Home-Based Dual-Task Balance Training for Parkinson’s Disease: A Feasibility Study of a Randomized Controlled Trial

Abstract

Purpose:

Dual-task training, incorporating both motor-motor and/or motor-cognitive elements, is recognized as a safe and effective strategy for enhancing balance in individuals with Parkinson’s disease. Despite this, limited research has explored the implementation of dual-task training within a home-based context or superiority of task types. This study aims to investigate the feasibility and acceptability of 2 novel, unsupervised home-based programmes; each adopting a comparable procedure yet with a focus on motor-motor and motor-cognitive dual-task training.

Methods:

Six participants with mild to moderate stage Parkinson’s disease were randomized to group (M-DTT/C-DTT). Each engaged in 30-minute training sessions 3 times per week over 6 weeks, accompanied by a non-professional ‘training buddy’. Attendance, adherence, and safety monitoring informed feasibility analysis, alongside post-intervention interviews. Balance outcomes were measured using the MiniBESTest and body sway analysis at baseline and post-intervention.

Results:

Overall trial design including intervention and assessment approaches were safe and generally well-received, with high attendance and adherence rates, yet key lessons were learnt: Both programmes could be further improved with diversifying secondary tasks, redesigning task combinations and improving the technical aspects of the training movies.

Conclusion:

Home-based motor-motor and motor-cognitive dual-task training interventions undertaken without therapists’ supervision can be an acceptable and inclusive approach for balance rehabilitation for people with Parkinson’s disease. Lessons learnt will inform intervention re-design ahead of a randomized controlled trial of effectiveness and superiority testing of these approaches.

ClinicalTrials.gov ID:

NCT05710588, https://clinicaltrials.gov/study/NCT05710588?term=NCT05710588&rank=1

Introduction

Parkinson’s disease (PD) is a chronic neurodegenerative condition marked by the degeneration of dopaminergic neurons in the substantia nigra and this degeneration results in diminished dopamine levels in the striatum, leading to disruptions in motor control. ¹ Postural instability and balance impairment are among the most disabling motor features; as the disease progresses, fall rates and the risk of dementia may increase, and worsening balance control can lead to a loss of independence, potentially resulting in the need for residential care. ² The Hoehn and Yahr (H&Y) scale is widely used to describe symptom progression in Parkinson’s disease. ³ Although clinically apparent balance impairments typically emerge between H&Y stages 2 and 3, indicating mild to moderate disease severity, posturography shows measurable impairment even in early disease.^4-6 Pharmacological treatments targeting dopaminergic deficits ⁷ improve symptoms such as bradykinesia, tremor, and rigidity ⁸ but have limited impact on functional balance, ⁴ and levodopa may worsen instability by increasing postural sway and fall risk. ⁵ Consequently, targeted balance-focused interventions are essential for supporting functional independence in mild to moderate PD.

Dual-task training (DTT) involves performing 2 attention-demanding tasks simultaneously—typically a balance-dependent motor task combined with a secondary motor task (M-DTT) or a cognitive task (C-DTT). ⁶ DTT is now widely adopted in rehabilitation settings as a safe and effective approach for improving functional and dynamic balance for pwPD; supported by recent systematic review and meta-analysis.^9,10 Despite effective, these traditional clinic-based approaches do not inclusively serve pwPD, an important concern highlighted in patient and public involvement and engagement activities (PPIE undertaken with pwPD in the South West England area of the United Kingdom in 2021). Adherence to training programmes, and in turn potential benefit to older adults with long-term conditions is influenced by factors such as adequate space, accessibility, and flexible scheduling, ¹¹ resulting in reduced participation when exercise classes are inconvenient or transportation is limited or unavailable. ¹² Consequently, re-design and validation of DTT programmes to enable home-based training is vitally needed to provide a more accessible and inclusive option for pwPD.

Isolating cognitive load in M-DTT from C-DTT is difficult because motor tasks inherently require cognitive attention, and cognitive tasks often involve motor responses; however, each targets distinct systems and may therefore produce different balance outcomes. Current knowledge of superiority of M-DTT and C-DTT is limited, ¹³ and no studies have presently evaluated home-based, unsupervised versions of DTT.

This study aimed to develop and assess the feasibility and acceptability of 2 novel home-based, video-delivered dual-task training programmes—motor-motor and motor-cognitive—designed to target balance in pwPD. Specifically, the study explored whether these 2 approaches could be delivered safely and practically in an unsupervised home setting, and whether participants were able to engage with and complete the motor and cognitive task components. The findings are intended to inform the design of a future randomized controlled trial evaluating their effectiveness.

Study Design

This feasibility study used a mixed-method design which was informed by extensive patient engagement activities and a prior qualitative study that gathered perceptions of DTT for pwPD; directing the home-based delivery, hybridized assessment approaches (including outcome measures adopted) and task-related training activities adopted. Mixed methods combine qualitative data with Likert-based patient-reported outcomes to evaluate the feasibility and acceptability of the newly developed DTT programmes and trial procedures. This study has been designed with the aspiration of conducting an onward RCT to compare the effectiveness of M-DTT and C-DTT, if feasibility is established and adopts progression criteria (with traffic light signaling) to inform next steps; to progress (total green lights), to modify and re-assess (mix of green and amber lights) or to stop (in the presence of any red lights). In the case of red light criteria a total redesign would be the appropriate step with re-evaluation of feasibility thereafter. Please see Supplementary Material 2 for a more detailed account of progression criteria employed. Quantitative data from outcome measures was collected to replicate the participant experience of an onward powered trial design. These measures were not inferentially analyzed as the initial focus on feasibility determines non-powering of analysis of effectiveness presently. The study was registered on ClinicalTrials.gov (NCT05710588) in accordance with CONSORT reporting standards.^14,15

Ethical Considerations

Ethical approval was granted by the Faculty of Health Ethics and Integrity Committee at the University of Plymouth (reference 2022-3332-2865).

Consent to Participate

Eligible participants provided written informed consent to participate to the study.

Materials and Methods

Participants, Recruitment, and Randomization

Participants were recruited between November 2022 and February 2023. South-West England Parkinson’s UK local support groups, and existing patient networks distributed advertisements and information about participation. Interested pwPD contacted the research team and were screened against their understanding of the study and their eligibility (using Zoom online Audio/Video Communications, Inc, CA, USA).

Eligibility criteria included a self-reported PD diagnosis, mild-to-moderate disease (based on scoring of the motor and cognitive items of the Movement Disorder Society Unified Parkinson Disease Rating Scale (MDS-UPDRS) modified into interview format, ability to follow instructions, to complete self-report outcomes, to use online platforms (independently or with support), availability of a supporter (aged ≥18) as a training buddy, and identification of a safe 2-m square training area in the home. The training area needed to be free from hanging objects or shelves and not featuring flaking wallpaper or plaster, with no trip hazards and with the potential to place a chair within the space (for seated rests). Exclusion criteria included medical conditions affecting balance (such as stroke, cerebellar disorders, a vestibular impairment, a skeletal fracture (occurring within the past 6 months), severe visual or hearing impairment, inability to communicate in English, or inability to stand independently for >1 minute.

Eligible participants and their training buddies provided informed consent and completed baseline assessments, including demographics, PD duration, falls history, MDS-UPDRS-II, and Mini Mental State Examination scores. Participants were then randomly and blindly assigned to groups using a minimization coin toss method (by L.B., blinded to any baseline assessment), balancing baseline MDS-UPDRS-II scores (12-17 mild, 17-30 moderate).

Intervention

The intervention design was informed by findings from a prior qualitative study involving pwPD, their supporters, and physiotherapists (see Supplementary Material 1). Participants indicated that a 3-times-per-week programme over 6 weeks was feasible within their daily routines, while sessions exceeding 30 minutes were considered potentially fatiguing and less acceptable due to the risk of off periods. Accordingly, the intervention was designed as 30-minute sessions delivered 3 times per week for 6 weeks.

Two intervention groups (M-DTT and C-DTT) were included in this randomized trial, with control of procedures to ensure group comparability. No true control group was involved as evidence has established effectiveness of DTT compared to no intervention and this programme of work seeks to inform future testing of superiority of approaches. Each 6-week training programme began with a baseline assessment and a researcher-supervised (N.C.), face-to-face session at the university. During this session, the participant and their designated ‘training buddy’ practiced the session structure. Participants then completed home-based sessions with their training buddy, 3 times per week for 6 weeks, without professional supervision. Each 30-minute session comprised 3 prerecorded videos: a 5-minute warm-up, a 20-minute DTT segment, and a 5-minute cool-down. The main DTT segment was divided into 2 concept-focused blocks separated by a 1-minute seated rest.

Training buddies were incorporated to support motivation and adherence. ¹⁶ Their responsibilities included documenting scores and task completion, assisting with tablet use, managing video navigation and setup, monitoring for fatigue or difficulties, and reporting any adverse events or near misses (see Supplementary Material for more information).

Training Materials

Training videos were hosted on the university server via Panopto, Inc (Washington, USA). Participants received personalized secure viewing links, enabling researchers to monitor individual engagement through digital records. During the initial session, participants and training buddies set up the training space and equipment with researcher assistance. Training buddies were briefed on their duties, and participants received workbooks with scoring instructions for performance, enjoyment, and challenge (rated at the end of each training session). All participants were provided tablets, non-damage wall mounts, and exercise mats. The M-DTT group also received an instructional setup video due to their additional wall-mounted equipment in this programme. The researcher conducted Zoom sessions at the end of every 2 weeks to check collection of training diary data, interim online Likert scale responses and adverse events or any near misses and assess readiness to progress to the next training phase. These sessions were used solely for monitoring and assessment purposes.

Training Content

Standardized primary motor balance tasks were selected from common rehabilitation exercises for pwPD such as stepping, marching, reaching and lunging.

Game-like features, steered by initial patient engagement, were added to enhance engagement. ¹⁷ C-DTT included memorization, calculation, and object-naming tasks; M-DTT incorporated purposeful use of fidget toys and ‘brick tasks’. Brick tasks specifically involved timed challenges for participants to affix either bricks onto pre-sited boards on the wall in front of them. Board positioning incorporated reach in upward, sideways and lower positions and a progression from the use of large (8 dot) to small (4 dot) bricks.

Personal training packs where scores could be documented in each training session by the training buddy adopted a error-less approach incorporating positive feedback and encouragement for participants to exceed personal bests and attempt higher difficulty levels. ¹⁸ Full training protocols are provided in Supplementary Material 2.

Progression, based on an intervention protocol, followed a consistent strategy for both groups, with identical balance tasks. ¹⁹ Difficulty was increased every 2 weeks during online meetings with the remote therapist-researcher. Individualized progression followed a standardized procedure, common to both groups, which was informed by remote re-assessment and included narrowing stance width and adjusting balance-related task speed.

Assessments and Outcome Measures

The study’s primary feasibility outcomes were adherence (informed by session attendance rates, training movie streaming percentage completion cross-referenced by training log completion), safety, and acceptability for both DTT programmes and the overall trial. Attendance is defined in the data analysis as the number of session views, indicating that participants accessed and streamed the training videos. Adherence is defined as the total minutes streamed. Panopto automatically captured attendance, session duration, and individual viewing times.

Safety was monitored through adverse event reporting, near misses, and reported difficulties documented in training workbooks or raised during bi-weekly online meetings and interviews.

Acceptability was evaluated through post-session enjoyment ratings and challenge ratings on separate 10-point scales (0 = none; 10 = maximum). To triangulate data, participants also completed bi-weekly reflective online 5-point Likert scales, which was adapted from prior survey tools.^20,21

Balance Assessments

In-person balance assessments were selected to meet participants’, supporters’, and physiotherapists’ expectations for valid and reliable evaluation of progress, as identified in the qualitative findings mentioned in the intervention section.

Balance assessments were scheduled and conducted during participants’ ON medication periods at baseline and within 1-week post-training. Functional balance was evaluated using the MiniBESTest, a planned primary outcome for a future RCT. Standing balance was measured with an XSens Awinda sensor (XSens, Enschede, Netherlands), located and affixed with double sided tape over participant’s C7 spinal vertebrae, recording angular sway at 100 Hz over 45 seconds, excluding the first and last 5 seconds. Sway velocity and root mean square (RMS) values of acceleration in lateral and anteroposterior directions were calculated across 3 conditions: eyes open, eyes closed in a comfortable stance, and eyes open with a narrowed standardized stance (feet 4 cm; as close as physically possible to challenge balance yet accommodative of differences in participant anthropometrics of the legs).

Interviews

Qualitative data were collected via face-to-face semi-structured interviews with pwPD during their second and final university visit, with training buddies optionally present to support comprehension or recall. Conducted by N.C. and audio-recorded, interviews averaged 40 minutes. An open-ended interview guide (Supplementary Material 3) explored experiences of the home-based training, assessment visits, and trial participation, including training difficulty and enjoyment, home comfort, travel perceptions, Likert scale completion, training confidence, and any adverse events such as falls or dizziness.

Data Analysis

Acceptability and enjoyment ratings were calculated per session from participants’ training workbooks. Attendance was measured by sessions attempted, and adherence by mean and total video viewing durations. Responses to the 5-point Likert scale were analyzed using descriptive frequencies and graphically displayed. Progression criteria were informed according to the thresholds listed for measures of each feasibility domain in Table 1 of Supplementary Material 4, with a traffic light system employed to provide a visual overview of study findings.

Changes in MiniBESTest scores, angular velocities, and RMS acceleration in lateral and anteroposterior directions were descriptively reported for each participant.

Qualitative Data Analysis

The data were analyzed using thematic analysis. ²² A structured, applied approach was chosen, rather than reflexive thematic analysis, as it is aligned with the descriptive aims of mixed-methods feasibility research and supports the systematic summarization of participant experiences to inform intervention and trial design refinement.^23,24

The qualitative data from the semi-structured interviews were analyzed thematically by the lead researcher (N.C.). Interviews were transcribed by a paid transcriber and cross-checked for accuracy by N.C. and L.B. Data analysis followed the stages of thematic analysis described by Braun and Clarke: familiarization with the data, initial coding (conducted independently by N.C. and L.B.), searching for themes, reviewing and refining themes, defining and naming themes, and producing the report. NVivo 12 (QSR International, 2018) was used to support data management and analysis.

Results

Characterization at Baseline

Twelve PwPD expressed interest in participating in the study, of whom 11 were screened via Zoom interviews, and 7 enrolled to the study. Three participants were randomly allocated to the M-DTT group, and 4 to the C-DTT group. One participant withdrew after completing the first 2 weeks of the training because of an unexpected diagnosis of a pulmonary health condition. An adapted CONSORT flow diagram provides a summary of recruitment and follow-up stages ²⁵ (Supplementary Material 5).

Baseline characteristics of participants are presented individually in Table 1 of Supplementary Material 4. The total average age of participant was 71 years and the average disease duration following diagnosis was 11.42 years. According to the self-report UPDRS-II score, 4 participants were mildly affected by PD (score <17), and 2 were moderately affected (scores ≥17) in terms of impact of the disease on daily life.

Process Data as Indicators of Feasibility

Attendance rate to the main exercise sessions was high (number of views ≥18 of a potential 18 viewing sessions); all participants in both groups viewed each movie of the main exercises at least once (Table 2 of Supplementary Material 6).

Overall, the adherence rate (minutes of watched sessions) was high, exceeding a target of 80% of content across the 6-week training programme in both groups. The average percentage of minutes of completed main exercise sessions in the M-DTT group over 18 sessions was 97.25%, and 96.78% in the C-DTT group (Table 2 of Supplementary Material 6). The number of views exceeded the total number of sessions, as 1 participant completed more than 18 sessions (n = 20). Two participants in the C-DTT group watched each session but neither completed each session.

Safety

There was 1 adverse event in M-DTT group. This was recorded as a diagnosis of a pulmonary health condition that was not related to the training and the participant withdrew at the point of diagnosis. No other adverse events or near misses were reported.

Patient Reported Outcome Measures

Median enjoyment scores over 6 weeks were 7.0 (IQR 6.75-8.0) for M-DTT and 5.0 (IQR 1.0-6.0) for C-DTT. Median challenge scores were high in both groups: 8.0 (IQR 7-8) for M-DTT and 8.0 (IQR 6.63-8.75) for C-DTT (Table 3 of Supplementary Material 6).

Challenge

Bi-weekly Likert data showed that 5 of 6 participants consistently found balance tasks challenging, with half reporting difficulty maintaining foot position on training mats.

M-DTT participants largely viewed secondary motor tasks as challenging across all time points, except for 1 instance in weeks 3 to 4. All C-DTT participants reported secondary cognitive tasks as challenging by weeks 5 to 6; 4 of 6 did so in earlier periods, while the others were unsure (weeks 1-2) or rated them mostly not challenging (weeks 3-4). Perceived progression difficulty was evenly split between mostly challenging and mostly not challenging across all time points and groups (Table 4 of Supplementary Material 6).

Enjoyment

Most participants (4/6) rated balance tasks as enjoyable in weeks 1 to 2. M-DTT ratings declined in weeks 3 to 4 and were evenly split in weeks 5 to 6, whereas C-DTT ratings varied by week, with only 1 participant reporting balance tasks as not enjoyable.

Secondary motor tasks (M-DTT) were generally rated enjoyable in weeks 1 to 2 and 5 to 6, with split ratings in weeks 3 to 4. Secondary cognitive tasks (C-DTT) showed the greatest variability enjoyment ratings across weeks and participants. Overall, enjoyment ratings for cognitive tasks were less stable than for motor tasks (Table 5 of Supplementary Material 6) and are further informed by qualitative interview data.

Burden

Participant fatigue varied: all M-DTT participants found training fatiguing in weeks 1 to 4 but not in weeks 5 to 6, while half of C-DTT participants reported fatigue throughout. Only 1 participant initially found the equipment setup challenging. No participants reported difficulties with the technology or workbooks, and all considered home-based DTT safe (Table 6 of Supplementary Material 6).

Functional Balance

MiniBESTest scores were collected at baseline and post-intervention (Table 7 of Supplementary Material 6). The measure was acceptable as a primary outcome during the site visit. Although the study was not powered for effectiveness, M-DTT participants showed consistent signal of balance improvements, while changes in the C-DTT group were more variable.

Standing Body Sway

Changes in total sway speed varied by participant and condition. Five participants improved in the fixed stance (feet 4 cm apart). One M-DTT and 1 C-DTT participant improved across all conditions, consistent with MiniBESTest patterns (Table 8 of Supplementary Material 6). Variability in antero-posterior versus mediolateral sway suggests that training did not selectively affect a single plane of motor control, warranting evaluation in a powered trial. Use of a body-worn sensor during the site visit was acceptable to participants, and the data were straightforward to analyse for a physiotherapist new to the technology (N.C.).

Qualitative Interview Data

All participants (n = 6) who completed their training were interviewed, face-to-face, at the end of the balance assessment. Four of the participant’s supporters contributed to the interviews. After data analysis, 7 themes emerged: (1) An acceptable DTT balances challenge and enjoyment, (2) A home-based intervention has both advantages and disadvantages, (3) Dealing with the technology, (4) There are advantages and disadvantages to training with a buddy, (5) Training characteristics are manageable, (6) Research assessments are familiar but are they acceptable? (7) Suggestions from pwPD and supporters. The themes/subthemes with participant’s quotations can be found in Supplementary Material 6.

Participants found the DTT interventions challenging yet interesting and enjoyable. Challenge was viewed as important for motivation and for improving performance and balance. Perceptions of challenge and enjoyment varied with personal interests and prior exercise habits, highlighting the need for varied task types. The home environment was considered an appropriate training setting, offering flexibility and low perceived safety risk. Accessing videos and completing online Likert scales were manageable due to prior technology experience but feedback suggested that this could be improved with better ‘How to. . .’ information resources. Training with a buddy was acceptable and enjoyed. Some participants felt the session duration was difficult to fit into their schedules, and several believed more frequent or longer home practice would further improve balance, inferring that dosage needs reviewing ahead on onward trial. Travel for face-to-face assessments was burdensome for some participants and supporters, yet all valued the opportunity to undertake face-to-face assessment with a physiotherapist. Most participants (n = 4) would recommend the DTT programmes to others but suggested improving technical aspects of the videos and expanding task variety.

Discussion

This study provides the first feasibility and acceptability analysis of 2 home-based DTT programmes targeting functional balance for pwPD. The trial determined a mixture of green and amber lights across studied feasibility domains, indicating a review/modify onward action. The mixed methods approach synthesized quantification and contextualization of process data and patient reported outcomes. Enjoyment and challenge as acceptability measures enabled a rich understanding of present limitations and onward adaptations required before proceeding with onward study.

Although interviews highlighted areas for refinement, both programmes were safe, feasible, and acceptable. This is an encouraging finding and contrasts with some prior reports relating to other home-based programmes where low adherence rates ²⁶ can risk the potential for successful outcomes. ²⁷ Rich qualitative data provided honest and constructive feedback with suggestions to guide forward modification before wider study.

Contextual factors such as setting and delivery influence adherence. ²⁸ In this study, training buddies supported performance monitoring, score recording, session assistance, and equipment setup, reinforcing motivation and progress. These findings align with evidence that motivation and support promote adherence and acceptability. ¹² Participants further recommended adding cueing or coaching roles for training buddies and integrating guidance, consistent with research showing that simple instructions and reminders improve adherence. ²⁹

Participants reported limited task variety, impractical task combinations, and technical issues in the training movies. Future iterations should diversify secondary tasks, refine task combinations, and improve video quality. Despite these limitations, participants generally described the training as challenging and enjoyable—factors linked to motivation, adherence, and acceptability. ²⁹ Enjoyment may relate to exercise self-efficacy, which is known to enhance engagement. ³⁰

Although concerns exist regarding unsupervised home-based training, this study reported no adverse events or near misses, which supports existing evidence that suggests DTT is safe for pwPD.^10,31

Assessment visits and outcome measures performed well. Participants found 2 in-person assessments acceptable and consistent with preferences reported elsewhere. ³² While home-based assessments may improve access, participants valued standardised face-to-face clinical testing, reported concerns related to travel or fatigue burden.

Online progress reviews via Zoom every 2 weeks were also acceptable, suggesting a hybrid assessment model may be suitable for future clinical services and trial design; offering reliable data while reducing burden. ³³

Physical outcome data showed variable changes with no clear trends between groups. Both the MiniBESTest and body sway measures were acceptable to undertake, were fast to administer and together provided detailed balance information. Whilst not powered within this trial, agreement between directional changes in these outcome measures offers some confidence in future determination of effectiveness or non-effectiveness of the intervention if used together. Combined use may also help determine whether benefit is achieved at impairment as well as functional levels of training. Despite encouragement related to the feasibility of use of these outcome measures, the minimal clinically important difference for the MiniBESTest is 4.0 points ³⁴ and change in sway speed RMS remains undetermined. The maximum change in score in this trial was 3.0, indicating no meaningful individual improvement was observed. Although effectiveness cannot be determined within the scope of feasibility trial, this may indicate the need to re-design the dosage of the programme. Current evidence suggests that achieving balance gains requires at least 150 minutes of home-based exercise per week for 6 weeks, which is a factor of 2 greater than what these programmes are designed to deliver. ³⁵ Whilst the decision to restrict training duration to 30 minutes/session and frequency to 3 times per week was informed by initial patient engagement, feedback from our participants suggests that this could be increased alongside an increase intensity within each 30-minute session.

Limitations

This feasibility study was not powered to assess effectiveness or compare M-DTT with C-DTT and inherently the small sample (n = 6) and limited demographic profile limits generalizability. The study was conducted under financial constraints that limited video production, trial duration and use of a follow-up period. While the training buddy model enhanced safety and engagement, reliance on informal social or family support may exclude individuals without such support, potentially limiting participation among socially isolated individuals. Recruitment was constrained by opting for early feasibility analysis outside of the National Health Service to reduce burden during Covid-19 and was similarly impacted by reduced by Covid-19 concerns among people with Parkinson’s, with some declining participation to avoid non-essential contact. Recruitment rate was impossible to calculate owing to the pragmatic recruitment strategy employing local Parkinson’s UK Charity group leaders as gatekeepers and snow-ball sampling leading to an unknown population and remains a key domain to pilot ahead of onward power calculation for RCT design. Thematic analysis performed well yet did not opt to embed reflexivity within the analysis process. However, retrospective reflexive engagement did highlight how researcher positioning may hold potential to influence data, presenting a limitation in this study and a priority consideration for future work.

Another limitation is that viewing metrics and streamed minutes do not ensure correct or complete protocol execution; although cross-referenced with training diaries, these measures are inherently less robust than adherence data from supervised or observed sessions yet provide frugal proxy measures in lieu of more expensive monitoring options.

Conclusion

This study indicates that home-based, unsupervised M-DTT and C-DTT holds potential to serve as an acceptable and feasible adjunct to conventional rehabilitation for pwPD, aligning with earlier evidence that DTT is safe for those with mild to moderate balance impairment. A blended delivery model combining in-person balance assessments, online monitoring, and support from training buddies appears feasible. The findings also demonstrate that low-cost DTT programmes and frugal trial designs can be implemented, though several refinements are advised before evaluating effectiveness: Improvements to training content and technology, researcher blinding, inclusion of an attention-control group, and addition of a follow-up period. A definitive comparative trial of M-DTT versus C-DTT remains essential to guide targeted balance training, improve therapeutic outcomes, broaden access for those unable to travel frequently, and reduce the burden and fatigue associated with exercise in this population.

Future Directions

Future research should evaluate the effectiveness of motor-motor versus motor-cognitive dual-task training in a fully powered randomized controlled trial, with sufficient dosage to achieve clinically meaningful improvement in balance. Follow-up assessments and hybrid outcome measurement models should be included to assess the sustainability of effects while maintaining accessibility. In preparation for conducting a fully powered effectiveness trial, refinements made to the training programme should be first piloted, ideally within a National Healthcare setting to determine recruitment rates and further inform sample size calculations.

Intervention refinement should engage pwPD and prioritize inclusivity, adherence, and fidelity by offering alternatives to informal training buddies, flexible delivery formats, and adaptable task progression. Adherence and fidelity may be further enhanced through structured cueing, regular feedback, increased task variety, and enhanced monitoring approaches. Low cost activity sensor-based logs, structured task checklists, and optional video-based verification may ensure accurate capture of adherence, as viewing metrics and streamed minutes alone cannot ensure correct or complete protocol execution.

Supplemental Material

sj-pdf-1-hhc-10.1177_10848223261423908 – Supplemental material for Home-Based Dual-Task Balance Training for Parkinson’s Disease: A Feasibility Study of a Randomized Controlled Trial

Supplemental material, sj-pdf-1-hhc-10.1177_10848223261423908 for Home-Based Dual-Task Balance Training for Parkinson’s Disease: A Feasibility Study of a Randomized Controlled Trial by Nesibe Cakmak, Jenny Freeman, Camille Carroll and Lisa Bunn in Home Health Care Management & Practice