Objectives To explore the cost-effectiveness of a supervised moderate-To-high intensity aerobic exercise programme in people diagnosed with Alzheimer's disease (AD) and estimate incremental cost-effectiveness ratios (ICER) using participant-reported and proxy-reported measures of health-related quality of life (HRQoL) Design A cost-effectiveness analysis of economic and HRQoL data from a randomised trial delivered over 16 weeks. Setting Memory clinics in Denmark. Participants 200 individuals with mild AD aged 50-90 years gave informed consent to participate in the study. Participants were randomised to control or intervention group. Interventions Control group received treatment as usual. The intervention group performed 1 hour of supervised moderate-To-high intensity aerobic exercise three times weekly for 16 weeks. Primary and secondary outcomes measures Different physical, functional and health measures were obtained at inclusion (baseline) and 4 and 16 weeks after. HRQoL (EuroQol-5 Dimensions-5 Levels/EQ-Visual Analogue Scale) was reported by the participants and the primary caregivers as proxy respondents. Differences in HRQOL as reported by the participant and caregiver were explored as were different values of caregiver time with respite from care tasks. Results The intervention cost was estimated at €608 and €496 per participant, with and without transport cost, respectively. Participants and caregivers in the intervention group reported a small, positive non-significant improvement in EQ-5D-5L and EQ-VAS after 16 weeks. The ICER was estimated at €72 000/quality-Adjusted life year using participant-reported outcomes and €87000 using caregiver-reported outcomes. Conclusions The findings suggest that the exercise intervention is unlikely to be cost-effective within the commonly applied threshold values. The cost of the intervention might be offset by potential savings from reduction in use of health and social care. Trial registration number https://clinicaltrials.gov/ct2/show/NCT01681602.