Just-in-time Adaptive Feedback Systems to Assist Individuals With Spinal Cord Injury

The Craig H. Neilsen Foundation, Northeastern University, Magee Rehabilitation Hospital, Jefferson Health, MossRehab: Einstein Healthcare Network

The objective of this study is to develop algorithms that incorporate each individual's automatically detected physical activity (PA) level and a clinician's PA level recommendation to provide a smartphone application that helps a person set safe and highly personalized PA goals. By adapting the goals in real-time based on the person's actual behavior, the system aims to keep the individual feeling positive and motivated towards a change in the PA behavior.

Lack of regular physical activity (PA) in the general population is a top public health concern, and this problem is even more acute among individuals with spinal cord injury (SCI). Research has shown that only a small percentage (<20%) of persons with SCI reported consistent PA. Individuals with SCI also experience secondary conditions such as pain, fatigue, weight gain, and deconditioning, conditions that are considered preventable through PA and exercise interventions. The objective of this proposed study is to develop algorithms that incorporate each individual's PA level and a clinician's PA level recommendation to provide a mobile phone application that helps a person set PA goals that are safe, but also highly personalized. By adapting the goals in real-time based on the person's actual behavior, the system aims to keep the individual feeling positive and motivated. Aim 1: Extend and utilize Physical Activity Monitor System (PAMS) to track PA levels, sedentary behavior, and secondary conditions such as pain, fatigue, and deconditioning in community settings. Aim 2: Extend and utilize PAMS to passively monitor PA and provide continuous, but passive feedback about PA levels to individuals with SCI in community settings. Aim 3: Extend and utilize PAMS to passively monitor PA and provide just-in-time persuasive and adaptive feedback to motivate individuals with SCI in community settings. Sample size: A total of 20 individuals with SCI will take part in the study. The sample size for this pilot study is based on budget constraints and other pilot studies. This study will provide the pilot data required to compute the power for future studies. Statistical Analysis: Univariate analysis will be performed to obtain a range of values and the central tendency for variables such as PA levels and sedentary behaviors. Sedentary behavior will be assessed by the time duration of non-movement of individuals with SCI and not just the total duration of being seated in their wheelchairs. The investigators hypothesize that the PA level of individuals with SCI in community will be low compared to the PA level recommendations for individuals with disabilities in general. Furthermore, the sedentary behavior of individuals with SCI will be high compared to the general population. Multiple regression analysis will be performed to assess a relationship between secondary conditions such as pain (scores), fatigue (scores), and deconditioning (reduced capacity scores) and PA levels. The investigators postulate that secondary conditions will be negatively correlated with the PA levels. Repeated measures general linear model (GLM) analysis will be performed to assess the change in PA levels, sedentary behaviors and secondary conditions. In addition, linear mixed model analysis will be performed to develop a personal intercept (and maybe slope) for each participant compared to the mean intercept for each group. Mixed effects model analysis will provide correct estimates of intervention (passive feedback and just-in-time adaptive feedback) and other fixed effects (within-subjects factor) in the presence of correlated data (each participant at different time points) that arise from a data hierarchy (group). Non-parametric tests will be performed if the assumptions for parametric tests are not met.

Spinal Cord Injuries, Physical Activity, Mobile-health, Just-in-time-adaptive intervention, Real-time feedback

Each individual will take part in the study for three phases, each with a duration of one month. The first month will involve collecting baseline PA levels. The second and third months of the study will provide the participants with passive PA level feedback and just-in-time persuasive feedback about their PA levels, respectively.

Passive feedback and JITAI Second Phase - PA Level Feedback Third Phase - PA Level Feedback and Just-in-time Adaptive Intervention (JITAI)

Second Phase - PA Level Feedback A PA recommendation handout based on PA guidelines for individuals with SCI will be provided. Participants continue to use the same sensing equipment as in phase one (baseline). The equipment consists of a smartwatch, a smartphone, and a wheel rotation monitor. Participants can view their feedback whenever they want, but will not be prompted with the information. Third Phase - PA Level Feedback and Just-in-time Adaptive Intervention (JITAI) The third phase of the study will be similar to the second phase of the study, but with an additional JITAI component. The JITAI component of the study will include providing proactively-prompted, real-time feedback through the smartphone and smartwatch when the participant performs a bout of moderate PA.

Energy expenditure in kilocalorie as measured by the physical activity monitor system for wheelchair users.

Minutes of light-intensity physical activity as measured by the physical activity monitor system for wheelchair users.

Minutes of moderate and vigorous-intensity physical activity as measured by the physical activity monitor system for wheelchair users.

Minutes of light-intensity leisure time physical activity per week measured by self-report of Leisure Time Physical Activity Questionnaire for people with SCI.

Minutes of moderate and vigorous-intensity leisure time physical activity per week measured by self-report of Leisure Time Physical Activity Questionnaire for people with SCI.

The pain level for wheelchair users is collected using the Wheelchair Users Shoulder Pain Index. The score ranges from a minimum pain level of 0 to a maximum pain level of 150.

The general pain level was collected using the Chronic Pain Grade scale. The pain intensity score ranges from a minimum pain level of 0 to a maximum pain level of 100.

The general pain level was collected using the Chronic Pain Grade scale. Disability scores range from a minimum of 0 to a maximum of 100.

The fatigue is collected using the Fatigue Severity Scale (FSS). Fatigue scores range from a minimum of 0 to a maximum of 63.

Inclusion Criteria: between 18-65 years of age have been diagnosed with spinal cord injury are at least 6 months post injury use a manual wheelchair as their primary means of mobility (>80% of the time) self-propel their wheelchair are medically stable have experience using a smartphone. Exclusion Criteria: have active pelvic or thigh wounds (pressure ulcers) have a history of cardiovascular disease are pregnant (self-report).

The datasets generated during the current study are not publicly available as the authors do not have the permission to share participants' data.

Rimmer JH, Schiller W, Chen MD. Effects of disability-associated low energy expenditure deconditioning syndrome. Exerc Sport Sci Rev. 2012 Jan;40(1):22-9. doi: 10.1097/JES.0b013e31823b8b82.

Williams TL, Smith B, Papathomas A. The barriers, benefits and facilitators of leisure time physical activity among people with spinal cord injury: a meta-synthesis of qualitative findings. Health Psychol Rev. 2014;8(4):404-25. doi: 10.1080/17437199.2014.898406. Epub 2014 Mar 26.

Tawashy AE, Eng JJ, Lin KH, Tang PF, Hung C. Physical activity is related to lower levels of pain, fatigue and depression in individuals with spinal-cord injury: a correlational study. Spinal Cord. 2009 Apr;47(4):301-6. doi: 10.1038/sc.2008.120. Epub 2008 Oct 21.

Hiremath SV, Intille SS, Kelleher A, Cooper RA, Ding D. Estimation of Energy Expenditure for Wheelchair Users Using a Physical Activity Monitoring System. Arch Phys Med Rehabil. 2016 Jul;97(7):1146-1153.e1. doi: 10.1016/j.apmr.2016.02.016. Epub 2016 Mar 11.

Hiremath SV, Intille SS, Kelleher A, Cooper RA, Ding D. Detection of physical activities using a physical activity monitor system for wheelchair users. Med Eng Phys. 2015 Jan;37(1):68-76. doi: 10.1016/j.medengphy.2014.10.009. Epub 2014 Nov 10.

Canori A, Amiri AM, Thapa-Chhetry B, Finley MA, Schmidt-Read M, Lamboy MR, Intille SS, Hiremath SV. Relationship between pain, fatigue, and physical activity levels during a technology-based physical activity intervention. J Spinal Cord Med. 2021 Jul;44(4):549-556. doi: 10.1080/10790268.2020.1766889. Epub 2020 Jun 4.