Printer Friendly

The make-or-buy decision.

In this three-part series, the author supports extensive send-outs in a myriad of situations. This month: how to determine whether it would be wise to do that new assay in-house.

The basic laboratory tests that were done 10 or 15 years ago no longer satisfy clinicians' needs. Remarkable developments in science and computer technology have coincided to provide the practitioner with a dazzling array of laboratory tests from which to choose.

Even a single result from a highly specialized test may carry substantial diagnostic weight for the clinician working in genetics, oncology, immunology, infectious disease, therapeutic drug monitoring, forensic medicine, or endocrinology. Genetic probes and the proliferating immunoassays, with their varied applications, hint a formerly unimagined possibilities.

DNA analysis via flow cytometry can provide critical evidence concerning the malignant potential of a breast carcinoma. In prenatal screening, the gene for sickle cell anemia may be detected through polymerase chain reaction (PCC). Identifying that gene provides the key information in decisions related to the outcome of pregnancy. In every such result, the highest order of confidence is crucial.

The clinician's desire for more sophisticated laboratory procedures quickly becomes a mandate. It is then up to the laboratory director to decide which tests to perform in-house and which to order from a referral laboratory. Figure I lists questions to consider in formulating the make-or-buy decision.

* Personnel. Does your existing staff have the requisite skills? If not, can they be trained? Will you have to hire anew person with certain abilities, and if so, can you afford to do so?

Because specialized tests are frequently new and difficult, the number of skilled people available in any geographic area is limited. A referral laboratory offers a talent pool of qualified people who have licensure requirements, mandated in some states.

Many specialized tests are requested for rate disorders. For each such test, the number of requests per month in any single facility is low. IT would not be cost-effective for the laboratory to hire a full-time technologist who is highly skilled in genetic probing, for example, if that skill will be in demand for only a few hours a day. Referral labs, on the other hand, maintain a relatively high volume of exotic tests per month by serving many health care organizations simultaneously. Technologists in such labs maintain their level of skill by performing rare tests often.

New equipment is subject to frequent breakdown. Trouble-shooting may require considerable supervisory time. Can your lab spare it?

* Equipment. How much specialized equipment would you have to buy in order to perform the tests you want to offer? Will the test volume be high enough to support efficient use of that equipment? Is the instrumentation so new that you will, in effect, be engaging in unpaid research and development activities on the manufacturer's behalf?

Models that are undergoing rapid improvements will by their nature experience early obsolescence. In addition, essential equipment for specialized tests is often too expensive to permit duplication. If your new equipment breaks down, what backup resources will be available?

Will your facility have to modify or increase existing space to accommodate the equipment you have in mind? Acquiring a highpriced flow cytometer, for example, might be impractical if test volume will be low.

* Reagents. Will the reagents used in the proposed equipment be readily available? Some immunoassays, for example, require a specific antibody or purified standard that may be of limited supply and distribution.

When dealing with vendors, be selective. An antibody chosen solely on the basis of titer and cost may prove unsatisfactory if its specificity and sensitivity are inadequate.

Some cutting-edge reagents require special on-site preparation. Certain radioactive materials, for example, are not usable directly from the supplier and must be purified in the lab before being used in an assay.

* Service. Most specialized tests involve considerable setup time, including a substantial expenditure of effort for standards and quality control. Total assay time may be a matter of days rather than hours. Such considerations lead laboratory directors to perform some such tests infrequently, in batches.

Will your volume be sufficient to justify setting up the assay regularly enough to minimize turnaround time? If not, this single factor may discourage routine performance ofan assay. No one would question the wisdom of setting up a thyroxine immunoassay that tended to be ordered dozens of times a day. Not necessary so with a growth hormone (GH) determination requested four or five times a week. To meet a cost analysis requirement, even a large hospital center may have to save specimens for as long as a week before running GH assay.s Delay may extend result reporting to a degree unacceptable to the physician.

The clinical requirements of an assay, on the other hand, may dictate a specific turnaround time that supersedes cost analysis. For some procedures, such as as therapeutic drug monitoring, turnaround time longer than a few hours invalidates clinical usefulness. Since the test must be performed daily, your decision to set up the procedure would include accepting financial loss when volume was low. The larger test volume and technical staff of referral laboratories allow tests to be performed more often and with a shorter turnaround time.

Have ancillary service requirements been taken into account? In a small lab, frequent client service calls can erode technologists' time. Might you solve the problem by establishing a general laboratory client service department or by assigning one or more persons to handle phone work?

* Economics. Will performance of the new assay suit the economic framework of your laboratory? Most labs function within the real world of cash flow. Rarely can a procedure be set up that won't pay for itself. Exceptions might include a research project that depended on a particular test or onganizational pride calling for entry into a new field.

The cost analysis that characterizes most normal business dealings begins with a review of equipment and reagents. What is already on hand? What must be bought? Invisible costs--for maintenance, utilities, facility, and administration--can easily be overlooked.

How often will the assay have to be performed? On average, how many specimens will be run per assay? The cost-volume ratio of various assays differs greatly. The individual attention to each specimen that estradiol and progesterone receptor assays require, for example, limits the number of specimens in a batch. Incresing from 15 to 30 specimens per batch doubles the variable cost. On the other hand, in a immunoassay such as serum prolactin, 100 specimens or more can be run per batch. In that situation the monetary difference between 15 and 30 specimens would likely be insignificant.

Profit must be deferred for several months during setup and validation of a new test. Setup time does not end with the first set of satisfactory assay results, but continues as the assay is performed repetitively on noncommercial specimens to determine reproductibility and to uncover any procedural deficiencies. Once the entire cost analysis is complete, inhouse cost per procedure may be compared withteh fee charged by the best referral laboratory available. The comparison should be modified by intangible but important factors such as convenience and degree of confidence.

All general economies of volume are realized in working with an efficiently designed referral laboratory: bulk purchase of rare chemicals, near-capacity filling of assay batches, optima utilization of equipment, and full use of specialized personnel.

* Longevity. Is the new test a fad? It would be a mistake to dedicate space, personne, and equipment to a short-lived procedure.

Serum insulin was the first radioimmunoassay (RIA) set up in many labs. It was simple and reliable; reagents were readily available. Today, the test's everydat clinical usefulness has been reduced to the differential diagnosis of hypoglycemia. The rarity of hyperinsulinism as a cause of hypoglycemia has caused many clinical laboratories to stop doing tests for serum insulin and to send them out instead.

* Setting up. Will the end result of the new assay compensate for the difficulties of setting it up? If the new procedure is to replace an existing one, can both be done simultaneously for a while to facilitate the transition?

Developing a procedure from a recipe in a journal is seldom satisfactory or efficient. The former standard, on-the-spot observation in an established laboratory, has been replaced by a vendor's training session. Complete test systems and commercial reagent kits are highly appealing to the lab manager.

Setting up a new assay requires about three to six months. It is important to analyze cost per specimen from the beginning. Some research protocols require only a few large assays for an entire experiment. High setup costs have caused an increasing number of scientific investigators to use referral laboratories.

* Time frame. How much time is available to develop a workable assay? To answer this vital question, identify the source of the pressure to start the test. The use of a new drug by a recently established acute-care unit often prompts development of an assay. The development of a new sub-specialty department or the consensus of an existing one creates such a demand as well. Perhaps the scattered requests of a few physicians caused the clinical lab manager to consider embarking on a new assay.

Does the clinical pathology department want to update an old procedure? How quickly this must be achieved depends on the quality of the protocol that will be superseded. A lab might switch from testing progesterone with RIA to using high-pressure liquid chromatography (HPLC) in a leisurely way, since RIA is satisfactory. Sensitive immunometric assays, on the other hand, are so far superior to first-generation thyroid-stimulating hormone (TSH) RIAs as to make the latter obsolete [1]--and physicians ordering tests know this. Referral laboratories can fill the gap during a difficult transition.

* Projections. When planning equipment purchases, try to see as far down the road as possible. Will setting up an entirely new assay fit in with the developmental goals of the clinical chemistry section? If the section intends to perform a number of chemiluminometric assays over the next few years, for example, the substantial initial capital outlay for luminometers will be justified. Otherwise, perhaps not.

Let's say that an entirely new procedure will soon be available. Learning this may change your mind about setting up another assay that is excellent but becoming obsolete. In addition, when the future of a lab section seems uncertain, any test undertaken by that section should be versatile. Buying reagents that were married to non-upgradable apparatus would be unwise.

Some vendors capture their clients by selling them instruments that require unique reagents. A reagent-lease contract allowing such equipment to be acquired with little or no capital outlay but a long-term commitment to reagent purchase can prove expensive to the lab once the equipment is out of date.

If a new laboratory procedure will require drastic modification in the near future, consider the impact on cost, physician reeducation, and the reestablishment of reference values. A referral laboratory, at least in the interim, may be the answer.

In the next month's issue of MLO, the second article in this series will examine ways to monitor the quality of service offered by referral laboratories.

[1] Nicoloff, J.T., and Spencer, C.A. Clinical review 12: The use and misuse of the sensitive thyrotropin assays. J. Clin. Endocrinol. Metab. 71(3):553-558, 1990.

Jerald C. Nelson, M.D. is a professor of medicine and associate professor of pathology, Loma Linda (Calif.) University School of Medicine, and medical director for endocrinology, Nichols Institute Reference Laboratories, San Juan Capistrano, Calif.
COPYRIGHT 1991 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Using Referral Labs Efficiently, part 1
Author:Nelson, Jerald C.
Publication:Medical Laboratory Observer
Date:Jun 1, 1991
Previous Article:Lab regulators gain penalty options under CLIA sanctions.
Next Article:Assessing linearity the easy way: laboratories can comply with new Federal regulations for linearity verification by applying simple calculations and...

Related Articles
Judge rejects key IG arguments in kickback case.
My useful niche as send-out technologist; the author's research and enterprise in ordering tests from referral labs have delighted doctors and...
How to get the best service from a referral laboratory.
Lab networks: models of regional cooperation.
Assessing the quality of management decisions.
Automating laboratory send-out tasks.
Clinical information systems hotlist.
Clinical information systems hotlist.

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