Psychiatric illnesses such as depression, anxiety, posttraumatic stress disorder (PTSD), and substance abuse are associated with significant psychosocial morbidity, excess mortality, and increased cost to society, yet these disorders frequently go undetected and untreated. Only a minority of patients with psychiatric disorders will ever receive treatment, whether from a generalist or specialist.
In 1975, Haggerty coined the term hidden morbidity as it was shown that psychiatric illness was one of the most common reasons for consulting a medical practitioner, and the bulk of psychiatric disorders (95%) were treated without specialist involvement.' This trend has persisted through the years, and psychiatric illnesses are still among the most common disorders presenting in primary care settings.
Regier and colleagues' term de facto mental health services system refers to the fact that 50% of the nation's mentally ill receive treatment solely from primary care practitioners.2 Recent epidemiological surveys confirm this trend, showing that psychiatric disorders, particularly depressive disorders, substance use disorders, and anxiety disorders, are routinely treated in primary medical care settings. They further contended that children and elderly patients were most likely to receive treatment from primary care physicians for symptoms associated with mental disorders. Reports by othershave continued to document the large and growing role that pediatricians play in the identification and treatment of children and adolescents with mental disorders.35
Internalizing and Externalizing Symptoms
Understanding the existing knowledge and current gaps in the delivery of service to this population requires a systemwide approach. This article reviews the research and clinical practice domains that are important to the development of effective services for this largely underserved and untreated population.
DETECTION OF MENTAL DISORDERS
Even though most patients with mental disorders are seen only in primary care settings, these disorders frequently go unrecognized by primary care practitioners. Between 15% and 40% of adult primary care patients have diagnosable mental disorders, yet less than 20% of these are treated by mental health specialists. 6
Adolescents have the lowest utilization of health care services of any age group and are the group least likely to seek care in traditional office-based settings; adolescents are also more likely to be uninsured than any other age group.7 This is especially true for ethnic minorities and the economically impoverished.8 These adolescents tend to have the most significant health problems, as poverty is associated with increased risk of disease and chronic illness. 9
The single leading cause of disability among adolescents is not physical disabilities, but mental disorders, which comprise 32% of all disabilities for this age group. Between 17% and 22% of youth younger than age 18 suffer from developmental, behavioral, or emotional problems. The former US Office of Technology Assessment (OTA) estimates 7.5 million youth younger than age 1 8 are in need of mental health services, but less than one third of these children receive services.9
Among adolescents, the suicide rate has increased more rapidly than among the general population.10 Most adolescents who commit suicide have suffered from a psychiatric disorder including affective disorders (especially depression), conduct disorders, substance abuse, anxiety disorders, eating disorders, and schizophrenia.10
During the past 60 years, the leading cause of death for adolescents has changed from natural causes to injury and violence." Overall mortality rates for young people rise by 239% when they reach ages 15 to 19, with violence responsible for this dramatic increase." The vast majority of the deaths in this age group are from accidents, homicide, or suicide.12 With such a high mortality rate among adolescents, it is imperative for primary care providers to be aware of symptoms that may indicate a need for mental health treatment.
Primary care physicians range widely in their ability to correctly identify children with behavioral or psychiatric disorders, identifying between 0.6% and 16% in different medical settings.4 Although psychosocial problems are both common and disabling, less than 2% of children and adolescents receive care from mental health specialists in any given year.4·5 This lack of mental health services for children creates a burden on other service sectors, including schools, child welfare services, correctional facilities, and nonpsychiatric health care providers, particularly pediatricians.
As with the adult and adolescent population, relatively few children are seen by mental health specialists. It is estimated that between 5% and 15% of all children younger than age 1 8 have significant mental health problems, yet only 2% of children with mental health difficulties are seen by specialists, while primary care physicians see approximately 75% of children with psychiatric disabilities.4,5
IDENTIFICATION AND DIAGNOSIS
One response to the increasing recognition by physicians of the need to treat psychosocial problems has been screening within primary care settings.13 Screening for mental health problems has been popular since 1985 when 46% of physicians reported using questionnaires in their practice. Screening may be especially important because primary care physicians range so widely in their ability to recognize children with behavioral or psychiatric problems, identifying a small fraction in different medical settings 14
Since pediatricians see the majority of children with disorders, researchers across the country have focused on screening in pediatric practices as a way to assist in proper identification of this population. Several studies have evaluated the ability of pediatricians to accurately assess psychiatric disability among their patients.4,15 In these studies, using an objective measurement of psychopathology, under-recognition by pediatricians was examined.
Results showed the pediatricians had high specificity and low sensitivity in their identification of psychiatric disorders - in other words, they failed to detect many positive cases but seldom misidentified negative cases. Pediatricians were more likely to identify emotional or behavioral problems in those children with high levels of service. 3,4
COST OF DISPLACED UTILIZATION
Lack of recognition of children's mental health problems may lead to high health care utilization and hide the costs of mental health in primary care. Children with mental health problems have been found to be high utilizers of health services in general.3,4,5,14,51,16 This relationship has been documented beginning in preschool children and extending into adolescence.
In a study on pediatric health care utilization in a private, nonmanaged care system in Chicago, Lavigne et al.5 found that Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition (DSM-ITI-R) diagnoses in preschoolers were associated with increased emergency department use. Dimensional ratings of psychopathology among preschoolers appeared to be more sensitive indicators and were associated with both increased primary care utilization and emergency department visits. Higher Child Behavior Symptom Checklist (CBCL) scores were associated with more utilization, suggesting a "doseresponse" relationship. Thus, behavior problems and family factors may be better predictors of health care utilization than diagnoses.
Another study examined psychosocial disability in 1 ,800 children in six Kaiser Permanente Northern California pediatric settings.14 This study included children aged 2 to 18 years (41% were 2 to 5 years old) from several ethnic groups (23% Latino, 14% black, 16% Asian, and 54% white). Measures of psychosocial disability included the Pediatric Symptom Checklist (PSC), the Beck Depression Inventory (BDI), and parental and pediatrician reporting. The PSC was used to classify positive cases as internalizers versus externalizers for purposes of cost comparison (Table). Outcome measures included utilization (number of outpatient visits) and costs.
Results from multiple regression models predicted that the average log visits was 7.6 for internalizers, 5.3 for externalizers, and 10.5 for children with chronic illness compared to 4.3 visits for the total sample. The average log health care costs for internalizers were higher than those for the total sample, but were still less than those for children with chronic illness.
Among internalizers, regression models of cost and age predicted high levels of medical costs and low levels of psychiatric cost at young ages. With increasing age, predicted medical costs of internalizers decreased, approaching medical costs of the entire sample while psychiatric costs of internalizers increased. These patterns of utilization and cost represented the cost effect of nonrecognition and delayed intervention. The same pattern was found for externalizers; however, they began with lower predicted medical costs and ended with higher psychiatric costs. Parents, teachers, and pediatric primary care providers identified externalizers more readily than internalizers.
The early increased costs among the internalizers imply there is a delay in the detection of mental illness and institution of psychiatric services for internalizers. This supports the notion that pediatric psychosocial morbidity is undetected early in life and results in displaced utilization in terms of medical visits. Internalizers thus overburden the pediatric primary care clinics before they are recognized and referred to psychiatry. Other studies with asthmatic children,17 within a large pediatric sample,4 with children with behavioral disorders, and with children and adolescents with attentiondeficit/hyperactivity disorder also have documented the impact of psychosocial morbidity on displaced utilization and increased cost of care.
Multiple studies have shown that depression affects adult primary care utilization. The utilization and costs of care for the parents of children in the study described above also were measured for 1 year prior to the screening of their children. Using the same multiple regression models, the predicted visit average was 4.6 for all parents in the sample. Depressed parents were predicted to have 0.2 visits more per point in their BDI score. The BDI status of the depressed parent predicted an additional $24 dollars for each point above the BDI cutoff in their primary care utilization in the prior year. The models also predicted additional 15 primary care visits per year for the parent of a chronically ill child.
Thus, psychosocial disability among children not only increases child utilization rates and the cost of care, but also is associated with an increase in parental utilization and health care costs. The impact of parental factors such as depression,18 lack of social support, low education and income, and single parent status also have been linked in other studies13,15 to psychosocial morbidity and decreased physical and psychological functioning in children.
It is well documented that chronically ill children, as a group, are high utilizers of primary care.15·19 Moreover, chronically ill children contribute to increased primary care utilization of their parents. These associations support a circular and interactive model of escalating problems in children with chronic illness and their parents, leading to increased health care utilization.
Parental depression increases the vulnerability of children with chronic physical illness for psychosocial dysfunction.18 Indeed, approximately 15% of the children in the aforementioned studies had chronic health conditions. These children were approximately two times more likely to be identified as experiencing psychosocial dysfunction based on their PSC scores.
Viewed from another perspective, more than one quarter of all children with psychosocial dysfunction have a chronic health condition. Furthermore, more than 40% of the top 10% of total healthcare utilizers were PSC positive and 25% of all PSC positives were high utilizers. Thus, there is a strong relationship between psychosocial dysfunction and health care utilization, and a substantial proportion of those with psychosocial dysfunction have a chronic illness.
In another study, Murphy et al .4 found 15% of the children had chronic health conditions. These children were approximately two times more likely to have positive screening scores on the PSC. In other words, more than one quarter of all positive PSC screens were children with chronic health conditions.
When patients are referred to mental health specialists, as many as 50% do not follow through and keep their appointment. The social stigma associated with depression, gaps in physician knowledge about efficacious treatment, and the lack of integration of mental health services into primary care constitute additional barriers to effective treatment. The frequency and prevalence of psychosocial issues are of great concern as identification of these disorders allows for the expethent provision of treatment.
Barriers to behavioral health treatment appear to be multifactorial. Economic allocations for behavioral health programs have decreased over the past decade.20 The literature on compliance to treatment addresses the psychological factors affecting the child, the parent, or the family as well as provider factors affecting compliance.21 Studies reviewing the follow-through on medical treatment indicate that provider factors influence treatment adherence in a pediatric setting.22 Services for this population are delivered in multiple settings, including primary care and specialty behavioral health care clinics. However, a large proportion of the population in need is not actively involved in care.23
Several authors have reviewed patterns of compliance and adherence in primary care and specialty care settings.23 Maternal education determines compliance with care at a general level. In specialty care, socioeconomic factors more than clinical severity appear to be the major determinants to followthrough.21 Treatment compliance affects large systems of care at the individual patient's level, both clinically and ethically. It affects service delivery at the organizational and economic level. One third to half of the patients in large behavioral care settings received one to three sessions regardless of severity or benefit limitations.20
MODELS OF CARE
Four models of mental health care are available within primary care settings.24-25
Psychological screening has arisen out of the recognition that rates of both case finding and referral are low within standard medical practice. Research with adults and children has documented that patients persist in making unnecessary appointments with medical doctors in an attempt to alleviate their suffering when mental health issues, such as depression or child behavior problems, are the actual cause of their difficulties.15
The use of a standardized and well-validated questionnaire that is selfadministered by the patient is one solution to this problem, and many physicians use instruments such as the PSC, the Parenting Stress Index, the BDI, and other assessment tools. For the past decade, there has been evidence mat psychological screening programs in pediatrics are associated with increased rates of mental health treatment.26 Recent studies have shown improved mental health outcomes for screened children who receive services.27
Early Preventive Intervention
Early prevention focuses on education. In this model, the physician retains sole responsibility for medical management of the patient while the mental health professional assumes the role of a teacher or informed colleague who provides advice on topics such as pregnancy prevention, smoking cessation, or the ongoing treatment and management of a chronic medical disease (eg, diabetes or asthma).
This model includes three collaborative approaches. The first is a client-centered approach in which the primary goal is for the consultant to develop a plan to help a patient who has been referred by a medical doctor. The second approach is consultée centered, with the primary focus to increase the practitioner's understanding or emotional mastery of issues involved in caring for a particular patient population.
The third approach is that of the collaborative team. The contribution of information from various disciplines and perspectives in search of a comprehensive solution to health problems is the main advantage of this approach. In the treatment of many conditions, a multimodal approach has been shown to be more effective than a single treatment modality.25
Referral to Mental Health Clinics
Referral to mental health clinics may have limited value given the concerns with the stigmatization of psychiatric diagnoses and the reluctance of patients to accept and act on these diagnoses. The authors' research has indicated that only 7% of parents contacted psychiatric departments within a health maintenance organization setting for mental health services for their children after referrals were made by pediatrician.25
In addition, a substantial number of psychiatric patients seen in community settings fail to comply with psychiatric treatment recommendations. Noncompliance takes many forms, including failure to keep initial appointments, complete outpatient referrals from emergency services, keep follow-up appointments after hospitalization, remain in treatment, or take prescribed medications. These aspects of noncompliance are highly associated with relapse, recidivism, and rehospitalization.
The 1999 Surgeon General's report on mental health28 concluded that stateof-the-art treatment for mental disorders is not being translated or disseminated into clinical practice. The report suggested certain groups such as children may be underrepresented and recommended more emphasis be placed on research regarding the identification and treatment of mental health issues in primary medical care. This recommendation was based on the recognition that large numbers of children have mental health problems but few receive services for these problems and the knowledge that primary care clinicians are the major providers of mental health services to children and adolescents. In addition, primary care clinicians provide most psychotropic drug prescriptions to children and adolescents.29
Horwitz et al.30 assessed the extent to which federal research agencies' portfolios of funded grants were consistent with the significant attention to primary care. They found that federal agencies' planning documents devote considerable attention to the need to understand the identification and treatment of children's behavioral and emotional issues within primary medical care settings. However, the authors discovered little evidence that such attention has resulted in aggressive programs of research in this area.
In fact, of a mere 63 grants examining children and adolescent issues in primary care, only 21 (0.059c of the total Computer Retrieval of Information on Scientific Projects (CRISP) portfolio) involved studies of behavioral and emotional issues in primary care.30 Even in the institutes with the largest number of these studies, this research represents less than 1% of all portfolio projects funded, and according to the content of the abstracts, few examined or attempted to modify usual practice.30 Considerable work in adults shows that care for a prevalent mental health problem, such as depression, can be provided in the primary care setting, but without appropriate incentives, innovative care models are not maintained.30
Horowitz et al. examined the distribution of child or adolescent and adult studies on the treatment of depression within primary care. They found that only six (5.8%) of 103 research abstracts on this topic had children or adolescents as their focus.30 On a topic that is important in children, adolescents, and adults, adults received more than 15 times the research attention compared with children. These findings have been consistent with those presented previously in the 1989 Institute of Medicine report on the National Institute of Mental Health's research on children and re-examined the interim progress in 19943U2;A11 reports may lead to the conclusion that albeit some gains had been, the gaps across all research areas had not been appreciably closed.30
IMPLICATIONS FORTHE FUTURE
Documenting the cost effectiveness of children's mental health programs in primary care has been problematic. Several authors have described how most of the returns on investments in child mental health are in the form of long-term outcomes, the impact on other systems of care (juvenile justice, schools, etc), and the effects on other populations.
Future studies on cost effectiveness need to include indicators and outcomes for both parents and children. Interventions need to be designed, developed, implemented, and evaluated in an integrated care system. Evaluation needs to involve systemwide indicators and outcomes. Finally, long-term studies that examine the effectiveness of early mental health intervention in preventing or reducing adult mental health problems and disability are needed.
Although the importance of primary care as a system for identifying and treating behavioral and emotional problems was recognized more than 20 years ago, not enough research has been conducted in these areas. Without an understanding of the opportunities and barriers that face physicians confronted with these and a host of other important issues, primary care will remain an underused venue for providing mental health services to children. Developing knowledge of the capacity for change in primary care necessitates creating a wellplanned research agenda.
Identifying whether adequate workforce is available, supporting integration and collaboration between clinical and research settings. If sufficient applications are submitted in response to requests for proposals and review committees include experts in children's mental health services, research will be critical for the development of solutions.
Without efforts to change the agenda, the research incentives, and the review process, the potential of primary care for identifying and treating behavioral and emotional problems in children is unlikely to ever be realized. Finally, the gap between research and clinical practice will require the ongoing support and endorsement from all groups to adhere to standards of practice based on scientific evidence.
1. Haggerty RJ, Roghmann KJ, Pless IB. Child Health and the Community. New York, NY: John Wiley & Sons; 1975.
2. Regier DA, Goldberg ID, Taube CA. The de facto US mental health services system: a public health perspective. Arch Gen Psychiatry. 1978;35:685-693.
3. Costello EJ, CostelloAJ, Edelbrook C, et al. Psychiatric disorders in pediatric primary care: prevalence and risk factors. Arch Gen Psychiatry. 1988;45:1107-1116.
4. Kelleher KJ, Mclrnerny TK, Gardner WP, Childs GE, Wassermann RC. Increasing identification of psychosocial problems: 1 979- 1 997. Pediatrics. 2000; 1 05:113-1321.
5. Lavigne JV, Binns HJ. Chritoffel KK, et al., and the Pediatric Practice Research Group. Behavioral and emotional problems among preschool children in pediatric primary care: prevalence and pediatrician's recognition. Pediatrics. 1 993:9 1 :649-655.
6. Regier DA, Goldberg ID, Bums BJ, Hankin J, Hoeper EW, Nycz GR. Specialist/generalist division of responsibility for patients with mental disorders. Arch Gen Psychiatry. 1982; 39:219-224
7. Kellener KJ, Child GE, Wasserman RC, Mclnerny TK, Nutting PA, Gardner WP. Insurance status and recoginition of psychosocial problems. A report from the Pediatric Research in Office Settings and the Ambulatory Sentinel Practice Networks. Archive of Pediatric and Adolescent Medicine. 1997;151 (11):1 109-1 115
8. McCormick, Kass, Elixhauser, et al. Annual report on access to and utilization of health care for children and youth in the Unites States-1999. Pediatrics. 2000; 105:219-230
9. Burns, Taube & Taube (1990). Background paper for U.S. Congress OTA Adolescent
10. Brandenburg NA, Friedman RM, Silver SE. The epidemiology of childhood psychiatric disorders: Prevalence findings from recent studies. J Am Acad Child Adolesc Psychiatry. 1990;29:76-83
11. Murphy SL. 2000 Deaths:Final data for 1 998. National Vital Statistic Report 48:11 (Dhs Publication No. PHS 2000-1120) Hyattsville.MD National Center for Health Statistics.
12. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARS) [Online]. (2002). National Center for Injury Prevention and Control, Centers for Disease Control and Prevention (producer). Available from: URL: www.cdc.gov/ncipc/wisqars. [2003 March 27].
13. Murphy JM, Reede J, Jellinek MS, Bishop SJ. Screening for psychosocial dysfunction in inner-city children: further validation of the Pediatric Symptom Checklist. J Am Acad Child Adolesc Psychiatry. 1 992;3 1 : 1 1 051111.
14. Horwitz SM, Leaf PJ, Leventhal JM, Forsyth B, Speechley KN. Identification and management of psychosocial and developmental problems in community-based, primary care pediatric practices. /Wiamcs. 1992;89:480485.
15. Bernal P, Bendell-Estroff D, Abdouraham JF, Murphy M, Keller A, Jellinek MS. Psychosocial morbidity. Arch Pediatr. 2000; 154:261266.
16. Zuckerman B, Moore KA, Glei D. Association between child behavior problems and frequent physician visits. Archives of Pediatrics & Adolescent Medicine. 1996; 150(2): 146153.
17. Lozano P, Fishman P, VonKorff M. Hecht J. Health care utilization and cost among children wim asthma who were enrolled in a health maintenance organization. Pediatrics. 1997:99:757-764.
18. Wallander JL, Thompson RJ Jr. Psychological adjustment of children with chronic physical conditions. In: Roberts MC, ed. Handbook of Pediatric Psychology. New York, NY: Guilford; 1995;124-141.
19. Newacheck PW, Taylor WR. Childhood chronic illness: prevalence, severity, and impact. Am J Public Health. 1992;82: 364371.
20. The Hay Group, document for the National Association of Psychiatric Health Systems and the Association of Behavioral Group Practices entitled "Health Care Plan Design and Cost Trends - 1988 through 1998.
21. Armbruster, P. & Kazdin, A.E. Attrition in child psychotherapy ,In TH Ollendick & RJ Prinz (Eds) Advances in Child Psychology.( Vol 16 pp.81-108) New York: Plenum.
22. Di Matteo, MR et al.(1993).Physicians' characteristics influence patients adherence to treatment: Results for the medical outcomes study. Health psychology, 12,93-102
23. U.S. Public Health Service, Report of the Surgeon General's Conference on Children's Mental Health: A National Action Agenda. Washington, DC: Department of Health and Human Services, 2000.
24. Knesper DJ, Riba MB, Schwenk TL. Primary Care Psychiatry. Philadelphia, Pa: WB Saunders Co; 1997.
25. Saravay SM. Psychiatric intervendons in the medically ill: outcome and effectiveness research. Psychiatr Clin North Am. 1996;19:467-480.
26. Jones, R., Latkowski, M., Green, D. & Ferre, R. 1996 Psychosocial assessment in the in the general pediatric population. Journal of Pediatric Health Care. 10: 10-16.
27. Jellinek MS, Murphy JM, Little M, Pagano ME, Comer D, Kelleher K. Use of the Pediatric Symptom Checklist (PSC) to screen for psychosocial problems in pediatric primary care: A national feasibility study. Archives of Pediatrics and Adolescent Medicine. In press.
28. Mental Health: A Report of the Surgeon General. Rockville, Md: US Dept of Health and Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Studies, National Institutes of Health, National Institute of Mental Health; 1999.
29. Kelleher KJ. Hohmann AA, Larson DB. Prescription of psychotropics to children in office-based practice. Am J Dis Child. 1989;143:855-859.
30. Horwitz SMC, Kelleher K, Boyce T, et al. Barriers to health care research for children and youth with psychosocial problems. JAMA. 2002;288:1508-1512.
31. Institute of Medicine. Research on Children and Adolescents With Mental, Behavioral and Developmental Disorders: Mobilizing a National Initiative. Washington, DC: National Academy Press; 1989.
32. Report Card on the National Plan for Research on Adolescent Mental Disorders: The Midway Point. Washington DC : Institue of Medicine : 1994
Internalizing and Externalizing Symptoms