A Novel Digital Tool Clinicians Can Use to Prescribe Exercise to Patients With Cardiovascular Disease Risk Factors

Background Approximately half of U.S. adults have 1 or more of the major CVD risk factors of obesity, hypertension, dyslipidemia, and diabetes, and 1 in 4 have 3 or more. Professional organizations such as the American College of Sports Medicine (ACSM) and the American Heart Association (AHA) recommend exercise as a first-line lifestyle therapy to prevent and treat CVD and its risk factors. The ACSM advocates clinicians to assess PA as a vital sign and prescribe exercise for every patient. However, only about 20% of U.S. adults and 44-56% of patients with CVD risk factors report being advised by their health care providers to exercise. Although clinicians are receptive to prescribing exercise to their patients, they encounter barriers to doing so such as a lack of time, training, and the tools. Digital health tools show promise as clinical decision support systems to guide clinicians in prescribing exercise to their patients. The investigators developed a clinical decision support tool named P3-EX. P3-EX includes: 1) the ACSM exercise preparticipation health screening recommendations to determine if there is a need for medical clearance; 2) an adapted AHA Life's Essential 8 cardiovascular health scoring system to determine the CVD risk factor posing the greatest risk; 3) the ACSM strategies for designing an ExRx for adults with multiple CVD risk factors; and 4) the ACSM Frequency, Intensity, Time, Type (FITT) ExRx for the prioritized CVD risk factor posing the greatest risk. Objectives The primary aim of this trial is to evaluate the feasibility and acceptability of P3-EX for clinicians to use to prescribe exercise to patients with CVD risk factors. The secondary aim is to explore the preliminary efficacy of P3-EX to improve patient CVD risk factors, PA levels, and exercise adherence. Study Procedures The investigators will recruit 24 clinicians, including physicians, physician assistants, or advanced practice registered nurses, from Hartford HealthCare or UConn Health clinics using email listservs, newsletters, university communication channels, flyers, and word of mouth. Clinicians will attend a virtual study orientation led by a UConn Graduate Research Assistant to provide informed consent, confirm eligibility, complete demographics, assess barriers to and confidence with ExRx, and receive brief ExRx delivery training to their patients. Clinicians will then recruit two of their patients (N=48) who have obesity, hypertension, dyslipidemia, and/or diabetes. Patients will attend two in-person study visits led by a UConn Graduate Research Assistant at a clinic in Hartford or Farmington, CT, to provide informed consent, confirm eligibility, complete demographics, and assess the AHA Life's Essential 8, anthropometrics, vitals, subjective PA levels via the Timeline Followback for Exercise, and objective PA levels via accelerometry. Patients will attend their local Quest Diagnostics Service Center to assess blood lipid-lipoproteins and blood glucose. Clinicians will be individually randomized with a 1:1 allocation ratio to deliver an P3-EX ExRx to one of their patients and the ACSM-PAVS ExRx to the other patient in a random sequence crossover design. Patients will attend their healthcare appointment with their clinicians and receive either a P3-EX ExRx (n=24) or the ACSM-PAVS ExRx (n=24). Within 48 hours following each healthcare appointment, clinicians and patients will complete the validated mHealth Application Usability Questionnaire and the System Usability Scale to rate the feasibility and acceptability of P3-EX or ACSM-PAVS. Patients will be asked to perform their ExRx and monitor their exercise adherence for 12 weeks using the Timeline Followback for Exercise with virtual oversight from UConn Graduate Research Assistants. Patients will receive a 12-week exercise program information packet, two exercise guidance virtual student visits led by UConn Graduate Research Assistants, and weekly progressive FITT exercise goals via email from UConn Graduate Research Assistants. UConn Graduate Research will also provide Timeline Followback for Exercise summary reports to patients weekly via email. At post-intervention, patients will attend two more in-person study visits led by a UConn Graduate Research Assistant at a clinic in Hartford or Farmington, CT to assess trial satisfaction, AHA Life's Essential 8, CVD risk factors and PA levels. Statistical Analysis Plan The investigators will conduct statistical analyses using Statistical Package for the Social Sciences Version 30. The investigators will first use descriptive statistics and graphical techniques to ensure all test assumptions are met, including the inspection for outliers, normal distributions, and homogeneity of variances. Missing values will be addressed using model-based approaches and/or multiple imputations when appropriate to include the entire randomized sample. If normality assumptions are not met for secondary outcomes, considerations will be made to transform the data to achieve a normal distribution or use alternative non-parametric approaches such as permutated tests. An alpha level of 0.05 will be used to determine statistical significance. The investigators will use the following statistical approaches to evaluate the feasibility and acceptability of P3-EX for clinicians to use to prescribe exercise to patients with CVD risk factors. The investigators will use a one-sided Wilcoxon signed-rank test (one sample case) to assess whether the clinicians mHealth Application Usability Questionnaire ratings of P3-EX and the ACSM-PAVS are above the null hypothesis middle score of 4.0 on the Likert scale, and whether System Usability Scale ratings are above the average score of 68/100. The investigators will use a linear mixed effects model adjusting for the delivery order of P3-EX and PAVS and the time between delivery as potential covariates and interactions between these two covariates to assess differences in clinician usability questionnaire scores between P3-EX and the ACSM-PAVS. The investigators will use normal linear regression to test the strength of relationships between the three domains on the mHealth Application Usability Questionnaire and the usage time of P3-EX. The investigators will use the following statistical approaches to explore the preliminary efficacy of P3-EX to improve patient PA levels, CVD risk factors, and exercise adherence. A one-way Analysis of Variance will test if pre-intervention values are equal between groups, indicating if there is a need to adjust for potential covariates related to demographics, medication use, and/or pre-intervention PA level and CVD risk factor values. The investigators will use a repeated measures two-way Analysis of Covariance using a linear mixed effects model to test patient differences in PA level and CVD risk factor changes over 12 weeks between the P3-EX and the ACSM-PAVS groups, adjusting for potential covariates related to demographics, medication use, and/or pre-intervention values. Scientific Rationale The novelty of P3-EX is supported by the investigators' systematic review which evaluated whether there are decision support tools on the market that utilize evidence-based ExRx standards of the ACSM and AHA to target CVD risk factors. The investigators evaluated 219 exercise apps that were rated ≥4 out of 5 overall with ≥1000 reviews, free to download, and not gender specific. Of the 219 apps, very few (0 to 4.3%) were evidence based, had a preparticipation screening protocol, framed exercise plans by the FITT of ExRx, specified special considerations, or focused on chronic diseases or health conditions, and only 28% built CVD risk factor profiles. The investigators concluded there are no evidence-based ExRx apps on the market like P3-EX. The potential usability and user satisfaction of P3-EX in the healthcare setting is further supported by the investigators' feasibility survey study. A total of 309 healthcare providers and allied health professionals, including 101 physicians, completed a timed case study using the P3-EX web-based algorithm, and then rated its satisfaction and usability using the Mobile Application Rating Scale. Most of the respondents (93%) agreed they would recommend P3-EX to their colleagues, the primary goal of any feasibility study, and 80% agreed P3-EX produced safe ExRx and were satisfied with P3-EX. Also, over 70% agreed P3-EX would make their patients healthier and could save them time, prescribing exercise in an average time of 4.6 minutes. Conclusions This protocol provides the scientific rationale and methodology to test P3-EX within a real-world clinical setting, to inform the feasibility of using P3-EX as a digital health support tool to be used by clinicians to prescribe personalized FITT ExRx to their patients with CVD risk factors, and the preliminary efficacy of P3-EX to improve patient cardiovascular health and PA levels. If successful, this trial could demonstrate that P3-EX is a solution for clinicians to overcome their barriers to ExRx, which includes lacking the tools, training, time, and confidence. The investigators intend to use the pilot data for secondary outcomes to power a larger clinical trial to evaluate the efficacy of P3-EX for improving PA levels and CVD risk factors.