Developing and understanding the use of patient reported outcomes in advanced renal disease

Introduction The number of patients receiving treatment for advanced chronic kidney disease (CKD) is increasing worldwide. Current treatments are imperfect and impose a number of restrictions on daily life, both physical and psychosocial. The psychosocial burdens of advanced CKD have been explored to varying degrees with the effects on quality of life and symptoms most extensively documented in cross sectional data. There are no data on changing symptoms over relatively short time periods or the factors that affect an individual patient’s interaction with the disease and its treatments. This thesis set out to explore the health related quality of life (HRQOL), symptom burden, the factors associated with understanding/decision making and the focus for outcome preferences in a cohort of patients with advanced CKD receiving a variety of different treatments. This was achieved with the use of a questionnaire-based study called the “Influence of illness beliefs and treatment outcome preferences in patients with end stage renal disease” (INTUITIVE) study. In addition I explored the specific question of frequent collection of symptom related data, with the design and implementation of a web based symptom scoring system. Methods The INTUITIVE study recruited 58 individuals from prevalent dialysis, transplant and pre dialysis populations in a single renal unit. The study comprised a number of patient reported outcome questionnaires that covered the areas of health related quality of life HRQOL, symptom and treatment burdens, impact on daily life and screening for depression. In addition questionnaires that explored individual’s perceptions of their illness, beliefs about aspects of treatment and their outcome preferences were completed. All participants were asked to complete all relevant sections and a subset were asked to complete the questionnaires again after one year. For the frequent symptom tracking we designed and implemented a web based system for frequent tracking of symptoms (general well being (GWB), pain, sleep, breathing, energy and appetite) on a visual analogue scale. Participants from dialysis and pre dialysis populations were asked to complete the online form daily. Outcomes of interest were uptake, response rates, inter-modality variation and change in symptoms corresponding to predefined events. Results I have shown that HRQOL is dependant to a large degree on the number and burden of symptoms but also on the burden of specific symptoms in particular restless legs, shortness of breath, tiredness, dry mouth and itching. Individuals in the pre dialysis setting had more symptoms than transplant recipients and attributed more of their symptoms to kidney disease than the other groups. People on haemodialysis (HD) attributed more symptoms to the dialysis than those on peritoneal dialysis (PD) In addition, participants had a complex psychological understanding through which they were able to interpret and make sense of their kidney disease and its treatments. Individuals on peritoneal dialysis had a greater feeling of personal control over their kidney disease and older individuals perceive less negative consequences from the condition. Stronger beliefs around the necessity of prescribed medicines were seen in the transplant (Tx) population and in dialysis patients who had lower serum phosphate. Individuals not on dialysis had a much stronger preference for survival than those on dialysis and all were prepared to trade improvements in survival or symptoms against increases in treatment burden. The results from the web-based system showed that we successfully introduced an online, patient acceptable, iterative patient reported outcome measure that discriminates symptom burden, cross-sectionally and longitudinally. Discussion I have been able to demonstrate that poor HRQOL is related to symptom burden in advanced CKD and to identify the symptoms that are of most relevance. At the same time I have developed a methodology for frequent symptom tracking that is able to detect changes associated with relevant clinical events. I have been able to demonstrate that individuals are able to choose between symptoms, survival and symptom burden and that this balancing can be quantified in a meaningful way. In addition I have shown that treatment beliefs about necessity relate to behaviour and this has suggested further avenues for investigation in future work. Future work will focus on the use of these tools to inform interventions aimed at improving outcomes including HRQOL