Background: Few randomized controlled trials (RCTs) investigated the effects of mindfulness intervention on affective and cognitive symptoms in older adults with mild cognitive impairment (MCI). Furthermore, no RCTs on mindfulness followed participants beyond two years. Objective: To examine the longitudinal effects of a mindful awareness practice (MAP) intervention on depressive, anxiety, and cognitive symptoms in MCI. Methods: In this parallel-arm and assessor-blinded RCT, 55 community-dwelling older adults with MCI were randomized into the MAP or active control, i.e., health education program (HEP). Intervention sessions were conducted weekly for three months and monthly for the subsequent six months. Assessments and follow-up were conducted at baseline, 3-month, 9-month, and 5-year time points. Depressive, anxiety, and cognitive symptoms were measured using the Geriatric Depression Scale-15 (GDS-15), Geriatric Anxiety Inventory-20 (GAI-20), and Mini-Mental State Examination (MMSE), respectively. Linear-mixed models, following the intention-to-treat principle, were used for data analyses. Results: A total of 55 participants aged 60 to 86 (Mean age: 71.3±6 years old) was recruited, with n = 28 allocated to the MAP arm and n = 27 allocated to the HEP arm. Compared to HEP, GDS-15, GAI-20, and MMSE scores did not differ significantly in MAP during follow-ups. Conclusion: Compared to HEP, MAP did not improve affective symptoms nor delay deteriorations in general cognition in community-dwelling older adults with MCI. Compared to our previous findings showing domain-specific improvements in MAP over HEP in attention and memory up to 9 months, this study highlights the importance of examining domain-specificity using detailed cognitive measures in non-pharmacological intervention with MCI.