The role of blood pressure (BP) changes in dementia is debatable. We aimed to analyse how resting and postural BP changes relate to incident dementia over a long-term follow-up. In the prospective population-based Malmö Preventive Project, 18,240 study participants (mean age: 45 ± 7 years, 63% male) were examined between 1974 and 1992 with resting and standing BP measurement, and re-examined between 2002 and 2006 at mean age of 68 ± 6 years with resting BP. A total of 428 participants (2.3%) were diagnosed with dementia through Dec 31, 2009. The association of resting and postural BP changes with risk of dementia was studied using multivariable-adjusted Cox regression models controlling for traditional risk factors. Diastolic BP (DBP) decrease on standing indicated higher risk of dementia [Hazard ratio (HR) per 10 mmHg: 1.22; 95% confidence interval (CI) 1.01-1.44, p = 0.036], which was mainly driven by increased risk in normotensive individuals. Higher systolic (SBP) and diastolic BP at re-examination was associated with lower risk of dementia (HR per 10 mmHg: 0.94; 95% CI 0.89-0.99, p = 0.011; and 0.87; 0.78-0.96, p = 0.006, respectively). Extreme decrease in SBP/DBP between baseline and re-examination (4th quartile; -7 ± 12/-15 ± 7 mmHg, respectively) indicated higher risk of dementia (HR 1.46; 95% CI 1.11-1.93, p = 0.008, and 1.54; 95% CI 1.14-2.08, p = 0.005; respectively) compared with reference group characterised by pronounced BP increase over the same period (1st quartile; +44 ± 13/+15 ± 7 mmHg). Diastolic BP decrease on standing in the middle age, decline in BP between middle-and advanced age, and lower BP in advanced age are independent risk factors of developing dementia.