Printer Friendly

A lab program to do bedside glucoses.

A lab program to do bedside glucoses

"Why can't I get a simple glucose test done on time?' the irritated doctor wanted to know. It didn't help to tell him about a rising demand for timed tests--150 to 180 at 6 a.m. alone. His treatment of diabetics and the appropriate insulin dosage depended on quick turnaround of accurate glucose results. An hour was too long to wait.

Delays occurred despite the fact that our 700-bed hospital laboratory has a highly effective and productive rapid response section for emergency testing in chemistry. The problem was that we were heavily outnumbered. There were nearly 500 staff physicians plus several hundred residents, and many of their special requests carried the same critical times.

The nurses lent a hand on glucoses, using dipsticks. Unfortunately, the methods and equipment varied from floor to floor, and quality control was lacking. If results were out of range, they ordered Stat glucoses, and at times that might strain our Stat capabilities and delay results.

"What would help?' the administrative technologist asked the clinician who had complained about glucose turnaround time. He plainly wasn't buying explanations.

"Why don't you perform the glucoses at bedside?' he wanted to know. "Patients do them at home, and the nurses use a dipstick on the floor. The laboratory could provide this service for both patients and nursing.'

"It's worth a try,' the administrative technologist said after a moment's thought. The physician, expecting yet another set of excuses, was stunned but delighted. He and the technologist began developing a plan. About this time, the head nurse of the hospital's busiest medical floor got caught up in the enthusiasm and offered her floor for a pilot project. The three new partners agreed to work on the mechanics in their respective areas and to meet within a week to finalize the trial program. Thus a physician's question, a technologist's open mind, and a nurse's willingness to help launched an exciting cooperative venture.

Semiquantitative glucose testing, conducted by phlebotomists with dipsticks and portable glucose reflectance meters, began on a single medical floor in October 1983 amid some skepticism from the medical staff. The phlebotomists performed more than 100 bedside readings that month and grew progressively busier thereafter. By November, clinicians were transferring some patients to this floor just for the bedside tests. When the program went hospitalwide in January 1984, average monthly volume rose to nearly 1,000 tests and kept climbing.

The program has cut turnaround time in half on follow-up quantitative glucose assays in the laboratory. We are now confident that requests for these assays are genuine Stats, in response to unexpected or out-of-range results, and we always treat them as such. The program has also trimmed 1.3 days off the average patient stay for diabetic ketoacidosis. Complaints are down substantially, and we have laid the foundation for a solid working relationship with nurses and physicians. As a result, lab morale has risen, especially among phlebotomists.

Several analyzers will determine a diabetic patient's glucose level from a single drop of whole blood. These instruments had long been used in critical care areas, at nursing stations, in physicians' offices, and by diabetics themselves for home monitoring.

What then was new and improved about our approach? Direct laboratory involvement, among other things:

The laboratory's business is providing accurate, precise values of such analytes as glucose. Centralized processing of requests and standardized testing would enhance efficiency and possibly minimize Stat abuse.

The bedside glucose analyses would be performed by properly trained personnel accustomed to the critical need for running controls and calibrating and maintaining instruments. They understood the importance of accuracy and precision and the need for correlations between an analyte's semiquantitative and quantiative values.

The test required only one venipuncture or fingerstick. If necessary, phlebotomy for this purpose could be combined with requests for other laboratory tests to reduce patient sticks.

If the pilot study demonstrated a good correlation between a patient's quantitative and semiquantitative glucose values, the laboratory would gain confidence in the bedside reading as a precise and accurate measure of blood glucose. That would mean availability of immediate results versus the delay built into quantitative glucose testing.

The program marked a joint effort on the part of a concerned physician, endocrinologist Stanley Von Hofe, M.D.; an energetic and forward-thinking head nurse, Laverne Smoke, RN; and a laboratory staff willing to try someting new--myself.

To start with, the laboratory invested $600 in four glucose reflectance meters and trained the phlebotomists on them. Although the test employs a fairly simple procedure, designed for home use, the training ran nearly one week. The lab's special functions section took responsibility for training.

They demonstrated the technique, gave phlebotomists a tube of blood, told them to run 20 consecutive tests, and then determined their coefficients of variation. Each phlebotomist had to achieve a coefficient of variation below 5 per cent, on 20 consecutive tests with the same specimen, before going onto the floor. Sometimes phlebotomists had to run as many as 50 tests in order to check out. A year later, we switched to a reflectance meter that was less dependent on the operator's technique.

While the phlebotomists practiced, the physician-nurse-laboratory team set up ordering procedures and the mechanism for determining the correlation between quantitative and semiquantitative results on the same specimen. This led to a new form (Figure I) used to report semiquantitative results and also to request follow-up quantitative testing as needed.

All specimens with semiquantitative results below 70 or above 300 mg/dl automatically go to the lab for Stat testing, but patients are not billed the usual Stat surcharge. Quantitative results that vary from bedside findings by amounts specified in the form are phoned to the physician or ward secretary. There is space on the form to document the call.

It took 10 days to get all the systems in place. During the first month of the study, phlebotomists drew 114 specimens for a paired data comparison of bedside and laboratory results.

We initially found that semiquantitative and quantitative results did not correlate well at glucose levels greater than 250 mg/dl. This still appeared to be the case after we had an additional 173 pairs of data, representing the November workload. Later, however, we changed the upper limit of usefulness of semiquantitative values to 300 mg/dl.

Bedside testing was reserved at first for insulin-dependent patients and performed only on the day shift. Dr. Von Hofe urged physicians with these patients to give semiquantitative glucose testing a try.

In an effort to win acceptance of the program and expand it to other floors, I addressed clinicians from the department of internal medicine on three occasions. Dr. Von Hofe stood by to answer questions.

Laverne Smoke, our team's mursing representative, was a gentle persuader. The laboratory nicknamed her saleslady of the year as she got physician after physician to put Stat requests on hold and try the semiquantitative option.

Bedside glucose testing expanded to a second major medical floor and to the intensive care unit in December 1983. It also became available on all three shifts. One month later, the program went hospitalwide. As Figure II shows, utilization rose at a steady pace throughout 1984 and reached a record high of 3,571 tests last March. Even now, it's not unusual to see requests jump by 50 per cent from one month to the next.

Phlebotomists have managed to handle the testing with only an increase of 3.5 FTEs, which was achieved through overtime and more use of part-timers. We have 18 phlebotomy FTEs on the first shift, 12 on the second, and 3 on the third. They are proud to be frontline providers of a popular service, and their productivity shows it.

Three laboratory areas were involved in establishing the program. The phlebotomists work out of the laboratory's specimen center. Chemistry technologists in the rapid response section process the follow-up quantitative tests. And a "heme' team--which is made up of selected phlebotomists, MLTs, medical technologists, and a supervisor--handles the heavy early morning collections, from 5:30 to 8:30 a.m. Specimen center techs then make routine sweeps throughout the hospital during the day.

The hospital's quality assurance coordinator conducted a patient stay audit to assess the impact of semiquantitative glucose analysis in insulin regulation for patients admitted with a primary diagnosis of diabetic ketoacidosis. The study indicated that the average length of stay for DKA patients dropped from 7.5 to 6.2 days after bedside testing was implemented.

While bedside readings aren't as reliable as we would like, especially in the upper ranges, Dr. Von Hofe explains that prompt turnaround is often more important than absolute accuracy in diabetic ketoacidosis cases. When a patient has had a glucose level in the 1500s, the clinician's main concern is whether the next reading after an insulin dose is still too high, not whether the glucose level is exactly 450 mg/dl. The semiquantitative test provides enough information to adjust therapy pending laboratory confirmation.

Not only is everyone in our hospital pleased with the bedside testing program, but hospitals as far as 100 miles away now offer the same option. That's due to our physicians, who demand the service wherever they practice.

A little bit of cooperation and communication proved to be more powerful than any of us thought possible. Watching something so simple and logical grow into such a popular service--and all because of our efforts--is probably the most exciting experience I have had as a technologist.

Photo: Figure I Bedside glucose result and request for follow-up lab assay

Photo: Figure II Rising volume of bedside glucose testing
COPYRIGHT 1986 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1986 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Eason, Melba H.
Publication:Medical Laboratory Observer
Date:Jan 1, 1986
Previous Article:Ten-hour shifts solved our turnover problem.
Next Article:Japanese QC techniques merit our attention.

Related Articles
Quality control in the new environment: lab testing near the patient.
The technologist's role in quality management of off-site testing.
Protocols for bedside testing.
Protocols for bedside testing.
Protocols for bedside testing.
Monitoring versus testing technologies: present and future.
Managing information from bedside testing.
A pain-free inauguration of point-of-care testing.
Point-of-care testing in pediatric hospitals.
How should a lab design a fail-safe system for point-of-care testing? (Liability and the lab).

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters