- Top of page
- Implications for further research
- Relevance to clinical practice
- Conflict of interests
Chronic inflammatory polyarthritis as rheumatoid arthritis (RA) and psoriatic arthritis (PsA) has a reported prevalence of 0·5% (Uhlig et al. 1998) and 0·2% (Madland et al. 2005), respectively. RA is most common in middle-aged women (3:1); common symptoms include pain, fatigue, inflammation of joints and irreversible joint damage with functional losses if the disease remains active and uncontrolled (Harris 2005). PsA usually occurs when patients are in their early forties (Zink et al. 2006); common symptoms are inflammation of peripheral joints and manifestations of skin, nails and spine (Madland et al. 2005, Mease 2011). Medical treatment options for patients with RA and PsA consist of nonsteroidal anti-inflammatory drugs (NSAIDs), disease-modifying antirheumatic drugs (DMARDs) and biological agents (Madland et al. 2005, Scott et al. 2010).
Patients with RA and PsA have similar complaints, they experience coping as a balancing process (Gronning et al. 2011), and the psychological and social consequences of living with chronic arthritis are often addressed in patient education interventions (Barlow et al. 2002, Sinclair & Blackburn 2008). The recommendations of European League Against Rheumatism (EULAR) for management of early arthritis focus on 12 key points including pharmacological and nonpharmacological interventions like education programmes (Combe et al. 2007).
Patient education is often offered to patients in groups or individually. Group education facilitates patients' learning from each other, while individual education is easier to tailor to patients' individual needs (Newman et al. 2004). In the field of rheumatology, individual patient education is often delivered by nurses (Firth 2011) and covers a great variety of aspects related to living with arthritis, for example medical treatment, motivation to carry out exercise programmes, joint protection, energy conserving and nonmedical pain management (Makelainen et al. 2009). Individual consultations with an expert nurse have shown to have a positive influence on patients' ability to control and cope with arthritis (Ryan et al. 2006).
Five reviews (Savelkoul 2003, Niedermann et al. 2004, Riemsma et al. 2004, Albano et al. 2010, Iversen et al. 2010, Knittle et al. 2010) have summarised that group-based patient education or self-management interventions show small short-term effects in patients with chronic inflammatory polyarthritis, but the evidence on long-term effects is inconclusive. Five relatively old studies of group-based interventions (Parker et al. 1988, 1995, Taal et al. 1993, Lindroth et al. 1997, Scholten et al. 1999) along with more recent studies (Hammond & Freeman 2004, Masiero et al. 2007, Hammond et al. 2008) have demonstrated beneficial long-term effects, while others have not (Brus et al. 1998, Helliwell et al. 1999, Giraudet-Le Quintrec et al. 2007, Mayoux-Benhamou et al. 2008, Rudd et al. 2009). One study showed improvements in patient coping, knowledge and satisfaction, but the intervention was not found effective on the primary outcomes after one year.
Very few studies have investigated whether a combination of group and individual education for patients with chronic inflammatory polyarthritis is effective. The randomised controlled study (RCT) by Nunez and colleagues (Nunez et al. 2006) showed that a combination of group and individual education in addition to pharmacological treatment had beneficial effects six months post-treatment. Another RCT (Hammond et al. 2004) of a combination of group and individual education did not find that intervention as effective one year post-treatment.
We have previously published an article on the short-term effect of a nurse-led intervention combining group and individual self-management education for patients with chronic inflammatory polyarthritis (Grønning et al. 2012). The intervention showed to have effect on the primary outcomes, general well-being and self-efficacy, compared with usual care. Given the lack of studies looking at long-term effects of patient education, the aim of this study was to investigate the long-term effects of this intervention.
- Top of page
- Implications for further research
- Relevance to clinical practice
- Conflict of interests
This intervention has already shown short-term effects on patients' Global Well-Being and Arthritis Self-Efficacy Other Symptoms Subscale (Grønning et al. 2012). This study showed that a small to moderate long-term effect, effect size = 0·4 (Cohen 1992), on global well-being still was present after 12 months. However, the CI was rather wide ranging from 1·6–14·8. The interpretation of this wide-ranging CI is that there may be a bit uncertainty related to the estimated value of well-being.
The short-term effect on patients' Arthritis Self-Efficacy Other Symptoms Subscale was no longer present at 12 months. The intervention appears to have prevented long-term deterioration in patients' arthritis self-efficacy as the control group had a significant deterioration, while the intervention group did not.
Lack of maintenance of the proven short-term effect on arthritis self-efficacy other symptoms may have several explanations. First of all, the intervention focused on the overall perspective of living with chronic inflammatory arthritis (Grønning et al. 2012) with lectures on common symptoms in RA and PsA, disease process and prognosis, how disease complaints influenced patients' lives and the impact of disease on patients' physical, mental and social well-being. Following the lectures, the participants discussed how they experienced living with arthritis. As a result of having more focus on the overall perspective of living with chronic inflammatory arthritis, it is possible that the intervention focused too little on the areas covered by the arthritis self-efficacy other symptoms scale; for example patients' self-efficacy in dealing with fatigue, regulating activity, coping with mood and frustrations, or dealing with pain during daily activities. According to self-efficacy theory (Bandura 1997), fostering self-care behaviours needs time and practice. Self-care behaviours involve acquisition of practical skills to manage pain, fatigue, physical limitations and coping (emotional challenges, symptom unpredictability, stress and anxiety) (Katz 2005). Although such topics were addressed, no practices of self-management skills were included in the intervention.
Second, length and content of the intervention may also be of relevance. This nurse-led intervention was developed in cooperation with patients and health professionals in terms of format, duration, content and delivery (Grønning et al. 2012) and addresses similar topics as other beneficial patient education programmes (Riemsma et al. 2003, Giraudet-Le Quintrec et al. 2007, Hammond et al. 2008). The unique feature of the intervention in our study was that the topics were addressed in a group-setting with a mixture of lectures and discussions before it was tailored to the individual participant in a one-to-one educational session. However, other interventions with beneficial outcomes such as increased self-efficacy (Riemsma et al. 2003, Hammond et al. 2008) and coping (Giraudet-Le Quintrec et al. 2007) are different in length and duration, and they were held over a longer period of time and included booster sessions. In our study, the intervention was delivered in 10 hours without boosters. One beneficial intervention includes three-two-hour booster sessions where the last booster was offered after nine months, and the boosters were given in addition to 10 hours of education delivered over a period of five weeks (Riemsma et al. 2003). Another intervention consisted of two modules; each module included four 2·5-hour meetings and a two-hour review meeting over a period ranging from three to nine months depending on the individual participants' requests (Hammond et al. 2008). The most comprehensive intervention included six hours of education once a week over eight consecutive weeks followed by a four-hour booster session after six months (Giraudet-Le Quintrec et al. 2007). On the other hand, a recent pilot study may indicate that amount of education is not necessarily the most important aspect, because a one-day educational programme showed beneficial effect one year later (Kennedy et al. 2011).
Other explanations could be related to socio-demographic characteristics. This study sample was highly educated. Level of education may have influenced the coping abilities because the baseline level of self-efficacy in this sample was higher than that in a comparable study (Hammond et al. 2008). It is thus possible that highly educated patients are more active in collecting relevant information (e.g. on consequences of having arthritis and how to cope with these) resulting in greater coping abilities. Furthermore, it is possible that patient education does not facilitate an additional increase in patients' self-efficacy if the baseline scores are high.
The short-term beneficial outcomes on patient activation and pain (Grønning et al. 2012) were no longer present at 12 months. There was a statistical trend for group difference in activation, but no statistically significant differences in VAS pain. The review of therapeutic patient education in RA (Albano et al. 2010) detected only few effective interventions, mainly with short-lived beneficial effects on pain, fatigue, functional status, anxiety, depression and health behaviours. However, a literature review on theoretical perspectives of arthritis pain (Keefe & Somers 2010) emphasised that emotions, cognitions and social context variables were important factors related to arthritis pain and that interventions focusing on pain coping skills training were beneficial in decreasing arthritis pain, psychological distress and arthritis-related disability.
The intervention and control groups' baseline exercise patterns in this study were also different. The controls reported more exercise than the intervention group. At 12 months, no statistical differences in exercise pattern between the groups were present. Nevertheless, fewer in the intervention group reported never to exercise or exercised less than once a week, while more exercised daily or two-three times a week. Motivation to exercise is described as an important content of patient education provided by rheumatology nurses (Makelainen et al. 2009). This intervention did not include exercise, but lectures and discussions about beneficial outcomes from exercise (the theme ‘healthy life style’) were discussed (Grønning et al. 2010). Moreover, this nurse-led patient education intervention is ‘protective’ by maintaining patients' health, because the controls had a statistically significant worsening in arthritis self-efficacy other symptoms, AIMS social and HADS total from baseline to 12 months. The intervention group maintained the same levels throughout the study period. Also, the qualitative responses from the participants in the intervention group revealed that some patients experienced increased confidence and disease control. However, a few participants thought it was difficult to cope (managing pain and judging their disease).
A statistically significant decrease in the intervention group's DAS28-3 score was observed, with a trend for a group difference. The decrease in DAS28-3 may have several explanations. It is known that early use of DMARD is crucial to avoid persistent and erosive arthritis (Donahue et al. 2007). Individual education is found to be suitable for better adherence to medical treatment (Hill et al. 2001, van den Bemt et al. 2009). In this intervention, the topic ‘medical information’ was addressed by a physician in a group session, the use of medication was discussed within the group, and it was individually tailored in the one-to-one educational session. Medical information covered aspects related to taking medication as prescribed, possible side effects and self-monitoring (blood tests and observe bodily changes). We did not measure adherence, but the participants' qualitative responses covered descriptions of increased knowledge about medicines after attending the patient education intervention. It is thus possible that increased knowledge entailed a more conscious attitude about taking medicines as prescribed to keep the arthritis under control.
It is important to identify for whom patient education is most beneficial (Newman et al. 2004, Riemsma et al. 2004, Hammond et al. 2008). We investigated whether men and women responded differently to this nurse-led patient education intervention. The results must be taken with precaution because the sample size was not calculated to detect gender differences. The sample consisted of fewer men than women (1:2), but was considered as representative for the population studied. The analysis showed that men and women responded similarly to the intervention in the majority of outcomes: global well-being, arthritis self-efficacy other symptoms, psycho-social health status and physical health status. However, women showed a statistically significant improvement in activation and DAS28-3, while men had a statistically significant deterioration in arthritis self-efficacy pain and a trend for improved DAS28-3. The interpretation of these results may be that the men in this sample were more reluctant about discussing their self-efficacy in coping with pain. It is suggested that men tend to avoid actions, which could evoke feelings of vulnerability (sharing concerns with others or revealing their physical disability) (Lack et al. 2011).
Strengths and limitations
The major strength of this study was the long-term follow-up and the randomisation. The low dropout rate (6%) decreased the chance of making a type II error. Furthermore, the dropout rate is lower than reported in comparable studies (Giraudet-Le Quintrec et al. 2007, Hammond et al. 2008). One limitation is the lack of blinding. However, blinding is difficult because the patients would know whether they took part in an educational programme or not. It could have strengthened the study if the assessors of clinical outcomes (joint counts) were blinded, but this was not feasible due to practical issues. Another limitation is that the study was only powered in relation to arthritis self-efficacy other symptoms and not the other primary outcome, patients' global well-being. Furthermore, a possible limitation is the chance of selection bias because the intervention was delivered during the daytime. It is possible that eligible patients in paid employment chose not to respond to the invitation because they prioritised their job. The employment rate in our sample was lower, twenty-nine per cent, compared with thirty-eight per cent reported in other studies (Hammond et al. 2008, Reinseth et al. 2010). Our population was similar on age, marital status, disease duration (Hammond et al. 2008, Reinseth et al. 2010), self-efficacy (Reinseth et al. 2010) and medical treatment (Hammond et al. 2008).