The aim of this study was to monitor long-term changes in bone mineral density (BMD) after menopause and factors affecting BMD. The study population consisted of a random sample of 3222 women from the Kuopio Osteoporosis Risk Factor and Prevention (OSTPRE) study, of which 62.1% were postmenopausal at the beginning of the study. This group of women underwent dual-energy X-ray absorptiometry (DXA) measurements at the femoral neck every 5 years from baseline (in 1989) up to 25-year follow-up. They also responded to risk-factor questionnaires at 5-year intervals. During the 25-year follow-up, the baseline cohort decreased to 686 women. The women were divided into quartiles based on their baseline BMD. Self-reported hormone replacement therapy (HRT) and corticosteroid use were divided into ever users and never users. Morbidity was assessed as the total number of self-reported diseases and BMD-affecting diseases. The mean 25-year BMD change was found to be −10.1%, p < 0.001. Higher baseline BMD was associated with higher bone loss rate;​ the reduction in the highest quartile BMD was 11.1% and in the lowest quartile 7.4% (p = 0.0031). Lower baseline body mass index (BMI) and a greater increase in BMI were found to protect against postmenopausal bone loss (p < 0.001). The lowest bone loss quartile included 15.2% more HRT users than the highest bone loss quartile (p = 0.004). The number of diseases/bone-affecting diseases, use of vitamin D/calcium supplementation, use of corticosteroids, smoking or alcohol use had no statistical significance for annual bone loss rate. This study presents hitherto the longest (25-year) BMD follow-up in postmenopausal women. The linear femoral neck bone loss of 10% was less than previously assumed. A 5-year BMD change appeared to predict long-term bone loss in postmenopausal women.