States. 2020. Accessed April 7, 2023. https://www.iqvia.com/insights/the-iqvia-institute/reports/prescription-opioid-trends-in-the-united-states

55.

CDC Injury Center. Drug overdose. United States opioid dispensing rate maps. 2022. Accessed April 7, 2023. https://www.cdc.gov/drugoverdose/rxrate-maps/index.html

56.

Baehren

Marco

Droz

Sinha

Callan

Akpunonu

A statewide prescription monitoring program affects emergency department prescribing behaviors. Ann Emerg Med. 2010;56(1):19-23.e1-e3. doi:10.1016/j.annemergmed.2009.12.011

57.

Bao

Pan

Taylor

, et al. Prescription drug monitoring programs are associated with sustained reductions in opioid prescribing by physicians. Health Aff (Millwood). 2016;35(6):1045-1051. doi:10.1377/hlthaff.2015.1673

58.

Bonnie

Ford

Phillips

, eds. Pain Management and the Opioid Epidemic: Balancing Societal and Individual Benefits and Risks of Prescription Opioid Use. National Academies Press; 2017. doi:10.17226/24781

59.

PDMP Center of Excellence. Briefing on PDMP Effectiveness. 2013. Accessed April 7, 2023. https://www.ojp.gov/pdffiles1/bja/247133.pdf

60.

Davis

Lieberman

Hernandez-Delgado

Suba

Laws limiting the prescribing or dispensing of opioids for acute pain in the United States: a national systematic legal review. Drug Alcohol Depend. 2019;194:166-172. doi:10.1016/j.drugalcdep.2018.09.022

61.

Reisman

Shenoy

Atherly

Flowers

CR.

Prescription opioid usage and abuse relationships: an evaluation of state prescription drug monitoring program efficacy. Subst Abuse. 2009;3:41-51.

62.

Urban Institute. Medicaid coverage of effective treatment for opioid use disorder. 2017. Accessed April 7, 2023. https://www.urban.org/research/publication/medicaid-coverage-effective-treatment-opioid-use-disorder

63.

Wen

Hockenberry

Borders

Druss

BG.

Impact of Medicaid expansion on Medicaid-covered utilization of buprenorphine for opioid use disorder treatment. Med Care. 2017;55(4):336-341. doi:10.1097/MLR.0000000000000703

64.

Wen

Soni

Hollingsworth

, et al. Association between Medicaid expansion and rates of opioid-related hospital use. JAMA Intern Med. 2020;180(5):753-759. doi:10.1001/jamainternmed.2020.0473

65.

Rep. Bonamici S [D O 1]. H.R.34—114th Congress (2015-2016): 21st Century Cures Act. 2016. Accessed April 7, 2023. https://www.congress.gov/bill/114th-congress/house-bill/34

66.

Rep. Connolly GE [D V 11]. Text—H.R.2617—117th Congress (2021-2022): Consolidated Appropriations Act, 2023. 2022. Accessed April 7, 2023. https://www.congress.gov/bill/117th-congress/house-bill/2617

67.

Rep. Walden G [R O 2]. H.R.6—115th Congress (2017-2018): SUPPORT for Patients and Communities Act. 2018. Accessed April 7, 2023. https://www.congress.gov/bill/117th-congress/house-bill/6

68.

Sen. Whitehouse S [D R]. Text—S.524—114th Congress (2015-2016): Comprehensive Addiction and Recovery Act of 2016. 2016. Accessed April 7, 2023. https://www.njleg.state.nj.us/bill-search/2022/S524

69.

Substance Abuse and Mental Health Services Administration. Removal of DATA Waiver (X-Waiver) Requirement. 2023. Accessed April 7, 2023. https://medi-calrx.dhcs.ca.gov/cms/medicalrx/static-assets/documents/provider/dur/educational-articles/dured_Removal_of_Data_Waiver_X-Waiver_Requirement.pdf

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.95 MB