Clinicians’ Utilization of Child Mental Health Telephone Consultation in Primary Care: Findings From Massachusetts

An estimated 20% of children and adolescents in the United States have a mental or behavioral disorder, and approximately 50%−80% of children and adolescents who may need psychiatric care are not receiving it (1,2). Many parents are unable or unwilling to access specialty mental health care for youths, in some cases because of the shortage of pediatric mental health providers (3). Thus many children and adolescents, hereafter referred to as children, with mental health problems are treated in primary care settings (4).

When treating children with mental health conditions, pediatricians (used here to describe any primary care clinician treating individuals ages 25 years or younger) often seek to collaborate with specialists (5,6), and a range of strategies has been developed to support pediatricians (5,7–9). The Massachusetts Child Psychiatry Access Project (MCPAP) is a publicly funded, novel consultation program that differs from traditional outpatient child and adolescent psychiatry consultation by providing almost immediate telephone access to mental health clinicians. MCPAP is intended to provide guidance for pediatricians during the clinical encounter (10), both to improve children’s access to mental health treatment and to enhance the quality of care provided by pediatricians.

Referral and treatment patterns for traditional mental health consultation services have been described previously (5), but there is a paucity of data regarding use of child mental health telephone consultation programs such as MCPAP. Previous research on MCPAP has provided descriptive information about initial consultations and pediatrician satisfaction (10), but little is known about the characteristics of children who are the subject of calls to consultation programs or the utilization of such programs after the initial contact. This retrospective study examined factors associated with initial contact with MCPAP by pediatricians and utilization of MCPAP services during the six months following the initial consultation. We hypothesized that similar to the pattern for use of traditional mental health consultation and referral methods (4–6), patients with more severe psychiatric diagnoses and complex psychotropic medication regimens would account for greater service utilization.

Methods

MCPAP is managed by the Massachusetts Behavioral Health Partnership and funded by the Massachusetts Department of Mental Health (DMH), but it is available for any child, regardless of insurance type or status. All Massachusetts primary care practices that treat children were invited to register with MCPAP, and it is estimated that more than 95% of primary care practices are enrolled, making MCPAP services available to more than 95% of children in Massachusetts (10). During regular business hours, MCPAP child psychiatrists respond within one hour of the pediatrician’s consultation request. Following the initial contact, the MCPAP team, made up of child psychiatrists, licensed therapists, and care coordinators, is available as needed for ongoing telephone support for pediatricians, facilitating a referral for evaluation by a MCPAP clinician within two weeks and providing interim treatment while coordinating referrals to mental health providers in the child’s community.

Following each pediatrician contact, the MCPAP clinician records encounter information, which is entered into a central database that is monitored and maintained by MCPAP staff. Using this database, we examined MCPAP utilization, defined as the mean number of contacts between MCPAP clinicians and the child’s pediatrician within 180 days of the initial contact for individuals younger than 25 years old, for contacts initiated between July 1, 2008, and June 30, 2009.

The initial contact was categorized by the topic of the consultation question, including clarification of the child’s psychiatric diagnosis; initiation or management of psychotropic medications; general information about psychotropic medications; access to community resources; or other issues, such as requests for guidance caring for a family in crisis or a child with academic problems. Children’s medication status at the time of the initial contact was categorized as no psychotropic medication, antipsychotic or mood stabilizer medication, antidepressant or anxiolytic medication, attention-deficit hyperactivity disorder (ADHD) medication, or polypharmacy (receiving medications from two or more of the categories listed above). The child’s primary mental health or behavioral problem at the time of the initial contact was categorized according to the following hierarchy from most to least severe: developmental disorder; psychosis or bipolar disorder; depression or suicidality; anxiety-related disorder; ADHD; behavioral problem; or other problem, such as substance abuse, eating disorder, or child abuse. For children presenting with more than one problem, the most severe problem was considered the primary problem.

Records of the consultation question, child’s medication status, and primary mental health or behavioral problem commonly were coded according to predefined categories, but in some cases alternative responses were added by MCPAP clinicians. These alternative responses were postcoded to appropriate predefined categories or categorized as “other.” Children’s insurance type was categorized as public insurance, private insurance, and other or unknown. The date of the initial contact was categorized as January–March, April–June, July–September, or October–December. This study was approved by and conducted in compliance with the Partners HealthCare Institutional Review Board.

We conducted bivariate analyses to describe associations between MCPAP utilization and each child and consultation characteristic. We also conducted multivariate regression analysis to examine if there was a significant association between these characteristics and MCPAP utilization. The final model included all available characteristics. All analyses were conducted by using SPSS 13.0 for Windows (Sept. 1, 2004 release).

Results

We identified 4,436 initial MCPAP contacts that occurred from July 1, 2008, through June 30, 2009 (Table 1). Sixty-one percent of the contacts were for boys. A majority of contacts were for children aged 12–17 years (41%) and children aged six to 11 years (38%). Only 14% of initial contacts were for children ages five years or younger.

Almost half (48%) of MCPAP contacts were to clarify the child’s psychiatric diagnosis, and consultations related to accessing mental health services (28%) were also common. At the time of the initial contact, a majority of children (66%) were not taking psychotropic medications, 15% were taking ADHD medication, 11% were taking antidepressant or anxiolytic medication, 7% were taking multiple psychotropic medications, and 3% were taking antipsychotic or mood stabilizer medications. Pediatricians’ initial contacts with MCPAP were most commonly for children with depression or suicidality (20%), followed by anxiety-related disorders (17%), and ADHD (17%). Slightly less than one-third of contacts were for children with other mental health problems (31%). Among the initial contacts with MCPAP, over half (56%) were for privately insured children, 24% were for publicly insured children, and 20% were for children who had other or unknown insurance. Initial MCPAP contacts were less frequent during July-September (19%) than during other seasons.

The mean±SD number of contacts between the pediatrician and MCPAP clinicians in the 180 days following the initial contact was 3.83±3.46. After adjustment for other child and consultation characteristics, the analysis showed that MCPAP utilization was significantly greater following initial contacts for initiation or management of psychotropic medications (p=.03) compared with contacts for clarification of the child’s psychiatric diagnosis. Utilization also was greater when the initial contact was for children prescribed polypharmacy (p=.04) compared with contacts for children prescribed no medications (Table 1). Conversely, there was significantly less MCPAP utilization following initial contacts for general information about psychotropic medications, resources and community access, or other topics (p<.001 for all) compared with contacts about the child’s psychiatric diagnosis. MCPAP utilization was greater for initial contacts regarding children with public (p=.03) or private insurance (p<.001) compared with those whose insurance status was other or unknown and for contacts initiated in October–December compared with July–September (p<.001). There was no significant difference in MCPAP utilization on the basis of the child’s sex, age, or primary mental health or behavioral problem.

Discussion

Given the national shortage of child mental health specialists (3) and the substantial unmet need for child mental health treatment (1,2), pediatricians are likely to continue playing a critical role in addressing the needs of children with mental health problems for the foreseeable future. Telephone consultation programs like MCPAP may provide necessary support to pediatricians, and our findings suggest that telephone consultation may be used for guidance in diagnosing and treating mental health conditions. Although MCPAP is publicly funded, it is accessed on behalf of children with any insurance type or status.

Many pediatricians report reluctance in prescribing psychotropic medications due to limited training, lack of time during the clinical encounter to address psychosocial issues, and safety concerns (11,12). Thus, when treating children with mental health problems, pediatricians may seek consultation with specialists. Our finding of increased program utilization for consultation regarding management of psychotropic medications, particularly polypharmacy, is consistent with this possibility and also with our original hypothesis. Notably, contrary to expectation, telephone consultation program utilization was not significantly associated with the child’s primary mental health or behavioral problem. Pediatricians may manage children with less severe mental health conditions independently, referring more complex cases for consultation regardless of the diagnosis (6).

Telephone consultation programs have a potential advantage over traditional models by providing almost immediate access to specialists while potentially enhancing pediatricians’ understanding of mental health treatment through conversation during the consultation (10). Furthermore, online resources provided by telephone consultation programs (www.mcpap.org, www.palforkids.org) may enhance pediatricians’ knowledge of psychiatric diagnosis and treatment.

Our findings are consistent with previous research indicating that a child’s insurance status does not determine pediatricians’ referral for psychiatric treatment (13). Our data did not allow us to determine whether the substantial use of MCPAP for privately insured children is due to the absence of similar consultation programs for privately insured youths, greater pediatrician awareness of or preference for MCPAP, or other factors. However, given that the service is used for privately insured children, public-private partnership models to support public programs such as MCPAP may be appropriate.

Previous research has shown that during autumn, there is a relative increase in the frequency of children presenting to emergency departments with mental health issues, likely because of school referrals and increased phone calls to crisis lines (14,15). We also found that MCPAP utilization was greater during autumn, and planning and allocation of resources for similar telephone consultation programs should account for seasonal variations in use.

The results of this study must be interpreted in light of its limitations. The data available to us did not contain potentially important information—such as race, ethnicity, socioeconomic status, existing community resources, or clinical severity—that likely explains some of the variation in MCPAP utilization. Nor did we have individual level information on pediatricians, preventing us from clustering contacts for statistical analysis, examining the extent of MCPAP utilization by pediatricians, or evaluating pediatrician characteristics associated with the choice of accessing the program. The nature of MCPAP encounters also was not specified, so we were unable to assess the type or intensity of MCPAP resources associated with each contact, nor were data available to allow assessment of the broader feasibility or acceptability of MCPAP.

Conclusions

Despite these limitations, this study provided empirical information on a collaborative consultation program designed to improve the quality of mental health care delivered to children in primary care settings. Telephone consultation models may offer an efficient approach for addressing workforce limitations (3), and this study can help inform the development of telephone consultation programs by identifying clinical and social factors associated with their use. Future research should include identification of essential elements of telephone consultation services, pediatrician-specific factors associated with use, and variations by state in program implementation. As telephone consultation services may enhance collaboration between clinicians and support the incorporation of mental health treatment into primary care, developing and improving such services may contribute to national efforts to improve the quality of child mental health treatment.