Abstract Background We reported that a declining slope in eGFR evaluated from 3-year observation could predict subsequent CKD among the relatively healthy population, even adjusted for baseline eGFR. However, its certainty may vary depending on the presence of comorbidities such as diabetes and hypertension, which are also recognized as the risk factors of CKD. Methods Annual health check-up data from a Japanese company spanning from 2009 to 2022 was used. The analyses included participants having 4 times continuous eGFR measurements for 3 years, undergoing health checkups at least once during a follow-up period, with negative for urinary protein (UP), and non-diabetic. Individual ‘eGFR slopes’ were calculated by the regression line with 4 times eGFR measurements. Participants with slopes of < -5 per year were categorized as the “decreased” group and -5 or more were categorized as the “stable” group. The cases of positive UP or less than 45mL/min/1.73m2 in eGFR during a follow-up period were defined as CKD. Cox regression analyses adjusted for sex, age, and eGFR at baseline were performed to calculate the hazard ratios and 95% CIs for CKD in the decreased group (reference; stable), stratified by hypertension. Results Out of 5598 study participants(men 36%, mean age 48.2±10.1years, mean follow-up period 4.3 years), 260 CKD cases (4.6%) were observed. CKD cases accounted for 4.5% (235/5218) in the stable group and 6.6% (25/380) in the decreased group. Multivariate-adjusted hazard ratios [95%CIs] of CKD by eGFR decline was 1.50[0.99-2.27]. The eGFR decline increased the risk of CKD only in participants with hypertension (2.09[1.09-4.04]), but not in those without hypertension (1.25[0.73-2.13]). After excluding individuals with eGFR slopes exceeding 10 per year to consider for hyperfiltration of eGFR, the results were consistent. Conclusions The eGFR decline over a 3-year observation may predict subsequent CKD only in individuals with hypertension. Key messages • With a relatively short-term evaluation period of three years, eGFR decline can predict subsequent CKD only in hypertensive individuals. • In middle-aged hypertensive patients, it would be important for the prevention of CKD to pay attention not only to low eGFR but also to the progression of eGFR decline.