PARIS—Primary care physicians who use a clinical decision support system (CDSS) may be able to avoid late referrals for renal replacement therapy (RRT), researchers reported at the 49th Congress of the European Renal Association-European Dialysis and Transplant Association.
Christopher Farmer, MD, a renal consultant at East Kent University Hospitals NHS Foundation in Kent, UK, and colleagues tested the impact of a CDSS on the incidence of late referral to their renal service.
“Timely referral is requisite for ensuring favorable outcomes in patients reaching end-stage renal failure,” Dr. Farmer said. “We know that patients who are referred late are about twice as likely to die within 90 days of starting dialysis, they are more likely to be denied the possibility of choosing their treatment, and they are much less likely to start peritoneal dialysis, and so they tend to start on hemodialysis. And, finally, there is no possibility for preemptive transplantation, which is clearly the best treatment for end-stage kidney disease. So, in short, you are limiting the patient’s choice.”
The investigators developed a CDSS that is an algorithm available as computer software that screens patients having serum creatinine estimates in primary care. Data are routinely extracted from primary care databases and patient-specific advice is given to primary care physicians about referrals, medical management of co-morbidities such as hypertension and diabetes, and additional investigations that might be needed to screen patients for anemia or disorders of calcium and phosphate.
Over a recent four-year period, Dr. Farmer’s team tested the tool in patients for whom estimated serum creatinine levels were available. “What we found was significantly fewer ‘crash landers’ when physicians used the system,” Dr. Farmer said. “By ‘crash landing,’ I am referring to patients who were referred less than 90 days before starting RRT.”
The “crash lander” rate in patients from primary care practices that did not use the CDDS was 25% versus 6% in patients from practices that had used the CDDS. “The difference between the two groups was huge and significant,” Dr. Farmer said.
He cautioned that the primary care practices included in the analysis were not randomly selected. “What this means is that there may be selection bias in that practices that are better may be more innovative and may be more likely to opt to participate in a managed care program that uses a CDSS system or the like,” he said.