OUP user menu

Current chronic kidney disease practice patterns in the UK: a national survey

A. Ahmad, P. Roderick, M. Ward, R. Steenkamp, R. Burden, D. O'Donoghue, D. Ansell, T. Feest
DOI: http://dx.doi.org/10.1093/qjmed/hcl029 245-251 First published online: 23 February 2006


Background: There is an increasing focus on improving the detection and management of patients with chronic kidney disease (CKD). Data on CKD prevalence based on population sampling are now available, but there are few data about CKD patients attending nephrology services or how such services are organized.

Aim: To survey services for CKD patients nationally.

Methods: A pre-piloted questionnaire was sent to all 72 renal units in the UK, referring to the situation in June 2004.

Results: Seventy units (97%) responded. The median ratio of prevalent CKD patients/prevalent renal replacement therapy (RRT) patients in the 25 units with data was 3.7 (IQR 2.7–5.7) and the median ratio of CKD stage 4 and 5 patients/prevalent RRT patients was 0.6 (IQR 0.4–1.1). This gives an estimated 140 000 CKD patients under the care of UK nephrologists, with 23 000 at CKD stage 4 or 5 (excluding those on RRT). Very few units had a full complement of the recommended multi-skilled renal team. Counsellors and psychologist were the most common perceived shortages. Of 70 responding units, 50 (74%) were using low clearance clinics for management of advanced CKD patients. Elective dialysis access services often had long delays, with median waiting time for vascular access ranging between 1 and 36 weeks, and for Tenchkoff catheter, between 0 and 12 weeks.

Discussion: CKD patients are a significant workload for UK nephrologists. Current provision of service is variable, and services need to be re-designed to cope with the expected future increase of referral of CKD patients.


The National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF-K/DOQI)1 guidance uses estimated glomerular filtration rate (GFR) to produce a chronic kidney disease (CKD) staging system. This system has been widely adopted, facilitating a standardized approach to estimating the prevalence of CKD (Table 1). In the US, the Third National Health and Nutrition Examination Survey (NHANES III) found that 4.7% of the population had a GFR <60 ml/min (stages 3–5), with a much higher prevalence in the elderly and people with diabetes and hypertension.2 In the UK, the NeoErica (New Opportunities for Early Renal Intervention by Computerised Assessment) study showed a higher prevalence (5.1%) of known stages 3–5 CKD in the general population, based on analysis of single serum creatinine results held on GP computers; however, the true prevalence may be higher, as only 25% of the population were tested.3

View this table:
Table 1

Definition and stages of chronic kidney disease according to NKF-K/DOQI guidelines, prevalence in the US and recommended action plan1

StageDescriptionGFR (ml/min/1.73 m2)Prevalence (%)Actions
At increased risk, e.g. known diabetes or hypertension⩾60 (with CKD risk factors)Screening; chronic kidney disease risk reduction
1Kidney damage* with normal or increased GFR⩾903.3%Diagnosis and treatment; treatment of co-morbid conditions; slowing progression; cardiovascular risk reduction
2Kidney damage* with mildly decreased GFR60–893.0%Estimating progression
3Moderately decreased GFR30–594.3%Evaluating and treating complications
4Severely decreased GFR15–290.2%Preparation for kidney replacement therapy
5Kidney failure<15 (or dialysis)0.2%Kidney replacement (if uraemia present)
  • *Kidney damage for ⩾3 months, defined by structural or functional abnormalities of the kidney, with or without decreased GFR, manifest by either pathological abnormalities or markers of kidney damage, including abnormalities of blood or urine or abnormalities in imaging tests.

These estimates are higher than many had appreciated; in consequence, policy has increasingly focussed on the identification and management of patients with CKD. The National Service Framework (NSF) for Renal Services for England4,,5 had sections on both CKD and on pre-established renal failure (pre-ERF). CKD referral guidelines jointly developed by the Renal Association, the Royal College of Physicians of London Specialty Committee on Renal Disease, the Royal College of General Practitioners and the Association of Clinical Biochemistry, have also just been published.6 Implementation of these initiatives will probably lead to an increase in the referral of patients with CKD to nephrology services, and specifically, to greater numbers with pre-ERF being managed by renal services. This conclusion is supported by a population study in east Kent, in which 85% of prevalent patients with GFR <30 ml/min were not referred to a nephrologist, although this was not always inappropriate.7

Service planning for care of patients with advanced CKD is difficult, as there are no data on the number of patients approaching ERF in the UK. Moreover, there is no information on the facilities available for the care of such patients before RRT. This contrasts with the situation for RRT, where the UK Renal Registry will shortly have near complete population coverage,8 and where there have been Department-of-Health-funded surveys on patient numbers and the organization of RRT care, which continue on a regular basis.9

Important markers of good practice in the Renal NSF for patients with advanced CKD were: (a) Referral to a multi-skilled renal team (MSRT) at least one year before the anticipated start of dialysis: the specific organizational structure (e.g. low clearance clinic, LCC) was not defined. (b) People with ERF being given information about all forms of treatment so that an informed choice could be made. (c) Early referral for assessment and investigation for best means of dialysis access. (d) People with ERF should have been on the transplant waiting list within 6 months prior to the anticipated start of dialysis. (e) Anaemia treatment to maintain an adequate haemoglobin level.

We present data from a national survey of UK renal units that aimed to provide a baseline position relevant to the Renal NSF, by determining the current numbers of CKD patients under nephrology care, the organization and delivery of pre-ERF care (particularly MSRTs and LCCs), and perceived barriers to the development of such services.


The questionnaire was developed within the Demographic Study Group of the UK Renal Registry. The questionnaire was piloted at five renal units before being circulated to all 72 renal units in the UK (England, Scotland, Wales and Northern Ireland) in June 2004. The questionnaire refers to the situation in June 2004, and focuses mainly on the management of patients with CKD stage 4 and 5, as these are the patients who are thought most likely to progress towards requiring RRT. For this report, CKD stage 5 refers only to patients with estimated glomerular filtration rate <15 ml/min who are not on RRT.

Data were entered and stored in a Microsoft Access database by a single researcher. Data were then checked by another researcher to ensure correct data transfer from the paper version. Any queries from the responses were then followed up with the relevant contact person for the renal unit concerned.

Data were analysed using the SAS statistical package.


Number of patients

Of the 72 renal units, 70 (97%) responded. Thirty-five centres (50%) were able to provide data on the number of CKD patients under nephrological follow-up, and 25 (36%) of these were able to provide estimated GFR for these patients. In 21 (30%) of the units, details of all CKD patients were kept in the same clinical database as RRT patients.

In these 35 centres, there were a total of 78 000 patients with CKD, with a median of 2000 patients per unit (range 275–5685, IQR 1225–2975). In the 25 centres with estimated GFR data, there were a total of 8912 CKD stage 4 and 5 patients; a median of 321 patients per unit (range 53–819, IQR 200–489).

The median ratio of prevalent CKD patients/prevalent RRT patients in the units with data was 3.7 (IQR 2.7–5.7) and the median CKD stage 4 and 5/prevalent RRT ratio was 0.6 (IQR 0.4–1.1). CKD stage 4 and 5 made up nearly 20% of the total CKD patients (median 19%, IQR 11–40%).

Using these ratios and extrapolating using the total number of prevalent RRT patients in the UK in 2004 (estimated to be 38 00010) indicates there were about 140 000 (IQR 102 000–216 000) (38 000 × 2.7, 3.7 and 5.7) CKD patients under the care of UK nephrologists, of whom 23 000 were CKD stage 4 and 5 patients.

Multi-skilled renal teams (MSRTs)

Few units had a complete MSRT (Table 2). While most had a dietitian for CKD patients, and a specific person providing dialysis education, only 33% had a counsellor. Some units were creating more specific nursing roles, e.g. for diabetic patients or blood pressure management. Assuming that each role was being allocated to a different individual, there was no correlation between the number of available MSRT personnel and the number of patients with CKD stage 4 and 5 in the unit.

View this table:
Table 2

Multi-skilled renal team (MSRT) composition within renal centres

Units with the following MSRT personnel for CKD patientsUnits without the following MSRT personnel that feel they are needed (numbers)Units where MSRT personnel attend clinics where CKD patients are seen (numbers)
Dietitian99%0% (0/1)94% (63/69)
Dialysis education provider87%89% (8/9)93% (54/61)
Anaemia coordinator76%31% (5/17)76% (38/53)
Pharmacist72%60% (9/19)15% (7/50)
Social worker64%91% (20/25)47% (20/44)
Access coordinator53%75% (21/33)66% (23/37)
Counsellor33%78% (32/47)52% (11/23)
Diabetic nurse29%58% (22/48)33% (6/20)
Occupational therapist28%40% (16/49)22% (4/19)
Psychologist25%67% (26/51)53% (9/17)
Physiotherapist22%27% (12/54)21% (3/15)
Blood pressure nurse9%30% (16/63)67% (4/6)

The most important perceived staff shortages (a combination of where staff were not in post and perceived importance) were counsellors and psychologists. Dietitians and dialysis education providers were the main MSRT personnel attending CKD clinics in over 95% of the renal units.

There were regular MSRT meetings in 47 of the 70 units (67%), the frequency varying from 1 meeting/week to 1/13 weeks, with the majority of units having a weekly (49%) or a monthly (36%) meeting. In 36 of the 47 units (77%), the regular MSRT meetings had been in place for more than 1 year.

Of the 70 units, 49 (70%) ran clinics for CKD patients in neighbouring district general hospitals (DGH); the median number of clinics per each of these units was two (range 1–11, IQR 1–4). In 15 of the 49 units (31%), CKD patients in the peripheral DGH could only access the MSRT services via the main unit.

Low clearance clinics (LCCs)

50 (71%) units held an LCC for managing patients approaching RRT. In 10 of these 50 units, not all the consultants were using the clinic. A further 10 centres (15%) were planning to set up similar clinics. The median number of patients under the care of these clinics was 120 (range 25–850, IQR 64–252). The ratio of number of CKD patients stage 4 and 5 to the number of patients in each LCC ranged from 0.2 to 3.7. This indicates a lack of specificity in the definition of LCC; in some centres, not all patients with stage 4 and 5 CKD were followed-up in an LCC, and in other centres such clinics had more liberal inclusion criteria.

In 84% of units with an LCC, there was a renal nurse involved in the organization and running of the LCC. The amount of nurse responsibility varied, the main role being to facilitate clear and efficient communication with other personnel involved in the care of CKD patients. Some were involved in the delivery of CKD education, counselling, transplant assessments and prescribing and altering prescription under medical supervision. In some units, patients were reviewed by a nephrologist and the renal nurse on an alternate basis.

Pre-dialysis education

Apart from a specific dialysis education provider, education was also delivered by a variety of other professionals such as dialysis nurses, transplant co-ordinators, dietitians and pharmacists. While subjects such as types of dialysis, dietary restrictions, fluid balance, CKD-related anaemia, renal bone disease were very well covered, aspects of cardiovascular risk factors, sexual matters and psychological support were less commonly reported as being covered in such programmes.

Some units had education materials available in audio and Braille for the blind, and translated into Bengali, Cantonese, Gujarati, Hindi, Punjabi, Somali, Urdu and Welsh (Table 3).

View this table:
Table 3

Education materials available in other languages

LanguageHospitals where leaflets are available in the language
BengaliMiddlesex, Sheffield
CantoneseGlasgow, Sheffield
GujaratiLeicester, Sheffield, Preston, Queen Elizabeth (Birmingham)
HindiLeicester, Middlesbrough, Sheffield, Preston, Queen Elizabeth (Birmingham)
PunjabiCoventry, Leicester, Queen Elizabeth (Birmingham), Stoke-on-Trent
UrduGlasgow, Reading, Sheffield, Queen Elizabeth (Birmingham)
WelshBangor, Cardiff, Rhyl, Swansea, Wrexham

Dialysis access services

Fifty-five units (79%) had a dedicated vascular access surgical team, with 41 (59%) providing clinics for pre-access assessments and 27 (39%) providing clinics for post-access formation follow-up. In the 55 units with a dedicated vascular access team, the median waiting time for elective fistula surgery was 6 weeks (range 1–36 weeks), compared with 12 weeks (range 4–26 weeks) in the 14 units without a dedicated team. There were dedicated theatre sessions for access formation in 51 units (73%), the number of sessions ranging between 1 session per month to 6 sessions per week.

In 28 units (40%), nephrologists performed Tenchkoff catheter insertions. In these centres. the median waiting time for catheter insertion was 2 weeks (range ‘within same week’−8 weeks), compared with a median of 4 weeks (range ‘within same week’−12 weeks) in centres where nephrologists do not perform insertions. No statistical comparison was made for the median waiting times, as these data were self-reported and may have involved some estimation from the reporting unit.

Forty units (57%) had a dedicated renal interventional radiologist. Thirty-seven centres (53%) used an access coordinator to organize and prioritize the waiting list. Thirty-two units (46%) were already entering information regarding access formation and problems into a database.


This national survey documents the prevalence of patients with CKD under the care of nephrologists and to assess the organization of nephrology services for their management. There were an estimated ∼140 000 CKD patients under the care of nephrologists in UK renal units, of whom ∼23 000 were CKD stage 4 and 5 (not on dialysis). These estimates were calculated from only a proportion of the renal units in the UK. The number of patients with CKD in the population could be substantially higher, as these figures exclude those with recognized CKD under the care of nephrologists not working in main renal units, and those with CKD (recognized or not) under the care of other specialists such as cardiologists, diabetologists, urologists and in primary care.

A large proportion of CKD nephrology workload was from early CKD; only about 20% of the CKD patients were in stages 4 and 5. This also suggests that a significant proportion of patients entering the RRT programme do so without prior substantive care by nephrologists. The NSF and the CKD referral guideline are expected to lead to greater detection of CKD patients through use of GFR estimation, which may increase referrals. However, there should be better shared management of CKD patients between nephrologists and primary care physicians, which will enable secondary nephrology services to focus more on patients with more advanced CKD, while still offering joint care with primary care for patients with less severe disease.

One of the important key markers of good practice in the NSF for pre-ERF care was referral to an MSRT at least 1 year before the anticipated start of dialysis.4 Identification of these patients can be difficult, as there is a high risk of cardiovascular mortality.11 Moreover, some patients opt for conservative management, which needs to be taken into account in the service planning. A recent national survey of renal units in the UK showed significant variation in the provision of palliative care services across the country, with only 39% of the units having specified personnel involved in the palliative care aspects of patients with ERF.12

A report by the British Renal Society (BRS)—The Renal Team: A Multi Professional Renal Workforce Plan for Adults and Children with Renal Disease13—outlined the personnel that should constitute a multi-skilled renal team. In the survey reported here, the availability of the recommended types of professional varied between the units, with few having the full complement. Notably lacking were social workers, psychologists and counsellors, suggesting that formal emotional and psychological support is often a relatively low priority, especially in a financially constrained, medically driven environment.

The Renal NSF recommended that people approaching dialysis be given information about all forms of treatment, so that an informed choice of modality can be made. Rather surprisingly, not all centres had a specific dialysis education provider to facilitate this: this is another area that needs to be improved. This can be a problem for patients whose first language is not English, and the availability of translated dialysis and CKD education materials in some centres is therefore welcomed. A multi-lingual dialysis education DVD is currently being jointly developed by the BRS and the NKRF (National Kidney Research Foundation) and should be available by autumn 2006.14

It is generally accepted, but not proven, that the multi-disciplinary approach is the best way to manage the complex needs of pre-ERF patients, as recommended by the Renal NSF.4 However, evidence for the effectiveness of pre-ERF care, and which model of care to use, is limited. Studies by Binik et al.15 and Devins et al.16 suggested a beneficial effect from a more enhanced intensive education programme for pre-ERF patients, by delaying the start of RRT by at least 3 months. In a study by Levin et al.,17 patients who were attending a pre-dialysis clinic programme had better clinical variables at the start of dialysis, were less likely to have a problem with symptomatic uraemia, less likely to start dialysis in an emergency manner and less likely to start dialysis as an in-patient. More recently, Curtis et al.18 showed that patients attending a multi-disciplinary clinic had a higher haemoglobin, calcium and albumin at the start of dialysis, and a better survival rate, compared to patients receiving standard nephrology care. In contrast, a study by Harris et al.19 showed that intensive multi-disciplinary management approach did not appear to offer any significant advantage, in terms of progression of renal disease or mortality rate, compared to standard nephrological care. The conflicting results of the study by Harris et al. highlight the need for further prospective research to identify the most effective methods of organizing care for CKD patients approaching dialysis. There are currently several prospective studies in progress in the USA20–22 and Canada23 aiming to address this question. At present, it is fashionable to deliver and streamline the multidisciplinary care required in an LCC setting: 71% of the units were using this approach.

Geographical accessibility to pre-ERF care varies, with some units not having any outreach clinics for managing patients with advanced CKD referred from further afield. Even in units where outreach clinics were already set up, some patients were not receiving the same MSRT input as those patients being followed up in the main unit. One of the issues raised by some of the units is that these MSRT personnel are not funded to work outside the remit of the employing hospital, and also that given the level of cost, it may not prove to be cost-effective. A more flexible way of providing outreach CKD care in neighbouring DGHs is therefore required; use of facilities and/or space in existing satellite dialysis units is an option that needs to be explored.

Preparation for dialysis in terms of timely provision of vascular or peritoneal access is still not optimum, with some centres still lacking a dedicated surgical team to perform these procedures. There are currently Department of Health funded pilot projects which aim to identify ways of improving vascular access service, and the UK Renal Association is surveying renal units to identify current practices in more detail. Data on this are already being collected electronically in 21 units. These data could be linked with the UK Renal Registry database, which would enable analysis of the impact of pre ERF care on RRT prognosis.

In conclusion, CKD represents a significant workload for UK nephrologists. The provision of pre-ERF services is variable: few units have the full complement of MSRT personnel, and there are some problems of geographical accessibility. Dialysis access services and socio-psychological support are the two main aspects of service delivery which need to be addressed. Improvement of CKD data collection will enable more investigations into the quality of pre-ERF care. The data in this study serve as a good baseline to gauge the impact of the implementation of the recommendations of the NSF on service delivery, both locally and nationally; such a survey could be repeated to gauge progress.


We thank all the units who participated in the survey. Funding for this survey was provided by the UK Renal Registry, through the annual capitation fee charged to renal units in England and Wales.


View Abstract