Printer Friendly

Interactive instruments to help labs 'work smarter.'

Analyzers with built-in instructions will boost the usefulness of technology for a decreasingly skilled work force.

As laboratory instruments grow more complex, they are beginning to be used by individuals without formal lab training. Skilled technologists are required to operate multiple complex instruments and cope with the complexities of instrument operation, daily quality control, and general troubleshooting. An investment in designing, operating, and supporting intuitive interactive interfaces between instrument and user would smooth operations, producing more consistently reliable test results and eventually reducing costs.

Much of today's technology seems to have been designed to prevent us from using it properly. Telephones that can perform numerous functions provide no apparent clues to the key sequences required, nor do they warn when a function has been performed incorrectly. Institutional doors confront us with unmarked hardware that flouts expectations: Suddenly bars must be pulled, handles pushed.

* Feedback. Consider, in contrast, our recent encounter with an automatic teller machine in a busy airport far from home. Unambiguous instructions in words and graphics guided us through an operation that rewarded us with cash and relief. The experience was not stressful, but gratifying.

Smart machines, an outgrowth of advances in artificial intelligence and graphic interface design, announce their readiness and explain what to do. Recognizing that different users have different needs, they provide the appropriate level of information, from rudimentary to streamlined. We need more smart instruments at work in the lab.

This article presents a description of how interactive machines could improve matters in the clinical laboratory. Some of the underlying ideas presented here were inspired by "The Psychology of Everyday Things" by Donald A. Norman (New York, Basic Books, 1988).

* Show and tell. User-friendly computer programs reduced training time by presenting choices in menus, an improvement on the memorized strings of commands formerly required by computers. Still, the user must learn how the program is organized and understand each menu selection. A substantial amount of training and referral to manuals is often needed to operate complex systems. With an intuitive interface, a novice may be able to use the device immediately.

Systems that have intuitive interactive software show available options to the user and interrupt their improper use. The interface shows the operator how to initiate allowable actions, indicates when the system is ready to perform them, and communicates their results.

Any action with potentially undesirable results should be readily reversible. When an important contemplated action is irreversible, the system should state its possible effects and ask whether to proceed, allowing enough time to cancel the plan, if desired.

* Out of the lab. Like organizations in other fields, laboratories have recognized the need to revalidate an individual's competence to perform a wide range of tasks or services. Proficiency testing of employees must be documented more than ever before. An area that could benefit from equipment-driven continual monitoring of competence combined with built-in remedial training is point-of-care testing, including tests done in physicians' office laboratories (POLs) and at the patient's bedside.

If, as anticipated, final CLIA '88 regulations give POLs more latitude in the tests they can perform, manufacturers will perhaps feel more inclined to accelerate research and development of intuitive interactive interfaces.

They could be equally effective for fully trained technologists operating complex analytic systems in the clinical lab. The coming generation of intuitive interactive interfaces should be able to guide users through training, system operation, QC, and troubleshooting.

* Training. Standardized instruction and testing modules run on smart instruments could reduce the high cost of training and competence testing. Such programs should:

[paragraph] Identify new operators and provide them with tutorials, enabling them to process control specimens. A successful trainee, obtaining control sample results within preset limits, could then be certified to process patient specimens. A trainee who failed such training could receive further instruction in the use of the system before being allowed to process patient care specimens.

[paragraph] Be paced to the user's ability to grasp the material. Fast learners and those with accumulated expertise would be presented with more challenging situations than others. (Unless shortcuts are provided for users who have mastered the technology, experts become inattentive as they tediously go through each step of the process every time.) Using context-sensitive help screens, operators could request information as needed.

[paragraph] Be available on demand during all shifts. Training could be done at the times of lightest workload.

[paragraph] Provide simulations of common problems or malfunctions. Users could learn to identify problems and make repairs.

[paragraph] Test operators for ongoing competence and maintain records needed for QA, accreditation, and licensing. Such documentation is especially important when non-technically trained personnel operate point-of-care testing systems.

Ongoing training and PT provide a cost-effective way to insure consistent result reliability. Smart interfaces can save money by standardizing training, reducing paid training staff, and making tutorials available 24 hours a day.

* Operation. Besides assisting neophytes, smart interfaces could assist experienced workers who were about to handle unaccustomed problems or perform procedures they do infrequently.

Many analytic systems alert the operator when the constituent level in a specimen is above the linearity range or if the result (action value) could be critical to the patient's care. Some systems notify users when the specimen needs to be diluted and retested.

The relatively untrained user performing testing in the laboratory or at the point of care could be guided through dilution and retesting procedures and the performance of appropriate calculations. The interface could also help reduce the need for outside technical support for POLs and bedside testing sites. "Simple" tasks, such as putting a new roll of paper into an instrument's printer, can be difficult for the new user. The procedure is usually covered in the instruction manual, which may well be gathering dust on a shelf. To make life easier, key parts of the printed manual should be available for on-line operator assistance. Any aspect of POL or bedside testing system operation that is beyond the understanding of its user is likely to reduce result reliability and increase the cost of risk management.

Periodic maintenance and calibration are usually required at specific time intervals, when reagent lots are changed, or when drift/shift is apparent in a quality control specimen result. An intuitive interactive interface could indicate when such tasks were required. The analyzer could record this work, once done, along with other relevant information: repairs, the occurrence of unscheduled downtime, and changes in reagent, control, and calibrator lots.

For many difficult operational problems that affect analytic systems, problem identification and resolution strategies are well known to manufacturers. Troubleshooting aids could be loaded into the instrument memory and made to appear instantaneously on a monitor. If the company's technical support staff had to be brought in, their job would be facilitated by having on-line maintenance and repair records available.

* Quality control. Running control specimens with known constituent levels is fundamental in determining whether equipment is working properly from day to day. Many laboratory workers, especially those with few technical qualifications, need help in understanding control specimen results and in dealing with problem confirmation, identification, and resolution.

Contemporary laboratory information systems can make the statistical analyses needed to develop intralaboratory limits for QC, flag results that fall outside preset limits, and report outliers to the operator. Few systems can interpret the report and formulate a plan for action, however; for this, human intervention is required.

The intuitive interactive instrument could advise the user of any result outside preset limits, indicate likely problems, and advise what to do next. Such information is frequently available off line--in a manual, for example--but could be available on line for immediate display as well.

* Troubleshooting. Figure 1 shows a hypothetical interchange between an interactive analyzer and a technologist during a quality control test. Such conversations could be saved in memory and used for two purposes. First, they would allow the manufacturer's technical assistance representative to review precisely what happened and what was done about it. Second, they could provide documentation for quality assurance and accreditation activities.

These analyzers could create and retain documentation of periodic equipment cleaning and maintenance, recalibration, and changes in QC specimen and reagent lot numbers. Data could be transferred periodically to a personal computer for comprehensive record keeping of a laboratory's quality management activities.

* Effective supervision. Intelligent interfaces could assist in supervising laboratorians in technical subjects. Both the experienced and the inexperienced would benefit.

Untrained personnel could learn more easily and quickly. When QC and QA of analytic systems used by operators with minimal technical training were overseen by qualified supervisors, records could provide a means of reviewing and identifying individuals, locations, shifts, and instruments having problems. In such situations, daily control specimen testing is one way to monitor, manage, and record the competence of each operator and to document the reliability of results produced.

Trained personnel, including supervisors, would benefit from the monitoring of users. In traditional clinical laboratories, smart analyzers could help supervisors identify technologists and equipment that produced more outliers than expected. Additional training and other corrective action would follow.

When technical support visits were required, the manufacturer's personnel could find needed information in the system memory, including the operator's actions and the system's response. Reviewing such material would provide the manufacturer with objective data on common problems.

Other advantages would accrue to the manufacturer. The time and cost of technical support would be lower because of the need for fewer on-site visits. The local user could, for example, connect the analytic system to the tech service representative by modem. The service rep could reconstruct the problem by running diagnostic analyses directly on the instrument, perhaps obviating its return for repair or exchange.

* Many pluses. Analyzers with intuitive and interactive user-instrument interfaces could boost the image of technology as partners to the staff in busy, cost-conscious laboratories. Holding a dialog with an analyzer to discuss equipment problems would reduce operator effort and minimize demand on manufacturers' expensive technical support services.

Providing users with standard procedures for identifying and correcting problems could increase productivity and reduce the time in which the system was unavailable for patient testing. By providing on-line guidance for troubleshooting, interactive interfaces would speed the return to service.

Besides the other attributes of smart instruments, their automatic record-keeping capability would lead to significant savings of time and money. Collecting, analyzing, compiling, and reporting data manually to produce documentation needed to satisfy accrediting agencies and inspectors require costly staff effort. The ultimate beneficiaries of faster and more accurate results are, as always, the patients we serve.

Figure 1

Talking to the machine

A hypothetical |conversation' between a technologist and a smart analyzer Operator tests quality control sample. Analyzer responds (on monitor): QC result outside expected limits. Most likely due to chance alone. Please retest the control sample. Operator runs QC again. Analyzer: The QC result was within limits. You may test patient care specimens now. (Or:) The second result is also outside preset limits. There may be trouble with the testing system. First, test a sample from a new, freshly opened vial of control material to be sure the problem is not with the control itself. Operator tests new QC material. Analyzer: Is this a new bottle of QC material? YES--Enter 1; NO--Enter 2. Operator (keys in): 1 Analyzer: The control sample result is still outside expected limits. There may be trouble with the system. Inspect and clean the optics (describes how to perform this task). Enter 1 when you have done this. Operator cleans optics; keys in: 1 Analyzer: Retest the control sample again. Operator runs QC sample. Analyzer: The control sample result is still outside preset limits, indicating that an instrument problem may exist. (Suggests several troubleshooting procedures; asks user to try them.) Control sample remains beyond preset limit range. Analyzer: We need help. Call the toll-free technical support phone number, 1 (800) 123-4567. Be ready to hook me to the modem (describes procedure) so I can talk with my designer, who will help us determine the source of the problem.

Dr. Richard Belsey, M.D. is professor and head of chemical pathology at Oregon Health Sciences University, Portland, Ore. Dr. Daniel M. Baer, M.D., a member of MLO's Editorial Advisory Board, is professor of pathology, Oregon Health Sciences University, and chief of lab services, Veterans Affairs Medical Center, also in Portland.
COPYRIGHT 1992 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Belsey, Richard; Baer, Daniel M.
Publication:Medical Laboratory Observer
Date:Feb 1, 1992
Previous Article:Final OSHA regs mandate HIV, HBV protection.
Next Article:How to use anger constructively.

Related Articles
Strengthening the weak links in lab services.
The role of a development lab in a hospital setting.
2004: A Lab Odyssey.
Real-time PCR companies eye clinical goldmine.
In the e-learning field with Bill Strelke of Knowledge Anywhere: promoting lab-tech careers will help maintain and energize the profession's work...
SMART launches new wide-screen interactive whiteboard in Middle East.
New SMART corporate solutions to change the way companies collaborate.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters