Piloting an Intervention for Improving Treatment Adherence

People living with renal failure have to endure a demanding treatment regimen to keep them alive including haemodialysis, a restricted diet and fluid intake and multiple drug treatments. These demands have a profound effect on the patient's quality of life. It is not surprising that many find it very difficult to manage some treatments even when they are known to be important for the patient's survival. Restricting fluid intake (often as little as 500ml per/day) is an essential treatment for many patients with advanced kidney failure receiving haemodialysis, but many find it difficult to adhere. Estimates of fluid restriction non-adherence range between 10-70%. Not adhering to the recommended maximum fluid intake leads to fluid overload, which is associated with pulmonary oedema, cardiovascular disease and premature death.

Understanding how to help patients improve their treatment adherence is challenging and multifaceted. A recent Cochrane Review concluded that: "For long-term treatments, no simple intervention, and only some complex ones, led to improvements in health outcomes.". The healthcare service has tried to help patients by promoting the patient's own role in their healthcare and in educating patients about their illness and the treatments. However, it is widely acknowledged that education alone is insufficient to change a patient's behaviour. Educating patients about the consequences of non-adherence is often unsuccessful as for some patients, they have become entrenched in poor health care behaviour over extended periods of time. In this case, repeated exposure of the patient to threatening health-risk information might be ineffective as the patient has likely become well rehearsed in discounting the information being provided, and the behaviour of the patient will not change.

The proposed study is designed to investigate a novel intervention to promote behaviour change that has been shown to be effective in changing behaviour in public health (e.g. unhealthy dietary intake, unsafe sun exposure, hazardous drinking and smoking cessation) The intervention is simple to implement, requires little training of staff and may have major benefits for the patients and the NHS. The intervention is based on self-affirmation theory. Central to this body of work is the recognition that most health-risk education confronts people with information that is threatening, both to survival, and to their sense of self. The person's response is to interpret the information in a way that reduces the threat to them, for example by denigrating the veracity of the information. In turn the bias with which the information is interpreted reduces its ability to change behaviour. The theory suggests that if the person's sense of their own worth (e.g. sense-of-self) is boosted at the time of receiving the threatening health-risk message, the person is less likely to interpret the information in a biased way, rendering them more likely to accept the message, have increased intention to change and believe that they can (self efficacy), and ultimately change their behaviour. For example, Armitage examined adults' reactions to alcohol education materials, and in a randomised controlled trial compared participants who self-affirmed by completing an experimental questionnaire with participants who completed a control questionnaire. Self-affirmation not only significantly increased participants' receptiveness to the message, but also improved the effectiveness of the education in that 29% (27/92) fewer people were engaged in hazardous drinking at the end of the study if they had self-affirmed.

Until recently, this behaviour change intervention method had not been tested in a patient group experiencing a long-term condition like renal failure. In a recent pilot study (UKCRN Portfolio study 74185), the team evaluated the potential for a self-affirmation intervention to promote improved phosphate medication adherence among non-adherent patients. Interim results from this study have highlighted key issues for which further research is necessary. Firstly, this intervention is known to be more effective in people with low motivation and this informed the recruitment strategy, targeting patients who appeared (on the basis of high serum phosphate levels) to be non-adherent in taking their phosphate medication. However, despite the patient sample's higher average phosphate levels, the majority of the patient group reported to be motivated to improve phosphate control. Therefore, it is likely that for many of these patients there are other reasons for their higher average phosphate levels, such as forgetting to take their medication. In the group of patients who did report low motivation to taking their medication (25% of the sample), the effects of the self-affirmation were observed. There was a trend for patients who self-affirmed, to report the health-risk message to be higher quality; greater self-efficacy and intention to improve phosphate control and ultimately their serum phosphate levels were significantly reduced at one month compared with the control group. It is important to now test the intervention with a larger group of patients reporting low motivation towards the required health behaviour. The team identified another treatment (fluid management) that most renal patients must follow and evaluated 93 patients' thoughts about managing their fluid intake. The results of this brief study suggest that a larger proportion of patients are less motivated to manage their fluid intake and therefore, assessing the feasibility and efficacy of the self-affirmation intervention would be achievable in this treatment group. Fluid management is a central adherence issue to address as it is not only clinically very important but it is a particularly difficult aspect of treatment for patients to manage.

The second issue that arose in the first study is the longevity of the intervention. The apparent effects of self-affirmation on behaviour change to improve phosphate control observed at one month were not sustained at three month follow-up which suggests that self-affirmation might require repeated interventions. An important feature of this new study would be to include repeated self-affirmations in the intervention group to investigate this further.

A randomised controlled pilot trial is proposed. The study aims to recruit 120 patients across a network of dialysis units, focussing on patients who have higher than average interdialytic weight gains (IDWG). The trial is designed to build upon routine education about fluid management during dialysis visits, and will compare a group of patients who have received a self-affirmation intervention to boost their sense-of-self (self-affirmation arm) to a control group (control arm). It is expected that patients in the self-affirmation arm should have a more positive evaluation of the health-risk information about fluid management, feel more able to, and intend to change their fluid control behaviour. The appropriate methods of analysis include logistic regression, and mixed multilevel models, but the relatively small sample size will limit the extent of the modelling that can be undertaken. Thus the aim of the analysis will be examine the size of the observed effects and to design a larger study. The study will inform the extent to which psychological and clinical factors contribute to the outcome and should be used in future studies.

As this intervention is targeting non-adherent patients (as defined by higher IDWG levels) and therefore not necessarily representative of all patients on dialysis, it is important to establish the true clinical context. Patients who appear to be managing their treatment well, based on their current clinical measures, might well be highly motivated. However, it is also possible that some of those patients also experience low motivation, which in time, as their illness becomes more severe, would be revealed by worsening adherence and clinical outcomes. Thus, the study also aims to examine the psychological and clinical factors associated with treatment adherence and non-adherence in a wider patient sample.