- Email: [email protected]
Patient and caregiver view on seizure detection devices: A survey study
Seizure 41 (2016) 179–181
Contents lists available at ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Short communication
Patient and caregiver view on seizure detection devices: A survey study Diego F. Tovar Quiroga a,*, Jeffrey W. Britton a, Elaine C. Wirrell a,b a b
Epilepsy Division, Department of Neurology, Mayo Clinic Rochester, USA Child and Adolescent Neurology, Mayo Clinic Rochester, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 June 2016 Received in revised form 2 August 2016 Accepted 15 August 2016
Purpose: Seizure detection devices (SDD) may reduce the potential for seizure-related injury, SUDEP or status epilepticus. We performed a survey of persons with epilepsy (PWE) and caregivers to assess their perspectives regarding the features and priorities that should be considered in the design of these devices. Methods: PWE/caregiver completed a survey which assessed the worry of undetected seizures, the impact of this concern on diurnal functioning/sleep and the level of interest in using SDD. Furthermore, questions regarding acceptable rates of false positive (FP)/negative (FN) alarms, acceptable time to generate an alarm and insurance coverage were asked. Results: 92 surveys were completed. Respondents expressed significant worry of undetected seizures. The impact of this concern on sleep and diurnal functioning was moderate. There is significant interest in using SDDs. Most would use SDD constantly. The acceptable FP-FN rate should be 25%. The time until caregivers are alerted should be 1 min. Regarding affordability, the majority would not use SDD unless covered by insurance and a few would use if not covered but affordable. Conclusions: The concern of undetected seizures is high among PWE and caregivers. Most expressed a high interest in using SDDs. Accuracy and affordability were key. ß 2016 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Seizure detection Devices Survey
1. Introduction The concern for sudden unexpected death in epilepsy (SUDEP) and the need for better seizure control in persons with epilepsy (PWE) have led to the implementation of seizure detection devices (SDD). Timely identification of seizures may lead to decreased SUDEP risk [1] and a reduction in the potential for seizure related injury and status epilepticus. Accurate detection of seizures and creation of alarms should improve monitoring of treatment efficacy, as patient reporting is not completely reliable [2–4]. Different methods for detection of biomarkers are under development [5–10]. There is no consensus on which system works for which seizure type or patient, and FDA approval is lacking for most.
Abbreviations: SDD, seizure detection device; PWE, persons with epilepsy; FN, false negative; FP, false positive. * Corresponding author at: 511, 6th Avenue SW, Rochester, MN 55902, USA. Fax: +1 5072664419. E-mail addresses: [email protected], [email protected] (D.F. Tovar Quiroga).
Understanding user needs and requirements is critical to product development. We performed a survey of PWE and their caregivers to assess their perspectives regarding the features and priorities that should be considered in design of seizure detection devices. 2. Methods This is a prospective non-interventional survey study approved by the institutional review board of Mayo Clinic and performed at a level 4 epilepsy center. Subjects: The study enrolled patients with uncontrolled epilepsy seen at our center. Adult and pediatric patients in the inpatient and outpatient settings participated. The patients were randomly selected. The survey was completed by the patient when capable or by family or caregivers if unable. The complete numbers of patients and families approached was not tabulated. Survey: The survey asked for current use of any type of SDD and the type of device being used. Additionally, it queried the degree of worry of undetected seizures, the impact of this worry on diurnal functioning, and the level of interest in using a SDD. These questions were answered using Likert scales ranging from 1 to 7, with lower
http://dx.doi.org/10.1016/j.seizure.2016.08.004 1059-1311/ß 2016 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
180
D.F. Tovar Quiroga et al. / Seizure 41 (2016) 179–181
values indicating less worry, less impact in daytime function/sleep, and less interest in using a device. Furthermore, questions regarding acceptable rates of FP and FN alarms, acceptable times until caregivers are alerted, and insurance coverage and cost were asked (Appendix C-Survey). Statistics: All statistical analyses were done using IBM1 SPSS1 Statistics version 22.
3.4. Acceptable rate of false positives (specificity) and false negatives (sensitivity)
3. Results
3.5. Maximal time to be alerted on cell phone
Ninety-two surveys were completed (27 pediatric, 65 adults) 42 by the patient and 50 by a family member or caregiver. Nearly half of subjects (n = 43) had seizures at least weekly [daily or more (n = 23), daily to weekly (n = 20), weekly to monthly (n = 24), every 6 months (n = 15), >every 6 months (n = 10)]. The majority of the patients (n = 85) had epilepsy refractory to 2 or more medications. All survey results summarized in Table 1 (Appendix B-Table 2, demographic information).
Most respondents (76.1%) reported an acceptable maximal time to be alerted on cell phone of 1minute, with majority (n = 39) preferring a responder time 30 s (Fig. 1C).
3.1. Degree of worry of undetected seizures Respondents expressed significant degree worry of the possibility of undetected seizures (Fig. 1A). The reported effect of this worry on the patient or caregiver sleep quality, as well as diurnal functioning, was moderate (Fig. 1B). 3.2. Interest in using a seizure detection device The majority of respondents manifested high interest in using a SDD. This interest correlated with the degree of worry of undetected seizures (Fig. 1A). A Spearman’s rank correlation was computed and there is moderate correlation between the degree of worry of undetected seizures and the level of interest in using a SDD (r = 0.489, p = <0.001). 3.3. Desired time for use of the device The majority of respondents (65.6%) indicated that the patient would use the device continuously, regardless of whether the patient is accompanied or not. Some indicated that the device [(Fig._1)TD$IG] would only be used when the patient is alone or sleeping (Fig. 1B).
The majority of respondents (76.1%) reported an expected FP detection rate 25% to be acceptable. Similarly, a majority (85.9%) reported an expected FN detection rate of 25% to be acceptable (Table 1).
3.6. Importance of insurance coverage Insurance coverage and cost were reported as determinant factors in the decision to use a seizure detector. Most respondents (67%) reported that the device would not be used unless covered by insurance and only a minority (6.6%) reported that they would use the device regardless of the cost or insurance coverage (Fig. 1D). 4. Discussion The objective of this study is to provide SDD developers with vetted insight on potential SDD users (patients and caregivers), including propensity of use and expectations around accessibility and design preferences. This study confirms that there is significant concern regarding undetected seizures and a high level of interest in using a SDD. Results suggest preference for continuous-use devices that can generate cell phone alerts in less than a minute after detection. Therefore, developers should envision smart designs that incorporate comfort and ease of use. Timely alerts to caregivers will prompt basic attention and administration of rescue medications. Although not specifically addressed by our survey, these devices should allow for effective communicate through different platforms and devices. These integrations could provide access to other tools which improve patient care. For example, Hope et al. [11] emphasized the need for not only seizure detection but also seizure registration. This would
Fig. 1. (A) Scatter plot showing correlation between worry of undetected seizures and interest in using SDD. (B) Preferred time for use of the device. (C) Reported maximum time to be alerted on cell phone. (D) Importance of insurance coverage and potential use of the device.
D.F. Tovar Quiroga et al. / Seizure 41 (2016) 179–181 Table 1 Survey results. MDN=Median, IQR=Interquartile range. Question
Result
Degree of worry of undetected seizures Effect of worry on sleep quality Effect of worry on diurnal function Interest in using a seizure detection device Desired time to use the device All the time When sleeping alone When alone (awake or asleep) Other Acceptable rate of false positive and false negative 0% 25% 50% 75% Other Maximum time to be alerted <30 s <1 min <3 min <5 min <15 min <30 min Other Importance of insurance coverage Very important (would not use unless covered) Would possibly use if not covered (unless cost is prohibitive) Would definitely use (even if not covered) Other
Mdn Mdn Mdn Mdn
6, 4, 4, 7,
IQR IQR IQR IQR
(5–7) (2–6) (2–5) (5–7)
n = 59 (65.6%) n = 15 (16.7%) n = 10 (11.1%) n = 6 (6.7%) False positive/false negative n = 7 (7.6%)/n = 25 (27.2%) n = 63 (68.5%)/n = 54 (58.7%) n = 9 (9.8%)/n = 4 (4.3%) n = 9 (9.8%)/n = 5 (5.4%) n = 4 (4.3%)/n = 4 (4.3%) n = 39 (42.4%) n = 31(33.7%) n = 13 (14.1%) n = 2 (2.2%) n = 1 (1.1%) n = 1 (1.1%) n = 4 (4.3%)
181
aim was to identify the basic needs of potential users of a SDD. Therefore, although the information from this survey does not provide holistic customer validation, it presents relevant information for the development and implementation of SDDs since it clarifies the views of patients and caregivers, and aids in incorporating their perspectives into a more user-focused design. Another limitation of the study is that it did not address the specific characteristics of the device. Other studies have addressed essential needs and desirable attributes of the ideal device including acceptable mode and site of sensors, availability of seizure prediction and the option of generating emergency calls as well as documentation of the seizures [11]. Nevertheless, no other study has addressed the questions that we asked to the potential users of such devices. Understanding the needs and expectations of the stakeholders is essential in the development of new technologies and this study helps fill that gap. 5. Conclusions
n = 6 (6.6%)
Wearable seizure detection devices are under development and the needs of potential customers should be considered in their design. PWE and their caregivers are interested in using these devices since there is significant worry and a real risk of SUDEP associated with undetected seizures. Data suggests a preference for these devices to be continuous-use, emphasizing the need for wearability or portability. These devices should generate an alarm in less than a minute after the seizure and be affordable or covered by insurance.
n = 2 (2.2%)
Conflict of interest statement
n = 61 (67%) n = 22 (24.2%)
None of the authors has conflict of interest to disclose. improve documentation rates which are ordinarily mediocre due to seizure induced unawareness. Our survey shows that overall, potential users of wearable SDDs expect moderately high (>75%) sensitivity and specificity. Interestingly, respondents gave more relevance to sensitivity than specificity, showing that the preference is to avoid false negative alarms. Published reports of alarm generating wearable SDDs have shown a wide range of sensitivity. A device using combined accelerometry and electrodermal activity for detection of convulsive seizures reported sensitivity of 88–94% and an average false detection rate of 1 seizure in 24 h [12]. Other devices also based on accelerometry, which currently on the market, have reported specificity of 2–41% for detection of different seizure types, including convulsive seizures [5,8]. The design of highly sensitive and specific wearable seizure detection is challenging since different seizure types can cause changes in various biomarkers, and the value of a SDD is not only in the capability of identifying convulsive episodes but also in the identification of clinically subtle seizures. Therefore, a highly accurate device should consider this variability and incorporate various seizure detection mechanisms into a single device. Accessibility was also a key factor. The majority indicated that a device would be used only if affordable or covered by insurance. Products for seizure detection currently available in the market range between US $200 and $600. For many, this is cost prohibitive. There is a need to validate the positive impact of seizure detection devices, not only for recognition of events that could lead to SUDEP, but also for seizure registration and monitoring. The latter would be useful during therapy changes and outcome documentation. Appropriate functionality validation is essential in paving the way for FDA approval, which will facilitate decisions regarding insurance coverage. This study did not include the view of other stakeholders such as medical professionals and other formal caregivers. However, the
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.seizure.2016.08.004. References [1] Van de Vel A, Cuppens K, Bonroy B, Milosevic M, Jansen K, Van Huffel S, et al. Non-EEG seizure-detection systems and potential SUDEP prevention: state of the art. Seizure 2013;22([30_TD$IF]5):345–55. [2] Blum DE, Eskola J, Bortz JJ, Fisher RS. Patient awareness of seizures. Neurology 1996;47([32_TD$IF]1):260–4. [3] Fisher RS, Blum DE, DiVentura B, Vannest J, Hixson JD, Moss R, et al. Seizure diaries for clinical research and practice: limitations and future prospects. Epilepsy Behav 2012;24([34_TD$IF]3):304–10. [4] Hoppe C, Poepel A, Elger CE. Epilepsy: accuracy of patient seizure counts. Arch Neurol 2007;64([35_TD$IF]11):1595–9. [5] Patterson AL, Mudigoudar B, Fulton S, McGregor A, Poppel KV, Wheless MC, et al. SmartWatch by SmartMonitor: [37_TD$IF]Assessment of [38_TD$IF]Seizure [39_TD$IF]Detection [40_TD$IF]Efficacy for [41_TD$IF]Various [38_TD$IF]Seizure [42_TD$IF]Types in [43_TD$IF]Children, a [4_TD$IF]Large [45_TD$IF]Prospective [46_TD$IF]Single-[47_TD$IF]Center [48_TD$IF]Study. Pediatr Neurol 2015;53([49_TD$IF]4):309–11. [6] Jeppesen J, Beniczky S, Johansen P, Sidenius P, Fuglsang-Frederiksen A. Exploring the capability of wireless near infrared spectroscopy as a portable seizure detection device for epilepsy patients. Seizure 2015;26:43–8. [7] Cogan D, Nourani M, Harvey J, Nagaraddi V. Epileptic seizure detection using wristworn biosensors. Conf Proc IEEE Eng Med Biol Soc 2015;2015: 5086–9. [8] Van de Vel A, Verhaert K, Ceulemans B. Critical evaluation of four different seizure detection systems tested on one patient with focal and generalized tonic and clonic seizures. Epilepsy Behav 2014;37:91–4. [9] Osorio I. Automated seizure detection using EKG. Int J Neural Syst 2014; 24([52_TD$IF]2):1450001. [10] Beniczky S, Polster T, Kjaer TW, Hjalgrim H. Detection of generalized tonic[54_TD$IF]clonic seizures by a wireless wrist accelerometer: a prospective, multicenter study. Epilepsia 2013;54([5_TD$IF]4):e58–61. [11] Hoppe C, Feldmann M, Blachut B, Surges R, Elger CE, Helmstaedter C. Novel techniques for automated seizure registration: [57_TD$IF]Patients’ wants and needs. Epilepsy Behav 2015;52([58_TD$IF]Pt A):1–7. [12] Poh MZ, Loddenkemper T, Reinsberger C, Swenson NC, Goyal S, Sabtala MC, et al. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor. Epilepsia 2012;53([60_TD$IF]5):e93–7.
Contents lists available at ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Short communication
Patient and caregiver view on seizure detection devices: A survey study Diego F. Tovar Quiroga a,*, Jeffrey W. Britton a, Elaine C. Wirrell a,b a b
Epilepsy Division, Department of Neurology, Mayo Clinic Rochester, USA Child and Adolescent Neurology, Mayo Clinic Rochester, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 June 2016 Received in revised form 2 August 2016 Accepted 15 August 2016
Purpose: Seizure detection devices (SDD) may reduce the potential for seizure-related injury, SUDEP or status epilepticus. We performed a survey of persons with epilepsy (PWE) and caregivers to assess their perspectives regarding the features and priorities that should be considered in the design of these devices. Methods: PWE/caregiver completed a survey which assessed the worry of undetected seizures, the impact of this concern on diurnal functioning/sleep and the level of interest in using SDD. Furthermore, questions regarding acceptable rates of false positive (FP)/negative (FN) alarms, acceptable time to generate an alarm and insurance coverage were asked. Results: 92 surveys were completed. Respondents expressed significant worry of undetected seizures. The impact of this concern on sleep and diurnal functioning was moderate. There is significant interest in using SDDs. Most would use SDD constantly. The acceptable FP-FN rate should be 25%. The time until caregivers are alerted should be 1 min. Regarding affordability, the majority would not use SDD unless covered by insurance and a few would use if not covered but affordable. Conclusions: The concern of undetected seizures is high among PWE and caregivers. Most expressed a high interest in using SDDs. Accuracy and affordability were key. ß 2016 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Seizure detection Devices Survey
1. Introduction The concern for sudden unexpected death in epilepsy (SUDEP) and the need for better seizure control in persons with epilepsy (PWE) have led to the implementation of seizure detection devices (SDD). Timely identification of seizures may lead to decreased SUDEP risk [1] and a reduction in the potential for seizure related injury and status epilepticus. Accurate detection of seizures and creation of alarms should improve monitoring of treatment efficacy, as patient reporting is not completely reliable [2–4]. Different methods for detection of biomarkers are under development [5–10]. There is no consensus on which system works for which seizure type or patient, and FDA approval is lacking for most.
Abbreviations: SDD, seizure detection device; PWE, persons with epilepsy; FN, false negative; FP, false positive. * Corresponding author at: 511, 6th Avenue SW, Rochester, MN 55902, USA. Fax: +1 5072664419. E-mail addresses: [email protected], [email protected] (D.F. Tovar Quiroga).
Understanding user needs and requirements is critical to product development. We performed a survey of PWE and their caregivers to assess their perspectives regarding the features and priorities that should be considered in design of seizure detection devices. 2. Methods This is a prospective non-interventional survey study approved by the institutional review board of Mayo Clinic and performed at a level 4 epilepsy center. Subjects: The study enrolled patients with uncontrolled epilepsy seen at our center. Adult and pediatric patients in the inpatient and outpatient settings participated. The patients were randomly selected. The survey was completed by the patient when capable or by family or caregivers if unable. The complete numbers of patients and families approached was not tabulated. Survey: The survey asked for current use of any type of SDD and the type of device being used. Additionally, it queried the degree of worry of undetected seizures, the impact of this worry on diurnal functioning, and the level of interest in using a SDD. These questions were answered using Likert scales ranging from 1 to 7, with lower
http://dx.doi.org/10.1016/j.seizure.2016.08.004 1059-1311/ß 2016 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
180
D.F. Tovar Quiroga et al. / Seizure 41 (2016) 179–181
values indicating less worry, less impact in daytime function/sleep, and less interest in using a device. Furthermore, questions regarding acceptable rates of FP and FN alarms, acceptable times until caregivers are alerted, and insurance coverage and cost were asked (Appendix C-Survey). Statistics: All statistical analyses were done using IBM1 SPSS1 Statistics version 22.
3.4. Acceptable rate of false positives (specificity) and false negatives (sensitivity)
3. Results
3.5. Maximal time to be alerted on cell phone
Ninety-two surveys were completed (27 pediatric, 65 adults) 42 by the patient and 50 by a family member or caregiver. Nearly half of subjects (n = 43) had seizures at least weekly [daily or more (n = 23), daily to weekly (n = 20), weekly to monthly (n = 24), every 6 months (n = 15), >every 6 months (n = 10)]. The majority of the patients (n = 85) had epilepsy refractory to 2 or more medications. All survey results summarized in Table 1 (Appendix B-Table 2, demographic information).
Most respondents (76.1%) reported an acceptable maximal time to be alerted on cell phone of 1minute, with majority (n = 39) preferring a responder time 30 s (Fig. 1C).
3.1. Degree of worry of undetected seizures Respondents expressed significant degree worry of the possibility of undetected seizures (Fig. 1A). The reported effect of this worry on the patient or caregiver sleep quality, as well as diurnal functioning, was moderate (Fig. 1B). 3.2. Interest in using a seizure detection device The majority of respondents manifested high interest in using a SDD. This interest correlated with the degree of worry of undetected seizures (Fig. 1A). A Spearman’s rank correlation was computed and there is moderate correlation between the degree of worry of undetected seizures and the level of interest in using a SDD (r = 0.489, p = <0.001). 3.3. Desired time for use of the device The majority of respondents (65.6%) indicated that the patient would use the device continuously, regardless of whether the patient is accompanied or not. Some indicated that the device [(Fig._1)TD$IG] would only be used when the patient is alone or sleeping (Fig. 1B).
The majority of respondents (76.1%) reported an expected FP detection rate 25% to be acceptable. Similarly, a majority (85.9%) reported an expected FN detection rate of 25% to be acceptable (Table 1).
3.6. Importance of insurance coverage Insurance coverage and cost were reported as determinant factors in the decision to use a seizure detector. Most respondents (67%) reported that the device would not be used unless covered by insurance and only a minority (6.6%) reported that they would use the device regardless of the cost or insurance coverage (Fig. 1D). 4. Discussion The objective of this study is to provide SDD developers with vetted insight on potential SDD users (patients and caregivers), including propensity of use and expectations around accessibility and design preferences. This study confirms that there is significant concern regarding undetected seizures and a high level of interest in using a SDD. Results suggest preference for continuous-use devices that can generate cell phone alerts in less than a minute after detection. Therefore, developers should envision smart designs that incorporate comfort and ease of use. Timely alerts to caregivers will prompt basic attention and administration of rescue medications. Although not specifically addressed by our survey, these devices should allow for effective communicate through different platforms and devices. These integrations could provide access to other tools which improve patient care. For example, Hope et al. [11] emphasized the need for not only seizure detection but also seizure registration. This would
Fig. 1. (A) Scatter plot showing correlation between worry of undetected seizures and interest in using SDD. (B) Preferred time for use of the device. (C) Reported maximum time to be alerted on cell phone. (D) Importance of insurance coverage and potential use of the device.
D.F. Tovar Quiroga et al. / Seizure 41 (2016) 179–181 Table 1 Survey results. MDN=Median, IQR=Interquartile range. Question
Result
Degree of worry of undetected seizures Effect of worry on sleep quality Effect of worry on diurnal function Interest in using a seizure detection device Desired time to use the device All the time When sleeping alone When alone (awake or asleep) Other Acceptable rate of false positive and false negative 0% 25% 50% 75% Other Maximum time to be alerted <30 s <1 min <3 min <5 min <15 min <30 min Other Importance of insurance coverage Very important (would not use unless covered) Would possibly use if not covered (unless cost is prohibitive) Would definitely use (even if not covered) Other
Mdn Mdn Mdn Mdn
6, 4, 4, 7,
IQR IQR IQR IQR
(5–7) (2–6) (2–5) (5–7)
n = 59 (65.6%) n = 15 (16.7%) n = 10 (11.1%) n = 6 (6.7%) False positive/false negative n = 7 (7.6%)/n = 25 (27.2%) n = 63 (68.5%)/n = 54 (58.7%) n = 9 (9.8%)/n = 4 (4.3%) n = 9 (9.8%)/n = 5 (5.4%) n = 4 (4.3%)/n = 4 (4.3%) n = 39 (42.4%) n = 31(33.7%) n = 13 (14.1%) n = 2 (2.2%) n = 1 (1.1%) n = 1 (1.1%) n = 4 (4.3%)
181
aim was to identify the basic needs of potential users of a SDD. Therefore, although the information from this survey does not provide holistic customer validation, it presents relevant information for the development and implementation of SDDs since it clarifies the views of patients and caregivers, and aids in incorporating their perspectives into a more user-focused design. Another limitation of the study is that it did not address the specific characteristics of the device. Other studies have addressed essential needs and desirable attributes of the ideal device including acceptable mode and site of sensors, availability of seizure prediction and the option of generating emergency calls as well as documentation of the seizures [11]. Nevertheless, no other study has addressed the questions that we asked to the potential users of such devices. Understanding the needs and expectations of the stakeholders is essential in the development of new technologies and this study helps fill that gap. 5. Conclusions
n = 6 (6.6%)
Wearable seizure detection devices are under development and the needs of potential customers should be considered in their design. PWE and their caregivers are interested in using these devices since there is significant worry and a real risk of SUDEP associated with undetected seizures. Data suggests a preference for these devices to be continuous-use, emphasizing the need for wearability or portability. These devices should generate an alarm in less than a minute after the seizure and be affordable or covered by insurance.
n = 2 (2.2%)
Conflict of interest statement
n = 61 (67%) n = 22 (24.2%)
None of the authors has conflict of interest to disclose. improve documentation rates which are ordinarily mediocre due to seizure induced unawareness. Our survey shows that overall, potential users of wearable SDDs expect moderately high (>75%) sensitivity and specificity. Interestingly, respondents gave more relevance to sensitivity than specificity, showing that the preference is to avoid false negative alarms. Published reports of alarm generating wearable SDDs have shown a wide range of sensitivity. A device using combined accelerometry and electrodermal activity for detection of convulsive seizures reported sensitivity of 88–94% and an average false detection rate of 1 seizure in 24 h [12]. Other devices also based on accelerometry, which currently on the market, have reported specificity of 2–41% for detection of different seizure types, including convulsive seizures [5,8]. The design of highly sensitive and specific wearable seizure detection is challenging since different seizure types can cause changes in various biomarkers, and the value of a SDD is not only in the capability of identifying convulsive episodes but also in the identification of clinically subtle seizures. Therefore, a highly accurate device should consider this variability and incorporate various seizure detection mechanisms into a single device. Accessibility was also a key factor. The majority indicated that a device would be used only if affordable or covered by insurance. Products for seizure detection currently available in the market range between US $200 and $600. For many, this is cost prohibitive. There is a need to validate the positive impact of seizure detection devices, not only for recognition of events that could lead to SUDEP, but also for seizure registration and monitoring. The latter would be useful during therapy changes and outcome documentation. Appropriate functionality validation is essential in paving the way for FDA approval, which will facilitate decisions regarding insurance coverage. This study did not include the view of other stakeholders such as medical professionals and other formal caregivers. However, the
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.seizure.2016.08.004. References [1] Van de Vel A, Cuppens K, Bonroy B, Milosevic M, Jansen K, Van Huffel S, et al. Non-EEG seizure-detection systems and potential SUDEP prevention: state of the art. Seizure 2013;22([30_TD$IF]5):345–55. [2] Blum DE, Eskola J, Bortz JJ, Fisher RS. Patient awareness of seizures. Neurology 1996;47([32_TD$IF]1):260–4. [3] Fisher RS, Blum DE, DiVentura B, Vannest J, Hixson JD, Moss R, et al. Seizure diaries for clinical research and practice: limitations and future prospects. Epilepsy Behav 2012;24([34_TD$IF]3):304–10. [4] Hoppe C, Poepel A, Elger CE. Epilepsy: accuracy of patient seizure counts. Arch Neurol 2007;64([35_TD$IF]11):1595–9. [5] Patterson AL, Mudigoudar B, Fulton S, McGregor A, Poppel KV, Wheless MC, et al. SmartWatch by SmartMonitor: [37_TD$IF]Assessment of [38_TD$IF]Seizure [39_TD$IF]Detection [40_TD$IF]Efficacy for [41_TD$IF]Various [38_TD$IF]Seizure [42_TD$IF]Types in [43_TD$IF]Children, a [4_TD$IF]Large [45_TD$IF]Prospective [46_TD$IF]Single-[47_TD$IF]Center [48_TD$IF]Study. Pediatr Neurol 2015;53([49_TD$IF]4):309–11. [6] Jeppesen J, Beniczky S, Johansen P, Sidenius P, Fuglsang-Frederiksen A. Exploring the capability of wireless near infrared spectroscopy as a portable seizure detection device for epilepsy patients. Seizure 2015;26:43–8. [7] Cogan D, Nourani M, Harvey J, Nagaraddi V. Epileptic seizure detection using wristworn biosensors. Conf Proc IEEE Eng Med Biol Soc 2015;2015: 5086–9. [8] Van de Vel A, Verhaert K, Ceulemans B. Critical evaluation of four different seizure detection systems tested on one patient with focal and generalized tonic and clonic seizures. Epilepsy Behav 2014;37:91–4. [9] Osorio I. Automated seizure detection using EKG. Int J Neural Syst 2014; 24([52_TD$IF]2):1450001. [10] Beniczky S, Polster T, Kjaer TW, Hjalgrim H. Detection of generalized tonic[54_TD$IF]clonic seizures by a wireless wrist accelerometer: a prospective, multicenter study. Epilepsia 2013;54([5_TD$IF]4):e58–61. [11] Hoppe C, Feldmann M, Blachut B, Surges R, Elger CE, Helmstaedter C. Novel techniques for automated seizure registration: [57_TD$IF]Patients’ wants and needs. Epilepsy Behav 2015;52([58_TD$IF]Pt A):1–7. [12] Poh MZ, Loddenkemper T, Reinsberger C, Swenson NC, Goyal S, Sabtala MC, et al. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor. Epilepsia 2012;53([60_TD$IF]5):e93–7.