This issue of CHP/PCOR's quarterly newsletter, which covers news from the summer 2005 quarter, includes articles about:

• our new core faculty member Grant Miller, a Harvard-trained health economist with an interest in improving health in developing countries;
• a discussion with center director Alan Garber on key issues and challenges facing the Medicare program;
• the fourth meeting of the Patient Safety Consortium, a group of more than 100 U.S. hospitals taking part in CHP/PCOR research on patient safety culture;
• core faculty member Jay Bhattacharya's research on HIV patients' perceptions of their lifespan as examined through viatical settlement transactions; and
• a research project on technology coverage decisions in the U.S. vs. the U.K., undertaken by Stirling Bryan, a U.K.-based Harkness Fellow in Health Care Policy who is spending the next academic year at CHP/PCOR.

Many stakeholders agree that the current model of U.S. health care competition is not working. Costs continue to rise at double-digit rates, and quality is far from optimal. One proposal for fixing health care markets is to eliminate provider networks and encourage informed, financially responsible consumers to choose the best provider for each condition. We argue that this "solution" will lead our health care markets toward even greater fragmentation and lack of coordination in the delivery system. Instead, we need markets that encourage integrated delivery systems, with incentives for teams of professionals to provide coordinated, efficient, evidence-based care, supported by state-of-the-art information technology.

The performance of medical groups is receiving increased attention. Relatively little conceptual or empirical work exists that examines the various dimensions of medical group performance. Using a national database of 693 medical groups, this article develops a scorecard approach to assessing group performance and presents a theory-driven framework for differentiating between high-performing versus low-performing medical groups. The clinical quality of care, financial performance, and organizational learning capability of medical groups are assessed in relation to environmental forces, resource acquisition and resource deployment factors, and a quality-centered culture. Findings support the utility of the performance scorecard approach and identification of a number of key factors differentiating high-performing from low-performing groups including, in particular, the importance of a quality-centered culture and the requirement of outside reporting from third party organizations. The findings hold a number of important implications for policy and practice, and the framework presented provides a foundation for future research.

Background: The reported accuracy of transbronchial needle aspiration (TBNA) for mediastinal staging in non- small cell lung cancer (NSCLC) varies widely. We performed a meta-analysis to estimate the accuracy of TBNA for mediastinal staging in NSCLC.

Methods: We searched Medline, Embase and the bibliographies of retrieved articles, with no language restriction, for studies evaluating TBNA accuracy. We used meta-analytic methods to construct summary receiver- operating characteristic curves and to pool sensitivity and specificity.

Results: Thirteen studies met inclusion criteria, including six studies that surgically confirmed all TBNA results and enrolled at least 10 patients with and without mediastinal metastasis (tier 1). Methodologic quality varied, but did not affect diagnostic accuracy. In tier 1 studies, the median prevalence of mediastinal metastasis was 34%. Using a random effects model, the pooled sensitivity and specificity were 39% (95% CI, 17% to 61%) and 99% (95% CI, 96% to 100%), respectively. Compared with tier 1 studies, median prevalence of mediastinal metastasis (81%; p=0.002) and pooled sensitivity (78%; 95% CI, 71% to 84%; p=0.009) were higher in non-tier 1 studies. Sensitivity analysis confirmed that the sensitivity of TBNA depends critically on the prevalence of mediastinal metastasis. The pooled major complication rate was 0.3% (95%CI, 0.01% to 4%).

Conclusions: When properly performed, TBNA is highly specific for identifying mediastinal metastasis in patients with NSCLC, but sensitivity depends critically on the study methods and patient population. In populations with a lower prevalence of mediastinal metastasis, the sensitivity of TBNA is much lower than reported in recent lung cancer guidelines.

The Patient Safety Consortium included a group of 26 diverse hospitals in or near California. In 2001 and 2002, many consortium hospitals were surveyed using the Patient Safety Climate in Healthcare Organizations (PSCHO) tool to present quantitative measures of hospital safety climate and qualitative reports on safety practices over 2 years. Investigators engaged in discussions with consortium hospitals to elicit reports about their patient safety activities. Overall quantitative measures of safety climate remained approximately the same over the 2 years, although in some specific survey areas climate appeared to improve. Hospitals reported a range and mix of patient safety activities. While considered an essential enabler of safety, cultural change takes time. Significant hospital efforts appear to be underway, and attention to a number of lessons from past patient safety efforts may benefit future undertakings.

Background: Low rates of technology utilization in hospitals with high proportions of black inpatients may be a remediable cause of healthcare disparities.

Objectives: Our objective was to determine how differences in technology utilization among hospitals contributed to racial disparity and if temporal reduction in hospital procedure rate variation resulted in decreased racial disparity for these technologies.

Methods: We identified 2,348,952 elderly Medicare beneficiaries potentially eligible for 1 of 5 emerging medical technologies from 1989-2000 and determined if these patients had received the indicated procedure within 90 days of their qualifying hospital admission. Initial multivariate regression models adjusted for age, race, sex, admission year, clinical comorbidity, community levels of education and income, and academic/urban hospital admission. The inpatient racial composition of each patient's admitting hospital and time-race interactions were added as covariates to subsequent models.

Results: Blacks had significantly lower adjusted rates (P 0.001) compared with whites for tissue replacement of the aortic valve, internal mammary artery coronary bypass grafting, dual-chambered pacemaker implantation, and lumbar spinal fusion. Hospitals with > 20% black inpatients were less likely to perform these procedures on both white and black patients than hospitals with 9% black inpatients, and racial disparity was greater in hospitals with larger black populations. There were no temporal reductions in racial disparities.

Conclusions: Blacks may be disadvantaged in access to new procedures by receiving care at hospitals that have both lower procedure rates and greater racial disparity. Policies designed to ameliorate racial disparities in health care must address hospital variation in the provision of care.

Purpose: The change in direct medical costs for schizophrenia patients who were started on olanzapine or risperidone and who were privately insured was studied.

Methods: A retrospective analysis of 1996-1999 data from the databases representing the health care experiences of individuals employed by large organizations and their dependents was performed. The sample included all individuals with a drug claim for olanzapine or risperidone, a claim with a schizophrenia diagnosis within 90 days of the drug claim, no claim for the same drug in the prior six months, and continuous health-plan enrollment for 12 months before and after the prescription.

Results: The sample included 162 patients initiated on olanzapine and 119 patients initiated on risperidone. Demographic and clinical profiles were not significantly different between groups. Annual schizophrenia-related prescription and outpatient costs increased following initiation on olanzapine or risperidone compared with the pre-initiation period. This was partially offset by a decrease in inpatient expenditures. Olanzapine initiators had higher outpatient drug expenditures than risperidone initiators in the 12 months following initiation (adjusted means, $2105 versus $1934) (p 0.05), but there was no significant difference between groups in total schizophrenia-related payments ($5251 versus $4950).

Conclusion: The total health care expenditure related to treating schizophrenia was similar between privately insured patients who were initiated on olanzapine and patients who were started on risperidone.