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.

NEW TOOLS FOR MONITORING GAMMA CAMERA UNIFORMITY

Detector uniformity is a fundamental performance characteristic of all modern gamma camera systems, and ensuring a stable, uniform detector response is critical for maintaining clinical images that are free of artifact. For these reasons, the assessment of detector uniformity is one of the most common activities associated with a successful clinical quality assurance program in gamma camera imaging. The evaluation of this parameter, however, is often unclear because it is highly dependent upon acquisition conditions, reviewer expertise, and the application of somewhat arbitrary limits that do not characterize the spatial location of the non-uniformities. Furthermore, as the goal of any robust quality control program is the determination of significant deviations from standard or baseline conditions, clinicians and vendors often neglect the temporal nature of detector degradation (1). This thesis describes the development and testing of new methods for monitoring detector uniformity. These techniques provide more quantitative, sensitive, and specific feedback to the reviewer so that he or she may be better equipped to identify performance degradation prior to its manifestation in clinical images. The methods exploit the temporal nature of detector degradation and spatially segment distinct regions-of-non-uniformity using multi-resolution decomposition. These techniques were tested on synthetic phantom data using different degradation functions, as well as on experimentally acquired time series floods with induced, progressively worsening defects present within the field-of-view. The sensitivity of conventional, global figures-of-merit for detecting changes in uniformity was evaluated and compared to these new image-space techniques. The image-space algorithms provide a reproducible means of detecting regions-of-non-uniformity prior to any single flood image’s having a NEMA uniformity value in excess of 5%. The sensitivity of these image-space algorithms was found to depend on the size and magnitude of the non-uniformities, as well as on the nature of the cause of the non-uniform region. A trend analysis of the conventional figures-of-merit demonstrated their sensitivity to shifts in detector uniformity. The image-space algorithms are computationally efficient. Therefore, the image-space algorithms should be used concomitantly with the trending of the global figures-of-merit in order to provide the reviewer with a richer assessment of gamma camera detector uniformity characteristics.

Maternal factors, fetal parameters and the risk of shoulder dystocia

Introduction. Shoulder dystocia is a serious complication of vaginal birth, with an incidence ranging from 0.15% to 2.1% of all births. There are approximately 4 million births per year in the United States and shoulder dystocia will be experienced by approximately 20,000 women each year. Although studies have been reported on shoulder dystocia, few studies have addressed both maternal and fetal risk factors. The purpose of this study was to identify maternal and fetal risk factors for shoulder dystocia while proposing factors that could be used to predict impending shoulder dystocia. ^ Material and methods. Articles were reviewed from Medline Pubmed using the search phrase "Risk factors of shoulder dystocia" and Medline Ovid using the search words "Dystocia", "Shoulder" and "Risk factors". Rigorous selection criteria were used to identify articles to be included in the study. Data collected from identified articles were transferred to STATA 10 software for trend analysis of the incidence of shoulder dystocia and the year of publication and a pair wise correlation was also determined between these two variables. ^ Results. Among a total of 343 studies identified, only 20 met our inclusion criteria and were retained for this review. The incidence of shoulder dystocia ranged from 0.07% to 2% and there was no particular trend or correlation between the incidence of shoulder dystocia and year of publication between 1985 and 2007. Pre-gestational and gestational diabetes, postdatism, obesity, birth weight > 4000g and fundal height at last visit > 40cm were identified as major risk factors in our series of studies. ^ Conclusion. Future strategies to predict shoulder dystocia should focus on pre-gestational and gestational diabetes mellitus, postdatism, obesity, birth weight > 4000g and fundal height at last visit > 40cm. ^

Evaluating the disparity of female breast cancer mortality among racial groups - a spatiotemporal analysis

Background The literature suggests that the distribution of female breast cancer mortality demonstrates spatial concentration. There remains a lack of studies on how the mortality burden may impact racial groups across space and over time. The present study evaluated the geographic variations in breast cancer mortality in Texas females according to three predominant racial groups (non-Hispanic White, Black, and Hispanic females) over a twelve-year period. It sought to clarify whether the spatiotemporal trend might place an uneven burden on particular racial groups, and whether the excess trend has persisted into the current decade. Methods The Spatial Scan Statistic was employed to examine the geographic excess of breast cancer mortality by race in Texas counties between 1990 and 2001. The statistic was conducted with a scan window of a maximum of 90% of the study period and a spatial cluster size of 50% of the population at risk. The next scan was conducted with a purely spatial option to verify whether the excess mortality persisted further. Spatial queries were performed to locate the regions of excess mortality affecting multiple racial groups. Results The first scan identified 4 regions with breast cancer mortality excess in both non-Hispanic White and Hispanic female populations. The most likely excess mortality with a relative risk of 1.12 (p = 0.001) occurred between 1990 and 1996 for non-Hispanic Whites, including 42 Texas counties along Gulf Coast and Central Texas. For Hispanics, West Texas with a relative risk of 1.18 was the most probable region of excess mortality (p = 0.001). Results of the second scan were identical to the first. This suggested that the excess mortality might not persist to the present decade. Spatial queries found that 3 counties in Southeast and 9 counties in Central Texas had excess mortality involving multiple racial groups. Conclusion Spatiotemporal variations in breast cancer mortality affected racial groups at varying levels. There was neither evidence of hot-spot clusters nor persistent spatiotemporal trends of excess mortality into the present decade. Non-Hispanic Whites in the Gulf Coast and Hispanics in West Texas carried the highest burden of mortality, as evidenced by spatial concentration and temporal persistence.

Policy analysis of Texas Senate Bill 532 and the regulation of nurse practitioners in the convenience care model

Background. Retail clinics, also called convenience care clinics, have become a rapidly growing trend since their initial development in 2000. These clinics are coupled within a larger retail operation and are generally located in "big-box" discount stores such as Wal-mart or Target, grocery stores such as Publix or H-E-B, or in retail pharmacies such as CVS or Walgreen's (Deloitte Center for Health Solutions, 2008). Care is typically provided by nurse practitioners. Research indicates that this new health care delivery system reduces cost, raises quality, and provides a means of access to the uninsured population (e.g., Deloitte Center for Health Solutions, 2008; Convenient Care Association, 2008a, 2008b, 2008c; Hansen-Turton, Miller, Nash, Ryan, Counts, 2007; Salinsky, 2009; Scott, 2006; Ahmed & Fincham, 2010). Some healthcare analysts even suggest that retail clinics offer a feasible solution to the shortage of primary care physicians facing the nation (AHRQ Health Care Innovations Exchange, 2010). ^ The development and performance of retail clinics is heavily dependent upon individual state policies regulating NPs. Texas currently has one of the most highly regulated practice environments for NPs (Stout & Elton, 2007; Hammonds, 2008). In September 2009, Texas passed Senate Bill 532 addressing the scope of practice of nurse practitioners in the convenience care model. In comparison to other states, this law still heavily regulates nurse practitioners. However, little research has been conducted to evaluate the impact of state laws regulating nurse practitioners on the development and performance of retail clinics. ^ Objectives. (1). To describe the potential impact that SB 532 has on retail clinic performance. (2). To discuss the effectiveness, efficiency, and equity of the convenience care model. (3). To describe possible alternatives to Texas' nurse practitioner scope of practice guidelines as delineated in Texas Senate Bill 532. (4). To describe the type of nurse practitioner state regulation (i.e. independent, light, moderate, or heavy) that best promotes the convenience care model. ^ Methods. State regulations governing nurse practitioners can be characterized as independent, light, moderate, and heavy. Four state NP regulatory types and retail clinic performance were compared and contrasted to that of Texas regulations using Dunn and Aday's theoretical models for conducting policy analysis and evaluating healthcare systems. Criteria for measurement included effectiveness, efficiency, and equity. Comparison states were Arizona (Independent), Minnesota (Light), Massachusetts (Moderate), and Florida (Heavy). ^ Results. A comparative states analysis of Texas SB 532 and alternative NP scope of practice guidelines among the four states: Arizona, Florida, Massachusetts, and Minnesota, indicated that SB 532 has minimal potential to affect the shortage of primary care providers in the state. Although SB 532 may increase the number of NPs a physician may supervise, NPs are still heavily restricted in their scope of practice and limited in their ability to act as primary care providers. Arizona's example of independent NP practice provided the best alternative to affect the shortage of PCPs in Texas as evidenced by a lower uninsured rate and less ED visits per 1,000 population. A survey of comparison states suggests that retail clinics thrive in states that more heavily restrict NP scope of practice as opposed to those that are more permissive, with the exception of Arizona. An analysis of effectiveness, efficiency, and equity of the convenience care model indicates that retail clinics perform well in the areas of effectiveness and efficiency; but, fall short in the area of equity. ^ Conclusion. Texas Senate 532 represents an incremental step towards addressing the problem of a shortage of PCPs in the state. A comparative policy analysis of the other four states with varying degrees of NP scope of practice indicate that a more aggressive policy allowing for independent NP practice will be needed to achieve positive changes in health outcomes. Retail clinics pose a temporary solution to the shortage of PCPs and will need to expand their locations to poorer regions and incorporate some chronic care to obtain measurable health outcomes. ^

Analysis of life expectancy using linear and nonlinear models

Life expectancy has consistently increased over the last 150 years due to improvements in nutrition, medicine, and public health. Several studies found that in many developed countries, life expectancy continued to rise following a nearly linear trend, which was contrary to a common belief that the rate of improvement in life expectancy would decelerate and was fit with an S-shaped curve. Using samples of countries that exhibited a wide range of economic development levels, we explored the change in life expectancy over time by employing both nonlinear and linear models. We then observed if there were any significant differences in estimates between linear models, assuming an auto-correlated error structure. When data did not have a sigmoidal shape, nonlinear growth models sometimes failed to provide meaningful parameter estimates. The existence of an inflection point and asymptotes in the growth models made them inflexible with life expectancy data. In linear models, there was no significant difference in the life expectancy growth rate and future estimates between ordinary least squares (OLS) and generalized least squares (GLS). However, the generalized least squares model was more robust because the data involved time-series variables and residuals were positively correlated. ^