Attention-deficit/hyperactivity disorder (ADHD) medication use and psychotherapeutic polypharmacy is increasing. This study was designed to assess annual rates of ADHD medication prescribing and psychotherapeutic polypharmacy among patients 2 to 24 years old in the United States, identify commonly prescribed ADHD medications and concomitant psychotropic agents, and assess if specific characteristics are associated with polypharmacy.
In this cross-sectional study, we used publicly available ambulatory health care data sets to evaluate ADHD and psychotropic polypharmacy use in patients 2 to 24 years old from 2006 to 2015. National rates were estimated by using sampling weights, and common ADHD and psychotropic drugs prescribed were identified. Multivariate logistic regression models were developed to assess the strength of association between polypharmacy and patient or provider characteristics.
Between 2006 and 2015, ADHD medication prescribing increased from 4.8% to 8.4%. ADHD polypharmacy increased from 16.8% to 20.5%, whereas psychotropic polypharmacy increased from 26.0% to 40.7%. The most common ADHD combinations were stimulants and α-2 agonists (67.1%), whereas the most common concomitant psychotropic agents were selective serotonin reuptake inhibitors (14.4%) and second-generation antipsychotics (11.8%). Factors associated with polypharmacy were age, female sex (psychotropic), nonprivate insurance, northeast and south regions (ADHD), receipt of mental health counseling or psychotherapy, and calendar year.
ADHD and psychotropic polypharmacy use is increasing and associated with specific patient characteristics. These patterns should spark further inquiry about the appropriateness, efficacy, and safety of psychotherapeutic polypharmacy in children and young adults, particularly within subgroups in which the use is high.
Patients with attention-deficit/hyperactivity disorder (ADHD) are frequently treated with medications such as stimulants, and they often have concomitant psychiatric disorders that are treated with psychotropic agents. Little information exists regarding the factors associated with ADHD and psychotropic polypharmacy in children and young adults.
Psychotherapeutic polypharmacy use in youth with ADHD is increasing and is associated with patient-specific factors not previously described in the literature. These subgroups can be targeted for additional education and research on treatment decision-making regarding polypharmacy in children with ADHD.
Attention-deficit/hyperactivity disorder (ADHD) is the most common neurobehavioral disorder in childhood, affecting ∼5% to 11% of children and adolescents.1 This chronic condition often presents early in life with attention difficulties, hyperactivity, and impulsiveness. Functional impairments in central neurotransmitter systems, particularly those involving dopamine and norepinephrine, disrupt executive functions such as attentional and inhibitory control and working memory.2 Centers for Disease Control and Prevention epidemiological surveys of parents in the United States reveal that ADHD diagnoses and medication treatment of youth have risen since 2003.1,3 The percentage of children 4 to 17 years of age with an ADHD diagnosis increased by 42% between 2003 and 2011, along with a 28% increase in ADHD medication therapy.3 Clinical guidelines emphasize behavioral or psychotherapy as either first-line treatment (preschool-aged children) or combined with medication (older children and adolescents).4,5 Parent-reported data from 2016 reveal that 62% of children 2 to 17 years of age with ADHD in the United States received ADHD medication, whereas fewer than half received behavioral treatment.6 Stimulant medications are prescribed in >70% of ADHD outpatient visits.7 Nearly two-thirds of patients with ADHD have concomitant psychiatric comorbidities such as aggressive and/or conduct disorders, autism spectrum disorder (ASD), mood and/or depressive disorders, learning disabilities, and sleep disorders. 1,6 These conditions frequently have complicated and similar presentations with overlapping symptomatology that result in medication therapy. In fact, the incidence and prevalence of polypharmacy in pediatric mental health conditions is on the rise. Among child and adolescent ambulatory office visits associated with a mental disorder diagnosis, the percentage of visits with multiclass psychotropic treatment increased from 22.2% in 1996–1999 to 32.2% in 2004–2007, with a sixfold increase in the coprescription of ADHD and antipsychotic medication.7 From 1993 to 1998, the proportion of pediatric office visits in which stimulants were coprescribed with ≥1 other psychotropic medication increased nearly fivefold.8 In an analysis of 5-year treatment trajectories for children newly diagnosed with ADHD, authors reported that younger age at diagnosis was a pronounced determinant of psychotropic polypharmacy and antipsychotic use, with continuous increases in each year of follow-up.9
The purpose of this study was to examine more recent trends in ADHD medication prescribing and polypharmacy. Specifically, this study was designed to describe annual rates of ADHD medication prescribing and the frequency of psychotherapeutic polypharmacy among patients 2 to 24 years old with ADHD in the United States between 2006 and 2015, identify the most commonly prescribed ADHD medications and concomitantly prescribed psychotropic agents, and assess if patient and provider characteristics are associated with an increased risk of ADHD and psychotropic polypharmacy.
Methods
The study data set was obtained from the National Ambulatory Medical Care Survey for the period 2006–2015 and the National Hospital Ambulatory Medical Care Survey for the period 2006–2011. These surveys are nationally representative samples of outpatient provider visits that collect cross-sectional data such as patient demographics, prescriber information, diagnosis codes, and medications and therapies prescribed. Each visit is weighted on the basis of 3-stage probability sampling of geographic areas, physician practices, and randomly sampled visits from each physician practice. From the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey databases, the study data set included visits from patients 2 to 24 years of age in which ≥1 ADHD medications were prescribed. Drug prescriptions were identified by using the Multum classification system (Cerner Corporation, Denver, CO). The Multum classification uses a 3-level nested category system in which drugs are coded in terms of their generic components and therapeutic classes. Medications that were defined as ADHD treatments were stimulants (amphetamine salt mixtures, dextroamphetamine, lisdexamfetamine, dexmethylphenidate, and methylphenidate) and nonstimulants (atomoxetine, guanfacine, and clonidine). Bupropion was excluded because of its common use in other psychiatric conditions and limited role in ADHD management. Medications defined as psychotropic agents were antidepressants, antipsychotics, mood stabilizers, anxiolytics, and hypnotics (Supplemental Table 4). Data on drug dosages and formulations were not available in this data set.
First, national annual rates of ADHD drug prescribing and polypharmacy were estimated. Polypharmacy was identified in 2 categories: (1) ADHD polypharmacy was defined as having ≥2 ADHD drug prescriptions during an outpatient visit, and (2) psychotropic polypharmacy was defined as having a psychotropic drug prescribed concurrently with an ADHD drug prescription. Second, frequently used ADHD drugs and commonly coprescribed psychotropic drugs were identified by using generic names (active pharmaceutical ingredients). As supplementary analyses, we estimated the rates of polypharmacy and common drug combinations in age-stratified samples. Lastly, differences in patient and provider characteristics between ADHD monotherapy and polypharmacy visits were analyzed by using χ2 tests. Logistic regression models were developed to identify factors associated with ADHD polypharmacy (first model) and psychotropic polypharmacy (second model). In both models, several patient and treatment characteristics such as age, sex, payer source, geographic location, and receipt of mental health counseling or psychotherapy were included as explanatory variables. All statistical tests were performed as 2 sided, and statistical significance was set at P < .05. From logistic regression analyses, adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were estimated. Sample weights were applied in all analyses, and Stata/SE 13 (Stata Corp, College Station, TX) was used for statistical analysis. The study was approved by the institutional review board, which oversees the ethical conduct of research at this institution.
Results
Between 2006 and 2015, there were 121 481 sample ambulatory visits made by patients 2 to 24 years of age. After applying sampling weights, ∼5.8% of visits included ≥1 ADHD medication prescription (95% CI 5.3%–6.3%). At visits with ADHD medication prescriptions, they most commonly involved children 6 to 11 years of age (40.6%) and 12 to 18 years of age (39.8%), with a majority involving male patients (65.3%). Rates of ADHD prescriptions and polypharmacy changed substantially during this time period (Figs 1 and 2). ADHD medication prescribing rates increased over the study period from 4.8% (95% CI 3.9%–5.9%) of all visits in 2006 to 8.4% (95% CI 5.4%–12.9%) in 2015. When focusing on specific therapeutic classes, stimulant prescription rates nearly doubled from 4.0% (95% CI 3.2%–5.0%) of all visits in 2006 to 7.8% (95% CI 4.9%–12.2%) in 2015. Rates of α-2 agonist (guanfacine and clonidine) prescriptions increased 10-fold during the study period, ranging from a low of 0.2% (95% CI 0.1%–0.4%) in 2007 to a high of 2.1% (95% CI 1.1%–3.9%) in 2014. Atomoxetine prescription rates fell by 75% over the study period, from 0.8% (95% CI 0.5%–1.3%) in 2006 to 0.2% (95% CI 0.08%–0.6%) in 2015. At office visits with an ADHD prescription, ∼15.2% included ≥2 ADHD medications (ADHD polypharmacy, 95% CI 12.9%–17.8%), and 31.5% included ≥1 concurrent psychotropic prescription (psychotropic polypharmacy, 95% CI 28.5%–34.7%). More specifically, from 2006 to 2015, ADHD polypharmacy rates increased nearly threefold, ranging from a low of 6.2% (95% CI 3.9%–9.7%) of visits in 2007 to a high of 23.6% (95% CI 15.8%–33.6%) in 2013. Psychotropic polypharmacy rates doubled, ranging from a low of 20.2% (95% CI 14.1%–28.1%) of visits in 2007 to a high of 40.7% (95% CI 29.3%–53.2%) in 2015. When stratified by age, the ADHD polypharmacy rate was the highest in patients 6 to 11 years of age, and the psychotropic polypharmacy rate was the highest in patients 19 to 24 years of age (Supplemental Table 5).
During the study period, the most frequently prescribed ADHD medications were stimulants (methylphenidate [2.2% of all visits], amphetamine and dextroamphetamine [1.6%], and lisdexamfetamine [1.0%]), followed by clonidine (0.6%). At visits with ADHD polypharmacy prescriptions, the most common combinations were stimulants combined with an α-2 agonist (67.1% of all ADHD polypharmacy; Table 1). At visits with psychotropic polypharmacy prescriptions, the most common combinations were an ADHD medication prescribed with a selective serotonin reuptake inhibitor (SSRI) (14.4% of all ADHD visits), second-generation antipsychotic (SGA) (11.8%), or a mood stabilizer (5.0%) (Table 1). At visits with ADHD medication prescriptions, concurrent SSRI prescription rates were the highest in patients 12 to 18 years of age, whereas concurrent SGA prescription rates were the highest in patients 2 to 5 years of age (Supplemental Table 6).
Patient and provider characteristics were compared between ADHD monotherapy and polypharmacy visits (Table 2). Among ADHD office visits, there was a higher percentage of patients 6 to 18 years of age among those who received ADHD polypharmacy than there was for those who received monotherapy. There was a higher percentage of patients 12 to 24 years of age among those who received psychotropic polypharmacy than there was for those who received ADHD monotherapy. There was a higher percentage of patients with private insurance among those who received ADHD monotherapy than there was for those who received ADHD or psychotropic polypharmacy. There were higher percentages of female patients among those who received psychotropic polypharmacy in comparison with ADHD monotherapy and ADHD polypharmacy. Psychiatrists prescribed a higher percentage of ADHD polypharmacy and psychotropic polypharmacy and a lower percentage of ADHD monotherapy in comparison with other prescriber types.
In the multivariate logistic regression models (Table 3), the variables that were significantly associated with ADHD polypharmacy were age, geographic region, calendar year, payer type, and receipt of mental health counseling or psychotherapy. School-aged children were more likely to have ADHD polypharmacy visits than young adults (4–5 years [aOR 3.02], 6–11 years [aOR 3.57], 12–18 years [aOR 3.52]). Patients who live in the northeast (aOR 1.99) or south regions (aOR 1.4) were more likely to have ADHD polypharmacy visits than patients who live in the Midwest region. Patients with Medicaid insurance (aOR 1.66) or other payer sources (aOR 1.92) were more likely to have ADHD polypharmacy visits than patients with private insurance. Other factors that were independently associated with ADHD polypharmacy were calendar year (aOR 1.12) and receipt of mental health counseling (aOR 2.33) or psychotherapy (aOR 1.99). Factors that were significantly associated with psychotropic polypharmacy visits were female sex (aOR 1.29), calendar year (aOR 1.07), Medicaid insurance (aOR 1.94) or other payer sources (aOR 1.59), and receipt of mental health counseling (aOR 2.88) or psychotherapy (aOR 3.36).
Discussion
The purpose of this study was to (1) assess annual rates of ADHD medication prescriptions and the frequency of psychotherapeutic polypharmacy among patients 2 to 24 years old with ADHD in the United States between 2006 and 2015, (2) identify the most commonly prescribed ADHD medications and concomitantly prescribed psychotropic agents, and (3) assess if patient and provider characteristics are associated with an increased risk of ADHD and psychotropic polypharmacy. Significant increases in prescription rates were observed in nearly all analyzed categories, such as rates of ADHD medication prescriptions, rates of individual medication class prescriptions, and the rates of both ADHD polypharmacy and psychotropic polypharmacy prescriptions. Earlier studies also describe rising ADHD medication treatment rates and psychotherapeutic polypharmacy.7–13 The observed trends may be driven by an increasing prevalence of ADHD diagnoses and subsequent medication treatment or by higher rates of medication use in those who are diagnosed.14 Diagnostic criteria are outlined in the Diagnostic and Statistical Manual of Mental Disorders, and although they have become more reliable over time, use of the 2013 Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, criteria may result in higher ADHD prevalence rates because of changes made to the minimum number of symptoms required and age of onset for diagnosis.15 Treatment guidelines emphasize the high response rates to medication therapy, particularly stimulants, among nearly all age groups that may lead to increased prescription rates over time.4,5 Increased rates of α-2 agonist prescriptions may relate to their Food and Drug Administration approval for ADHD treatment in 2009–2010. Parental desire to maximize their child’s academic potential by using medication may also lead to increased prescription rates.14 Another potential driving force is governmental effort to increase access to and affordability of medication therapy and mental health care, such as the Patient Protection and Affordable Care Act, which expanded access to mental health services and improved prescription coverage.16 The potential costs and lack of access to professionally administered cognitive behavioral therapy and psychotherapy or counseling could also increase medication prescription rates because of the relative ease of medication access and administration by nonpsychiatric medical providers.17 Lastly, a burgeoning pediatric population that is not well balanced with a growing pediatrician supply may contribute to increases in medication prescription rates caused by shortened physician visit times spent with individual patients.18
Other observable changes in the rates of ADHD medication and psychotropic polypharmacy prescribing are the moderate decreases from 2009 to 2010 and 2009 to 2012, respectively. In 2006 and 2007, the Food and Drug Administration released warnings concerning the cardiovascular and neuropsychiatric effects associated with ADHD medications, culminating in 2009 with a study linking stimulants with sudden death.19 Additional warnings about risks of suicidal ideation with antidepressants and antipsychotics were also released during this time. However, in late 2011, a large retrospective cohort study assuaged the concerns about the cardiovascular risks associated with stimulants.20
In this study, we also analyzed the most commonly prescribed ADHD and psychotropic polypharmacy combinations from 2006 to 2015. The most frequently prescribed ADHD combinations were stimulant medications plus an α-2 agonist. Such combinations have moderate evidence to support their use and are frequently used in patients with incomplete responses to stimulants, to improve tolerability by using low doses of 2 drugs rather than high stimulant doses, or to counter the adverse effects of stimulants (eg, insomnia can be managed with the sedating effects of clonidine).4,12 Concomitant mental health conditions such as behavioral disorders, anxiety, depression, and ASD occur in as many as two-thirds of patients with ADHD.6 Therefore, some psychotherapeutic drug combinations are necessary and are supported by guidelines despite an overall lack of published data to support the safety and efficacy of most combinations.4,5,12 In this study, SSRIs, SGAs, and mood stabilizers were the medication classes most frequently coprescribed with an ADHD medication, similar to earlier findings.7 Surprisingly, SGAs were coprescribed with ADHD medications most frequently at visits in the youngest patients (2–5 years of age). This could be explained by commonly coprescribing α-2 agonists for sleep difficulties in young children with behavioral disorders or ASD (who are often treated with SGAs) rather than a high usage of SGAs in young children with ADHD.21 Outpatient visits in this study data set were characterized as ADHD visits if a medication that is commonly used for ADHD management was prescribed at the visit and the rationale or indication for each drug was not captured in the data set. Nevertheless, the high use of multiclass drug combinations in children and adolescents is concerning, given their vulnerability to adverse effects and the unknown impact of long-term neurologically altering substances on the development of an immature brain and nervous system.
Several factors were correlated with an increased likelihood of receiving ADHD and psychotropic polypharmacy in this study. It is reassuring that the strongest associations with polypharmacy were found in visits in which patients were also noted to be receiving mental health counseling or psychotherapy. These services are frequently provided by psychiatric and provider specialists who care for patients with ADHD and concomitant disorders and, thus, prescribe the medications used to treat these conditions. Other factors that were significantly associated with ADHD and psychotropic polypharmacy were consecutive study year (increased odds with time over the study period), region of the United States, payer type and/or reimbursement, sex (higher odds in female patients for psychotropic polypharmacy only), and age (higher odds of ADHD polypharmacy for 4–18-year-old patients and lower odds of psychotropic polypharmacy in 2–18-year-old patients). Psychotropic polypharmacy has been previously linked to older child age (13–17 years), specific mental health disorder diagnoses, public insurance, and psychiatrist physicians.7 These prescription trend findings necessitate further analysis in future studies to determine why certain children and youth have higher rates of psychotherapeutic drug combination regimens, possibly placing them at risk for adverse effects from potentially inappropriate prescribing practices. These could involve large-scale provider and patient surveys to collect data on rationale for prescribed polypharmacy combinations that may provide insight on appropriateness of prescribing patterns.
This study did have several limitations that should be noted. Survey data rely on completion by physicians or data recorders, resulting in the potential for inherent inaccuracies. The data set provided information on randomly selected outpatient visits and not individual patients, so it is possible that some individuals could be represented more than once in the sample. This study included data through 2015, so the most current prescribing trends are not encompassed in this analysis. The unweighted counts for several analyses were small (eg, <30 patients for certain medication combinations; Supplemental Table 7) before statistical weighting, which may result in unreliable annual estimates regarding polypharmacy regimens. The cross-sectional nature of the data set prevents analysis of the appropriateness, duration, and impact of long-term polypharmacy in specific patients. Data on diagnoses were not collected because this study was focused on medication prescribing trends over time. It was assumed that ADHD medications were prescribed to only patients with an ADHD diagnosis, but it is possible that these prescriptions may have been for reasons other than ADHD. Similarly, a large proportion of psychotropic polypharmacy could be used to treat diagnosed comorbid conditions other than ADHD; nonetheless, the findings from this study reflect a growing rate of psychotherapeutic polypharmacy in youth despite the underlying diagnoses. Lastly, the data only reflect rates of prescribing at ambulatory visits and do not account for adherence to therapy or short-term therapeutic overlap, so drug use may be overestimated.
Conclusions
Between 2006 and 2015, rates of ADHD and psychotropic polypharmacy increased among children and young adults. The most commonly used combinations were stimulants with α-2 agonists and ADHD medications with either SSRIs or SGAs. Factors significantly associated with both ADHD and psychotropic polypharmacy were age, payer type, geographic region, receipt of mental health counseling or psychotherapy, and advancing calendar year. These prescribing patterns raise questions about the rationale and evidence for polypharmacy use and can be used to spark discussion and further inquiry about the appropriateness, efficacy, and safety associated with ADHD and psychotropic polypharmacy in children and young adults, particularly within subgroups in which polypharmacy use is high.
Dr Girand conceptualized and designed the study and analyzed and interpreted the study data; Dr Litkowiec analyzed and interpreted the study data and drafted the initial manuscript; Dr Sohn conceptualized and designed the study, acquired the study data, and analyzed and interpreted the study data; and all authors critically reviewed the manuscript for important intellectual content, revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
References
Competing Interests
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.