JMIR Publications

ABSTRACT

Background:

Little is currently known regarding the feasibility of using a self-guided, remote, web-based platform as the basis for a longitudinal study of aging in community-dwelling older adults (OA’s). The current study describes the feasibility, and risk factors for participant drop out, using this approach as part of the web-based Louisiana aging brain study (web-LABrainS).

Objective:

This study used data from 402 participants in the web-LABrainS effort to determine the feasibility of using a self-guided, remote, and web-based platform as the basis for conducting longitudinal research in community-dwelling OA’s. Additionally, we sought to determine the risk factors associated with participant dropout over a 12-month period in web-LABrainS, and determine whether the same risk factors associated with dropout in in-clinic longitudinal studies were observed in web-LABrainS dropouts.

Methods:

Participants in web-LABrainS were enrolled on a rolling basis using grass roots and word of mouth promotional efforts. Participants used the web-LABrainS platform to provide electronic consent, demographic and health information, answer questionnaires, and complete assessments as part of a self-guided and web-based effort offsite of the clinic (remote). Following the completion of the baseline battery the participants were contacted by email every 6-months and asked to complete another round of the web-LABrainS battery. The data in the current study was collected from 402 participants, 217 of whom participated at baseline, 6-months, and 12-months (adherent participants) and 185 of whom participated in only the baseline web-LABrainS battery (dropout participants).

Results:

Our pilot studies indicate that even with limited participant outreach/retention efforts, it is feasible to conduct longitudinal clinical research studies in community-dwelling OA’s using a self-guided, remote, and web-based approach. In contrast to traditional in-clinic longitudinal studies, the dropouts from web-LABrainS were not observed to be significantly different with respect to with age, education, gender, marital status, or living alone. Similar to traditional longitudinal studies, dropouts in web-LABrainS had significantly higher use of depression medication, decreased self-reported mobility, and decreased memory (delayed recall). Interestingly, no differences in technology use, comfort with technology, time of assessment, or consent to be contacted about future research were observed between adherents and dropouts. Predictive binary logistic regression yielded a moderately accurate model and further supported a negative association between cognitive ability and dropout.

Conclusions:

Our studies provide some of the first clinical evidence for the feasibility of conducting longitudinal human research using a self-guided, remote, and web-based approach. Additionally, these data highlight the similarities and differences in key factors associated with participant dropout using this type of approach, as compared to traditional longitudinal study formats. The findings from this study may help guide the design and deployment of future longitudinal studies in OA’s that are focused on using self-guided, remote, and/or web-based approaches.