Introduction and objective: The aim of the presented research was to analyse the differentiation of clinical diagnoses of autism spectrum disorder, attention-deficit/hyperactivity disorder and post-traumatic stress disorder with the use of diagnostic tools. Materials and methods: The study involved 110 carers of 2–10-olds, patients of a psychiatric treatment ward. Medical records were assessed for the diagnosis obtained by the child in the course of treatment, then parents were interviewed. The usefulness of individual methods for differentiating clinical diagnoses of autism, Asperger’s syndrome, attention-deficit/hyperactivity disorder and post-traumatic stress disorder was assessed using ROC (receiver operating characteristic) curves. Results: In many cases, the diagnoses obtained by the children were found to be different from those indicated by the tools. In addition, the tools themselves were in some cases found to be ineffective in diagnosing the disorder for which they were designed. The analyses also showed that crisis experiences can moderate the symptoms observed in children, thus influencing the diagnosis and the clinical picture. Conclusions: The presented analyses confirmed the thesis that the differential diagnosis of children is not an easy task, and that a potentially traumatic factor can affect the clinical picture of a disorder, as well as be its cause. The issues of unreliability of some tools used in the diagnosis and/or their inconsistency with the diagnoses obtained in the course of treatment were also found to be significant.