Psychiatry and clinical psychology are the two dominant disciplines in mental health research, but the structure of scientific influence and information flow within and between them has never been mapped.

Citations among 96 of the highest impact psychiatry and clinical psychology journals were examined, based on 10 052 articles published in 2008. Network analysis explored patterns of influence between journal clusters.

Psychiatry journals tended to have greater influence than clinical psychology journals, and their influence was asymmetrical: clinical psychology journals cited psychiatry journals at a much higher rate than the reverse. Eight journal clusters were found, most dominated by a single discipline. Their citation network revealed an influential central cluster of ‘core psychiatry’ journals that had close affinities with a ‘psychopharmacology’ cluster. A group of ‘core clinical psychology’ journals was linked to a ‘behavior therapy’ cluster but both were subordinate to psychiatry journals. Clinical psychology journals were less integrated than psychiatry journals, and ‘health psychology/behavioral medicine’ and ‘neuropsychology’ clusters were relatively peripheral to the network.

Scientific publication in the mental health field is largely organized along disciplinary lines, and is to some degree hierarchical, with clinical psychology journals tending to be structurally subordinate to psychiatry journals.

Cognitive behaviour therapies (CBTs) have through several trials been demonstrated to reduce symptoms and disability in irritable bowel syndrome (IBS) patients, but the mechanisms responsible for the changes are still unknown. The aim of this study was to test a theoretical model of CBT and investigate if cognitions and/or behaviour mediated the changes seen in CBT for IBS.

To assess for possible mediating effects, we applied path analysis to the dataset of 149 diagnosed participants randomized to mebeverine hydrochloride plus CBT or mebeverine hydrochloride alone. Primary outcome was symptom severity, while secondary outcomes were work and social adjustment and anxiety.

The path analyses supported mediational paths for all outcomes. Changes in behaviour and cognitions mediated all three outcomes, with models placing behaviour change ‘upstream’ of cognition change having best fit. The analyses of model fits revealed best fit for the anxiety model and hence provide increased confidence in the causal model of anxiety.

Changes in behaviour and cognitions mediate the change in CBT given to IBS patients. The results strengthen the validity of a theoretical model of CBT by confirming the interaction of cognitive, emotional and behavioural factors in IBS.