Increasing activity in patients with Parkinson’s disease

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f1429 (Published 06 March 2013) Cite this as: BMJ 2013;346:f1429
  1. E D Playford, reader in neurological rehabilitation
  1. 1UCL Institute of Neurology, London WC1N 3BG, UK
  1. d.playford{at}ion.ucl.ac.uk

Crucially important but requires complex long term support

There is no doubt about the benefits of exercise. Physical activity is associated with a reduced risk of coronary heart disease, obesity, type 2 diabetes, and other chronic diseases and conditions.1 Despite the wealth of information about the benefits of exercise, sedentary adults often find it hard to change their behaviour in a sustainable manner. The development of interventions that support behaviour change is therefore of general interest. Social cognitive theory,2 which explains why people behave as they do, has been used to develop strategies that support changes in behaviour. Examples of such strategies include encouraging people to choose from several options, goal setting, and encouraging clients to monitor their progress through record keeping. In a linked paper (doi:10.1136/bmj.f576), van Nimwegen and colleagues use such strategies to encourage a long term increase in participation in activity among patients with Parkinson’s disease.3

It is more important and more difficult for patients with Parkinson’s disease (and other long term neurological conditions) to increase exercise levels than it is for healthy adults. More important because, in addition to the cardiovascular benefit, exercise may have a positive effect on the non-motor symptoms of Parkinson’s disease, such as fatigue, insomnia, depression, and constipation. In addition, it may even be neuroprotective. However, it is more difficult because the motor symptoms of Parkinson’s are often associated with cognitive deficits such as difficulties with attention, executive functions, working memory, and visuospatial functions.

Van Nimwegen and colleagues therefore developed a multifaceted intervention with a strong theoretical basis to facilitate an increase in activity among these patients, which is clearly described in this well designed trial. Participants recorded changes in activity in an activity diary, and triaxial acclerometers were used to supply objective data on activity. The study found a positive effect of the intervention for these measures but not for the primary outcome measure—the difference between groups in hours spent in physical activity as estimated using the LASA physical activity questionnaire (the control group spent 12 hours a week in physical activity and the intervention group 12.5 hours).

The study represents an important advance in our understanding of sustainable change in people with long term neurological conditions. People with a disability experience several barriers to exercise. These include the fear and embarrassment of exercising (especially in the presence of fit and healthy people) and perceptions that fitness instructors will lack knowledge about their condition and how to help them participate in exercise safely and effectively.4 Conversely, individually tailored gym programmes, an exercise environment that actively supports people with similar conditions and disabilities, and exercise programmes that consider individual motivators for exercise facilitate exercise in such groups.4

A United Kingdom based randomised control trial that looked at these factors as part of the intervention demonstrated its acceptability to people with disabilities (mean 14; standard deviation 9; range 0-39 gym attendances per participant) over 12 weeks, but it was unable to show an overall increase in activity.5 At the end of the study only 34% of the participants were sure they would continue going to the gym, with 43% being undecided.5 Similarly, a recent systematic review of physiotherapy interventions encompassing general physiotherapy, exercise, treadmill training, cueing, dance, and martial arts found that physiotherapy improved mobility but not function over the short term (less than three months).6

In contrast to these short term changes, van Nimwegen and colleagues’ study demonstrated engagement in the exercise programme over two years. Their intervention had several key elements. Physiotherapy and physical activity coaching were delivered by physiotherapists. Patients were given specific information about the benefits of exercise and a large number of elements informed by social cognitive theory were incorporated into the programme. These included the identification of potential barriers and means to overcome them, systematic goal setting with a health contract and log book, encouragement to participate in group exercise, an ambulatory monitor with automated feedback, and the ability to upload data to a website where patient and coach could monitor progress. Importantly, physiotherapy and coaching sessions occurred regularly (19 and 14 times, respectively) over the two years.

Van Nimwegen and colleagues acknowledge that their intervention was complex and suggest that the next steps may include identifying the active elements. However, there is a danger in deconstructing the intervention. The whole intervention may be greater than the sum of its parts—individually each of the elements may not be effective. Two factors stand out, however. One is the focus on behaviour change and the other is the long term engagement over two years. In most healthcare systems access to treatments for people with long term conditions is poor. In the Netherlands, 63% of people with Parkinson’s disease have had physiotherapy, 9% occupational therapy, and 14% speech therapy.7 In a UK study of 340 people with Parkinson’s disease, fewer than a third had seen an allied health professional.8 In addition, in many healthcare systems interventions are typically short, lasting about six weeks and focusing on teaching new approaches. Long term support will demand greater resources. Justifying such change demands convincing health economics, but before this can be attempted there has to be clear evidence that these interventions work.

The interpretation of this and other studies has been complicated by the outcome measures. People with Parkinson’s disease and other long term conditions value therapeutic input highly, and it may be unwise to ignore their intuition. In the LIFE study, activity levels did not seem to increase despite people going to the gym more often. In the current study, there was no significant effect on the primary outcome measure, the LASA physical activity questionnaire score, despite diary records and accelerometers suggesting an increase in overall activity. There are two possibilities. The first is that the diary records and accelerometers are correct and participants did increase their activity. The second is that, despite this effort, achieving behaviour change is extremely difficult in a motor and cognitively impaired population. Technologies that are sensitive to movement and activity—both inside and outside the home—may offer a way forward, but we still need psychometrically robust patient reported outcomes that capture the domains examined in this and similar studies.


Cite this as: BMJ 2013;346:f1429


  • Research, doi:10.1136/bmj.f576
  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Commissioned; not externally peer reviewed.