The ability to perform two or more tasks together is impaired in Parkinson's Disease (PD). Based on pilot work we hypothesize that dual tasking is amenable to training in PD and will not compromise safety. In the proposed study we will therefore collect high level evidence on whether 6 weeks of integrated dual task gait training is more effective than consecutive task practice. The investigators will use a randomized, single blind study design and conduct the same protocols in two academic centers (Katholieke Universiteit Leuven and Radboud Universiteit Nijmegen). The investigators intend to test an experimental condition which consists of focused dual task training in the home using a novel program of personalized cognitive tasks. The aim of dual task training is to teach motor-cognitive task integration and achieve optimal levels of automaticity and functionality. The control arm will receive gait practice and separate cognitive training of the same intensity, but offered consecutively. We will test the hypothesis that dual task training (integration) will have more pronounced effects on complex gait than consecutive task training.

Sub-analysis will be conducted on patients with and without freezing of gait. The investigators expect that integrated dual task training may be less effective in patients with freezing, due to the increased fall risk and impaired cognitive profiles.

Overall, this project will provide evidence to support future directions for motor learning and innovative rehabilitation targets.

Purpose:

The ability to perform two or more tasks together - for instance walking and talking - is impaired in Parkinson's Disease (PD). As a result, dual tasking might evoke gait deterioration with an increased risk of falling in PD. In the European guideline for physiotherapy, it is therefore recommended to advise PD patients to avoid dual tasking. However, based on clinical experience and pilot work we argue that rather than avoiding, dual tasking should be trained and this may enhance safety.

At present, there is little clinical evidence about the effectiveness of attention-related exercise in PD. The RESCUE-trial showed sustained reductions of dual task interference after a short period of combined single and dual task cued gait training without increasing fall risk. These effects were cue-related and not contrasted with an active control group. Dual task training with a variable locus of attention showed very effective in comparison to single task training in achieving retention and transfer effects in balance-impaired older people. This randomized study also showed that dual tasking led to greater improvements of cognitive function compared to single task intervention. In addition, recent study showed dual task training to be surprisingly effective in mild to moderately affected patients with dementia, who were randomly assigned to dual tasking or unspecific exercise.

Given the positive effects of these clinical studies, the effectiveness of dual task training needs to be urgently addressed in PD patients, in whom both motor and cognitive problems are apparent. The fact that patients with freezing of gait (FOG) are associated with a surplus decline of attention and have a higher fall risk, begs the question whether dual task training is equally effective in this important clinical subgroup. Therefore, we propose to obtain high level evidence on whether dual task gait training is as effective and safe in PD-patients with FOG. The overall purpose of this project is to guide future directions for innovative rehabilitation targets.

Research objectives:

1. To determine whether intensive and integrated dual task gait training leads to sustained improvements in complex gait performance and is more effective than consecutive task training. The study will focus on influencing the ability to resist dual task interference, the ability to transfer learning effects to an untrained but related task and the ability to retain the learning increments after a period without training (12 weeks).
2. To test whether integrated training of dual task walking is a safe intervention in PD and does not increase fall risk.
3. To determine whether disease stage or disease profile (with and without FOG) and cognitive factors predict the effect of dual task training.

Design:

The design will consist of a multicentered single blind randomized controlled trial in which PD patients are either allocated to a control group (Consecutive Task Training, "CTT") or to an experimental group (Integrated Dual Task Training, "IDT"). Patients in both arms of the study will continue their usual care (including standard physiotherapy) as normal. In case patients receive physiotherapy, we will inform the therapists involved and make sure that we ascertain the contents of normal care and avoid overlap. The trial will be conducted in two countries Belgium and the Netherlands, which are similar in terms of language and healthcare facilities.

Randomization:

The overall timeline of the study is 3 years. Subjects will undergo a control period of six weeks without intervention. This allows us to investigate the effects of repeated measures. Subsequently, patients will be randomized to one of the two groups within each center: 1) IDT or 2) CTT. All subjects will participate in an intensive, progressive training protocol consisting of 18 sessions (3 per week for 6 weeks) and up to one hour per session. Assessors will be unaware of group allocation. Randomization will be stratified with respect to subgroups (i.e. with or without FOG and Hoehn & Yahr stage II and III) to ensure adequate representation and diversity. A computerized block randomization will be performed per centre. Freezers will be defined by a score >1 on the New Freezing of Gait Questionnaire.

Power calculation:

Based on the literature and our previous experience, we anticipate that DTT will improve with an average of 20% and STT with 5% of baseline dual task gait speed (during a novel task). Assuming an alpha of 0.05, we calculated our sample size with dual task gait speed as the main outcome. On the basis of the RESCUE data, (including patients with >5% DT interference) we found a mean dual task gait speed of 0.766 m/s and SD=0.21. Based on these figures we estimated that the total sample size would be N=108 (54 per centre). Incorporating a loss to follow-up of about 5% brings the final figure up to 60 patients per centre to be recruited over a period of two years (N=120).

Intervention:

Integrated dual task training will be conducted at home by a physiotherapist, especially trained in the intervention. It will consist of gait training (straight-line gait, turning and manoeuvring round the house) while using a standardised cognitive exercise protocol with an MP3 player. To address the correct level of learning, the levels of task complexity will be matched to the patient's baseline characteristics, considering the following components: baseline DT interference, previous fall risk and presence of FOG. Given the cognitive deterioration in PD, implying a reduced learning capacity, a fixed priority training (attention priority on both tasks) will be implemented, aimed to improve better and safe walking when distracted by other attention demands. Every session will also include training of a functional dual task, relevant for each patient. Progression of treatment will be standardized. Treatment will consist of 18 sessions over a period of 6 weeks. Homework will be conducted with the partner for safety reasons. Intervention will be additional to regular care (including physiotherapy). A follow-up period of 12 weeks without training will be used to test retention.

Consecutive task training will be conducted at home by a physiotherapist. It will consist of traditional gait training(straight-line gait, turning and maneuvering). Patients will get separate training of cognitive tasks while sitting, using the same program via an MP3 player as described above. Every session will also include training of a functional task, relevant for each patient, but dual tasking will be positively discouraged. Functional training will thus consist of carrying out task components separately. Progression and intensity of treatment will be the same as IDT.

Co-interventions:

During the 6 weeks of IDT or CTT, all other interventions (medication, allied health care etc) will be kept as stable as possible. Any changes in treatment will be monitored.

Subjects:

Patients will be recruited by a movement disorders specialist at each medical center involved. The following clinical descriptive outcomes will be collected: Disease duration, Hoehn & Yahr stage, Unified Parkinson's Disease rating scale, New Freezing of Gait Questionnaire, Cognitive scores, Beck's Depression Index, and general gait performance (2 minute walk test).

Test procedure:

Testing will be conducted by an independent therapist at each centre. Patients will be followed up over a period of 24 weeks in total. Four testing moments will be foreseen: 1) one baseline test at study entry and the second baseline tests after a 6 week control period without intervention; 2) one test after 6 weeks of training and 3) one test after 12 weeks follow-up without training. All spatiotemporal gait outcomes with and without DT will be measured during normal speed conditions and therefore will require transport of the patients to the gait laboratory. Clinical testing will also be conducted in this environment. Both centers use the same GAITRite Walkway System in their gait laboratory, which is considered to be the 'gold standard' instrumented system for measuring spatiotemporal gait parameters. The untrained secondary task will be an Auditory Stroop task delivered by a wireless headphone. A microphone will record the verbal responses of the patients, allowing detailed measurement of the reaction times of patients' responses. This system is currently used in one of the centers (Katholieke universiteit Leuven). Secondary task performance will be measured during sitting as well as walking.

Tests:

1. DT gait performance will be evaluated by measuring gait speed while performing the untrained cognitive task, i.e. the Auditory Stroop.
2. Safety will be assessed by determining fall rate ascertained by a weekly telephone call.
3. We will also test cognitive performance during a trained dual task (Digit Span Backward) and during a functional task (using a mobile phone). At baseline, the appropriate level of cognitive challenge for each patient will be determined. We will measure gait and secondary task performance separately and calculate dual task interference outcomes.

Other outcomes will be

• Single task gait performance: speed, stride length and cadence;
• Single task cognitive performance (during sitting);
• Activities-specific Balance Confidence scale (measure of function);
• Tests of executive function and attention;
• Freezing of Gait Questionnaire
• ADL scale
• PDQ39 Quality of LIfe scale

Inclusion Criteria:

1. Hoehn & Yahr stage II-III in the on-phase
2. Able to walk for 10 minutes continuously
3. Dual task interference of at least 5% using a cognitive secondary task
4. Without cognitive impairment (MMSE > 24)
5. On stable medication
6. Diagnosis of Parkinson's disease made by a movement disorders specialist

Exclusion Criteria:

1. Deep Brain stimulation
2. Living alone
3. Severe medical conditions affecting gait
4. Hearing problems
5. Likely to change medication regimen.