Patient education and self-management programs in arthritis

For funders of healthcare, the allure of patient education and self-management programs includes the promise of reduced costs alongside higher patient satisfaction. This has placed self-management on to the agenda of local and international health authorities. Local authorities are currently discussing whether large-scale organised self-management programs should be implemented in Australia.1 This agenda is consistent with the goal of the World Health Organization’s Bone and Joint Decade that “. . . the informed and empowered patient must be a critical part of any therapy”.2 Furthermore, international clinical practice guidelines recommend that self-management be part of routine clinical practice.3 Despite this enthusiasm for self-management, the potential benefits of such programs may not be immediately apparent for healthcare professionals and their patients.

There are a variety of models of both formal and informal self-management. However, the core concepts involve engagement in self-care, improved self-monitoring, interactions with healthcare professionals and coping with disease (Box 1). The dominant model in Australia is the Stanford University School of Medicine chronic disease self-management program (Box 2). This course runs for 2.5 hours per week over 6 weeks and is led by peers or health professionals. The arthritis self-management program (ASMP) is the core business of the Australian Arthritis Foundation (www.arthritisfoundation.com.au).

Outside the research setting, self-management programs are mostly provided to people with arthritis and other chronic diseases through community organisations such as Arthritis Foundations (see Box 3). In the United States, courses are also sometimes available to members of Health Maintenance Organisations (HMOs), and are reputedly highly endorsed.4 In the United Kingdom, the Stanford model is being used as a vehicle for healthcare reform through the Expert Patients Program.6 This program is designed to integrate user-led self-management courses for people with chronic diseases into the existing National Health Service (NHS) frameworks. Over 150 Primary Care Trust sites across England are currently taking part, and an estimated 20 000 patients will receive training in this initial phase. From 2004 to 2007, it is proposed that the program will become part of mainstream services throughout the NHS.

Existing evidence for the “clinical” effectiveness of self-management is summarised in Box 4, using information from two recent meta-analyses.7,8 While many studies have been published, they can be difficult to interpret for several reasons:

• many of the intended benefits do not fit into the traditional medical model and are difficult to measure and qualify (eg, “empowerment”);

• many trials include “chronic diseases” without disease-specific subanalyses (ie, for arthritis);

• patients studied are often from selected subpopulations; and

• self-management interventions are delivered in addition to usual care (which is often of a very high standard in clinical research centres).

The meta-analyses for both rheumatoid arthritis and osteoarthritis indicate small average benefits for clinical outcomes (Box 4). Larger benefits have been reported for “illness distress” and “self-efficacy” (confidence in undertaking healthful behaviours), which are critical components of quality of life for patients with a chronic disease.4,5,9 Most controlled trials have a relatively short follow-up period, as participants are usually randomly allocated to wait-list control groups who receive the intervention within 4–6 months.4,9 Our 2-year longitudinal study of people with arthritis attending the ASMP found that the benefits that were apparent at 6 months (eg, less pain and fatigue, increase in physical activity) were still apparent at 2 years.5

The association between self-management and health service use is more complex. As patient education and self-management programs aim to improve the efficiency and quality of the healthcare process, it may be unrealistic to expect a reduction in doctor visits, hospital attendances or medication use. For some individuals, an improved healthcare process could include more frequent use of services because of improved compliance. In addition to better compliance, a better “mix” of healthcare resources might eventuate, including reduced use of unplanned services (eg, emergency presentations). Healthcare use outcomes from controlled trials of self-management reflect this complexity, with somewhat inconsistent outcomes for attendances at GPs, emergency presentations, and medication use.4,5-10

“Program logic” assists in conceptualisation of complex programs, and helps to identify measurable targets in the “causal” chain.11 Through workshops with stakeholders, we have developed a program logic of how the outputs of self-management, if delivered widely across the community, might lead to effects on individuals, the healthcare system and the population (see Box 5). Immediate effects involve education and patient empowerment, including improved collaboration within the healthcare system (see Box 2). These are the most basic effects, consistent with course content and are the most easily measured. Standardised indicators are being developed (see below), and will assist in ascertaining whether course participants actually receive these benefits. The difficulty of demonstrating outcomes increases as one progresses down the model (eg, clinical outcomes such as pain, quality of life and disability). Some of these outcomes can be measured reliably by questionnaire (see the non-italicised components in Box 5). Benefits to communities and ultimately to public health are the most difficult to demonstrate, and such outcomes are highly dependent on the coverage of the program in patient groups most likely to benefit.

In an attempt to standardise the quality of patient education programs, the National Arthritis and Musculoskeletal Conditions Advisory Group12 has funded the development of a national quality assurance program. A standardised questionnaire to measure outcomes, the Health Education Impact Questionnaire (HEI-Q), has been developed to assess eight proximal indicators — positive and active engagement in life, health behaviour change, skill and technique acquisition, constructive attitude shift, self-monitoring, healthcare service navigation, social integration, and emotional wellbeing. These indicators were developed through consultation with stakeholders and workshops. In addition to data on course quality, the HEI-Q will provide systematic data on where and to whom self-management programs are being delivered.

In Australia, there is interest in the potential value of patient education and self-management programs, particularly the ambitious UK Expert Patients Program.1 This program (Box 3) is a lay-leader model; it is unknown whether program quality will be maintained or whether it will reach those who might benefit most. A current Australian policy initiative is the Commonwealth Sharing Health Care Initiative,13 which is part of the Enhanced Primary Care (EPC) package. Over 4 years, $14.2 million was allocated to the development and implementation of a variety of chronic disease education models, including self-management. This program is implemented through the Divisions of General Practice, various non-government health organisations, academic centres and some state and territory health authorities. The first stage of this program should be completed during 2004 and will yield data on the applicability of a range of models in a variety of Australian settings and health conditions, including people with Aboriginal or Torres Strait Islander backgrounds. A further $21.8 million over 4 years was provided by the Federal Government in the 2003–2004 Budget to further integrate chronic condition self-management into the Australian healthcare system.

Preliminary Australian work on 2-year outcomes in 452 people attending ASMPs shows that, in a self-selected population, most individuals are female and educated.5 Our current nationwide studies are designed to be sensitive enough to identify a broad range of small intervention effects using new, sensitive questionnaires for community-based interventions. We have found that only about 6% of referrals to self-management courses come from GPs or hospitals. These studies, as well as the findings of the 12 demonstration projects in the Commonwealth Sharing Healthcare Initiative,14 will provide much-needed data to inform any future integration of patient education programs in Australia.