Large-Scale Data Mining to Optimize Patient-Centered Scheduling at Health Centers.

Patient-centered appointment access is of critical importance at community health centers (CHCs) and its optimal implementation entails the use of advanced data analytics. This study seeks to optimize patient-centered appointment scheduling through data mining of Electronic Health Record/Practice Management (EHR/PM) systems. Data was collected from different EHR/PM systems in use at three CHCs across the state of Indiana and integrated into a multidimensional data warehouse. Data mining was performed using decision tree modeling, logistic regression, and visual analytics combined with n-gram modeling to derive critical influential factors that guide implementation of patient-centered open-access scheduling. The analysis showed that appointment adherence was significantly correlated with the time dimension of scheduling, with lead time for an appointment being the most significant predictor. Other variables in the time dimension such as time of the day and season were important predictors as were variables tied to patient demographic and clinical characteristics. Operationalizing the findings for selection of open-access hours led to a 16% drop in missed appointment rates at the interventional health center. The study uncovered the variability in factors affecting patient appointment adherence and associated open-access interventions in different health care settings. It also shed light on the reasons for same-day appointment through n-gram-based text mining. Optimizing open-access scheduling methods require ongoing monitoring and mining of large-scale appointment data to uncover significant appointment variables that impact schedule utilization. The study also highlights the need for greater "in-CHC" data analytic capabilities to re-design care delivery processes for improving access and efficiency.

Identifying Successful Practices to Overcome Access to Care Challenges in Community Health Centers: A "Positive Deviance" Approach.

BACKGROUND: Despite health care access challenges among underserved populations, patients, providers, and staff at community health clinics (CHCs) have developed practices to overcome limited access. These "positive deviant" practices translate into organizational policies to improve health care access and patient experience. OBJECTIVE: To identify effective practices to improve access to health care for low-income, uninsured or underinsured, and minority adults and their families. PARTICIPANTS: Seven CHC systems, involving over 40 clinics, distributed across one midwestern state in the United States. METHODS: Ninety-two key informants, comprised of CHC patients (42%) and clinic staff (53%), participated in semi-structured interviews. Interview transcripts were subjected to thematic analysis to identify patient-centered solutions for managing access challenges to primary care for underserved populations. Transcripts were coded using qualitative analytic software. RESULTS: Practices to improve access to care included addressing illiteracy and low health literacy, identifying cost-effective resources, expanding care offerings, enhancing the patient-provider relationship, and cultivating a culture of teamwork and customer service. Helping patients find the least expensive options for transportation, insurance, and medication was the most compelling patient-centered strategy. Appointment reminders and confirmation of patient plans for transportation to appointments reduced no-show rates. CONCLUSION: We identified nearly 35 practices for improving health care access. These were all patient-centric, uncovered by both clinic staff and patients who had successfully navigated the health care system to improve access.

A Multidimensional Data Warehouse for Community Health Centers.

Community health centers (CHCs) play a pivotal role in healthcare delivery to vulnerable populations, but have not yet benefited from a data warehouse that can support improvements in clinical and financial outcomes across the practice. We have developed a multidimensional clinic data warehouse (CDW) by working with 7 CHCs across the state of Indiana and integrating their operational, financial and electronic patient records to support ongoing delivery of care. We describe in detail the rationale for the project, the data architecture employed, the content of the data warehouse, along with a description of the challenges experienced and strategies used in the development of this repository that may help other researchers, managers and leaders in health informatics. The resulting multidimensional data warehouse is highly practical and is designed to provide a foundation for wide-ranging healthcare data analytics over time and across the community health research enterprise.

Effects of data transformation methods on classification of patients diagnosed with myocardial infarction.

Large datasets may contain redundant data. Variable selection methods that select most relevant variables in the data set, fail to consider the interaction between the variables. Data transformation methods are used to transfer the original data to a new dimension and capture the most significant information within the data set. The data set used in this study was based on 45 clinical variables collected from 697 patients diagnosed as either having myocardial infarction (MI) or not. Principal component analysis (PCA) and independent component analysis (ICA) were applied prior to classification of patients to MI or Non-MI groups using support vector machines (SVM).

Automated blood sampling and glucose sensing in critical care settings.

BACKGROUND: Tight glycemic control (TGC) studies in intensive care units (ICU) have shown substantial improvements in clinical outcomes. However, implementation of TGC in ICU practice is partly constrained by the lack of automated continuous blood glucose monitoring systems that can facilitate clinically accurate feedback of glycemic data. The aim of this work is to develop a portable automated blood sampling system for integration with a glucose sensor for use in critical care settings. METHODS: clinical prototypes for glucose sensing in blood were developed based on two distinct technologies: mid-infrared laser absorption spectroscopy and electrochemistry. Concurrently, an automated peripheral venous blood sampling system was developed for integration with the glucose sensing system. RESULTS: The glucose sensing prototypes were validated clinically with various biological samples in a continuous mode. A customized micropump was employed in conjunction with a novel peripheral venous catheter system to automatically sample blood from the subject's forearm. Microvolumes of blood were sampled in continuous and intermittent modes at clinically relevant user-defined frequencies. The clinical feasibility of blood sampling, along with continuous glucose sensing, was demonstrated. CONCLUSION: Cascade's automated peripheral venous blood sampling system, in combination with a flow-through glucose sensor system, offers several advantages over current state-of-the-art systems. This includes the potential for significantly improved workflow in the ICU, minimal discomfort to the patient, and accurate glucose measurement in whole blood, thus helping achieve tight glycemic control.