Efficacy of an Individually Tailored,Internet-Mediated Physical Activity Intervention in Older Adults: A Randomized Controlled Trial

Abstract

Objective: This study determined the effectiveness of an individually tailored, Internet-mediated physical activity (PA) intervention for increasing walking behavior in inactive older adults. Method: This 12-week randomly controlled intervention divided participants (N = 170) into three groups: control (CON, n = 51), pedometer only (PED, n = 62), and an individually tailored, Internet-mediated pedometer (TI-PED, n = 57) group. The PED group was instructed to increase weekly step count by 10% until 10,000 steps per day was achieved. The TI-PED group was given the same goal and received tailored feedback via an online platform. Changes in average step count pre-to-post were assessed. Results: Total retention rate at postintervention was 75.3%. PED (p < .001) and TI-PED (p < .001) increased step count pre-to-post, which was higher than the CON group at 12 weeks (PED, p < .001; TI-PED, p < .001). The TI-PED group had a higher step count at 12 weeks than the PED group (p < .001). Discussion: Individually tailored, Internet-mediated PA interventions are an effective way to significantly increase PA in older adults.

Keywords

web walking pedometer exercise elderly

Introduction

Despite the well-established benefits of regular physical activity (PA), older adults are less likely to meet the 2008 PA guidelines compared with younger adults (Centers for Disease Control and Prevention, 2014). Traditional PA intervention delivery methods fall short in reaching large numbers of individuals. In an attempt to extend this reach, Internet-mediated PA interventions have been created that target larger groups of people and have shown to effectively improve health behaviors such as walking (Marcus et al., 2006). Internet-based interventions provide a platform for quick and immediate interactions, as well as an opportunity for individually tailored feedback (Compernolle, Vandelanotte, Cardon, De Bourdeaudhuij, & De Cocker, 2015). When guided by behavioral theories, individually tailored feedback has shown to increase individual PA levels (Noar, Benac, & Harris, 2007).

Pedometers have been shown to be an effective intervention strategy to increase PA behavior in previously inactive adults (Kang, Marshall, Barreira, & Lee, 2009), with meta-analyses specifically showing the efficacy of both small weekly increases in step goals and 10,000-steps-per-day targets (Bravata et al., 2007). A number of studies have partnered pedometer feedback with Internet delivery, guided by behavioral theory within the general population with success to increase PA levels (Compernolle et al., 2015). Given the current changing demographics of older adults and their high inactivity levels, exploring the utility of far-reaching technological interventions such as through Internet delivery, employing behavioral theory, goals, and interaction are warranted. Thus, the purpose of this study was to compare the effectiveness of an individually tailored, Internet-mediated PA intervention to a standard PA intervention for increasing walking behavior in previously inactive older adults.

Method

Adults aged 55 to 80 years, who were inactive or insufficiently active, defined as achieving less than 7,500 steps per day (Tudor-Locke & Bassett, 2004), were recruited to participate in this study (see Figure 1 for study flow). In addition, participants were included if they had no orthopedic limitations to walking, expressed interest in starting a walking program, and had access to a computer and the Internet. Participants were recruited from the community via mass media announcements, mailed invitations, and recruitment postings. Procedures were approved by the Institutional Review Board (IRB) at a major Midwestern university (IRB Protocol Number 07.02.317). All measures were collected in a laboratory on the university campus. Upon completing the intervention, participants received a free pedometer and a US$50 honorarium.

Figure 1.

Screening criteria, randomization, and completion information for the study.

This was a 12-week randomized controlled trial. Baseline demographic and anthropometric measures were collected on the first day. Height and weight were assessed to calculate body mass index (BMI), and a health history questionnaire was administered. Participants (N = 170) were randomized into three different groups: control (CON, n = 51), pedometer only 10,000-step intervention group (PED, n = 62), and a tailored Internet-mediated pedometer intervention group (TI-PED, n = 57). The CON group was instructed to maintain their usual PA levels over the 12-week intervention period. Participants in the CON group were invited to participate in the website intervention following the initial 12 weeks as a benefit of participating in the study. The PED group was administered an Omron HJ-720ITC pedometer (Omron Corporation, Kyoto, Japan), shown previously to be valid in older adults (Dondzila, Swartz, Miller, Lenz, & Strath, 2012), and given the goal to increase their daily step count by 10% each week until they met 10,000 steps per day after which they were instructed to maintain 10,000 steps per day. Daily step counts were recorded on paper logs and mailed back to researchers each week. Ten thousand steps per day were prescribed, as it is an attainable goal, which can be objectively measured (i.e., via pedometer; Choi, Pak, & Choi, 2007). In addition, this step goal is feasible to incorporate into an active lifestyle with the goal of meeting moderate intensity PA guidelines (Choi et al., 2007).

The TI-PED group received the same pedometer and was instructed to log into a secure, interactive website weekly. The interactive website employed key strategies to increase PA systematically in older adults (Oman & King, 1998). These included education and goal setting (King, Taylor, Haskell, & Debusk, 1988), self-regulation, and frequent feedback and rewarding (Artinian et al., 2010; Chao, Foy, & Farmer, 2000; King et al., 1988). Briefly, the interactive platform required each subject to log in once per week with the intervention segregated into two phases. Phase 1, Weeks 1 to 3, was used to provide cognitive understanding of the benefits of PA, education on national recommendations, and self-awareness of current activity levels. Uploaded steps per day were graphically represented for each day for the prior week, and plotted against nationally recommended amounts. Phase 2, Weeks 4 to 12, continued to provide educational information, and now required each individual to, intrinsically, set manageable daily step targets for themselves, such as increasing weekly steps by 10% increments. Pedometer uploading took place at the end of each week. Graphical representations of daily steps were provided along with information on how well they corresponded with intrinsically set goals. At this stage of the program, each individual was either in compliance with set goals (defined as meeting walking step targets 5 out of 7 days) or they were not in compliance. If a participant was in compliance, the user was guided through a series of congratulatory screens and a directive for setting the upcoming week’s PA step goal. If the participant was not in compliance, the user was guided through a series of interactive screens that were designed to collect barriers to accomplishing the goal and then deliver motivational messages tagged and retrieved from a database library. The motivational messages were designed to offer strategies for overcoming user-identified barriers to attaining their PA step goal (example motivational messages can be seen in Table 1). Each week of the interactive program, users were also guided by an ongoing discussion forum, posing questions and solutions to increase PA, and access to “ask the expert” (a trained behaviorist and member of the research team) for any points of clarification, such as instructions on the website use. Once and if a participant reached the 10,000-steps-per-day goal, the interactive platform encouraged the user to maintain this level of activity through continued tailored messaging. Postintervention surveys were administered to the PED and TI-PED groups to garner brief overall study feedback (see Appendix). Specifically, the surveys inquired about the overall study enjoyment and usability of the Internet-mediated platform.

Table 1.

Baseline Socioeconomic and Anthropometric Characteristics of Participants.

	CON (n = 51)	PED (n = 62)	TI-PED (n = 57)	p
Age (SD)	66.1 (4.9)	68.3 (7.1)	67.4 (6.4)	.712
Female, %	78.6	78.5	81.3	.935
Race, %				.818
Caucasian	88.2	87.1	86.0
Hispanic	2.0	1.6	3.5
African American	9.8	11.3	10.5
Income, %				.456
<US$30,000	39.2	33.8	29.4
>US$30,000 to US$49,000	43.2	53.8	49.8
≥US$50,000	17.6	12.4	20.8
Education, %				.512
High school or less	10.6	9.2	11.5
Some college	32.4	44.6	48.1
College/Grad school	57.0	46.4	40.4
BMI, kg/m² (SD)	29.6 (3.1)	30.7 (4.3)	29.1 (3.6)	.824

Note. CON = control; PED = pedometer only; TI-PED = tailored Internet-mediated pedometer; BMI = body mass index.

Analyses of step counts across group and time were completed in R (R Core Team, 2016) using the nlme package (Pinheiro, Bates, DebRoy, Sarkar, & R Core Team, 2016) and multicomp package (Hothorn, Bretz, & Westfall, 2008). Missing postintervention data were assumed Missing at Random and were kept as missing in the multilevel modeling analysis. Model 1 was a random intercept model that included only the main effects of group and time. Model 2 included Group × Time interaction effects in addition to all Model 1 effects. Maximum likelihood method was used to obtain the estimates. The two models were compared with a likelihood ratio test to determine whether the interaction effects were present. Tukey’s post hoc analyses were used for multiple comparisons between groups and time points. Alpha was set at .05.

Results

There were no significant demographic differences at baseline between groups (Table 2). A total of 129 participants completed the 12-week protocol. The dropout rate for the CON group (37%) was higher than the PED (18%, p = .03) and TI-PED (19%, p = .05) groups. Demographic and baseline step count did not differ between the dropouts and study completers (p > .05). There were no significant differences in step count between the three groups at baseline (p > .05). There was a significant group, time, and Group × Time interaction (p < .0001, Figure 2). The CON group did not experience significant changes in step count from baseline (4,690 ± 1,475) to 12 weeks (4,654 ± 1,447, p = .999). The PED group significantly increased average step count by 62.1%, from 4,853 ± 1,455 steps at baseline to 7,869 ± 2,118 steps postintervention (p < .0001). The TI-PED group experienced a significant 119.4% increase in step count, from 4,688 ± 1,475 steps at baseline to 10,286 ± 3,022 steps postintervention (p < .0001). In addition, the PED group (p < .0001) and TI-PED group (p < .0001) had significantly higher step counts than the CON group at 12 weeks by 3,215 and 5,632 steps, respectively. Finally, the TI-PED group had significantly higher step counts after 12 weeks compared with the PED group (p < .0001).

Table 2.

Example Barriers and Motivational Messages From the Web Platform.

Example barriers	Example corresponding motivational messages
I did not have much energy to be active this week.	It is okay to feel tired some days, but remember, being physically active can actually increase the amount of energy you have. You can also try to be active in the morning when you have more energy, rather than leaving it to later in the day.
The weather was cold outside this week.	If you found the weather too cold outside this week, remember that you can walk inside. You could identify a local indoor area (such as a shopping mall), or even walk inside your home. For instance, walk around your home 10 times in the morning, afternoon, and evening—you will be surprised how many steps you accumulate.
My motivation was low this week to be physically active.	Is there a friend close by who you could invite to walk with you? You could both plan a daily walk, schedule it in your calendars.Try to schedule your physical activity early in the day—reward yourself if you accomplish your activity.You could try and not allow yourself to do something you like (such as read the newspaper, or watch the morning News Station) until after you have taken a walk.
I had a lot of other obligations this week, and struggled for time.	This is a common problem. Try and work physical activity into other obligations. Remember, if grocery shopping or running errands, walk a little further on purpose while doing so.If visiting friends, go for a walk and talk.If babysitting the grandkids, play with them outside, or better still, take them for a walk.If you miss a goal, don’t despair, try again tomorrow.

Figure 2.

Pre- and post-12-week intervention step count comparison.

PED postsurvey results revealed that 85% suggested the need for social interaction, and 71% reported that daily reminders would be helpful. TI-PED postsurvey results revealed that 82% enjoyed the web-mediated platform. Across both groups, 92% reported enjoying wearing the pedometer.

Discussion

This study compared the effect of a theoretically grounded, individually tailored, Internet-delivered daily step intervention to a traditionally guided 10,000-steps intervention group and nonintervention control group, in older adults. Key findings were that the TI-PED group significantly increased average daily step count more so than both the CON and PED groups, and that high levels of enjoyment were reported for the web-based intervention platform. The finding that an Internet-mediated, individually tailored PA intervention was successful in increasing daily step count supports previous literature (Moy et al., 2010; Richardson et al., 2007), but the present study extends those findings to specifically target inactive or insufficiently active older adults. In addition, we found that 82% of the TI-PED group reported high enjoyment utilizing the Internet-based platform, making this a feasible intervention option for this population. Alternatively, the PED group reported a desire for social interaction and daily reminders to engage in the walking activity, which could have been the possible cause for less of an increase in average daily step count. Still, it should be noted that the PED group was able to increase step counts without any specialized equipment, supporting what has been reported in previous literature (Kang et al., 2009).

This study is not without limitations. Our sample consisted primarily of females, who had some college experience or were college/graduate school graduates. This affects the application of results to the general, older adult population. During this study, we did not capture specific features of use pertaining to the interactive web platform; these data would have been beneficial to identify what specific features led to step increases. Despite these limitations, the results of this study show promise for future studies targeting PA in older adults by means of using interactive, web-based platforms. In addition, future studies should evaluate health outcomes resulting from the methods described in this study. In conclusion, these results show that individually tailored, web-based, interactive interventions are an effective way to increase PA in older adults.

Footnotes

Appendix

Author Contributions

S.J.S. conceptualized and planned the study, and supervised data collection, data analysis, and manuscript writing. T.W.R. drafted the manuscript and assisted in data analysis and interpretation. E.K.L. and N.E.M. carried out the intervention, managed the study, and assisted in manuscript revision and data interpretation. A.M.S. helped to plan the study and contributed to manuscript writing. H.M. carried out data management and data analysis and interpretation and assisted in manuscript writing.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant through the National Institute on Aging (5K01AG025962).

References

Artinian

N. T.

Fletcher

G. F.

Mozaffarian

Kris-Etherton

Van Horn

Lichtenstein

A. H.

, . . . American Heart Association Prevention Committee of the Council on Cardiovascular Nursing. (2010). Interventions to promote physical activity and dietary lifestyle changes for cardiovascular risk factor reduction in adults: A scientific statement from the American Heart Association. Circulation, 122, 406-441. doi:10.1161/CIR.0b013e3181e8edf1

Bravata

D. M.

Smith-Spangler

Sundaram

Gienger

A. L.

Lin

Lewis

. . . Sirard

J. R.

(2007). Using pedometers to increase physical activity and improve health: A systematic review. JAMA: Journal of the American Medical Association, 298, 2296-2304.

Centers for Disease Control and Prevention. (2014). Facts about physical activity. Retrieved from http://www.cdc.gov/physicalactivity/data/facts.htm

Chao

Foy

C. G.

Farmer

(2000). Exercise adherence among older adults: Challenges and strategies. Controlled Clinical Trials, 21(5, Suppl. 1), S212-S217.

Choi

B. C.

Pak

A. W.

Choi

J. C.

(2007). Daily step goal of 10,000 steps: A literature review. Clinical and Investigative Medicine, 30, 146-151.

Compernolle

Vandelanotte

Cardon

De Bourdeaudhuij

De Cocker

(2015). Effectiveness of a web-based, computer-tailored, pedometer-based physical activity intervention for adults: A cluster randomized controlled trial. Journal of Medical Internet Research, 17(2), e38.

Dondzila

C. J.

Swartz

A. M.

Miller

N. E.

Lenz

E. K.

Strath

S. J.

(2012). Accuracy of uploadable pedometers in laboratory, overground, and free-living conditions in young and older adults. International Journal of Behavioral Nutrition and Physical Activity, 11(9), Article 143.

Hothorn

Bretz

Westfall

(2008). Simultaneous inference in general parametric models. Biometrical Journal, 50, 346-363.

Kang

Marshall

S. J.

Barreira

T. V.

Lee

(2009). Effect of pedometer-based physical activity interventions: A meta-analysis. Research Quarterly for Exercise and Sport, 80, 648-655.

10.

King

A. C.

Taylor

C. B.

Haskell

W. L.

Debusk

R. F.

(1988). Strategies for increasing early adherence to and long-term maintenance of home-based exercise training in healthy middle-aged men and women. The American Journal of Cardiology, 61, 628-632.

11.

Marcus

B. H.

Williams

D. M.

Dubbert

P. M.

Sallis

J. F.

King

A. C.

Yancey

A. K.

, . . . Interdisciplinary Working Group on Quality of Care and Outcomes Research. (2006). Physical activity intervention studies: What we know and what we need to know: A scientific statement from the American Heart Association council on nutrition, physical activity, and metabolism (subcommittee on physical activity); council on cardiovascular disease in the young; and the interdisciplinary working group on quality of care and outcomes research. Circulation, 114, 2739-2752.

12.

Moy

M. L.

Janney

A. W.

Nguyen

H. Q.

Matthess

K. R.

Cohen

Garshick

Richardson

C. R.

(2010). Use of pedometer and Internet-mediated walking program in patients with chronic obstructive pulmonary disease. Journal of Rehabilitation Research & Development, 47, 485-496.

13.

Noar

S. M.

Benac

C. N.

Harris

M. S.

(2007). Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological Bulletin, 133, 673-693.

14.

Oman

R. F.

King

A. C.

(1998). Predicting the adoption and maintenance of exercise participation using self-efficacy and previous exercise participation rates. American Journal of Health Promotion, 12(3), 154-161.

15.

Pinheiro

Bates

DebRoy

Sarkar

, & R Core Team. (2016). nlme: Linear and nonlinear mixed effects models. Retrieved from http://CRAN.R-project.org/package=nlme

16.

R Core Team. (2016). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available from https://www.R-project.org/

17.

Richardson

C. R.

Mehari

K. S.

McIntyre

L. G.

Janney

A. W.

Fortlage

L. A.

Sen

. . . Piette

J. D.

(2007). A randomized trial comparing structured and lifestyle goals in an internet-mediated walking program for people with type 2 diabetes. International Journal of Behavioral Nutrition and Physical Activity, 4, Article 59.

18.

Tudor-Locke

Bassett

D. R.

Jr. (2004). How many steps/day are enough? Sports Medicine, 34(1), 1-8.