To the Editors:
Lists
Only you see this
Abstract
I am a young clinical investigator using the internet hourly on a daily basis. I believe that the Internet will play a major role in the establishment of a large network among epileptologists as well as recruitment of research participants in the future. The advantage of Internet-based research includes speedy data collection from geographically dispersed subjects with a small cost (Asano, 2007). One may assume that prospective data collection can be done online using self-report by participants, if “seizure-freedom” is the outcome of interest as suggested by Bergin et al. Such online self-report can potentially reduce the running cost of study and allow more frequent outcome measurements. A repeated measurement approach will be useful to study epilepsy syndromes of which sample size is limited. Here, I will propose a hypothetical study to compare the efficacies of two treatments in children with focal epilepsy associated with tuberous sclerosis complex. Comparisons will be made between two medical treatments or between surgical and further medical treatments. Participants who meet the study criteria will be randomized to each treatment group (Wiebe et al., 2001). Each guardian of a patient will report whether the patient is free from seizures or not. If a single binary outcome measurement is applied, the required sample size will be 93 in each group to detect a 20% difference in efficacy at a two-tailed significance level of 0.05 and a power of 0.80, supposing that the chance of seizure-freedom is expected to be 50% in one treatment and 30% in the alternative treatment. If a repeated measurement analysis is employed to the outcome measures reported four times per patient, the required sample size will be as small as 52 to 80 in each group, depending on the degree of clustering among observations within patient (Leon, 2004). It would take five years or more for a single institute to recruit more than 100 new patients with tuberous sclerosis complex. I believe that a multicenter program with a repeated measurement approach will increase the chance of successful recruitment of participants enough to complete the study. Is online data collection dependent on repeated self-reports feasible? I agree with the authors that some of the problems may not be apparent until somebody begins to test the approach. To explore what internet users think about the research program with health data collected via the Internet, I have recently conducted a small survey in Japanese on my personal blog (http://blogs.yahoo.co.jp/eishi_asano_md_phd/44835508.html). Concerns described by responders in this survey include a threat to confidential data and uncertain validity of self-reports. Nonresponse bias will be another concern in a web-based research as reported previously (Harewood et al., 2003; Partridge et al., 2004; Brod et al., 2006; Asano, 2007). I would like to introduce some of the potential solutions proposed by internet users. For example, it was proposed that a participant should be able to choose not only online- but also offline-data-entry according to her/his preference, offline-data can be saved on her/his ID card or cellular phone, and offline-data will be transferred to the main computer when she/he sees the reporting physician in hospital. It was also proposed that participants should not enter any identifiable personal data onto the research database, and that such personal data should be separately saved. It was also suggested that some forms of rewards and feedback may motivate patients to participate in the program for a long period. Regardless of the method of data collection, I hope that the web-based multicenter research program will be established by the epilepsy community. I am willing to participate in the program.
BibTeX
@inproceedings{asano2007i,
author = {Eishi Asano},
title = {{<i>To the Editors:</i>}},
booktitle = {Epilepsia},
year = {2007},
doi = {10.1111/j.1528-1167.2007.01060\_2.x},
}Loading related content...