Background The effectiveness of identifying and monitoring early-stage chronic kidney disease (CKD) is not fully recognised. This study quantified people with undiagnosed CKD among the middle-aged Japanese population and clarified potential risks of untreated CKD.

Methods We included 71 233 individuals who underwent annual health check-ups (AHC) in 2014 for both baseline and follow-up proteinuria and serum creatine measurements. CKD was identified by AHC data as proteinuria or estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m². We differentiated undiagnosed from diagnosed CKD using the medical claims database. In undiagnosed CKD, we assessed risk differences for disease progression, defined as an eGFR decline slope >3 mL/min/1.73 m²/year or proteinuria incidence over 3 years, between those who visited a physician for CKD treatment within 6 months after AHC and those who did not.

Results CKD prevalence was 5.7% (5.2% undiagnosed and 0.5% diagnosed). Only 2.1% of the patients with undiagnosed CKD visited a physician for CKD treatment within 6 months after AHC. Between-group risk differences in instrumental variable adjustment models showed that those left untreated progressed to kidney diseases 16.3% more often than those who visited physicians for CKD treatment.

Conclusion CKD was undiagnosed in 5.2% of the middle-aged general population. Only a few people visited physicians for CKD treatment. Visiting physicians for CKD treatment during the first 6 months after screening may be associated with a lower risk of kidney disease progression.