Abstracts

Rationale: Research examining patients with psychogenic nonepileptic seizures (PNES) has identified several personality features associated with this patient population. Before the results of these studies can be put into practice, however, the validity of these findings must be established. One identified threat to validity is selection/volunteer bias. Previous research has suggested that personality characteristics are significant correlates of willingness to participate in research. Some studies have suggested low participation rates for PNES patients, but this methodological issue has not been examined systematically before. The purpose of the present study is to document the frequency with which PNES patients decline participation in research examining emotional and personality factors associated with their diagnosis. Methods: Patients admitted to the Epilepsy Monitoring Unit (EMU) at the University of California, San Francisco (UCSF) during a twelve-month period were approached for a Quality of Life (QOL) study examining personality and emotional functioning. Approval from the UCSF Committee on Human Research was obtained in order to collect diagnostic and demographic variables from those patients who declined participation in the study. Additionally, the diagnoses of those patients who consented, but who were excluded from the study due to incomplete or invalid protocols were also examined. Results: One hundred thirty nine patients were approached regarding participation in the QOL study. Of the 16 patients who declined to participate, 4 (25%) were diagnosed with PNES upon discharge versus 8 (50%) of those who were diagnosed with epileptic seizures (ES). Four subjects who declined received neither PNES nor ES diagnoses. Of the remaining 123 subjects approached, 36 (29%) were diagnosed with PNES, 75 (61%) were diagnosed with ES, 1 was diagnosed with ES+PNES, and 11 (9%) were diagnosed with other paroxysmal disorders. Six (17%) of the PNES subjects who consented had either invalid or incomplete protocols and were not included in the final QOL analysis, whereas 17 (22%) ES subjects were excluded for these reasons. There was no difference in gender between subjects who declined versus those who participated, but those who declined tended to be older (p<.05). Conclusions: Methodological problems such as volunteer bias can compromise and limit the generalizability of clinical research findings. In our ongoing research addressing QOL in epilepsy and PNES, a concern was raised that PNES patients may have been declining participation at a higher rate than ES patients, which would have implications both for diagnostic and intervention studies with these populations. Our results demonstrate that the rates of participation and nonparticipation by PNES subjects in clinical research are not significantly different from ES subjects. Furthermore, the percentage of PNES subjects who consented but who could not be included due to invalid or incomplete protocols did not differ from that of ES subjects. These findings argue against a volunteer bias in research involving EMU-referred PNES patients.