Can an Online Community Help Prevent Attrition from an Internet-Mediated Walking Program?
|A picture of two computers with gold men coming out of the screens, shaking hands.|
Richardson, C. R., Buis, L. R., Janney, A. W., Goodrich, D. E., Sen, A., Hess, M. L., et al. (2010). An online community improves adherence in an Internet-mediated walking program. Part 1: Results of a randomized controlled trial. Journal of Medical Internet Research, 12(4):e71.
Blythe Hiss, M.S., RCEP
While the Internet makes it easier and faster to get knowledge and information right to your doorstep, and with only lifting a finger (literally), it takes away one crucial piece of the pie, especially when we're talking about seeking behavior change - human contact. Yes, more information can get to me faster, but what if I want that personal support in order to put that information to good use? This study by Richardson, et al., took what we know about Internet-based behavior change interventions (that they can be widely disseminated and at low cost) and added a little piece of the pie (social support) back into the equation by allowing participants to communicate with each other in their own "online communities," therefore attempting to decrease attrition from an Internet-mediated walking program.
The population used in this study consisted of 324 sedentary and ambulatory adults who regularly e-mailed and had at least one of the following:
- BMI > 25 ("overweight")
- Type 2 diabetes
- Coronary heart disease
Exclusion criteria included pregnancy within one year, not being able to walk one block on their own, or not being able to make their own medicolegal decisions. Other than that, it does not seem that various other disabilities or activity limitations were identified and could possibly have been part of the cohort.
Participants in this 16-week randomized, controlled trial wore pedometers to upload step-count to the study server and could log-in to the study website to watch their progress, receive individually tailored motivational messages, and calculate weekly goals. One randomized group could post and read messages to their online community while the other group could not. The main outcomes were attrition and average daily step count.
Both groups significantly increased their average daily steps, and those who participated in the online communities stayed engaged in the walking program longer than those who did not. The percentage of those who uploaded at least 20 valid days of pedometer data during the final month of the 16-week intervention ("completers") was 13% higher in the group with the online community but the step-count increases between the two groups of "completers" were nearly identical; therefore, online community access did not significantly change the average daily step count of people who remained in the program. In addition, those who had lower baseline social support posted more messages and viewed more posts in the online community than those who reported higher social support at baseline.
Adding these online communities to the Internet-mediated walking program did not increase average daily step counts but did reduce attrition. In addition, participants who started out with lower social support at baseline used the online features more than those with initial high social support. These online communities may prove to be useful for targeting attrition from online programs and interventions seeking health behavior change, particularly in people who have low social support.
One of the barriers to physical activity that people with disabilities have reported in the literature is lack of support, so the findings suggest that people with disabilities could benefit from online communities. However, one of the limitations of the study is that baseline social support was measured with a single-item survey question that was designed to provide data to a message-tailoring algorithm rather than to actually measure social support, so analyzing this relationship must be done carefully. In addition, these communities were created specifically for this study and therefore weren't based on preexisting relationships or friendships or existing online community affiliations, which both could prove to be even more beneficial in future online programming.
The entire intervention was automated, with no face-to-face interaction whatsoever, which does increase its potential for a large volume of participants. The inclusion criteria were also somewhat broad, as described in the Participants section above, and this does strengthen its generalizability.
There were some additional limitations, as with any study, including the fact that (due to randomization) the online community group just happened to be more active at baseline than the group without an online community. The researchers suggested that stratified randomization could correct this issue. Also, since research staff had to contribute to the online community in order to create this "virtual reality", it may be possible that staff-generated content differed from content one would get from general participation. Length of the study (4 months) was also a limitation, as well as lack of information about activity levels when the pedometer was not worn.
This study is one of the first to use an RCT to document benefits of an online community. Previous studies (i.e., Glasgow et al., Takahashi et al.) have found no or possibly negative results from online communities, so these findings do contrast with many others, though is also consistent with some current findings that show positive results (i.e., Lorig et al.) from online community usage. This tells me . . . . that we still have a lot to learn
This study is actually only one in a series that took this Internet-mediated walking program and examined the effects of various components separately in order to help develop the evidence base for creating future similar interventions. Other components included participant safety, goal setting, and group competition on the effects of outcomes. I think I've found my research topic for next month . . .
Please send any questions or comments to Blythe Hiss at Blythe Hiss.