Depression treatment during outpatient visits by U.S. children and adolescents

Transcription

1 Journal of Adolescent Health 37 (2005) Original article Depression treatment during outpatient visits by U.S. children and adolescents Jun Ma, M.D., R.D., Ph.D., Ky-Van Lee, Ph.D., and Randall S. Stafford, M.D., Ph.D.* Program on Prevention Outcomes and Practices, Stanford Prevention Research Center, Stanford University, Palo Alto, California Manuscript received April 19, 2005; manuscript accepted July 28, 2005 Abstract Keywords: Purpose: Depression affects approximately 2 8% of all children and adolescents, and treatment of depression in children and adolescents has been the center of recent serious debates. We examined national trends in depression visits and treatment among outpatients aged 7 to 17 years. Methods: We analyzed visit-based data between 1995 and 2002 in two national ambulatory care surveys. Results: The number of visits by children and adolescents during which depression was reported more than doubled from (1.44 million) to (3.22 million). The proportion of these visits during which antidepressants were prescribed rose slightly from 47% in to 52% in , whereas the proportion during which psychotherapy or mental health counseling was provided declined from 83% to 68%. Selective serotonin reuptake inhibitors (SSRI) represented 76% of all antidepressants prescribed in and 81% in In absolute terms, SSRIs were reported in 1.35 million visits in , reflecting a 2.6-fold increase from Fluoxetine was prescribed in 207,914 visits in and increased 100% to 415,580 visits in The use of sertraline increased by 62% to 345,576 visits and paroxetine by 269% to 279,275 visits. Conclusions: We observed a declining trend in the provision of psychotherapy/mental health counseling during outpatient visits by children and adolescents diagnosed with depression. Although the likelihood of receiving antidepressants remained essentially unchanged, the number of children and adolescents whose visits involved prescription of antidepressants, particularly SSRIs, has increased markedly through Although fluoxetine remained the most commonly prescribed, other SSRIs were increasingly prescribed through These trends raise concerns regarding the widespread off-label use of antidepressants lacking reliable evidence of safety and efficacy for use in children and adolescents Society for Adolescent Medicine. All rights reserved. Depression; Psychotherapy; Antidepressants; Children; Adolescents; NAMCS; NHAMCS Depression is a major risk factor for suicide, which ranks third as a cause of death among teenagers in the United States, and is often accompanied by other psychiatric disorders, poor social functioning, and a high risk of substance abuse [1]. The prevalence of major depressive disorder, the most serious form of all depression diagnoses, is estimated to be approximately 2% in primary school-aged children and 4% to 8% in adolescents [2,3]. Under-diagnosis and *Address correspondence to: Dr. Randall S. Stafford, Stanford Prevention Research Center, Stanford, CA address: jun.ma@stanford.edu under-treatment of depression in children and adolescents has been a national and historical problem [4]. The treatment of children and adolescents with depression has been the center of serious debates in the past two years, particularly because of a suspected increased risk of suicidality associated with selective serotonin reuptake inhibitors (SS- RIs), a dominant class of antidepressants [5]. In the meantime, epidemiological and ecological data suggest a positive relationship between increased prescribing of SSRIs and decreased adolescent suicide in the last decade [3]. It is the position of the American Academy of Child and Adolescent Psychiatry that psychotherapy is appropriate X/05/$ see front matter 2005 Society for Adolescent Medicine. All rights reserved. doi: /j.jadohealth

2 J. Ma et al. / Journal of Adolescent Health 37 (2005) treatment for all depressed children and adolescents whereas antidepressant medications are indicated for those with severe, psychotherapy-resistant symptoms [3,6]. In practice, however, antidepressants became the second most commonly used psychotropic medications after stimulants by 1996, due in large part to increases in SSRI prescriptions [7]. Since their market entry, SSRIs have quickly emerged as the leading antidepressants prescribed to children and adolescents because of their relatively favorable adverseeffect profile, low lethality after overdose, and simplified dosing [8]. Studies have found that during the mid-1990s, SSRIs comprised 43% to 50% of all antidepressants prescribed to children and adolescents [3,7]. The marked increased use of SSRIs by prescription in children and adolescents occurred in the absence of evidence of safety and efficacy for all but one SSRI (i.e., fluoxetine). In 1997, a randomized controlled trial first showed the efficacy of fluoxetine in the treatment of depression in children and adolescents [9]. Six years later, the Food and Drug Administration (FDA) approved fluoxetine for use by 7- to 17-year-olds [10]. To date, fluoxetine is the only antidepressant approved by the FDA for use in patients younger than 18 years of age; the prescription of any other antidepressants or SSRI constitutes off-label use. In June 2003, regulatory agencies in the United States and the United Kingdom issued safety warnings concerning the use of paroxetine in children and adolescents [5]. In October 2004, the FDA directed manufacturers of all antidepressants to print a black-box warning label alerting health care providers about the increased risk of suicidality in children and adolescents treated with these agents and the need for close patient monitoring [11]. These recent regulatory actions are expected to have a dampening effect on the usage of SSRIs and possibly other antidepressants in children and adolescents, although it is contended that it would be imprudent to indiscriminately withhold pharmacological treatment for pediatric depression [4]. Data describing historical trends of antidepressant use in children and adolescents are needed to gauge the impact of recent regulatory actions through comparisons between current and future prescribing practices. The current literature tracks national trends of antidepressant use in U.S. children and adolescents only through the mid-1990s. Zito et al [7] found that antidepressant prevalence for youths age 20 and younger increased from 2% in 1987 to 10% in 1991 to 21% in 1996, with the highest prevalence for 15- to 19-year-olds in Nearly 60% of children and adolescents who received a prescription for antidepressants in were between the ages of 7 and 17 and 54% were female [12]. Similar trends were found in other countries. For example, in England, antidepressant prescriptions increased by 1.7-fold from 1992 to 2001, with SSRI prevalence increasing 10 times from.5 to 4.6 per 1000 persons 18 years [13]. Our study aims to document national trends of antidepressant usage, as well as provision of psychotherapy and mental health counseling services, among depressed children and adolescents seeking ambulatory care from 1995 through We also examine patient and physician factors that are significantly associated with the use of pharmacotherapy and psychotherapy/mental health counseling in the treatment of depressed children and adolescents. Methods Data sources Annual data from1995 through 2002 were obtained from the National Ambulatory Medical Care Survey (NAMCS) and the Outpatient Department (OPD) component of the National Hospital Ambulatory Medical Care Survey (NHAMCS). At the time of the study, 2002 data were the latest release from NAMCS and NHAMCS. The National Center for Health Statistics (NCHS) provides complete descriptions of both surveys and yearly data at cdc.gov/nchs/about/major/ahcd/ahcd1.htm. These surveys, particularly NAMCS, have been validated against other data sources [14,15] and have also been utilized in past research examining depression treatment for adults [16]. In brief, NAMCS captured health care services provided by private office-based physicians, whereas NHAMCS captured services offered at hospital outpatient departments. The sampling universe for NAMCS was office-based, patient-care physicians in 15 specialty strata from the master files maintained by the American Medical Association and American Osteopathic Association. The sampling frame for NHAMCS included short-stay ( 30 days) hospitals or hospitals whose specialty was general (medical or surgical) or children s general. Both surveys utilized multistage probability sampling procedures, which enable researchers to generate nationally representative estimates. Between 1995 and 2002, annual participation rates among physicians selected for NAMCS averaged 68% (ranging from 63% in 1999 to 73% in 1995), resulting in the number of participating physicians being between 1087 and 1883 per annum. For NHAMCS, the participation rate of a fixed panel of 600 hospitals per year ranged from 94% in 1995 to 98% in Physicians and/or staff completed standard encounter forms for a systematic random sample of patient visits during randomly assigned reporting periods. Yearly encounter forms varied slightly between NAMCS and NHAMCS and both the NAMCS and NHAMCS forms were revised every two years. This study relied on variables common to NAMCS and NHAMCS over time, including patient demographics, visit characteristics, reasons for visit (up to three), diagnoses (up to three), new and continuing medications (up to six), and specific medical services (e.g., psychotherapy and mental health counseling) provided at the visit. Item nonresponse rates were mostly 5% or less in both surveys for all years.

3 436 J. Ma et al. / Journal of Adolescent Health 37 (2005) In addition to NAMCS and NHAMCS data, 2000 U.S. census data were used to calculate population-based prevalence of depression visits per 100 children and adolescents by age group, gender and race/ethnicity. Subjects Diagnosis of depression and depression treatment Depression-related diagnoses were categorized as major depressive disorder (International Classification of Disease [ICD-9] codes and 296.3), dysthymia (300.4), depressive adjustment reaction (309.0 and 309.1), and depression not otherwise specified (311.0). We focused our analysis on patient visits for children and adolescents between the ages of 7 and 17 years. The number of all patient visits by 7- to 17-year-olds was 6148 in 1995, 5530 in 1996, 5507 in 1997, 5682 in 1998, 4241 in 1999, 5497 in 2000, 6266 in 2001, and 6568 in Approximately two-thirds of the study sample that involved a depressive disorder had the disorder as the primary diagnosis for all the years. We did not differentiate patient visits with depression as the primary diagnosis from those with depression as a secondary diagnosis in analysis because the visits showed no differences on any key response variable. To assess trends in depression treatment over time, patterns of both pharmacological and psychotherapeutic treatments were examined. Antidepressant use was identified if corresponding NAMCS and NHAMCS drug codes were listed among the six possible new or continuing medications on the encounter form. Antidepressants were categorized as SSRIs, tricyclic antidepressants (TCAs), or other antidepressants (i.e., bupropion, trazodone, nefazodone, mirtazapine). Psychotherapeutic treatment was identified if psychotherapy or mental health counseling was reported during the patient visit via checkboxes on the encounter form, including psychotherapy/mental health counseling that was ordered by a physician but provided by a nonphysician (e.g. mental health professional). Neither NAMCS nor NHAMCS captured the form, content or frequency of psychotherapy and mental health counseling. Patient visit characteristics In addition to depression diagnosis, patient visit characteristics also included patient age, gender, race/ethnicity, medical insurance, visit status, U.S. census region, metropolitan area status, and physician specialty. Examined racial and ethnic categories included non-hispanic Caucasian, non-hispanic African American, and Hispanic. Sample size was too small ( 30 according to NCHS analytical guidelines) for other racial/ethnic categories. Medical insurance was classified as private/commercial insurance, public insurance (i.e., Medicare and Medicaid), and other insurance (e.g., workers compensation and self-pay). Visit status distinguished first-time visits from return visits. Physician specialty was available only from NAMCS, which contributed more than 90% of the total visits for each of the study years. We compared psychiatrists with physicians in other specialties. Measures Of primary interest were national trends in visit rates for depression and for pharmacotherapy, psychotherapy/mental health counseling, and combination therapy in the study population. The visit rate for pharmacotherapy was defined as the proportion of patient visits diagnosed with depression where antidepressants were prescribed, and the visit rate for psychotherapy/mental health counseling as the proportion where such services were noted. Combination therapy referred to the reporting of both therapies for a visit. Patient visits with a diagnosis of obsessive-compulsive disorders (ICD-9 code 300.3), accounting for 2% or less of patient visits that involved a depressive disorder, were excluded from the numerator and denominator for calculating visit rates for pharmacotherapy. We also analyzed the association of patient visit characteristics with the likelihood of receiving pharmacotherapy, psychotherapy/mental health counseling, and combination therapy in the study population. Analyses Statistical analyses were performed using SAS for Windows software (SAS Institute, Cary, North Carolina) and SAS-callable SUDAAN software (RTI, Research Triangle Park, North Carolina). Both NAMCS and NHAMCS provided sampling weights accounting for unequal selection probabilities and nonresponse. Data from 1995 through 2002 also contained masked sample design information that takes into account the complex sample designs (clustering and stratification) of the surveys when used with appropriate statistical tools. The unit of analysis in both surveys is the patient visit. NCHS analytical guidelines permit the combination of data from NAMCS and NHAMCS and across multiple years. We combined NAMCS and NHAMCS data to obtain a wider range of outpatient settings and a broader socioeconomic spectrum of patients seeking ambulatory care. In addition, to minimize random fluctuations between years, data were combined in two-year groupings for trend analysis using the SAS procedure SURVEYMEANS. This procedure generated national estimates of visit rates and corresponding 99% confidence intervals (CI). Per NCHS analytical guidelines, we reported 99% CIs due to the use of masked sample design variables. Chi-square tests were performed using PROC CROSSTAB in SUDAAN to examine the association of depression treatment by type (i.e., psychotherapy/mental health counseling, pharmacotherapy, and combination therapy) with each individual patient visit characteristic for combined NAMCS and NHAMCS data. Three multivariate logistic regression models were tested using PROC RLOGISTIC in SUDAAN to examine the independent effect of each patient visit characteristic on the probability of receiving depression treatment by

4 J. Ma et al. / Journal of Adolescent Health 37 (2005) Fig. 1. Trends in ambulatory care visits for depression in children and adolescents 7 to 17 years of age. Sources: NAMCS/NHAMCS, type after controlling for other characteristics. Each model assumed the probability of receiving pharmacotherapy, psychotherapy, or combination therapy as the dependent variable and patient visit characteristics as independent variables. In each model, the independent variables included depression diagnosis, patient age, gender, race/ethnicity, physician specialty, and any other patient visit characteristics that were significantly associated with the treatment type in the chisquare tests. The models produced adjusted odds ratios and 99% CIs. Results Volume and composition of depression visits The number of annual visits by children and adolescents aged 7 to 17 years who were diagnosed with a depressive disorder increased markedly from 1.44 million (M) (99% CI: M) in to 3.22 M (2.50 M 3.94 M) in (Figure 1). During this period, there were fewer diagnoses of major depressive disorder compared with other depression diagnoses. The number of visits with a diagnosis of depression not otherwise specified (suggesting moderate severity) increased by 138% from 693,084 (416, ,503) to 1.64 M (1.08 M 2.20 M). At the same time, the number of visits with a dysthymia diagnosis (lesser severity) almost tripled from 304,632 (171, ,702) to 851,663 (526, M), whereas the number of major depressive disorder diagnoses (greater severity) increased only modestly from 392,160 (239, ,248) to 649,969 (415, ,977). Over the eight-year study period, depression visits comprised % of total outpatient visits in the target population of 7 17-year-olds. As shown in Table 1, depression visits accounted for 1.8% of all visits by 7 17-year-olds in , including 2.9% of the total visits for year-olds, 2.0% for year-olds, and.6% for 7 10-year-olds. Differences in the proportion of depression visits by gender (2.0% for girls and 1.6% for boys) and race/ethnicity (1.9% for non- Hispanic Caucasians, 1.7% for Hispanics, and 1.1% for non- Hispanic African Americans) did not reach statistical significance. Population-based prevalence of depression visits in the study population averaged three visits per 100 persons, with higher prevalence (4 6 visits/100 persons) among teens, girls, and non-hispanic Caucasians. Depression treatment patterns The proportion of depression visits by children and adolescents receiving pharmacotherapy or psychotherapy/ mental health counseling was generally high, ranging from

5 438 J. Ma et al. / Journal of Adolescent Health 37 (2005) Table 1 Number and percentage of patient visits for 7 17-year-olds with depression in Sample data (No. of visits) National estimates (No. of visits in thousands) % of Total visits (99% CI) Annual number of visits per 100 persons a Overall ( ) 3.4 Age group (years) ( ) ( ) (.2 1.0).9 Gender Female ( ) 4.1 Male ( ) 2.8 Race/ethnicity Non-Hispanic white ( ) 3.9 Non-Hispanic African-American (.4 1.9) 1.4 Hispanic (.3 3.1) 2.9 a Population-based visit rates were based on the 2000 U.S. census data. a low of 80% (69 92%) in to a high of 95% (91 99%) in (Figure 2). The prevalence of psychotherapy/mental health counseling was above 80% from 1995 through 1998, but it dropped considerably to 54% (41 68%) in and stabilized at 68% (58 78%) in Provision of psychotherapy or counseling was associated with longer visit durations: 20.6 ( ) minutes vs ( ) minutes. The proportion of depression visits during which pharmacotherapy was prescribed increased modestly from 47% (34 59%) in to 52% (41 63%) in , with a peak in (56% [43 69%]). Lastly, the proportion of depression visits during which both pharmacotherapy and psychotherapy/mental health counseling were provided re- Fig. 2. Proportion of depression visits for children and adolescents during which psychotherapy/mental health counseling and/or pharmacotherapy treatment were provided. Source: NAMCS/NHAMCS,

6 J. Ma et al. / Journal of Adolescent Health 37 (2005) Fig. 3. Share of total antidepressants prescribed for children and adolescents with depression by SSRIs, TCAs, and other antidepressants. Source: NAMCS/NHAMCS mained relatively steady, ranging between 30% and 40% across the study years. Similar treatment patterns were noted for patient visits diagnosed with major depressive disorder (MDD) vs. those with other depression diagnoses, with the former being associated with generally but statistically insignificantly higher proportions of being treated. Selective serotonin reuptake inhibitors accounted for 76% (63 88%) in and 81% (70 91%) in of total antidepressants prescribed for depressed children and adolescents aged between 7 and 17 years (Figure 3). At the same time, TCA s share of total antidepressant use declined from 16% (6 26%) in to 2% (0 6%) in The use of bupropion, trazodone, nefazodone, and mirtazapine together peaked at 32% (14 50%) in (up from 11% [1 21%] in ) and declined to 22% (11 32%) by In , trazodone was the leading drug in the other antidepressant category, whereas by bupropion became the leading non-ssri drug. The number of patient visits during which an SSRI was prescribed reached 1.35 M (.90 M 1.80 M) in , reflecting a 2.6-fold increase compared with 510,342 (298, ,813) in Fluoxetine, sertraline, and paroxetine were the most commonly prescribed SSRIs in all study years (Figure 4). The usage increased for all three SSRIs over the study period. The number of patient visits during which fluoxetine was prescribed doubled from 207,914 (83, ,870) in to 415,580 (132, ,278) in ; sertraline increased by 62% from 214,105 (57, ,945) to 345,576 (160, ,490) visits; and paroxetine increased from 75,781 (0 158,938) to 279,275 (84, ,049) visits. The share of total SSRI use accounted for by these three medications declined from 98% to 77% as health care providers switched to newer SSRIs as they became available. Other SSRIs including fluvoxemine, citalopram and venlafaxine together were prescribed for 331,427 (70, ,879) patient visits in Correlates of depression treatment The results of our multivariate logistic regressions showed that patient visits diagnosed with depressive adjustment reaction were less likely to receive pharmacotherapy (odds ratio [OR].11; 99% CI:.01.90) than those with MDD. The likelihood of being treated pharmacologically or psychotherapeutically did not differ significantly for patient visits involving dysthymia or depression not otherwise specified relative to those having MDD. Compared with physicians in other specialties, psychiatrists were more likely to prescribe antidepressant medications alone (OR 3.42; 99% CI: ) or in combination with psychotherapy/mental health counseling (OR 5.87; 99% CI: ). Also, psychotherapy and mental health counseling were more common among girls than boys (OR 2.06; 99% CI: ) and less common in vs (OR.31; 99% CI:.11.88). The likelihood of receiving pharmacotherapy, psychotherapy/mental health coun-

7 440 J. Ma et al. / Journal of Adolescent Health 37 (2005) Fig. 4. Number of depression visits for children and adolescents treated with various SSRIs. Source: NAMCS/NHAMCS, seling, or combination therapy did not differ significantly by patient age, race/ethnicity, medical insurance, visit status, geographic region, metropolitan area status, or site of care. Discussion Our study supports other research that finds children and adolescents to be increasingly seen and diagnosed with depressive disorders [3,7,17,18]. In concordance with the underlying epidemiology [19], depression visits were most common among year-olds, girls, and non-hispanic Caucasians. The absolute increase in depression visits was much greater than expected when compared with the moderate increase in the population of children and adolescents during these years. The greater number of depression visits consisted primarily of an increase in the number of children and adolescents diagnosed with depression not otherwise specified. The depression not otherwise specified diagnosis is a broad category that may reflect more frequent reporting of patient visits for mild to moderate depressive symptoms, or a declining specificity in diagnostic practices relating to mood disorders. In concordance with previous research [3,17,20], our results suggest that a vast majority of children and adolescents seen for depression were treated psychotherapeutically or pharmacologically. The visit rates for psychotherapy/mental health counseling showed a declining trend from 83% of visits in to 54% in and 68% in , whereas the visit rates for pharmacotherapy rose from 47% in to 56% in and 52% in The mirroring pattern of these two types of treatment across time suggests that pharmacotherapy is used to substitute, rather than complement, psychotherapy/ mental health counseling. Even though the increase in the proportion of visits resulting in antidepressant medication prescriptions is small, against the backdrop of dramatic increases in visits by children and adolescents for depression, the absolute increase in pharmacotherapy has been substantial. We also found that between 42% and 52% of depression visits with pharmacotherapy did not also include psychotherapy or mental health counseling. Although the data sources used preclude us from assessing treatment appropriateness, the American Academy of Child and Adolescent Psychiatry recommends that psychotherapy or mental health counseling be used for all children and adolescents [6,8] whereas antidepressants, preferably in combination with psychotherapy/mental health counseling, be prescribed to those with severe, psychotherapy-resistant symptoms [21,22]. Recently, the Treatment for Adolescents With Depression Study (TADS) demonstrated that compared with monotherapy of fluoxetine or cognitive behavioral therapy (CBT), the combination of fluoxetine and CBT

8 J. Ma et al. / Journal of Adolescent Health 37 (2005) provides the most favorable benefit-risk balance for treating MDD in adolescents [22]. CBT is only one form of psychotherapeutic intervention. The effectiveness of other forms of psychotherapy/mental health counseling is less clear and should be studied. In light of the seriousness of depression and its incapacitating effects on young lives, appropriate prescribing of medications as part of a well-monitored psychopharmacological treatment regimen can tremendously benefit the individual patient as well as society as a whole. Concerns have been raised by other studies that show a sizable proportion of youngsters seen for depression receiving pharmacotherapy as the first choice of treatment even when the diagnosis is not major depression [7,17]. We found that SSRIs remained the leading class of antidepressants used to treat depressed children and adolescents through 2002 and that fluoxetine, sertraline and paroxetine were the most prescribed SSRIs. Fluoxetine had been prescribed frequently before it was shown to be effective for treating child and adolescent depression in 1997 and approved by the FDA for such use in Also, increased off-label use of other SSRIs has occurred despite the lack of evidence supporting their safety and efficacy in this population and the apparent superiority of evidence supporting the use of fluoxetine. The large numbers of depression visits during which SSRIs were prescribed in our study imply that many children and adolescents may have been facing unknown risks associated with off-label use of SSRIs. The epidemiological literature suggests an association between increased SSRI usage and declined suicide rates in adolescents. Yet, several recent studies have reported increased agitation, suicide ideation, and violent behaviors in youths taking SSRIs [23 25]. The FDA issued a directive in October 2004 requiring manufacturers of all antidepressants to print a black-box warning label [11]. Future studies are needed to determine how the prescribing of antidepressants change in response to these latest evidence and regulatory actions. The current knowledge supports that, when medication is indicated after a comprehensive evaluation and an accurate diagnosis, fluoxetine should be the initial treatment for child and adolescent depression in conjunction with psychotherapeutic interventions such as CBT. Only for refractory patients should other antidepressant medications, including other SSRIs, be considered. Close monitoring is necessary for all children and adolescents treated with antidepressant medication, especially early in the course of treatment and whenever a change takes place in the medication or dosage. Several data limitations of this study must be acknowledged. Due to their per-patient visit nature, NAMCS and NHAMCS may disproportionately represent more frequent users of health care, leading to an overestimation of visit rates for pharmacotherapy and psychotherapy/mental health counseling. On the other hand, underestimation is possible, particularly for psychotherapy/mental health counseling, because of the inability to account for multiple visits by the same patient and services provided during nonsampled visits or by nonparticipating professionals. Visit-based estimates may not be directly comparable to population-based national estimates. Although item nonresponse rates are generally low, nonresponse rates for ethnicity ranged from 21 27% in NAMCS and 14 24% in NHAMCS. In our analyses, any patient visits for which no ethnicity was indicated were categorized as non-hispanic Caucasian, which may not be accurate. Lastly, we could not assess diagnostic accuracy or treatment appropriateness due to the lack of detailed clinical information. In conclusion, youth in the United States are increasingly being diagnosed with and treated for depression. The observation of the increasingly prevalent off-label use of SSRIs, as well as possibly inappropriate use of medications in substitution of psychotherapy/mental health counseling as first-line therapy, raises concerns about physicians adherence to evidence-based medicine. The importance of proper treatment of child and adolescent depression warrants more clinical and community-based research and also intervention studies for effectively translation of evidence into practice. Acknowledgment This study was supported by a research grant from Agency for Healthcare Research and Quality (AHRQ) (R01- HS11313). References [1] Vitiello B, Swedo S. Antidepressant medications in children. N Engl J Med 2004;350(15): [2] Kessler R, Avenevoli S, Ries Merikangas K. Mood disorders in children and adolescents: an epidemiological perspective. Biol Psychiatry 2001;49: [3] Olfson M, Gameroff M, Marcus S, Waslick BD. Outpatient treatment of child and adolescent depression in the United States. Arch Gen Psychiatry 2003;60: [4] Brent DA. Antidepressants and pediatric depression the risk of doing nothing. N Engl J Med 2004;351(16): [5] March J, Silva S, Petrycki S, et al. Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression: Treatment for Adolescents With Depression Study (TADS) randomized controlled trial. JAMA 2004;292(7): [6] Birmaher B, Brent D, Benson R. Practice parameters for the assessment and treatment of children and adolescents with depressive disorders. J Am Acad Child Adolesc Psychiatry 1998;37:63S 83S (10 suppl). [7] Zito J, Safer D, dosreis S, et al. Psychotropic practice patterns for youth: a 10-year perspective. Arch Pediatr Adolesc Med 2003;157(1): [8] Renaud J, Axelson D, Birmaher B. A risk-benefit assessment of pharmacotherapies for clinical depression in children and adolescents. Drug Safety 1999;20(1): [9] Emslie G, Weinberg A, Kowatch R, et al. A double-blind, randomized, placebo-controlled trial of fluoxetine in children and adolescents with depression. Arch Gen Psychiatry 1997;54(11): [10] US Food and Drug Administration. FDA Statement Regarding Anti- Depressant Paxil for Pediatric Population. June 19, 2003 [cited 2004

9 442 J. Ma et al. / Journal of Adolescent Health 37 (2005) Aug 21]. Available from: /ans01230.html [11] US Food and Drug Administration. Suicidality in Children and Adolescents Being Treated With Antidepressant Medications. October 15, 2004 [cited 2005 Jan 14]. Available from: cder/drug/antidepressants/ssripha htm [12] Goodwin R, Gould MS, Blanco C, et al. Prescription of psychotropic medications to youths in office-based practice. Psychiatr Serv 2001; 52(8): [13] Murray ML, de Vries CS, Wong IC. A drug utilisation study of antidepressants in children and adolescents using the General Practice Research Database. Arch Dis Child 2004;89(12): [14] Zell ER, McCaig LF, Kupronis BA, et al. A comparison of the National Disease and Therapeutic Index and the National Ambulatory Medical Care Survey to evaluate antibiotic usage. In: Proceedings of the Section on Survey Research Methods American Statistical Association; Alexandria, VA: 2000: [15] Gilchrist VJ, Stange KC, Flocke SA, et al. A comparison of the National Ambulatory Medical Care Survey (NAMCS) measurement approach with direct observation of outpatient visits. Med Care 2004; 42(3): [16] Stafford RS, Ausiello JC, Misra B, et al. National Patterns of Depression Treatment in Primary Care. Prim Care Companion J Clin Psychiatry 2000;2(6): [17] Olfson M, Marcus S, Weissman M, et al. National trends in the use of psychotropic medications by children. J Am Acad Child Adolesc Psychiatry 2002;41(5): [18] Olfson M, Marcus S, Pincus H, et al. Antidepressant prescribing practices of outpatient psychiatrists. Arch Gen Psychiatry 1998;55(4): [19] Kessler R, Walters E. Epidemiology of DSM-III major depression and minor depression among adolescents and young adults in the national comordibity survey. Depress Anxiety 1998;7:3 14. [20] Lewinsohn P, Rohde R, Seeley J. Treatment of adolescent depression: frequency of services and impact on functioning in young adulthood. Depress Anxiety 1998;7: [21] Keller M, McCullough J, Klein D, et al. A comparison of nefazodone, the cognitive behavioral-analysis system of psychotherapy and their combination for the treatment of chronic depression. N Engl J Med 2000;342: [22] March J, Silva S, Petrycki S, et al. Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression: Treatment for Adolescents with Depression Study (TADS) randomized controlled trial. JAMA 2004;18(7): [23] US Food and Drug Administration. Reports of Suicidality in Pediatric Patients Being Treated with Antidepressant Medications for Major Depressive Disorder (MDD). October 27, 2003 [cited 2004 Aug 29]. Available at: html [24] Jick H, Kaye J, Jick S. Antidepressants and the risk of suicidal behavior. JAMA 2004;292(3): [25] Vorstman J, Lahuis B, Buitelaar J. SSRIs associated with behavioral activation and suicidal ideation. J Am Acad Child Adolesc Psychiatry 2001;40(12):