Concomitant Psychotropic Medication for Youths

We were able to locate two national population assessments that measured use of concomitant psychotropic medication in U.S. children. One revealed that in 1997–1998, 24.7% of the representative physician office visits for youths in which a stimulant prescription was written were also associated with the use of concomitant psychotropic medication. This represented a fivefold increase over the 1993–1994 rate of 4.8% (19). The other was based on national parent report surveys in 1987 and 1996 pertaining to medication use by children and adolescents. Olfson et al. (11), by comparing these 1-year estimates, reported that the use of multiple psychotropic medications increased eightfold. For youths (younger than 19) who were given at least one psychotropic medication, the rate of receipt of additional psychotropic medication rose 2.5-fold (4.7% to 11.7%) over that decade. Outside the United States, a recent prevalence study from the Netherlands (20) indicated that psychotropic therapy for youths is almost exclusively monotherapeutic. Published data on the rate of use of concomitant psychotropic medication in youths come mainly from selected groups, and the findings vary considerably by diagnosis and treatment setting. Medical records of youths from psychiatric outpatient and inpatient settings have been the source of data for most of these studies.

The majority of concomitant psychotropic medication studies have focused on youths with attention deficit hyperactivity disorder (ADHD). Rappley and colleagues (21) found the proportional rate of use of concomitant psychotropic medication to be 20% among 2- and 3-year-old children who were enrolled for 15 consecutive months in a Medicaid insurance program in 1995–1996 and were diagnosed with ADHD (N=223). Ghuman et al. (22), using data from 1995–1999, reported a concomitant psychotropic medication rate of 26% for 3- to 5-year-old children referred to a developmental disorder clinic who were diagnosed with ADHD and who received stimulant medication. In an older cohort of youths in a health maintenance organization (HMO) setting, use of concomitant psychotropic medication in 1997–1998 was identified in 21% of 5–12-year-old children who were diagnosed with ADHD and who were receiving stimulant medication (23). A 1996 U.S. practice research network survey compiled by Zarin and colleagues (24) found that 49% of 166 youths (aged less than 15 years) who were treated by private psychiatrists for ADHD were receiving concomitant psychotropic medication, with 12% of that total receiving three or more psychotropic medications.

Medical chart audits from specialty outpatient clinics in Massachusetts revealed relatively high rates of use of concomitant psychotropic medication. In 1990, additional psychotropic medication was prescribed for 47% of youths diagnosed with ADHD (25) who were being treated with nortriptyline. A medical chart audit in 1991 (26) revealed the use rate of concomitant psychotropic medication to be 39% for youths diagnosed with obsessive-compulsive disorder (OCD). In some specialty clinics, even higher rates of concomitant psychotropic medication were observed. For example, Cohen et al. (27) reported a use rate of concomitant psychotropic medication of 68% from 1992–1996 among youths with ADHD, which was similar to the 71% use rate of concomitant psychotropic medication that Biederman and colleagues (28) found from 1991–1995 among youths with bipolar disorder.

Reports of use of concomitant psychotropic medication are also high among youths diagnosed with schizophrenia. In an early 1990s treatment program for youths with schizophrenia, Kumra et al. (29) noted that the mean number of psychotropic medications prescribed for each applicant youth was 2.1.

On the basis of parent reporting from a group referred to a center for social learning disabilities in July through December 1997, Martin et al. (30) reported a 29% rate of concomitant psychotropic medication among 109 youths (mean age=13.9 years) who had a pervasive developmental disorder. Similarly, in the early 1990s, Aman et al. (31) mailed a survey to parents of autistic youths (a 53% response) and reported that 28% of the identified youths (median age=13 years) given psychotropic medication were receiving concomitant psychotropic medication.

Concomitant psychotropic medication use findings based primarily on medical chart audits also vary considerably by treatment site and setting. In 1990, a systematic review of medical charts in community mental health centers (CMHCs) located in Maryland, New York, and Ohio revealed that 9% (9), 11%, and 22% (32), respectively, of treated youths were receiving concomitant psychotropic medication. More recently, medical chart reviews conducted in different CMHCs in Maryland found that concomitant psychotropic medication was prescribed for 21% of treated youths in 1994 (9) and 22% of treated youths in 1997 (33).

The rate of use of concomitant psychotropic medication has been consistently higher in inpatient than in standard outpatient psychiatric settings. In 1976–1977, 30% of 100 Canadian psychiatric inpatient youths received concomitant psychotropic medication (34). In a 1991 medical chart audit from three hospitals, Kaplan and Busner (35) similarly reported a use rate of concomitant psychotropic medication of 36%. A slightly higher rate of concomitant psychotropic medication of 42% was noted with 1994 data from seven youth inpatient units in Maryland (9).

The use of concomitant psychotropic medication among youths increased substantially during the 1990s. For example, the rate of use of concomitant psychotropic medication rose 133% (9% to 21%) from 1990 to 1994 in four Maryland CMHCs (9). In a national sample of physician office visits for youth (under age 18), the rate of combined antidepressant and stimulant use increased from 4% in 1994 to 29% in 1997, reflecting a sevenfold increase (36). Rushton and Whitmire (10) noted a similar increase from 1992 through 1998 for combined stimulant and selective serotonin reuptake inhibitor (SSRI) treatment among youths aged 1–19 years. Although most rates of concomitant psychotropic medication for psychiatric clinic youths in the early and mid-1990s were below 50%, several posters presented in 2001 at national psychiatric meetings showed an increase in this respect to over 50% (37–39).

An important trend to note is that the higher rate of use of concomitant psychotropic medication among youths has occurred simultaneously with the overall increase in psychotropic use. On the basis of findings from an HMO and two state Medicaid databases covering 1987 through 1996, the prevalence of psychotropic use among youths less than 20 years old increased 4–10-fold for antidepressants, 36–153-fold for alpha agonists, and three- to sevenfold for stimulants (40). Likewise among youths (younger than age 20) in one state Medicaid program from 1996 through 2000, there was a threefold increase in neuroleptic treatment (41). The increase was most prominent after atypical neuroleptics were introduced in the early 1990s (42).

The use of concomitant psychotropic medication has been found to be highly associated with aggressive behavior in youths, particularly the coadministration of neuroleptic medications (32, 49). Ahsanuddin et al. (34) noted that aggressive behavior disorders were the primary diagnoses associated with the use of concomitant psychotropic medication. Likewise, in a study of youths in residential treatment, Connor et al. (48) found that the major symptom pattern associated with use of concomitant psychotropic medication was aggressive behavior.

Insomnia is another target symptom associated with the use of concomitant psychotropic medication in youths. Wilens et al. (50) and Prince et al. (51) both reported that clonidine is frequently added to stimulant treatment to aid sleep. In recent years, risperidone has been added to other psychotropic medications to both aid sleep and lessen aggressiveness (52).

A particularly common concomitant psychotropic medication combination for youths has been methylphenidate and clonidine. In the early 1990s, Swanson et al. (53) estimated from a national pharmaceutical market source that 41% of surveyed youths in 1994–1995 who were receiving clonidine were also receiving methylphenidate. Prince et al. (51), in a psychiatric clinic survey from 1992 to 1995, reported the rate of this combination to be 68%. In a 2–3-year-old Medicaid population, Rappley et al. (54) similarly found this to be the most common psychotropic combination.

Another common concomitant psychotropic medication prescribing pattern involves antidepressants with stimulants. Pathiyal et al. (55), in an 18-month longitudinal analysis of pharmacy claims data, reported that 22% of youths receiving methylphenidate in 1993–1995 also received antidepressants concomitantly. On the basis of 1998 North Carolina Medicaid claims data, Rushton and Whitmire (10) reported that 30% of youths receiving an SSRI in 1998 also had a stimulant prescription during the same year. Similarly, Zito et al. (56) found that one-third of Medicaid youths in one state who were receiving antidepressant medication in 1994 had also been given a stimulant that year. It should be noted that these prevalence findings for a 1-year period may not reflect the use of concomitant psychotropic medication. However, Rushton and Whitmire (10) reported that 83% of stimulant/SSRI prescriptions were filled in the same month, suggesting concomitant use.

A fundamental underlying concern with concomitant psychotropic medication for youths is the virtual lack of rigorous systematic research to support its use and to understand its potential risks (99, 100). This and certain publicized cases of use of extreme concomitant psychotropic medication led the Council of the American Academy of Child and Adolescent Psychiatry (101) to recommend judicious and cautious use of concomitant psychotropic medication for youths.

The cases of concomitant psychotropic medication cited by Wagner (100), Sallee et al. (89), and Behr (102) of prepubertal youths receiving four to seven psychotropic medications concomitantly that were associated with adverse drug events understandably “raised eyebrows” in the field. In part, this was due to the consistent finding in the adult literature indicating that the risk of adverse drug events increases with the number of concomitant medications administered (103, 104). Turner et al. (105) and Martinez-Mir et al. (104), evaluating pediatric inpatients, likewise reported a significant increase in adverse drug events in relation to the number of concomitant medications used.

Other safety concerns with use of concomitant psychotropic medication in youths include 1) the greater possibility of untoward drug interactions (106, 107) and 2) the creation of drug-induced behavioral toxicity after the addition of another psychotropic medication—a consequence not often recognized as such, which can then lead to even more complex drug therapy to treat that side effect (100).

Table 4 presents a brief list of reports of adverse drug events with the use of concomitant psychotropic medication in youths. It is noteworthy that in nearly all cases, three or more psychotropic medications were administered concomitantly. The majority of the youths in these reports were boys younger than 13 years old.

The risk of adverse drug events with the use of concomitant psychotropic medication in relation to age can be appreciated from data on valproate (used primarily for seizure control). From 1978 to 1984, the rate of fatal hepatotoxicity due to valproate monotherapy in youths under the age of 11 was five in 42,618. The rate of hepatic fatality when valproate was administered with one or more additional anticonvulsants increased prominently to 22 in 47,864, a fourfold increase. Moreover, during that survey period, there were no hepatic deaths among youths aged 11–21 years that were associated with valproate monotherapy, compared to five with valproate polytherapy (108). One should also consider that, overall, youths aged 9 years or younger are substantially more at risk for adverse drug events than are older youths (109).

Although differences in prescribing patterns in concomitant psychotropic medication in relation to age are incompletely studied, there are a number of findings in this area that are of interest. The use of concomitant psychotropic medication is more common with advancing age, in female adults, and in male youths (110). Also, compared to adults, youths are prescribed different concomitant psychotropic medication combinations, have fewer multiple providers, and have different problematic behaviors targeted by concomitant psychotropic medication (13, 110–112).

Adding a second psychotropic medication to augment a partial response to the first could be useful if the first had been given at a customary therapeutic dose for a reasonable period of time and if that treatment and its subsequent augmentation received a systematically monitored assessment. The evaluation of the first medication and of the combination of medications should optimally be performed with reliable measures, baseline and subsequently rated targeted assessments, and preferably with at least one blind independent rater. “N of 1” research protocols for evaluating single patients in an experimental study might be developed to evaluate use of concomitant psychotropic medication more rigorously, beginning in teaching hospital settings under rigid research conditions and moving out to the community setting when methods are better developed.

Treating two comorbid psychiatric disorders with two or more different medications has been recommended if each of the disorders has a unique pattern and an optimal treatment. This rationale applies in certain instances (e.g., with tics and ADHD), but in most other circumstances, the diagnostic patterns are far less specific. In practice, there is often a sizable symptom overlap in comorbid diagnoses in child psychiatry (e.g., bipolar disorder and ADHD, disruptive behavior disorders and ADHD) (28, 113). In addition, although a 100% treatment response would obviously be optimal, it is uncommon (particularly in chronic cases), and “doctors who cannot accept this level of imperfection put their patients at risk in a futile search for the holy grail through tactics like adding more drugs” (114, p. 17). Green (16, p. xi) similarly advised clinicians to “resist the tempting but erroneous notion that the right combination of drugs will solve any problem.” Furthermore, a number of investigators have reported successfully treating comorbid disorders with only one psychotropic medication by exploiting its range of effects, as has been shown by Klein et al. (115), Spencer et al. (116), Steingard et al. (117), and Scahill et al. (118).

Some writers in the field identify rational as opposed to irrational use of concomitant psychotropic medication (119, 120). Essentially, a rational concomitant psychotropic medication approach is based on the weight of scientific evidence in its favor. However, as one notes from the available reports (Table 1, Table 2, and Table 3), the evidence base for the use of concomitant psychotropic medication is profoundly limited. Consequently, the use of the term “rational” for the use of concomitant psychotropic medication applied to youths appears to be premature.

Almost without exception, leaders in the field support use of concomitant psychotropic medication for specific indications in child psychopharmacology. Biederman (121, p. 11) stated, “Examples of the rational use of combined treatment include the use of an antidepressant plus a stimulant for ADHD and comorbid depression, the use of clonidine to ameliorate stimulant-induced insomnia, and the use of a mood stabilizer plus an anti-ADHD agent to treat ADHD comorbid with bipolar disorder.” Popper (122, p. 497) noted that “multiple agents are needed for treating comorbid presentations often seen in children, including ADHD with Tourette’s syndrome or obsessive-compulsive disorder (OCD), major depression with psychotic features, and certain depressive/anxiety disorders.” Although generally critical of many instances of use of concomitant psychotropic medication, Wagner (100, p. 78) found it “very appropriate for co-morbid conditions in which there is a first-line treatment for each disorder.” She cited using an SSRI for major depression with a stimulant for ADHD as an appropriate example.

There have also been statements of support for specific combinations of concomitant psychotropic medications. For example, Carlson et al. (123, p. 409) noted that “certain combinations appear to be useful.” These include 1) antidepressants with lithium for bipolar disorder, 2) lithium augmentation for refractory depression, 3) lithium and neuroleptics for acute psychoses, 4) lithium and anticonvulsants for refractory mania, 5) sustained-release methylphenidate with short-acting methylphenidate to increase potency at drug onset, and 6) methylphenidate and a late afternoon or evening neuroleptic for severely hyperactive children. Pliszka et al. (113, p. 112) recommended use of concomitant psychotropic medication for youths with disruptive behavior disorders who are partially responsive to stimulant treatment, King (124, p. 41S) recommended clonazepam or a low dose of a neuroleptic to augment the effects of SSRIs for youths with treatment-resistant OCD, and Emslie et al. (125, p. 828) supported one of four psychotropic drug groups to augment inadequate medication responses to an SSRI for depressed youths.

Essentially, Biederman (121) and Wagner (100) promoted treating comorbid disorders as the major justification for the use of concomitant psychotropic medication in youths, whereas Carlson et al. (123) recommended concomitant psychotropic medication to augment the treatment of partial medication responders. Others, such as Popper (122), Theesen (126), and Wilens et al. (127), indicated that both augmentation and comorbidity justify the use of concomitant psychotropic medication in youths. Still other reasons to use concomitant psychotropic medication include allowing a lower dose of one agent to be used, reversing side effects, and alleviating symptoms while waiting for another medication to become effective (16, 114, 120, 128).

The admission of youths to inpatient psychiatric facilities in the United States is precipitated largely by aggressive/disruptive behavior (34, 129–131). These youths receive concomitant psychotropic medication at a significantly higher rate (34) and proportionally more neuroleptic medication (32, 34, 49, 132) than nonaggressive youths. In fact, Connor et al. (48, p. 27) wrote that “neuroleptics were used primarily for aggression regardless of diagnosis,” a finding previously noted (32, 34). In addition to neuroleptics, psychotropic drugs used individually and concomitantly to treat aggressive behavior in youths included stimulants, lithium, antidepressants, anticonvulsants, and alpha agonists (133).

Child psychopharmacologists specifically aim to ameliorate discrete pathologically impairing target symptoms associated with diagnostic entities (134). Generally, sleep impairment, enuresis, tics, depressive symptoms, hyperactivity, obsessive-compulsive features, intermittent impulsive behavior, and delusions are commonly targeted for reduction with psychotropic medication. The usual treatment approach is to focus on the reduction of one symptom or one symptom complex by using a low dose of one psychotropic medication initially (13). However, when presented with an array of symptoms, one can easily extend this approach and add a different medication for each target symptom. A review of the adverse drug event reports comprising Table 4 reveals that a very common aim of concomitant psychotropic medication in these patients had been to treat multiple target symptoms.

Kutcher (135, p. 257) noted that “in child and adolescent psychopharmacology, augmentation strategies are poorly researched, and in many cases, clinicians use adult literature as guidelines.” Sanchez et al. (136, p. 639) similarly noted that most concomitant psychotropic medication for youths is “still based on studies in adults.” Extrapolation from adult studies to children and adolescents has numerous limitations because emotional disorders and psychotropic medication responses are often quite different in children and adolescents than in adults (13, 137). Even mixing groups of late adolescents with younger youths can blur age-specific outcomes (138). Likewise, concomitant psychotropic medication prescribing patterns for adults and youths prominently differ.

After reviewing the various methods for establishing evidence of efficacy, we propose that a systematic evaluation of concomitant psychotropic medication should take place on three levels. First, a national survey using probability sampling techniques can be conducted to estimate the prevalence and prescribing patterns of concomitant psychotropic medication across a spectrum of child and adolescent behavioral and emotional disorders as well as in different treatment settings, including primary care, psychiatry, and subspecialty practices. Second, the effectiveness, safety, and level of satisfaction with certain widely used drug combinations can be systematically evaluated by using multiple informants as well as standardized rating scales and questionnaires. Obtaining long-term outcome data on concomitant psychotropic medication in behavioral, academic, and social domains will, of course, depend greatly on the ability to retain subjects for extended periods. Third, patients receiving complex concomitant psychotropic medication could be addressed by using “N of 1” methods (139). The methods for evaluating single patients in an experimental design originated in clinical psychology and randomly assigns the treatment interventions under double-blind conditions to successive time periods. The results of these trials generalize to the single patient under study; the goal is to produce an objective assessment of the benefit of the various interventions (drug 1 alone, drug 1 plus placebo, drug 1 plus drug 2). Applications of this approach have already been reported in the adult psychiatry and pediatric literature (140, 141). If the approach is applied to concomitant psychotropic medication, careful monitoring of functioning and side effects could determine if the outcome is altered after a series of sequential evaluations of the concomitant medication or a placebo substitution. In regard to improving methods for assessing safety for rarely occurring events (less than 1 per 10,000 exposures), much work is needed. This will initially require an examination of weaknesses in the existing infrastructure for reporting adverse drug events and in the methods for producing evidence on risk assessment (142, 143).