BACKGROUND: Few studies have evaluated the long-term effects of disasters on older adults. We examined if older individuals experienced increased mental health (MH) effects immediately and long-term after Hurricane Sandy and if these effects varied by socioeconomic status (SES), disease subtypes, number of comorbidities, and length of stay. METHODS: We identified older patients (≥64 years) with hospital admissions and Emergency Department (ED) visits (2001-2015) with primary diagnosis of MH diseases using the New York State discharged data. We quantified both short-term (immediately post-Sandy) and long-term effects (3-month, 1-year, 2-year, and 3-year) following Hurricane Sandy, and used the pre-Sandy period in the affected counties as the control period. Poisson regression was used to compare daily counts of MH cases overall and by multiple strata pre-/post Sandy. FINDINGS: Older individuals had significantly increased risk of MH ED visits immediately (32%), and 3-months, 1, 2, and 3-years after Sandy (2%, 9%, 15%, and 10%, respectively). MH hospital admissions did not increase immediately, but significantly increased by 8% a year later. Males and those with low SES had delayed, but increased risks of MH 1-3 years after Sandy (RRs range: 1.14-1.71). The top MH subtypes after Sandy were psychosis, mood disorders, substance abuse, suicide, and anxiety (RRs range: 1.12-2.62). After Sandy, patients with ≥8 comorbidities increased from 15% to >25%, along with their length of hospital stay. CONCLUSION: We found long-term adverse effects of MH after Sandy, especially among vulnerable populations, which may help plan future disaster preparedness and recovery efforts.