Evidence suggests that the quality and frequency of bedside clinical examination have declined. We undertook the study to (1) determine whether intensive instruction in physical examination enhances medical student skills and (2) develop a tool to evaluate those skills using a modified observed structured clinical examination (OSCE). This was a randomized, blinded, prospective, year-long study involving 3rd year students at the Albert Einstein College of Medicine. Students were randomized to receive intensive instruction in physical examination [study group (n = 46)] or usual instruction [control group (n = 75)] and evaluated by a modified OSCE. The OSCE consisted of 6 real patient stations: Head, ears, eyes, neck, throat; pulmonary; cardiovascular; gastrointestinal; neurology; musculoskeletal; and 2 computer imaging stations: genitourinary and dermatology. A faculty member present at each patient station evaluated student performance. Data were analyzed using t-tests for comparison of the mean scores between the two groups for each station and for average scores across stations. A total of 121 students were tested. The study group performed significantly better than the control group in the gastrointestinal station (p = 0.0004), the combined average score across the six real patient stations (p = 0.0001), and the combined average score across all eight stations (p = 0.0014). Intensive physical diagnosis instruction enhances physical examination skills of 3rd year medical students. The modified OSCE is a useful tool to evaluate these skills.