Search

The N-of-1 approach has been applied by patients and providers in the management of a number of recurrent medical conditions, such as recurrent low back pain, migraine headaches, and depression. The basis of this approach is that through careful assessment of environmental exposures and/or medications, often in a trial-and-error methodology, factors can be identified that are positively or negatively associated with recurrence. When combined with technology advances in the form of mHealth applications, this approach holds great promise to improve the efficiency of treatment, and to provide the opportunity for patients to become active participants in their own care. However, much work is needed to understand which types of conditions lend themselves to such an approach, based on patterns of recurrence and association with exposures. In this study, we examined the statistical properties of an mHealth application called the iMTracker, designed for self-management of chronic recurrent medical conditions. We examined the impact of duration of data collection, patterns of recurrence of the primary outcome, and strength of association of a possible associated risk factor (trigger or suppressor), and impact of an intervention on the power of our approach to identify a significant association for possible modification. Using simulation studies of varying effect sizes and durations of data collection, we found that ~90 days of data collection was sufficient to identify associated risk factors with odds ratio (OR > 5.0) at power of 80%, with an absolute event rate of 50% being optimal. We then examined power calculations based on Fishers exact test for proportions to examine the power of this approach to assess the impact of an intervention on the proportion of days with the recurrent outcome, and found that 90 days was also sufficient to detect a decrease of 20% in the rate of the primary outcome, but that shorter data periods could be used to identify stronger effect sizes, down to 15 days with a 90% reduction in rate. We also examined use of an analysis based on Bayesian statistics, which showed general agreement with the frequentist approach, and we created a web-based tool to allow users to perform their own power calculations prior to using the app. Although work remains to be done to explore the effect of autocorrelation and seasonality and trends in data, this study demonstrates that the N-of-1 approach employed in the iMTracker app for self-management of recurrent medical conditions is statistically feasible, given the right conditions.

Management of chronic recurrent medical conditions (CRMC), such as migraine headaches, chronic pain and anxiety/depression, is a major challenge for modern providers. The fact that often the most effective treatments and/or preventative measures for CRMCs vary from patient to patient lends itself to a platform for self-management by patients. However, to develop such an mHealth app requires an understanding of the various applications, and barriers, to real-world use. In this pilot study with internet-based recruitment, we conducted an assessment of user satisfaction of the iMTracker iOS (iPhone) application for CRMC self-management through a self-administered survey of subjects with CRMCs. From May 15, 2019 until March 27, 2020, we recruited 135 subjects to pilot test the iMTracker application for user-selected CRMCs. The most common age group was 31–45 (48.2%), followed by under 30 (22.2%) and 46–55 (20%). There were no subjects over 75 years old completing the survey. 38.8% of subjects were college graduates, followed by 29.6% with a Master’s degree, and 25.9% with some college. No subjects had not graduated from high school, and only 2 (1.5%) did not attend college after high school. 80.7% of subjects were self-identified as Caucasian, and 90.4% as not Hispanic or Latino. The most common CRMC was pain (other than headaches) in 40% of subjects, followed by mental health in 17.8% and headaches in 15.6%. 39.3% of subjects experienced the condition multiple times in a day, 40.0% experienced the condition daily, and 14.8% experienced the condition weekly, resulting in a total of 94.1% of subjects experiencing the condition at least weekly. Among the concerns about a self-management app, time demands (54.8%) and ineffectiveness (43.7%) were the most prominent, with privacy (24.4%) and data security (25.2%) also noted. In summary, we found internet-based recruitment identified primarily Caucasian population of relatively young patients with CRMCs of relatively high recurrence rate. Future work is needed to examine the use of this application in older, underrepresented minorities, and lower socioeconomic status populations.