Search

Although much has been done to reduce the transmission of HIV, there are still 36.7 million individuals living with the virus worldwide, with 1.1 million of those living in the US1. At the University of Alabama at Birmingham (UAB) there is a specialty HIV clinic, the 1917 clinic, which has helped over 12,000 patients in their HIV healthcare journey since 1988. One way to reduce the prevalence of HIV is by engaging patients in their care. Patients that arrive at their clinic visits regularly have been shown to have improved medication adherence, reduced mortality, and improved health outcomes2. Our clinic experienced a 19% visit no show rate in fiscal year 2018. Identifying patients at risk for no show visits is important for clinic scheduling, staffing and resource allocation. Identifying those at risk is the first step to the implementation of strategies to decrease no shows. The data used for this project consisted of appointment data for 4,950 HIV positive patients between October 2015 and July 2019 at UAB's HIV clinic. The data source utilized was UAB's scheduling software, IDX®. To ensure data accuracy, automate data pre-processing, and provide a list of patients to be contacted, a relational database was developed in MS SQL Server using IDX® data and an automated SSIS package was utilized for the extract, transfer, and load (ETL) processes. A predictive logistic regression model (predicting no show Yes vs. No) was developed using the data from the database. The data was split into training (80% or 64,354 records) and testing (20% or 21,450 records) sets prior to deploying the model. A probability score for each patient was generated based on the trained logistic regression model with the independent variables of: previous no show/total appointments ratio, age, race, lag days (the days between scheduled and appointment dates) and gender. A threshold value, above which the patient was more likely to no show, was selected on the basis of analysis of the receiver operating curve (ROC) which showed the distribution of probability scores. The model was then tested for precision, specificity, and the weighted average (F1). Among the 21,450 testing records, the distribution of no-show probability scores ranged from 0.09 to 0.83. A threshold of 0.3 was chosen based on its comparatively lower false negative (FN) error of 1,549. The true positive (TP), or those patients predicted to no show and actually did no show, was 6,435. The model works with 44% precision, or specificity, 80% sensitivity, with an F1 of 57%. A cutoff point was chosen that emphasized sensitivity in our model as we preferred to bring the potential of an intervention to more patients than what was needed. By using predictive algorithms, such as ours, healthcare organizations will be better positioned to identify and have the opportunity to offer interventions to engage patients in longitudinal clinic care and the ensuing improved morbidity and mortality. [ABSTRACT FROM AUTHOR]

This study aims to: 1) examine the relationship between the level of hospital high-technology (high-tech) services and total expenses; and 2) explore whether this relationship varies depending on the measure of high-tech services. The study population consisted of medical/surgical hospitals operating in the US between 1997 and 2014 for a total of 57,116 hospital-year observations. The dependent variable, total expenses per inpatient day, along with predictor variables of four measures of high-tech services and control variables were analyzed in facility & year fixed effects regression. Spearman correlation results indicated high level of correlation among high technology indices ranging from 0.72 to 0.99. The regression results indicated that for every unit increase in high-tech services, there are a 30% (Saidin Index and Count Index), and 180% (Limited Count Index) increase in total expenses per inpatient day for all high-tech indices (p<0.001) except the high-tech-health referral region (HRR) index, which exhibited a 10% decrease (p<0.001). Based on three different measures of high-tech services, our findings suggest that an increase in high-tech services leads to higher total expenses. However, taking into account the rareness of high-tech services within the HRR shows that high-tech services result in lower total expenses. [ABSTRACT FROM AUTHOR]

Colorectal cancer researchers spend a substantial amount of effort performing integration, cleansing, interpretation, and aggregation of raw data from multiple sources, including health records and clinical research data. These efforts are often replicated for each project, with investigators running up against the same challenges and experiencing the same pitfalls discovered by those before them. The overall objective of this project is to design and implement a colorectal cancer data warehouse infrastructure to improve acquisition, management, and analysis of relevant health records, clinical research, and tumor registry data from our institution and state. The current data preparation processes, at best, are inefficient, costly, time-consuming, and cumbersome (Puppala et al. 2015). Moreover, without the needed information technology (IT) infrastructure, the potential of the ever-growing heterogeneous data accumulated in disparate data sets would be still untapped. Our previous colorectal cancer work included discovery and validation of biomarkers, the roles of tumor location and race/ethnicity, treatment efficacy, and prognostic/predictive models that considered the effect of molecular, demographical, epidemiological, and clinico-pathologic features on outcomes, such as mortality, relapse, and survival (Bovell et al. 2013). The data sources for these projects included data exports from clinical records and spreadsheet files created for each research project. Data management for each project is usually performed in an ad-hoc manner, involving manual processes of data entry, matching, and merging. This process is error-prone and inefficient for data reuse, and not suitable to incorporate additional data sources. This work proposes to initially design and implement a colorectal cancer data warehouse infrastructure that incorporates existing molecular-level and patient-level research data with continuous data feed from institutional enterprise data warehouse (EDW) in a multidimensional database format. We then plan to expand the scope of this data infrastructure to include social determinants of health (SDH), geospatial census, and state level tumor registry data. Such a data management platform will enable us to efficiently analyze disparities among various populations, create state-wide map projections and dashboards, and analyze certain outcomes (e.g. risk and aggressiveness of the disease) to identify differences between rural and urban populations. Creation of a colorectal cancer data warehouse infrastructure would allow us to store, clean, and manage the existing data sources efficiently, increase the quality and reliability of underlying data for our research, and incorporate new data sources to facilitate future research. [ABSTRACT FROM AUTHOR]