Bayesian Forecasting of Mortality Rates for Small Areas Using Spatiotemporal Models

Estimation and prediction of subnational mortality rates for small areas are essential planning tools for studying health inequalities. Standard methods do not perform well when data are noisy, a typical behavior of subnational datasets. Thus, reliable estimates are difficult to obtain. I present a Bayesian hierarchical model framework for prediction of mortality rates at a small or subnational level. By combining ideas from demography and epidemiology, the classical mortality modeling framework is extended to include an additional spatial component capturing regional heterogeneity. Information is pooled across neighboring regions and smoothed over time and age. To make predictions more robust and address the issue of model selection, a Bayesian version of stacking is considered using leave-future-out validation. I apply this method to forecast mortality rates for 96 regions in Bavaria, Germany, disaggregated by age and sex. Uncertainty surrounding the forecasts is provided in terms of prediction intervals. Using posterior predictive checks, I show that the models capture the essential features and are suitable to forecast the data at hand. On held-out data, my predictions outperform those of standard models lacking a regional component.

Transcriptional Profile Changes Induced by Rapamycin in Breast Cancer Maintained in Organ Culture

Background: mTOR (mammalian target of rapamycin) is a serine threonine kinase, member of the AKT/PI3K pathway, which is involved in multiple biologic functions such as transcription, translation, protein degradation and ribosome biogenesis. The activation of this protein results in phosphorilation and activation of S6K1 and 4EBP1. Rapamycin is a potent fungicide, immunossupressive and anti-cancer agent that inhibits the mTOR pathway and is an emerging cancer therapeutic. Objective: The aim of this study was to determine genes regulated by rapamycin in samples of ductal invasive breast cancer to identify novel markers of rapamycin response. Material and Methods: Inhibition of some elements of AKT pathway was assessed by immunohistochemistry in samples of breast carcinoma maintained in organ culture of fresh cut slices that preserves the interaction between epithelium and stroma before and after treatment with rapamycin. Additionally changes of the gene expression profile was analyzed by microarray in these samples subdivided in positive and negative Erb-B2. Results: Immunohistochemistry analysis indicated a significant decrease of 4EBP1 in the samples of tumor treated with rapamycin compared with the control cases. Microarray analysis revealed that few common genes were regulated in both tumor types but rapamycin affected the expression of genes mainly involved in cellular transcription and translation in both groups. To confirm the results obtained through microarray technique, RT-PCR analysis was performed in independent samples and we validated the genes WWOX, EXT1 and GTF2E2 in 60% of our cases. Over expression of the autophagy related beclin 1 indicated a contribution of autophagy to cytotoxicity of rapamycin. Conclusion: The organ culture represents a simple method to determine the effects of rapamycin and using gene profile analysis, novel genes that possible could be used as markers of mTOR inhibition were identified. Supported by FAPESP nº 04/04607-8

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6129.

Breast Cancer Gene Expression Profile in Post-Menopausal Patients Supplemented with Vitamin D

Vitamin D supplementation is indicated for post-menopausal women to prevent osteoporosis and lower 25(OH)2D3 or 1,25(OH)2D3 serum levels have been associated with breast cancer incidence or prognosis (metastasis). The antiproliferative effects of vitamin D are observed in breast cancer cell lines exposed to phamacological doses of calcitriol (1,25(OH)2D3, 100nM) but whether physiological doses are sufficient to produce growth inhibition in vivo is not known. The aim of our study was to investigate gene expression profile changes of breast cancer samples from patients supplemented with calcitriol, presenting an anti-proliferative effect on the tumor. Post-menopausal women diagnosed with breast cancer were instructed to take one (0.25ug/day, n=8) or two (0.50ug/day, n=8) tablets of calcitriol after tumor biopsy. Median time of supplementation was 30 days. Sixteen tumor samples were collected during biopsy (before supplementation) and breast surgery (after supplementation). Proliferation index was evaluated by tumor Ki-67 immunohistochemistry (IHC) expression in breast cancer samples before and after calcitirol supplementation and 1000 cells were counted by three observers (p < 0,001, Pearson's correlation). After establishing a counting error between observers, the cut-off value was fixed in 26.53% (percentile 60%) to define response. Patients were then categorized as responsive (a reduction of more than 26.53% on positive Ki-67 positive cells) and non-responsive (difference between samples before and after supplementation not exceeding 26.53%). Among our sixteen patients, all were categorized as responsive to calcitriol supplementation except for two. So far, gene expression profile of two patients (both categorized as responsive) has been analyzed using the U133 Plus 2.0 Affymetrix Gene Chips from 100ng of total RNA. To verify signalling pathways possibly involved in response to vitamin D exposure, additional five samples from a parallel study, in which breast cancer samples from post menopausal patients were collected at surgery and treated in vitro with a low concentration of calcitriol, 0.5nM (that can be attained with subcutaneous administration of doses of 8ug calcitriol, without hypercalcemia) for 24h, were included in the analysis. All samples had RNA hybridized to the same gene chips. Results were normalized and analyzed using RMA and Mev.TM4 softwares. CYP24A1, a target gene of vitamin D, presented a positive regulation after calcitriol supplementation in all samples analyzed. Differentially expressed genes were involved in the regulation of cell cycle [SMAD2, cyclin E, YWHAQ (14-3-3 family] and calcium signalling (HTR7, PTGER1 and PTGER2). Our results indicate that the tumor proliferation index is reduced upon calcitriol supplementation. Moreover, potentially regulated pathways in breast cancer specimens after administration of low doses of calcitriol are regulation of cell cycle and calcium signaling.Supported by FAPESP 2007/01111-0 – 2007/04799-2

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6128.

Why integrated health networks have failed

Over the last two decades, hospitals, physicians, and nursing homes have rushed to merge or partner with one another, cheered on by consultants, academics, and experts who claimed that such networking was imperative for these organizations to survive in an increasingly competitive environment. In the 1990s, however, reports of the major difficulties encountered by the merged entities began to dampen the industry's enthusiasm for integration, causing some to even question the value of the integration "revolution." This article will review the primary rationale behind integration and will assess its success to date. Based on field reports and the available evidence, the authors conclude that IHNS have not delivered on their promise. While each network should be considered in its unique market and contextual situation, the enormous financial, human, and clinical resources devoted to integration have not borne much fruit. Evidence of quantifiable, sustained financial or clinical value is scant.

Successful non-invasive ventilatory support in a patient with regimen-related toxicity during allogeneic bone marrow transplantation

A 13-year-old patient with transfusion-dependent beta thalassemia major developed acute regimen-related lung toxicity after the conditioning regimen but before allogeneic bone marrow transplantation. He was successfully managed on non-invasive ventilatory support. Advances in non-invasive ventilatory support may drastically improve the outlook of this subset of patients who otherwise have a grim prognosis.

Is quality in the eye of the beholder?

BACKGROUND

Relatively ignored in the literature on quality improvement (QI) in health care is the question of who defines the issues in quality of care and decides how those issues will be addressed.

METHODS

In late 1992, 669 hospital administrators, quality assurance (QA) coordinators, physicians, and nurses in 72 hospitals in six states were surveyed by telephone. Respondents were asked to identify the most serious issue related to quality of care in their hospital.

RESULTS

Of all the respondents, hospital administrators were most likely to identify quality issues related to organizational/institutional issues. QA coordinators, most of whom had nursing backgrounds, were most likely to identify organizational and patient care issues. Physician responses were distributed approximately evenly across issues related to physicians, to the organization, and to patient care. Nurses were most likely to identify issues related to patient care and patient satisfaction.

DISCUSSION

The findings validate the viewpoint that "quality is in the eye of the beholder" and that the eye is conditioned by the discipline in which hospital staff have been trained and socialized. As total quality management and continuous quality improvement approaches are implemented, and participants from many disciplines are involved in QI, a wide range of issues that have both a direct and indirect effect on patient care are likely to be addressed.

CONCLUSION

The findings of this study suggest that surveys that specifically target certain disciplinary groups may yield important information about issues related to quality of care.