Statistical analysis

Participants were divided into two groups based on baseline electrocardiographic findings: individuals with and without electrocardiographic LVH (LVH group and non-LVH group, respectively). Continuous variables are expressed as the mean and standard deviation and were analyzed using the t test unless otherwise noted. All categorical variables were expressed as raw numbers and percentages and were compared using the chi-square test.

Kaplan–Meier cumulative incidence curves were compared by the log-rank test. Multivariate Cox proportional hazards models were used to calculate the hazard ratios for the incidence of renal function decline. For adjusting potential confounding factors, we developed three models. Model 1 included sex and age. Model 2 additionally included baseline eGFR and baseline urinary protein (positive or negative). Model 3 further included body mass index (BMI), current smoking status, diagnosis of hypertension, diagnosis of diabetes mellitus, history of cardiovascular disease including heart disease or stroke, systolic BP, LDL-C, uric acid, and HbA1c at baseline. To determine whether the association between electrocardiographic LVH and renal function decline was affected by sex, age, BMI, diagnosis of hypertension, systolic BP, HbA1c, eGFR, and urinary protein, subgroup analyses were additionally performed using model 3. A two-sided P value < 0.05 indicated statistical significance. All the analyses were conducted with SAS 9.4 (SAS Institute Inc., Cary, NC, USA).