IntroductionGeneral characteristics of the HR studiesNeurological and motor developmentCognitive functioningGeneral intelligenceAttentionOther cognitive functionsBehavior and social adjustmentSocial adjustmentMother-infant interactionPsychiatric symptomsOther findingsStability of rearing environmentStructural brain abnormalitiesPhysical anhedoniaDiscussionAcknowledgementsReferencesReviewChildhood developmental abnormalities in schizophrenia:evidence from high-risk studiesLaura T. Niemi*, Jaana M. Suvisaari, Annamari Tuulio-Henriksson,Jouko K. Lo¨nnqvistDepartment of Mental Health and Alcohol Research, KTL, National Public Health Institute, Mannerheimintie 166,FIN-00300 Helsinki, FinlandReceived 14 September 2001; accepted 13 February 2002AbstractAccording to cohort studies, individuals who develop schizophrenia in adulthood show developmental abnormalities inchildhood. These include delays in attainment of speech and motor milestones, problems in social adjustment, and pooreracademic and cognitive performance. Another method of investigating developmental abnormalities associated withschizophrenia is the high-risk (HR) method, which follows up longitudinally the development of children at high risk forschizophrenia. Most HR studies have investigated children who have a parent with schizophrenia. This review summarizesfindings concerning childhood and adolescent development from 16 HR studies and compares them with findings from cohort,conscript, and family studies. We specifically addressed two questions: (1) Does the development of HR children differ fromthat of control children? (2) Which developmental factors, if any, predict the development of schizophrenia-spectrum disordersin adulthood? While the answer to the first question is affirmative, there may be other mechanisms involved in addition tohaving a parent with schizophrenia. Factors which appear to predict schizophrenia include problems in motor and neurologicaldevelopment, deficits in attention and verbal short-term memory, poor social competence, positive formal thought disorder-likesymptoms, higher scores on psychosis-related scales in the MMPI, and severe instability of early rearing environment.D 2002 Elsevier Science B.V. All rights reserved.Keywords: Childhood developmental abnormalities; Schizophrenia; High-risk method1. IntroductionThe hypothesis that schizophrenia is a neurodeve-lopmental disorder is supported by several lines ofevidence (Weinberger, 1995). Abnormalities in braindevelopment and maturation seem to begin prenatally,but may continue th roughout childhood (Woods,1998). Minor physical anomalies, manifesting asslight anatomical de fects of the head, hair, eyes,mouth, hands, and feet, and thought to be caused bysome injury during the first or second trimester of fetallife, are more common among patients with schizo-phrenia than in their unaffected siblings or the generalpopulation (Murphy and Owen, 1996; Ismail et al.,1998). Obstetric complications, particularly hypoxic-ischemia-related complications, increase the risk for0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.doi:10.1016/S0920-9964(02)00234-7*Corresponding author. Tel.: +358-9-4744-8894; fax: +358-9-4744-8478.E-mail address: [email protected] (L.T. Niemi).www.elsevier.com/locate/schresSchizophrenia Research 60 (2003) 239 – 258later development of schizophrenia (Zornberg et al.,2000; Geddes and Lawrie, 1995). Neuropathologicalfindings suggest abnormalities in brain developmentamong patients with schizophrenia; absence of gliosissuggests, although does not prove, that they may be offetal origin (Heckers, 1997; Dwork, 1997). Infectionsand malnutrition during pregnancy may also increasethe risk for later developing schizophrenia (Susser etal., 1996; Mednick et al., 1988; Barr et al., 1990;Brown et al., 2000).If schizophrenia is a consequenc e of abnormalneurodevelopment, it seems reasonable to assume thatabnormality in development would somehow manifestitself during childhood. Follow-back, cohort and con-script studies have demonstrated associations betweenadult-onset schizophrenia and delays or abnormalitiesin childhood or adolescent emotional, cognitive,motor and /or social development. These includedelays in the attainment of early childhood motormilestones (Jones et al., 1994), problems in motorcoordination (Crow et al., 1995; Rosso et al., 2000) aswell as other problems in neuromotor development(Fish et al., 1992; Walker et al., 1999; Cannon et al.,1999; Rosso et al., 2000), w orse performance incognitive tests or lower IQ than among control chil-dren (Crow et al., 1995; David et al., 1997; Kremen etal., 1998; Davidson et al., 1999; Cannon et al., 2 000),problems with speech (DeLisi et al., 1991; Jones etal., 1994; Bearden et al., 2000), and difficulties insocial adjustment (Walker et al., 1993; Crow et al.,1995; Malmberg et al., 1998; Davidson et al., 1999;Bearden et al., 2000). However, two Finnish cohortstudies found no difference in school marks in aca-demic subjects between children who later developedschizophrenia and other cohort members (Isohanni etal., 1998; Cannon et al., 1999), the other even foundthat excellent school marks were more commonamong males who later developed schizophrenia thanamong males with no psychiatric disorders (Isohanniet al., 1999).There are certain weaknesses in these studydesigns. Deficient data, sampling problems, and thedifferent practices of doctors when examining patientsand recording findings, may have caused bias infollow-back studies. Although cohort and conscriptstudies are prospective and the data reliable, they cannever provide very detailed information because thou-sands of individuals are typically assessed. As themorbidity of schizophrenia in the general populationis quite low, a study method has been developed toenrich the sample with indi viduals who later developschizophrenia. This method is called the High-Risk(HR) method.High-Risk (HR) research refers to a method ofstudying the etiology of a disorder by investigatingindividuals who have an increased risk for developingit (Cornblatt and O buchowski, 1997).Themostimportant risk factors for schizophrenia are genetic:heritability estimates from the most recent twin stud-ies are as high as 83% (Cannon et al., 1998; Car dno etal., 1999). Because conclusive evidence for any par-ticular environmental factor being a risk factor forschizophrenia has been lacking, it has not beenpossible to identify children