Showing papers by "Kevin E. Vowles published in 2009"

Technologically-assisted behaviour change: a systematic review of studies of novel technologies for the management of chronic illness.

Abstract: A systematic review was conducted to investigate the use of technology in achieving behaviour change in chronic illness. The areas reviewed were: (1) methods employed to adapt traditional therapy from a face-to-face medium to a computer-assisted platform; (2) targets of behaviour change; and (3) level of human (e.g. therapist) involvement. The initial literature search produced 2032 articles. A total of 45 articles reporting 33 separate interventions met the inclusion/exclusion criteria and were reviewed in detail. The majority of interventions reported a theoretical basis, with many arising from a cognitive-behavioural framework. There was a wide range of therapy content. Therapist involvement was reported in 73% of the interventions. A common problem was high participant attrition, which may have been related to reduced levels of human interaction. Instigating successful behaviour change through technological interventions poses many difficulties. However, there are potential benefits of delivering therapy in this way. For people with long-term health conditions, technological self-management systems could provide a practical method of understanding and monitoring their condition, as well as therapeutic guidance to alter maladaptive behaviour.

Classification of health level from chronic pain self reporting

Abstract: This paper proposes an approach to identify patients' health levels based on the information gathered following a process of self reporting based on the patient's current condition. The goal of approach is the accurate provision of information to assist with self management of chronic pain. Four supervised classifiers, namely decision tree, naive Bayes, support vector machine and multilayer perceptron, have been applied to classify the health level of patients suffering from chronic pain based on information collected from self reports from three treatment stages - pre-treatment stage, post-treatment stage and 3-month follow-up stage. Three binary classification problems, i.e. pre-treatment vs. post-treatment, pre-treatment vs. 3-month follow-up and post-treatment vs. 3-month follow-up, were investigated. The classification accuracy and area under Receiver Operating Characteristics (ROC) curve ranged from 66.7% 94.7% and 0.689 0.989 respectively. The multilayer perceptron classifier achieved the best performance with a classification accuracy of 94.7% and area under ROC curve of 0.981 for the pre-treatment vs. post-treatment classification. The results from this study have demonstrated that it is feasible to apply automated classification techniques to identify patients' health level from their self reports. This data may be used as an important indicator in automated approaches to chronic disease self management, an area which is currently receiving much attention. Further work will investigate the presence of optimal features derived from questionnaires to improve the classification performance.

The experience of pain and suffering from acute and chronic pain

Abstract: Pain can create huge problems, whether this pain is transient "everyday pain," acute post surgical pain, recurrent disease-related pain, or long-standing persistent pain. These problems include disturbed daily functioning, emotional suffering, poor general health, high healthcare use, and high healthcare costs, among others (e.g., Breivik, Collett, Ventafridda, Cohen, & Gallacher, 2006). However, while pain is ubiquitous in human experience, these consequences do not occur in all cases. For researchers and healthcare providers to better understand and manage pain, it is important to accurately discriminate differing pain experiences, identify their initiating causes, and identify the influences that both maintain the pain itself and determine impacts of pain on the daily lives of pain sufferers.

Progressing towards acceptable treatment outcomes.

Abstract: In this issue of Pain, Thorne and Morley [11] describe the amount of improvement necessary to achieve an ‘‘acceptable” outcome in pain severity, pain impact/interference, and activity in a sample of individuals seeking treatment for chronic pain. This paper has several noteworthy findings and raises a number of issues deserving extended discussion, of which three will be focused upon here. First, perhaps the most striking finding of the study pertains to the amount of change deemed necessary by participants to achieve an acceptable treatment result. For all four measures assessed, even the lower end of the 95% confidence interval for effect sizes was quite far beyond what is typically defined as a large effect size. To our knowledge, there is no intervention available for chronic pain that is able to achieve such sizable treatment effects. It would be easy to conclude, therefore, that acceptable treatment outcomes are not possible with the current approaches; however, such a conclusion would be ill informed given the literature that suggests good satisfaction with the treatment of chronic pain generally [5,8]. An alternate conclusion, which the authors themselves allude to, is that one purpose of treatment is to alter patient criteria for acceptable outcomes such that they are clinically feasible – essentially, a ‘‘recalibration” of what is deemed acceptable. Evidence of this recalibration has been demonstrated where patient criteria for success became less strict with treatment [2]. It is even possible to take this argument one step further: It may be that successful treatment depends in some way on this recalibration. Clinically speaking, demoting the importance of pain reduction and promoting the importance of other factors, such as progress towards meaningful living, may be crucial. This point is discussed more fully later in this editorial. A second key finding of Thorne and Morley pertains to the fact that a number of outcome domains were assessed, presumably as the authors were aware that pain reduction is not the only method to achieve patient satisfaction (e.g., [6]). In fact, a rough, and admittedly statistically imprecise, conclusion that can be drawn from the ‘‘desired change” data displayed in Table 2 is that less pain reduction was desired than reduction in the interference pain was having on functioning. The work of Robinson et al. [10] revealed a similar result, where the amount of change required to meet an acceptable reduction in pain’s intensity was significantly lower than desired reductions in its interference. These findings echo the decades old work of Fordyce and colleagues [4], as well as more recent consensus panels [3] – pain reduction is but one measure of treatment outcome and may be neither strongly related to other outcomes nor the most important area for improvement. This does not necessarily suggest that pain intensity should be ‘‘demoted” in

Integration of assistive technology to support self management of chronic disease

Abstract: Assistive technology has been used to assist people with different types of chronic disease, to monitor, remind and guide them through activities of daily living (ADL) in the home environment. This research aims to consider the integration of assistive technology to support self management for three chronic diseases, namely stroke, chronic pain and congestive heart failure. A personalised self-management system (PSMS) is under development. This paper outlines the PSMS infrastructure and presents the current work on the system design. (4 pages)

Feature selection on chronic pain self reporting data

Abstract: Chronic pain is a common long-term condition that changes patients' physical and emotional functioning. Currently, the integrated biopsychosoical approach is the mainstay treatment for patients with chronic pain. Self reporting (the use of questionnaires) is one of the most common methods to evaluate treatment outcome. Nevertheless, a large number of questions (for example 329 questions in this study) may be required and as such may be viewed as not being convenient for patients to complete. This paper has applied the theory of information gain to rank the questions in addition to investigating important factors related to the treatment outcome. Analysis within the study ranked the questions from 1 to 329 based on information gain (highest to lowest). Results showed that questions related to chronic pain coping strategies and value-based actions had high information gain. Four supervised learning classifiers were used to investigate the correlations between feature numbers and classification accuracy. The results showed classifier that a Multi-layer Perceptron classifier obtained the highest classification accuracy (96.05%) on an optimized subset which consisted of 133 questions.