Both hypertension and diabetes in middle-aged individuals have been suggested to be predictive indicators of cognitive decline. However, the association of hypertension, diabetes and their combination with cognitive functioning is still controversial in older people. The purpose of this study was to investigate the association between cognitive decline and hypertension, diabetes, and their combination in 70-year-old people based on a 3-year longitudinal analysis. Four hundred and fifty-four people aged 70 (±1) years who participated in the Japanese longitudinal cohort study of Septuagenarians, Octogenarians and Nonagenarians Investigation with Centenarians (SONIC) were recruited randomly from a general population and were monitored for 3 years. The data, including most of the demographics, cognitive functioning measured by the Montreal Cognitive Assessment Japanese version (MoCA-J), blood pressure, blood chemistry and other medical histories, were collected at baseline and during the follow-up. The prevalence of hypertension noted in the follow-up survey was significantly higher than than noted at baseline. The mean MoCA-J score at follow-up was not significantly different from the score obtained at baseline. However, the participants with diabetes, especially combined with hypertension at baseline, had significantly lower MoCA-J scores than those without lifestyle-related diseases. The combination of hypertension and diabetes was still a significant risk factor for cognitive decline, considering the MoCA-J scores obtained during the follow-up after adjustments at baseline, relative to sex, body mass index, dyslipidemia, smoking, excessive alcohol intake, antihypertensive treatment and education level (β=-0.14; P<0.01). Our findings indicate that diabetes and the combination of hypertension and diabetes are clear risk factors for future cognitive decline in elderly individuals who are 70 years of age.