There is little information about the reproducibility of the white coat effect, which was treated as a continuous variable. To investigate a long-term interval reproducibility of the white-coat effect as a continuous variable. We selected 153 participants without antihypertensive treatment (men, 22.9%; age, 64.4 years) from the general population of Ohasama, Japan, to assess the repeatedly measured white-coat effect (the difference between blood pressures at the office and home) in a 4-year interval. The reproducibility was assessed by testing the intraclass correlation coefficient (two-way random effect model-single measures). The white-coat effect for systolic/diastolic blood pressure slightly decreased by 0.17/1.56 mmHg at the 4-year visit on average. The Bland–Altman plots showed no significant systemic error for the white-coat effects (P ≥ 0.24). The intraclass correlation coefficient (95% confidence interval) of the white-coat effect for systolic blood pressure, office systolic blood pressure, and home systolic blood pressure were 0.41 (0.27–0.53), 0.64 (0.52–0.74), and 0.74 (0.47–0.86), respectively. Change in the white-coat effect was mainly affected by a change in office blood pressure. Long-term reproducibility of the white-coat effect is limited in the general population without antihypertensive treatment. The change in the white-coat effect is mainly caused by office blood pressure variation.