5Education and self-management

Publication Details

5.1. Patient information

5.1.1. Clinical introduction

There is limited disease-specific evidence on the benefits of providing information for osteoarthritis. It is essential that the consultation is one of information sharing and achieving concordance in the treatment regimes suggested (Cox et al. 2004; Elwyn et al. 1999). Recognising that the patient should be treated as an individual and not as a disease state is imperative in improving communication and outcomes (Donovan and Blake 2000).

People will vary in how they adjust to their condition or instigate changes as a result of the information and advice provided. This is likely to depend on a number of factors:

  • the disease severity and levels of pain, fatigue, depression, disability or loss of mobility
  • prior knowledge and beliefs about the condition
  • the social and psychological context at the time
  • health beliefs and learnt behaviours.

5.1.2. Methodological introduction

We looked for studies that investigated:

  • the effectiveness of patient information provision/education methods compared with each other or to no information/education
  • the effectiveness of patient self-management programmes compared with each other or no self-management
  • both with respect to symptoms, function, quality of life.

Due to the large volume of evidence, studies were excluded if they used a mixed arthritis population of which less than 75% had osteoarthritis or if population was not relevant to the UK.

Two systematic reviews and meta-analyses (MA) (Chodosh et al. 2005; Superio-Cabuslay et al. 1996), eight RCTs (Heuts 2005; Calfas et al. 1992; Nunez et al. 2006; Victor and Triggs 2005; Buszewicz et al. 2006; Maisiak et al. 1996; Pariser et al. 2005; Keefe and Blumenthal 2004), one implementation study (De Jong et al. 2004) and one observational study (Hampson et al. 1993) were found on patient education and self-management methods. Two of these studies (Pariser et al. 2005; Keefe 2004) were excluded due to methodological limitations.

The first MA (Chodosh et al. 2005) included 14 RCTs on osteoarthritis self-management programmes compared with usual care or control programmes (attending classes which were unrelated to osteoarthritis self-management). Follow-up was between 4–6 months for all studies. The quality of the included RCTs was assessed but the results of this are not mentioned. The MA pooled together all data for the outcomes of pain and function.

The second MA (Superio-Cabuslay et al. 1996) included ten RCTs/CCTs on osteoarthritis patient education (information about arthritis and symptom management) compared with control (types of controls not mentioned). Quality of the included RCTs was not assessed. The MA pooled together all data for the outcomes of pain and functional disability. Studies differed with respect to sample size and duration.

The six RCTs not included in the systematic reviews were all randomised, parallel group studies but differed with respect to:

  • osteoarthritis site (two RCTs knee, two RCTs hip and/or knee, two RCTs not specified)
  • treatment (five RCTs group sessions of self-management/education programmes, one RCT telephone intervention – treatment counselling and symptom monitoring)
  • comparison (two RCTs usual care, two RCTs waiting list, one RCT education booklet, one RCT education lecture)
  • trial size, blinding and length.

The implementation study (De Jong et al. 2004) was methodologically sound and compared the effects of a 6-week knee osteoarthritis self-management programme (N=204 patients) and a 9-week hip osteoarthritis self-management programme (N=169 patients) with pretreatment values in patients from urban and semi-rural communities.

The observational-correlation study was methodologically sound and consisted of giving questionnaires to, and interviewing, N=61 osteoarthritis patients in order to assess their use of self-management methods to deal with the symptoms of osteoarthritis.

5.1.4. From evidence to recommendations

There is a significant body of evidence in the field of social and psychological research on health behaviours in the context of information-giving and health-seeking behaviours and subsequent attitudes to treatments offered (Ajzen and Fishbein 1980; Carr and Donovan 1998; Donovan et al. 1989). Evidence has demonstrated that patients fail to retain all the information provided during a consultation. Lay health beliefs, perceived threat of the condition or treatments prescribed as well as time taken to adjust to the diagnosis all have an effect on an individual’s ability to retain information and make changes to their health behaviours of concordance with treatments.

Although it is clear that many patients want more information than they currently receive, not all individuals will wish this. The degree to which people may wish to be involved in decisions about their treatment is likely to vary. Evidence suggests individuals may adopt one of three approaches when asked to make treatment decisions on their own (Coulter and Ellins 2006), wishing to:

  • select their own treatment
  • choose to collaborate with the healthcare professionals in making a decision
  • delegate this responsibility to others.

Patient education is an information-giving process, designed to encourage positive changes in behaviours and beliefs conducive to health (Ramos-Remus et al. 2000). Patient education varies in content, length and type of programme (planned group sessions or tailored one-to-one sessions).

There are three components to patient education.

  • General information given to provide an overview of the condition to aid understanding and enable discussions about changes in health status.
  • Specific information given to encourage positive health-seeking behaviours that can improve patient self management and outcomes – for example, exercise in osteoarthritis.
  • Information given about benefits and risks to aid informed consent.

There is a professional responsibility to ensure that patients are provided with sufficient and appropriate information about their condition. Patient education is an integral part of informed decision-making. In addition, within the wider context patient education has been advocated as a way of limiting the impact of a long-term condition (Department of Health 2005).



Healthcare professionals should offer accurate verbal and written information to all people with osteoarthritis to enhance understanding of the condition and its management, and to counter misconceptions, such as that it inevitably progresses and cannot be treated. Information sharing should be an ongoing, integral part of the management plan rather than a single event at time of presentation.

5.2. Patient self-management interventions

5.2.1. Clinical introduction

Self-management can be defined as any activity that individuals do to promote health, prevent disease and enhance self-efficacy. Individuals who are able to recognise and believe in their ability to control symptoms (self-efficacy) can become more active participants in managing their condition and thus potentially improve their perceived control over their symptoms. This may improve concordance with treatment options offered and reduce reliance on healthcare interventions (Cross et al. 2006; Cox et al. 2004).

Providing a framework for patients that encourages self-management is now considered an integral aspect of care for all long-term conditions. Self-management principles empower the patient to use their own knowledge and skills to access appropriate resources and build on their own experiences of managing their condition. Not all patients will wish to self-manage or be able to achieve effective strategies and practitioners should be aware of the vulnerable groups who may require additional support.

5.2.2. Evidence base

The evidence for this self-management section was searched and appraised together with that for patient information (section 5.1)

5.2.3. From evidence to recommendations

Educational initiatives that encourage self-management strategies should be encouraged, although it has to be recognised that such support appears to have limited effectiveness from eligible UK studies to date. This may relate to a number of limitations including the range and diversity of outcomes measured and disparities in severity and site of osteoarthritis. Studies exploring key concepts such as self-efficacy and wider psychological and social factors were lacking. There are also important additional factors in the context of osteoarthritis as lay expectations – and to some extent healthcare professionals’ expectations – of good outcomes are somewhat negative and access to readily accessible support and advice are generally poor. These perspectives are likely to influence outcomes.

The members of this working group have considered these limitations yet accept that with the expected changes in the population – with a doubling of chronic disease and elderly patients by 2020 – the healthcare system has to consider encouraging a greater degree of self-management principles in line with current health policy. If longer-term outcomes are to be achieved, such as reduction in the use of health resources, effective use of therapeutic options and more adequately prepared and informed patients seeking interventions such as joint replacement surgery, then self-management may be an appropriate and cost-effective tool.

There will be a range of providers including voluntary and independent sectors who will be offering self-management programmes. These programmes will require a thorough evaluation of outcomes achieved at a time when primary care will also be enhancing the infrastructures and support for those with osteoarthritis requiring healthcare support.



Individualised self-management strategies should be agreed between healthcare professionals and the person with osteoarthritis. Positive behavioural changes such as exercise, weight loss, use of suitable footwear and pacing should be appropriately targeted.


Self-management programmes, either individually or in groups, should emphasise the recommended core treatments (see Fig 3.2) for people with osteoarthritis, especially exercise.

5.3. Rest, relaxation and pacing

5.3.1. Clinical introduction

It would seem sensible if something hurts to rest it. This may only be true in acute situations and may not hold for chronic conditions. It is counter productive to give rheumatoid arthritis patients bed rest. Muscle loss is a feature of both rheumatoid and osteoarthritis. Pain does not mean harm in many musculoskeletal conditions. We have looked at the effect of exercise on osteoarthritis especially of the knee, but where do rest, relaxation and coping strategies fit?

5.3.2. Methodological introduction

We looked for studies that investigated the efficacy and safety of rest and relaxation compared with no treatment or other interventions with respect to symptoms, function and quality of life. Three RCTs (Gay et al. 2002; Garfinkel et al. 1994; McCaffrey and Freeman 2003) were found on relaxation, yoga and listening to music. One RCT (Garfinkel et al. 1994) was excluded due to methodological limitations. No relevant cohort or case-control studies were found.

Two RCTs did not document blinding or intention to treat (ITT) analysis. One RCT (Gay et al. 2002) compared Erikson hypnosis with Jacobson relaxation technique or no treatment in N=41 patients with knee and/or hip osteoarthritis over 2 months with follow-up at 3–6 months. The second RCT (McCaffrey and Freeman 2003) compared listening to music with sitting quietly in N=66 patients with osteoarthritis. The interventions lasted for 14 days.

5.3.3. Evidence statements

Symptoms: pain, knee and/or hip

One RCT (Gay et al. 2002) (N=41) found that Jacobson relaxation was significantly better than control (no treatment) for pain (VAS) at 8 weeks, end of treatment (p<0.05), but there was no significant difference between the two groups at 4 weeks (mid-treatment) and at 3 months and 6 months post-treatment. (1+)

Symptoms: pain, unspecified site

One RCT (McCaffrey 2003) (N=66) found that rest and relaxation (sitting and listening to music) was significantly better than the control (sitting quietly and/or reading) for pre-post test changes of SF-MPQ pain (VAS) and SF-MPQ pain-rating index at day 1, day 7 and at 2 weeks (end of treatment), all p=0.001. Mean differences: SF-MPQ pain 23.4 18.9 and 17.3 respectively, all p=0.001; SF-MPQ pain-rating index −5.1, +3.8 and +2.2 respectively, all p=0.001. (1+)

Withdrawals: knee and/or shoulder

One RCT (Gay et al. 2002) (N=41) found that Jacobson relaxation and control (no treatment) were similar for total number of study withdrawals (N=3 21% and N=4, 31% respectively). (1+)

5.3.4. From evidence to recommendations

There was little evidence in this area. Many of the studies were about modalities not relevant to the NHS (for example therapeutic touch, playing music).

The GDG felt that it was important to emphasise the role of self-management strategies. As this is done in section 5.2, no recommendation is made here.

5.4. Thermotherapy

5.4.1. Clinical introduction

Thermotherapy has for many years been advocated as a useful adjunct to pharmacological therapies. Ice is used for acute injuries and warmth is used for sprains and strains. It seems appropriate to use hot and cold packs in osteoarthritis.

5.4.2. Methodological introduction

We looked for studies that investigated the efficacy and safety of local thermotherapy versus no treatment or other interventions with respect to symptoms, function and quality of life in adults with osteoarthritis. One systematic review and meta-analysis (Brosseau et al. 2003), one RCT (Evcik et al. 2007) and one non-comparative study (Martin et al. 1998) were found on thermotherapy. No relevant cohort or case-control studies were found. The RCT (Evcik et al. 2007) was excluded due to methodological limitations.

The meta-analysis assessed the RCTs for quality and pooled together all data for the outcomes of symptoms and function.

The meta-analysis included three single blind, parallel group RCTs (with N=179 participants) on comparisons between ice massage, cold packs and placebo, electroacupuncture (EA), short wave diathermy (SWD) or AL-TENS in patients with knee osteoarthritis.

Studies included in the analysis differed with respect to:

  • types of thermotherapy and comparisons used (one RCT ice application; one RCT ice massage)
  • type of comparison used (1 RCT SWD or placebo SWD; 1 RCT EA, AL-TENS or placebo AL-TENS)
  • treatment regimen (3 or 5 days/week)
  • trial size and length.

The non-comparative study (Martin et al. 1998) looked at pre- and post-treatment effects of liquid nitrogen cryotherapy (3 weeks of treatment) in N=26 patients with knee osteoarthritis.

5.4.4. From evidence to recommendations

The evidence base on thermotherapy is limited to three small RCTs, only one of which assesses pain relief. All the thermotherapy studies in osteoarthritis are on applying cold rather than heat. The RCT looking at pain found no significant difference between cold thermotherapy and control. The results in the RCTs assessing function are mixed when compared with controls, with electroacupuncture and with AL-TENS. There is no economic evidence available on the subject.

Despite the scarcity of evidence, in the GDG’s experience, local heat and cold are widely used as part of self-management. They may not always take the form of packs or massage, with some patients simply using hot baths to the same effect. As an intervention this has very low cost and is extremely safe. The GDG therefore felt that a positive recommendation was justified.



The use of local heat or cold should be considered as an adjunct to core treatment.