Days spent in acute care hospitals at the end of life of cancer patients in four Swiss cantons: a retrospective database study (SAKK 89/09).

Number of days spent in acute hospitals (DAH) at the end of life is regarded as an important care quality indicator for cancer patients. We analysed DAH during 90 days prior to death in patients from four Swiss cantons. Claims data from an insurance provider with about 20% market share and patient record review identified 2086 patients as dying of cancer. We calculated total DAH per patient. Multivariable generalised linear modelling served to evaluate potential explanatory variables. Mean DAH was 26 days. In the multivariable model, using complementary and alternative medicine (DAH = 33.9; +8.8 days compared to non-users) and canton of residence (for patient receiving anti-cancer therapy, Zürich DAH = 22.8 versus Basel DAH = 31.4; for other patients, Valais DAH = 22.7 versus Ticino DAH = 33.7) had the strongest influence. Age at death and days spent in other institutions were additional significant predictors. DAH during the last 90 days of life of cancer patients from four Swiss cantons is high compared to most other countries. Several factors influence DAH. Resulting differences are likely to have financial impact, as DAH is a major cost driver for end-of-life care. Whether they are supply- or demand-driven and whether patients would prefer fewer days in hospital remains to be established.

Financial impact of the Medicare prospective payment system on long term acute care hospitals

Objective. Long Term Acute Care Hospitals (LTACs) are subject to Medicare rules because they accept Medicare and Medicaid patients. In October 2002, Medicare changed the LTAC reimbursement formulas, from a cost basis system to a Prospective Payment System (PPS). This study examines whether the PPS has negatively affected the financial performance of the LTAC hospitals in the period following the reimbursement change (2003-2006), as compared to the period prior to the change (1999-2003), and if so, to what extent. This study will also examine whether the PPS has resulted in a decreased average patient length of stay (LOS) in the LTAC hospitals for the period of 2003-2006 as compared to the prior period of 1999-2003, and if so, to what extent. ^ Methods. The study group consists of two large LTAC hospital systems, Kindred Healthcare Inc. and Select Specialty Hospitals of Select Medical Corporation. Financial data and operational indicators were reviewed, tabulated and dichotomized into two groups, covering the two periods: 1999-2002 and 2003-2006. The financial data included net annual revenues, net income, revenue per patient per day and profit margins. It was hypothesized that the profit margins for the LTAC hospitals were reduced because of the new PPS. Operational indicators, such as annual admissions, annual patient days, and average LOS were analyzed. It was hypothesized that LOS for the LTAC hospitals would have decreased. Case mix index, defined as the weighted average of patients’ DRGs for each hospital system, was not available to cast more light on the direction of LOS. ^ Results. This assessment found that the negative financial impacts did not materialize; instead, financial performance improved during the PPS period (2003-2006). The income margin percentage under the PPS increased for Kindred by 24%, and for Select by 77%. Thus, the study’s working hypothesis of reduced income margins for the LTACs under the PPS was contradicted. As to the average patient length of stay, LOS decreased from 34.7 days to 29.4 days for Kindred, and from 30.5 days to 25.3 days for Select. Thus, on the issue of LTAC shorter length of stay, the study’s working hypothesis was confirmed. ^ Conclusion. Overall, there was no negative financial effect on the LTAC hospitals during the period of 2003-2006 following Medicare implementation of the PPS in October 2002. On the contrary, the income margins improved significantly. ^ During the same period, LOS decreased following the implementation of the PPS. This was consistent with the LTAC hospitals’ pursuit of financial incentives.^

Financial impact of the Medicare prospective payment system on long term acute care hospitals

Objective. Long Term Acute Care Hospitals (LTACs) are subject to Medicare rules because they accept Medicare and Medicaid patients. In October 2002, Medicare changed the LTAC reimbursement formulas, from a cost basis system to a Prospective Payment System (PPS). This study examines whether the PPS has negatively affected the financial performance of the LTAC hospitals in the period following the reimbursement change (2003–2006), as compared to the period prior to the change (1999–2003), and if so, to what extent. This study will also examine whether the PPS has resulted in a decreased average patient length of stay (LOS) in the LTAC hospitals for the period of 2003–2006 as compared to the prior period of 1999-2003, and if so, to what extent. ^ Methods. The study group consists of two large LTAC hospital systems, Kindred Healthcare Inc. and Select Specialty Hospitals of Select Medical Corporation. Financial data and operational indicators were reviewed, tabulated and dichotomized into two groups, covering the two periods: 1999–2002 and 2003–2006. The financial data included net annual revenues, net income, revenue per patient per day and profit margins. It was hypothesized that the profit margins for the LTAC hospitals were reduced because of the new PPS. Operational indicators, such as annual admissions, annual patient days, and average LOS were analyzed. It was hypothesized that LOS for the LTAC hospitals would have decreased. Case mix index, defined as the weighted average of patients’ DRGs for each hospital system, was not available to cast more light on the direction of LOS. ^ Results. This assessment found that the negative financial impacts did not materialize; instead, financial performance improved during the PPS period (2003–2006). The income margin percentage under the PPS increased for Kindred by 24%, and for Select by 77%. Thus, the study’s working hypothesis of reduced income margins for the LTACs under the PPS was contradicted. As to the average patient length of stay, LOS decreased from 34.7 days to 29.4 days for Kindred, and from 30.5 days to 25.3 days for Select. Thus, on the issue of LTAC shorter length of stay, the study’s working hypothesis was confirmed. ^ Conclusion. Overall, there was no negative financial effect on the LTAC hospitals during the period of 2003–2006 following Medicare implementation of the PPS in October 2002. On the contrary, the income margins improved significantly. ^ During the same period, LOS decreased following the implementation of the PPS. This was consistent with the LTAC hospitals’ pursuit of financial incentives. ^

The impact of Medicare, Medicaid and Schip Extension Act of 2007 (MMSEA) on the long term acute care hospitals

The Long Term Acute Care Hospitals (LTACH), which serve medically complex patients, have grown tremendously in recent years, by expanding the number of Medicare patient admissions and thus increasing Medicare expenditures (Stark 2004). In an attempt to mitigate the rapid growth of the LTACHs and reduce related Medicare expenditures, Congress enacted Section 114 of P.L. 110-173 (§114) of the Medicare, Medicaid and SCHIP Extension Act (MMSEA) in December 29, 2007 to regulate the LTCAHs industry. MMSEA increased the medical necessity reviews for Medicare admissions, imposed a moratorium on new LTCAHs, and allowed the Centers for Medicare and Medicaid Services (CMS) to recoup Medicare overpayments for unnecessary admissions. ^ This study examines whether MMSEA impacted LTACH admissions, operating margins and efficiency. These objectives were analyzed by comparing LTACH data for 2008 (post MMSEA) and data for 2006-2007 (pre-MMSEA). Secondary data were utilized from the American Hospital Association (AHA) database and the American Hospital Directory (AHD).^ This is a longitudinal retrospective study with a total sample of 55 LTACHs, selected from 396 LTACHs facilities that were fully operational during the study period of 2006-2008. The results of the research found no statistically significant change in total Medicare admissions; instead there was a small but not statistically significant reduction of 5% in Medicare admissions for 2008 in comparison to those for 2006. A statistically significant decrease in mean operating margins was confirmed between the years 2006 and 2008. The LTACHs' Technical Efficiency (TE), as computed by Data Envelopment Analysis (DEA), showed significant decrease in efficiency over the same period. Thirteen of the 55 LTACHs in the sample (24%) in 2006 were calculated as “efficient” utilizing the DEA analysis. This dropped to 13% (7/55) in 2008. Longitudinally, the decrease in efficiency using the DEA extension technique (Malmquist Index or MI) indicated a deterioration of 10% in efficiency over the same period. Interestingly, however, when the sample was stratified into high efficient versus low efficient subgroups (approximately 25% in each group), a comparison of the MIs suggested a significant improvement in Efficiency Change (EC) for the least efficient (MI 0.92022) and reduction in efficiency for the most efficient LTACHs (MI = 1.38761) over same period. While a reduction in efficiency for the most efficient is unexpected, it is not particularly surprising, since efficiency measure can vary over time. An improvement in efficiency, however, for the least efficient should be expected as those LTACHs begin to manage expenses (and controllable resources) more carefully to offset the payment/reimbursement pressures on their margins from MMSEA.^

Differences in performance for quality of care indicators for acute myocardial infarction (AMI) across various hospital categories in Texas

The study analyzed Hospital Compare data for Medicare Fee-for-service patients at least 65 years of age to determine whether hospital performance for AMI outcome and processes of care measures differ amongst Texas hospitals with respect to ownership status (for profit vs. not-for-profit), academic status (teaching vs. non-teaching) and geographical setting (rural vs. urban). ^ The study found a statistically significant difference between for-profit and not-for-profit hospitals in four process-of-care measures (aspirin at discharge, P=0.028; ACE or ARB inhibitor for LSVD, P=0.048; Smoking cessation advice: P=0.034; outpatients who got aspirin with 24 hours of arrival in the ED, P=0.044). No significant difference in performance was found between COTH-member teaching and non-teaching hospitals for any of the eight process-of-care measures or the two outcome measures for AMI. The study was unable to compare performance based on geographic setting of hospitals due to lack of sufficient data for rural hospitals. ^ The results of the study suggest that for-profit Texas hospitals might be slightly better than not-for –profit hospitals at providing possible heart attack patients with certain processes of care.^

TIME SERIES ANALYSIS AS INPUT FOR PREDICTIVE MODELING: PREDICTING CARDIAC ARREST IN A PEDIATRIC INTENSIVE CARE UNIT

The first manuscript, entitled "Time-Series Analysis as Input for Clinical Predictive Modeling: Modeling Cardiac Arrest in a Pediatric ICU" lays out the theoretical background for the project. There are several core concepts presented in this paper. First, traditional multivariate models (where each variable is represented by only one value) provide single point-in-time snapshots of patient status: they are incapable of characterizing deterioration. Since deterioration is consistently identified as a precursor to cardiac arrests, we maintain that the traditional multivariate paradigm is insufficient for predicting arrests. We identify time series analysis as a method capable of characterizing deterioration in an objective, mathematical fashion, and describe how to build a general foundation for predictive modeling using time series analysis results as latent variables. Building a solid foundation for any given modeling task involves addressing a number of issues during the design phase. These include selecting the proper candidate features on which to base the model, and selecting the most appropriate tool to measure them. We also identified several unique design issues that are introduced when time series data elements are added to the set of candidate features. One such issue is in defining the duration and resolution of time series elements required to sufficiently characterize the time series phenomena being considered as candidate features for the predictive model. Once the duration and resolution are established, there must also be explicit mathematical or statistical operations that produce the time series analysis result to be used as a latent candidate feature. In synthesizing the comprehensive framework for building a predictive model based on time series data elements, we identified at least four classes of data that can be used in the model design. The first two classes are shared with traditional multivariate models: multivariate data and clinical latent features. Multivariate data is represented by the standard one value per variable paradigm and is widely employed in a host of clinical models and tools. These are often represented by a number present in a given cell of a table. Clinical latent features derived, rather than directly measured, data elements that more accurately represent a particular clinical phenomenon than any of the directly measured data elements in isolation. The second two classes are unique to the time series data elements. The first of these is the raw data elements. These are represented by multiple values per variable, and constitute the measured observations that are typically available to end users when they review time series data. These are often represented as dots on a graph. The final class of data results from performing time series analysis. This class of data represents the fundamental concept on which our hypothesis is based. The specific statistical or mathematical operations are up to the modeler to determine, but we generally recommend that a variety of analyses be performed in order to maximize the likelihood that a representation of the time series data elements is produced that is able to distinguish between two or more classes of outcomes. The second manuscript, entitled "Building Clinical Prediction Models Using Time Series Data: Modeling Cardiac Arrest in a Pediatric ICU" provides a detailed description, start to finish, of the methods required to prepare the data, build, and validate a predictive model that uses the time series data elements determined in the first paper. One of the fundamental tenets of the second paper is that manual implementations of time series based models are unfeasible due to the relatively large number of data elements and the complexity of preprocessing that must occur before data can be presented to the model. Each of the seventeen steps is analyzed from the perspective of how it may be automated, when necessary. We identify the general objectives and available strategies of each of the steps, and we present our rationale for choosing a specific strategy for each step in the case of predicting cardiac arrest in a pediatric intensive care unit. Another issue brought to light by the second paper is that the individual steps required to use time series data for predictive modeling are more numerous and more complex than those used for modeling with traditional multivariate data. Even after complexities attributable to the design phase (addressed in our first paper) have been accounted for, the management and manipulation of the time series elements (the preprocessing steps in particular) are issues that are not present in a traditional multivariate modeling paradigm. In our methods, we present the issues that arise from the time series data elements: defining a reference time; imputing and reducing time series data in order to conform to a predefined structure that was specified during the design phase; and normalizing variable families rather than individual variable instances. The final manuscript, entitled: "Using Time-Series Analysis to Predict Cardiac Arrest in a Pediatric Intensive Care Unit" presents the results that were obtained by applying the theoretical construct and its associated methods (detailed in the first two papers) to the case of cardiac arrest prediction in a pediatric intensive care unit. Our results showed that utilizing the trend analysis from the time series data elements reduced the number of classification errors by 73%. The area under the Receiver Operating Characteristic curve increased from a baseline of 87% to 98% by including the trend analysis. In addition to the performance measures, we were also able to demonstrate that adding raw time series data elements without their associated trend analyses improved classification accuracy as compared to the baseline multivariate model, but diminished classification accuracy as compared to when just the trend analysis features were added (ie, without adding the raw time series data elements). We believe this phenomenon was largely attributable to overfitting, which is known to increase as the ratio of candidate features to class examples rises. Furthermore, although we employed several feature reduction strategies to counteract the overfitting problem, they failed to improve the performance beyond that which was achieved by exclusion of the raw time series elements. Finally, our data demonstrated that pulse oximetry and systolic blood pressure readings tend to start diminishing about 10-20 minutes before an arrest, whereas heart rates tend to diminish rapidly less than 5 minutes before an arrest.

Noninvasive monitoring of serial changes in pulmonary vascular resistance and acute vasodilator testing using cardiac magnetic resonance

Objectives The study sought to evaluate the ability of cardiac magnetic resonance (CMR) to monitor acute and long-term changes in pulmonary vascular resistance (PVR) noninvasively. Background PVR monitoring during the follow-up of patients with pulmonary hypertension (PH) and the response to vasodilator testing require invasive right heart catheterization. Methods An experimental study in pigs was designed to evaluate the ability of CMR to monitor: 1) an acute increase in PVR generated by acute pulmonary embolization (n = 10); 2) serial changes in PVR in chronic PH (n = 22); and 3) changes in PVR during vasodilator testing in chronic PH (n = 10). CMR studies were performed with simultaneous hemodynamic assessment using a CMR-compatible Swan-Ganz catheter. Average flow velocity in the main pulmonary artery (PA) was quantified with phase contrast imaging. Pearson correlation and mixed model analysis were used to correlate changes in PVR with changes in CMR-quantified PA velocity. Additionally, PVR was estimated from CMR data (PA velocity and right ventricular ejection fraction) using a formula previously validated. Results Changes in PA velocity strongly and inversely correlated with acute increases in PVR induced by pulmonary embolization (r = –0.92), serial PVR fluctuations in chronic PH (r = –0.89), and acute reductions during vasodilator testing (r = –0.89, p ≤ 0.01 for all). CMR-estimated PVR showed adequate agreement with invasive PVR (mean bias –1.1 Wood units,; 95% confidence interval: –5.9 to 3.7) and changes in both indices correlated strongly (r = 0.86, p < 0.01). Conclusions CMR allows for noninvasive monitoring of acute and chronic changes in PVR in PH. This capability may be valuable in the evaluation and follow-up of patients with PH.

The association between older age and receipt of care and outcomes in patients with acute coronary syndromes : a cohort study of the Myocardial Ischaemia National Audit Project (MINAP)

Peer reviewed

Endogenous tumor necrosis factor protects the adult cardiac myocyte against ischemic-induced apoptosis in a murine model of acute myocardial infarction

Previous studies have shown that proinflammatory cytokines, such as tumor necrosis factor (TNF), are expressed after acute hemodynamic overloading and myocardial ischemia/infarction. To define the role of TNF in the setting of ischemia/infarction, we performed a series of acute coronary artery occlusions in mice lacking one or both TNF receptors. Left ventricular infarct size was assessed at 24 h after acute coronary occlusion by triphenyltetrazolium chloride (TTC) staining in wild-type (both TNF receptors present) and mice lacking either the type 1 (TNFR1), type 2 (TNFR2), or both TNF receptors (TNFR1/TNFR2). Left ventricular infarct size as assessed by TTC staining was significantly greater (P < 0.005) in the TNFR1/TNFR2-deficient mice (77.2% ± 15.3%) when compared with either wild-type mice (46.8% ± 19.4%) or TNFR1-deficient (47.9% ± 10.6%) or TNFR2-deficient (41.6% ± 16.5%) mice. Examination of the extent of necrosis in wild-type and TNFR1/TNFR2-deficient mice by anti-myosin Ab staining demonstrated no significant difference between groups; however, the peak frequency and extent of apoptosis were accelerated in the TNFR1/TNFR2-deficient mice when compared with the wild-type mice. The increase in apoptosis in the TNFR1/TNFR2-deficient mice did not appear to be secondary to a selective up-regulation of the Fas ligand/receptor system in these mice. These data suggest that TNF signaling gives rise to one or more cytoprotective signals that prevent and/or delay the development of cardiac myocyte apoptosis after acute ischemic injury.