Domestic production vs international immigration: Options for the US physician workforce

OBJECTIVE: To determine alternate combinations for reductions in US medical school graduates (USMGs), international medical graduate (IMG) immigration, and graduate medical education (GME) residencies, based on future physician supply targets. DESIGN: A demographic projection model of the physician supply was constructed and calibrated to fit observed American Medical Association Physician Masterfile data and current supply forecasts. Total annual input to GME was backcast from given future supply targets, adjusting for the portion of IMGs in GME who do not enter the US workforce. MAIN OUTCOME MEASURES: The annual number of new physicians added to supply from domestic or international sources needed to reach future physician-to-population ratio targets. RESULTS: Because of the low rate of attrition from the physician supply, it takes up to 50 years for workforce policy to effectively stabilize the physician-to-population ratio at a target level. All target ratios considered here would require immediate reductions in the total number of GME positions. These reductions must be followed by gradual annual increases to account for population growth. The size of USMG and IMG reductions are interrelated and depend critically on the percentage of IMG trainees who remain to practice in the United States. CONCLUSIONS: Reductions in future physician supply can come from either the IMG or USMG component of physician production, or both. The model developed here allows the estimation of multiple combinations of both GME components.     

Deep venous thrombosis: incidence on admission to a brain injury rehabilitation program

OBJECTIVE: To determine the incidence of deep venous thrombosis (DVT) in brain injured individuals at time of admission to a brain injury (BI) rehabilitation program. DESIGN: Prospective study, sequential case series. SETTING: University tertiary care BI rehabilitation center. DATA SET: Eighty-two traumatic brain injury (TBI) and 71 atraumatic brain injury (ABI) patients were consecutively admitted to our BI unit over a 12-month period and screened within 24 hours of admission for a lower extremity DVT with color flow duplex Doppler ultrasonography. All patients had been prophylaxed with either subcutaneous heparin anticoagulation therapy or intermittent compression devices, and all patients were within 2 months of the original BI. MAIN OUTCOME MEASURES: Evidence of intrinsic venous occlusion by duplex Doppler. RESULTS: DVTs were detected and treated prior to rehabilitation admission in three patients (2%), and these persisted at rehabilitation admission. New DVTs were detected at time of rehabilitation admission in 17 patients (11%). All were occult DVTs; none of the 17 patients had clinical findings indicative of acute DVT. No significant differences were noted in the TBI group when age, highest 24-hour Glasgow Coma Scale score, length of acute hospitalization, type of DVT prophylaxis, or presence of an extremity fracture were compared for individuals with and without DVT. No significant differences were noted in the ABI group when age, length of acute hospitalization, and type of DVT prophylaxis were compared for individuals with and without DVT. CONCLUSION: The overall incidence of DVTs was 13% and the incidence of occult DVT was 11%. Individuals with TBI had an overall incidence of DVTs of 20% and an occult DVT incidence of 18%. Individuals with ABI had an overall incidence of DVT's of 6% and an occult DVT incidence of 4%. These findings indicate the importance of baseline screening for DVT in this patient population.     

Distribution of antigen specific memory T cells in lymph nodes after immunization at peripheral or mucosal sites

The distribution of antigen-specific memory T cells in different lymph nodes of sheep was determined using an antigen-specific in vitro proliferation assay. Lymph nodes were collected from sheep immunized simultaneously with avidin or ovalbumin in a peripheral tissue site (hind leg muscle) and keyhole limpet haemocyanin (KLH) in an intestinal tissue site (gut wall or colonic mucosa). The results showed a consistently high proliferative response in typical peripheral lymph nodes (popliteal and prescapular) and a low or negative response in gastrointestinal lymph nodes (abomasal and jejunal) while the response in other nodes was variable. The low proliferative response in the gastrointestinal lymph nodes was not due to the presence of suppressor CD8- lymphocytes and the proliferative response could not be raised to peripheral lymph nodes levels with the addition to cultures of IL-2 or mitomycin-C treated peripheral lymph node cells. The high proliferative response in the peripheral lymph nodes was not suppressed by the addition of mitomycin-C-treated gastric lymph node cells but was dramatically reduced by the addition of mAb against the IL-2-receptor or by depletion of CD4- T cells. The results suggest that antigen-specific proliferative memory T cells, which may be Th1-like memory cells, preferentially migrate to peripheral lymph nodes independent of their site of induction.