Sensitivity testing of the Seizure Severity Questionnaire (SSQ) Simon Borghs a, ⁎, Christine de la Loge b , Yves Brabant c , Joyce Cramer d a UCB Pharma, Slough, UK b UCB Pharma, Brussels, Belgium c Business and Decision Life Sciences, Brussels, Belgium d Houston, TX, USA abstract article info Article history: Received 22 July 2013 Revised 11 September 2013 Accepted 8 October 2013 Available online 22 November 2013 Keywords: Seizure severity Seizure Severity Questionnaire (SSQ) Sensitivity Patient-reported outcome Epilepsy The sensitivity of the Seizure Severity Questionnaire (SSQ) was evaluated using pooled data from open-label extensions of two clinical trials in patients with partial-onset seizures. The SSQ includes questions relating to frequency and helpfulness of warning signs as well as frequency, severity, and bothersomeness of ictal and postictal effects. Differences between mean change from baseline for each SSQ item for responders and non- responders were described and compared between patients solely with complex partial seizures (CPSs: responders, n = 166; nonresponders, n = 127) and those solely with secondarily generalized partial seizures (SGPSs: responders, n = 26; nonresponders, n = 24) at baseline. Seizure Severity Questionnaire total score and individual SSQ items related to ictal movement, consciousness, bothersomeness of postictal effects, and frequency of postictal emotional effects showed differentiation between seizure type responders. These data provide further validation of the SSQ by demonstrating its sensitivity in describing treatment effects. © 2013 Elsevier Inc. All rights reserved. 1. Introduction Clinical trials typically evaluate the efficacy of antiepileptic drugs (AEDs) based upon the extent to which seizure frequency is reduced [1,2]. A change in the severity of seizures is also a useful index of clinical effect; conversion to a less severe seizure type, diminished duration, and severity and/or bothersomeness of seizures and of postictal (after seizure) effects have practical clinical value, particularly for those pa- tients who are not able to achieve full seizure control [3]. Despite the importance of seizure severity as an aspect of epilepsy, validated and standardized instruments to assess seizure severity had been lacking [1,4,5]. The development of such an instrument requires an iterative process, building on initial conceptual model design and item/question generation through validity, reliability, and, ultimately, sensitivity testing [1,4,6]. The usefulness and validity of an instrument cannot be fully established without having demonstrated all of these aspects. Though a number of scales to assess seizure severity have been de- veloped over the past two decades (e.g., the VA Seizure Frequency and Severity Scale [VA Scale] [7], the Chalfont Seizure Severity Scale [8] later renamed to National Hospital Seizure Severity Scale [NHS3] [9], the Liverpool Seizure Severity Scale [LSSS] [10], and the Hague Seizure Severity Scale [HASS] [11]), limitations such as complexity in reporting, inappropriate weighting, a lack of distinction among seizure types, and, in all cases, a lack of confirmed sensitivity have hindered their use in both the clinic and in clinical trial settings [1]. Most recently, Cramer et al. described the development and initial reliability and validity testing of a new seizure severity scale (Seizure Severity Questionnaire [SSQ] [5]) with development based on current standards for patient-reported outcome instruments [6]. The SSQ was developed to evaluate seizure severity as a treatment response by characterizing changes in severity and bothersomeness of specific seizure characteristics (e.g., cognitive effects or shifts from complex to simple partial seizures). Based on the scale's characteristics and purpose, it would be expected to differentiate among seizure types. The SSQ, which was initiated in 1999, underwent initial evaluation for reliability and validity as reported in 2002 [5]. The instrument has since been increasingly used in large clinical trials to assess the severity of seizures. The final version of the SSQ is a 24-item questionnaire that includes questions for frequency and helpfulness of warning signs, se- verity and bothersomeness of ictal movement and loss of consciousness, and the frequency, severity, and bothersomeness of postictal cognitive, emotional, and physical effects [12]. Item scores range from 1 to 7, with higher scores indicating greater severity, except the warning item. An essential component of the overall validation of an instrument such as the SSQ is to establish sensitivity to clinical change following a therapeutic intervention. In this report, we describe the sensitivity of individual SSQ items and SSQ total score to clinical response according to seizure type using pooled data from open-label extensions of two clinical trials. Epilepsy & Behavior 31 (2014) 281–285 ⁎ Corresponding author at: UCB Pharma, 208 Bath Road, Slough, Berkshire SL1 3WE, UK. E-mail address: Simon.Borghs@ucb.com (S. Borghs). 1525-5050/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.yebeh.2013.10.010 Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh