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.

Analysis and interpretation of cost data in randomised controlled trials: review of published studies

Objective To review critically the statistical methods used for health economic evaluations in randomised controlled trials where an estimate of cost is available for each patient in the study.

Spontaneous rupture of malarial spleen: two case reports and review of literature

Malaria has long been among the most common diseases in the southeast Anatolia region of Turkey. In 1992, 18676 cases were diagnosed in Turkey, and Diyarbakir city had the highest incidence (4168 cases), followed by SanliUrfa city (3578 cases). Malaria was especially common during 1994 and 1995, with 84 345 and 82 094 cases being diagnosed in these years, respectively. Spontaneous rupture of malarial spleen is rare. We saw two cases during 1998, which are reported herein. Both patients were male, and were receiving chloroquine treatment for an acute attack of malaria. One of the patients had developed abdominal pain and palpitations, followed by fainting. The other patient had abdominal pain and fever. Explorative laparotomy revealed an enlarged spleen in both patients. Splenectomy was performed in both patients. We have identified 15 episodes of spontaneous rupture of the spleen in the English language literature published since 1961. Because of increased travel to endemic areas and resistance to antimalarial drugs, malaria is a major medical problem that is becoming increasingly important to surgeons worldwide. Malaria is a particularly important problem in the southeast Anatolia region of Turkey. Prophylactic precautions should be taken by tourists who travel to this region, especially during the summer.

Structural interpretation of the mutations in the beta-cardiac myosin that have been implicated in familial hypertrophic cardiomyopathy.

In 10-30% of hypertrophic cardiomyopathy kindreds, the disease is caused by > 29 missense mutations in the cardiac beta-myosin heavy chain (MYH7) gene. The amino acid sequence similarity between chicken skeletal muscle and human beta-cardiac myosin and the three-dimensional structure of the chicken skeletal muscle myosin head have provided the opportunity to examine the structural consequences of these naturally occurring mutations in human beta-cardiac myosin. This study demonstrates that the mutations are related to distinct structural and functional domains. Twenty-four are clustered around four specific locations in the myosin head that are (i) associated with the actin binding interface, (ii) around the nucleotide binding site, (iii) adjacent to the region that connects the two reactive cysteine residues, and (iv) in close proximity to the interface of the heavy chain with the essential light chain. The remaining five mutations are in the myosin rod. The locations of these mutations provide insight into the way they impair the functioning of this molecular motor and also into the mechanism of energy transduction.