Objectives: This retrospective analysis was conducted to derive national estimates of typical, atypical, and combination (typical-atypical) antipsychotic use and to examine factors associated with their use at adult (age ≫-18 years) ambulatory care visits by patients with mental health disorders in the United States. Methods: Data on adult visits with an International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code for a mental health disorder were extracted from the office-based National Ambulatory Medical Care Survey and the outpatient facilitybased National Hospital Ambulatory Medical Care Survey from 1996 through 2003. The visits were categorized according to whether use of a typical, atypical, or combination antipsychotic was mentioned (either prescribed, supplied, administered, ordered, or continued at the visits). Total weighted visit estimates, weighted visit percentages, and 95% CIs were calculated across the 3 types of visit groups. Bivariate analysis was performed on the association between selected characteristics and the 3 visit groups. Multivariate logistic regression was performed on factors associated with atypical versus typical antipsychotic use. Results: During the 8-year period, there were an estimated 47.7 million adult ambulatory care visits involving a mental health disorder and mention of an antipsychotic (weighted percent: 0.83%; 95% CI, 0.73-0.93). From 1996/1997 to 2002/2003, visits involving atypical and combination antipsychotics increased by 195% and 149%, respectively, and visits involving typical antipsychotics decreased by 71%. Men, blacks, and those with public insurance made more visits in which combination antipsychotics rather than typical or atypical antipsychotics were mentioned. Relative to typical or combination antipsychotic visits, more atypical antipsychotic visits involved antide-pressants (weighted percent: 61.23% atypical, 37.29% typical, and 38.32% combination). Fewer atypical antipsychotic visits compared with typical or combination antipsychotic visits involved psychotic disorders (weighted percent: 32.94%, 51.23%, and 69.93%, respectively) and medications for extrapyramidal symptoms (weighted percent: 6.69%, 29.95%, and 36.64%). In multivariate analyses controlling for sex, race, diagnosis of schizophrenia, region, diagnosis of anxiety, and recent years, atypical versus typical antipsychotic use was significantly less likely at visits by those aged 41 to 64 years compared with those aged 18 to 40 years (adjusted odds ratio [OR] = 0.63; 95% CI, 0.47-0.84; P = 0.002); significantly less likely at visits by those with public compared with private insurance (Medicare OR = 0.59 [95% CI, 0.40-0.88], P = 0.010; Medicaid OR = 0.44 [95% CI, 0.28-0.69], P < 0.001); and significantly more likely at visits associated with depression compared with those not associated with depression (OR = 1.92; 95% CI, 1.26-2.93; P = 0.003) and those associated with bipolar disorder compared with those not associated with bipolar disorder (OR = 2.10; 95% CI, 1.32-3.36; P = 0.002). Conclusions: This retrospective analysis found more atypical than typical or combination antipsychotic use at US ambulatory care visits by adults with mental health disorders other than schizophrenia or psychoses in the period studied. Atypical versus typical antipsychotic use was significantly less likely at visits by adults aged 41 to 64 years and those with public insurance, but significantly more likely at visits by those with depression or bipolar disorder.