Do Medicare HMOs cost shift?

We tested the hypothesis that health maintenance organizations (HMOs) increase their commercial premiums when Medicare pays less. Such a linkage would be taken as evidence of "cost shifting." Other studies have tested the cost-shifting hypothesis among health care providers, but this is the first to examine the HMO industry. Our data consisted of annual observations on all HMOs that operated in the United States between 1990 and 1995 and had a Medicare risk contract. A comparison group of HMOs that had no Medicare contract during that period also was analyzed. The main finding from this study is that HMOs have not shifted costs from Medicare to commercial premiums. This results supports the skeptical consensus that is developing toward the cost-shifting hypothesis. Additional findings include the negative effects of competition and for-profit status on HMOs' commercial premiums.     

Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women

CONTEXT: Raloxifene is a selective estrogen receptor modulator that has estrogen-agonistic effects on bone and estrogen-antagonistic effects on breast and uterus. OBJECTIVE: To identify the effects of raloxifene on markers of cardiovascular risk in postmenopausal women, and to compare them with those induced by hormone replacement therapy (HRT). DESIGN: Double-blind, randomized, parallel trial. SETTING: Eight sites in the United States. PARTICIPANTS: 390 healthy postmenopausal women recruited by advertisement. INTERVENTION: Participants were randomized to receive 1 of 4 treatments: raloxifene, 60 mg/d; raloxifene, 120 mg/d; HRT (conjugated equine estrogen, 0.625 mg/d, and medroxyprogesterone acetate, 2.5 mg/d); or placebo. MAIN OUTCOME MEASURES: Change and percent change from baseline of lipid levels and coagulation parameters after 3 months and 6 months of treatment. RESULTS: At the last visit completed, compared with placebo, both dosages of raloxifene significantly lowered low-density lipoprotein cholesterol (LDL-C) by 12% (P < .001), similar to the 14% reduction with HRT (P < .001). Both dosages of raloxifene significantly lowered lipoprotein(a) by 7% to 8% (P < .001), less than the 19% decrease with HRT (P<.001). Raloxifene increased high-density lipoprotein-2 cholesterol (HDL2-C) by 15% to 17% (P < .05), less than the 33% increase with HRT (P < .001). Raloxifene did not significantly change high-density lipoprotein cholesterol (HDL-C), triglycerides, or plasminogen activator inhibitor-1 (PAI-1); whereas HRT increased HDL-C by 11% and triglycerides by 20%, and decreased PAI-1 by 29% (for all, P < .001). Raloxifene significantly lowered fibrinogen by 12% to 14% (P < .001), unlike HRT, which had no effect. Neither treatment changed fibrinopeptide A or prothrombin fragment 1 and 2. CONCLUSIONS: Raloxifene favorably alters biochemical markers of cardiovascular risk by decreasing LDL-C, fibrinogen, and lipoprotein(a), and by increasing HDL2-C without raising triglycerides. In contrast to HRT, raloxifene had no effect on HDL-C and PAI-1, and a lesser effect on HDL2-C and lipoprotein(a). Further clinical trials are necessary to determine whether these favorable biochemical effects are associated with protection against cardiovascular disease.     

Insulin-like growth factor-binding protein-5 is cleaved by physiological concentrations of thrombin

Insulin-like growth factor (IGF)-binding protein-5 (IGFBP-5) is cleaved by a serine protease that is secreted by fibroblasts and porcine smooth muscle cells (pSMC) in culture. To investigate whether other serine proteases could cleave this substrate at physiologically relevant concentrations, we determined the proteolytic effects of thrombin on IGFBP-5. Human alpha-thrombin (0.0008 NIH U/ml) cleaved IGFBP-5 into 24-, 23-, and 20-kDa non-IGF-I-binding fragments. Cleavage occurred at a physiologically relevant thrombin concentration. The effect was specific for IGFBP-5, as other forms of IGFBPs, e.g. IGFBP-1, IGFBP-2, and IGFBP-4 were not cleaved by thrombin. Although IGFBP-3 was cleaved by thrombin, this effect required a 50-fold greater thrombin concentration. [35S]Methionine labeling followed by immunoprecipitation confirmed that IGFBP-5 that was constitutively synthesized by pSMC cultures was also degraded by thrombin into 24-, 23-, and 20-kDa fragments. The binding of IGF-I to IGFBP-5 partially inhibited IGFBP-5 degradation by thrombin, and an IGF analog that does not bind to IGFBP-5 had no effect. Thrombin did not account for the serine protease activity that had been shown previously to be present in pSMC-conditioned medium. This was proven by showing that 1) no immunoreactive thrombin could be detected in the pSMC-conditioned medium; 2) the IGFBP-5 fragments that were generated by thrombin showed three cleavage sites (Arg192-Ala193, Arg156-Ile157, and Lys120-His121), whereas the serine protease in conditioned medium cleaves IGFBP-5 at a different site; and 3) hirudin had no effect on IGFBP-5 cleavage by the protease in pSMC medium; however, it inhibited IGFBP-5 degradation by thrombin. To determine the physiological significance of IGFBP-5 cleavage, the effect of an IGFBP-5 mutant that is resistant to cleavage by the pSMC protease and has been shown to inhibit IGF-I actions in pSMC was determined. This mutant inhibited IGF-I-stimulated DNA synthesis, but if thrombin was added simultaneously, IGF-I was fully active. In summary, physiological concentrations of thrombin degrade IGFBP-5. Degradation can be blocked by hirudin and is partially inhibited by IGF-I binding. Generation of active thrombin in vessel walls may be a physiologically relevant mechanism for controlling IGF-I bioactivity.     

Recombinant hepatitis A virus antigen: improved production and utility in diagnostic immunoassays

Hepatitis A virus (HAV) immunoassays use cell culture-derived HAV antigen to detect HAV-specific antibodies. The current method of production of HAV antigen in tissue culture is time-consuming and expensive. We previously expressed the HAV open reading frame in recombinant vaccinia viruses (rV-ORF). The recombinant HAV polyprotein was accurately processed and was assembled into subviral particles. These particles were bound by HAV-neutralizing antibodies and were able to elicit antibodies which were detected by commercial immunoassays. The present investigation compared the production of HAV antigen by standard tissue culture methods to the production of HAV antigen with the recombinant vaccinia virus system. In addition, HAV and rV-ORF antigens were assessed for their utility in diagnostic immunoassays. Serum or plasma samples from HAV antibody-positive and antibody-negative individuals were evaluated by immunoassay that used either HAV or rV-ORF antigen. All samples (86 of 86) in which HAV antibody was detected by a commercial enzyme-linked immunosorbent assay (ELISA) also tested positive by the recombinant antigen-based immunoassay (VacRIA). Similarly, all samples (50 of 50) that were HAV antibody negative also tested negative by the VacRIA. The lower limit of detection of HAV antibody was similar among immunoassays with either HAV or rV-ORF antigen. Thus, in the population studied, the sensitivity and specificity of the VacRIA were equivalent to those of the commercial ELISA. Since production of recombinant antigen is faster and less expensive than production of traditional HAV antigen, the development of diagnostic HAV antibody tests with recombinant HAV antigen appears warranted.     

A program of quality improvement in transfusion safety. Experience at Saint-Camille Hospital

Because of concerns about competence and voluntariness, the mentally disordered constitute a vulnerable population in the context of nontherapeutic biomedical research and, as such, are in need of protection. Despite others' concern about protecting the mentally disordered, their decision-making potential should also be respected and maximized, allowing such individuals to consent to participate in experiments subject to an evaluation of their competence to make such a decision. Competent mentally disordered persons who anticipate future incapacity should be able to issue research directives or durable powers of attorney whereby they can provide explicit consent to participate in nontherapeutic research. When he or she becomes incompetent, a substitute decision maker should be able to provide consent on behalf of the mentally disordered person within established parameters. Nontherapeutic experimentation with the mentally disordered should be permitted, but only within the boundaries of ethical permissibility delineated by legislated guidelines. At present, the legal status of substituted consent for nontherapeutic procedures is uncertain and requires legislation, which in addition to legalizing such consent, would provide guidelines for substitute decision makers and for the creation of research directives. These guidelines should include restrictions on the scope of research, obligations of researchers, rights of subjects, and responsibilities of research ethics committees (RECs). In all cases, the voluntary and informed consent of the person or substitute decision maker must be obtained.     

Determinants of the pulmonary artery pressure rise in left ventricular dysfunction

Pulmonary artery hypertension in patients with left ventricular dysfunction is related to poor outcome but the role of cardiac functional abnormalities in the genesis of pulmonary hypertension remains unknown. The aim of this prospective study was to identify the determinants of pulmonary hypertension in 102 consecutive patients with primary left ventricular dysfunction (ejection fraction < 50%). Systolic pulmonary artery pressure was measured by continuous wave Doppler. Left ventricular systolic and diastolic function, severity of functional mitral regurgitation, cardiac output, and left atrial volume were assessed using Doppler echocardiography. In patients with left ventricular dysfunction, systolic pulmonary artery pressure was increased (51 +/- 14 mmHg, range 23 to 87 mmHg). Mitral deceleration time (r = -0.61; p = 0.0001) and mitral effective regurgitant orifice (r = 0.50; p = 0.0001) were the strongest parameters related to systolic pulmonary artery pressure. Multivariate analysis identified these two variables as the strongest predictors of systolic pulmonary artery pressure in association with the mitral E/A ratio (p = 0.006) and age (p = 0.005). In conclusion, pulmonary hypertension is common and variable in patients with left ventricular dysfunction. It is closely related to diastolic dysfunction and severity of functional mitral regurgitation but not independently to the degree of left ventricular systolic dysfunction. These findings underline the importance of assessing diastolic function and quantifying mitral regurgitation in patients with left ventricular dysfunction.     

Induction of MHC class I-restricted CTL response by DNA immunization with ubiquitin-influenza virus nucleoprotein fusion antigens

DNA vaccines have been shown to be an effective means of inducing cytotoxic T-lymphocyte (CTL) responses in both young and aged mice. Better understanding of the pathways by which antigens encoded by DNA vaccines are processed and presented to CTL may allow for improvements in CTL responses in older animals. Since CTL recognize short peptides presented by MHC class I molecules, and since ubiquitin-dependent proteolysis is widely believed to be responsible for degradation of endogenously synthesized antigens and generation of these peptide ligands, we sought to use ubiquitin (Ub) conjugation to target influenza virus nucleoprotein (NP) antigen into the Ub-proteasome degradation pathway for MHC class I-restricted antigen processing and presentation. However, the addition of the Ub moiety did not affect the half-life of Ub-NP protein in transiently transfected human rhabdomyosarcoma (RD) cells. Moreover, the modifications of NP DNA vaccine with Ub conjugation did not affect their ability to induce a CTL response specific for the H-2Kd-restricted NP147-155 epitope, as assessed by both percent cytolysis in bulk CTL culture and by CTL precursor (CTLp) frequency in limiting dilution analysis (LDA). In contrast, the anti-NP antibody (Ab) responses were dramatically reduced in mice immunized with low doses (1 microgram) of Ub-NP constructs, compared with mice immunized with wild-type NP DNA. These results demonstrate that Ub conjugation alone does not guarantee targeting of endogenously synthesized antigens for rapid degradation by proteasomes. Furthermore, the ability of ubiquintination to reduce Ab responses to NP without affecting CTL responses suggests that the Ub modifications result in a lower availability of full-length NP from transfected cells in vivo. The implications of these data on antigen presentation and cross-priming are discussed.