The aim of this study was to determine the changes in prevalence of xerostomia in subjects from 50 to 65 yr of age. Questionnaires were sent to all subjects who were born in 1942 and were living in two Swedish counties in 1992, 1997, 2002, and 2007. The analyses focused on those who answered the questionnaires both in 1992 and in 2007. The response rate was 71.4% (n = 6,346) in 1992 and 73.1% (n = 6,078) in 2007. Of those who answered the questionnaire in 1992, 74.3% (n = 4,714) also responded in 2007. There was an almost linear increase in the prevalence of xerostomia at the four study time-points (i.e. when the subjects were 50, 55, 60, and 65 yr of age). Xerostomia was more prevalent at night than during the day. The pooled prevalence of night-time and daytime xerostomia was 6% at 50 yr of age and 15% at 65 yr of age, and it was higher in women than in men on both occasions. Logistic regression analyses showed that impaired health and smoking were significantly associated with daytime xerostomia but not with night-time xerostomia. Despite the increase in prevalence of xerostomia from 50 to 65 yr of age, there was considerable variation during the observation period. The incidence rate was 13% (507/4,015) and the disappearance rate was 42% (104/250) (dichotomized answers).