Inclusion Criteria:

1. Participants must have a clinical diagnosis of MS according to the revised McDonald Criteria. Participants with an EDSS score of 2.0-6.5 will be included but must have a stable disease course without worsening more than 1.0 EDSS point over the last 3 months and no MS exacerbation within the preceding 4 weeks, as determined by interview and review of medical records. 2. Participants will have stable MS disease treatments. All medications will be consistent for at least 1 month prior to enrollment. No corticosteroids for at least 1 month and no botulinum toxin injections above the knee for at least 3 months prior to enrollment. 3. Participants will be between the age of 18 and 65 and have a body mass of less than 135kg (maximum mass for treadmill equipment). 4. Participants will be able to follow three commands, as determined by the three step command test of the Mini Mental State Exam. 5. Participants must have had no myocardial infarction in the past month, must not have uncontrolled hypertension (blood pressure must be < 190/110 mmHg at rest), must not have a symptomatic fall in blood pressure when standing and must not have documented, uncontrolled diabetes. 6. Participants will be medically stable, with absence of concurrent severe medical illness including: existing infection, known significant cardiovascular or metabolic disease that limits exercise participation, significant osteoporosis (as indicated by known history of fractures), known history of vascular claudication or pitting edema, and known history of pulmonary complications that limits exercise capacity, including significant obstructive and/or restrictive lung diseases. 7. All participants must be able to perform walking training with passive range of motion within the limits of normal locomotor function, including: 0-30 +/- 10 degrees ankle plantarflexion, knee flexion from 0 to 90 +/- 10 degrees, hip flexion to 0-90 +/- 10 degrees. 8. Individuals who are undergoing concurrent physical therapy or supervised exercise by a trained professional will be excluded from the study to eliminate confounding effects of additional physical interventions. Other therapies, such as occupational or speech therapies, will be allowed as prescribed by their physician. Individuals will be allowed and encouraged to continue their normal exercise routines during the course of the intervention. 9. Participants must have no other concomitant neurological diseases, no history of epileptic seizures, peripheral nerve injury in lower legs or traumatic brain injury. 10. Participants must have adequate hearing (whisper test) and vision (minimum 20/80 corrected vision on a Snellen chart). 11. Participants must be able to walk for 10 meters at their preferred walking speed. For participants that require assistive devices to walk overground, minimal assistance will be provided to enable training until participants recover enough that they are not needed. The use of braces or orthoses is allowed in the proposed study to assure orthopedic safety. Participants will be excluded if they have factors that preclude stepping exercise, such as severe spasticity, excessive fatigue or exercise intolerance. 12. Women of childbearing potential will not be excluded, although women who are pregnant or who are considering becoming pregnant will be excluded due to the trunk and pelvis restraints required for safety during treadmill training. 13. Participants must be willing to commit to the treadmill training program schedule and participate in all of the assessments. They must be able to provide informed written consent and willing to be randomized to any of the 4 study arms.

Exclusion Criteria:

See inclusion criteria.

This phase I/II clinical trial examines the effects of exercise intensity and balance perturbations on treadmill training in people with multiple sclerosis (MS). People with mild to moderate MS will be assigned to one of four treadmill training groups. One group will conduct treadmill training at a high exercise intensity with balance perturbations applied to the treadmill, a second group will undergo the same training at a low exercise intensity, a third group will conduct treadmill training at a high intensity with no perturbations and the fourth group will train at low intensity with no perturbations. The fourth group resembles typical clinical treadmill training paradigms. Each participant will undergo an initial screening to measure impairment (based on EDSS) and ability to follow three step directions. Demographic information will also be collected on each participant. Twenty participants will be assigned to each treatment group, with group assignment conducted randomly, balanced by severity of MS symptoms so that the impairment across groups is similar. The target sample size for this study was based on extensive previous studies examining high-intensity treadmill training in people with stroke or spinal cord injury, which show large effects compared to conventional treadmill training. In addition, data from a pilot study in people with MS demonstrated that perturbations applied during treadmill training have a large effect on postural stability while walking. The investigators expect that training intensity will impact gait speed and endurance, the use of perturbations will improve balance while walking, and both intensity and perturbations will improve stepping in the community. The proposed training interventions will consist of up to 30 sessions of up to one hour of treadmill training over approximately 10 weeks. Each training session will consist of a warmup period and assessment of intensity, followed by four 10-minute intervals of treadmill stepping (as tolerated) with 2-3 minute rest intervals in between and a cool down at the end. In the groups receiving perturbations, movements will be applied to the treadmill every 7-20 seconds in random direction: right, left, forward or backward. Safety will be assured using a fall arrest harness, which is worn at all times. Perturbation size will be determined using a four-accelerometer system (right foot, left foot, sacrum and C7) to obtain an estimate of margin of stability for right/left perturbations and dynamic stability index for anterior/posterior directions. Training perturbation size will be increased until the stability reaches the stability threshold. This perturbation size will then be used for the training session. The intensity of the training will be controlled by monitoring heart rate. The target heart rate will be 70-80% of age-adjusted heart rate reserve for high intensity training and 30-40% of age-adjusted heart rate reserve for low intensity training. Care will be taken to maintain a cool environment, with monitoring of heart rate, rating of perceived exertion and step count. Depending on initial conditioning, participants may need to start with lower levels of exercise at the beginning of the training period and gradually increase until the target dose is achieved. Assessments of walking function will be made at four times. The first assessment (Baseline) will be made before training begins, a second assessment will be made after 15 training sessions (Mid-Training), a third assessment will be made after training (Post-Training) and a follow up assessment will be made six months after training (Follow-Up). Each assessment will take approximately one hour to complete a series of clinical function tests and laboratory measures of gait function. Specifically, each assessment will consist of measurements of self-selected gait speed, gait endurance, peak treadmill speed, cardiorespiratory fitness and balance confidence. In addition, dynamic balance during gait will be measured using the response to balance perturbations. These measurements will be made using full body motion capture during a set of perturbations in each of the four perturbation directions. Kinematics of the response to the treadmill perturbations will be quantified including measures of foot placement, center of mass movement, with calculations of margin of stability and dynamic gait stability. Additional details of trunk and joint kinematics will be made as secondary measurements. At Baseline, Post-Training and Follow-Up, the investigators will also measure community stepping by issuing each participant a step counter for a 14-day period. The step counter will record the timing of steps throughout the day. When the participant returns, the steps count data will be downloaded and analyzed. Falls will be assessed at Baseline using a survey and each participant will be provided a journal to document falls, with phone contact every 4 weeks through Follow-Up. The study staff making the assessments will be blinded to the treatment group. The training staff and participant will not be able to be blinded to the treatment, since intensity of training and treadmill perturbations cannot be hidden from them. The statistician will be unblinded, as they will assign the participants to the treatment groups and conduct the statistical analysis on the measurements provided by the study staff. The effects of training intensity will be assessed using measurements of gait speed and endurance. The investigators expect significant increases in these measurements with high-intensity compared to low-intensity training. Perturbation training is not expected to affect gait speed or endurance. Instead, the investigators expect that perturbation training will improve balance while walking. Training intensity is not expected to significantly affect balance. Both training intensity and perturbation training are expected to improve community mobility, measured by the number of steps per day. The investigators also expect secondary effects from high-intensity training to include improved cardiorespiratory fitness. Similarly, perturbation training is expected to have positive secondary effects on falls and balance confidence. All together, the investigators anticipate that improvements in community mobility provided by high-intensity perturbation treadmill training will improve quality of life in people with MS.

Arms

Experimental: High-Intensity With Pertubations

30 sessions of high-intensity treadmill training will be conducted. Perturbations that disrupt balance will be applied during the training.

Experimental: High-Intensity No Perturbations

30 sessions of high-intensity treadmill training will be conducted on a stable treadmill.

Experimental: Moderate-Intensity With Perturbations

30 sessions of moderate-intensity treadmill training will be conducted. Perturbations that disrupt balance will be applied during the training.

Active Comparator: Moderate-Intensity No Perturbations

30 sessions of moderate-intensity treadmill training will be conducted on a stable treadmill.

Interventions

Procedure: - High-Intensity Treadmill Training

Participants will walk on a treadmill at a speed that produces 70-80% of age-adjusted heart rate reserve or a rating of perceived exertion of 17/20. A total of 30, one hour sessions will be conducted. Training will be conducted in 10 minute bouts, with 2-3 minutes rest between bouts.

Procedure: - Treadmill Training with Perturbations

Participants will walk on a treadmill and a perturbation of the treadmill will be produced every 20s. Perturbations will randomly occur in the forward, backward, right or left directions, with a perturbation size set to 80% of stability threshold. A total of 30, one hour sessions will be conducted. Training will be conducted in 10 minute bouts with 2-3 minutes rest between bouts.

Procedure: - Standard Treadmill Training

Participants will walk on a treadmill at a speed that produces 30-40% of age-adjusted heart rate reserve. A total of 30, one hour sessions will be conducted. Training will be conducted in 10 minute bouts, with 2-3 minutes rest between bouts.