Outcome in patients with religious delusions.

We read with interest the recent report on the definition, diagnosis, and clinical implications of religious delusions (RD).1 In our sample of 929 delusional schizophrenia patients who had been admitted to two psychiatric hospitals in Germany between 2010 and 2014, 138 patients (15%) reported RD. In 569 cases, information on religious affiliation was available. Patients with religious affiliation did not differ from patients without religious affiliation in the frequency of RD [χ2(1,569)= 0.02, p= 0.885]. Furthermore, patients with RD did not differ from patients with other types of delusion (OD) in the duration of hospitalisation [t(924)= -0.39, p= 0.695], or the number of hospitalisations [t(927)= -0.92, p= 0.358]. Additionally, in 185 cases, information on Clinical Global Impressions (CGI) and Global Assessment of Functioning (GAF) was available at the beginning and end of the hospital stay. By CGI-scores, no difference was seen in morbidity of subjects with RD relative to subjects with OD on admission [t(183)= -0.78, p= 0.437] and discharge t(183)= -1.10, p= .273 . Likewise, GAF-scores on admission did not differ in these groups [t(183)= 1.50, p= 0.135]. However, a trend was noted for lower GAF-scores on discharge in subjects with RD [t(183)= 1.91, p= .057, d= 0.39, CI 95% (-0.12-0.78)]. While RD have often been associated with a poorer prognosis in schizophrenia,2,3 we argue that this need not apply to all domains. Mohr et al4 reported that patients with RD were less likely to maintain psychiatric treatment, but did not have a more severe clinical status than patients with OD. Iyassu et al5 found higher levels of positive, but also lower levels of negative symptoms in patients with RD compared to patients with OD. Groups did not differ in terms of length of illness or level of medication. Siddle et al6 reported higher symptom scores in patients with RD at their first presentation, but a similar response to treatment when compared to patients with OD after 4 weeks of treatment. Furthermore, Ellersgaard et al7 iindicated that first-episode psychosis patients with RD at baseline were more likely to be non-delusional at follow-ups conducted after years 1, 2 and 5 when compared to patients with OD at baseline. We conclude that RD may thus interfere with short-term clinical outcome. With regard to long-term effects more favourable observations exist8 and the interplay of psychotic delusions with non-psychotic beliefs still warrants further research.