Post-stroke gait deviations contribute to significant functional disability, impaired walking ability and poor quality of life. Prior studies suggest that gait training with paretic lower limb loading may improve gait parameters and walking ability in post-stroke. However, most gait training methods used in these studies are not readily available, and studies using cheaper methods are limited.

Stroke is the second leading cause of death and a major public health problem in both developed and developing countries. It is one of the most common causes of long-term disabilities among adults.

Stroke survivors exhibit varying neurological deficits and impairments including sensory, cognitive, perceptual, emotional, speech and motor such as gait impairments. The most important motor deficit after a stroke is the paresis of the affected side, which reduces the capacity of the paretic lower limb to maintain normal functions such as balance, initiation, and control of movements, and this negatively affects their gait.

Gait is altered post-stroke due to basic motor impairments such as altered muscle tone, lack of selective motor control, abnormal reflexes, poor balance, and muscle weakness. Gait impairment is one of the hallmarks of stroke, and many survivors are reported to have slow motor recovery in the affected lower extremity years after the stroke, which impairs their gait quality and contributes to persistent gait asymmetry. Gait asymmetry implies a significant imbalance of functional activities between two sides of the body or lower limb.

Gait symmetry is a valuable feature of gait quality, and its abnormality negatively affects gait parameters. Post-stroke, 44%-62% of the survivors showed asymmetry in the spatial gait parameters, while 48%-82% developed asymmetry in the temporal gait parameters.

After a stroke, the paretic limb exhibits a prolonged period of swing and a reduced period of stance; as a result, decreased swing time and increased stance time are seen in the non-paretic limb. These changes are believed to result from decreased walking speed as well as disturbances in other gait parameters between the paretic and non-paretic limbs, which may limit gains from rehabilitation. Despite rehabilitation efforts, self-loading and self-initiation of gait are typically ineffective in stroke patients. Stroke rehabilitation now focuses on improving load transfer onto the paretic limb to enhance proprioception and gait symmetry by providing an external source of limb loading to augment patient effort. Many survivors usually do not have significant improvement in gait function. This is often due to inefficient weight application (limb loading) on the affected side, causing non-linear limb load transfer.

Though there have been numerous interventions to manage gait in stroke survivors, using weight during over-ground walk training on stroke survivors has not been thoroughly evaluated. However, a cross-sectional study has found that weight application at the knee and ankle were both found to influence the spatiotemporal gait parameters and gait symmetry, with ankle loading having more influence on more parameters. This study is hoped to serve as a follow-up RCT to the cross-sectional study by Bala (2019).

Inclusion Criteria:

1. Unilateral ischemic or hemorrhagic stroke occurring at least 3 months before enrolment
2. Male or female between the age of 20 and 70 years
3. Having a Modified Rankin Scale (MRS) scores of 1, 2, or 3, and also able to walk at least 10 meters independently without an assistive device
4. Ability to follow verbal instructions.

Exclusion Criteria:

1. Cognitive impairments
2. Visual impairments
3. Musculoskeletal disorders that may affect gait such as arthritis
4. Other neurological disorders such as Parkinson's disease, multiple sclerosis
5. Cardio-respiratory conditions that may limit participation such as atelectasis
6. Concurrent participation in other interventional clinical trials.

Upon completion of all the baseline assessments, consenting participants will be stratified by MRS scores (i.e., 1, 2, and 3) and randomized into two groups: over-ground walking with lower limb loading or over-ground walking with lower limb non-loading in a 1:1 ratio by a third party not involved in other aspects of the study.

The allocation of participants will be concealed using consecutively numbered, sealed, and opaque envelopes which will be prepared using a web-based randomization tool (available at https://www.sealedenvelope.com) with permuted block sizes of 6.

All outcome assessors will be blinded to participants' group allocation. However, it will be impossible to blind the therapists and participants considering the nature of the interventions.