Abstract
Objective.
There are potential risks and benefits of combining benzodiazepine (BZD) and selective-serotonin reuptake inhibitor (SSRI) therapy at anxiety disorder treatment onset. We investigated how often adolescents and young adults with anxiety disorders simultaneously initiate BZD treatment with SSRI treatment and examined whether SSRI treatment duration varies by simultaneous BZD initiation.
Methods.
In a United States commercial claims database (1/2008–12/2016), we identified adolescents (10–17) and young adults (18–24) with ICD-9-CM/ICD-10-CM anxiety disorder diagnoses initiating SSRI treatment, without past-year SSRI and BZD treatment. We defined simultaneous initiation as filling a new BZD prescription on the date of SSRI initiation. We estimated time to SSRI treatment discontinuation and used stabilized inverse probability of treatment weighting for adjusted estimates.
Results.
The study included 94,399 adolescents and 130,971 young adults initiating SSRI treatment with an anxiety disorder. Four-percent of adolescents and 17% of young adults simultaneously initiated BZD treatment, varying by age, anxiety disorder, comorbidities, healthcare utilization, and provider type. Simultaneous BZD initiation among SSRI initiators declined from 2008 to 2016. SSRI treatment duration was similar in initiators of simultaneous therapy vs. SSRI monotherapy: ≥6 months in adolescents (55% vs. 56%, respectively) and in young adults (39% vs. 40%). Nine-percent of simultaneous initiators continued BZDs for ≥6 months.
Conclusion.
Simultaneous initiation of BZD and SSRI treatment is relatively common in young adults with anxiety disorders and was not associated with longer SSRI persistence. Given risks of BZD treatment, potential benefits and risks of adding a BZD at SSRI treatment initiation must be carefully weighed.
INTRODUCTION
Antidepressants are typically recommended as the first-line pharmacotherapy for anxiety disorders in children and adults.1–6 Benzodiazepines (BZDs) are prescribed alone or with antidepressants for acute treatment to rapidly reduce moderate to severe anxiety symptoms or as an augmentation strategy to treat residual anxiety symptoms.1,7,8 In children, there is no evidence that BZDs are efficacious in treating anxiety disorders, with only a few trials having been conducted.5,9 The recently released 2020 American Academy of Child and Adolescent Psychiatry (AACAP) guideline for pediatric anxiety calls for more research on the benefits and harms of BZDs given insufficient information.6
The co-prescribing of BZDs in combination with antidepressants at treatment onset has become a pharmacological treatment practice for adult depression. An estimated 11% of adults with depression starting antidepressant treatment simultaneously initiated BZD treatment (2001–2014).10 For adults, there are limited studies on the benefits of combination BZD and antidepressant therapies at treatment onset. In a review of randomized trials in adults with depression, those with combination therapy were less likely to drop out and had improved short-term response than those on antidepressants alone.11 In patients with panic disorder, one study found greater improvement in symptoms in the first weeks of treatment with combined treatment.8 In patients with comorbid anxiety and depression, it has been suggested that adding a BZD to selective-serotonin reuptake inhibitors (SSRI) treatment may reduce SSRI-induced anxiety or agitation or improve SSRI adherence.7 It is unknown how common the practice of co-prescribing is in children and young adults with anxiety disorders.
Whether SSRI treatment duration is longer in those simultaneously initiating BZD treatment is unclear but could inform decisions around co-prescribing. After initiating antidepressant treatment for an anxiety disorder, it is recommended that a trial period of 8–10 weeks, if effective, be followed by maintenance treatment for 9–12 months.3 However, SSRI adherence is often low; 43% of adults12 and 58% of children13 with anxiety disorders were adherent at 6 months.
While BZDs may offer benefit in managing anxiety disorder symptoms, they are recommended less frequently than SSRIs because of potential harms including side effects, such as drowsiness or memory problems, risk of dependency, risk of overdose, and misuse.1,14–17 Young adults commonly misuse BZDs, or take them outside medical direction, with approximately 5% reporting past-year misuse.18 Further, many use BZDs with other depressants, such as alcohol or opioids,18,19 raising increased safety concerns for prescribing BZD treatment. In September 2020, the US Food and Drug Administration announced an anticipated update to the Boxed Warning for BZDs on the potential for abuse, addiction, and other serious risks.20
Among adolescents (10–17 years) and young adults (18–24 years) with anxiety disorders initiating SSRI treatment, we aimed to a) estimate how many patients simultaneously initiate prescription BZD treatment and whether this varied from 2008 to 2016 and b) estimate whether SSRI treatment duration varies by whether a patient simultaneously initiates BZD treatment and how long BZD treatment is continued in simultaneous initiators.
METHOD
Dataset & study population.
We used the 2008–2016 MarketScan Commercial Claims and Encounters database, which covers individuals with employer-sponsored health insurance and their dependents across the United States. The MarketScan dataset is owned by IBM and can be accessed for research.21,22 Patient-level details were available on insurance enrollment, outpatient and inpatient services, and outpatient dispensed prescriptions.
We identified adolescents (10–17) and young adults (18–24) initiating SSRI treatment, who had not filled BZD or SSRI prescriptions for at least one year. Continuous insurance enrollment was required during the prior year and SSRI treatment was defined with records of dispensed prescriptions for citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, sertraline, and vilazodone. We required SSRI initiators to have a recent anxiety disorder diagnosis (<=30 days) defined with ICD-9-CM and ICD-10-CM codes (Table 1 and 2 footnotes). We excluded SSRI new users with diagnoses for bipolar disorder, schizophrenia, other psychoses, personality disorder, epilepsy/convulsions, or autistic disorder in the prior year or with an inpatient admission in the prior 30 days. Adolescents and young adults with any other comorbid diagnosis were allowed.
Table 1.
Adolescents (10–17 years) with anxiety disorders initiating SSRI treatment by simultaneous initiation with BZD treatment
| Patient characteristic | SSRI + BZD N=4,068 | SSRI alone N=90,331 | Standard difference | SSRI + BZD Row % (of total) |
|---|---|---|---|---|
| No. (%) | No. (%) | |||
| Male | 1,450 (35.6) | 32,415 (35.9) | 0.0 | 4.3 |
| Female | 2,618 (64.4) | 57,916 (64.1) | 0.0 | 4.3 |
| Age at SSRI initiation, mean (SD) | 15.2 (1.9) | 14.4 (2.2) | 0.4 | |
| 10–13 y | 757 (18.6) | 27,785 (30.8) | −0.3 | 2.7 |
| 14–17 y | 3,311 (81.4) | 62,546 (69.2) | 0.3 | 5.0 |
| Anxiety disorder diagnosis (prior 30d)a | ||||
| Unspecified anxiety | 1,464 (36.0) | 37,867 (41.9) | −0.1 | 3.7 |
| Generalized anxiety disorder | 950 (23.4) | 24,251 (26.8) | −0.1 | 3.8 |
| Panic disorder | 562 (13.8) | 3,622 (4.0) | 0.3 | 13.4 |
| Adjustment disorder with anxiety | 318 (7.8) | 11,977 (13.3) | −0.2 | 2.6 |
| Acute stress disorder | 59 (1.5) | 1,194 (1.3) | 0.0 | 4.7 |
| Social phobia | 93 (2.3) | 2,530 (2.8) | 0.0 | 3.5 |
| Other anxiety disorder | 116 (2.9) | 3,040 (3.4) | 0.0 | 3.7 |
| Multiple specific diagnoses | 506 (12.4) | 5,850 (6.5) | 0.2 | 8.0 |
| Provider of anxiety diagnosis | ||||
| Psychiatry | 957 (23.5) | 19,438 (21.5) | 0.0 | 4.7 |
| Psychology | 359 (8.8) | 13,780 (15.3) | −0.2 | 2.5 |
| Family practitioner | 898 (22.1) | 14,528 (16.1) | 0.2 | 5.8 |
| Pediatrician | 532 (13.1) | 17,138 (19.0) | −0.2 | 3.0 |
| Other general providerb | 436 (10.7) | 7,913 (8.8) | 0.1 | 5.2 |
| Other provider type | 479 (11.8) | 10,664 (11.8) | 0.0 | 4.3 |
| Unknown or multiple | 407 (10.0) | 6,870 (7.6) | 0.1 | 5.6 |
| Any comorbid psychiatric diagnosis, 1y | 2,011 (49.4) | 54,846 (60.7) | −0.2 | 3.5 |
| Depression | ||||
| Recent (0–30d) | 1,062 (26.1) | 27,820 (30.8) | −0.1 | 3.7 |
| Past (31–365d), no recent | 156 (3.8) | 4,023 (4.5) | 0.0 | 3.7 |
| Sleep disorder | ||||
| Recent (0–30d) | 172 (4.2) | 2,790 (3.1) | 0.1 | 5.8 |
| Past (31–365d), no recent | 92 (2.3) | 2,311 (2.6) | 0.0 | 3.8 |
| ADHD | ||||
| Recent (0–30d) | 224 (5.5) | 11,978 (13.3) | −0.3 | 1.8 |
| Past (31–365d), no recent | 145 (3.6) | 5,348 (5.9) | −0.1 | 2.6 |
| PTSD, 1y | 67 (1.6) | 1,211 (1.3) | 0.0 | 5.2 |
| Adjustment disorder, 1y | 281 (6.9) | 7,476 (8.3) | −0.1 | 3.6 |
| Conduct disorder, 1y | 90 (2.2) | 4,088 (4.5) | −0.1 | 2.2 |
| OCD, 1y | 123 (3.0) | 2,911 (3.2) | 0.0 | 4.1 |
| Substance use disorder, 1y | 103 (2.5) | 1,887 (2.1) | 0.0 | 5.2 |
| Tobacco/nicotine dependence, 1y | 28 (0.7) | 467 (0.5) | 0.0 | 5.7 |
| Suicidal ideation, self-harm, 1y | 54 (1.3) | 1,588 (1.8) | 0.0 | 3.3 |
| Recent (90d) prescriptions | ||||
| Stimulant | 225 (5.5) | 11,442 (12.7) | −0.3 | 1.9 |
| Antidepressant (excludes SSRIs) | 190 (4.7) | 4,618 (5.1) | 0.0 | 4.0 |
| Hydroxyzine | 77 (1.9) | 3,014 (3.3) | −0.1 | 2.5 |
| Hypnotic | 38 (0.9) | 355 (0.4) | 0.1 | 9.7 |
| Anticonvulsant | 52 (1.3) | 1,386 (1.5) | 0.0 | 3.6 |
| Opioid | 314 (7.7) | 4,576 (5.1) | 0.1 | 6.4 |
| Psychotherapy claims, recent (0–30d) | 1,420 (34.9) | 38,918 (43.1) | −0.2 | 3.5 |
| Inpatient admissions, 31–365dc | ||||
| Psychiatric-related | 30 (0.7) | 794 (0.9) | 0.0 | 3.6 |
| Non-psychiatric related | 98 (2.4) | 1,252 (1.4) | 0.1 | 7.3 |
| ED visit, recent (0–30d) | 662 (16.3) | 9,688 (10.7) | 0.2 | 6.4 |
| Preventative outpatient visit, 1y | 2,071 (50.9) | 49,757 (55.1) | −0.1 | 4.0 |
BZD: benzodiazepine; SSRI: selective serotonin reuptake inhibitor; d: day; y: year; SD: standard deviation; ED: emergency department; OCD: Obsessive–compulsive disorder; PTSD: Post-traumatic stress disorder; ADHD: Attention deficit hyperactivity disorder
Sample includes adolescents (10–17 years) identified from a commercial claims database (2008–2016); BZD+SSRI row percent by geographical region: Northeast=5.4%, North Central=4.0%, South=3.7%, West=4.9%
Anxiety disorder diagnoses: ICD-9-CM: 293.84, 300.0, 300.2, 308, 309.21, 309.24, 309.28, 313.23, 799.21; ICD-10-CM: F06.4, F40, F41, F43.0, F43.22, F43.23, F93.0, F94.0, R45.0, R45.7
Other general provider categories, BZD+SSRI row percent: internal medicine=5.9%, nurse practitioner=3.2%, physician assistant=5.6%
Adolescents with a recent inpatient admission (0–30d) were excluded
Table 2.
Young adults (18–24 years) with anxiety disorders initiating SSRI treatment by simultaneous initiation with BZD treatment
| Patient characteristic | SSRI + BZD N=21,902 | SSRI alone N=109,069 | Standard difference | SSRI + BZD Row % (of total) |
|---|---|---|---|---|
| No. (%) | No. (%) | |||
| Male | 7,578 (34.6) | 37,974 (34.8) | 0.0 | 16.6 |
| Female | 14,324 (65.4) | 71,095 (65.2) | 0.0 | 16.8 |
| Age at SSRI initiation, mean (SD) | 21.1 (1.9) | 20.7 (2.0) | 0.2 | |
| 18–20 y | 8,710 (39.8) | 55,188 (50.6) | −0.2 | 13.6 |
| 21–24 y | 13,192 (60.2) | 53,881 (49.4) | 0.2 | 19.7 |
| Anxiety disorder diagnosis (prior 30d)a | ||||
| Unspecified anxiety | 10,653 (48.6) | 56,185 (51.5) | −0.1 | 15.9 |
| Generalized anxiety disorder | 4,466 (20.4) | 25,702 (23.6) | −0.1 | 14.8 |
| Panic disorder | 3,025 (13.8) | 5,541 (5.1) | 0.3 | 35.3 |
| Adjustment disorder with anxiety | 1,019 (4.7) | 8,321 (7.6) | −0.1 | 10.9 |
| Acute stress disorder | 468 (2.1) | 2,580 (2.4) | 0.0 | 15.4 |
| Social phobia | 226 (1.0) | 2,210 (2.0) | −0.1 | 9.3 |
| Other anxiety disorder | 675 (3.1) | 4,496 (4.1) | −0.1 | 13.1 |
| Multiple specific diagnoses | 1,370 (6.3) | 4,034 (3.7) | 0.1 | 25.4 |
| Provider of anxiety diagnosis | ||||
| Psychiatry | 2,271 (10.4) | 11,096 (10.2) | 0.0 | 17.0 |
| Psychology | 725 (3.3) | 7,956 (7.3) | −0.2 | 8.4 |
| Family practitioner | 9,150 (41.8) | 41,392 (38.0) | 0.1 | 18.1 |
| Pediatrician | 400 (1.8) | 4,959 (4.5) | −0.2 | 7.5 |
| Other general providerb | 4,433 (20.2) | 20,257 (18.6) | 0.0 | 18.0 |
| Other provider type | 2,529 (11.5) | 13,195 (12.1) | 0.0 | 16.1 |
| Unknown or multiple | 2,394 (10.9) | 10,214 (9.4) | 0.1 | 19.0 |
| Any comorbid psychiatric diagnosis, 1y | 8,536 (39.0) | 48,908 (44.8) | −0.1 | 14.9 |
| Depression | ||||
| Recent (0–30d) | 4,846 (22.1) | 27,581 (25.3) | −0.1 | 14.9 |
| Past (31–365d), no recent | 568 (2.6) | 3,643 (3.3) | 0.0 | 13.5 |
| Sleep disorder | ||||
| Recent (0–30d) | 1,415 (6.5) | 5,361 (4.9) | 0.1 | 20.9 |
| Past (31–365d), no recent | 408 (1.9) | 2,380 (2.2) | 0.0 | 14.6 |
| ADHD | ||||
| Recent (0–30d) | 641 (2.9) | 6,613 (6.1) | −0.2 | 8.8 |
| Past (31–365d), no recent | 636 (2.9) | 3,766 (3.5) | 0.0 | 14.4 |
| PTSD, 1y | 279 (1.3) | 1,113 (1.0) | 0.0 | 20.0 |
| Adjustment disorder, 1y | 560 (2.6) | 3,470 (3.2) | 0.0 | 13.9 |
| Conduct disorder, 1y | 70 (0.3) | 631 (0.6) | 0.0 | 10.0 |
| OCD, 1y | 245 (1.1) | 1,439 (1.3) | 0.0 | 14.5 |
| Substance use disorder, 1y | 854 (3.9) | 4,509 (4.1) | 0.0 | 15.9 |
| Tobacco/nicotine dependence, 1y | 1,064 (4.9) | 4,458 (4.1) | 0.0 | 19.3 |
| Suicidal ideation, self-harm, 1y | 93 (0.4) | 656 (0.6) | 0.0 | 12.4 |
| Recent (90d) prescriptions | ||||
| Stimulant | 867 (4.0) | 7,299 (6.7) | −0.1 | 10.6 |
| Antidepressant (excludes SSRIs) | 907 (4.1) | 6,984 (6.4) | −0.1 | 11.5 |
| Hydroxyzine | 362 (1.7) | 4,779 (4.4) | −0.2 | 7.0 |
| Hypnotic | 419 (1.9) | 1,980 (1.8) | 0.0 | 17.5 |
| Anticonvulsant | 240 (1.1) | 1,724 (1.6) | 0.0 | 12.2 |
| Opioid | 2,517 (11.5) | 9,852 (9.0) | 0.1 | 20.3 |
| Psychotherapy claims, recent (0–30d) | 2,472 (11.3) | 19,380 (17.8) | −0.2 | 11.3 |
| Inpatient admissions, 31–365dc | ||||
| Psychiatric-related | 66 (0.3) | 536 (0.5) | 0.0 | 11.0 |
| Non-psychiatric related | 777 (3.5) | 3,614 (3.3) | 0.0 | 17.7 |
| ED visit, recent (0–30d) | 3,228 (14.7) | 12,059 (11.1) | 0.1 | 21.1 |
| Preventative outpatient visit, 1y | 8,580 (39.2) | 46,949 (43.0) | −0.1 | 15.5 |
BZD: benzodiazepine; SSRI: selective serotonin reuptake inhibitor; d: day; y: year; SD: standard deviation; ED: emergency department; OCD: Obsessive–compulsive disorder; PTSD: Post-traumatic stress disorder; ADHD: Attention deficit hyperactivity disorder
Sample includes young adults (18–24 years) identified from a commercial claims database (2008–2016); BZD+SSRI row percent by geographical region: Northeast=16.3%, North Central=16.5%, South=16.0%, West=19.1%
Anxiety disorder diagnoses: ICD-9-CM: 293.84, 300.0, 300.2, 308, 309.21, 309.24, 309.28, 313.23, 799.21; ICD-10-CM: F06.4, F40, F41, F43.0, F43.22, F43.23, F93.0, F94.0, R45.0, R45.7
Other general provider categories, BZD+SSRI row percent: internal medicine=19.3%, nurse practitioner=13.1%, physician assistant=13.5%
Young adults with a recent inpatient admission (0–30d) were excluded
Simultaneous BZD initiation.
We defined simultaneous SSRI and BZD initiation as SSRI initiators who filled a BZD prescription on the same date as SSRI initiation. BZD treatment was defined with records of dispensed prescriptions for alprazolam, chlordiazepoxide, clobazam, clonazepam, clorazepate, diazepam, estazolam, flurazepam, halazepam, lorazepam, midazolam, oxazepam, prazepam, quazepam, temazepam, and triazolam. For a secondary analysis we examined how many youth filled a BZD prescription up to 90 days following SSRI initiation.
Treatment duration.
We estimated SSRI treatment duration based on prescription refill dates and prescription days supply. If there was no prescription refill 30 days after the previous prescription’s days supply ended, the individual was considered to have discontinued treatment. In simultaneous initiators, we also estimated BZD treatment duration.
Additional measures.
Patient characteristics were identified from inpatient and outpatient records (ICD-9-CM, ICD-10-CM codes, CPT codes) and records of dispensed prescriptions in the year before SSRI initiation. We included measures for age, sex, anxiety disorder, comorbid depression, other mental health diagnoses, provider type on the anxiety diagnosis, psychotherapy claims, healthcare utilization, prescription medication, and non-psychiatric diagnoses (Tables 1 and 2).
Analysis.
We described adolescents and young adults by simultaneous BZD initiation status. Per year, we estimated the proportion of SSRI initiators who simultaneously initiated BZD treatment. For a sensitivity analysis, we standardized trends by age, sex, and geographical region (New England, Middle Atlantic, South Atlantic, East South Central, West North Central, West South Central, Mountain, Pacific) to account for possible changes in the underlying source population. We estimated the survival function of SSRI treatment persistence at 3-, 6-, and 9-months with Nelson–Aalen estimator stratified by simultaneous BZD initiation status, censoring at insurance disenrollment or end of data. For adjusted analyses of treatment persistence, we estimated propensity scores for simultaneous BZD initiation with logistic regression including factors associated with treatment type and adherence12,23–25 and applied stabilized inverse probability of treatment weights (SIPTW). Analyses were completed separately in adolescents and young adults. Results were consistent under limited and more inclusive variable selection; we present results from the more inclusive propensity score model (sex, age, year of SSRI initiation, mental health diagnoses, provider type diagnosing anxiety, psychotherapy claims, other medication use, other medical diagnoses, inpatient admissions, emergency department visits, outpatient visits, geographical region; Table 3 footnote).
Table 3.
SSRI treatment persistence by simultaneous BZD initiation status
| Adolescents (10–17 years) | Young adults (18–24 years) | |||
|---|---|---|---|---|
| SSRI + BZD N=4,068 | SSRI alone N=90,331 | SSRI + BZD N=21,902 | SSRI alone N=109,069 | |
| Proportion remaining on SSRI treatment at: | ||||
| 3 months | 74% (73–75) | 78% (78–79) | 64% (63–65) | 66% (66–67) |
| 6 monthsa | 52% (50–54) | 57% (56–57) | 38% (37–38) | 40% (40–41) |
| 9 months | 40% (38–42) | 43% (43–43) | 25% (25–26) | 28% (28–28) |
| Weightedb | ||||
| 3 months | 76% (75–78) | 78% (78–78) | 65% (65–66) | 66% (66–66) |
| 6 months | 55% (53–57) | 56% (56–57) | 39% (39–40) | 40% (40–40) |
| 9 months | 43% (41–45) | 43% (43–43) | 27% (26–27) | 28% (27–28) |
| Proportion remaining on BZD treatment at: | ||||
| 3 months | 25% (23–26) | - | 23% (23–24) | - |
| 6 monthsc | 9% (9–11) | - | 9% (9–10) | - |
| 9 months | 6% (5–6) | - | 6% (5–6) | - |
BZD: benzodiazepine; SSRI: selective serotonin reuptake inhibitor
SSRI continuation at 6 months in persons with comorbid depression: adolescents: SSRI + BZD: 56% (53–59); SSRI alone: 58% (58–59) and young adults: SSRI + BZD: 39% (37–40); SSRI alone: 42% (42–43)
Stabilized inverse probability of treatment weighting, variables in propensity score model: sex, age, year of SSRI initiation, anxiety disorder diagnosis, provider type diagnosing anxiety, psychiatric and non-psychiatric diagnoses, psychotropic and other prescription medication use, psychotherapy claims, treated injuries, suicidality, provider contact, inpatient admissions, emergency department visits, outpatient visits, geographical division
BZD continuation at 6 months in persons with comorbid depression: adolescents: 12% (10–15) and young adults: 11% (10–12)
In simultaneous initiators, we also estimated the survival function of BZD treatment persistence in the same manner. Further, in simultaneous users with 6 months of follow-up, we examined the percent with long-term BZD-treatment overall and by specific patient factors (initial BZD agent, SSRI agent, BZD days supply, age, sex, mental health diagnoses, psychotropics and opioid prescriptions, provider type, psychotherapy claims). Long-term BZD use was defined as at least 6 months of continued BZD use. We used logistic regression to examine whether factors were independently associated with long-term BZD use. Adolescents and young adults were combined for this analysis given the limited number adolescents with long-term BZD use. The study was approved by the Rutgers University Institutional Review Board.
RESULTS
Simultaneous BZD initiation.
The study cohort included 94,399 adolescents and 130,971 young adults initiating SSRI treatment with a recent anxiety disorder diagnosis. Overall, 4.3% of adolescents and 16.7% of young adults simultaneously initiated BZD treatment, with a monotonic increase by age (Figure 1). Alprazolam was the most common BZD dispensed to simultaneous initiators (adolescents: 37%, young adults: 49%) with a median 15 days supply on the initial prescription (Supplemental Table 1). An additional 4% of adolescents and 7% of young adults started BZD treatment 1 to 90 days after SSRI initiation.
Figure 1. Percentage of SSRI initiators with anxiety disorders simultaneously initiating benzodiazepine treatment stratified by age at SSRI initiation.
Privately insured adolescents (10–17 years) and young adults (18–24 years) newly initiating SSRI treatment after an anxiety disorder diagnosis, identified from a commercial claims database (2008–2016)
The highest prevalence of simultaneous initiation among SSRI initiators was in adolescents (13%) and young adults (35%) with panic disorder (row %s, Tables 1 and 2). The prevalence of simultaneous initiation was similar in adolescents (5%, 6%) and young adults (17%, 18%) diagnosed by a psychiatrist or family practitioner, with lower prevalence of simultaneous initiation seen in adolescents (3%) and young adults (8%) diagnosed by a pediatrician. Considering region, simultaneous prescribing was more common in young adults treated in the West (19% vs. 16%) and was similar by region in adolescents (4–5%; footnotes Tables 1 and 2).
Overall, simultaneous BZD initiators were older, more likely to have panic disorder, a recent ED visit, or sleep disorder, and to have been recently prescribed an opioid (adolescents: Table 1, young adults: Table 2). Simultaneous BZD initiators were less likely to have an ADHD diagnosis, a psychotherapy claim, or hydroxyzine prescriptions. Comorbid depression was slightly less common in simultaneous initiators vs. SSRI monotherapy in adolescents (26% vs. 31%, respectively) and young adults (22% vs. 25%).
Trends in simultaneous BZD initiation.
In adolescents, there was a steady decline over time in the proportion of SSRI initiators simultaneously initiating BZD treatment from 6% in 2008 to 3% in 2016 (Figure 2). From 2008–2012, 19–20% of young adult SSRI initiators simultaneously initiated BZD treatment, which declined to 12% by 2016. Trends were consistent after standardizing by age, sex, and geographical region.
Figure 2. Percentage of adolescent and young adult SSRI initiators simultaneously initiating benzodiazepine treatment stratified by year and comorbid depression.
Privately insured adolescents (10–17 years) and young adults (18–24 years) newly initiating SSRI treatment after an anxiety disorder diagnosis, identified from a commercial claims database (2008–2016)
SSRI Treatment duration.
Seventy-eight percent of adolescents (BZD+SSRI: 75%, SSRI: 78%) and 64% of young adults (BZD+SSRI: 63%, SSRI: 65%) refilled their SSRI prescription. Overall, SSRI treatment duration was similar between SSRI initiators with or without simultaneous BZD initiation (Table 3). After weighting, in adolescents, 55% (95% CI: 53–57) of simultaneous BZD initiators and 56% (95% CI: 56–57) of SSRI monotherapy initiators remained on SSRI treatment after 6-months. In young adults, after weighting, 39% (95% CI: 39–40%) of simultaneous BZD initiators and 40% (95% CI: 40–40%) of SSRI monotherapy initiators remained on SSRI treatment after 6-months.
BZD treatment duration.
Approximately one third of adolescent (32%) and young adult (30%) simultaneous initiators refilled the initial BZD prescription. From 2008 to 2016, the proportion of simultaneous initiators refilling the initial BZD prescription was highest in 2011 for adolescents (36%) and in 2012 for young adults (33%), followed by declines to 29% in adolescents and 30% in young adults in 2015 (the last year with full follow-up data).
Among simultaneous initiators, 9% (95% CI: 9–11) of adolescents and 9% (95% CI: 9–10) of young adults continued BZD treatment for 6+ months (Table 3). In the analysis of factors associated with long-term BZD use among simultaneously initiators, females were less likely to become long-term BZD users than males (aOR=0.60, Supplemental Table 2). Simultaneous initiators with a longer days supply value were more likely to become long-term BZD users as were those initiating with clonazepam, seen by a psychiatrist, with a prior substance use disorder diagnosis, and with recent use of other psychotropics.
DISCUSSION
Among adolescents and young adults with anxiety disorders initiating SSRI treatment, the simultaneous initiation of prescription BZD and SSRI treatment has declined, but remained relatively common in young adults and was not associated with lengthened SSRI treatment. Given risks of BZD treatment, the potential benefit and risk of adding BZD treatment to SSRI treatment should be carefully weighed. While the practice of BZD and SSRI co-prescribing became less frequent among SSRI initiators from 2008 through 2016, detailing this practice will hopefully encourage dialogue and development of specific recommendations for co-prescribing in young populations.
The proportion of SSRI initiators simultaneously initiating BZD treatment was higher with increasing age from 10 to 24 years (2% to 21%), with the largest increase from age 17 to 18 years. This mirrors broader BZD prescribing distributions in young adults and adolescents.26 Several factors may explain the age difference, including variation in evidence and guidelines between children and adults.1,2,5 There is evidence that BZDs are efficacious in treating anxiety disorders in adults.27 In the limited trials in children, BZDs were not efficacious in treating anxiety disorders.5,9,28 However, few trials in young populations exist and there are no trials with BZDs as adjunctive treatment with SSRIs at treatment onset. While simultaneous prescribing was more common in young adults across provider types, provider types more commonly seen by young adults (i.e. family practitioners) had higher prevalence of simultaneous initiation. Simultaneous BZD initiation may also be more common in young adults due to increased comorbidity, presentation acuity, or for targeting of specific symptoms.
We observed a decline in the proportion of SSRI initiators simultaneously initiating BZD treatment. One reason for the decline may be increasing concerns surrounding BZD-related overdoses, particularly when combined with opioids or alcohol, which may limit BZD prescribing.29 The FDA added a Boxed Warning to BZDs in 2016 for a risk of death when combining BZDs and opioids.30 Additionally, in findings from a study of adolescents with treatment-resistant depression, adjunctive BZD treatment (non-randomly assigned) was associated with suicidal adverse events (n=6/10) compared to adolescents without adjunctive BZD treatment (n=42/324).31 Observed associations between BZD treatment and suicide attempts could be related to confounding by indication,32,33 yet these concerns may have nevertheless affected BZD co-prescribing in youth with anxiety disorders. Rates of substance use in young people34 may have also raised concerns for prescribing physicians. Further, the decline could be, at least in part, a function of the base population (adolescents and young adults starting SSRI treatment for anxiety) shifting over the study period. For example, this could include a decline in the average severity of SSRI initiators.
Other concerns surrounding BZDs such as dependency, misuse, or side effects may influence prescribing practices. Given dependency concerns, BZDs are not recommended for routine use27 and adolescents receiving BZD treatment should be monitored for BZD misuse.16 In our study, the majority of simultaneous initiators discontinued BZD treatment early, as recommended. However, 9% of simultaneous initiators continued BZD treatment for at least 6 months. The factors associated with long-term BZD use apply specifically to simultaneous initiators with anxiety disorders and likely differ from factors associated with long-term BZD use in general. While persistent BZD treatment may be a key component of treatment for certain patients3 and it may reflect severity of psychiatric symptoms and an evolving course of clinical presentation, the subset progressing to long-term BZD treatment warrants closer examination.
BZDs are typically used at the initiation of antidepressant treatment for rapid, short-term relief of symptoms and SSRI-induced agitation, and it has been suggested that they may improve antidepressant adherence by ameliorating these troublesome symptoms.7 In randomized trials for adult depression, simultaneous users were less likely to drop out due to side effects than antidepressant monotherapy.11 However, in older adults with GAD, those prescribed BZDs with escitalopram treatment were more likely to drop out of the study than those without BZD treatment, with no difference in treatment response or adherence.35 We cannot evaluate symptoms or adherence in the first days and weeks, when BZD treatment may have the largest potential for benefit. However, our results suggest that improving SSRI treatment persistence after 3 months may not be a benefit of simultaneous treatment, as we did not observe longer SSRI treatment in adolescents and young adults who simultaneously initiated BZD treatment. This was also observed in a similar study of adults with depression, where simultaneous antidepressant and BZD initiation was not associated with longer antidepressant treatment.10 Other modifiable factors12,24,25 may be more effective in improving SSRI persistence with lower risk than BZDs.
Simultaneous BZD treatment initiation was highest in adolescents (13%) and young adults (35%) with panic disorder initiating SSRI treatment. This is consistent with stronger evidence for BZDs as a treatment for panic disorder.1 A small trial of adults with moderate to severe panic disorder observed greater improvement in symptoms in the first weeks of treatment with combined clonazepam and sertraline treatment compared with sertraline monotherapy; still, both groups had similar levels of clinical improvement by study end (12 weeks).8 Practice guidelines for adult panic disorder state that combined BZD and antidepressant treatment can help with rapid symptom control;1 however, not all guidelines recommend BZDs for treating panic disorder given risks of BZD treatment.4
In addition to age and anxiety disorder, simultaneous initiation varied by comorbidity, prescription medications, and healthcare utilization and was slightly less common in adolescents and young adults with comorbid depression. There is limited information on the potential benefits to BZD and SSRI co-prescription for adults with major depressive disorder11 or comorbid anxiety and depression.7 In adults with anxiety disorders, psychiatrists are more likely to prescribe BZDs than primary care providers.36 As compared to youth seen by psychiatrists, we found youth diagnosed by family practitioners were more likely and those diagnosed by pediatricians were less likely to receive BZD and SSRI treatment. Adolescents and young adults who received psychotherapy prior to SSRI initiation were less likely to be co-prescribed BZD treatment, perhaps an indication of an alternative treatment strategy. Individuals receiving psychotherapy prior to SSRI initiation may be less likely to have an acute need for rapid symptom reduction, which is a situation in which BZD treatment may be more likely to be prescribed. In our sample, there was an increased prevalence of simultaneous BZD initiation in youth with a recent ED visit; research examining ED visits in which anxiety was the primary diagnosis (all ages), found anxiolytics and BZDs to be commonly prescribed (53% of visits when medication was provided).37 The higher co-prescribing could be related to a greater need for acute symptom remission in individuals recently presenting to the ED. Overall, the variation in simultaneous BZD initiation by psychotherapy receipt, provider type, and ED visit is likely at least in part related to anxiety disorder severity. Data resources with more detailed anxiety symptom measures can inform the role of severity in the decision to co-prescribe.
Despite recommendations that BZD treatment should be avoided in persons with substance use,3,5 3% of adolescents and 4% of young adults simultaneously initiating BZD and SSRI treatment had a recorded substance use disorder diagnosis within the past year. These individuals were also more likely to become long-term BZD users. While we cannot determine if the prescriber knew of the prior diagnosis, this suggests prescribing improvements are needed in youth with substance use disorders. A closer examination of alcohol and drug use prior to SSRI initiation could direct future efforts, particularly as substance use disorders are almost certainly underdiagnosed in our sample and we lack details on alcohol and drug use.
Limitations of this work should be considered. BZD and SSRI prescriptions could have been prescribed by separate providers and not intended for concurrent use; we do not have details on the prescriber, but instead the provider who diagnosed the anxiety disorder. We have access only to dispensed prescriptions and do not know how many patients were co-prescribed BZD treatment but did not fill the BZD prescription; primary adherence to new anti-anxiety medications was estimated to be 77%, with higher primary adherence (94%) in children.38 Further, we cannot determine whether the BZD was intended for as-needed or regular use and the setting in which the SSRI or BZD was prescribed. While we required an anxiety disorder diagnosis at least 30 days prior to SSRI initiation, we cannot be certain the SSRI or BZD was prescribed for anxiety, especially in patients with comorbid diagnoses. Psychiatric diagnoses were based on diagnostic codes. We examined treatment length adjusting for potential confounders; however, we lack clinical measures, such as anxiety severity, that may influence treatment duration. We detailed BZD and SSRI simultaneous initiation in a commercially insured population; findings may not generalize to adolescents and young adults who are uninsured or insured by Medicaid. Further, race and ethnicity data are unavailable in the dataset. Our study design focused on same-day BZD and SSRI initiation; concurrent SSRI and BZD use at any point during the treatment course will be higher. Our study focused on SSRI initiators; results may differ for treatment initiation with SNRIs or other antidepressants.
The decision to co-prescribe BZD and SSRI treatment for anxiety disorders is multifactorial and there is limited evidence on potential benefits of, or the sub-population of adolescents and young adults who will benefit the most from, this treatment approach. While these data show that the practice of simultaneous prescribing has declined, in everyday clinical practice in those seen by both generalists and specialty care providers, BZD and SSRI treatments continue to be simultaneously prescribed to young people with anxiety disorders, particularly to young adults. Still, it remains unclear if simultaneous BZD treatment improves SSRI treatment persistence.
Supplementary Material
CLINICAL POINTS.
Simultaneously initiation of benzodiazepine and SSRI treatment, and its effect on SSRI treatment duration, had not been detailed in young people with anxiety disorders.
Simultaneous benzodiazepine and SSRI initiation was relatively common in young adults, but not associated with longer SSRI treatment.
Potential benefits and risks of adding a benzodiazepine at SSRI initiation should be carefully considered.
Funding.
Research reported in this publication was supported by the National Institute of Mental Health (Bethesda, MD) under Award Number T32MH013043. Dr. Crystal’s work is supported by AHRQ 1R01HS026001-01A1, PCORI HHSA2902010000101, and NIH 1R01DA047347-01 and UL1TR003017. This content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Role of the sponsor.
The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Potential conflict of interest.
Dr. Rynn is a DSMB consultant for Allergan. Dr. Gerhard reports grants from NIA and NIMH during the conduct of the study; grants and personal fees from Bristol-Myers Squibb; personal fees from Eisai, Merck, Pfizer, Lilly, and IntraCellular Therapies outside the submitted work. Drs. Bushnell, Crystal, and Olfson report no conflicts of interest relevant to this article to disclose.
REFERENCES
- 1.American Psychiatric Association. Practice Guideline for the Treatment of Patients with Panic Disorder. 2009.
- 2.Anxiety and Depression Association of America. Clinical Practice Review for GAD. Silver Spring, MD. [Google Scholar]
- 3.Stein MB, Craske MG. Treating Anxiety in 2017: Optimizing Care to Improve Outcomes. JAMA. 2017;318(3):235–236. [DOI] [PubMed] [Google Scholar]
- 4.National Institute for Health and Care Excellence. Generalised anxiety disorder and panic disorder in adults: management. 2011. [PubMed]
- 5.Connolly SD, Bernstein GA, Work Group on Quality Issues. Practice parameter for the assessment and treatment of children and adolescents with anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry. 2007;46(2):267–283. [DOI] [PubMed] [Google Scholar]
- 6.Walter HJ, Bukstein OG, Abright AR, et al. Clinical Practice Guideline for the Assessment and Treatment of Children and Adolescents with Anxiety Disorders. Journal of the American Academy of Child & Adolescent Psychiatry. 2020. [DOI] [PubMed] [Google Scholar]
- 7.Dunlop BW, Davis PG. Combination treatment with benzodiazepines and SSRIs for comorbid anxiety and depression: a review. Prim Care Companion J Clin Psychiatry. 2008;10(3):222–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Goddard AW, Brouette T, Almai A, Jetty P, Woods SW, Charney D. Early coadministration of clonazepam with sertraline for panic disorder. Arch Gen Psychiatry. 2001;58(7):681–686. [DOI] [PubMed] [Google Scholar]
- 9.Dobson ET, Bloch MH, Strawn JR. Efficacy and Tolerability of Pharmacotherapy for Pediatric Anxiety Disorders: A Network Meta-Analysis. J Clin Psychiatry. 2019;80(1). [DOI] [PubMed] [Google Scholar]
- 10.Bushnell GA, Sturmer T, Gaynes BN, Pate V, Miller M. Simultaneous Antidepressant and Benzodiazepine New Use and Subsequent Long-term Benzodiazepine Use in Adults With Depression, United States, 2001–2014. JAMA Psychiatry. 2017;74(7):747–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Furukawa TA, Streiner D, Young LT, Kinoshita Y. Antidepressants plus benzodiazepines for major depression. The Cochrane database of systematic reviews. 2001(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Stein MB, Cantrell CR, Sokol MC, Eaddy MT, Shah MB. Antidepressant adherence and medical resource use among managed care patients with anxiety disorders. Psychiatr Serv. 2006;57(5):673–680. [DOI] [PubMed] [Google Scholar]
- 13.Bushnell GA, Brookhart MA, Gaynes BN, Compton SN, Dusetzina SB, Sturmer T. Examining Parental Medication Adherence as a Predictor of Child Medication Adherence in Pediatric Anxiety Disorders. Med Care. 2018;56(6):510–519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jann M, Kennedy WK, Lopez G. Benzodiazepines: a major component in unintentional prescription drug overdoses with opioid analgesics. J Pharm Pract. 2014;27(1):5–16. [DOI] [PubMed] [Google Scholar]
- 15.Goodman & Gillman’s The Pharmacological Basis of Therapeutics. 12th ed: The McGraw-Hill Companies, Inc.; 2011. [Google Scholar]
- 16.McCabe SE, West BT. Medical and nonmedical use of prescription benzodiazepine anxiolytics among U.S. high school seniors. Addict Behav. 2014;39(5):959–964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gaither JR, Shabanova V, Leventhal JM. US National Trends in Pediatric Deaths From Prescription and Illicit Opioids, 1999–2016. JAMA Netw Open. 2018;1(8):e186558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Maust DT, Lin LA, Blow FC. Benzodiazepine Use and Misuse Among Adults in the United States. Psychiatr Serv. 2019;70(2):97–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Schepis TS, West BT, Teter CJ, McCabe SE. Prevalence and correlates of co-ingestion of prescription tranquilizers and other psychoactive substances by U.S. high school seniors: Results from a national survey. Addict Behav. 2016;52:8–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.U.S. Food and Drug Administration. FDA requiring Boxed Warning updated to improve safe use of benzodiazepine drug class. FDA Drug Safety Communication 2020; https://www.fda.gov/drugs/drug-safety-and-availability/fda-requiring-boxed-warning-updated-improve-safe-use-benzodiazepine-drug-class. [Google Scholar]
- 21.IBM. IBM MarketScan Research Databases. 2021; https://www.ibm.com/products/marketscan-research-databases. Accessed 3/24/2021.
- 22.IBM Watson Health™. IBM MarketScan Research Databases for Health Services Researchers. White Paper 2019; https://www.ibm.com/downloads/cas/6KNYVVQ2. [Google Scholar]
- 23.Liu X, Chen Y, Faries DE. Adherence and persistence with branded antidepressants and generic SSRIs among managed care patients with major depressive disorder. Clinicoecon Outcomes Res. 2011;3:63–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353(5):487–497. [DOI] [PubMed] [Google Scholar]
- 25.Fontanella CA, Bridge JA, Marcus SC, Campo JV. Factors associated with antidepressant adherence for Medicaid-enrolled children and adolescents. Ann Pharmacother. 2011;45(7–8):898–909. [DOI] [PubMed] [Google Scholar]
- 26.Paulozzi LJ, Strickler GK, Kreiner PW, Koris CM, Centers for Disease C, Prevention. Controlled Substance Prescribing Patterns--Prescription Behavior Surveillance System, Eight States, 2013. MMWR Surveill Summ. 2015;64(9):1–14. [DOI] [PubMed] [Google Scholar]
- 27.Bandelow B, Reitt M, Rover C, Michaelis S, Gorlich Y, Wedekind D. Efficacy of treatments for anxiety disorders: a meta-analysis. Int Clin Psychopharmacol. 2015;30(4):183–192. [DOI] [PubMed] [Google Scholar]
- 28.Wang Z, Whiteside SPH, Sim L, et al. Comparative Effectiveness and Safety of Cognitive Behavioral Therapy and Pharmacotherapy for Childhood Anxiety Disorders: A Systematic Review and Meta-analysis. JAMA Pediatr. 2017;171(11):1049–1056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Slee A, Nazareth I, Bondaronek P, Liu Y, Cheng Z, Freemantle N. Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis. Lancet. 2019;393(10173):768–777. [DOI] [PubMed] [Google Scholar]
- 30.U.S. 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. Silver Spring, MD: 2016. [Google Scholar]
- 31.Brent DA, Emslie GJ, Clarke GN, et al. Predictors of spontaneous and systematically assessed suicidal adverse events in the treatment of SSRI-resistant depression in adolescents (TORDIA) study. Am J Psychiatry. 2009;166(4):418–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Neutel CI, Patten SB. Risk of suicide attempts after benzodiazepine and/or antidepressant use. Ann Epidemiol. 1997;7(8):568–574. [DOI] [PubMed] [Google Scholar]
- 33.Cato V, Hollandare F, Nordenskjold A, Sellin T. Association between benzodiazepines and suicide risk: a matched case-control study. BMC Psychiatry. 2019;19(1):317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Schulenberg JE JL, O’Malley PM, Bachman JG, Miech RA, Patrick ME,. Monitoring the Future national survey results on drug use, 1975–2017: Volume II, College students and adults ages 19–55. Ann Arbor: Institute for Social Research, The University of Michigan; 2018. [Google Scholar]
- 35.Altmann H, Stahl ST, Gebara MA, et al. Coprescribed Benzodiazepines in Older Adults Receiving Antidepressants for Anxiety and Depressive Disorders: Association with Treatment Outcomes J Clin Psychiatry. 2020;81(6). [DOI] [PubMed] [Google Scholar]
- 36.Weisberg RB, Dyck I, Culpepper L, Keller MB. Psychiatric treatment in primary care patients with anxiety disorders: a comparison of care received from primary care providers and psychiatrists. Am J Psychiatry. 2007;164(2):276–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Dark T, Flynn HA, Rust G, Kinsell H, Harman JS. Epidemiology of Emergency Department Visits for Anxiety in the United States: 2009–2011. Psychiatr Serv. 2017;68(3):238–244. [DOI] [PubMed] [Google Scholar]
- 38.Fischer MA, Stedman MR, Lii J, et al. Primary medication non-adherence: analysis of 195,930 electronic prescriptions. J Gen Intern Med. 2010;25(4):284–290. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.