Task-Oriented vs Conventional Balance Training on Balance and Mobility in Chronic Stroke: A Randomised Controlled Trial

Balance impairment is one of the strongest predictors of falls, dependence and poor community participation after stroke. Many patients discharged from Indian tertiary hospitals continue physiotherapy in an OPD for months, where balance training is often a set of static weight-shifting and standing exercises repeated session after session. Motor-learning theory argues that practice of meaningful, variable, goal-directed tasks — reaching for objects, stepping over obstacles, sit-to-stand from different heights, turning, walking while carrying — transfers better to daily life.

This MPT dissertation is a two-arm, parallel-group, assessor-blinded randomised controlled trial in adults with chronic stroke (more than six months post-onset). The experimental group receives task-oriented circuit balance training; the control group receives conventional balance training of matched dose — 45 minutes, five days a week, for six weeks. Outcomes are the Berg Balance Scale (primary), the Timed Up and Go and the Fugl-Meyer Assessment – Lower Extremity, measured at baseline and at six weeks by an assessor who does not know group allocation.

The trial uses computer-generated block randomisation, allocation concealment with sequentially numbered opaque sealed envelopes, prospective registration on the Clinical Trials Registry – India (CTRI), Institutional Ethics Committee approval and written informed consent, and reports participant flow using a CONSORT 2010 diagram. It follows the synopsis → ethics → dissertation → viva route of RGUHS / KUHS / MUHS MPT regulations.

Abstract

Background: Balance impairment after stroke limits independence and increases fall risk. Task-oriented training applies motor-learning principles, but evidence comparing it with dose-matched conventional balance training in Indian chronic-stroke populations is limited. Aim: To compare the effect of task-oriented and conventional balance training on balance, mobility and lower-limb motor function in chronic stroke. Methods: Parallel-group, assessor-blinded RCT; 40 participants randomised 1:1; 45-minute sessions, 5 days/week for 6 weeks. Primary outcome BBS; secondary outcomes TUG and FMA-LE; analysis by intention to treat.

Introduction

Stroke is a leading cause of adult disability in India. Post-stroke balance deficits arise from weakness, abnormal tone, sensory loss, impaired anticipatory postural adjustments and fear of falling. Conventional balance programmes emphasise isolated static and dynamic exercises; task-oriented training organises practice around functional goals, with variability and progression in environmental demand, which is expected to promote motor learning and transfer.

Review of literature and gap

Systematic reviews report that task-specific and circuit class training improve walking and balance after stroke, but many trials compare against no training or unequal dose, have small samples, or lack concealed allocation and blinded assessment. Indian trials frequently omit trial registration and CONSORT reporting. A dose-matched, registered, assessor-blinded RCT addresses these weaknesses.

Hypotheses

• H0: there is no significant difference in BBS, TUG and FMA-LE change scores between task-oriented and conventional balance training.
• H1: task-oriented training produces significantly greater improvement in BBS, TUG and FMA-LE.

Feasibility

• Setting: neuro-physiotherapy OPD of a tertiary teaching hospital seeing adequate chronic-stroke caseload.
• Equipment: steps, cones, obstacles, chairs of varying height, balance pads, stopwatch — all low-cost.
• Time: 6 months for recruitment and intervention within the Year-2 dissertation window.
• Ethics: low-risk exercise; IEC approval, CTRI registration and informed consent before the first enrolment; a trained attendant guards every participant against falls.

Adults with chronic stroke frequently continue to have impaired balance, slow functional mobility and a high risk of falls long after discharge, even when they attend outpatient physiotherapy regularly. Conventional balance training in many Indian OPDs relies on repetitive, isolated exercises that may not transfer to everyday tasks such as turning in a small kitchen, stepping over a threshold or rising from a floor-level seat. Task-oriented training is theoretically better suited to transfer, but local evidence from rigorously designed trials — with concealed randomisation, dose-matched comparison, blinded assessment and prospective registration — is scarce.

This dissertation asks: in adults with chronic stroke, is six weeks of task-oriented balance training more effective than dose-matched conventional balance training in improving Berg Balance Scale score, Timed Up and Go time and Fugl-Meyer lower-extremity score?

In scope

• Adults aged 40–75 with first-ever unilateral stroke more than six months ago, able to walk 10 m with or without an aid.
• Two dose-matched interventions over six weeks in one hospital OPD.
• BBS, TUG and FMA-LE at baseline and six weeks; adverse events and adherence.
• Intention-to-treat analysis with a CONSORT diagram.

Out of scope

• Acute and sub-acute stroke, cerebellar or brainstem stroke with ataxia, and bilateral lesions.
• Long-term follow-up for falls over months.
• Instrumented posturography or gait laboratory measures.

Study design: prospective, two-arm, parallel-group, assessor-blinded randomised controlled trial with 1:1 allocation.

Setting: neuro-physiotherapy OPD of a tertiary-care teaching hospital.

Sample size: computed in G*Power for the difference between two independent means (two-tailed), using the expected between-group difference in BBS change and its SD from a published trial or your own pilot (for example, effect size d = 0.95, α = 0.05, power 80% gives 18 per group); adding about 10% for attrition gives 40 participants (20 per group). Replace these values with the effect size you justify in the synopsis.

Inclusion criteria: first-ever unilateral ischaemic or haemorrhagic stroke confirmed by CT/MRI; more than 6 months since onset; age 40–75; BBS between 21 and 45 (medium fall risk); able to walk 10 m independently with or without an aid; MMSE ≥ 24; medically stable.

Exclusion criteria: cerebellar or brainstem lesions; other neurological or vestibular disorders; lower-limb fracture or surgery in the last 6 months; severe visual or hearing loss; uncontrolled hypertension, unstable angina or recent myocardial infarction; participation in another trial.

Randomisation and concealment: a person not involved in recruitment generates a block-randomised sequence (block size 4) by computer and places allocations in sequentially numbered, opaque, sealed envelopes, opened only after baseline assessment.

Blinding: the outcome assessor is blinded to allocation; participants and the treating therapist cannot be blinded, which is acknowledged.

Interventions (45 min, 5 days/week, 6 weeks):

• Task-oriented group: circuit of 8 stations — sit-to-stand from varying heights, reaching in standing to targets at different distances, stepping forward/sideways onto steps, stepping over obstacles, walking with turns and direction changes, walking while carrying objects, tandem walking, standing on foam while performing a functional task; progression by height, distance, speed and surface.
• Conventional group: static and dynamic weight shifting, single-leg stance with support, heel–toe raises, balance-board exercises and standard gait training.
• Both groups receive the same 10-minute warm-up and cool-down.

Outcome measures: BBS (primary), TUG, FMA-LE, measured at baseline and end of week 6.

Ethics and registration: IEC approval; prospective CTRI registration before enrolment; participant information sheet and written informed consent in the participant's language following the ICMR National Ethical Guidelines (2017); a trained attendant guards every session; falls and adverse events recorded and reported to the IEC.

Statistical analysis: Shapiro–Wilk for normality; baseline comparability by independent t-test / chi-square; within-group change by paired t-test or Wilcoxon; between-group change by independent t-test or Mann–Whitney U (ANCOVA with baseline as covariate as a sensitivity analysis); mean differences with 95% CI and effect sizes; intention-to-treat with last observation carried forward, per-protocol as sensitivity; p < 0.05; SPSS.