A new model of workforce projections, based on physician supply and utilization, predicts an impending physician shortage, which the nation cannot afford to ignore. The hypothesis put forward by Cooper and colleagues comes under serious challenge in a set of invited perspectives that follow. We thought that it was important to gather a variety of views because the debate over health care workforce issues has been long neglected and deserves closer scrutiny.

A B S T R A C T Platelets from individuals with familial hypercholesterolemia show increased sensitivity to the aggregating agents, epinephrine and ADP. Since the mechanism of this abnormal sensitivity is unknown, we examined, in vitro, the influence of the plasma lipid environment on the function of platelets. The composition of plasma lipids was altered by the addition of sonicated cholesterol-dipalmitoyl lecithin liposomes which were "cholesterol normal" (cholesterol-phospholipid mole ratio [C/P] = 1.0), "cholesterol rich" (C/P = 2.2), or "cholesterol poor" (C/P = 0). Cholesterolnormal liposomes had no influence on platelet lipids or platelet function. In contrast, after incubation for 5 h at 370C with cholesterol-rich liposomes, normal platelets acquired 39.2% excess cholesterol with no change in phospholipids or protein. The percent increase in platelet membrane cholesterol was threefold that of the granule fraction. The acquisition of cholesterol by platelets was associated with a 35-fold increase in sensitivity to epinephrine-induced aggregation (P < 0.001) and 15-fold increase to ADP aggregation (P <0.01), as determined both by aggregometry and by [14C]serotonin release. Response to thrombin or collagen was unchanged. Platelets incubated with cholesterol-poor liposomes underwent a selective loss of 21.4% cholesterol and this was associated with an 18-fold reduction in their sensitivity to epinephrine. (5,6). These studies have demonstrated that platelets from individuals with type II hyperlipoproteinemia have an increase in both platelet factor 3 availability and phospholipid content (5) and an increased sensitivity to aggregating agents, particularly epinephrine (6). However, it is not clear whether these findings are due to an intrinsic platelet defect or are related to the platelet's lipid-rich plasma environment. Furthermore, the relevance of these observations to the in vivo situation is unknown.The abnormal lipid composition of lipoproteins in certain individuals with alcoholic cirrhosis profoundly influences erythrocyte membrane lipid composition and membrane function leading to the disorder, spur cell anemia (7,8 METHODSPlatelet preparation. All blood donors were fasting for 12 h, had abstained from medications for at least 2 wk before blood donation, and had normal serum lipoproteins according to standard criteria (10), except for one who conformed to type Ha. Venous blood was collected through siliconized needles into plastic syringes and anticoagulated by mixing 9 vol of blood with 1 vol citrate-phosphate-dextrose (trisodium citrate 0.0894 M, citric acid 0.0156 M, monobasic sodium phosphate 0.0161 M, dextrose 0.1418 M). All blood processing was carried out in plastic-ware at room temperature. Platelet-rich plasma was obtained by centrifugation of samples for 10 min at 100 g. The remaining blood was centrifuged for 15 min at 1,800 g to obtain platelet-poor plasma which contained less than 20,000 platelets per A1. Platelets were counted (11) and their volume distributions measured using a Coul...

Cholesterol and phospholipid are the two major lipids of the red cell membrane. Cholesterol is insoluble in water but is solubilized by phospholipids both in membranes and in plasma lipoproteins. Morever, cholesterol exchanges between membranes and lipoproteins. An equilibrium partition is established based on the amount of cholesterol relative to phospholipid (C/PL) in these two compartments. Increases in the C/PL of red cell membranes have been studied under three conditions: First, spontaneous increases in vivo have been observed in the spur red cells of patients with severe liver disease; second, similar red cell changes in vivo have been induced by the administration of cholesterol-enriched diets to rodents and dogs; third, increases in membrane cholesterol have been induced in vitro by enriching the C/PL of the lipoprotein environment with cholesterol-phospholipid dispersions (liposomes) having a C/PL of greater than 1.0. In each case, there is a close relationship between the C/PL of the plasma environment and the C/PL of the red cell membrane. In vivo, the C/PL mole ratio of red cell membranes ranges from a normal value of 0.09--1.0 to values which approach but do not reach 2.0. In vitro, this ratio approaches 3.0. Cholesterol enrichment of red cell membranes directly influences membrane lipid fluidity, as assessed by the rotational diffusion of hydrophobic fluorescent probes such as diphenyl hexatriene (DPH). A close correlation exists between increases in red cell membrane C/PL and decreases in membrane fluidity over the range of membrane C/PL from 1.0 to 2.0; however, little further change in fluidity occurs when membrane C/PL is increased to 2.0--3.0. Cholesterol enrichment of red cell membranes is associated with the transformation of cell contour to one which is redundant and folded, and this is associated with a decrease in red cell filterability in vitro. Circulation in vivo in the presence of the spleen further modifies cell shape to a spiny, irregular (spur) form, and the survival of cholesterol-rich red cells is decreased in the presence of the spleen. Although active Na-K transport is not influenced by cholesterol enrichment of human red cells, several carrier-mediated transport pathways are inhibited. We have demonstrated this effect for the cotransport of Na + K and similar results have been obtained by others in studies of organic acid transport and the transport of small neutral molecules such as erythritol and glycerol. Thus, red cell membrane C/PL is sensitive to the C/PL of the plasma environment. Increasing membrane C/PL causes a decrease in membrane fluidity, and these changes are associated with a reduction in membrane permeability, a distortion of cell contour and filterability and a shortening of the survival of red cells in vivo.

A B S T R A C T Cholesterol-rich membranes are the hallmark of "spur" red cells. Spur cells accumulate cholesterol from cholesterol-rich serum lipoproteins. Previous studies suggested that this added cholesterol is responsible for both the altered morphology and the destruction of spur cells. To examine this process in the absence of other serum factors, cholesterol-lecithin dispersions with varying amounts of unesterified cholesterol (C) relative to phospholipid (P) were prepared, and their influence on normal human red cells was studied.Cholesterol-rich lipid dispersions (C/P mole ratio > 1.0) transferred cholesterol to both red cell membranes and serum lipoproteins, and cholesterol-poor dispersions (C/P mole ratio < 1.0) depleted red cells of cholesterol. Changes in membrane cholesterol paralleled changes in membrane surface area, as calculated from osmotic fragility, with a 0.22% variation in surface area per 1.0% variation in cholesterol content. Cold-induced compression of membrane surface area was increased in cholesterol-poor red cells (C/P = 0.4), whereas the surface area of cholesterol-rich membranes (C/P = 1.80) underwent no compression. Although the Na and K permeability of red cells severely depleted of cholesterol was increased, lesser degrees of depletion had no effect, and the permeability of cholesterol-rich cells was normal. However, increasing membrane cholesterol caused a progressive decrease in red cell deformability, as measured by filtration.Cholesterol-poor red cells were spherocytic in appearance and cholesterol-rich cells were broad and flat, indicative of their surface areas. In addition, cholesterol-

A B S T R A C T The influxes of Na+ and K+ into the human red cell appear to be interrelated. This relationship was investigated under conditions in which either Na4 or K+ concentration outside the cell was varied or one cation was replaced by Mg2", choline+, or Li'. The effects of furosemide on Na+ and K+ movements were studied in the presence of ouabain.When ouabain was present, Na4 influx was higher with K4 ions externally than with other cations externally. Furosemide inhibited this K+-stimulated Na4 influx, but it had little effect when K+ was absent. Ouabain-insensitive K4 influx was stimulated two-fold by external Nae compared with other cations. Furosemide also inhibited this stimulation, but it had little effect when Mg`4 or choline4 replaced external Nae. Thus it was confirmed that synergism exists between the ouabain-insensitive influxes of Na4 and K4 and it was demostrated that furosemide inhibits this cooperative effect. The ouabaininsensitive influx of both K+ and Na4 showed a hyperbolic "saturating" dependence on the external concentration of the transported cation. Furosemide therefore eliminates a saturable component of influx of each cation.The net uptake of Na+ in the presence of ouabain was stimulated by K4 ions. A similar effect was observed with red cells, in which Li4 replaced nearly all the internal Nae plus K4 ions. In these cells, net Nae uptake was stimulated by external K+, and net K+ uptake was stimulated by external Na+. Furosemide inhibited this mutual stimulation of net cation entries.The inhibitory action of furosemide was not limited to inward flux and net movement of Na4 and K+. Furose-This work has appeared in preliminary form (Clin. Res.