Comparative Analysis of Point-of-C care vs Central Lab Hematology Outcomes

Thesis Abstract

This study addresses the reliability and clinical impact of Point-of-Care (POC) hematology testing compared with Central Laboratory (CL) hematology results in acute care settings, where rapid decision-making may influence patient outcomes. The problem centers on potential discordances between POC and CL hematology measurements, which could affect time-to-treatment, diagnostic accuracy, and patient safety. The aim is to evaluate equivalence and agreement between POC and CL hematology outputs and to identify determinants of any discordance that influence clinical decisions. Specific objectives are (1) to quantify the level of agreement between POC and CL complete blood counts (CBCs) across diverse patient cohorts; (2) to assess the impact of measurement differences on clinical decision-making, including transfusion thresholds and anticoagulation management; (3) to identify pre-analytical, analytical, and operational factors contributing to discrepancies; (4) to compare turnaround times (TAT) and subsequent patient outcomes such as length of stay and 30-day readmission; and (5) to propose a decision-support framework integrating POC results with CL data. The study adopts a cross-sectional, observational design conducted in a tertiary-care hospital over 12 months, enrolling a census sample of 1,200 adult inpatients who require serial CBC testing. A stratified sampling approach ensures representation across emergency, medical, surgical, and intensive care units. Data collection uses paired CBC measurements obtained concurrently by trusted POC devices (e.g., hematology analyzers within point-of-care carts) and standard CL analyzers, with linked clinical data extracted from electronic health records. Instruments include calibrated POC hematology analyzers, CL hematology modules, a data capture software for time stamps and results, and a structured clinical outcomes form. Validity and reliability are addressed through device calibration logs, proficiency testing of operators, inter-device concordance checks, and pilot testing of data extraction protocols. Analytical methods comprise descriptive statistics to summarize central tendencies and dispersion, Bland-Altman analysis to assess agreement and systematic bias between POC and CL CBC parameters (WBC, RBC, Hb, Hct, Platelets), and Cohen’s kappa for categorical interpretations (anemia, thrombocytopenia thresholds). Regression analyses investigate associations between discordance magnitude and variables such as pre-analytical time delays, sample type (arterial vs venous), hemolysis indices, ambient temperature, operator experience, and device lots. ANOVA and post hoc tests compare TAT and downstream outcomes across units and patient categories. Multivariate logistic regression identifies predictors of clinically significant discordance that would alter management decisions, while survival analyses examine the impact on length of stay and 30-day readmissions. The theoretical framework draws on the Theory of Measurement Validity and the Technology Acceptance Model to interpret how workflow, user interaction, and perceived trust influence the integration of POC data into clinical decision-making. Anticipated findings include a high but not perfect agreement between POC and CL CBC values, with mean differences within clinically acceptable ranges for Hb and platelets in stable patients, but potential discrepancies in WBC and hematocrit under conditions of sample handling and device calibration drift. It is expected that discordance will correlate with pre-analytical delays, sample handling, and operator proficiency, and that TAT advantages of POC will partially offset measurement variability by enabling faster initial therapeutic decisions. The study will contribute to knowledge by delineating specific contexts in which POC hematology is reliable, identifying thresholds for acceptable discordance, and informing guidelines for when CL confirmation is warranted before critical interventions. The conclusion will likely recommend an integrated testing pathway that leverages the rapid turnaround of POC data for urgent decisions, supplemented by CL confirmation in cases with flagged discordance or high-risk patients. Practical recommendations include standardized operator training, routine cross-validation of POC devices with CL results at defined intervals, post-analytic interpretive rules within electronic ordering systems, and targeted quality assurance measures to minimize pre-analytical variability. The study aims to provide evidence-based adoption criteria for POC hematology in hospital workflows, contributing to optimized patient care, improved timeliness of interventions, and safer reliance on hematology results in acute care environments.