Printer Friendly

A checklist of vendor information for a better lab computer system.

Although computer functions and features are very important factors, inadequate performance is often the primary reason laboratory information systems fail to satisfy users. An LIS that otherwise meets the laboratory's functional requirements may not, for example, offer prompt CRT responses or print reports fast enough.

Usually the vendor is blamed for "under-configuring" the system-that is, for installing too small a computer to provide acceptable performance levels. The vendor, however, may not have received appropriate data describing lab workload and operations.

In many cases, the underlying problem is that the laboratory does not know its computer-critical workload-the activity that affects computer performance, such as testing on an interfaced automated instrument-and the distribution of that workload during the day. Without such information, LIS configurations may be based on incomplete or irrelevant operational and workload statistics.

It is clear that system performance depends on hardware and software as well as computer workload. Not quite so obvious, but critical in impact on performance, is the way a laboratory will use its LIS. An understanding of the laboratory's operational environment is one of the most important things a vendor must have in order to prepare a valid proposal for an information system.

How can laboratories provide appropriate data to help their prospective LIS vendors establish a system configuration of the fight size? (The right size means neither too large to be cost-effective nor too small to accomplish all the required functions efficiently.)

The answer lies in specifying the key operational and workload parameters that are the best predictors

of required LIS performance capabilities and providing these data to prospective LIS vendors as part

of a request for quotation or request for proposal.

Such parameters should be established for the current laboratory environment and projected into the future. With the modular growth potential built into most modern computers, it may be desirable to acquire a system just large enough initially to meet current and short-term (one to two years') needs, thus deferring expenditures on expanded capability until it is actually required.

Eleven U.S. vendors of complete laboratory systems, the majority of those in the industry, responded to a survey we conducted on this subject. They identified the following parameters that influence LIS configuration, and they grouped them according to relative importance.

Very important

Number of accessions, or specimen numbers, assigned on a typical day and on a busy day. Deten-nine this by section or workstation and in total for the lab.

Number of active workstations by section and in total for the laboratory. Ideally, figures for accessions and number of results reported should correspond to these identified workstations.

Number of current and projected reference clients. Indicate the number of larger clients for whom a remote, on-site printer may be justified.

Number of laboratory outpatient visits annually and on a typical and a busy day.

The period of on-line retention of data, expressed

in days after 1) discharge for inpatients, 2) patient visit for outpatients, and 3) initial report for reference testing. Although the cost of data storage is declining rapidly, a very large active database can prolong retrieval, processing, and archiving of selected data.

Important

Test counts, including controls and standards, by section and in total for the lab. The number of billable procedures and CAP workload units are less useful than test counts for determining how much work the computer has to do. A chemistry profile can be a single billable procedure, even if it consists of 20 tests for which the computer has to process data. Similarly, CAP units may be relatively low in a highly automated lab, though test volume and computer workload are high. .

Number of results, including controls, reported on a typical day, and on a busy day, by section or workstation and in total for the lab. Count each component of a multiple-result test, such as a chemistry profile or complete blood count with differential.

Number of automated instruments to interface, with their schedule of operation or periods of peak, use.

Number of patient cumulative reports produced daily. Remember, these reports are prepared only for patients who have new lab results on any given day.

Projected number and scheduled times for the printing of computer-generated work sheets and load lists. Assume such printouts will follow major collection rounds or incoming courier deliveries and that they will precede scheduled instrument runs.

Number of licensed beds and active beds and average census in the hospital served.

Less important

If a hospital information system will be interfaced to the LIS, a description of HIS hardware and software and the number of HIS/LIS transactions by type-admissions, transfers, discharges, demographic changes, lab orders, and so on. Estimate the distribution of such transactions by hour throughout the day. Scheduled times of specimen collection or courier specimen deliveries and an estimate of the number Estimated per collection/delivery cycle.

Estimated frequency and number of interim result reports produced daily for inpatients, outpatients, andreferral clients.

Number of admissions or discharges. The number of inpatient demographic records has an impact on on-line disk storage and archiving requirements.

Also mentioned by several vendors was the number of lab personnel or the full-time equivalent count.

These importance rankings are subjective. Information about every factor we have listed can contribute to a vendor's understanding of laboratory workload and workload distribution.

As we have seen, some commonly available datanumber of beds, CAP workload units, and number of billable procedures-are actually of secondary importance in sizing a lab computer. Their main value is providing relative comparisons among labs. Lacking better data, vendors often use these items to compare a lab with other labs where they have installed an LIS and thus extrapolate the "appropriate" system configuration. This approach often proves unsatisfactory, however, particularly if the lab in question is larger than any that the vendor has previously computerized.

Laboratories do not normally accumulate much of the information useful for predicting computer performance needs. They will have to take such steps as extracting data from section log books for typical and busy days. In some cases, special data collection procedures will have to be set up.

This additional effort will pay dividends. The availability of more pertinent data will help LIS vendors propose cost-effective system configurations appropriate to a lab's workload and work flow, with the highest assurance of good performance.
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 * END OF ARTICLE *
 * *
 * *
 * *
 * *
 *
 * END OF ARTICLE *
COPYRIGHT 1988 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Clinical Laboratory Reference
Author:Winsten, Dennis
Publication:Medical Laboratory Observer
Date:Jan 1, 1988
Words:1117
Previous Article:Using a spreadsheet for discounted cash flow analysis.
Next Article:MTs needed: sunny clime, ocean view.
Topics:


Related Articles
Why is this hospital administrator smiling?
Anatomy of a lab transition: retaking in-house control.
Developing an LIS-clinic interface.
Saving time with combined microcomputer applications.
Joining the bar code revolution.
Leveling the playing field: the economics of robotics in the hospital clinical labs.
Intranets: virtual procedure manuals for the pathology lab.
Information technology treads subtly buy powerfully through labs.
Mike Dugan and Maryam Sadri talk about their Specialty.
No more STAT testing: improve critical results management and lab efficiency through automation and data management.

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