It is well documented that approximately 20% of children in the United States have behavioral health issues (1–4), that approximately one-quarter to one-half of all pediatric primary care office visits involve psychosocial concerns (5,6), and that most children who experience psychosocial issues seek treatment in primary care settings (7). In addition, there is evidence that pediatric primary care providers write as many as 85% of all prescriptions for psychotropic medications for children and adolescents (8–10).

This Open Forum reviews literature on patterns of psychotropic medication prescribing for children and adolescents and the role of pediatric primary care providers. After describing two model programs for collaborative care, we offer recommendations for improving professional training and comanagement of behavioral health care for this population.

Prevalence of psychotropic drug use

Numerous studies have addressed the psychiatric diagnoses for which psychotropic medications are prescribed for children and adolescents (9,11); the frequency of prescriptions associated with outpatient, office-based visits (11–13); and, in particular, the increase in the number and percentage of children being treated with various classes and combinations of psychotropic drugs (9,11,14–17). These increases have been observed across various populations, including youths enrolled in Medicaid, youths in foster care, preschoolers, and adolescents (13,16,18,19).

An analysis of data from nearly 900,000 children and adolescents found a two- to threefold increase from 1987 to 1996 in the percentage of patients taking any psychotropic medication; 1996 rates ranged from 5.9% to 6.3% across three sites (20). Using data from the National Ambulatory Care Survey (NAMCS), another study found that for adolescents age 14 to 18, the percentage of office visits that resulted in a psychopharmacological prescription increased from 3.4% in 1994–1995 to 8.3% in 2000–2001 (13). The authors further noted that 10% of office visits for males in 2001 resulted in such prescriptions. Using more recent data from the NAMCS (1996–2007), Comer and colleagues (11) reported that 8.8% of office-based visits for children age six to 17 involved a prescription for one of five types of psychotropic medications (medications for attention-deficit hyperactivity disorder [ADHD], antidepressants, antipsychotics, mood stabilizers, and sedative-hypnotics). In addition, Aparasu and Bhatara (12) reported that 1% (approximately two million) of all office visits by children and adolescents in 2003–2004 involved prescriptions for antipsychotic medications, primarily second-generation antipsychotics, most frequently risperidone.

In regard to specific disorders, ADHD is one of the most common psychiatric disorders diagnosed among children and adolescents (21). As Mayes and colleagues (15) noted, as many as half of all youths treated at child psychiatry clinics have received an ADHD diagnosis. Comer and colleagues (11) reported that 42.9% of visits by children and adolescents that involved a prescription for psychotropic medications were by youths with disruptive behavior diagnoses, including ADHD. Olfson and colleagues (22) noted a fourfold increase in the likelihood of stimulant use among children between 1987 and 1996, and Zito and colleagues (23) found that two-thirds of office visits for ADHD medication were with primary care providers. Consistent with these findings, Shatin and Drinkard (9) reported that primary care providers prescribed approximately 70% of stimulants used by children and adolescents.

Depression and mood disorders are also relatively common among children and adolescents. Ma and colleagues (24) noted that various sources have cited prevalence rates of 2% to 8%, the latter in adolescent populations. Olfson and colleagues (22) reported an increase in antidepressant use among children and adolescents from .3% to 1.0% from 1987 to 1996, and Zito and colleagues (20) documented prevalence rates from 1.7% to 2.1% in 1996. Others have also noted increases (14,25), particularly among adolescents, and Comer and colleagues (11) found that 21.5% of office visits for psychotropic medications by children and adolescents were related to mood disorders.

In 2004, the U.S. Food and Drug Administration issued a black-box warning that antidepressants may increase the risk of suicidality among children and adolescents (26). After the warning, researchers noted a decrease in use of antidepressants among children, nationally (27) and in one state's Medicaid program (28). This decrease was accompanied by a decrease in the diagnosis of depression among children (28).

Antipsychotic medications are also prescribed to children and adolescents (29), including second-generation antipsychotics (29–31). According to one report, prescription of such drugs to youths increased nearly 500% from 1993 to 2002 (29). In addition, from 2000 to 2002, antipsychotics were involved in nearly one out of ten mental health office visits by children and adolescents and in over 18% of visits to psychiatrists (30). The authors noted that “second-generation agents composed 92.3% of the antipsychotic medications prescribed in office-based practice to children and adolescents” (30). Endocrine, metabolic, and other potential side effects have been noted with the use of second-generation antipsychotics (30–33), and at least one author has attempted to “provide a practical guide for the evaluation and management of antipsychotic-related adverse effects in this vulnerable population” and also noted that “the widespread use exceeds the database regarding efficacy as well as safety and tolerability in this population” (34).

In the context of the complexities associated with the use of second-generation antipsychotics, the American Academy of Child and Adolescent Psychiatry (AACAP) has posted a recently approved practice parameter for use of these drugs with children and adolescents that provides guidelines for screening and assessment, dosing, and monitoring (35). It also addresses risks, potential side effects, and safety issues related to these drugs and provides information about their efficacy and related research. The practice parameter includes the following caution: “While there is a growing body of evidence that has evaluated the use of atypical antipsychotics in youths, there remains a compelling need for methodologically rigorous trials assessing the efficacy and the acute and long-term safety of these drugs” (35).

Not surprisingly, given the marked increase in psychotropic prescriptions, polypharmacy—or the simultaneous prescription of two or more psychotropic medications—is not uncommon (11,18), and there are indications that it is increasing (11). In their recent review of polypharmacy among children and adolescents, Comer and colleagues (11) reported that according to 2004–2007 data, 20.2% of office-based visits that involved a psychotropic medication “mention” included two different classes of medication; in the context of a diagnosed psychiatric disorder the percentage rose to 32.3%. Further, they found that antidepressants were the psychotropic medications most commonly prescribed and that drugs to treat ADHD were the second most frequently prescribed during visits in which multiple medications were prescribed. With regard to specific combinations, antidepressants are frequently prescribed with stimulants for children with ADHD; clonidine is another drug prescribed along with stimulants, as are antipsychotics (18,23). Antidepressants and antipsychotics are combined in a substantial proportion of instances of polypharmacy (18), and some combinations also include mood stabilizers (11).

All psychotropic medications for children must be prescribed with care and caution, and this is especially the case with polypharmacy. The AACAP has made it clear that currently there is “limited support” for coprescribing psychotropic medications and that a “clear rationale” is required for doing so (36). Further, the AACAP has noted that before prescribing more than one medication, the provider must develop a plan both for the intervention and for supervising the intervention, must provide information and education to the child and family, must ensure assent from the youth and consent from the parent or guardian, and must conduct the medication trial according to AACAP guidelines.

Pediatric primary care providers and psychotropic medication

Studies examining numbers of patients and numbers of outpatient office visits have found that primary care providers prescribe most of the psychoactive medications used by children and adolescents (8,9,11,37). These findings are related largely to medications commonly prescribed for ADHD and depression. However, one study reported that more than three-quarters of office visits associated with prescriptions for anxiolytics, antipsychotics, and mood stabilizers for youths occurred with primary care providers, not psychiatrists (8).

Although the substantial role that pediatric providers play in prescribing psychotropic medications may not initially appear to be problematic, several related findings raise concerns and suggest that too much may be expected of these providers when they do not have the benefit of extensive training in behavioral health or the support of behavioral health specialists. First, research suggests that these providers do not always identify the disorders that their patients have (1,7,38). If the providers are not identifying these disorders, then it is unlikely that they are prescribing medications that may be beneficial to address the disorders or they may be prescribing medications that are not appropriate to a child's psychiatric diagnosis.

Second, pediatric primary care providers have reported substantial variations in their comfort level with diagnosing psychiatric disorders among children and with intervening with children who have various behavioral health problems; in general, they have reported greater comfort with problems such as ADHD, depression, and anxiety than with other disorders (39–42). Consequently, and as a recent study suggests, they may also be more comfortable prescribing stimulants and less comfortable prescribing second-generation antipsychotics and combinations of medications (Pidano AE, Honigfeld L, Bar-Halpern M, et al., unpublished manuscript, 2011). However, some pediatric primary care providers do prescribe classes of medications such as antipsychotics and coprescribe multiple psychotropics. In addition, although Harpaz-Rotem and Rosenheck (37) stated that they found marked similarity in the prescribing practices of primary care providers and psychiatrists (for example, in dosage and types of medications), they also reported that retention rates beyond the first visit were higher for psychiatrists, which can be critical in terms of monitoring medication compliance and response. “One approach to improving retention among patients treated by primary care providers,” they suggested, “may thus be to promote collaborative models of care which coordinate the work of primary care providers and nonprescribing mental health professionals (i.e., clinical social workers and psychologists” (37).

Lack of training is a third potential concern and may be one of the reasons for pediatric providers' limited comfort with identifying and diagnosing behavioral health disorders and managing children with them. Many pediatric primary care providers do not have extensive specialized training and may have no specialized training related to developmental and behavioral pediatrics (39,38).

Several strategies to improve prescription and management of psychotropic medications for children exist, notably development of treatment guidelines, engagement of pediatric providers in learning collaboratives, and use of electronic decision support with feedback and profiling systems (43). One solution to improving primary care providers' care in this area may be found in models of comanagement. A limited but growing literature suggests that comanagement of psychotropic medications can improve care for children and adolescents with behavioral health disorders. The Massachusetts Child Psychiatry Access Project (MCPAP) (44–47) and Washington State's Partnership Access Line (PAL) (palforkids.org) (48) provide evidence that comanagement is both feasible and promising.

The development of the MCPAP grew out of and was significantly influenced by an earlier project, Targeted Child Psychiatric Services (TCPS) (44). TCPS provided telephone consultation and support to primary care providers by child psychiatric providers, as well as psychiatric assessment and short-term counseling (one to four sessions) for patients when indicated.

The MCPAP, inaugurated in July 2005, has enrolled most pediatric primary care providers in Massachusetts, with the result that 95% of pediatric patients in the state are covered by the project. MCPAP services are available to all children and adolescents regardless of insurance status, at a cost to the Commonwealth of approximately $2 per child per year (44–46). Although the Massachusetts Behavioral Health Partnership “envisioned MCPAP as a service that could and should be supported by health insurers” (45), it was fully funded with $2.5 million allocated by the legislature to the Department of Mental Health (44).

MCPAP consists of six regional teams, funded by the Commonwealth, each of which includes the full-time equivalent (FTE) child psychiatrist, 1.0 or 1.5 FTE licensed behavioral health providers, and a care coordinator; some teams also have a psychiatric advanced practice nurse (45,46). The principal services provided to pediatric primary care providers and their patients through the MCPAP include telephone consultation, outpatient appointments with patients, referral when needed to behavioral health providers in the community, care coordination, and education for primary care providers with respect to mental health-related issues. A critical aspect of the project is regular communication with the pediatric primary care providers, which occurs both via the teams (especially the care coordinators) and in written form.

Another important aspect of the program is that pediatric primary care providers provide a substantial amount of the services themselves. After an MCPAP contact, primary care providers managed a third of patients (44). In addition, MCPAP psychiatric providers do not write prescriptions for patients. Rather, the primary care providers are responsible for prescribing, with consultation and communication with the psychiatric provider as needed (44). It is important to note that education and training are integral to the MCPAP. As Sarvet and colleagues (46) observed, “All consultative communications are intended to function as individualized, case-based education for PCCs [primary care clinicians]. MCPAP clinicians are encouraged to include discussion of topics such as relevant research, best-practice guidelines, and interviewing/assessment methods in telephone consultations.” In addition, MCPAP provides Web-based resources and regional conferences that offer continuing education (47).

From July 1, 2008, to June 3, 2009, nearly 30,000 contacts occurred between pediatric primary care providers and the MCPAP teams; the contacts involved clinical issues, evaluations, and referrals for behavioral health services in the community (45). The most frequently provided services during that period were telephone consultations and care coordination or requests to facilitate services in the community (45). Satisfaction data collected from 514 (38%) of 1,341 primary care providers indicated that more than 90% of respondents found the consultation service to be helpful; in follow-up surveys 90% again reported that the service was helpful (46).

The PAL program, launched in Washington State in 2008, was originally meant to duplicate the MCPAP. However, because of contextual differences, including the larger geographic area of the state and the population distribution, a number of modifications have been made. In addition, although PAL telephone services are available to providers who treat all children, the program focuses on children insured by Medicaid; if requested by the pediatric provider the consulting psychiatrist may conduct a televideo or in-person evaluation of the child or adolescent (48). Like MCPAP, PAL depends substantially on a telephone consultation service. PAL consultants are required to provide responses and recommendations consistent with evidence-based information. In addition, they are expected to provide advice and suggestions consistent with a guide for pediatric primary care providers that PAL has made available online (49). The PAL team includes a social worker and psychiatrists who are on call; services include full in-person or telemedicine evaluations and assistance in accessing ongoing behavioral services in the community (48). During its first two years of operation, PAL received more than 1,200 phone calls from pediatric primary care providers, with an average of 3.17 calls per participating primary care provider. Results from 133 satisfaction surveys showed that the providers overwhelmingly believed that the service improved access to care for their patients and increased their own skills in addressing the mental health needs of their patients (50).

In addition to patient-related services, PAL also provides educational opportunities for pediatric primary care providers. Hilt and colleagues (48) noted that these encompass both lectures and full-day trainings that are available at little or no cost to the primary care providers. The content addresses “best practice assessment and treatment of mental health disorders in children, as well as education about community resources and use of the PAL program” (48).

In summary, Massachusetts and Washington have used policy and system reform to develop statewide strategies to enable improved comanagement of medications between psychiatrists and pediatric primary care providers. Although the available published data from MCPAP and the PAL are limited, they suggest the possible benefits of such systems. However, continued and increasingly rigorous evaluation of the programs is necessary before more confident assertions can be made about their impact and effectiveness. In addition, the potential of these programs to move beyond consultation and provide a more integrated form of primary care and behavioral health care may contribute to improved management of psychotropic medications. Stroul (51) has described the potential of integrated services to address children's mental health issues.

On the basis of reports from these two state initiatives, the literature reviewed here, and our own experience and that of colleagues, we offer several recommendations.

It is clear that children use a significant amount of psychotropic medication, which is prescribed both by psychiatrists and by pediatric primary care providers. Primary care providers have consistently reported lack of knowledge about prescribing psychotropic medications and limited comfort in being the sole professional managing their patients' use of these drugs (39,40).

To benefit children who require psychotropic medications, a multipronged approach is necessary. Educational programming for child health providers at all levels of training needs to address pediatric psychopharmacology. Such training programs also should include education about comanaging with mental health colleagues children who have behavioral health conditions.

A full range of mental health services, including case management, should be available to augment psychopharmacological interventions. Not every child with a behavioral health disorder either requires medication or benefits from it. Many nonpharmaceutical, evidence-based treatments for children and adolescents are available, and it is critical that pediatric primary care providers have some familiarity with them, even if they are not trained and qualified to implement them.

Consultation, collaboration, and comanagement are promising approaches to improving the prescription and management of psychotropic medication for children and therefore need further study. These practice strategies need to be documented and rigorously evaluated. Although the results reported for MCPAP and PAL focus on utilization and provider-related variables, future research should include stakeholders, especially youths and families, in defining meaningful parameters for the assessment of efforts to reform comanagement practices. In addition, analyses of return on investment and patient outcomes are also essential in order to garner support for comanagement programs.

In conclusion, we concur with Kelleher and colleagues (52): “The promise of pediatric mental healthcare will not be fulfilled unless primary-care clinicians and behavioral health specialists forge new collaborative relationships that enhance the delivery of evidence-based care to affected children and their families.”

The authors thank the Child Health and Development Institute of Connecticut, which provided support for the initial report on which this Open Forum is based.

The authors report no competing interests.