Summary measures for the data are shown in Table 1 . There were 63.2% males and 36.8% females, a male to female RRT ratio of 1.7:1. The median age for patients on RRT was 56 years, peritoneal dialysis 58 years, haemodialysis 66 years and transplantation 50 years.
The result of our extensive data cleaning and validation exercise was as follows. The database contained 2434 records with unique dates of birth, which could also be fully matched to current or historical Welsh postcodes. An additional 167 records could not be matched to a Welsh LA area; 110 of these records clearly represented English postcodes; an additional 35 were probable English postcodes, and 22 records had invalid postcodes, which could potentially have been Welsh postcodes. The maximum number of missing Welsh records in our analysis would therefore be <1% (22/2434). In conclusion, the 167 unmatched and consequently excluded records identified during data validation largely represented English and not Welsh patients and their exclusion would consequently not have affected our analysis.
Comparison with the Welsh NHS AR indicated that only 10 records (0.45%) represented deceased patients that had not been removed from the renal registers in various NHS Trusts. A handful of patients may have been double counted as a result of errors in dates of birth or postcode changes. Detailed analysis also suggested that a handful of errors might also have occurred because patients moved house during the months preceding the census. Once again, our data validation indicated that these issues did not significantly affect our data.
A significant issue was identified in relation to one English NHS Trust. Comparison between NHS AR data and data on patients provided with a transplant by this Trust suggested around 3% of the patients still on the trust's renal register were deceased. These patients were removed from our database and consequently would not have affected our results. However, our data validation exercise indicated that other data collection exercises in the UK may have potentially significant errors due to inadequate removal of deceased patients from some trust renal registers.
Age-specific prevalence rates for the three modalities of RRT had significantly different peak rates and distributions. RRT age-specific prevalence rates (Figure 1) peaked at around 70 years with a rate of around 1790 pmp; haemodialysis age-specific prevalence rates peaked at around 80 years with a rate of around 1080 pmp; peritoneal dialysis rates did not have a clear peak; and transplantation age-specific prevalence rates peaked at around 60 years with a rate of around 924 pmp.
Age-specific prevalence rates (moving average pmp) of RRT for the renal stock in Wales on 30 June 2004
Age-specific incidence rates for RRT (Figure 2) peaked at a rate of around 488 pmp at 79 years. This was very heavily influenced by the high incidence rates for haemodialysis, which peaked at the same age. Age appears to have much less effect on initiation of peritoneal dialysis, which had a broad plateau between the early fifties and late seventies. Rates of new transplantation were highest in the mid-fifties, but were markedly absent in old age. This is in contrast to the renal transplant prevalence rates, which indicates that a number of patients in old age have functioning transplanted kidneys. These patients appear to have received their transplants at a much younger age.
Age-specific incidence rates (moving average pmp) of renal patients alive at the census date and starting RRT in the year prior to 30 June 2004 in Wales
Figure 3 explores the relative probability of being on a given modality of RRT at any given age. The graph indicates that the probability of being on peritoneal dialysis remains at around 10–15% for all age groups. Until the age of 60 years the probability of being on haemodialysis remains at around 25%, after which it rises steeply and over the age of 80 years an individual on RRT had around an 80% probability of being on haemodialysis. In contrast, the probability of having a transplant falls steadily from the age of 45 years.
Age-specific probability (moving average pmp) of being on a given modality of RRT for the renal stock in Wales on 30 June 2004
Figure 4 indicates the probability, based on an individual's age, of commencing a specified modality of RRT when RRT is first initiated. The graph suggests that physicians prefer giving younger patients renal transplantation or starting them on peritoneal dialysis. With increasing age, and particularly after the mid-fifties, physicians appear to have an increasing preference for starting patients on haemodialysis. However, some physicians seem to consider transplantation for occasional patients up until their late sixties, and some physicians started patients on dialysis into their late eighties. The extent to which patients have contributed to these treatment decisions is unclear.
Age-specific probability (moving average pmp) of renal patients alive at the census date having started a given modality of RRT in the year before 30 June 2004 in Wales
The effect of gender with age is examined in Figure 5. The graph indicates that at every age, men receive rates of RRT that are higher, or at least as high, as the rates for women. This gender difference applies to all three modalities of RRT. The difference is most marked for transplantation in middle age and for haemodialysis in old age.
Age-specific prevalence rates (moving average pmp) of RRT for the renal stock in Wales on 30 June 2004 for males and females
Geographical location also has an effect on RRT rates of incidence (acceptance) and prevalence (Figures 6 and 7). North Wales, Swansea, Gwynedd and Wrexham area rates appear to be higher than the all Wales rates. Cardiff and Powys area rates appear to be lower than the all Wales rates. There is wide variation in the acceptance rate, ranging from under 50 pmp to over 200 pmp.
Geographical location also influences the relative proportion of patients on peritoneal or haemodialysis (Figure 8). The proportion of patients on haemodialysis in the LHB areas at the extremes of Figure 8 (Wrexham and Flintshire) is not statistically significantly different from the proportion of patients on haemodialysis in Wales as a whole (73.1%). However, there is a 25.1% (95%CI 8.0–40.6%) difference between the proportions of patients on haemodialysis between these two areas.
BMC Nephrology © 2007 van Woerden et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The author(s) declare that they have no competing interests.
Cite this: The Effect of Gender, Age, and Geographical Location on the Incidence and Prevalence of Renal Replacement Therapy in Wales - Medscape - Jan 11, 2007.