The authors of a recent study concluded that it ‘gave no support to child sexual abuse being associated with schizophrenic disorders later in life’ (Reference Spataro, Mullen and BurgessSpataro et al, 2004). Despite numerous acknowledged limitations that ‘reduce the probability of finding a positive association between [child sexual abuse] and mental disorders’, males who had suffered child sexual abuse were 1.3 times, and abused females 1.5 times, more likely to have been subsequently treated for schizophrenic disorders than the general population. However, the researchers missed a crucial additional limitation. Because the abused subjects were drawn from police and court records many will have been removed from the abusive situation and received early support. The researchers warned, specifically in relation to schizophrenia: ‘Care must be taken in interpreting this and other negative findings’; we agree.

The researchers also claimed ‘the findings to date do not support an association between child sexual abuse and schizophrenia’, adding that this hypothesis ‘has claimed considerable public, if not professional, attention’. It seems professional attention has been somewhat selective.

There are many studies demonstrating the powerful relationship between child abuse (sexual and otherwise) and schizophrenia (reviewed by Reference Read, Goodman, Morrison, Read, Mosher and BentallRead et al, 2004). Studies of specific psychotic symptoms reveal that the relationship is particularly strong with hallucinations (Reference Hammersley, Dias and ToddHammersley et al, 2003; Read et al, Reference Read, Agar and Argyle2003, Reference Read, Goodman, Morrison, Read, Mosher and Bentall2004). When mediating variables are controlled for, the relationship, with both clinician-rated symptoms (e.g. Reference Read, Agar and ArgyleRead et al, 2003) and research measures of psychosis (e.g. Reference Janssen, Krabbendam and BakJanssen et al, 2004), remains significant.

One of the most robust of these studies was a prospective general population study (n=4045), controlling for age, gender, education, unemployment, urbanicity, ethnicity, discrimination, marital status, drug use, and psychotic symptoms or psychiatric care in first-degree relatives. On the three measures of psychosis, people who had suffered child abuse were 2.5, 7.3 and 9.3 times more likely to have psychosis. As in previous studies (e.g. Reference Read, Agar and ArgyleRead et al, 2003), there was a ‘dose–response’ relationship. Those who had experienced severe child abuse were 48 times more likely than the general population to have ‘pathology level’ psychosis (Reference Janssen, Krabbendam and BakJanssen et al, 2004).