The epidemiology of scaphoid fractures has been described in limited populations, and incidence reports have been inconsistent. We investigated the nationwide incidence of scaphoid fractures by evaluating data on 34,377 patients in the Swedish National Patient Register for the years 2006–2015 regarding diagnosis, age, sex and treatment. The data were validated in 300 random patients, and incidence rates were adjusted accordingly. Forty-one per cent of the initially diagnosed fractures were false positives. The adjusted true fracture incidence rate was 22 per 100,000 person-years. During the decade studied incidence rates decreased in younger men and increased in middle-aged women. The incidence of surgical treatment vs. non-operative treatment did not change over time. Men were treated surgically more often than women (6% vs. 3%) and had a greater risk for nonunion (3% vs. 1%).