Panic disorder : symptomatology, medical utilisation and treatment.

Abstract

The overall aim of this project was to investigate the nature and structure of the physiological symptoms of panic attacks and the relationship between these symptoms and use of the health care system by people with a clinical diagnosis of panic disorder. Cioffi's model of somatic interpretation was explored in relation to this issue as it had been previously applied to predominantly physiological conditions and appeared to offer a potentially useful framework for understanding the behaviour of people with panic disorder.The first study consisted of a principal components analysis of 153 panic attack symptom checklists from the Anxiety Disorders Interview Schedule - Third Edition - Revised (ADIS-III-R).Five separate physiological components emerged from the analysis which mirrored common medical conditions. A cluster analysis of the symptoms of 153 individuals indicated that the sample formed five separate groups corresponding to the five physiological components identified. The results of this study supported suggestions put forward in the literature regarding the possible clustering of the physiological symptoms of panic attacks. The study also found evidence to suggest that individuals with panic disorder can be identified in distinct sub-groups according to the most predominant physiological symptoms reported.The second study was made up of two parts. Part one investigated the health utilisation behaviour and associated costs for people with panic disorder and compared them with people with social phobia. Significantly higher costs and rates of utilisation were found for the panic disorder group compared to the group with social phobia. Part two of this study investigated the relationship between a person with panic disorder's most predominant physiological panic symptoms and the type of medical specialists consulted. Fifty three individuals with panic disorder were included in the study and the proposed relationship was analysed using a bi-partial regression analysis. The respiratory group was significantly related to the type of specialist seen.The third study was aimed at clarifying the interpretation of ambiguous symptoms in panic disorder. Thirty eight people with panic disorder completed a questionnaire requiring them to give explanations as to the cause of a number of ambiguous somatic sensations. It was hypothesised that there would be a relationship between the persons highest component score (as identified in the first study) and the interpretation of threat made in response to the items on the questionnaire. No such relationship was found although significantly more threat-interpretations were made when the individual's cognitive threat schema was activated.Study four investigated the influence of the type of panic recording measure upon the severity and number of panic symptoms reported. A secondary aim was to compare panic symptoms recorded following a panic provocation procedure in the clinic with those recorded following naturally occurring panic attacks. Thirty seven people with panic disorder recorded the symptoms of panic attacks experienced in the natural environment and those induced via hyperventilation in the clinic. It was hypothesised that there would be an effect for recording measure on the dependent variables of symptom severity and number. This hypothesis was supported with the structured recording measure producing significantly more symptoms of a greater severity than the unstructured or descriptive measure. An interaction effect was found for the neurological group of symptoms whereby the severity of symptoms was significantly higher in the clinic setting than in the natural environment with the descriptive measure resulting in significantly greater severity ratings.The final study investigated the efficacy of information-giving as an intervention for panic disorder. Forty individuals with panic disorder were randomly assigned to either receive two sessions of information-giving as well as self-monitoring of their symptoms or self-monitoring only. As hypothesised the group receiving information as well as self- monitoring had significantly lower levels of general anxiety and depression as well as anticipatory anxiety at the end of the intervention period.Several important implications emerge from these results. The finding that people with panic disorder can be identified according to the predominant set of physiological symptoms they report provides some useful information for identification of the problem in general medical settings. This project demonstrated the need for a screening measure for panic disorder in Australian medical settings as well as the potential effectiveness of the provision of information relating to anxiety and panic. In addition, Cioffi's model of somatic interpretation was found to be a useful framework with which to consider underlying processes relating to the interpretation of panic sensations.