Physical activity-only intervention
This intervention will consist of three group teleconference sessions and four one-to-one calls. The overall purpose of the group teleconferencing will be to teach people how to engage in physical activity and set personalized goals to increase physical activity. The overall purpose of the one-to-one telephone calls will be to provide feedback and encouragement on achieving physical activity goals. Participants will first receive the three one-hour weekly group teleconference sessions from a licensed occupational therapist (OT). We selected OTs to deliver the teleconference sessions because of their expertise and training in managing group processes to facilitate behavior change. Participants will also be provided with four 10 to 20-minute, one-to-one, structured, follow-up telephone calls with a graduate assistant every other week. Both the group teleconference sessions and one-to-one follow-up telephone calls will be consistent with goal-setting theory
 and increasing self-efficacy using verbal persuasion, teaching how to set measurable and reasonable goals, proving feedback on goals, and using role modeling (e.g., group members sharing stories)
The first intervention session will focus on discussing the safety precautions for and benefits of physical activity specifically for persons with MS, and then teaching participants how to use a pedometer to monitor physical activity levels. Homework will be assigned to self-monitor walking behavior using the pedometer to generate a baseline level of physical activity for subsequent goal-setting. The second session will focus on using the pedometer to set goals. Participants will be encouraged to achieve personal goals related to the accumulation of pedometer steps by at least 1% every week. Participants will have the option to set additional goals related to exercise, which they can keep track of in their exercise log. Homework will be assigned to set personalized physical activity goals. The final session will focus on the personalized goals, and feedback from the OT and other group members on the appropriateness of goals. There will be a discussion and summary of the guiding principles that will help persons with MS safely engage in physical activity and overcoming barriers to engagement. The one-to-one follow-up telephone calls will focus on providing feedback on physical activity goals. Goals will be reviewed and adjusted biweekly under the guidance of a trained research assistant. The aim will be to teach the participant how to set and adjust goals independently.
Contact-control social support intervention
The contact control group will consist of the same number of contacts as the FM+, i.e., six weekly one-hour teleconference calls with four biweekly one-to-one phone calls. Thus, this condition will allow us to determine whether it is the actual content being taught in the goal-setting physical activity intervention and FM+ intervention that is beneficial, rather than simply the contacts with the OT or social support provided by other group members. The teleconference calls and the one-to-one calls will involve a discussion of topics commonly addressed in support groups. Topics will include information on MS, disease modifying medications, preventive screening, community organizations, nutrition and supplements, cognitive problems, and recognizing symptoms of depression and chronic stress. For the teleconference sessions, participants will be given the opportunity to share problems and provide peer support and solutions for others. For the one-to-one calls, participants will be asked about topics of interest, and the graduate student will gather educational materials and send them to the participant. This information will be the basis for the discussion in the calls. When issues related to fatigue and physical activity are raised, participants will be directed to publically available educational pamphlets. This type of social support group has been used successfully in other clinical trials as a control condition to evaluate the efficacy of fatigue management programs in persons with MS
Ensuring the fidelity of the interventions
Guidelines suggested by Bellg
 will be implemented to promote uniform application of all three interventions. For example, a manual of operating procedures will be provided to interventionists and incorporated into training. Interventionists’ adherence to the intervention protocol will be verified using a checklist and episodic monitoring of the sessions over the phone. Fidelity of the intervention “dose” will be further monitored by tracking attendance for each session, monitoring the completion of worksheets, and testing comprehension.
Primary dependent variables
All dependent variables will be administered at baseline, at the end of the 12-week intervention period, and at a 24-week follow-up period through the mail.
Fatigue impact will be measured using the Fatigue Impact Scale
. This 40-item scale evaluates the perceived impact of fatigue on everyday life, and is valid and reliable among people with MS. It was developed by interviewing patients with MS, and has discriminant validity between patients with MS, general fatigue, and essential hypertension
. Respondents rate each statement using a five-point Likert-type scale ranging from 0 (no problem) to 4 (extreme problem). A total score and three subscale scores (physical, social, cognitive) can be produced from participants responses. The FIS was used to detect changes in fatigue impact resulting from a fatigue management teleconference program
Physical activity will be evaluated using the Godin Leisure-Time Exercise Questionnaire (GLTEQ)
 and an Actigraph tri-axis accelerometer. The GLTEQ is a self-administered measure of usual physical activity. The GLTEQ includes three items that measure the frequency of strenuous, moderate, and mild exercise for periods of more than 15 minutes during one’s free time in a typical week. Participants will be sent an accelerometer to wear daily over three seven-day periods (baseline, post-test, and follow-up). Data are retrieved for analysis via a computer and software provided with the unit. We will use an epoch of one minute. Participants will be instructed to wear the accelerometers all day, except while showering and swimming, for each of the three seven-day measurement periods. The participant will use a log to record the time that the accelerometer is worn, and this will be verified by inspection of the minute-by-minute accelerometer data. Motl et al. have demonstrated the validity and reliability of both the GLTEQ and the accelerometer in the population with MS
Secondary dependent variables
Health-related quality of life, mental health, and participation in life roles will be measured with the SF-12, Mental Health Inventory, and Community Participation Indicator (CPI), respectively. The SF-12 and Mental Health Inventory are valid and reliable measures, and are commonly used in the population with MS
. The CPI is a new measure developed by Heinemann and colleagues. The measure was developed with input from multiple stakeholders, and has been validated through Rasch analysis in a sample of 1,163 individuals with a variety of disabling conditions
[31, 32]. In a subsequent online survey study, the measure had good test-retest reliability over a 2.5 month period in people with MS (t = 0.12, p = 0.91; r = 0.84, p < 0.01). For each item, respondents rate the frequency of engagement (either in days, hours, or times per week, depending on the type of activity), whether it was important (yes/no), and whether they were doing it not enough, enough, or too much. Examples of the 20 items included are: work for money, provide care for a loved one, volunteer, and participate in learning activities.
Six questionnaires will be implemented to identify underlying mechanisms of the interventions’ possible effectiveness for increasing physical activity levels and reducing fatigue impact. Psychosocial construct measures for physical activity will be the Self-Efficacy for Exercise Scale
, Exercise Goal-Setting Scale
[12, 45], and Multidimensional Outcome Expectations Scale
. All three questionnaires are validated and have been used in previous MS research on physical activity
[45, 47]. Psychosocial construct measures for fatigue management will be the validated Self-Efficacy for Performing Energy Conservation Strategies
 along with two additional measures on goal-setting and outcome expectations, which will be developed and validated during the study.
To address potential confounders on the interventions’ effects, information will be collected from participants for the purposes of post-hoc sub-group comparisons and potential use as control variables. These variables will include age, sex, educational level, medications, exacerbations, functional level, and involvement in any other wellness, rehabilitation, or exercise program. Information about changes in MS-related disability, exacerbations, medications, use of rehabilitation programs and other types of exercise/educational programs, and reasons for attrition will be gathered over the phone at 12 weeks and at the 24-week follow-up.
Data management and analysis
A secure online application, Research Electronic Data Capture
, will be used to manage recruitment efforts and facilitate data collection. The first step in the analysis process will be to compute summary statistics, calculate outcome scores, and conduct quality control assessments. The second step in the analysis will be to test whether the randomization process resulted in equivalent groups at baseline. If differences are found at baseline, these variables will be used as covariates during the hypothesis testing phase. The third step in the analysis process will be addressing normality assumptions, and the potential for missing data, before hypothesis testing begins. If the normality assumption is violated, then an appropriate data transformation will be used. Missing data will be treated conservatively. If appropriate, maximum likelihood, multiple imputation, or last observation carried forward will be used to estimate the missing values. After addressing issues of attrition and missing data, the fourth step will be to conduct intent-to-treat analysis to test the hypotheses. Linear mixed-effects models will be used as the primary analysis tool for testing hypotheses, which include condition, time, and interactions as the fixed covariates, and a random intercept that models the correlation among outcomes collected from the same subject. The mixed models will allow us to determine whether there are significant differences between the three conditions and whether there are significant differences across time. All tests will be two-sided, and p = 0.05 will serve as the criterion for significance. As is typically the case for mixed-effects analyses, the results will not be controlled for multiple comparisons. The generalized estimating equation approaches will be used if the normality distributional assumption is violated for the random effects in the mixed models. To test for mediation, we will use a linear regression model. This analysis will first involve regressing changes in fatigue impact or physical activity on the intervention conditions, and then regressing changes in fatigue or physical activity on the intervention conditions plus changes in the potential mediator. Mediation will be based on the interventions’ association with physical activity or fatigue impact being reduced in magnitude and non-significant after controlling for the mediator variable.