Kelleher and co-workers' (Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012) recent contribution addresses crucial issues in research on psychosis-proneness with potential implications for prevention. By revising the prevalence studies on psychotic-like experiences (PLEs) in children and adolescents, Kelleher et al. (Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012) provide a comprehensive map of mainstream empirical approaches in the field, but also reveal major flaws in the clinical and epidemiological validity of the results.

For example, two studies (Poulton et al. Reference Poulton, Caspi, Moffitt, Cannon, Murray and Harrington2000; Welham et al. Reference Welham, Scott, Williams, Najman, Bor, O'Callaghan and McGrath2009) are quoted in support of the higher risk of psychosis in adulthood following reports of childhood or adolescent psychotic symptoms. However, in Poulton et al. (Reference Poulton, Caspi, Moffitt, Cannon, Murray and Harrington2000) the data show that out of the 654 individuals screened at follow-up only 13 received a diagnosis of schizophreniform disorder. Of these only nine out of 95 participants were considered with ‘weak’ symptoms at age 11, and three out of 12 of those with ‘strong’ symptoms at age 11. Overall, of the 107 children considered at risk of psychosis from their screening, only 12 (11%) developed psychosis in the following years. In Welham et al. (Reference Welham, Scott, Williams, Najman, Bor, O'Callaghan and McGrath2009) the evidence in favour of early screening procedures is based on 56 adult positive cases out of a birth cohort (n=3573) assessed at ages 5 and 14. Unfortunately, the paper does not report a detailed description of the symptoms experienced and their distribution. In both studies the figures are statistically significant, but rather evanescent for epidemiological and clinical purposes. The low sensitivity of the assessment measures have also aroused clinical concerns including exposing young individuals to unnecessary assessment procedures and potential risk of stigmatization (Corcoran et al. Reference Corcoran, Malaspina and Hercher2005).

Moreover, the large part of the empirical literature reviewed by Kelleher et al. (Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012) assumes that children younger than 14 years old can unequivocally understand a question on ‘hearing voices no one else can hear’ with the same semiological subtlety of a clinician. This issue becomes evident when considering the question ‘Do you ever hear voices when you are alone?’ In this case, the prevalence estimates (i.e. 22% in youths aged 12–16 and 32% in those aged 12–18) are plausibly over-inflated for the very fact that children may have replied thinking about a real experience (i.e. really hearing voices when alone) rather than guessing the research's focus on hallucination-like experiences. Several studies have estimated the lifetime prevalence of any mental disorder to range between 10% and 36% (Alonso et al. Reference Alonso, Angermeyer, Bernert, Bruffaerts, Brugha, Bryson, de Girolamo, Graaf, Demyttenaere, Gasquet, Haro, Katz, Kessler, Kovess, Lépine, Ormel, Polidori, Russo, Vilagut, Almansa, Arbabzadeh-Bouchez, Autonell, Bernal, Buist-Bouwman, Codony, Domingo-Salvany, Ferrer, Joo, Martínez-Alonso, Matschinger, Mazzi, Morgan, Morosini, Palacín, Romera, Taub and Vollebergh2004; Kessler et al. Reference Kessler, Aguilar-Gaxiola, Alonso, Chatterji, Lee and Ustün2009). It seems unrealistic that prevalence figures for auditory hallucinations in children equates to the lifetime prevalence of mental disorders in the general population and it may be reasonable to suggest that some of the items contained in the current assessments might result in a gross overestimation of psychosis risk. Indeed, a growing empirical literature (e.g. Simonds et al. Reference Simonds, Demetre and Read2009) indicates that the specificity of some of the PLEs might be very limited and that they probably intercept a broad spectrum of sub-threshold psychopathology not necessarily forerunning psychosis, but also anxiety or affective disorders.

The appraisal of unusual subjective experiences and the role of emotions have been suggested to play a more crucial role in exacerbating and consolidating psychotic symptoms (Yung & McGorry, Reference Yung and McGorry1996; Freeman & Garety, Reference Freeman and Garety2003; Cella et al. Reference Cella, Cooper, Dymond and Reed2008; Preti et al. Reference Preti, Cella and Raballo2011). These research findings suggest that the role of associated features (e.g. appraisal and emotional context) may be as important for the development of psychosis as the frequency and content of the unusual experience per se. Without such caveats, stating that ‘psychotic symptoms are common in childhood and adolescence’ has the same heuristic value of the sentence ‘rain [i.e. a potential prodrome for hydrological disasters] is common in the northern hemisphere’.