Authors (including presenting author) :
L M LEUNG (1), T K CHU (1), R S Y CHENG (1), F TSUI (1), M L CHAN (1), J LIANG (1), Y S NG (1)
Affiliation :
(1)Department of Family Medicine and Primary Health Care, New Territories West Cluster
Introduction :
Diabetic kidney disease (DKD) causes significant burden to individuals and the health care system. Improving standard of care for diabetes in primary care could reduce the incidence and progression of DKD. Overseas studies showed improving disease coding for chronic kidney disease was associated with better standard of care in primary care setting.
Objectives :
To assess association between disease coding in medical record and clinical outcomes of DKD in a primary care clinic in Hong Kong.
Methodology :
A registry of patients having regular follow up for type 2 diabetes and chronic kidney disease, defined as persistent reduction in eGFR (CKD-EPI) for period of at least 90 days (<60 ml/min/1.72m2) or presence of persistent albuminuria confirmed with 2 or more urine specimens at least 3 months apart (urine ACR >2.5mg/mmol in male or >3.5mg/mmol in female), was developed. A retrospective review of medical records of all patients with DKD followed up in the clinic in the period from July 2020 to June 2021 was performed based on the registry, to evaluate the performance of disease coding for chronic kidney disease (International Classification of Primary Care-2 codes of U98 and/or U99) and clinical outcomes (HbA1c, blood pressure, unplanned admission and casualty attendance).
Result & Outcome :
Medical records of 794 patients with DKD were reviewed. Only 17% of them had been coded chronic kidney disease. Mean age of the sample was 60. Female to male ratio was 1:1.6. More than 90% of them had urine ACR >3.5mg/mmol and 29% had eGFR <60 ml/min/1.72m2. Two third of the sample had HbA1c <7%. Adjusted for age, gender, eGFR and urine ACR by regression, the group without chronic kidney disease codes were less likely to achieve target HbA1c <7% (p=0.033), while there was no statistically significant difference in terms of proportion achieving target blood pressure (<130/80), number of unplanned admission and casualty attendance between the groups.
Conclusion
Less proportion of achieving target HbA1c was associated with missing disease coding of DKD in medical record. Improving DKD awareness and its coding practice in medical record may help to achieve better disease outcome in primary care setting.
