Chronic kidney disease in Australian general practice: its prevalence, and associated patient characteristics and comorbidities

Aim: To provide an estimate of the prevalence and associations of CKD in Australian general practice, based on routinely collected electronic record data.

Background: Whilst most chronic kidney disease (CKD) patients are identi- fied and managed in Australian general practice, the characteristics of this population, based on a large general practice dataset, have not been described. MedicineInsight, a primary care quality improvement initiative from NPS MedicineWise, collects de-identified patient information, including prescriptions and pathology results, from over 600 participating general practices to create a longitudinal database of over 3.5 million Australians.

Method: MedicineInsight data (01/01/2013 to 01/06/2016) was used to identify CKD (stages 3 to 5). Patients with two or more eGFR results <60ml/ min/1.73m2, at least 90 days apart, were identified and characterized.

Result: Of 1,483,416 patients in the general cohort, 61,102 (4.1%) had CKD (stages 3 to 5). Of these, 56% were female, 28% were ≤ 70 yrs, 14% living in outer regional or remote locations, and 5% were smokers. The most common co-morbid conditions included hypertension (78% of CKD cohort v 22% of general cohort), CVD (39% v 7%), diabetes (31%v 8%) and AF (15% v 2%, all P<0.001). Most commonly prescribed drugs in the CKD cohort were PPIs, statins, opioids and benzodiazepines. Diagnoses of CKD were recorded with more than 30 different terms, including renal “failure”, “insufficiency” or “disease”, but only 11,289 (18.4%) with CKD3-5 had this diagnosis recorded. Urine ACR was more likely to be tested if a CKD diagnosis was recorded (64.6% v 47.3%).

Conclusion: Routinely collected electronic record data from Australian general practice provide a large, valuable primary care-based dataset on CKD, which could inform national quality improvement initiatives.