BACKGROUND: Qualitative data show negative impacts of menstruation on health and education in many settings, but there are few longitudinal quantitative studies of the impact of menstruation. We analyse associations with menstrual anxiety and school attendance in a study of Ugandan secondary school students. METHODS: Data were from a longitudinal pilot study of a menstrual health intervention (MENISCUS), conducted in two secondary schools in Entebbe sub-district, Uganda. Self-completed menstrual-related data, including menstrual anxiety, were collected from 232 participants pre- and post-intervention. A sub-cohort of 100 randomly-selected post-menarcheal girls were asked to self-complete daily diaries during 10 months of follow-up, with data on menstrual flow, pain, and school attendance. We used multivariable logistic regression to estimate associations with menstrual anxiety among all girls at baseline, and random-effects logistic regression to estimate associations of menstrual characteristics with school non-attendance for 3 months pre-intervention in the sub-cohort, adjusting for within-girl clustering. RESULTS: Overall, 130/222 (58.6%) of menstruating girls reported being anxious about their next period. Menstrual anxiety was higher in those not living with their mother (adjusted odds ratio (OR) = 1.91; 95% confidence interval (CI) 1.01-3.60), believing menstrual myths (aOR = 1.83; 0.95-3.50 for not agreeing that it is healthy for a girl to run, dance or cycle during her period; aOR = 1.97; 1.04-3.73 for agreeing that when a girl has her period she is unclean), lower menstrual confidence (aOR = 2.49; 1.33-4.65 for avoiding physical activity during her period; aOR = 1.68; 0.89-3.17 for not feeling comfortable to talk to other girls about her period; aOR = 2.89; 1.28-6.54 for agreeing that boys/girls tease them about their periods; and aOR = 2.27; 1.21-4.27 for worrying about being teased during her period). Those with lower knowledge about menstruation were less likely to report anxiety (aOR = 0.44; 0.23-0.84). During the pre-intervention period of the sub-cohort, school non-attendance was associated with menstrual pain, with 21.7% of girls missing school on days when they reported pain vs. 8.3% on days when no pain was reported (aOR = 3.82; 1.66-8.77). CONCLUSIONS: Menstruation causes substantial anxiety in Ugandan schoolgirls, and menstrual pain is associated with missing school on period-days. Menstrual health interventions should address socio-cultural aspects of menstruation to reduce anxiety, and provide education on pain management strategies to support school attendance.