Renal centre characteristics and physician practice patterns associated with home dialysis use.

BACKGROUND: There is a wide variation in home dialysis use (peritoneal dialysis and home haemodialysis) between renal centres. This study identifies which centre characteristics and practice patterns are associated with home dialysis use. METHODS: An observational study of all UK patients starting renal replacement therapy (RRT) in 2007-2008 using patient characteristics from the UK Renal Registry (UKRR) and renal centre characteristics ascertained from a national survey. Multilevel logistic regression was used to examine the association between patient and centre characteristics and home dialysis uptake. RESULTS: Twenty-six per cent of 11 913 patients used home dialysis and survey responses were available from every renal centre. After taking into account patient factors, several centre factors were associated with a higher probability of home dialysis: physicians aspiring to a higher 'ideal' peritoneal dialysis rate (odds ratio, OR 1.21, 95% CI 1.06-1.37, P = 0.003 per 10% increase in 'ideal' percentage), early use of peritoneal dialysis (PD, OR 1.52, 95% CI 1.18-1.95, P < 0.001), use of home visits to educate patients pre-dialysis (OR 1.39, 95% CI 1.05-1.83, P = 0.02) and to provide trouble-shooting advice for existing home dialysis patients (OR 1.63, 95% CI 1.11-2.42, P = 0.01). Using videos/DVDs as part of the pre-dialysis education programme was associated with a lower probability of home dialysis, but this was correlated with lower levels of physician enthusiasm (r = -0.48, P < 0.001). After adjustment for this, the association disappeared (OR 0.77, 95% CI 0.55-1.07, P = 0.1). CONCLUSIONS: Home dialysis use is associated with modifiable centre factors as well as individual patient characteristics.

How much of the regional variation in RRT incidence rates within the UK is explained by the health needs of the general population?

BACKGROUND: Variation in end-stage renal disease treatment rates in the UK persist after adjustment for socio-demographic factors. METHODS: UK-wide ecological study using population socio-demographic factors, health status characteristics and access to health services factor in to explain the incidence of renal replacement therapy (RRT). RESULTS: There was a 6% higher incidence rate of RRT per standard deviation (SD) increase in area diabetes prevalence after adjustment for area level socio-economic deprivation status and the proportion of non-white residents [incidence rate ratio adjusted (IRR adjusted) 1.06 (95% confidence interval 1.03,1.09), P < 0.001]. A 3% lower-adjusted RRT incidence rate was seen with each SD higher proportion of diabetics achieving an HbA1c of <7.5% [IRR 0.97 (0.94, 1.00), P = 0.03]. Hypertension prevalence was independently associated with an 8% higher RRT incidence rate per SD increase [IRR adjusted 1.08 (1.04, 1.11), P < 0.001] and an SD increase in life expectancy in an area was independently associated with 7% lower RRT incidence rate [IRR adjusted 0.93 (0.91, 0.96), P < 0.001]. An SD increase in premature cardiovascular (CV) mortality rate in an area was also independently associated with RRT incidence rates [IRR adjusted 1.06 (1.03, 1.09), P < 0.001]. Rates of coronary artery bypass grafting (CABG)/angioplasty and knee replacement were positively associated with RRT incidence, but mammography uptake was not associated. In total, 31% of the regional variation in RRT incidence could be explained by these factors. CONCLUSIONS: Diabetes prevalence, the proportion of diabetics achieving good glycaemic control, hypertension prevalence, life expectancy, premature CV mortality, CABG/angioplasty and knee replacement rates were all associated with RRT incidence. A third of the regional variation in RRT incidence between areas can be explained by these demographic, health and access to health services factors.