Background: Increased somatostatin plasma concentration has been found in patients with vascular dementia. However, it is unknown whether or not somatostatin levels may predict dementia development in the general population. To this end, we sought to assess the association of circulating N-terminal prosomatostatin (NT-proSST) with incident dementia among community-dwelling older adults. Methods: In the prospective population-based Malmö Preventive Project, 5,347 study participants (mean age: 69 ± 6years; 70% men) provided plasma for the determination of NT-proSST concentration. Of these, 373 participants (7%) were diagnosed with dementia (120 Alzheimer's disease, 83 vascular, 102 mixed, and 68 other aetiology) during a follow-up period of 4.6 ± 1.3 years. The association of NT-proSST with the risk of dementia and its subtypes was studied using multivariable-adjusted Cox regression models controlling for age, gender, body mass index, systolic blood pressure, antihypertensive treatment, smoking, diabetes, lipid levels and prevalent stroke. Results: Higher levels of NT-proSST were significantly associated with an increased risk of vascular dementia (hazard ratio [HR] per 1 SD: 1.29; 95% CI 1.05-1.59; p = 0.016), whereas no association was observed with Alzheimer's disease (HR per 1 SD: 0.99; 95% CI 0.81-1.20; p = 0.91), all-cause dementia (HR per 1 SD: 1.04; 95% CI 0.94-1.16; p = 0.44), and mixed dementia (HR per 1 SD: 0.98; 95% CI 0.79-1.21; p = 0.84). Levels of NT-proSST above 563 pmol/L (highest quartile) conferred distinctly increased risk of vascular dementia (HR 1.66; 95% CI 1.05-2.63; p = 0.029) compared with lower values. Conclusions: Higher levels of circulating N-terminal-prosomatostatin are associated with increased incidence of vascular dementia. Our findings might be of importance for the understanding of dementia development in older adults.