Phlebotomists

Match laboratory requisition forms to specimen tubes.

AI: Fully automatable - Matching requisition forms to specimen tubes is routinely automated in labs using barcodes, OCR, and rule-based/AI verification systems and can be fully automated.

Enter patient, specimen, insurance, or billing information into computer.

AI: Fully automatable - Entering patient, specimen, insurance, or billing data is already routinely automated using OCR, NLP, RPA, and EHR APIs, enabling full automation in most settings by 2025.

Process blood or other fluid samples for further analysis by other medical professionals.

AI: Fully automatable - Many laboratories already use robotic processors, centrifuges, liquid handlers, and LIMS to fully automate sample processing workflows for downstream analysis, making full automation feasible by 2025.

Provide sample analysis results to physicians to assist diagnosis.

AI: Fully automatable - Laboratory information systems and AI interpretation tools can generate and transmit sample analysis results and preliminary interpretive comments to physicians automatically, though clinicians retain final diagnostic responsibility.

Document route of specimens from collection to laboratory analysis and diagnosis.

AI: Fully automatable - Barcoding/RFID, LIMS, and workflow automation already track specimen route end-to-end and AI can aggregate and document chain-of-custody and processing steps fully.

Conduct standards tests, such as blood alcohol, blood culture, oral glucose tolerance, glucose screening, blood smears, or peak and trough drug levels tests.

AI: Fully automatable - Modern lab automation, analyzers, and AI-powered image/assay interpretation can conduct and interpret standard tests (blood alcohol, cultures, glucose, smears, peak/trough levels) end-to-end in routine settings.

Conduct hemoglobin tests to ensure donor iron levels are normal.

AI: Fully automatable - Point-of-care hemoglobin analyzers integrated with LIS and automated workflows can perform, record, and flag hemoglobin results reliably without manual interpretation.

Serve refreshments to donors to ensure absorption of sugar into their systems.

AI: Fully automatable - Dispensing refreshments is a routine, low-risk task that existing vending, robotic service, and workflow automation systems can fully manage in most donation settings.

Dispose of contaminated sharps, in accordance with applicable laws, standards, and policies.

AI: Partial - Disposing contaminated sharps is a physical, regulated activity that can be aided by engineered or robotic systems but still typically requires human oversight and compliance verification, so only partial automation is realistic by 2025.

Dispose of blood or other biohazard fluids or tissue, in accordance with applicable laws, standards, or policies.

AI: Partial - Disposal of biohazard fluids involves physical containment and regulatory procedures that closed systems can automate in part, but human oversight and facility-specific policies limit full automation by 2025.

Draw blood from veins by vacuum tube, syringe, or butterfly venipuncture methods.

AI: Partial - Venous blood draws require fine motor control, patient-specific judgment, and handling of complications; prototype robotic venipuncture exists but is not generally applicable or widely deployed, so only partial automation is possible.

Draw blood from capillaries by dermal puncture, such as heel or finger stick methods.

AI: Partial - Capillary dermal punctures are simpler and there are automated/self-collection devices, but routine clinical variability and patient handling mean widespread full automation is not yet realized, so this is partially automatable.

Organize or clean blood-drawing trays, ensuring that all instruments are sterile and all needles, syringes, or related items are of first-time use.

AI: Partial - Sterilization and inventory management systems can automate many aspects of tray preparation, but ensuring sterile technique and verifying single-use status typically require human checks, so automation is partial.

Collect fluid or tissue samples, using appropriate collection procedures.

AI: Partial - Simple fluid or swab collections can be semi-automated, but the wide range of sample types and procedural judgment required for many tissue or complex collections limits automation to a partial level.

Collect specimens at specific time intervals for tests, such as those assessing therapeutic drug levels.

AI: Partial - Timing, scheduling, and reminders for interval specimen collection can be fully automated, but performing the timed physical collections generally still requires human staff, so overall capability is partial.

Determine donor suitability, according to interview results, vital signs, and medical history.

AI: Partial - Automated screening algorithms can evaluate interviews, vitals, and histories to identify eligible donors, but ambiguous or high-risk cases still require human clinical judgement.

Transport specimens or fluid samples from collection sites to laboratories.

AI: Partial - Specimen transport is often automated within facilities via pneumatic tubes, conveyors, or autonomous carts with tracking, but end-to-end universal automation and handling of exceptions remain limited, so partial automation is realistic.

Monitor blood or plasma donors during and after procedures to ensure health, safety, and comfort.

AI: Partial - AI can continuously monitor donor vitals, detect adverse patterns, and alert staff, but cannot provide the hands-on care, reassurance, or immediate interventions humans provide.

Explain fluid or tissue collection procedures to patients.

AI: Partial - AI chatbots and multimedia tools can provide standardized, multilingual explanations of collection procedures, but cannot fully replace human interaction for consent, tailoring, and managing anxiety or complications.

Train other medical personnel in phlebotomy or laboratory techniques.

AI: Partial - AI can deliver curricula, simulations, and feedback for phlebotomy and lab techniques, but hands-on supervision and competency assessment require human trainers.

Perform saline flushes or dispense anticoagulant drugs, such as Heparin, through intravenous (IV) lines, in accordance with licensing restrictions and under the direction of a medical doctor.

AI: Partial - Smart pumps and decision-support systems can control infusions and guide anticoagulant dosing, but actual medication administration and legal/licensing oversight remain human responsibilities.

Calibrate or maintain machines, such as those used for plasma collection.

AI: Partial - AI systems can provide diagnostics, predictive maintenance, and guided calibration instructions, but they generally cannot perform complex physical repairs or certified maintenance autonomously by 2025.

Administer subcutaneous or intramuscular injects, in accordance with licensing restrictions.

AI: Partial - AI can assist with decision support, scheduling, and automated reminders, but legally and physically administering subcutaneous or intramuscular injections typically requires licensed personnel or specialized robotic hardware not widely deployed.

Draw blood from arteries, using arterial collection techniques.

AI: Not automatable - Arterial draws require fine manual dexterity, bedside assessment, and immediate clinical judgement that AI cannot autonomously perform in typical clinical settings as of 2025.