Titel: Development of an instrument to measure general and seizure-related self-efficacy and mindfulness in people with seizures (SEMPS)
ID: FV 07
Art: Abstractvortrag
Session: Freie Vorträge II

Referent: Rosa Michaelis (Herdecke/DE)

Abstract - Text

Background: Mindfulness, general and seizure-related self-efficacy have been shown to be important mediators of quality of life in patients with seizures. The purpose of the present study is to summarize the development of a questionnaire which measures these concepts in people with seizures.

Methods: Epilepsy-specific questionnaires, generic questionnaires relating to these concepts and the outcome of previous qualitative research were reviewed to develop a candidate item set. In a quantitative-descriptive study, clinical experts assessed content validity. In a qualitative-descriptive study, cognitive debriefing was conducted with patients. The factorial structure and the confirmatory validity of this questionnaire were examined in a pilot study.

Results: The candidate item set comprised 28 items to probe the concepts of mindfulness, general and seizure-related self-efficacy. Seven clinical experts assessed content validity. Based on the initial content validity rating, 15 items remained unchanged, ten items underwent revisions, three items were excluded, and six items were added. Cognitive debriefing was conducted with 25 patients. Cognitive debriefing revealed problems regarding the interpretation of items and missing items. According to the patients' feedback, 24 items remained unchanged, five items were reworded, two items were excluded and four items were added. The resultant questionnaire consisted of 33 items; seven of these items included a follow-up question.

To examine the factorial structure and the confirmatory validity of this questionnaire, data was acquired from 31 patients. In default of the feasibility of the factorial analysis, the internal constancies of the subscales were examined and revealed acceptable results. The correlations with similar constructs were also unexpected in direction, but still significant. An analysis of the questionnaire on item-level revealed that there was a high number of inverted items in this questionnaire and a possible problem with the one-dimensionality of the subscales. Based on these results, five items were reworded which reversed the inversion in three items. One follow-up item was split in two items and another follow-up item was decoupled due to rewording of the item it had originally referred to. Four items were eliminated. Answering options were simplified in three items and four follow-up questions i.e. the original 5-point Likert-Scale was collapsed to a dichotomous answering format (yes/no).

Conclusion: Based on these results, the SEMPS seems to be valid in terms of content. Prior to its wider clinical or research application, the validity and reliability as well as the psychometric properties and factorial structure of the new questionnaire will need to be assessed in future clinical studies.