Skip to main content

Hematology

Data Management

Data management extends beyond the simple reporting of a CBC result. The modern laboratory is a repository of massive datasets (“Big Data”) that can be mined for research, quality improvement, and population health management. Effective data management involves the aggregation, analysis, and interpretation of this information to drive operational efficiency and improve patient outcomes. The laboratory scientist must transition from being a data producer to a data analyst

Utilization Management (UM)

One of the primary uses of laboratory data is to monitor and control how lab tests are utilized by clinicians. The goal is to reduce waste (unnecessary testing) while ensuring appropriate testing

  • Duplicate Testing: Data queries can identify how often a CBC is ordered twice on the same patient within 24 hours. The lab can implement “soft stops” in the CPOE (Computerized Physician Order Entry) system to alert the doctor: “A normal CBC was performed 4 hours ago. Do you really want to order another one?”
  • Obsolete Testing: Data analysis identifies orders for outdated tests (e.g., Bleeding Time). The lab can use this data to remove these tests from the order menu or create educational pop-ups suggesting better alternatives (e.g., PFA-100)
  • Formulary Management: Similar to a pharmacy, the lab can restrict expensive genetic tests (e.g., Molecular coagulopathy panels) to specific specialists (Hematologists) to prevent general practitioners from ordering them inappropriately

Quality Assurance & Outcome Studies

Laboratory data is linked to patient outcomes to prove the value of the laboratory. This is crucial for demonstrating that “expensive” tests actually save money downstream

  • Turnaround Time (TAT) Analysis
    • Data: The LIS tracks the time from “Collection” to “Receipt” to “Verification.”
    • Outcome: The lab analyzes this data to find bottlenecks. For example, if the “Collection-to-Receipt” time is high, the problem is transport (couriers/tube system), not the lab itself
    • Clinical Impact: Improving Troponin or Stroke Panel TAT directly correlates with reduced length of stay in the ER
  • Correlation Studies
    • Research: Comparing Hematology data (e.g., Immature Platelet Fraction - IPF) with clinical outcomes (e.g., platelet recovery after chemotherapy/transplant)
    • Goal: Validating that a new parameter (IPF) allows doctors to discharge patients earlier or withhold unnecessary platelet transfusions
  • Blood Management (PBM)
    • Data: Analyzing Hgb triggers for transfusion. Are doctors transfusing RBCs when Hgb is > 7.0 g/dL?
    • Outcome: Providing this data to the Transfusion Committee allows for targeted education to reduce inappropriate blood usage, saving money and reducing transfusion reaction risks

Population Health & Epidemiology

The Hematology lab serves as a sentinel for public health trends. Aggregated LIS data is often reported to state or federal agencies

  • Prevalence Studies: The lab can query LIS data to determine the prevalence of anemia (Low Hgb) or lead poisoning in specific zip codes or age groups (pediatrics). This informs public health interventions
  • Oncology Registries: Pathology and Hematology data (Bone Marrow reports, Flow Cytometry leukemia phenotypes) are mined to populate cancer registries. This data tracks the incidence of specific leukemias (e.g., CML vs. AML) and survival rates
  • Reference Range Determination
    • Research: The lab can use its own “normal” patient data (thousands of outpatient results) to calculate its own reference intervals using non-parametric statistics. This is more accurate than using the manufacturer’s generic ranges because it reflects the specific local population

Operational Analytics (Business Intelligence)

Laboratory managers use data dashboards to run the “business” side of the lab

  • Test Volume Analysis: Tracking the number of tests performed by hour of the day and day of the week
    • Application: Staffing. If data shows a spike in ER workload at 2:00 AM, the manager justifies moving a day-shift laboratory scientist to the night shift or hiring a mid-shift float
  • Productivity Metrics
    • Billable Tests per FTE: A calculation of efficiency. (Total Tests / Total Full-Time Equivalents)
    • Goal: To benchmark the lab’s efficiency against similar hospitals. If the lab’s productivity is low, it indicates overstaffing or inefficient workflow
  • Analyzer Performance
    • Data: Tracking the frequency of QC failures or “User Maintenance” logs
    • Outcome: Identifying “Lemon” instruments that break down constantly. This data is used to negotiate service contract credits or justify capital replacement

Data Storage & Retention

The management of data also involves its physical and digital preservation. Regulatory agencies (CAP/CLIA) mandate how long records must be kept

  • Retention Schedules
    • General Lab Records (QC, Maintenance): Typically 2 years
    • General Patient Reports: Typically 2 years (though most hospitals keep EMR data indefinitely)
    • Blood Bank/Immunohematology: 10 years (or indefinitely for deferred donors)
    • Pathology/Bone Marrow Slides: 10 years
    • Blocks: 10 years
    • Wet Tissue: 2 weeks after final report
  • Data Retrieval: The system must be capable of retrieving archived data within a reasonable time (e.g., 24 hours) for audits or legal discovery
  • Backup Systems: Data must be backed up daily (tape drives or cloud) to prevent loss during server failures. Redundancy (mirror servers) ensures that if the main LIS crashes, data is not lost