Sources of selection bias in evaluating social programs: An interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation method

This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.

Crystallographic comparison of the estrogen and progesterone receptor’s ligand binding domains

The 2.8-Å crystal structure of the complex formed by estradiol and the human estrogen receptor-α ligand binding domain (hERαLBD) is described and compared with the recently reported structure of the progesterone complex of the human progesterone receptor ligand binding domain, as well as with similar structures of steroid/nuclear receptor LBDs solved elsewhere. The hormone-bound hERαLBD forms a distinctly different and probably more physiologically important dimer interface than its progesterone counterpart. A comparison of the specificity determinants of hormone binding reveals a common structural theme of mutually supported van der Waals and hydrogen-bonded interactions involving highly conserved residues. The previously suggested mechanism by which the estrogen receptor distinguishes estradiol’s unique 3-hydroxy group from the 3-keto function of most other steroids is now described in atomic detail. Mapping of mutagenesis results points to a coactivator-binding surface that includes the region around the “signature sequence” as well as helix 12, where the ligand-dependent conformation of the activation function 2 core is similar in all previously solved steroid/nuclear receptor LBDs. A peculiar crystal packing event displaces helix 12 in the hERαLBD reported here, suggesting a higher degree of dynamic variability than expected for this critical substructure.

Occupational class and cause specific mortality in middle aged men in 11 European countries: comparison of population based studies

Objectives: To compare countries in western Europe with respect to class differences in mortality from specific causes of death and to assess the contributions these causes make to class differences in total mortality.

DNA vaccine for hepatitis B: evidence for immunogenicity in chimpanzees and comparison with other vaccines.

Vaccination of two chimpanzees against hepatitis B virus (HBV) by intramuscular injection of plasmid DNA encoding the major and middle HBV envelope proteins induced group-, subtype- and preS2-specific antibodies. These were initially of IgM isotype, and then they were of IgG (predominantly IgGl) isotype. The chimpanzee injected with 2 mg of DNA attained >100 milli-international units/ml of anti-HBs antibody after one injection and 14,000 milli-international units/ml after four injections. A smaller dose (400 microg) induced lower and transient titers, but a strong anamnestic response occurred 1 year later. Comparison with responses in 23 chimpanzees receiving various antigen-based HBV vaccines suggests that the DNA approach is promising for prophylactic immunization against HBV.

Studies of group B streptococcal infection in mice deficient in complement component C3 or C4 demonstrate an essential role for complement in both innate and acquired immunity.

Group B streptococci (GBS) cause sepsis and meningitis in neonates and serious infections in adults with underlying chronic illnesses. Specific antibodies have been shown to be an important factor in protective immunity for neonates, but the role of serum complement is less well defined. To elucidate the function of the complement system in immunity to this pathogen, we have used the approach of gene targeting in embryonic stem cells to generate mice totally deficient in complement component C3. Comparison of C3-deficient mice with mice deficient in complement component C4 demonstrated that the 50% lethal dose for GBS infection was reduced by approximately 50-fold and 25-fold, respectively, compared to control mice. GBS were effectively killed in vitro by human blood leukocytes in the presence of specific antibody and C4-deficient serum but not C3-deficient serum. The defective opsonization by C3-deficient serum in vitro was corroborated by in vivo studies in which passive immunization of pregnant dams with specific antibodies conferred protection from GBS challenge to normal and C4-deficient pups but not C3-deficient pups. These results indicate that the alternative pathway is sufficient to mediate effective opsonophagocytosis and protective immunity to GBS in the presence of specific antibody. In contrast, the increased susceptibility to infection of non-immune mice deficient in either C3 or C4 implies that the classical pathway plays an essential role in host defense against GBS infection in the absence of specific immunity.