ARTICLE. Metabolic Screening in Children Receiving Antipsychotic Drug Treatment

Transcription

1 ARTICLE Metabolic Screening in Children Receiving Antipsychotic Drug Treatment Elaine H. Morrato, DrPH, MPH; Ginger E. Nicol, MD; David Maahs, MD; Benjamin G. Druss, MD, MPH; Daniel M. Hartung, PharmD, MPH; Robert J. Valuck, PhD; Elizabeth Campagna, MS; John W. Newcomer, MD Objectives: To estimate metabolic screening rates, predictors of screening, and incidence of metabolic disturbances in children initiating second-generation antipsychotic (SGA) drug treatment. Design: A retrospective, new-user cohort study (between July 1, 2004, and June 30, 2006) using Medicaid claims data. Settings: California, Missouri, and Oregon. Patients: A total of 5370 children (aged 6-17 years) without diabetes mellitus taking SGA drugs and children without diabetes taking albuterol (control) but no SGA drugs. Intervention: Findings 1 year after recommendations from the American Diabetes Association and American Psychiatric Association called for metabolic screening of patients receiving SGA drugs. Outcome Measures: Serum glucose and lipid testing, 6-month incidence of diabetes, and dyslipidemia disturbances. Results: Glucose screening was performed in 1699 (31.6% [95% confidence interval (CI), 30.4%-32.9%]) SGA-treated children vs 1891 (12.6% [12.1%-13.2%]) control individuals. Lipid testing was performed in 720 (13.4% [95% CI, 12.5%-14.4%]) SGA-treated children vs 458 (3.1% [2.8%-3.3%]) controls. In multivariate logistic regression analysis, children with serious and/or multiple psychiatric diagnoses and those who used health care services more intensively were more likely to receive metabolic screening. The case incidence of glucose and lipid disorders was higher in SGA-treated vs albuterol-treated children (8.9 per 1000 children [95% CI, 6.6%-11.8%] vs 4.9 per 1000 children [3.9%- 6.2%]; and 9.7 per 1000 children [95% CI, 7.2%- 12.7%] vs 4.6 per 1000 children [95% CI, 3.6%-5.8%], respectively). Conclusion: Most children starting treatment with SGA medications in this public sector sample did not receive recommended glucose and lipid screening. Arch Pediatr Adolesc Med. 2010;164(4): Author Affiliations are listed at the end of this article. ANTIPSYCHOTIC DRUGS ARE approved for the treatment of schizophrenia and other neuropsychiatric disorders, 1,2 including indications for the treatment of bipolar disorder, 3-7 adjunctive therapy for the treatment of depression, 3 and treatment of children and adolescents. 8 Children with mental disorders have been increasingly prescribed antipsychotic medication, 9-13 primarily second-generation (atypical) antipsychotics (SGAs). One multistate Medicaid study 14 found that one-quarter of new prescriptions for antipsychotics were filled for children. In a national study, 10 1% to 2% of all outpatient visits by children to office-based physicians included a prescription for an antipsychotic medication. During the past decade, use of antipsychotic medications has increased an estimated 5-fold in US children 9,10 and is mainly being used for behavioral control (in autistic and nonautistic children). 10 Although there has been a parallel increase in antipsychotic prescription rates in Denmark, 15 Germany, 16 the United Kingdom, 17 the Netherlands, 18 Italy, 19 and France, 20 the rates of prescription of antipsychotics in these countries have remained consistently lower than in the United States. 9,10,16 In 2003, the Food and Drug Administration required that a class warning be added to product labeling for all SGAs regarding the risk of hyperglycemia and diabetes mellitus, including the potential for extreme hyperglycemia associated with ketoacidosis, hyperosmolar coma, or death. 21 The warning further stated that glucose levels should be monitored in patients with an established diagnosis of diabetes, risk factors for diabetes, or symptoms of hyperglycemia. Since then, greater specificity has been added to some SGA product labeling regarding adverse effects on weight gain and lipid profiles. 5,22,23 In addition, the American Diabetes Associa- 344

2 tion, the American Psychiatric Association, the North American Association for the Study of Obesity, and the American Association of Clinical Endocrinologists issued a joint consensus statement in early recommending metabolic screening and monitoring, consisting of weight and body mass index (calculated as weight in kilograms divided by height in meters squared), waist circumference, blood pressure, and fasting serum glucose and lipid profiles for all patients receiving these agents, regardless of age. The evidence base supporting metabolic testing for SGA-treated patients consisted of a synthesis of the literature and presentations from 14 experts drawn from the areas of psychiatry, obesity, and diabetes and from the Food and Drug Administration and representatives from the pharmaceutical industry. 24 Relatively little information exists regarding the shortterm and long-term metabolic effects of antipsychotic use in children, 25 although initial observations indicate that metabolic adverse effects of antipsychotic treatment in pediatric populations are readily detected Compared with adults, children who take olanzapine, risperidone, and quetiapine fumarate are particularly at risk for weight gain Type 2 diabetes and cardiovascular adverse events have also been identified in children and adolescents exposed to antipsychotics, especially when multiple antipsychotics are prescribed or antipsychotics are coprescribed with mood stabilizers and antidepressants. 27 Moreover, the association between antipsychotic medication use and diabetes has been stronger in children and adolescents than adults. 26 However, secondary analyses in some recent studies 14,32-34 suggest that children are less likely to receive metabolic screening and monitoring compared with adults. The present study aimed to estimate populationbased rates of serum glucose and lipid testing in children from 3 state Medicaid programs who initiated SGA drug treatment compared with a pediatric reference population after the availability of Food and Drug Administration warnings and the American Diabetes Association, American Psychiatric Association, North American Association for the Study of Obesity, and American Association of Clinical Endocrinologists joint recommendations. 24 Using this sample, we also sought to identify demographic and clinical predictors of glucose and lipid testing in SGA-treated children and to estimate the incidence of early detection of metabolic disorders, with the goal of informing strategies for improving metabolic monitoring of this at-risk population. METHODS STUDY POPULATION A new-user cohort was defined as Medicaid fee-for-service clients (aged 6-17 years) enrolled in California, Missouri, and Oregon who had started taking antipsychotic medication in Children were included if they had a prescription claim for 1 of 5 commonly prescribed SGA drugs (aripiprazole, olanzapine, quetiapine, risperidone, or ziprasidone hydrochloride) and continuous enrollment eligibility 180 days before and after the date of the first (index) antipsychotic pharmacy claim. Use of clozapine in children was so infrequent that it was not included in this study. Study participants had a unique encrypted identifier from which to identify their complete medical, pharmacy, and laboratory claims for the 180 days before and after the index date (representing an observation range of July 1, 2004, through June 30, 2006). Children were not included in the study if they were Medicare dual eligible or in a managed care plan because complete laboratory and medical claims were not available. Because our aim was to estimate rates of glucose and lipid screening rather than monitoring related to existing disease, children with preexisting diabetes or dyslipidemia (n=100), defined by a diagnosis in the medical claims or a pharmacy claim for an oral antidiabetic, insulin, or antidyslipidemic drug in the 180 days before antipsychotic drug therapy initiation, were not included in the study. To provide a relative comparison, we compared absolute rates in our sample of SGA-treated children with a reference population of chronically ill children for whom metabolic screening is not clinically indicated. A comparison cohort of children initiating albuterol, but not receiving antipsychotic medication, was identified (N=15 000). Albuterol is a commonly prescribed medication for children with asthma and has been used to identify a control group for other claims-based research in psychiatry 14,35 because it selects for individuals across a broad age range typically receiving health care in a nonpsychiatric setting. We hypothesized that rates of screening in SGAtreated children would be higher than in an age-matched comparison group of albuterol-treated children receiving care during the same period. MAIN OUTCOME MEASURES Glucose testing was defined as a medical claim with an American Medical Association Current Procedural Terminology (CPT) code for a metabolic or general health panel (80048, 80050, or 80053) or glucose-specific serum test (82947, 82948, 82950, or 82951) occurring 30 days before through 180 days after the index date. 14 LipidtestingwasdefinedasaclaimwithaCPTcodeforalipidpanel (80061) or lipid-specific serum test (82465, 83700, 83701, 83715, 83716, 83721, or 84478), using the same time frame. Diagnoses were recorded in the medical claims according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). 36 Incident glucose disturbances were identified when a diagnosis in the medical claims or a prescription claim for an oral antidiabetic drug or insulin was present in the 180 days after the index date. Incident lipid disturbances were identified, using the Clinical Classifications Software (CCS) classification scheme developed by the Agency for Healthcare Research and Quality, 37 when a prescription claim for an antidyslipidemic drug or a diagnosis for disorders of lipid metabolism (CCS code 53) was present in the 180 days after the index date. Studies 38,39 that examine the validity of using claims data to identify patients with chronic disease suggest our algorithm would yield specificity levels greater than DEMOGRAPHIC AND CLINICAL CHARACTERISTICS OF PATIENTS Age (at the index date), sex, state of residence, and race/ ethnicity (white, black, and other [including Hispanic and Asian]) were available from the Medicaid beneficiary records from the state in question for each child. Physiologic and developmental differences associated with puberty were accounted for by categorizing the study population into 2 age groups (6-12 and years). Patients with serious mental disorders have been a particular focus for cardiometabolic screening and intervention. 40,41 Se- 345

3 rious mental illness was defined as having schizophrenia (CCS code 70), other psychoses (CCS code 71), or affective disorders (CCS code 69), which include bipolar disorder and major depression. Patients with medical claims associated with 2 or more diagnosed mental disorders (ICD-9-CM codes ) were classified as having a comorbid mental disorder. 10 Coprescribed psychotropic medications during the study period were identified from the pharmacy claims records and categorized as stimulants, antidepressants, mood stabilizers, and anxiolytics and hypnotics. 10 Persistent antipsychotic users, defined as a maximum gap in therapy no greater than 30 days during the first 180 days of therapy, 42 were identified to evaluate whether they had higher rates of metabolic monitoring because they would theoretically be at greater risk for developing metabolic adverse effects than patients receiving timelimited therapy. As an indicator of the degree to which each child accessed general medical care, the number of medical claims for office visits (CPT codes , , or ) occurring in the 180 days after drug initiation were determined and categorized as none, 1, or 2 or more. Children with a medical claim for an emergency department visit or a hospitalization were also identified because this may indicate greater use of medical services, greater severity of disorder, and increased access to general laboratory testing. STATISTICAL ANALYSIS The distribution of patient demographic and clinical characteristics and health care use patterns were stratified by age group, and proportions were compared between SGA-treated and albuterol-treated children using 2 tests. The percentage of SGAtreated children who received glucose and lipid testing was stratified by age group and compared with the albuterol-treated comparison group using 2 tests. Multivariate logistic regression was used to identify predictors of the likelihood of receiving testing among SGA-treated children after adjusting for age group, ethnicity, state, presence of serious mental illness and mental disorder comorbidity, persistent antipsychotic use, use of other psychotropic medications, number of medical office visits, emergency department use, and hospitalization. Glucose testing and lipid testing were modeled independently. Incidence rates of glucose and lipid disturbances per 1000 treated children were calculated by age group and compared between the SGA-treated and albuterol-treated comparison groups using 2 tests. All analyses were performed with SAS statistical software, version 9.2 (SAS Institute Inc, Cary, North Carolina). The study received approval from the Colorado Multiple Institutional Review Board and the California Committee for the Protection of Human Subjects. RESULTS CHARACTERISTICS OF SGA-TREATED CHILDREN Among the 5370 children initiating SGA therapy, the most commonly prescribed antipsychotic was risperidone, which was prescribed to 2714 children and adolescents (50.5%). New prescriptions to children aged 6 to 12 years totalled 1729 (60.7%), and new prescriptions to adolescents aged 13 to 17 years totalled 985 (39.1%), followed by quetiapine, aripiprazole, olanzapine, and ziprasidone. Median days of index drug supplied for all children initiating antipsychotic therapy was 30 days (interquartile range, days). Persistent use at 84 days was 62.7%, which decreased to 37.8% at 180 days. Persistence rates varied by 3.3% for each index antipsychotic drug. Children treated with SGAs were more often male, white, and hospitalized during the study period than the age-matched group of albuterol-treated children (Table 1). A total of 38.9% of children aged 6 to 12 years and 58.6% of adolescents aged 13 to 17 years who were initiating antipsychotic drug treatment had a recorded diagnosis for a serious mental illness. As indicated in Table 1, other psychotropic drug use was 4 to 6 times more prevalent among antipsychotic-treated children. RATES OF SERUM GLUCOSE AND LIPID TESTING Glucose screening was performed in 1699 (31.6% [95% confidence interval (CI), 30.4%-32.9%]) SGA-treated children vs 1891 (12.6 [12.1%-13.2%] control individuals [P.001]). Lipid testing was performed in 720 (13.4% [95% CI, 12.5%-14.4%]) SGA-treated children vs 458 (3.1% [2.8%-3.3%]) controls. Glucose and lipid testing rates for the antipsychotic and albuterol cohorts, stratified by age group, are presented in the Figure. Glucose screening was 2 to 3 times more common than lipid screening in both age groups, and screening was more common in older children. PREDICTORS OF GLUCOSE AND LIPID TESTING Table 2 presents the results from the multivariate logistic regression analysis of patient characteristics associated with the likelihood of receiving glucose and lipid screening among SGA-treated children. The strongest determinants of glucose testing were having an emergency department visit (adjusted odds ratio [AOR], 2.52), having multiple mental health comorbidities (2.42), and having been hospitalized (2.15). The strongest determinants of lipid testing were having multiple mental health comorbidities (2.85), having 2 or more medical office visits (1.88), and having serious mental illness (1.51). Children receiving antipsychotic treatment in Missouri and Oregon were less likely to receive glucose testing (0.62 and 0.81, respectively) and lipid testing (0.29 and 0.57, respectively) than those in California. The adjusted likelihood of metabolic testing was not associated with the index antipsychotic. INCIDENCE OF DETECTED GLUCOSE AND LIPID DISTURBANCES The incidence of glucose disturbances during the first 6 months of treatment were more common in antipsychotictreated compared with albuterol-treated children (0.9% vs 0.5%, P=.001) as were lipid disturbances (1.0% vs 0.5%, P.001). The incidence of metabolic disturbances was not statistically different between children prescribed olanzapine, risperidone, or quetiapine (glucose, 0.8%; lipids, 1.2%) compared with children prescribed aripiprazole or ziprasidone (glucose, 0.9%; lipids, 1.3%). Table 3 presents the incidence of glucose and lipid disturbances and source of ascertainment stratified by the child s age. 346

4 Table 1. Select Patient Characteristics a Patient Characteristic Aged 6 to 12 Years, No. (%) Aged 13 to 17 Years, No. (%) Antipsychotic Cohort (n=2850) Albuterol Cohort (n=9000) Antipsychotic Cohort (n=2520) Albuterol Cohort (n=6000) Male sex 2000 (70.2) 4802 (53.4) 1430 (56.7) 2865 (47.8) Race/ethnicity White 1883 (66.1) 4707 (52.3) 1665 (66.1) 3450 (57.5) Black 330 (11.6) 792 (8.8) 330 (13.1) 649 (10.8) Other 625 (21.9) 3494 (38.8) 518 (20.6) 1893 (31.6) Unknown 12 (0.4) 7 (0.1) 7 (0.3) 8 (0.1) State California 1739 (61.0) 6021 (66.9) 1604 (63.7) 4108 (68.5) Missouri 814 (28.6) 2558 (28.4) 654 (26.0) 1562 (26.0) Oregon 297 (10.4) 421 (4.7) 262 (10.4) 330 (5.5) Mental disorder diagnosis Serious mental illness b 1108 (38.9) 294 (3.3) 1476 (58.6) 502 (8.4) Mental disorder comorbidity c 2136 (74.9) 1030 (11.4) 1973 (78.3) 1075 (17.9) Other psychotropic drug use Any use 1672 (58.7) 841 (9.3) 1466 (58.2) 754 (12.6) Stimulants 1090 (38.2) 564 (6.3) 428 (17.0) 270 (4.5) Antidepressants 484 (17.0) 132 (1.5) 807 (32.0) 323 (5.4) Mood stabilizers 404 (14.2) 183 (2.0) 535 (21.2) 199 (3.3) Anxiolytics and hypnotics 70 (2.5) 83 (0.9) 160 (6.3) 137 (2.3) Health care use 180 days after index date Frequency of medical office visits None 1477 (51.8) 5020 (55.8) 1247 (49.5) 3340 (55.7) (18.0) 1617 (18.0) 466 (18.5) 1046 (17.4) (30.1) 2363 (26.3) 807 (32.0) 1614 (26.9) Emergency department visit 496 (17.4) 1409 (15.7) 688 (27.3) 1275 (21.3) Hospitalization 182 (6.4) 157 (1.7) 294 (11.7) 203 (3.4) a Data from California, Missouri, and Oregon state Medicaid programs of children initiating drug treatment in Antipsychotic- and albuterol-treated patients had no recognized diabetes or dyslipidemia at the start of therapy. b Serious mental illness was defined as having schizophrenia, affective disorder, or other psychoses. c Patients with medical claims associated with 2 or more diagnosed mental disorders (International Classification of Diseases, Ninth Revision, Clinical Modification codes ) were classified as having mental disorder comorbidity. Glucose and lipid disturbances were more common in older than younger children. The overall difference in observed rates of metabolic disturbances between the antipsychotic-treated and albuterol-treated children appears to be primarily attributable to differences in rates of new diagnoses rather than rates of new antidiabetic and antidyslipidemic drug prescriptions. COMMENT Received Metabolic Screening, % Glucose Lipids Patients Aged 6 to 12 y Atypical antipsychotic cohort Albuterol (control) cohort Glucose Lipids Patients Aged 13 to 17 y Figure. Serum glucose and lipid screening among children receiving second-generation antipsychotic and albuterol drug treatment. Mean estimates and 95% confidence intervals (error bars) are presented. Data are from California, Missouri, and Oregon state Medicaid programs of children initiating drug treatment in Antipsychotic-treated (n=5370) and albuterol-treated (n=15 000) patients had no recognized diabetes mellitus or dyslipidemia at the start of therapy. The results of this multistate study of publicly insured children indicate that most children prescribed antipsychotic medication did not receive metabolic screening. Screening rates were approximately one-quarter to one-half lower than rates reported for adults. 32,33 To our knowledge, this is the first study to compare rates of metabolic screening in antipsychotic-treated children with screening rates in an age-matched comparison pediatric population. As hypothesized, rates of glucose and lipid testing were significantly higher among antipsychotic-treated children; however, those rates fell short of consensus screening recommendations and they point to a substantial missed opportunity for metabolic screening in the vulnerable population of antipsychotic-treated children. This finding is disturbing, given the general environment of public health concern surrounding increasing rates of obesity and type 2 diabetes in adolescents and given specific concerns about metabolic risk for children with psychiatric conditions. Recent approvals of pediatric indications for 2 additional SGA drugs, 43,44 with approval of a third drug still pending, 45 sug- 347

5 Table 2. Factors Predicting Glucose and Lipid Testing Among Children Receiving SGA Drug Treatment a AOR of Testing (95% CI) Patient Characteristic Glucose Lipid Age group, y [Reference] 1 [Reference] ( ) 1.25 ( ) Race/ethnicity White NA 1 [Reference] Black NA 0.71 ( ) Other NA 0.86 ( ) State California 1 [Reference] 1 [Reference] Missouri 0.62 ( ) 0.29 ( ) Oregon 0.81 ( ) 0.57 ( ) Mental disorder diagnosis Serious mental illness b 1.68 ( ) 1.51 ( ) Mental disorder comorbidity c 2.42 ( ) 2.85 ( ) Drug use Persistent SGA medication use 1.23 ( ) 1.43 ( ) Other psychotropic medication use 1.23 ( ) NA Frequency of medical office visits None 1 [Reference] 1 [Reference] ( ) 1.66 ( ) ( ) 1.88 ( ) Emergency department visit 2.52 ( ) 1.23 ( ) Hospitalization 2.15 ( ) NA Abbreviations: AOR, adjusted odds ratio; CI, confidence interval; NA, not applicable (factor excluded from multivariate model because of statistical insignificance [P.10] at the bivariate level); SGA, second-generation antipsychotic. a Data from California, Missouri, and Oregon state Medicaid programs of children initiating drug treatment in Antipsychotic-treated (n=5370) patients had no recognized diabetes or dyslipidemia at the start of therapy. Glucose and lipid models are adjusted for shown factors. b Serious mental illness was defined as having schizophrenia, affective disorder, or other psychoses. c Patients with medical claims associated with 2 or more diagnosed mental disorders (International Classification of Diseases, Ninth Revision, Clinical Modification codes ) were classified as having mental disorder comorbidity. Table 3. Six-Month Incidence of Metabolic Disorders Among Children Receiving Second-Generation Antipsychotic and Albuterol Drug Treatment a Aged 6 to 12 Years, % (95% CI) Aged 13 to 17 Years, % (95% CI) Metabolic Disturbance Antipsychotic Cohort (n=2850) Albuterol Cohort (n=9000) Antipsychotic Cohort (n=2520) Albuterol Cohort (n=6000) Glucose Diabetes diagnosis 18 (6.3 [ ]) 38 (2.4 [ ]) b 21 (8.3 [ ]) 28 (4.7 [ ]) c Antidiabetic medication use 4 (1.4 [ ]) 16 (1.8 [ ]) 7 (2.8 [ ]) 22 (3.7 [ ]) Either diagnosis or medication use 20 (7.0 [ ]) 33 (3.7 [ ]) c 28 (11.1 [ ]) 41 (6.8 [ ]) c Lipids Dyslipidemia diagnosis 19 (6.7 [ ]) 28 (3.1 [ ]) c 32 (12.7 [ ]) 34 (5.7 [ ]) c Antidyslipidemic medication use 0 4 (0.4 [ ]) 2 (0.8 [ ]) 4 (0.7 [ ]) Either diagnosis or medication use 19 (6.7 [ ]) 31 (3.4 [ ]) c 33 (13.1 [ ]) 38 (6.3 [ ]) b Abbreviation: CI, confidence interval. a Data from California, Missouri, and Oregon state Medicaid programs of children initiating drug treatment in Antipsychotic and albuterol users had no recognized diabetes or dyslipidemia at the start of therapy. Metabolic disturbances were ascertained for 6 months after the index start date. b P.01 for the comparison between the antipsychotic and albuterol cohorts. c P.05 for the comparison between the antipsychotic cohort and albuterol controls. gest that the number of children receiving antipsychotic medication, and therefore requiring metabolic monitoring, will increase. The current results further indicate that children with serious and/or multiple psychiatric diagnoses and those who use health care services more intensively (eg, multiple medical office visits, emergency department use, and hospitalization) are the most likely to receive metabolic screening. This finding is consistent with determinants of metabolic testing in adults, 46,47 which indicate that patients with greater severity of mental illness, higher levels of medical and psychiatric comorbidity, and higher use of health care services are more likely to receive metabolic testing. However, even among children with mental disorder comorbidity, who were persistent in taking their antipsychotic medication and who had 2 or more medical office visits after drug initiation, only 45.2% received glucose testing and only 22.1% received lipid test- 348

6 ing. One patient-level factor that notably did not influence screening rates in this study was race/ethnicity, despite certain racial/ethnic groups clearly being at higher risk for developing type 2 diabetes 48 and evidence that race/ethnicity can influence risk of antipsychoticinduced metabolic adverse events. 49 More work is needed to confirm this finding because race/ethnicity measures in claims data can be imperfect. In addition to these patient-level factors, the observation of higher testing rates for glucose and lipids in children in the California Medicaid systems (vs in Missouri and Oregon) suggests that system-level or system-wide physician-level factors may also influence screening rates. More research is needed to discern the root causes of the low metabolic screening rates observed in our study population; however, it may be useful to consider several potential obstacles to metabolic monitoring in children. First, pediatricians and primary care physicians may remain unconvinced of the need to monitor children receiving antipsychotic medication. Instead, some physicians may have decreased antipsychotic prescribing and/or switched to lower-risk agents in response to the metabolic warnings, as has been shown in some studies. 13,14 Alternatively, the low rates of monitoring found in this study and others may reflect adoption of targeted screening for those with clinical changes (eg, those with significant weight gain) rather than mass screening. This hypothesis should be evaluated in future studies because our claims data do not include weight or body mass index measures. Furthermore, many physicians might order metabolic testing, but many families may be noncompliant with follow-through because of the child s refusal to have blood drawn or the added hassle factor of fasting and traveling to another health care facility for testing. Finally, challenges to interpreting metabolic measures in children and disagreement in general about what constitutes a metabolic abnormality requiring intervention 50 could dissuade physicians from collecting laboratory data altogether. We also found evidence that the development of glucose and lipid disturbances among antipsychotictreated children during the first 180 days after drug therapy initiation was more common than in our pediatric reference group. The findings are consistent with results from controlled trials and epidemiology studies 31 that demonstrated early onset and rapid progression of metabolic risk factors after antipsychotic treatment in children. The fact that only a few children were screened suggests that actual rates of metabolic abnormalities in these settings may be higher and that claimsbased estimates available to state policy makers may underestimate true rates of these problems. A recent Food and Drug Administration Psychopharmacologic Drugs Advisory Committee evaluating antipsychotic use in children strongly recommended that registries, or some other means, be established toward the goal of better safety surveillance in children. 45 Although psychiatrists have expertise in prescribing antipsychotic medications, increased attention to metabolic screening by all health care professionals is required. Pediatricians and other primary care physicians are uniquely positioned to identify early signs of cardiometabolic dysfunction related to antipsychotic medications. Increasing numbers of primary care physicians are treating children with psychotropic medications; some have reported nearly 85% of all prescriptions for psychotropic medications are provided by pediatricians and primary care physicians 51 and as many as one-third of prescriptions for antipsychotic medications are provided by nonpsychiatrists. 52 Primary care physicians are likely to encounter such children in their practices because data regarding effectiveness for treatment of major childhoodonset mental illnesses continue to accumulate and antipsychotic drug use increases. Several limitations of this study deserve mention. For the purposes of this analysis, testing refers to laboratory testing alone and does not encompass other clinical measures of risk commonly performed in the primary care setting, such as weight, height, body mass index percentile, and family history of diabetes and cardiovascular disease. Furthermore, this study assessed only whether laboratory tests were ordered and obtained, not what the results were or what the clinical response was in terms of adjusting treatment plans to mitigate medication-associated risk. Furthermore, the results of testing performed more than 30 days before the index prescription or more than 180 days after prescription were not evaluated. However, monitoring recommendations call for baseline evaluation, which should fall within the period studied. Caution should also be exercised when interpreting these results because they may not be generalizable to other populations or reimbursement settings, even though they represent a large sample from 3 states. Theoretically, metabolic screening rates may have improved since the time of data collection in response to either the prevalence of high body mass index among children and adolescents in the United States 53 or specific communication about the metabolic risks associated with antipsychotic medication. However, little evidence is available with regard to substantial improvements over time in available Medicaid or managed care data sets. 14,33,54 In conclusion, children who are treated with antipsychotic medications are recognized as a higher-risk group for metabolic disorders because of potential adverse metabolic effects of these medications and other predisposing factors. Results from this study suggest that serum glucose and lipid monitoring, recommended tools for assessing metabolic risk, are greatly underused in pediatric populations receiving antipsychotic drug treatment. Despite the result that use of glucose and lipid screening was higher in the antipsychotic-treated group than in an age-matched reference pediatric population, screening rates are markedly low, are lower than rates reported in antipsychotic-treated adults, and are well below the recommended goal that all patients 24 taking antipsychotics be monitored. Given evidence of shortterm adverse metabolic changes and the potential longterm metabolic risk burden, greater effort is needed to ensure consistent screening for all children receiving antipsychotic medications. Accepted for Publication: November 12, Author Affiliations: Department of Health Systems, Management, and Policy, Colorado School of Public Health 349

7 (Dr Morrato), Department of Pediatrics, School of Medicine (Drs Morrato and Maahs), Department of Clinical Pharmacy, School of Pharmacy (Drs Morrato and Valuck), University of Colorado at Denver, and Children s Outcome Research Program, The Children s Hospital (Dr Morrato and Ms Campagna), Denver, and Barbara Davis Center for Childhood Diabetes (Dr Maahs), Aurora, Colorado; Departments of Psychiatry and Medicine and the Center for Clinical Studies, School of Medicine, and Department of Psychology, Washington University, St Louis, Missouri (Drs Nicol and Newcomer); Department of Health Policy and Management, Rollins School of Public Health, Emory University, Atlanta, Georgia (Dr Druss); and Department of Pharmacy Practice, College of Pharmacy, Oregon Health & Science University Campus, Oregon State University, Portland (Dr Hartung). Correspondence: Elaine H. Morrato, DrPH, MPH, University of Colorado at Denver, Mailstop B119, Bldg 500, E 19th Pl, Aurora, CO Author Contributions: Drs Morrato and Druss had full access to all data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Morrato, Druss, Valuck, Campagna, and Newcomer. Acquisition of data: Hartung and Campagna. Analysis and interpretation of data: Morrato, Nicol, Maahs, Druss, Hartung, Valuck, Campagna, and Newcomer. Drafting of the manuscript: Morrato, Nicol, Campagna, and Newcomer. Critical revision of the manuscript for important intellectual content: Morrato, Nicol, Maahs, Druss, Hartung, Valuck, Campagna, and Newcomer. Statistical analysis: Valuck, Campagna, and Newcomer. Obtained funding: Morrato. Administrative, technical, and material support: Morrato, Nicol, and Hartung. Study supervision: Morrato and Newcomer. Financial Disclosure: Dr Morrato has received past research funding from the National Institutes of Health, Pfizer Inc, and Eli Lilly and Company and is a consultant to the Food and Drug Administration. Dr Nicol has received research funding from the National Alliance for Research on Schizophrenia and Depression. Dr Maahs has received research funding from Merck/Schering- Plough for an investigator clinical trial of lipid-lowering medications and is supported by grant K23 DK Dr Druss has received grant funding from the National Institute of Mental Health and the Agency for Healthcare Research and Quality and has consulted with Pfizer Inc. Dr Valuck has received research grant support from the Agency for Healthcare Research and Quality, the American Foundation for Suicide Prevention, Eli Lilly and Company, Pfizer Inc, and Ortho-McNeil-Janssen Pharmaceutical Ltd and has served as a consultant to Eli Lilly and Company, Forest Laboratories, and H. Lundbeck A/S. Dr Newcomer has received research grant support from the National Institute of Mental Health, the National Alliance for Research on Schizophrenia and Depression, the Sidney R. Baer, Jr Foundation, Ortho-McNeil-Janssen Pharmaceutical Ltd, Bristol-Myers Squibb, Wyeth Pharmaceuticals, and Pfizer Inc; has served as a consultant to AstraZeneca Pharmaceuticals, Bristol-Myers Squibb, Janssen Pharmaceuticals, Pfizer Inc, Solvay Pharmaceuticals, Otsuka Pharmaceutical Group, Wyeth Pharmaceuticals, Forest Laboratories, Sanofi Aventis, H. Lundbeck A/S, Tikvah Therapeutics Inc, Otsuka Pharmaceuticals, and Vanda Pharmaceuticals; has been a member of data and safety monitoring boards for Organon Pharmaceuticals, Merck/Schering-Plough, Dainippon Sumitomo Pharma Co Ltd, and VIVUS Inc; has been a consultant to litigation; and has received royalties from Compact Clinicals/Jones and Bartlett Publishers for a metabolic screening form. Additional Contributions: We acknowledge the collective efforts of our Medicaid collaborators from California, Missouri, and Oregon for their assistance in data acquisition and manuscript review. REFERENCES 1. Zuvekas SH. Prescription drugs and the changing patterns of treatment for mental disorders, Health Aff (Millwood). 2005;24(1): Walton SM, Schumock G, Lee KV, Alexander G, Meltzer D, Stafford R. Prioritizing future research on off-label prescribing: results of a quantitative evaluation. Pharmacotherapy. 2008;28(12): Abilify (aripiprazole): US prescribing information. Accessed February 8, Geodon (ziprasidone HCl): US prescribing information. Pfizer Inc Web site. http: // Accessed February 8, Zyprexa (olanzapine): US prescribing information. Eli Lilly and Company Web site. Accessed February 8, Seroquel (quetiapine fumarate): US prescribing information. AstraZeneca Pharmaceuticals LP Web site. /2009/020639s045s046lbl.pdf. Accessed February 8, Risperdal (risperidone): US prescribing information. Ortho-McNeil-Janssen Pharmaceuticals Inc Web site. /risperdal_0.pdf. Accessed February 21, Pediatric Exclusivity Labeling Changes. United States Food and Drug Administration Web site. /PediatricTherapeuticsResearch/UCM pdf. Accessed February 21, Aparasu RR, Bhatara V. Patterns and determinants of antipsychotic prescribing in children and adolescents, Curr Med Res Opin. 2007;23(1): Olfson M, Blanco C, Liu L, Moreno C, Laje G. National trends in the outpatient treatment of children and adolescents with antipsychotic drugs. Arch Gen Psychiatry. 2006;63(6): Zito JM, Safer DJ, dosreis S, et al. Psychotropic practice patterns for youth: a 10-year perspective. Arch Pediatr Adolesc Med. 2003;157(1): Zito JM, Safer DJ, Sai D, et al. Psychotropic medication patterns among youth in foster care. Pediatrics. 2008;121(1):e157-e Accessed February 8, Constantine R, Tandon R. Changing trends in pediatric antipsychotic use in Florida s Medicaid program. Psychiatr Serv. 2008;59(10): Morrato EH, Druss B, Hartung DM, et al. Metabolic testing rates in three state Medicaid programs after FDA warnings and ADA/APA recommendations for second-generation antipsychotic drugs. Arch Gen Psychiatry. 2010;67(1): Deurell M, Weischer M, Pagsberg AK, Labianca J. The use of antipsychotic medication in child and adolescent psychiatric treatment in Denmark: a crosssectional survey. Nord J Psychiatry. 2008;62(6): Zito JM, Safer DJ, Berg LT, et al. A three-country comparison of psychotropic medication prevalence in youth. Child Adolesc Psychiatry Ment Health. 2008; 2(1): Rani F, Murray ML, Byrne PJ, Wong IC. Epidemiologic features of antipsychotic prescribing to children and adolescents in primary care in the United Kingdom. Pediatrics. 2008;121(5): Schirm E, Tobi H, Zito JM, de Jong-van den Berg LT. Psychotropic medication in children: a study from the Netherlands. Pediatrics. 2001;108(2):e25. http: //pediatrics.aappublications.org/cgi/reprint/108/2/e25. Accessed February 8, Clavenna A, Rossi E, Derosa M, Bonati M. Use of psychotropic medications in Italian children and adolescents. Eur J Pediatr. 2007;166(4): Sevilla-Dedieu C, Kovess-Masfety V. Psychotropic medication use in children and adolescents: a study from France. J Child Adolesc Psychopharmacol. 2008; 18(3): Rosack J. FDA to require diabetes warning on antipsychotics. Psychiatr News. 2003;38(20):1. 350

8 22. Fanapt (iloperidone): US prescribing information. Vanda Pharmaceuticals Inc Web site. %20Fanapt%20full%20Master.pdf. Accessed February 8, Invega (paliperidone): US prescribing information. Ortho-McNeil-Janssen Pharmaceuticals Inc Web site. Accessed February 8, American Diabetes Association; American Psychiatric Association; American Association of Clinical Endocrinologists; North American Association for the Study of Obesity. Consensus development conference on antipsychotic drugs and obesity and diabetes. Diabetes Care. 2004;27(2): Vitiello B, Correll C, van Zwieten-Boot B, Zuddas A, Parellada M, Arango C. Antipsychotics in children and adolescents: increasing use, evidence for efficacy and safety concerns. Eur Neuropsychopharmacol. 2009;19(9): Hammerman A, Dreiher J, Klang SH, Munitz H, Cohen AD, Goldfracht M. Antipsychotics and diabetes: an age-related association. Ann Pharmacother. 2008; 42(9): McIntyre RS, Jerrell JM. Metabolic and cardiovascular adverse events associated with antipsychotic treatment in children and adolescents. Arch Pediatr Adolesc Med. 2008;162(10): Correll CU, Carlson HE. Endocrine and metabolic adverse effects of psychotropic medications in children and adolescents. J Am Acad Child Adolesc Psychiatry. 2006;45(7): Arango C, Robles O, Parellada M, et al. Olanzapine compared to quetiapine in adolescents with a first psychotic episode. Eur Child Adolesc Psychiatry. 2009; 18(7): Khan RA, Mican LM, Suehs BT. Effects of olanzapine and risperidone on metabolic factors in children and adolescents: a retrospective evaluation. J Psychiatr Pract. 2009;15(4): Correll CU, Manu P, Olshanskiy V, Napolitano B, Kane JM, Malhotra AK. Cardiometabolic risk of second-generation antipsychotic medications during firsttime use in children and adolescents. JAMA. 2009;302(16): Morrato EH, Newcomer JW, Allen RR, Valuck R. Prevalence of baseline serum glucose and lipid testing in users of second-generation antipsychotic drugs: a retrospective, population-based study of Medicaid claims data. J Clin Psychiatry. 2008;69(2): Haupt DW, Rosenblatt LC, Kim E, Baker RA, Whitehead R, Newcomer JW. Prevalence and predictors of lipid and glucose monitoring in commercially insured patients treated with second-generation antipsychotic agents. Am J Psychiatry. 2009;166(3): Meltzer HY. Putting metabolic side effects into perspective: risks versus benefits of atypical antipsychotics. J Clin Psychiatry. 2001;62(suppl 27):35-41, discussion Ma J, Vaillancourt R, Boddam R, Auger S, Sampalis J. Association between antidepressant use and prescribing of gastric acid suppressants. Can J Psychiatry. 2006;51(3): National Center for Health Statistics and Centers for Medicare and Medicaid Services. International Classification of Diseases, Ninth Revision Clinical Modification (ICD-9-CM). 6th ed. Hyattsville, MD: National Center for Health Statistics; Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project (HCUP). Clinical Classifications Software (CCS) for ICD-9-CM Fact Sheet. Healthcare Cost and Utilization Project Web site /toolssoftware/ccs/ccsfactsheet.jsp. Accessed December 29, Rector TS, Wickstrom SL, Shah M, et al. Specificity and sensitivity of claimsbased algorithms for identifying members of Medicare Choice health plans that have chronic medical conditions. Health Serv Res. 2004;39(6, pt 1): Quam L, Ellis LB, Venus P, Clouse J, Taylor CG, Leatherman S. Using claims data for epidemiologic research: the concordance of claims-based criteria with the medical record and patient survey for identifying a hypertensive population. Med Care. 1993;31(6): Newcomer JW, Hennekens CH. Severe mental illness and risk of cardiovascular disease. JAMA. 2007;298(15): Marder SR, Essock SM, Miller AL, et al. Physical health monitoring of patients with schizophrenia. Am J Psychiatry. 2004;161(8): Sikka R, Xia F, Aubert RE. Estimating medication persistency using administrative claims data. Am J Manag Care. 2005;11(7): Seroquel labelling with pediatric indication. AstraZeneca International Web site. Accessed February 8, Zyprexa labelling with pediatric indication. Eli Lilly and Company Web site. http: // Accessed February 8, Kuehn BM. FDA panel OKs 3 antipsychotic drugs for pediatric use, cautions against overuse. JAMA. 2009;302(8): Daniels SR, Greer FR; Committee on Nutrition. Lipid screening and cardiovascular health in childhood. Pediatrics. 2008;122(1): American Diabetes Association. Standards of medical care in diabetes Diabetes Care. 2008;31(suppl 1):S12-S American Diabetes Association. Standards of medical care in diabetes Diabetes Care. 2009;32(suppl 1):S13-S Ader M, Garvey WT, Phillips LS, et al. Ethnic heterogeneity in glucoregulatory function during treatment with atypical antipsychotics in patients with schizophrenia. J Psychiatr Res. 2008;42(13): Correll CU. Monitoring and management of antipsychotic-related metabolic and endocrine adverse events in pediatric patients. Int Rev Psychiatry. 2008;20 (2): Goodwin R, Gould MS, Blanco C, Olfson M. Prescription of psychotropic medications to youths in office-based practice. Psychiatr Serv. 2001;52(8): Cooper WO, Arbogast PG, Ding H, Hickson GB, Fuchs DC, Ray WA. Trends in prescribing of antipsychotic medications for US children. Ambul Pediatr. 2006; 6(2): Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, JAMA. 2010;303(3): Morrato EH, Newcomer JW, Kamat S, Baser O, Harnett J, Cuffel B. Metabolic screening after the American Diabetes Association s consensus statement on antipsychotic drugs and diabetes. Diabetes Care. 2009;32(6): There are only 2 things a child will share willingly communicable diseases and his mother s age. Benjamin Spock, MD, Dr Spock s Baby and Child Care,

9 primarily from limitations imposed by conducting an epidemiologic study in the context of a cluster investigation. For example, the differential interval to interview for case and control mothers was a reflection of complete case ascertainment and careful information gathering, which is described in phases 1 and 2 of the Centers for Disease Control and Prevention guidelines. Also, the number of potential controls was quite high (n=389) because of budget and time constraints, which prevented us from sending letters or making additional telephone calls to inquire about participation. (Although Kirby and Marshall suggest that only 57 of the 389 chose to participate, the Methods section states that 312 women [80%] were never contacted because of disconnected telephones or no answer after 4 calls.) We acknowledge the imperfections of our study and encourage readers to critically evaluate the findings. Nevertheless, we believe that the study makes a contribution to the growing body of literature regarding gastroschisis etiology. Leslie Elliott, MPH, PhD Author Affiliation: School of Community Health Sciences, University of Nevada Reno, Reno. Correspondence: Dr Elliott, School of Community Health Sciences, University of Nevada Reno, Lombardi 215, Mailstop 274, Reno, NV (lelliott@unr.edu). Financial Disclosure: None reported. 1. Elliott L, Loomis D, Lottritz L, Slotnick RN, Oki E, Todd R. Case-control study of a gastroschisis cluster in Nevada. Arch Pediatr Adolesc Med. 2009;163(11): Guidelines for investigating clusters of health events. MMWR Recomm Rep. 1990;39(RR-11): Kucik JE, Bitsko RH, Williams L, Lazarus C, Jarman DW, Correa A. Birth defects cluster study: a national approach to birth defects cluster investigations. Birth Defects Res A Clin Mol Teratol. 2008;82(11): Correction Errors in Table and Text. In the Original Article titled Metabolic Screening in Children Receiving Antipsychotic Drug Treatment by Morrato et al, published in the April issue of the ARCHIVES (2010;164[4]: ), errors occurred in the Abstract section on page 344. The Patients section should have read: A total of 5370 children (ages 6-17 years) without diabetes mellitus taking SGA drugs and children without diabetes taking albuterol (control individuals) but no SGA drugs. The first sentence of the Results section should have read, Glucose screening was performed with 1699 (31.6% [95% CI, 30.4%-32.9%]) SGA-treated children vs 1891 (12.6% [12.1%-13.2%]) controls. The final sentence of that section should have read, The case incidence of glucose and lipid disorders was higher in SGA-treated vs albuterol-treated children (8.9 per 1000 children [95% CI, ]; vs 4.9 per 1000 children [ ]; and 9.7 per 1000 children [ ]; vs 4.6 per 1000 children [ ], respectively). Also, in Table 3, errors occurred in 2 column headings on page 348. The top-centermost and top-rightmost column headings should have read, Ages 6 to 12 Years, no.; rate per 1000 children (95% CI) and Ages 13 to 17 Years, no.; rate per 1000 children (95% CI), respectively. (REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 164 (NO. 6), JUNE