Creating School^Age Versions ofSemistructured Interviews for the Prodrome to

Schizophrenia: Lessons From Case Reviewsby Randal Q* Ross, John Schaeffer, Nina Compagnon, Shari Heinlein,

Carol Beresford, and Qordon Farley

Abstract

There is an increasing emphasis on identifying individ-uals with schizophrenia earlier and earlier in their dis-ease process, with the assumption that earlier identifi-cation translates into earlier treatment, whichtranslates into improved outcome. Unfortunately, oneage cohort, children under 13 years of age, have beenexcluded from this critical alteration in clinical inter-vention strategy, and its associated improved clinicaloutcome. One of the barriers to inclusion of youngerchildren is the lack of knowledge about diagnosticissues related to attenuated psychotic symptoms in thisage sample. This report focuses on our experience withevaluating attenuated psychotic symptoms in youngchildren, in particular subthreshold hallucinations anddelusions, using semistructured interviews. The inclu-sion of both Caregiver and Child report sections andthe addition of concrete, detailed examples of clear-conscience, non-stress-related subthreshold psychoticsymptoms are likely to be necessary.

Keywords: childhood-onset schizophrenia, pro-drome, diagnostic interviews.

Schizophrenia Bulletin, 29(4):729-735,2003.

Introduction

Schizophrenia is generally characterized as a neurodevel-opmental disorder, in which the full clinical syndrome isassumed to be the result of abnormal alterations in braindevelopment, perhaps years before the onset of the fullclinical disorder (Weinberger 1987). This neurodevelop-mental hypothesis has an important corollary; specifi-cally, the earlier in the disease process treatment begins,the more likely the intervention is to stabilize or evennormalize brain development, and therefore the greaterthe benefit of treatment. Although far from conclusive, areview of the available evidence supports that the earliertreatment begins after onset of the full clinical disorder,

the better the long-term outcome (Wyatt and Henter2001).

The past several years have seen a concerted effort toextend these clinical findings by initiating antipsychotictreatment even before the full clinical disorder is present.For schizophrenia, and other psychotic disorders such asbipolar mood disorder with psychotic features, the fullclinical syndrome is often preceded by a period of sub-threshold psychosis and thought disorder, termed the pro-drome or at-risk mental state. A major advance in recentyears is the development of structured diagnostic inter-views, such as the Comprehensive Assessment of At RiskMental States (CAARMS; Phillips et al. 2000) and theStructured Interview for Prodromal Syndromes (SIPS;Miller et al. 2002), to identify individuals in the late pro-dromal period. These interviews identify a group ofultra-high-risk subjects, where 41 to 54 percent convert toan Axis I psychosis within the following 12 months(Miller et al. 2002; Phillips et al. 2000). In an attempt todecrease the duration of untreated psychosis and improvelong-term outcome (McGorry 2000), a number of double-blind treatment trials during the prodromal period areunderway. Early results (McGorry et al. 2002) support thevalue of a 6-month antipsychotic regimen in decreasingconversion to psychosis over a 1-year period, although itremains to be seen if the improvement in outcome contin-ues with longer followup.

The most common ages of onset for psychosis, and inparticular schizophrenic psychosis, are late adolescenceand early adulthood. Research protocols have generallyfocused on these ages of highest risk and have thus limitedparticipation to subjects 12 years of age and older. Child-hood-onset psychosis has been described in children asyoung as 4 years of age (Russell 1994; Spencer and Camp-bell 1994). Childhood-onset schizophrenia is similar toadolescent- and adult-onset forms of the disorder (Nicol-

Send reprint requests to: Dr. R.G. Ross, Bldg CPH, Rm 3F10, BoxC268-31, 4200 East 9th Ave., Denver, CO 80262; E-mail:randy.ross@uchsc.edu.

729

Schizophrenia Bulletin, Vol. 29, No. 3, 2003 R.G. Ross et al.

son et al. 2000), with the prepsychotic phases of child-hood-onset schizophrenia associated with significantimpairment, psychiatric care, and pharmacological inter-ventions (Alaghband-Rad et al. 1995; Schaeffer and Ross2002). Although rare, childhood-onset psychosis is oftenmore severe than adolescent- and adult-onset psychosis(Beitchman 1985; Alaghband-Rad et al. 1995) and createsa significant burden for affected children and their fami-lies. The ethical implications of treating false positivesrequires constant review of treatment trials; however, if aschool-age prodrome has the high levels of current mor-bidity seen in adolescent patients (Preda et al. 2002), theninvolvement in treatment trials of school-age children vul-nerable to psychosis may be warranted.

One of the factors hampering school-age enrollment inprodromal studies is the unknown applicability of relevantsemistructured interviews to this age group. The prodromalperiod is identified in adolescents and young adults by (a)the onset of attenuated (either lower frequency or lowerseverity) positive symptoms or (b) an increase in generalpsychopathology in a genetically vulnerable individual.When considering the application of these criteria toschool-age populations, at least three concerns can beraised: (1) children will be uncomfortable talking withunknown adults and thus will under-report; (2) childrenwill both under-report (because they don't understand thequestions) and over-report (because they have difficultyseparating real from nonreal), leading to higher levels ofboth false negatives and false positives; and (3) the typesand pattern of symptoms will vary with age, and combina-tions of symptoms used to identify adolescents and adultsas prodromal will not be applicable to school-age children.The purpose of this report is to use case examples to illus-trate these issues, so that empirical studies can be designed.

Subject Pool

The Colorado Childhood-Onset Schizophrenia ResearchProgram began in 1994 and was expanded to become theColorado Childhood-Onset Psychosis Research Programin 1998. The Program is focused on genetic etiology(Buervenich et al. 2000; Leonard et al. 2002), physiologi-cal correlates (Ross et al. 1999; Ross 2003), developmen-tal phenomenology (Ross and Compagnon 2001; Schaef-fer and Ross 2002), and treatment (Schaeffer and Ross2002) of children, ages 12 years and younger, with andvulnerable to psychosis. Children are referred to the Pro-gram because either (a) the treating clinician believes thechild has a psychotic illness or (b) the child has a 1stdegree relative with psychosis. The former group has a 4:1male to female ratio; the latter group consists of approxi-mately equal numbers of each gender. The average age ofchildren referred to the program is 10 years of age,

although children as young as 3 years of age have beenevaluated. All children referred to the program are diag-nosed using DSM-FV (American Psychiatric Association1994) criteria, using a structured diagnostic interview(Kaufman et al. 1997). All children involved in the pro-gram have received the parent and child versions of theKiddie-SADS-PL, although for children younger than 8years of age, the child interview is limited to affective andpsychotic symptomology. Over the last 6 months we havepiloted the inclusion of a modified Comprehensive Assess-ment of At-Risk Mental States (Phillips et al. 2000) in theassessment battery (n = 8 subjects to date). Semistructuredinterviews are completed by experienced research clini-cians with advanced degrees (M.S.W., D.O., or M.D.), andmedical records are reviewed. Final diagnosis is a consen-sus diagnosis from the research team. Age at onset ofsymptoms is based on parental recall and review of med-ical records. All children given an Axis I diagnosis of apsychotic disorder are assessed using high-resolution cyto-genetic screening and flourescent in situ hybridization(FISH) for 22qll deletion. To date, one child diagnosedwith schizophrenia had abnormalities on high-resolutioncytogenetic screening (a mosaic deletion distal to 16q22),and no child had 22qll deletions.

This Program has developed into the regional referralcenter for childhood psychosis for the state of Colorado andportions of surrounding states. Over the past 8 years, wehave completed structured interviews on approximately 130subjects 12 years of age and younger, who were referred aschildren with or vulnerable to psychosis. Approximately 45percent met diagnostic criteria for an Axis I psychotic disor-der: schizophrenia, schizoaffective, bipolar mood disorderwith psychosis, or major depression with psychosis. We takeextensive effort to differentiate psychosis associated withpost-traumatic stress disorder (PTSD) as a separate category,and only 5 to7 percent of our subjects given an Axis I psy-chotic disorder have a history of significant trauma.

Issue 1: Children Are Uncomfortable Talking WithUnknown Adults and Thus Will Under-Report

Case A. This 5-year-old male is the offspring of amother with bipolar-I disorder. He has a 2 1/2-year historyof unusual behavior, including inserting foreign objects inhis rectum, ingesting dirt, stabbing the family dog, andkilling the pet hamster, as well as unprovoked frequentaggression. Child A has threatened his siblings with scis-sors, threatened his mother with a knife, and described, indetail, his plans for murdering his brother with a gun.There was a 1-month history of irritability and lethargy,but no sleep or appetite changes nor pressured speech. OnMental Status Exam, Child A was a cheerful youngschool-age child with good eye-contact, age-appropriatevocabulary, and bright but anxious affect. Affect and eye

730

Prodrome in School-Age Children Schizophrenia Bulletin, Vol. 29, No. 4,2003

contact remained cheerful and positive while Child A dis-cussed his plans to murder his brother.

Direct questioning of Child A for psychotic symp-toms produced responses that were not clearly psychotic.The majority of questions concerning the presence ofpositive symptoms from the Kiddie Schedule for Affec-tive Disorders and Schizophrenia (KSADS) receivedeither a shrug of the shoulders, an initial response of "Idon't know," or a disorganized response that related littleto the question being asked. Conversely, when Child A'smother was interviewed, she repeated a number of psy-chotic symptoms that Child A had repeatedly demon-strated across settings, including command auditory hal-lucinations to kill himself and his peers. At these times,he got mad and hit his mother if she said she couldn'thear the voices. Child A had informed his mother ofvisual hallucinations where, when alone in the middle ofa field, he would start trying to swing at "children thatare trying to hit me." Child A also reported to his mothergrandiose delusions such as that "he can get the signal tochange from red to green just by thinking it." When thesespecific examples were repeated to Child A, he endorsedthem.

Case B. Child B is an 8-year-old male, adopted atbirth, who had been diagnosed with attention deficithyperactivity disorder (ADHD) and dyslexia at the ageof 6 and came to the clinic because of ineffectiveness ofmethylphenidate in reducing his symptoms. IQ is in thelow normal range. He has been expelled from a succes-sion of schools because of oppositional behavior, pri-marily refusing to do any schoolwork. Child B hadongoing difficulties with sustaining attention, was easilydistracted, had difficulties with staying in his seat,showed impulsivity, made frequent careless mistakes,had difficulty listening to directions, and was forgetfulin daily activities. In addition, Child B is notable forhigh levels of anxiety, with frequent concerns aboutwhether he will arrive at places on time, and fears thathis mother's plane will crash anytime she travels. ChildB's behavior was much calmer when within eyeshot ofhis mother. There were no reports of trauma, or symp-toms consistent with PTSD, depression, or mania. Whenquestioned about psychotic symptoms, Child B's parentsreported that, at 4 years of age, he would complain ofbeing afraid of ghosts, but that that had not occurredsince that time. Since then there had been no history ofhallucinations or delusions. In contrast, Child B pro-vided a rich, and consistent over time, description of theghosts that he saw "most of the time," including duringthe interview. The ghosts had "red eyes," could only beseen by him, and served to protect him from other indi-viduals. The ghosts controlled and guided his behavior,were responsible for all "nice" behaviors by Child B,

and provided him with answers to his homework. ChildB frequently felt "things walking on me,*' which heattributed to a variety of secret animals from a "privatepet shop." [Child B has no pets]. Child B believes thatolder students at school are constantly staring at andtalking about him, and planning to kill him. Affect wasflat, eye contact was intermittent, and Child B intermit-tently put his hands in his pants to masturbate through-out the exam.

Comment on Cases A and B. Many young chil-dren have difficulty conversing with unknown adults(strangers) and have difficulty with self-recognition ofbehavior as different from peers or outside of acceptednorms. Information obtained directly from Child Awould not have suggested a psychotic illness. In con-trast, inclusion of a parental interview provided clearevidence supporting the diagnosis of a DSM-IV Axis Ipsychotic disorder. Most diagnostic interviews ofyounger children require parental report, and interviewsfor prodromal symptoms should follow this model. Inour setting, with children under 12 years of age, parentsare always interviewed before the children.

While Case A was notable for the importance ofparental report in symptom extraction, Case B is anexample of the importance of direct child interview. Aninterview with the parents alone would have led to adiagnosis of ADHD and generalized anxiety disorder,while inclusion of child report led to a diagnosis of schiz-ophrenia. Jensen and colleagues (Jensen et al. 1999) havestressed the general value in most childhood psychiatricdisorders of utilizing the "either/or" criteria in childhoodassessment. Specifically, if either the child or the care-giver endorses a symptom, it should be considered aspresent, although the reliability of the responses shouldalso be considered. Because of the rarity of childhoodpsychosis, the work of Jensen et al. did not specificallyaddress psychotic symptoms; however, as illustrated byChildren A and B, both child and caregiver interviews arenecessary in evaluation of psychosis. Although symp-toms endorsed by the child should be discussed with thecaregivers to rule out alternative explanations, theeither/or criteria should, in general, be extended to thisdiagnostic group.

Issue 2. Children Will Both Under-Report (becausethey don't understand the questions) and Over-Report(because they have difficulty separating real from non-real), Leading to Higher Levels of Both FalseNegatives and False Positive Cases

Case C. The mother of this 8-year-old femalereported that this child began, at age 4 years, to describeunusual experiences, like seeing a jaguar run next to thecar (age 4), seeing colored spots in the air (age 6), or see-

731

Schizophrenia Bulletin, Vol. 29, No. 3, 2003 R.G. Ross et al.

ing big green worms (age 7). Child C also had, for severalyears, intermittently complained that adults in publicplaces were watching her and insisted on wearing a boxover her head for entire trips to the grocery store. In directquestioning of Child C, her responses to standard psy-chosis interview questions, such as "Do you hear voicesthat other people don't hear?" or "Do you see things thatother people don't see?" were almost universallyresponded to in the negative. However, when specificexamples were given, they were more likely to elicit posi-tive responses. For example, "Have you ever heard peopletalking and looked around and nobody was there?"resulted in Child C discussing a woman's voice callingher in the grocery store, or "Have you ever seen a ghost?"resulted in discussion of disembodied hands that followher. Positive response content to psychosis questions,whether maternal or Child C report, were notable for aconsistent pattern of low frequency events, occurring afew times per month or less.

Case D. Child D is a 7-year-old male who has beenfollowed longitudinally in our Program, receiving struc-tured diagnostic interviews at 4 years, 5 years, and 7years of age. He is the offspring of a father who hasschizoaffective disorder. Child D was exposed to domes-tic violence in the first 6 months of life and reportedlywas shaken once at age 6 months. There has been nocontact with the biological father since 6 months of age,and no history of notable trauma since that age. At allthree time points, attention and impulsivity have beenidentified as concerning symptoms. Both Child D andhis mother have denied, at all three time points, anysymptoms consistent with anxiety disorders (includingPTSD), mood disorders, or psychosis. While Child Dappeared to understand and respond appropriately toaggression and depression questions, the direct question-ing of Child D concerning psychotic symptoms wasmore problematic. Frequently, Child D would respondaffirmatively to general questions concerning psychoticsymptoms; however, a quizzical facial expression or anunusual response suggested incomplete understanding ofthe expression. For example, when asked if he everheard or saw anyone talking to him that no one elsecould see or hear, he responded "Yes"; additionalrequests for clarification elicited "I hear your voice and Isee you too." Despite the problems with Child D's com-prehension of general psychotic questions, concreteexamples, such as those described for Child C, weregenerally understood and associated with a negativeresponse.

Comment on Cases C and D. Two of the threeclusters of symptoms generally included in prodromaldiagnostic schemas are based primarily on the presenceof subthreshold positive symptoms (e.g., where halluci-

nations and/or delusions are present but are either notsevere enough or not present often enough to meet diag-nostic threshold). Thus, if prodromal schemas are to beapplicable to school-age populations, subthreshold posi-tive symptoms must be distinguishable from other formsof school-age hallucinations. Children have incompletedevelopment of cognition, emotional understanding,social interpretation, and verbal skills and may be at riskof both under-reporting true prodromal symptoms aswell as over-reporting normal phenomena such as imagi-nary companions or hypnagogic hallucinations (seeEgdell and Kolvin 1972 for a review). Supporting thelatter concern, hallucinations have been reported in 1percent of outpatient (Garralda 1984a) and 5 percent ofinpatient (Egdell and Kolvin 1972) child-psychiatry pop-ulations, while longitudinal followup suggests that con-version to a psychotic illness in children with hallucina-tions is no greater than the general population (Garralda19846).

Breslau (1987) has suggested that the validity of posi-tive psychotic symptoms can dramatically be improvedwith longer, more concrete questions. Children C and Dprovide examples of this phenomenon. Children withmore severe psychosis, like Child B, respond affirmativelyto psychosis questions even in the absence of specificexamples, even if they have never been psychiatricallyinterviewed before. However, for Child C, where psy-chotic symptoms are subthreshold for a psychotic diagno-sis, concrete examples of psychotic symptoms improveddetection. Especially for children who may in the prodro-mal period, where psychotic symptoms are of lower sever-ity, frequency, and duration, utilizing specific examples ofeach type of psychotic symptom may be necessary to elicitpositive responses. This is true for children under 10 yearsof age and particularly true for children under 8 years ofage. For children who have not themselves experiencedpsychotic symptoms, like Child D, there is an even greaterdifficulty comprehending psychosis questions. While gen-eral questions can lead to false positive responses, detailedconcrete examples may improve comprehension and mini-mize false positives. Examples likely should also include adescription of clear consciousness and presence of symp-toms unrelated to acute stressors (Egdell and Kolvin 1972)and determine that the child believes that the hallucina-tions or delusions are held with at least reasonable convic-tion (Yung et al. 2003). There is a theoretical possibilitythat repetition of a number of concrete examples providedby the examiner might suggest that this is what the exam-iner wants to hear, resulting in over-reporting by nonpsy-chotic children for secondary gain. Our experience withchildren below 12 years of age suggests that the frequencyof this event is low, even with children who come from afamily with psychosis.

732

Prodrome in School-Age Children Schizophrenia Bulletin, Vol. 29, No. 4, 2003

Issue 3: The Types and Pattern of Symptoms Will VaryWith Age and Combinations of Symptoms Used toIdentify Adolescents and Adults as Prodromal Will NotBe Applicable to School-Age Children

Continuation of Case C. As discussed above, this 8-year-old female endorsed hallucinations and delusions,but only for short durations a few times each month. Thischild and parent were re-interviewed 4 months after theinitial interview. Both parent and child gave a history ofmore frequent and severe psychotic symptoms. Thisincrease in the pervasiveness of the psychotic symptomswas associated with greater capacity by Child C toendorse psychotic symptoms in response to a more gen-eral psychosis question.

Case E. Child E was first evaluated as a 10-year-old male while living with his maternal grandmother;both Child E's mother and 12-year-old brother hadschizophrenia. Child D had mildly delayed speech andmotor milestones as an infant, met criteria for ADHD by5 years of age, and, since the age of 7 years, "worriesabout everything," including future performance andwhat others think of his behavior. There was no knownhistory of trauma; care by the maternal grandmother wasinitiated at birth. Child E was brought in for evaluationbecause he had on three occasions, without obviousprovocation, screamed and ran around the house fright-ened, believing a stranger was in the house trying to hurthim. Two of these events occurred waking up from asound sleep in the middle of the night, one occurred dur-ing daylight hours. Child E reported a single episode ofhearing voices for a few hours but being unable to makeout what they said. Child E believed the voices must behis imagination playing tricks on him. No other halluci-natory or delusional symptoms were endorsed. No moodsymptoms or substance use were present.

Over the following two months, Child E developed acase of severe trichotillomania and intermittent irritability.Six months after that, Child E heard multiple male voicesconversing frequently, disrupting attention and thoughtprocesses, with the experiences associated with screaming,hiding under his desk, and running out of the classroom.

Comment on Cases C and E. Child C demon-strates the prospective 4-month progression from sub-threshold to super-threshold psychosis consistent withwhat we have retrospectively reported for the prodromalperiod in school-age psychosis (Schaeffer and Ross2002) and is similar to prospective reports in adolescentand older subjects (Rosen et al. 2002). Child E presentsan alternative cluster of prodromal symptoms. Whileclear sub-threshold psychosis could not be established, arapid increase in general psychopathology in a geneti-cally vulnerable individual was followed by onset ofsuper-threshold psychotic symptoms, consistent with

one of the proposed adolescent presentations (Phillips etal. 2002). In at least some cases of school-age psychosis,onset is preceded by a prodromal period with sympto-matic presentation similar to that identified for adoles-cents and adults.

Discussion

Most psychiatric disorders—including ADHD, depres-sion, and psychosis—occur across the lifespan, fromearly school-age years to later adulthood. For many psy-chiatric disorders, symptom presentation is sufficientlysimilar in children, adolescents, and adults that the sameor very similar diagnostic criteria can be used overalmost the entire lifespan (American Psychiatric Associa-tion 1994). Empirical studies are necessary to establish ifthis pattern holds for the prodrome to psychosis diagnos-tic category. However, case review suggests that prodro-mal symptoms can be sensitively elicited in school-agechildren.

Despite the similarities across ages for symptom pre-sentation in many neuropsychiatric disorders, the rela-tionship between age and response to treatment is oftenmore unpredictable. While treatment response to stimu-lants is remarkably similar in children, adolescents, andadults with ADHD, the effect of age on response to anti-depressants in depression is more complex. Thus, it isinappropriate to assume that treatment trials in adoles-cents and adults can inherently be extended to youngerchildren without empirical evidence. If treatment forneuropsychiatric disorders in children is to be empiri-cally based, it is critical that children be included in treat-ment trials. However, to be included in treatment trials,age-appropriate diagnostic tools must be available. Foridentifying school-age children with the prodrome topsychosis, we suggest modification of current diagnostictools, such as the CAARMS and SIPS. Initial modifica-tions should include a more extensive list of examplepsychotic symptoms, which are detailed, concrete, whichoccur independent of acute stressors, and which occurduring clear consciousness. In addition, the diagnostictools should include both child and caregiver interviews.The either/or strategy for symptom assessment hasproved useful in younger children for other symptomdomains. Thus, for each symptom, the worst severityreported by either source should be considered as theseverity level of record.

References

Alaghband-Rad, J.; McKenna, K.; Gordon, C.T.; Albus,K.E.; Hamburger, S.D.; Rumsey, J.M.; Frazier, J.A.;

733

Schizophrenia Bulletin, Vol. 29, No. 3, 2003 R.G. Ross et al.

Lenane, M.C.; and Rapoport, J.L. Childhood-onset schiz-ophrenia: The severity of premorbid course. Journal ofthe American Academy of Child and AdolescentPsychiatry, 34:1273-1283, 1995.

American Psychiatric Association. DSM-IV: Diagnosticand Statistical Manual of Mental Disorders. 4th ed.Washington, DC: APA, 1994.

Beitchman, J.H. Childhood schizophrenia: A review andcomparison of adult-onset schizophrenia. PsychiatricClinics of North America, 8:793-814, 1985.

Breslau, N. Inquiring about the bizarre: False positives indiagnostic interview schedule for children (DISC) ascer-tainment of obessions, compulsions, and psychotic symp-toms. Journal of the American Academy of Child andAdolescent Psychiatry, 26:639-644, 1987.

Buervenich, S.; Carmine, A.; Arvidsson, M.; Xiang, R.;Zhiping, Z.; Sydow, O.; Jonsson, E.G.; Sedvall, G.C.;Leonard, S.; Ross, R.G.; Freedman, R.; Perlmann, T.;Anvret, M.; and Olson, L. NURR1 mutations in schizo-phrenia and manic depressive disorder. American Journalof Medical Genetics, 96:808-813, 2000.

Egdell, H.G., and Kolvin, I. Childhood hallucinations.Journal of Child Psychology and Psychiatry, and AlliedDisciplines, 13:279-287, 1972.

Garralda, M.E. Hallucinations in children with conductand emotional disorders: I. The clinical phenomena.Psychological Medicine, 14:589-596, 1984a.

Garralda, M.E. Hallucinations in children with conductand emotional disorders: II. The followup study.Psychological Medicine, 14:597-604, 19842?.

Jensen, P.S.; Rubio-Stipec, M.; Canino, G.; Bird, H.R.;Dulcan, M.K.; Schwab-Stone, M.E.; and Lahey, B.B.Parent and child contributions to diagnosis of mental dis-order: Are both informants always necessary? Journal ofthe American Academy of Child and AdolescentPsychiatry, 38:1569-1579, 1999.

Kaufman, I ; Birmaher, B.; Brent, D.; Rao, U.; Flynn, C ;Moreci, P.; and Williamson, D. Schedule for affective dis-orders and schizophrenia for school-age children—Present and lifetime version (K-SADS-PL): Initial relia-bility and validity data. Journal of the American Academyof Child and Adolescent Psychiatry, 36:980-988, 1997.

Leonard, S.; Gault, J.; Hopkins, J.; Logel, J.; Vianzon, R.;Short, M.; Drebing, C ; Berger, G.; Olincy, A.; Ross,R.G.; Adler, L.E.; and Freedman, R. Promotor variants inthe al nicotinic acetylcholine receptor subunit gene areassociated with an inhibitory deficit found in schizophre-nia. Archives of General Psychiatry, 59:1085-1090, 2002.

McGorry, P.D. Evaluating the importance of reducing theduration of untreated psychosis. Australian and New

Zealand Journal of Psychiatry, 34 (Suppl):S145-S149,2000.

McGorry, P.D.; Yung, A.R.; Phillips, L.J.; Yuen, H.P.;Francey, S.; Cosgrave, M.A.; Germano, D.; Bravin, J.;McDonald, T.; Blair, A.; Adlard, S.; and Jackson, H.Randomized controlled trial of interventions designed toreduce the risk of progression to first-episode psychosis ina clinical sample with subthreshold symptoms. Archivesof General Psychiatry, 59:921-928, 2002.

Miller, T.J.; McGlashan, T.H.; Rosen, J.L.; Sornjee, L.;Markovich, P.J.; Stern, K.; and Woods, S.W. Prospectivediagnosis of the initial prodrome for schizophrenia basedon the Structured Interview for Prodromal Syndromes:Preliminary evidence of interrater reliability and predic-tive validity. American Journal of Psychiatry,159:863-865, 2002.

Nicolson, R.; Lenane, M.; Hamburger, S.D.; Fernandez,T; Bedwell, J.; and Rapoport, J.L. Lessons from child-hood-onset schizophrenia. Brain Research. BrainResearch Reviews, 31:147-156, 2000.

Phillips, L.J.; Yung, A.R.; and McGorry, P.D.Identification of young people at risk of psychosis:Validation of Personal Assessment and Crisis EvaluationClinic intake criteria. Australian and New ZealandJournal of Psychiatry, 34(Suppl):S164-S169, 2000.

Phillips, L.J.; Yung, A.R.; Yuen, H.R; Pantelis, C ; andMcGorry, P.D. Prediction and prevention of transition topsychosis in young people at incipient risk for schizophre-nia. American Journal of Medical Genetics(Neuropsychiatric Genetics), 114:929-937, 2002.

Preda, A.; Miller, T.J.; Rosen, J.L.; Somjee, L.;McGlashan, T.H.; and Woods, S.W. Treatment histories ofpatients with a syndrome putatively prodromal to schizo-phrenia. Psychiatric Services, 53:342-344, 2002.

Rosen, R.L.; Woods, S.W.; Miller, T.J.; and McGlashan,T.H. Prospective observations of emerging psychosis.Journal of Nervous and Mental Disease, 190:133-141,2002.

Ross, R.G. Early expression of a pathophysiologicalfeature of schizophrenia: Saccadic intrusions intosmooth pursuit eye movements in school-age childrenvulnerable to schizophrenia. Journal of the AmericanAcademy of Child and Adolescent Psychiatry,42:468-476, 2003.

Ross, R.G., and Compagnon, N. Diagnosis and treatmentof psychiatric disorders in children with a schizophrenicparent. Schizophrenia Research, 50:123-131, 2001.

Ross, R.G.; Olincy, A.; Harris, J.G.; Radant, A.D.;Hawkins, M.; Adler, L.E.; and Freedman, R. Evidence forbilineal inheritance of physiological indicators of risk in

734

Prodrome in School-Age Children Schizophrenia Bulletin, Vol. 29, No. 4, 2003

childhood-onset schizophrenia. American Journal ofMedical Genetics, 88:188-199, 1999.

Russell, A.T. The clinical presentation of childhood-onsetschizophrenia. Schizophrenia Bulletin, 20(4):631-646,1994.

Schaeffer, J., and Ross, R.G. Childhood-onset schizophre-nia: Premorbid and prodromal diagnosis and treatmenthistories. Journal of the American Academy of Child andAdolescent Psychiatry, 41:538-545, 2002.

Spencer, E.K., and Campbell, M. Children with schizo-phrenia: Diagnosis, phenomonology, and pharmacother-apy. Schizophrenia Bulletin, 20(4):713-725, 1994.

Weinberger, D.R. Implications of normal brain develop-ment for the pathogenesis of schizophrenia. Archives ofGeneral Psychiatry, 44:660-669, 1987.

Wyatt, R.J., and Henter, I. Rationale for the study ofearly intervention. Schizophrenia Research, 51:69-76,2001.

Yung, A.R.; Phillips, L.J.; Yuen, H.P.; Francey, S.M.;McFarlane, C.A.; Hallgren, M.; and McGorry, P.D.Psychosis prediction: 12-month follow up of a high risk("prodromal") group. Schizophrenia Research,60(l):21-32, 2003.

Acknowledgments

This research was funded in part by Public Health Servicegrants R01 MH56539 and T32 MH 15442, the NationalAlliance for Research in Schizophrenia and Depression,Eli Lilly and Company, the Institute for Children's MentalDisorders, the Developmental Psychopathology ResearchGroup, and the Child and Adolescent Psychiatry ResidentTraining Program of the Department of Psychiatry, Uni-versity of Colorado Health Sciences Center, Denver, CO.We also gratefully acknowledge the involvement of thechildren and their families who have participated in ourresearch programs.

The Authors

Randal G. Ross, M.D., is an Associate Professor. JohnSchaeffer, D.O., is a resident in Child and Adolescent Psy-chiatry. Nina Compagnon, M.S.W., is a professionalresearch assistant. Carol Beresford, M.D., is a post-doc-toral fellow. Shari Heinlein, B.A., is a professionalresearch assistant. Gordon Farley, M.D., is a Professor. Allauthors are with the Department of Psychiatry, Universityof Colorado Health Sciences Center, Denver, Colorado.

735