This was a pre-post design study with 2-3 months of follow-up. Online surveys and a mobile-enabled web app were used to collect participant data.

All participants were recruited online. Through our collaborative partnership with a cancer survivor's organization, emails were sent out to their list-serve inviting eligible cancer survivors to participate. Inclusion criteria for participation included (1) ≥60 years, (2) having ever been diagnosed with cancer, (3) reporting the ability to do physical activity, and (4) having access to the Internet. Potential participants were instructed to the iCanFit website [21]. On the homepage, there was a screening survey for potential participants. Individuals who completed the screening survey (including the provision of contact information) received a phone call from our research staff to re-verify their eligibility. Eligible participants received instruction on how to participate in the study. Instructional videos on how to use iCanfit were also available on the “Help” page.

Project information was sent out to participants via email with a link to an online survey. Prior to the survey, an informed consent with a detailed project description, benefits, and potential risks of participation was provided. Participants who completed the informed consent were directed to the baseline survey, which took approximately 10 minutes to complete. After the survey, participants were directed to the project site and instructed to create a username and password. The “Help” page also contained instructional videos on how to use the web app. All participants were instructed to use iCanFit for about 8-12 weeks. During that time, 4-6 emails were sent to participants reminding them to continue using iCanFit. The emails were sent automatically and contained personalized information such as participant’s name, duration of the study, and when the follow-up survey would be sent. After 8-12 weeks, participants were sent a link to the follow-up survey, which took approximately 10 minutes to complete. Participants who completed the baseline survey received a $15 gift card and those who completed the follow-up survey received an additional $35 gift card. The study protocol was approved by the Institutional Review Board (IRB) of the Texas A&M University.

The baseline survey included participant demographics, use of mobile tools, quality of life, and current physical activity. The quality of life was measured with 7 items, including self-rated health (1 being poor and 5 being excellent), overall quality of life, fatigue, pain, shortness of breath, stress, and sleep. The response options ranged from 0-10, with 0 being the worst and 10 the best. The Cronbach alpha value for the 7 quality of life items was .79. We also asked the participants how many days in the past 30 days did they not have good physical health, good mental health, or could not do usual activities.

Participants were asked about their current engagement in regular physical activity with the following 5 response options (1) not engaged in physical activity and have no plan of doing so, (2) not engaged but plan to do so in 3 months or less, (3) engaged occasionally but not on a regular basis, (4) engaged in regular physical activity but started less than 3 months, and (5) engaged in regular physical activity and has been doing so for more than 3 months.

In the follow-up survey, participants were asked about the same outcomes on quality of life and engagement in regular physical activity, as well as their experience with the program. Some open-ended questions were included to solicit their feedback and suggestions for iCanFit. In addition to the above self-reported subjective data, users’ activities on iCanFit including the number of goals set and achieved were captured by the web app.

All data from the online surveys were downloaded and stored in SPSS 22.0. Descriptive statistics were used for data analysis; paired t test was used to compare pre-post changes and assess level of significance at the P<.05 level. Open-ended responses were entered into Microsoft Word to identify ranges and patterns of responses.

Of the total eligible participants who created an online account and completed the baseline survey (N=30), 87% (26/30) completed the follow-up survey. These 26 participants were the sample to test the efficacy of iCanFit.

As shown in Multimedia Appendix 1, the median age of the 26 participants in the baseline sample was 69 years (range 60-78). About 70% (18/26) of them were female, 73% (19/26) were white, 77% were married (20/26), 38% (10/26) had some college education, and 42% (11/26) had completed college. Only 2 participants (7%, 2/26) were still in cancer treatment, while the remaining majority weremajority was not, including 46% (12/26) that were being monitored but not in treatment, and 46% (12/26) that were post-treatment survivors. All participants had other chronic conditions, including high blood pressure (46%, 12/26), heart disease (23%, 2/26), arthritis (31%, 8/26), osteoporosis (23%, 6/26), sciatica (12%, 3/26), diabetes (12%, 3/26), chronic back pain (12%, 3/26), and depression or anxiety (8%, 2/26). About 46% (12/26) of participants talked to their health care providers about physical activity almost every time they saw the provider, 27% (7/26) said sometimes they did, and 23% (6/26) said they rarely or never did.

All of the participants had high-speed Internet access at home, and most of them used multiple mobile tools for Internet access. Specifically, 69% (18/26) used a desktop, 46% (12/26) used a laptop, 46% (12/26) used a tablet, and 81% (21/26) used a mobile phone with app capabilities; and as many as 23% (6/23) of participants used a mobile phone for Internet access every day. Their weekly online hours varied from 1-60 hours (median 10 hours). Most older cancer survivors frequently searched health information online, for example, only 1 person (4%, 1/26) never searched health information online, 8% (3/26) did so twice a month, 19% (2/26) did it once a week, 27% (5/26) did it 2-3 times a week, and 15% (6/26) did it every day. Most participants were active users of social media. For instance, 65% (17/26) of them were on Facebook, 50% (13/26) subscribed to an email listserv, 31% (15/26) used LinkedIn, and 19% (5/26) used an online messenger.

The differences in quality of life and physical activity at baseline and follow-up are shown in Table 1. The mean self-rated health score (1-5) increased from 3.23 to 3.81, the overall quality of life score increased from 8.12 to 8.50, fatigue-related quality of life score increased from 6.04 to 6.77, pain-related quality of life score increased from 8.85 to 9.11, shortness of breath-related quality of life score increased from 8.65 to 9.04, stress-related quality of life score increased from 6.27 to 7.04, and sleep quality also increased from 6.19 to 7.00. All of the quality of life-related changes were significant (P<.05). The number of days without good physical health, good mental health, or reporting they could not do usual activities did not have significant changes due to flooring effects of the numbers.

In terms of self-reported physical activities at baseline, 3 participants (12%, 3/26) were not engaged in physical activity and had no plan of doing so and 5 participants (19%, 5/26) were not engaged in physical activity but planned to do so in less than 3 months. All of these 8 participants were engaged in physical activity in the follow-up. The number of participants engaged in regular physical activity also increased from 11 (42%, 11/26) to 15 (58%, 15/26).

Participants’ use of and experience with iCanFit is depicted in Table 2. Most participants accessed iCanFit with multiple mobile devices including 58% (15/26) via desktop, 38% (10/26) via laptop, 23% (6/26) via tablet, and 27% (7/26) via mobile phones. Participants used iCanFit with varying frequencies; 12% (3/26) used it once every 2 weeks, 62% (16/26) used it once a week, 19% (5/26) used it 2-3 times per week, and 8% (2/26) used it 4-5 times a week. Their weekly median time on iCanFit was 25 minutes (range 5-60). About 65% (17/26) of participants had talked to their family and friends about iCanFit, and 27% (7/26) used iCanFit with their family or friends.

Table 2 also demonstrates that participants reported an overall positive experience with iCanFit. On the scale of 0-5, the mean (SD) value of overall ease of using iCanFit was 4.1 (1.7), 3.9 (2.1) for "Goals", 4.2 (0.9) for "Healthy Tips", 4.5 (0.8) for "Resources", and 4.1 (0.9) for "Community".

Participants also shared their experience of using iCanFit. Most participants made comments like “Great program, it helps me track my exercise.” Their favorite functions included the graphic display of their progress and the personalized feedback of their physical activities. For example, one shared “I like it sends me feedback right after I enter my activities,” and “It’s great to see a bar chart comparing my goals and my activities.” Participants who did not exercise regularly or have an exercise plan prior to the intervention reported using iCanFit to set goals and track their activities. For example, one participant reported, “It motivates me to exercise more so I can meet or exceed my goals,” and “After some weeks of using it, I know how much I can do on a weekly basis.”

Participants also shared suggestions on further improving the programs. Nearly half of the participants had frustration with forgetting to enter activities. “I often forget to enter my activities, and at the end of the week, I would totally forget what I’ve done,” and “It’d be better if I don’t need to enter my activity because I easily forget.” Some suggested using automatic function; for example, a participant had proposed a pedometer “If you could link my pedometer to iCanFit, then I don’t need to enter activity.” Participants also commented that they mostly used the "Goals" function and didn’t use the other functions (Healthy tips, Resources and Community) very often.

Within2-3 months of using iCanFit, participants set a total of 289 goals, with a mean of 11.1 per participant. Nearly half of these goals were met. The most typical activities included walking, jogging, aerobics, and gardening.

To our knowledge, there has been little systematic research of mobile-enabled, Web-based apps for older cancer survivors. As one of the first pilot studies, the initial efficacy of iCanFit as a web app has been demonstrated in this study. Significant improvement in quality of life and engagement in physical activity all indicate that this, and similar mobile-enabled, web apps may, be able to have positive effects for older cancer survivors. Positive user experiences with iCanFit, coupled with improvements in health-related items all point to the promising utility of this interactive web app. The already large, parallel advances in similar technology including mobile and web apps, fitness trackers, and other devices, may potentially benefit millions of older adults with chronic conditions.

Our study has the following limitations. First, the use of convenience sampling, the sample being mostly drawn from a cancer survivor's organization in Texas, and the relatively highly educated sample (eg, most with some college education or above) limits generalizability to all older cancer survivors in the country. In addition, there was a voluntary bias in the sampling as the older cancer survivors who signed up for this study might have been those who were already physically active. Second, our measures of quality of life were mainly developed from our own research and might not be comparable to other studies with validated quality of life scales. The pilot study described here aimed to test the initial efficacy of iCanFit and it took participants <10 minutes to complete the online survey. Third, the quality of life measures focused on physical and mental health functioning and not a more broad application of multiple quality of life domains. Even so, lifestyle factors such as physical activity were the focus of this study. Future studies need to include validated, multi-dimensional qualify of life scales to increase its comparability to other studies. Fourth, the measure of physical activity was dependent on self-entry and self-report, which may incur recall or social desirability biases. Future research should thus include alternative forms of data entry. For example, the recent popularity of accelerometer and sensors can be used to capture users’ physical activity to reduce users’ burden and potential errors and biases in data entry. Finally, our study did not have a control group and had only one follow-up. Future studies need to include a larger sample, a control group, and long-term follow-ups to further establish the efficacy of the intervention. While the vast majority of older adults have at least one chronic condition, nearly three quarters have ≥2 [22]. Thus, effective interventions must be multi-faceted and recognize that older adults are likely to have co-morbidities which bring greater complexity to adopting and maintaining healthy lifestyles [23,24]. In this study, all older cancer survivors had multiple chronic conditions, which is important to consider when determining their specific needs and barriers to physical activity [25,26]. For example, more tailored messages about how to manage chronic conditions and tips on easy access to opportunities for physical activity and enhanced communication with healthcare providers may be helpful [27-29]. We also need to explore innovative multicomponent interventions to tailor the needs of older adults [30]. Online and mobile health programs designed for older adults need to tailor for their literacy level, cognitive function, and physical abilities [31].

Since the population of older cancer survivors is expected to grow; interventions to improve quality of life among this population are timely and will continue to be needed at even greater levels. The high rates of Internet access and mobile phone ownership throughout the United States and globally have driven greater searches for health-related information online via various mobile devices. Thus, the utility of this timely research is of global interest. Our next steps will be focused on two aspects: building the mobile app with support for communication with smart devices (eg, devices with sensors), and disseminating this mobile-enabled app to greater numbers of older cancer survivors.