Introduction: There is an increasing number of Canadians living with end stage renal disease (ESRD). Kidney transplantation is currently the best treatment for ESRD but long-term outcomes remain suboptimal. Identifying factors associated with better outcomes may lead to interventions or practice change that could improve patient survival or quality of life. The objectives of this thesis were to: i) systematically review the literature to examine centre variation in kidney transplantation outcomes and identify centre and provider level factors that may contribute to variation in outcomes; ii) describe differences that may exist at the patient, centre and provider level at the time of kidney transplantation across the six transplant centres in Ontario, Canada; iii) examine variation in graft and patient survival rates across transplant centres in Ontario; and iv) examine whether patient, centre and provider level characteristics contribute to variation in graft and survival rates across transplant centres. Methods: The first objective of this thesis was met by conducting a systematic review of the literature according to a predefined protocol. The last three objectives of the thesis were met by conducting a population based retrospective cohort study using administrative data from Ontario. Differences at the patient, centre and provider level were described at the time of kidney transplantation. Outcomes of interest included total graft loss; graft loss with follow-up censored at death; death with graft function; and total mortality. All outcomes were assessed at one year post transplantation and at the end of study follow up. Cox proportional hazards regression was used to obtain hazard ratios (HR) for each centre relative to the average across all centres. The independent effect of centre volume and provider characteristics on outcomes was also examined. Results: The systematic review identified 24 eligible studies. Outcomes included graft survival (n=24) and patient survival (n=9). The main characteristics evaluated were centre volume (n=17) and provider volume (n=2). Centre variation in graft survival was described in 80% (12/15) of studies, while less than half of studies (8/17) found a significant association between volume and graft survival. The population based retrospective cohort included 5092 adults (≥18 years) who received a primary solitary kidney transplant across 6 transplant centres in Ontario between January 1st 2000 and December 31st 2013. Variation in patient, centre and provider level factors existed across centres at the time of transplantation. At the end of study follow-up, case-mix adjusted HRs for total graft loss ranged from 0.84 (95% CI 0.53-1.33) to 1.16 (95% CI 1.00-1.34) across centres (p-value for between centre variation 0.46). After adjusting for centre and provider factors, differences across centres persisted. Centre volume, provider experience and provider type were not independently associated with either short or long-term outcomes (all p>0.05) with the exception of graft loss with follow-up censored at death. Discussion: This thesis suggests that there is variation in clinical outcomes across transplant centres in Ontario which is not explained by patient factors, centre volume or provider characteristics at the time of transplantation. Additionally centre volume, provider type and experience were not independently associated with outcomes. Future prospective studies with a larger sample size of transplant centres that examine follow-up care after discharge from hospital (e.g. frequency of visits) are required to better understand this phenomenon.

This thesis examines the effect on height determination of the antenna phase centre variation of GPS user segment antennae. A discussion of the various antenna types in common use is followed by an explanation of the problem at hand. In particular the effect of the antenna's environment on the phase centre variation is covered more fully, since the phase variation phenomenon itself is largely unexplained in the engineering community to date. A number of examples of the heighting errors caused by this phenomenon are presented, followed by specifically designed experiments, which quantify the effect. Finally the phase centre variation itself is modelled for a particular GPS antenna in common use by surveyors in Australia. The overall conclusion, arrived at by demonstration, is that the antenna phase centre offsets and the variation model are very important for high accuracy determinations of height.