OBJECTIVES: Identify the health profiles of older nursing home residents with and without at-admission self-reported suicidal ideation (SI) and examine the association between the identified profiles and self-reported SI at 90 days. METHODS: Using the Minimum Data Set 3.0 and the ninth Patient Health Questionnaire-9 (PHQ-9) item, we identified 15,277 older residents with and 562,184 without self-reported SI at nursing home admission. Latent class analysis, using frailty, cognitive impairment, palliative care index, pain, and remaining PHQ-9 items as indicators, identified health profiles by at-admission SI and the BCH method estimated their association with SI at 90 days. RESULTS: Profiles identified for residents without at-admission SI were: (1) frail and depressed(NoSI) (prevalence: 33.9%); (2) frail and severe cognitive impairment(NoSI) (38.1%); (3) pre-frail(NoSI) (28.0%). Residents in the frail and depressed(NoSI) group had greater odds [adjusted OR: 2.80; 95% Confidence Interval: 2.60–3.00] while those in the frail and severe cognitive impairment(NoSI) group had lower odds [aOR: 0.79; 95% CI: 0.71–0.86] of 90-day SI than those in the pre-frail(NoSI) group. Profiles identified for residents with at-admission SI were: (1) frail and all depressive symptoms(SI) (22.8%); (2) frail and some depressive symptoms(SI) (32.2%); (3) frail and severe cognitive impairment(SI) (22.9%); (4) pre-frail(SI) (22.0%). Compared to those in the pre-frail(SI) group, residents in the frail and all depressive symptoms(SI) group had greater odds of continuing reporting SI at 90 days [aOR: 1.22; 95% CI:1.09–1.35]. CONCLUSIONS: Findings indicated unique health profiles of nursing home residents at higher risk of new onset of or continued SI.