**Self-Reporting Technologies for Supporting Epilepsy Treatment Treatment**

**Self-Reporting Technologies for Supporting Epilepsy** 

DOI: 10.5772/intechopen.70283

Jonathan Bidwell, Eliana Kovich, Cam Escoffery, Cherise Frazier and Elizabeth D. Mynatt Cam Escoffery, Cherise Frazier and Elizabeth D. Mynatt

Additional information is available at the end of the chapter Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/intechopen.70283

Jonathan Bidwell, Eliana Kovich,

#### **Abstract**

Epilepsy diagnosis and treatment relies heavily on patient self-reporting for informing clinical decision-making. These self-reports are traditionally collected from handwritten patient journals and tend to be either incomplete or inaccurate. Recent mobile and wearable health tracking developments stand to dramatically impact clinical practice through supporting patient and caregiver data collection activities. However, the specific types and characteristics of the data that clinicians need for patient care are not well known. In this study, we conducted interviews, a literature review, an expert panel, and online surveys to assess the availability and quality of patient-reported data that is useful but reported as being unavailable, difficult for patients to collect, or unreliable during epilepsy diagnosis and treatment, respectively. The results highlight important yet underexplored data collection and design opportunities for supporting the diagnosis, treatment, and self-management of epilepsy and expose notable gaps between clinical data needs and current patient practices.

**Keywords:** health tracking, patient self-reporting, clinical data indicators, neurocognitive, neurophysiological, implications for design

#### **1. Introduction**

Health tracking technologies such as wrist-worn seizure detection devices stand to play an increasingly important role in epilepsy diagnosis and treatment as tools that can assist patients with reporting physical [1] and psychiatric symptoms [2]. Neurologists rely on patient selfreporting to document a wide range of symptoms and triggers [3, 4] such as medication

Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons

intake, sleep, exercise, missed meals and stress levels. In practice, most patients struggle with reporting these types of information [5], and these reports are known to be highly inaccurate [6–10].

providers to collect high priority patient self-reports that are not well supported by current health reporting mechanisms and technologies. This research is unique in that we investigate both the information needs of neurologists during clinical treatment and the performance

Self-Reporting Technologies for Supporting Epilepsy Treatment

http://dx.doi.org/10.5772/intechopen.70283

71

Health information can also be challenging for patients to collect and therefore be unreliable even when available, or it may not be collected frequently enough to be informative [9, 11]. Handwritten and electronic patient "seizure diaries" tends to be either incomplete or inaccurate [5, 12]. In practice, many patients (30–50%) fail to report seizures during the day [6–9] while most patients fail to report seizures at night (85.8%) [9]. Eyewitness accounts often disagree on important details of how a seizure presents [9, 13], and observation is often difficult at night [5, 14]. Reminding patients to fill in reports may be ineffective as consciousness can be impaired both during and following a seizure [9]. These collective challenges present problems for clinicians as important patient information is often unavailable, unreliable, or not collected frequently enough to be informative during diagnosis and treatment [9, 11].

The **types** and **priorities** of patient self-reporting information are therefore important to

**1. Reporting availability:** Self-reports may not be available due to patient non-compliance, inability to document or observe the requested data [9], or limited awareness regarding the types of information that would be most relevant to collect in preparation for appointments.

**2. Reporting usefulness:** Self-reports may not be useful for clinical management. For example, patients may be unsure of what data to collect and fail to report important indicators [15] while devices may report data in a manner that requires considerable interpretation

**3. Reporting reliability:** Self-reports may not be reliable due to issues such as recall bias [16]. Moreover, self-reporting performance can be difficult to assess given the absence of readily

**4. Reporting difficulty:** Self-reporting may be too difficult or burdensome for patients to collect between appointments [17]. Neurologists often ask patients to document and report data such as the time, date, and a description of symptoms before, during, and after clinical

**5. Reporting frequency:** Finally, patient health data must be sampled or collected at an appropriate frequency for clinical interpretation. For example, clinicians require frequent and detailed seizure and developmental reports when treating patients with infantile

understand for informing care along with the following self-reporting **characteristics**:

statistics that technologists need for guiding development efforts.

**2. Related work**

**2.1. Patient self-reporting challenges**

for answering clinical questions.

presentations of symptoms [5, 18–20].

spasms [24].

quantified measurements or validated study designs [5].

These shortcomings present two sets of challenges for technology developers and healthcare providers alike. First, the type, priority, and characteristics of clinically relevant patient data are not well documented. This presents problems for technology developers, who must understand these needs for establishing appropriate technology design requirements. Second, the performance of health tracking devices is not well explored with respect to clinical seizure reporting needs. This presents problems for healthcare providers, who require an understanding of current reporting capabilities for recommending appropriate patient selfreporting tools.

In this chapter, we present our research with clinicians to first establish clinical patient selfreporting needs with respect to clinical decision-making during epilepsy diagnosis and treatment, and second to investigate the extent that the performance of current seizure detection and classification devices may be suitable for addressing these needs.

Our study included a literature review and interviews with clinicians to identify relevant epilepsy-related symptoms and triggers; a card sorting session to prioritize these symptoms and triggers; a technology review of seizure detection devices; and finally, a pair of online surveys was administered for establishing further characteristics of clinical patient self-reporting needs during epilepsy diagnosis and treatment, respectively.

Our results include a consensus on the types, priorities, and characteristics of clinical patient self-reporting needs during anti-epileptic drug (AED) selection and treatment along with a comparison of seizure detection devices and patient self-reporting performance. These findings are intended to provide a useful reference for both developing future patient self-reporting tools and to highlight the extent that current technologies may be suitable for addressing the clinical needs that we identified within the practice. Finally, we conclude with a discussion of technology recommendations that could stand to benefit mainstream epilepsy treatment.

The main contribution of this work is a "roadmap" for developing technologies that support epilepsy treatment. The results include:


Moreover, our study demonstrated the importance of stakeholder engagement. The role of clinical patient self-reporting is important yet often undocumented in literature. In our findings, we help to address this gap by presenting specific data collection strategies to help providers to collect high priority patient self-reports that are not well supported by current health reporting mechanisms and technologies. This research is unique in that we investigate both the information needs of neurologists during clinical treatment and the performance statistics that technologists need for guiding development efforts.
