2 The Case for a Commissioning Framework

2.1 Acceptance and Prevalence Rates in Renal Replacement Therapy (RRT)
2.2 Historical Basis of Inequity and Congestion
2.3 Current Problems in the Renal Service


2.1.1
Clarification of Terms

Renal Replacement Therapy (RRT) is a convenient term which comprises all life saving treatments for end stage renal failure (ESRF). It includes dialysis and renal transplantation. Acceptance of newly treated patients into the RRT programme is expressed as a rate per million of the population per annum (pmp.pa). The 'per annum' is often implied and acceptance rate is therefore more usually quoted pmp. Since 'pre-emptive' transplantation before patients require dialysis is rare, acceptance into RRT approximates to the acceptance rate for dialysis. The two are often used synonymously.

The prevalence of patients refers to the number currently receiving RRT and is usually expressed as a rate pmp. The term 'stock' which equates to prevalence is often used to better emphasise the expansion of numbers receiving treatment which most directly impacts on total cost. Because of the different acceptance and survival rates, for a given percentage expansion of RRT there may be very different rates of expansion (or contraction) of its component parts i.e. transplants and the different dialysis modalities, haemodialysis (HD) and peritoneal dialysis (PD).

2.1.2 Incidence and Acceptance Rates

It is not possible to accurately determine the incidence of ESRF in the population which is predominantly a disease of the elderly and a natural cause of death in this age group. Twenty years ago few patients over the age of 60 started dialysis treatment. Since then the numbers of elderly patients accepted for treatment has increased markedly. The second National Renal Registry report (ref.1) showed that at the start of RRT 46% of patients were aged 65 or more and 33% aged 70 or over in England and Wales in 1998. Median age was 63 years.

The acceptance rate into the RRT programme is the main determinant of how many patients are supported on treatment. Rates have been rising and have surpassed the 80 pmp 'target' figure quoted by the Renal Association in 1991(ref.2). T his was a figure based on treatment of patients under 80 years without adjustment for higher incidence in ethnic sub groups.

There persists a commonly held view in some quarters outside the UK that British healthcare professionals are restrictive as to whom they will or will not accept for RRT. In reality it is likely that practitioners in the UK differ little from colleagues abroad in their attitude to the individual patient. The relatively low numbers on RRT in the UK are more likely explained by continuing difficulties in accessing the service (see Section 2.2).

The annual acceptance rate of new patients onto dialysis rose from 67 pmp in 1991/1992 to 82 pmp in 1995, and rose further to 90 pmp in 1998 (ref.1). The figure for Scotland was 101 pmp. Between 1990 and 1995, the annual increase in acceptance rate was about 8% (ref.3). The West Midlands Regional Audit Group which examines trends in provision across 9 renal units serving a population of 7 million people showed acceptance rate for RRT increasing from 109 to 121 pmp between 1996 and 1999 with a particularly marked increase in those aged over 80 years (ref.4). Acceptance rate for new ESRF patients in the 1.7 million population of Northern Ireland was 105 pmp in 1999, (Doherty C - personal communication).

There continue to be marked differences in acceptance rates in different areas. Some differences are partly explained by the age and ethnic structure of the population (see Section 2.1.6). While the UK Renal Registry is not yet accurately population based, calculations based on estimated catchment areas show acceptance rates varying from 50 to 150 pmp per year (ref.1).

2.1.3 Transplant and Death Rates

Transplantation is the preferred outcome for many patients reaching ESRF. Organs are in short supply, the numbers of transplants having diminished slightly over the last 10 years (ref.5). Transplantation rates which rarely exceed 30 pmp per year compare poorly with acceptance rates of new patients. Also a substantial minority of transplant recipients will have lost their transplant kidneys after 10 years. In planning terms therefore it is important to recognise that dialysis remains the default therapy for all end stage renal disease.

The crude death rate i.e. the proportion of patients dying per year on the RRT programme is difficult to ascertain but will soon be available from the UK Renal Registry. Estimates would suggest that this is about 20% annually despite the increasing age and co  of the RRT population.

2.1.4 Rising Prevalence

Since acceptance continues to exceed death rate, the stock receiving RRT increases year on year.
This has been the case since dialysis was introduced in the mid 1960's and transplantation in 1968-70. In monetary and management terms the rise in dialysis stock is the biggest issue. While second nature to those in the specialty, the concept of rising stock has often proved difficult to appreciate amongst commissioners. This prompted the Medical Director of the NHS to warn in 1996 that "Purchasers should be aware that the total patient population is likely to continue to rise even if there is no further increase in acceptance rate" (ref.6).

The prevalence rate of RRT patients (dialysis and transplant) in England increased from 396 pmp in 1993 to 476 pmp in 19953 and rose further to 539 pmp (England, Scotland and Wales) in 1998 (ref.1). [Note: these figures differ slightly from those quoted in Appendix II. Only when European countries develop national registries will the figures be entirely reliable]. The prevalence of RRT in Northern Ireland at the end of 1999 was 602 pmp. Despite clear increases, the UK RRT prevalence still lags behind the estimated average (1998) for Germany, Netherlands, Spain, Italy and France of 747 pmp (Appendix II). The percentage of RRT patients with functioning grafts has diminished from 56% in 1988 to 50% in 1998.

Mathematical modelling to estimate future demand for RRT has consistently shown that steady-state (i.e. no growth in prevalence) is not likely to be reached within the next 10 years, and that the overall dialysis numbers are likely to increase by 50% - 100% and HD numbers by 150% (ref.7). There is considerable experience using a model which was developed in Sheffield (ref.8). The Renal Review Group in the Northern Region, taking 1998 as the baseline and using the Sheffield model showed an increase of 23% of patients on RRT up to the year 2004 (ref.9). The model showed great sensitivity to the acceptance rate. If this rose from 104 to 120 pmp the numbers of patients on dialysis alone in 2004 would increase by 58%.

Extrapolation from recent activity offers an alternative approach. In an interim report, the Renal Sub-Committee of the Eastern Region Specialities Services Commissioning Group (RSSCG) used provider-based data from the past 5 years to predict growth. They advised Health Authorities to anticipate continued growth in dialysis stock at an annual rate of between 7% and 10% (ref.10). The West Midlands Regional Audit Group showed patient numbers climbing from 2297 in 1996 to 2809 in 1999 (ref.4). This equates to an annual growth of dialysis stock of 7.4% (simple interest).

2.1.5 Differential Expansion of Dialysis Modalities

Haemodialysis (HD) and peritoneal dialysis (PD) are complementary therapies, the ratio of patients on each varying historically and by provider. While patient preference plays a part, this is often overridden by other factors which include availability of HD stations, technique failure of peritoneal dialysis and the proportion of patients presenting late as 'uraemic' emergencies. The vast majority of late referrals are ultimately treated by HD (see Section 4.2). The availability of automated peritoneal dialysis (APD) which employs machinery to allow overnight cycling of fluids rather than the manual exchanges of continuous ambulatory peritoneal dialysis (CAPD) is a recent development which may allow patients to receive PD for longer.

The relative prevalence of PD in the UK has been declining since its peak in the early 90's. Chief movers have been the availability of satellite haemodialysis and the recognition that in some patients PD provides sufficient dialysis only as long as it is supplemented by natural renal function which declines with time on dialysis. There is a higher rate of transfer from PD to HD in the first two years than HD to PD (ref.1). In addition, 6% of peritoneal dialysis patients (60% of those who died) have a brief period of HD immediately prior to death. These figures emphasise the need for an adequate HD programme to support the PD programme.

The percentage of dialysis patients on HD was 50% in 1993 climbing to 62% in 1998 (ref.1). Since 1997 there has been an annual increase of 10% in HD patients, 4% in PD patients and 2.5% in the transplant stock providing an overall 5.3% increase in the total number of patients on RRT. A survey of Eastern Region renal units showed the percentage of dialysis patients on HD climbed from 66% to 69% between 1996 and 1999. HD in the Eastern Region is expanding by 9% pa and peritoneal dialysis stock declining by 2% pa (ref.10). In the West Midlands the percentage of patients treated by HD expanded from 50% to 65% between the years 1995 and 1999. While between 1996 and 1999 the PD population remained static at approximately 1000 patients, HD increased from 1245 to 1806 patients, an increase of 15% p.a. This disparity between HD and PD growth in the West Midlands resulted from an increased proportion of patients choosing HD as their initial modality and a high level of PD technique failure. However, there was a three-fold increase in automated PD (APD) (ref.4).
In Northern Ireland the annual percentage growth in HD during 1995 - 1999 averaged 17.2%
with zero growth in PD numbers. (Doherty C - personal communication).

It would be prudent therefore for commissioners to plan for PD prevalence to approach those found in Europe and the United States which are generally less than a quarter of the total managed on dialysis (ref.11,12). The corollary is that the planners should accept the inevitable growth of haemodialysis. This remains the default modality for all patients with ESRF: the numbers are increasing despite the successes of CAPD/APD and transplantation.

2.1.6 Smouldering Demographic Changes

Age
Population projections in the UK are for an increase in the percentage over 65. Since ESRF is age related, the ageing of the population will increase need.

Liberalisation of acceptance and poor rates of transplantation resulting from co-existing comorbid disease in the elderly and ageing of patients surviving on RRT are increasing the mean age of patients on dialysis, a trend which is likely to continue. This increases the overall burden of co-morbidity which has major implications for complementary services, particularly cardiac and vascular surgery. Higher numbers of older and more disabled patients will require far more assessments of need for rehabilitative care and support in the community under the 1990 NHS and Community Care Act. Social work input will be significant.

Diabetes
Twenty years ago, diabetics who developed nephropathy were considered to be in the terminal stages of their disease and were rarely accepted for RRT. Diabetics now comprise the most common disease group entering RRT (ref.1). Although rates vary, over 15% of new patients accepted are diabetic and the proportion is likely to increase further given the worldwide epidemic of type 2 diabetes.

In the United States the figure exceeds 50% (ref.12). Diabetics who reach ESRF typically have a full house of complications including hypertension, retinopathy (eye disease) and often coronary and peripheral vascular disease.

Ethnic Sub Groups

In the UK, Afro-Caribbeans and Asians from the Indian Subcontinent have higher levels of hypertension and diabetes than the white population which lead to a higher incidence of renal failure and the need for RRT (ref.13). These populations have three to fourfold higher acceptance rates onto RRT and in some districts comprise the majority of patients treated (ref.14). Their greater need for RRT is accompanied by difficulties in blood group and tissue matching in cross racial transplants and the shortage of donor organs. An important demographic feature of Afro-Caribbeans and Asians in the UK is their relatively young age. The ageing of ethnic minority groups will increase local demand for RRT services significantly (ref.15,16). Regions which have high proportions of people from ethnic minorities will need to build in assumptions of growth of dialysis prevalence above 10% per year.