The meaning of childhood attention-deficit hyperactivity symptoms in patients with a first-episode of schizophrenia-spectrum psychosis

Schizophrenia Research 126 (2011) 28–35

Contents lists available at ScienceDirect

Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s

The meaning of childhood attention-deficit hyperactivity symptoms in patients with a first-episode of schizophrenia-spectrum psychosis Victor Peralta ⁎, Elena García de Jalón, Maria S. Campos, Maria Zandio, Ana Sanchez-Torres, Manuel J. Cuesta Psychiatry Section B, Complejo Hospitalario de Navarra, Pamplona, Spain

a r t i c l e

i n f o

Article history: Received 5 May 2010 Received in revised form 16 August 2010 Accepted 15 September 2010 Keywords: Schizophrenia ADHD Neurodevelopment Familial liability Obstetric complications

a b s t r a c t Objective: To examine childhood ADHD symptoms in regard to their association with a number of illness-related variables including risk factors, early neurodevelopment, premorbid functioning and clinical characteristics in patients with schizophrenia-spectrum psychoses. Methods: One-hundred and twenty-two first-episode patients with DSM-IV schizophreniaspectrum disorders were retrospectively assessed by means of their biological mothers for childhood ADHD symptoms. Using correlational analyses and hierarchical regression models, the severity of ADHD symptoms was examined in relation to familial liability to schizophrenia, obstetric complications, milestones attainment delay, premorbid functioning during childhood and adolescence, age at illness onset, episode psychopathology and response to treatment after one-month trial with antipsychotic medication. Results: Twenty-one patients (17%) met DSM-IV criteria for childhood ADHD. Univariate analyses showed that severity of childhood ADHD symptoms was related to male gender, obstetric complications, delayed milestones attainment, poor school functioning and an earlier age of onset of psychotic symptoms. Hierarchical regression analyses showed that severity of childhood ADHD symptoms was independently predicted by obstetric complications and neurodevelopmental delay, with no further variables entering in the regression models. Path analyses showed that obstetric complications had both direct and indirect effects, through neurodevelopmental delay, on ADHD symptoms. Conclusions: These findings are consistent with a neurodevelopmental model of schizophrenia and with the hypothesis of shared environmental risk factors between ADHD and schizophrenia-spectrum disorders. Childhood ADHD symptoms in schizophrenia-spectrum disorders appear to be an epiphenomenon of obstetric complications and early neurodevelopment delay with no further influence on the clinical expression of the illness. © 2010 Elsevier B.V. All rights reserved.

1. Introduction

⁎ Corresponding author. Psychiatry Section B, Complejo Hospitalario de Navarra, Irunlarrea 3, 31008 Pamplona, Spain. Tel.: +34 848422488; fax: +34 848429924. E-mail addresses: [email protected] (V. Peralta), [email protected] (E.G. de Jalón), [email protected] (M.S. Campos), [email protected] (M. Zandio), [email protected] (A. Sanchez-Torres), [email protected] (M.J. Cuesta). 0920-9964/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2010.09.010

The characteristic symptoms of schizophrenia and spectrum disorders, which typically emerge in late adolescence or early adulthood, are preceded in a substantial proportion of individuals by a broad range of childhood behavioral abnormalities (Mehl, 1962; Gittelman-Klein and Klein, 1969; Fish et al., 1992; Cannon et al., 2009), which are central to a neurodevelopmental model of schizophrenia (Weinberger, 1987; Murray and Lewis, 1987). The neurodevelopmental theory of schizophrenia is relevant because it has directed

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

scientific attention to the origins of the disorder in early life, many years before the illness can be diagnosed. Developmental psychopathology of schizophrenia-related disorders likely arises from the dynamic interaction between genetic and environmental factors, some specific and some non-specific. In support of this, evidence has accumulated indicating that the emergence of certain psychopathologies during childhood and adolescence including neurodevelopmental deviance, cognitive dysfunction, and a broad range of behavioural disturbances parallels age-dependent anomalies of maturation processes in the brain (Keshavan et al., 2004; Paus et al., 2008). In recent years, an increasing focus of interest has been the study of symptoms of attention-deficit hyperactivity disorder (ADHD) during childhood in subjects with schizophrenia (Elman et al., 1998; Rubino et al., 2009) or who are at risk of developing the disorder (Niemi et al., 2003; Keshavan et al., 2002, 2008). These studies have consistently shown that a history of childhood ADHD is more frequent in patients with schizophrenia-spectrum disorders and in subjects at risk of developing these disorders than in the general population. However, the reported prevalence rate of ADHD in these conditions has varied widely, ranging between 5% (Stahlberg et al., 2004) and 84% (Ross et al., 2006). These divergent prevalence figures are likely due to differences in sample composition, assessment methodology and timing for diagnosing ADHD. For example, while high-risk studies have consistently reported that about one-fourth of at-risk subjects present with ADHD (Rieder and Nichols, 1979; Keshavan et al., 2008), cohort studies of schizophrenia patients have reported variable prevalence rates, with studies of earlyonset schizophrenia reporting the highest ones (Ross et al., 2006; Karatekin et al., 2007), which likely reflects the higher developmental impairment in these patients. Although evidence for the presence of ADHD symptoms in children at risk for, or who eventually develop schizophreniaspectrum disorders is well established, the correlates of these symptoms with illness-related characteristics are less clear. High-risk studies have shown that, compared with subjects without ADHD, those with ADHD display more neurocognitive dysfunction (Rieder and Nichols, 1979; Öner and Munir, 2005), more functional impairment (Öner and Munir, 2005), more psychotic-like symptoms (Keshavan et al., 2002) and an increased risk for developing schizophrenia-spectrum disorders (Rieder and Nichols, 1979; Keshavan et al., 2008). Only two previous studies, which were conducted in early-onset patients, have compared schizophrenia patients with and without a history of ADHD. These studies yielded contradictory results regarding differences among groups across a number of clinical variables (Elman et al., 1998; Karatekin et al., 2007). Given that studies of both at-risk subjects and early-onset schizophrenia are problematic regarding generalizability of findings, along with the possibility that associations between ADHD symptoms and clinical features are due to other intervening variables, the meaning of childhood ADHD symptoms in patients with schizophrenia and related disorders remains largely unknown. In this study we retrospectively examined ADHD symptoms during childhood in never-treated patients with a first-episode of schizophrenia-spectrum disorders and analyzed their unique associations with well-known risk factors for schizophrenia such as familial liability and obstetric complications (OCs), and with a

29

number of relevant illness-related variables including early developmental impairment, premorbid adjustment during childhood and adolescence, age at illness onset, index episode psychopathology and response to treatment. Based on previous findings, we hypothesized that the prevalence of ADHD largely exceeds that reported in the general population, and that ADHD symptoms are related to familial liability to schizophrenia, poor premorbid adjustment, more severe psychopathology and poor response to treatment. Given the lack of previous data regarding the relationship of ADHD symptoms with developmental delay and age at onset of the psychotic disorder, no specific hypotheses regarding these relationships were formulated. 2. Method 2.1. Participants This study is part of an ongoing study on antecedents, phenomenology and response to treatment of first-episode, drug-naive patients with schizophrenia-spectrum disorders. Full details of the original study design, sampling strategy and assessment methodology are available elsewhere (Peralta et al., 2010). The study population for the present study was made of 122 patients who were admitted for their first treatment and met a DSM-IV diagnosis of schizophrenia (n = 85), schizophreniform disorder (n = 28) or schizoaffective disorder (n = 9). The study was approved by the local ethics committee and all patients or their legal representatives gave written informed consent to participate. Inclusion criteria were: age 15–65 years, no previous exposition to antipsychotic drugs, and available biological mother. Exclusion criteria were: a history of drug dependence, evidence of organic brain disorder, IQb 70, or meaningful medical illness. Patients' biological mother provided information about the retrospectively collected variables, and only those patients whose mothers provided reliable information, as judged by the interviewer, were included in the study. The rationale for the exclusion criteria was to avoid potential secondary sources of neurological dysfunction, which was one of the main aims of the stem study. The mean age of the patients was 27.8 years (s.d. = 8.35); 84 were men and 38 women, and 111 were single. The average years of education was 12.2 (s.d. = 3.78), and 87 patients (71%) were full or part time employed before the index admission. 2.2. Assessments The subjects were administered the Comprehensive Assessment of Symptoms and History (CASH) (Andreasen et al., 1992), which served to assess demographics, diagnosis, illness-related variables and psychopathology. Diagnoses were made by consensus between the two senior psychiatrists (VP and MJC) at the end of the hospitalization period using all the available information. 2.2.1. Outcome variables The primary outcome measure of interest was the 10-item parents' version of the Wender Utah Rating Scale (WURS) (Ward et al., 1993). This scale is a condensed version of the of the 61item original scale to retrospectively assess ADHD symptoms in childhood. The two versions of the WURS have shown good

30

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

convergent validity (Ward et al., 1993). The abbreviated WURS rates 10 typical ADHD symptoms, according to their repercussion, in a severity range from 0 (not at all) to 3 (very much). The WURS was filled out by the mothers considering the age period between 6 and 10 years. A secondary outcome measure was a DSM-IV diagnosis of ADHD. It was based on information provided by the mothers who filled out a check-list for the presence during the same age period of the 18 items included in the DSM-IV criteria for ADHD. After completing the check-list, it was reviewed by MZ with the patients' mother in order to evaluate duration and level of impairment associated with the symptoms along with the age at onset of the symptoms causing impairment. On the basis of all this information, a DSM-IV diagnosis of ADHD was made by MZ who was blind to the clinical status of the patients. Mothers also provided information about previous history of diagnosis or treatments of ADHD, which was supplemented by clinical records.

2.2.2. Independent variables For the purposes of the present study, the independent measures of interest were grouped according to their temporal sequence and included 1) risk factors, 2) developmental milestones achievement, 3) childhood premorbid functioning, 4) adolescent premorbid functioning and premorbid deterioration, 5) age at illness onset, 6) index episode psychopathology, and 7) treatment response. Antecedent variables were all assessed through interviews with the patients' mother, which was supplemented by information provided by other close relatives and medical records.

2.2.2.1. Risk factors. Familial loading was assessed by means of the Family History-Research Diagnostic Criteria (Andreasen et al., 1977) in the first-degree relatives of the patients. To estimate the familial loading for schizophrenia we used the familial loading score designated by Pak Sham (Verdoux et al., 1996), which takes account of number of relatives affected, family size and age structure. This is a log-transformed score, which is intended to summarize the extent of psychiatric morbidity in the family by using a continuous measure of liability. OCs were assessed by means of the McNeil–Sjöström scale (McNeil and Sjöström, 1995). The scale provides a systematic evaluation and weighting of several hundred specific pregnancy, labor-delivery and neonatal factors, which are scored according to a 6-point scale reflecting the potential somatic damage in the offspring. In this study we used the Type C score, which reflects the highest severity level of any existing OC, and has proved to be highly sensitive to neurological damage in the offspring (Peralta et al., 2006). 2.2.2.2. Neurodevelopmental delay. The assessment of developmental milestones attainment was based on Jones et al. (1994), and included the assessment of 6 specific milestones not attained at the expected age: sitting, standing, walking, talking words, talking sentences and urine/feces control. We derived a neurodevelopmental delay score ranging from 0 (all milestones attained at the expected age) to 6 (none of the milestones attained at the expected age).

2.2.2.3. Premorbid functioning. The Premorbid Social Adjustment scale (Cannon et al., 1997) was used to assess social and academic adjustment each rated on a 7-point scale ranging from 1 (excellent adaptation) to 7 (extremely poor adaptation). The social and academic areas of adjustment were rated separately for childhood (5–11 years) and adolescence (12– 16 years). A global change score for premorbid adjustment was computed by subtracting the childhood score from the adolescent score. This score is intended to reflect premorbid deterioration with age, a condition with potentially relevant clinical implications regarding childhood ADHD symptoms. 2.2.2.4. Age at illness onset. Age at illness onset was rated using the Symptom Onset in Schizophrenia scale (Perkins et al., 2000), which assess three different aspects of symptom onset: age at first non-specific symptom, age at first psychotic symptom and age at first continuous psychotic symptom. 2.2.2.5. Index episode psychopathology. Episode psychopathology was assessed through at least two interviews with the patient plus the information provided by close relatives and clinical records, and rated as its worst during the month previous to hospitalization. Episode psychopathology included the assessment of reality distortion, disorganization, and negative symptom domains, which were assessed with the CASH and each scored according to a 0–5 global rating (0 = absence, 1 = doubtful, 2 = mild, 3 = moderate, 4 = marked, and 5 = severe). 2.2.2.6. Response to treatment. Of the 122 patients assessed at baseline, 116 were re-assessed for psychopathological ratings after a 4-week trial with antipsychotic medication. Change scores for reality distortion, disorganization and negative domains were computed by subtracting the 4-week score from the baseline score. Higher values for change scores indicate greater improvement and higher scores for the rest of independent measures indicate greater impairment. Inter-rater reliability for all the rater-administered measures was good: mean intraclass correlation coefficient = 0.87, range = 0.78–0.96. 2.3. Statistics Variables included in regression analyses were inspected for normality, and because the normality assumption could not be held for most of them, all the variables were log transformed for use in subsequent analyses. Univariate statistics including the Pearson χ2 test, the Student's t test, and the Pearson correlation coefficients were used as appropriate. A Receiver Operator Characteristic (ROC) curve analysis was used to examine the diagnostic performance of the WURS total score (a global measure of severity of ADHD symptoms) against the DSM-IV criteria for ADHD. In modelling the relationships of ADHD symptoms with the illness-related variables we used hierarchical regression analyses to account for the temporal ordering among the variables. This modelling approach may help to delineate the risk factors for ADHD symptoms and to understand the role of these symptoms in the clinical expression of schizophrenia. The hierarchical regression model examined the association of the WURS score with 1) risk factors, 2) neurodevelopmental delay

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

score, 3) childhood premorbid functioning, 4) adolescence premorbid functioning and premorbid deterioration, 5) age of illness onset, 6) index episode psychopathology, and 7) response to treatment, in that order. To avoid multicollinearity among the various definitions of age at onset, only that for the first presentation of psychotic symptoms was entered in the regression model. The predictors of ADHD symptoms were specifically modelled according to their temporal ordering with the purpose of examining the direct and indirect (mediating) effects of the predictor variables on ADHD symptoms. In modelling these relationships we used the multiple regression analysis procedure as described by Baron and Kenny (1986) and Frazier et al. (2004). Using this procedure, a variable may be considered as dependent or independent as appropriate, and the standardized partial correlation of a predictor variable with its dependent variable is displayed as a path coefficient (beta weight). Age, gender and diagnosis (1 = schizophrenia, 0 = schizophreniform or schizoaffective disorder) were included at the first step of each regression model. Despite performing multiple tests, we choose not to correct for multiple comparisons in order to avoid the increase in Type II errors and because we used a priori hypotheses stated in a clear direction (Perneger, 1998). All tests were two-tailed, with statistical significance set at p b 0.05. 3. Results 3.1. Prevalence and severity of childhood ADHD symptoms Twenty-one patients (17%) met DSM-IV criteria for childhood ADHD, and the male/female prevalence ratio for that diagnosis was 4:1. Of the 21 patients with ADHD, 8 presented the predominantly inattentive subtype, 3 the predominantly hyperactive-impulsive subtype and 10 the combined subtype. Six further patients had meaningful ADHD symptoms not meeting criteria for ADHD. The mean WURS total score was 3.22, s.d.=4.47, range=0– 23, and the mean number of DSM-IV ADHD symptoms was 2.74, s.d.= 3.31, range= 0–16. In those patients with a DSM-IV diagnosis of ADHD, the mean WURS total score was 11.2, s.d.=4.90, range 5–23, and the mean number of DSM-IV ADHD symptoms was 8.6. s.d.= 2.81, range =6–16. According to clinical records and mothers' information, of the 21 patients with a history of ADHD only 5 were formally diagnosed as having such a disorder during childhood, four of them were treated only with psychoeducation and one patient received psychoeducation plus methylphenidate. 3.2. Correspondence between severity of ADHD symptoms and a diagnosis of ADHD The Cronbach α for the WURS and the DSM-IV symptom check-list were 0.85 and 0.91, respectively, thus suggesting that the two instruments are measuring homogeneous constructs. The WURS total score and the number of DSM-IV ADHD symptoms were closely related (r= .84, p b 0.001). ROC analysis showed that the WURS total score performed very well against a DSM-IV diagnosis of ADHD: area under the ROC curve= 0.98, s.e.= 0.01, p b 0.001. A WURS cut-off score ≥7 showed the best diagnostic efficiency: sensitivity = 0.91, specificity = 0.95.

31

3.3. Univariate analyses The WURS total score was inversely related to the age of the patients, but this association was non-significant (r=−0.16, p=0.078). The mean WURS score was significantly higher in men that in women (3.79±4.93 vs. 1.97±3.0, t=2.52, df=111,2, p=0.013), as was the mean number of DSM-IV symptoms, although at a trend level (3.09±3.54 vs. 1.97±2.61, t=1.74, df=120, p=0.083). Univariate analyses showed significant associations of the WURS score with OCs, neurodevelopmental delay score, poor academic functioning during childhood and adolescence, and two of the three definitions of age at illness onset (Table 1). Table 2 shows the distribution of the illness-related variables scores across patients with and without a childhood DSM-IV diagnosis of ADHD. The results were very similar to that reported for the WURS score excepting that age of onset of psychotic symptoms did not significantly differ across groups. 3.4. Multivariate analyses The WURS score was independently predicted by the McNeil–Sjöström scale score (beta = .39, s.e. = .12, t = 3.34, p = 0.006), and the neurodevelopmental delay score (beta = .46, s.e. = .14, t = 3.34, p b 0.001). After step 3, which included neurodevelopmental delay, no further variables entered in the regression models (Table 3). Fig. 1 illustrates the full path model for the development of ADHD symptoms. Path analysis indicated two independent pathways to ADHD symptoms: OCs and delayed milestone achievement. In addition, delayed milestone achievement Table 1 Mean (s.d.) scores of illness-related variables and their Pearson correlation coefficients with the Wender Utah Rating Scale score in 122 patients with schizophrenia-spectrum disorders. Mean s.d. Risk factors Familial loading score for schizophrenia a 1.16 McNeil–Sjöströn scale, type C score 1.42 Early neurodevelopment Neurodevelopmental delay score 0.93 Premorbid adjustment in childhood PSA, social adjustment score 7.10 PSA, academic adjustment score 6.06 Premorbid adjustment in adolescence PSA, social adjustment score 8.39 PSA, academic adjustment score 7.22 PSA, change score 2.45 Illness onset Age at first unspecific symptom 24.9 Age at first psychotic symptom 25.2 Age at first continuous 26.3 psychotic symptom Index episode psychopathology Reality-distortion symptoms 3.78 Disorganization symptoms 1.69 Negative symptoms 1.70 Symptom change b Reality-distortion symptoms 2.58 Disorganization symptoms 1.25 Negative symptoms 0.53 a b

Log-transformed variable. n = 116.

r

p

0.55 2.02

0.12 0.107 0.31 b 0.001

1.43

0.36 b 0.001

4.51 2.70

0.14 .23

5.10 2.98 3.53

0.10 0.273 0.32 b 0.001 0.07 0.433

0.130 .010

− 0.17 − 0.18 − 0.20

0.066 0.047 0.026

1.10 1.40 1.40

0.10 0.05 0.04

0.282 0.612 0.627

1.33 1.35 1.10

0.05 0.04 −.04

0.623 0.659 .680

8.4 8.0 8.1

32

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

Table 2 Distribution of illness-related variables in patients with and without a DSM-IV diagnosis of childhood attention-deficit hyperactivity disorder.

Risk factors Familial loading score for schizophrenia a McNeil–Sjöströn scale, type C score Neurodevelopment Neurodevelopmental delay score Premorbid adjustment in childhood PSA, social adjustment score PSA, academic adjustment score Premorbid adjustment in adolescence PSA, social adjustment score PSA, academic adjustment score PSA, change score Illness onset Age at first unspecific symptom Age at first psychotic symptom Age at first continuous psychotic symptom Index episode psychopathology Reality-distortion symptoms Disorganization symptoms Negative symptoms Symptom change b Reality-distortion symptoms Disorganization symptoms Negative symptoms a b

Without ADHD (n = 101)

With ADHD (n = 21)

tdf

Mean (s.d.)

Mean (s.d.)

1.17 (0.52) 1.29 (2.06)

1.15 (0.69) 2.09 (1.76)

0.14120 −2.33120

0.886 0.021

0.73 (1.27)

1.90 (1.76)

−3.67120

b 0.001

7.02 (4.45) 5.79 (2.56)

7.48 (4.93) 7.33 (3.02)

−0.44120 −2.10120

0.657 0.037

8.19 (5.04) 6.89 (2.90) 2.27 (2.97)

9.33 (5.37) 8.81 (2.94) 3.33 (5.52)

−1.00120 −2.62120 −0.74120

0.325 0.010 0.460

25.4 (8.82) 25.7 (8.52) 26.7 (8.48)

22.9 (6.13) 23.4 (5.00) 24.3 (5.33)

0.94120 0.81120 0.94120

0.350 0.417 0.350

3.80 (1.10) 1.62 (1.38) 1.63 (1.40)

3.67 (1.15) 2.00 (1.52) 2.05 (1.40)

0.45120 −1.01120 −1.34120

0.653 0.317 0.181

2.55 (1.36) 1.19 (1.38) 0.52 (1.06)

2.67 (1.24) 1.52 (1.21) 0.62 (1.32)

−0.42114 −1.16114 −0.1124.6

0.674 0.249 0.916

p

Log-transformed variable. n = 116.

that only the first two variables, OCs and neurodevelopmental delay, were independently related to childhood ADHD symptoms. Furthermore, path analysis revealed that neurodevelopmental delay was a mediator variable between obstetric complications and ADHD symptoms, or in other words, OCs had both direct and indirect effects on ADHD symptoms. Thus, OCs appear to play a key role in producing ADHD symptoms in schizophrenia-spectrum disorders, a finding not reported previously. Associations of childhood ADHD symptoms with variables corresponding to the same and subsequent time periods in which the ADHD symptoms were assessed (i.e. poor premorbid academic functioning and earlier age of onset of psychotic symptoms) were accounted for by risk factors and early neurodevelopment. These findings support the view of childhood ADHD symptoms in patients with schizophrenia-related disorders as the expression of an altered neurodevelopment and may be of importance for the further understanding of the

served as a mediator between OCs and ADHD symptoms. Another way to describe the amount of mediation is in terms of the proportion of the total effect that is mediated. Using the standardized regression coefficients, we get 0.23 × 0.27 / 0.30 = 0.21. Thus, about 21% of the total effect of OCs on ADHD symptoms is mediated by neurodevelopmental delay. 4. Discussion 4.1. Main findings We found that 17% of first-episode schizophrenia-spectrum patients met DSM-IV criteria for childhood ADHD. Correlational analyses showed that severity of childhood ADHD symptoms was related to OCs, delay of milestones attainment, poor academic functioning during childhood and adolescence and an earlier age of onset of psychotic symptoms. Multivariate analyses showed

Table 3 Model summary of hierarchical multiple regression analyses predicting symptoms of childhood attention-deficit hyperactivity disorder in 122 patients with schizophrenia-spectrum disorders.

Step Step Step Step Step Step Step Step a

1 (confounders) 2 (risk factors) 3 (neurodevelopmental delay) 4 (childhood adjustment) 5 (adolescence adjustment) 6 (age at illness onset) 7 (index episode psychopathology) 8 (symptom change) a

n = 116.

R2

R2 change

F change

df

p

0.06 0.15 0.23 0.24 0.28 0.30 0.31 0.31

0.06 0.09 0.08 0.01 0.04 0.02 0.01 0.00

2.38 5.60 11.2 1.21 2.31 0.19 0.43 0.04

3112 2110 1109 2107 3104 1103 3100 3970

0.073 0.005 b 0.001 0.301 0.099 0.661 0.731 0.983

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

.23 (p=0.01)

Obstetric complications

Neurodevelopmental delay

.30 (p<0.001)

33

.27 (p=0.002)

Attention-deficit hyperactivity symptoms

Fig. 1. Path analytic model for the development of attention-deficit hyperactivity symptoms in 122 patients with schizophrenia-spectrum disorders. The values in the model are standardized beta coefficients (and the associated p value).

childhood precursors of adult psychotic disorders. The lack of association between childhood ADHD symptoms and adult psychotic symptoms suggests that they arise from different neurobiological mechanisms, which does not exclude the possibility that the former are a risk factor for the latter. In summary, childhood ADHD symptoms in schizophrenia-spectrum disorders appear to be, at least in part, an epiphenomenon of obstetric complications and early neurodevelopment delay with no further influence on the clinical expression of the illness. 4.2. Comparison with the literature Our 17% prevalence rate of a history of childhood ADHD in schizophrenia-spectrum disorders was in the lower range of that reported in previous studies. This lower rate may be partially accounted for by the exclusion of patients with lifetime drug dependence, a diagnosis highly comorbid with ADHD, along with possible underreporting by the mothers. Gomez et al. (1981) using a shortened version of the WURS, reported a very similar prevalence rate of 15% in psychotic inpatients. Notwithstanding, our prevalence rate for childhood ADHD is still notably higher than the 5% prevalence rate reported in the general population (Polanzyczyk et al., 2007), and adds to the evidence that childhood ADHD is overrepresented in patients with schizophrenia-spectrum disorders. We found a lack of association between ADHD symptoms severity and with familial liability to schizophrenia, which is at odds with studies reporting an elevated prevalence of ADHD in at-risk individuals (Rieder and Nichols, 1979; Keshavan et al., 2008). Given the nature of our data we are cautious about drawing any further conclusions at this point, although it might also be that the degree of such a relationship varies as a function of the study sample. Thus, it is possible that development impairments are more likely to reflect a genetic vulnerability to schizophrenia in high-risk samples but not in low-risk samples. In line with previous reports of early-onset schizophrenia we found an association of childhood ADHD symptoms with more developmental deviance (Elman et al., 1998), poor school functioning (Schaeffer and Ross, 2002), an earlier age at illness onset (Elman et al., 1998; Karatekin et al., 2007), and a lack of association with symptoms of schizophrenia (Elman et al., 1998; Karatekin et al., 2007); however, we could not confirm previous reports of an association with poorer response to treatment (Elman et al., 1998). Findings from early-onset studies may not generalize to all patients with the disorder, since early-onset schizophrenia is a rare, severe, more genetically loaded and more developmentally impaired form of the disorder (Alaghband-Rad et al., 1995).

4.3. Explanations about the relationship between childhood ADHD and adult schizophrenia-spectrum disorders A number of explanations may be advanced to understand the controversial nature of such a relationship: A) the association between ADHD symptoms and schizophrenia is the result of shared risk factors, B) ADHD symptoms are the expression of a developmental alteration inherent to the familial liability to schizophrenia (Keshavan et al., 2008), C) children with ADHD may represent a subgroup with increased risk for schizophrenia-spectrum disorders (Keshavan et al., 2008), and D) patients with schizophrenia and ADHD symptoms represent a distinct clinical condition, the so called ADHD-psychosis, which is characterized by distinctive clinical features and poor response to antipsychotic drugs (Bellak and Charles, 1979). Overall, our findings provide considerable support for explanation A, some support for explanation C, and no support for explanations B and D. Explanation A is also supported by a wealth of data indicating that many of the risk factors identified in schizophrenia are shared with other psychiatric illnesses (Weiser et al., 2005; Pasamanick et al., 1956; Verdoux and Sutter, 2002) and with psychiatric symptoms in the general population (Breetvelt et al., 2010). More specifically, the association between OCs and ADHD symptoms reported in this study parallels that reported in ADHD patients (Milberger et al., 1997; Biederman and Faraone, 2005); likewise, the association between OCs and developmental delay also parallels that reported in ADHD patients (Bathia et al., 1991) and in the general population (Hediger et al., 2002). Furthermore, early neurodevelopmental delay has been also related to a range of psychiatric disorders (Sørensen et al., 2010). Therefore, it can be hypothesized that the observed links among OCs, developmental delay and ADHD symptoms probably lack diagnostic specificity with regard to schizophrenia-spectrum disorders, as these associations have been also described in a number of neuropediatric and psychiatric disorders. Despite this lack of specificity, the higher prevalence of ADHD in schizophrenia-spectrum disorders, relative to the general population, may indicate an increased risk for psychosis (Keshavan et al., 2008). This possibility needs to be examined in prospective follow-up studies, which would be of considerable relevance to early diagnosis of schizophrenia-spectrum disorders. 4.4. Strengths and limitations To our knowledge, this is the first study examining the associations of childhood ADHD symptoms with a number of relevant illness-related variables in a relatively representative

34

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35

sample of patients with a first episode of schizophreniaspectrum disorders. A further advantage of this study is that we simultaneously examined the independent association of ADHD symptoms with a number of variables including risk factors, premorbid functioning, age at onset, index episode psychopathology and response to treatment, as well as their incremental association in hierarchical regression analyses to account for the temporal ordering of the independent variables. A number of limiting factors and weaknesses exist for this study. First, the main study's limitation was that we relied on retrospective information for assessing variables antedating the index episode; thus we are limited by the constraints inherent in any retrospective assessment. Without prospective data recording, we do not know the extent to which the findings reported here are subject to recall bias. This concern is attenuated somewhat, as most retrospective assessments were reliable and based on multiple information sources in addition to that provided by the mother. Second, findings cannot generalize to the less severe patients not requiring hospitalization. Third, the inclusion of psychoses other than schizophrenia limits generalization of study's findings to schizophrenia. Fourth, patients with childhood ADHD symptoms may have been underrepresented because our exclusion criteria. Notwithstanding, our study sample still can be considered as the most representative one examining the correlates of ADHD symptoms in schizophrenia-spectrum disorders. Fifth, we examined a limited number of variables potentially relating to ADHD symptoms, thus we cannot exclude associations with other risk factors or clinical parameters. More specifically, a family history of ADHD was not recorded and as a consequence the putative contribution of familial ADHD to our data remains unknown. Lastly, we did not correct for multiple comparisons, and whereas spurious findings cannot be disregarded, they are not likely because all the statistically significant associations in multivariate analyses were within the 1% probability level. Role of funding source Funding for this study was provided by the Departmento de Salud del Gobierno de Navarra (grants 946/2005 and 55/2007) and the Ministerio de Educación y Ciencia (SAF2008-05674-C03-02). The two funding agencies had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. Contributors Victor Peralta and Manuel J. Cuesta designed the study, and wrote the protocol and the first draft of the manuscript. Maria S. Campos and Elena García de Jalón assessed the patients and managed the literature searches and analyses. Maria Zandio and Ana Sanchez-Torres contributed to design the study and undertook the statistical analysis. All authors contributed to and have approved the final manuscript. Conflict of interest All the authors declare that they have not conflicts of interest. Acknowledgement The authors thank the reviewers for many suggestions which have been incorporated in this article.

References Alaghband-Rad, J., McKenna, K., Gordon, C.T., Albus, K.E., Hamburger, S.D., Rumsey, J.M., Frazier, J.A., Lenane, M.C., Rapoport, J.L., 1995. Childhood onset schizophrenia: the severity of premorbid course. J. Am. Acad. Child Adolesc. Psychiatry 34, 1273–1283.

Andreasen, N.C., Endicott, J., Spitzer, R.L., Winokur, G., 1977. The family history method using diagnostic criteria. Arch. Gen. Psychiatry 34, 1229–1235. Andreasen, N.C., Flaum, M., Arndt, S., 1992. The Comprehensive Assessment of Symptoms and History (CASH): an instrument for assessing diagnosis and psychopathology. Arch. Gen. Psychiatry 49, 615–623. Baron, R.M., Kenny, D.A., 1986. The moderator–mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J. Pers. Soc. Psychol. 51, 1173–1182. Bathia, M.S., Nigam, V.R., Bohra, M., Malik, S.C., 1991. Attention deficit disorders with hyperactivity among paediatric outpatients. J. Child Psychol. Psychiatry 32, 297–306. Bellak, L., Charles, E., 1979. Schizophrenia syndrome related to minimal brain dysfunction: a possible neurologic subgroup. Schizophr. Bull. 5, 480–489. Biederman, J., Faraone, S.V., 2005. Attention-deficit hyperactivity disorder. Lancet 366, 237–248. Breetvelt, E.J., Boks, M., Numans, M., Selten, J.-P., Sommer, I.E.-C., Grobbee, D.E., Kahn, R.S., Geerlings, M.I., 2010. Schizophrenia risk factors constitute general risk factors for psychiatric symptoms in the population. Schizophr. Res. 120, 184–190. Cannon, M., Jones, P., Gilvarry, C., Rifkin, L., McKenzie, K., Foerster, A., Murray, R.M., 1997. Premorbid social functioning in schizophrenia and bipolar disorder: similarities and differences. Am. J. Psychiatry 154, 1544–1550. Cannon, M., Tarrant, C.J., Huttunen, M.O., Jones, P.B., 2009. In: Murray, R.M., Jones, P.B., Susser, E., van Os, J., Murray, R.M. (Eds.), Childhood development and later schizophrenia: evidence from genetic- high-risk and birth cohort studies. : The Epidemiology of Schizophrenia. Cambridge University Press, Cambridge, pp. 100–123. Elman, I., Sigler, M., Kronenberg, J., Lindenmayer, J.P., Doron, A., Mendlovic, S., Gaoni, B., 1998. Characteristics of patients with schizophrenia successive to childhood attention deficit hyperactivity disorder (ADHD). Isr. J. Psychiatry Relat. Sci. 35, 280–286. Fish, B., Marcus, J., Hans, S.L., Auerbach, J.G., Perdue, S., 1992. Infants at risk for schizophrenia: sequelae of a genetic neurointegrative defect: a review and replication analysis of pandysmaduration in the Jerusalem Infant Development Study. Arch. Gen. Psychiatry 49, 221–235. Frazier, P.A., Tix, A.P., Barron, K.E., 2004. Testing moderator and mediator effects in counseling psychology. J. Counsel. Psychol. 51, 115–134. Gittelman-Klein, R., Klein, D.F., 1969. Premorbid social adjustment and prognosis in schizophrenia. J. Psychiatr. Res. 7, 35–53. Gomez, R.I., Janowsky, D., Zetin, M., Huey, M., Clopton, P.L., 1981. Adult psychiatric diagnosis and symptoms compatible with the hyperactive child syndrome: a retrospective study. J. Clin. Psychiatry 42, 389–394. Hediger, M.L., Overpeck, M.D., Ruan, W.J., Troendle, J.F., 2002. Birthweight and gestational age effects on motor and social development. Paediatr. Perinat. Epidemiol. 16, 33–46. Jones, P., Rodgers, B., Murray, R., Marmot, M., 1994. Child developmental risk factors for adult schizophrenia in the British 1946 birth cohort. Lancet 344, 1398–1402. Karatekin, C., White, T., Bingham, C., 2007. Shared and nonshared symptoms in youth-onset psychosis and ADHD. J. Att. Dis. 20, 1–20. Keshavan, M., Sujata, M., Mehra, A., Montrose, D., Sweeney, J.A., 2002. Psicosis proneness and ADHD in young relatives of schizophrenia patients. Schizophr. Res. 59, 85–92. Keshavan, M., Kennedy, J., Murray, R., 2004. Neurodevelopment and Schizophrenia. Cambridge University Press, London. Keshavan, M., Montrose, D.M., Rajarethinam, R., Diwadkar, V., Prasad, K., Sweeney, J.A., 2008. Psychopathology among offspring of parents with schizophrenia: relationship to premorbid impairments. Schizophr. Res. 103, 114–120. McNeil, T.F., Sjöström, K., 1995. McNeil–Sjöstrom Scale for Obstetric Complications. Lund University Department of Psychiatry, Malmö University Hospital, Malmö, Sweden. Mehl, P.E., 1962. Schizotaxia, schizotypy, schizophrenia. Am. Psychol. 17, 827–838. Milberger, S., Biederman, J., Faraone, S.V., Guite, J., Tsuang, M.T., 1997. Pregnancy, delivery and infancy complications and attention deficit hyperactivity disorder: issues of gene–environment interaction. Biol. Psychiatry 41, 65–75. Murray, R.M., Lewis, S.W., 1987. Is schizophrenia a neurodevelopmental disorder? BMJ 295, 681–682. Niemi, L.T., Suvissaari, J.M., Tuulio-Henriksson, A., Lönnqvist, J.K., 2003. Childhood developmental abnormalities in schizophrenia: evidence from high-risk studies. Schizophr. Res. 60, 239–258. Öner, O., Munir, K., 2005. Attentional and neurocognitive characteristics of highrisk offsprings of parents with schizophrenia compared with DSM-IV attention deficit hyperactivity syndrome. Schizophr. Res. 76, 293–299. Pasamanick, B., Rogers, M., Lilienfeld, A., 1956. Pregnancy experience and the development of behavior disorder in children. Am. J. Psychiatry 112, 613–618. Paus, T., Keshavan, M., Giedd, J.N., 2008. Why do many psychiatric disorders emerge during adolescence? Nat. Rev. Neurosci. 9, 947–957.

V. Peralta et al. / Schizophrenia Research 126 (2011) 28–35 Peralta, V., Cuesta, M.J., Serrano, J.F., 2006. Obstetric complications and neurological abnormalities in neuroleptic-naive psychotic patients. Eur. Arch. Psychiatry Clin. Neurosci. 256, 407–413. Peralta, V., Campos, M.S., de Jalon, E.G., Cuesta, M.J., 2010. DSM-IV catatonia signs and criteria in first-episode, drug-naive, psychotic patients: psychometric validity and response to antipsychotic medication. Schizophr. Res. 118, 168–175. Perkins, D.O., Leserman, J., Jarskog, L.F., Graham, K., Kazmer, J., Lieberman, J.A., 2000. Characterizing and dating the onset of symptoms in psychotic illness: the Symptom Onset in Schizophrenia (SOS) inventory. Schizophr. Res. 44, 1–10. Perneger, T.V., 1998. What's wrong with Bonferroni adjustments. BMJ 316, 1236–1238. Polanzyczyk, G., Silva de Lima, M., Horta, B.L., Biederman, J., Rhode, L.A., 2007. The worldwide prevalence of ADHD: a systematic review and metaregression analysis. Am. J. Psychiatry 164, 942–948. Rieder, R.O., Nichols, P.L., 1979. Offspring of schizophrenics III. Hyperactivity and neurological soft signs. Arch. Gen. Psychiatry 36, 665–674. Ross, R.G., Heinlein, S., Tregellas, H., 2006. High rates of comorbidity are found in childhood-onset schizophrenia. Schizophr. Res. 88, 90–95. Rubino, I.A., Frank, L., Nanni, R.C., Pozzi, D., di Scalea, T.L., Siracusano, A., 2009. A comparative study of axis I antecedents before age 18 of unipolar depression, bipolar disorder and schizophrenia. Psychopathology 42, 325–332.

35

Schaeffer, J.L., Ross, R.G., 2002. Childhood-onset schizophrenia: premorbid and prodromal diagnostic and treatment histories. J. Am. Acad. Child Adolesc. Psychiatry 41, 538–545. Sørensen, H.J., Mortensen, E.L., Schiffmann, J., Reinish, J.M., Maeda, J., Mednick, S.A., 2010. Early developmental milestones and risk of schizophrenia: a 45-year follow-up of the Copenhagen perinatal cohort. Schizophr. Schizophr. Res. 118, 41–47. Stahlberg, O., Soderstrom, H., Rastam, Gillberg, C., 2004. Bipolar disorder, schizophrenia, and other psychotic disorders in adults with childhood onset AD/HD and /or autism spectrum disorders. J. Neural. Transm. 111, 891–902. Verdoux, H., Sutter, A.-L., 2002. Perinatal risk factors for schizophrenia: diagnostic specificity and relationships with maternal psychopathology. Am. J. Med. Gen. (Neuropsychiat. Gen.) 114, 898–905. Verdoux, H., van Os, J., Sham, P., Jones, P., Gilvarry, K., Murray, R., 1996. Does familiality predispose to both emergence and persistence of psychosis? A follow-up study. Br. J. Psychiatry 168, 620–626. Ward, M.F., Wender, P.H., Reimherr, F.W., 1993. The Wender Utah Rating Scale: an aid in the retrospective diagnosis of childhood attention deficit hyperactivity disorder. Am. J. Psychiatry 150, 885–890. Weinberger, D.R., 1987. Implications of normal brain development for the pathogenesis of schizophrenia. Arch. Gen. Psychiatry 44, 660–669. Weiser, M., van Os, J., Davidson, M., 2005. Time for a shift in focus in schizophrenia: from narrow phenotypes to broad endophenotypes. Br. J. Psychiatry 187, 203–205.