BACKGROUND: Because of overcrowding and cost-control efforts, emergency departments are under pressure to refer patients with nonacute conditions to other settings. However, no validated systematic methods exist for safely performing such referrals. OBJECTIVE: To determine the effects on health status and access to care of systematically referring patients with nonacute conditions to next-day primary care. DESIGN: Randomized, controlled trial. SETTING: An emergency department in a public hospital. PATIENTS: 156 adults who used the emergency department on weekdays from 7:00 a.m. to 3:00 p.m. and met criteria for deferred care. The criteria applied to three symptom complexes that account for 33% of U.S. emergency department visits by adults. INTERVENTION: Next-day care at the study site's primary care center or usual same-day care. MEASUREMENTS: Self-reported health status and use of health services during 1-week follow-up. RESULTS: Patients assigned to next-day care did not demonstrate clinically important disadvantages in health status or physician visits compared with usual care patients. In each group, more than 95% of patients were evaluated at least once by a physician, 4% sought health services after their initial evaluation, and no patients were hospitalized or died. At follow-up, both groups reported improved health status and fewer days in bed or with disability, although the deferred care group reported less improvement on all three measures. The 95% CIs were sufficiently narrow to exclude a clinically significant difference in self-reported health status. However, the possibility of 1 additional day in bed or with disability in the deferred care group could not be excluded. CONCLUSIONS: Clinically detailed standardized screening criteria can safely identify patients at public hospital emergency departments for referral to next-day care. However, larger studies are needed to assess the possibility of adverse effects.