THE DEVELOPMENT OF AN EXPERT RENAL SYSTEM TO HIGHLIGHT AND MANAGE CKD IN THE COMMUNITY

B Klebe1, R Cooley2, S Grindle2, P Stevens1, D O’ Donoghue3, C Farmer1

1Kent and Canterbury Hospital, 2University of Kent, 3Hope Hospital, Salford,

PROBLEM: Population based studies have highlighted a high prevalence of chronic kidney disease (CKD). The majority of CKD patients are not known to renal services and are far more likely to die than to progress to established renal failure. Early identification and management is the key to reduce the burden of CKD, however, given the large numbers of patients and lack of resources, it would not be feasible to refer all patients with CKD to renal services.

PURPOSE: Phase one of the NEOERICA project has already demonstrated that GP databases can be used to analyse and screen populations for CKD. The next phase was to develop an Expert Renal System to facilitate the management of CKD.

DESIGN: In a pilot study, a computer program was developed to identify patients with stage 3-5 CKD and facilitate their management in the community. The Expert System worked on a static dataset of patients from GP databases in Kent, Surrey and Greater Manchester. It estimated GFR and further categorised patients depending on their renal function and co-morbidity.

FINDINGS: Over a 15 month period 30,759 patients were identified and categorised by the program. The expert system highlighted 1139 new patients with stage 3-4 CKD and 14 patients with stage 5 CKD per million population per month. Taking a GFR decline of > 5ml/min/year as evidence of unacceptable progression, 2861 patients per million per month were highlighted requiring more tests to clarify their GFR and rate of renal progression.

CONCLUSION: The expert system was able to categorise CKD patients from the GP databases and highlight those with progressive renal disease needing Nephrology referral as well as those requiring risk factor modification. This confirms that the burden of CKD and its associated co-morbidity is too great to be managed solely by secondary care and dictates a requirement for redesign of the management of CKD.

RELEVANCE: A good IT infrastructure should be part of the effective delivery of a CKD care-plan. In order to improve CKD outcomes and delay progression, early identification of renal disease and risk factor modification is essential. Expert Systems can be utilized as management tools to assist doctors in targeting CKD and highlighting those patients requiring disease management. The Expert System will also serve as a tool to describe co-morbidity and stratify risk in CKD patients.