Limits...
Feasibility of the "Bring Your Own Device" Model in Clinical Research: Results from a Randomized Controlled Pilot Study of a Mobile Patient Engagement Tool.

Pugliese L, Woodriff M, Crowley O, Lam V, Sohn J, Bradley S - Cureus (2016)

Bottom Line: The BYOD subgroup showed significantly greater engagement than study-provided phone users, as shown by higher application use frequency and duration over the study period.Although group differences in medication adherence decline were insignificant, the greater trend of decline in provisioned device users warrants further investigation to determine if trends reach significance over time.Significantly higher application engagement rates and effect of engagement on medication adherence in the BYOD subgroup similarly imply that greater application engagement may correlate to better medication adherence over time.

View Article: PubMed Central - HTML - PubMed

Affiliation: Innovation Research, HITLAB.

ABSTRACT

Background: Rising rates of smartphone ownership highlight opportunities for improved mobile application usage in clinical trials. While current methods call for device provisioning, the "bring your own device" (BYOD) model permits participants to use personal phones allowing for improved patient engagement and lowered operational costs. However, more evidence is needed to demonstrate the BYOD model's feasibility in research settings.

Objective: To assess if CentrosHealth, a mobile application designed to support trial compliance, produces different outcomes in medication adherence and application engagement when distributed through study-provisioned devices compared to the BYOD model.

Methods: 87 participants were randomly selected to use the mobile application or no intervention for a 28-day pilot study at a 2:1 randomization ratio (2 intervention: 1 control) and asked to consume a twice-daily probiotic supplement. The application users were further randomized into two groups: receiving the application on a personal "BYOD" or study-provided smartphone. In-depth interviews were performed in a randomly-selected subset of the intervention group (five BYOD and five study-provided smartphone users).

Results: The BYOD subgroup showed significantly greater engagement than study-provided phone users, as shown by higher application use frequency and duration over the study period. The BYOD subgroup also demonstrated a significant effect of engagement on medication adherence for number of application sessions (unstandardized regression coefficient beta=0.0006, p=0.02) and time spent therein (beta=0.00001, p=0.03). Study-provided phone users showed higher initial adherence rates, but greater decline (5.7%) than BYOD users (0.9%) over the study period. In-depth interviews revealed that participants preferred the BYOD model over using study-provided devices.

Conclusions: Results indicate that the BYOD model is feasible in health research settingsĀ and improves participant experience, calling for further BYOD model validity assessment. Although group differences in medication adherence decline were insignificant, the greater trend of decline in provisioned device users warrants further investigation to determine if trends reach significance over time. Significantly higher application engagement rates and effect of engagement on medication adherence in the BYOD subgroup similarly imply that greater application engagement may correlate to better medication adherence over time.

No MeSH data available.


Related in: MedlinePlus

Pill count ratio by study group and period
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4835151&req=5

FIG1: Pill count ratio by study group and period

Mentions: Study-provided phone users and BYOD users did not have a statistically significant difference in rate of medication adherence (p=0.20) as assessed by pill count at midline and endline. The study-provided phone group exhibited higher medication at midline compared to the BYOD group as shown in Figure 1. However, they also displayed a directional trend (p=0.20) of greater decline in medication adherence than the BYOD group from the midline to endline assessment periods, where the study-provided phone group saw a 5.7% decline in medication adherence from midline to endline, and the BYOD group saw a 0.9% decline in medication adherence over the same period. Table 1 describes pill count ratios for both groups at midline and endline.


Feasibility of the "Bring Your Own Device" Model in Clinical Research: Results from a Randomized Controlled Pilot Study of a Mobile Patient Engagement Tool.

Pugliese L, Woodriff M, Crowley O, Lam V, Sohn J, Bradley S - Cureus (2016)

Pill count ratio by study group and period
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4835151&req=5

FIG1: Pill count ratio by study group and period
Mentions: Study-provided phone users and BYOD users did not have a statistically significant difference in rate of medication adherence (p=0.20) as assessed by pill count at midline and endline. The study-provided phone group exhibited higher medication at midline compared to the BYOD group as shown in Figure 1. However, they also displayed a directional trend (p=0.20) of greater decline in medication adherence than the BYOD group from the midline to endline assessment periods, where the study-provided phone group saw a 5.7% decline in medication adherence from midline to endline, and the BYOD group saw a 0.9% decline in medication adherence over the same period. Table 1 describes pill count ratios for both groups at midline and endline.

Bottom Line: The BYOD subgroup showed significantly greater engagement than study-provided phone users, as shown by higher application use frequency and duration over the study period.Although group differences in medication adherence decline were insignificant, the greater trend of decline in provisioned device users warrants further investigation to determine if trends reach significance over time.Significantly higher application engagement rates and effect of engagement on medication adherence in the BYOD subgroup similarly imply that greater application engagement may correlate to better medication adherence over time.

View Article: PubMed Central - HTML - PubMed

Affiliation: Innovation Research, HITLAB.

ABSTRACT

Background: Rising rates of smartphone ownership highlight opportunities for improved mobile application usage in clinical trials. While current methods call for device provisioning, the "bring your own device" (BYOD) model permits participants to use personal phones allowing for improved patient engagement and lowered operational costs. However, more evidence is needed to demonstrate the BYOD model's feasibility in research settings.

Objective: To assess if CentrosHealth, a mobile application designed to support trial compliance, produces different outcomes in medication adherence and application engagement when distributed through study-provisioned devices compared to the BYOD model.

Methods: 87 participants were randomly selected to use the mobile application or no intervention for a 28-day pilot study at a 2:1 randomization ratio (2 intervention: 1 control) and asked to consume a twice-daily probiotic supplement. The application users were further randomized into two groups: receiving the application on a personal "BYOD" or study-provided smartphone. In-depth interviews were performed in a randomly-selected subset of the intervention group (five BYOD and five study-provided smartphone users).

Results: The BYOD subgroup showed significantly greater engagement than study-provided phone users, as shown by higher application use frequency and duration over the study period. The BYOD subgroup also demonstrated a significant effect of engagement on medication adherence for number of application sessions (unstandardized regression coefficient beta=0.0006, p=0.02) and time spent therein (beta=0.00001, p=0.03). Study-provided phone users showed higher initial adherence rates, but greater decline (5.7%) than BYOD users (0.9%) over the study period. In-depth interviews revealed that participants preferred the BYOD model over using study-provided devices.

Conclusions: Results indicate that the BYOD model is feasible in health research settingsĀ and improves participant experience, calling for further BYOD model validity assessment. Although group differences in medication adherence decline were insignificant, the greater trend of decline in provisioned device users warrants further investigation to determine if trends reach significance over time. Significantly higher application engagement rates and effect of engagement on medication adherence in the BYOD subgroup similarly imply that greater application engagement may correlate to better medication adherence over time.

No MeSH data available.


Related in: MedlinePlus