The Influence of Hospital Variability on the Management of Cancer-Associated Comp

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The quality of treatment for cancer-associated complications, those complications directly attributable to the cancer itself or that result from chemotherapy and/or radiation, have received little attention. This is particularly problematic as cancer-associated complications are associated with significant morbidity, are often accompanied by pain and discomfort, and require the expenditure of substantial healthcare resources for treatment. The management of cancer-associated complications is compounded by the availability of a myriad of treatment options, many of which are unproven and associated with significant cost. Little is known about the factors that influence allocation of care for thes complications and what factors influence outcome. The health care a patient receives and its associated outcomes are influenced by the interplay of a number of patient, physician, and hospital factors. For acute inpatient medical conditions the characteristics of the hospital in which a patient is treated play a crucial role. There is growing recognition that there is widespread between-hospital variation in how care is allocated and that these differences in practice patterns explain a large portion of treatment variability. Uninsured and minority patients are particularly vulnerable to these hospital level disparities as these patients are disproportionately more likely to receive care at facilities with limited clinical and financial resources that may adversely affect quality. In this proposal we will determine the quality of treatment for hospitalized patients admitted with a primary diagnosis of: febrile neutropenia, hypercalcemia of malignancy, esophagitis, and acute cancer-associated pain. For each cancer-associated complication we will determine patterns of treatment, the factors that underlie treatment decisions, and the effect of prompt initiation of guideline-based therapy on outcomes. For each condition we will first examine the utilization of guideline-based and non-indicated treatments. We will then comprehensively examine the influence of patient, physician, and hospital characteristics as well as between- hospital variations on the allocation of guideline-based care. We hypothesize that there is substantial variation in treatment between hospitals that is unexplained by traditional measures. Finally, we will perform novel causal inference and mediation analyses to determine the direct and indirect effects operating through a mediator of prompt guideline-based treatment on outcomes (length of stay, readmission, cost, non-routine discharge and mortality) for each complication. In addition to including a number of novel statistical methodologies, this proposal will utilize a unique database ideally suited to the study of these complications. The data from these studies will lead to immediately actionable results that can be used to guide the implementation pragmatic hospital-based interventions to improve the quality of care for patients with cancer-associated complications.