| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
University of Nebraska Medical Center, Department of Pathology and Microbiology, 983135 Nebraska Medical Center, Omaha, NE 68198-3135.
2
Aultman Hospital, 2600 Sixth St. SW, Canton, OH 44710.
a Author for correspondence. Fax 402-559-6018; e-mail rmarkin{at}unmc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
To provide a broad overview, several important questions must be asked, including (a) what changes will affect delivery of laboratory results? (b) how can automation improve laboratory services? (c) how can automation decrease operating costs? and (d) how is the approach to laboratory automation changing? For the purposes of presentation and organization, the answers to these broad questions can be organized into three categories: (a) trajectory, including healthcare trends and clinical laboratory trends; (b) technology, including automation approaches and automation payback; and (c) tactics, including automation trends and the path forward for laboratorians.
|
Trajectory |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
There also appear to be changes or shifts in the diseases in our current population. Rheumatologic diseases continue to increase and, as expected, will parallel the increase in average age (2). Neurological diseases are also on the increase, as are age-related disorders (3). Autoimmune disease currently affects ~5% of the population, and it is estimated that up 50% of the population may be tested for autoimmune diseases.
Another major trend in healthcare is the continual pressure to reduce costs. Medicare and Medicaid have announced a $116 billion decrease in spending administered through the Balanced Budget Act of 1997 (4). The Balanced Budget Act contains language that will lead to a decrease in hospital reimbursement of approximately $44 billion over a 5-year period from 1998 through 2003. These decreases produced inpatient capital reductions of ~18% for the 1998 fiscal year, a decrease in the disproportionate payment system, a decrease in indirect medical education costs, and a decrease in the prospective payment of diagnosis-related groups. Our institution (Nebraska Health System) has projected a decrease of approximately $34 million, the cumulative effect of 5 years of the Balanced Budget Act modeled on a hospital system with approximately $400 million of revenue annually.
Another important healthcare trend is disease and health management. Several disease and health management companies have been created over the last several years, including HealthMagic, Inc. and Creative Health Care Management. HealthMagic, Inc., an e-health (electronic health) company, has several Internet-based products that allow the patient to maintain a lifelong history of health events, diet, exercise activity, and health history, all which can be used to assess health risks and provide input about health issues.
Focused outpatient therapies are also an important disease management trend. Asthma and allergy and diabetes appear to be the two main categories of focused outpatient therapy optimization.
Technology drivers also appear to have an important effect on global healthcare trends. Technology usually leads to substantial expenditures for health systems, and only when properly implemented does it increase efficiency.
outcomes optimization
During the past several years, outcomes optimization has been an
important focus of patient care. The concept of outcomes optimization
centers around the management of a course of patient care, either
inpatient or outpatient. The continuum of a patient’s care is
maximized for clinical benefit while striving to minimize the cost and
use of invasive treatment. As we learn more about clinical
laboratory results and incorporate them in the outcomes optimization
schemes, the laboratory will play a more pivotal role in both the
management of patients and their eventual outcomes. Laboratory results
will become the focus of managed care, health maintenance, and disease
management companies within the next 5 years.
Information technology has changed dramatically over the past 20 years. The most recent technology drivers include the Internet and Worldwide Web technology (5). The implications for laboratory automation center around the processing power and database schemes necessary to control the real-time clinical, business, and operational needs of the clinical laboratory.
Transplantation has also been an influential healthcare driver and will more than likely continue to be a driver in the future. Bone marrow transplant procedures are now being performed on an outpatient basis (6), and many other transplant programs have moved to the concept of cooperative care (7), where a small portion of the patient’s care is provided as an inpatient and the rest is provided in a step-down or outpatient environment. Advanced technology tied closely with outcomes optimization will allow transplant patients to move into the outpatient arena.
Genetic therapies and genetic testing have raised healthcare and social issues that should not be overlooked or minimized (8). The fundamental concept of genetic testing is to be able to predict the occurrence of disease, the behavior of therapeutic modalities, and outcomes. Genetic testing may lead to a shift of the uncertainty in healthcare financing. Indemnity insurance and other plans that cover a patient for a fixed fee will now statistically be able to more accurately determine that fee and shift costs away from the bulk of the population to individuals based on their "genotype".
Phenotypically targeted therapies may also have some impact on the clinical laboratory as the result of additional testing that may occur to determine whether a particular drug or pharmaceutical preparation has an optimal effect in each individual patient (9). These healthcare trends may lead to decreased revenue and decreased expense budgets for clinical laboratories as well as demands for new testing and support technologies.
clinical laboratory trends
The effect of cost containment pressures on the clinical
laboratory is shown in Fig. 1
A. In 1990, the fully burdened cost per test was approximately
$24.00. In 1995, the cost per test was reduced to approximately $16.00
and is estimated to fall below $10.00 by the year 2005. This continuous
decline in cost per test has forced major changes in the clinical
laboratory in terms of its relationship to the healthcare delivery
system and its profitability as a stand-alone operation. Large margins
are almost impossible to achieve in this financial environment. The
financial pressures are one of many issues that have catalyzed the
mergers and acquisitions within the clinical laboratory business.
|
The change in relative distribution of the dollars spent from 1991 to
1999 is highlighted in Fig. 1B
. In 1991, ~43% of each laboratory
dollar was expended for labor, 35% for equipment and reagents, and
22% for overhead (10). In 1999, data from US Bancorp Piper
Jaffray suggested that 65% of the laboratory’s dollar is spent for
labor, 15% for equipment and reagents, and 20% for overhead. This
large decline from 35% to 15% in the amount of each dollar spent for
equipment and reagents is a clear result of the healthcare economy’s
effect on in vitro diagnostic
(IVD)1
manufacturers. Over the past 9 years, there has been an
~58% decrease in the amount of money expended on reagents and
equipment, on average, by laboratories. This decrease of ~58% has
had an enormous impact on IVD manufacturers.
The relative distribution of worldwide diagnostic sales by target site
is shown in Fig. 1C
. The central laboratory and point-of-care testing
account for approximately $15 billion, the consumer market of home
testing and the other non-central laboratory diagnostics account for
approximately $3 billion, and the blood banking industry accounts for
approximately $775 million. This relative distribution of dollars is
important because it shows that the bulk of the laboratory testing
still remains within the healthcare delivery system and is probably
subject to the application of some form of laboratory automation.
The projected sales for IVD manufacturers from 1997 through 2001 is
shown in Fig. 1D
. Sales for 1999 are estimated to reach approximately
$20 billion, with sales in the year 2001 estimated to reach
approximately $22 million.
|
Technology |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
The philosophy of automation system design rests with the inherent understanding of the designer. Implementation of strictly mechanical concepts into the clinical laboratory may override the overall mission of the clinical laboratory and its integral involvement in the delivery of patient care. To develop a philosophy, we believe that the following should be understood: (a) how the laboratory relates to healthcare; (b) the process of the clinical laboratory; and (c) the business of the clinical laboratory. From a structural standpoint, either software or hardware can be made the primary focus of an automation system. As with the early development of information technology and other similar advances, hardware technology has had a prominent place in early automation systems designs. It is our contention that the design of a LAS should be centered on the patient, with a software design that allows patient-related information and laboratory process to be under the control (direction) of the software. The hardware then serves the function of appendages or end-actuators similar to the application of technology in a parallel environment: computer-integrated manufacturing.
Process control software requires several important components and functionalities, including the following: (a) a basis in modern information technology, which requires hardware and operating systems that are vertically upgradable; (b) transportation system management at both the local (device) and overall system levels; (c) specimen container tracking so that any specimen can be identified in its physical location on or in the automation system; (d) repeat testing so that a specimen that may yield a certain result can be rerouted using the rules embedded in the software to repeat the test on another instrument using a different methodology or to confirm the test on the same or another instrument; (e) reflex testing where an additional test can be performed at the same workcell/instrument or a specimen can be trafficked onto another workcell/instrument for subsequent testing that is the result of applying a rule against the result of the first test; and (f) information systems integration so that LISs and other information components of IVD equipment (analyzers) can be combined to make a functional automated laboratory where the instrument can be managed using rules and other software-driven parameters to replace the technologist at the individual instrument. For example, the system software would "know" through the information passed by the hospital or LIS that the patient with a high urea value is from the dialysis unit and that the test does not need to be repeated. A rule can provide the functionality necessary to make the determination.
There are several important dependencies between software and hardware. If the software functionality is absent, the hardware cannot be expected to perform. Similarly, if the hardware functionality is absent, the software cannot be expected to actuate that hardware function; hence, the hardware and the software functionality are interdependent. It has become clear that to allow random access, one must have a single tube per carrier design so that each specimen has individual real-time access to any workcell or device in the LAS. To allow reflex testing, there must be real-time control of hardware devices and instruments by the software that manages the overall operation; to allow routing, there must be more than one transportation path to move a specimen to one or many instruments.
Several software systems now include functionality for both the procedure and the process. At the procedural level, rules can be applied that allow only the performance of specific tests on an identified matrix, e.g., only perform sodium on serum or plasma, only perform a complete blood count on EDTA-treated or heparinized whole blood. The rules processing in the software component of an automation system should provide the following functionality: (a) the ability to monitor quality using the process control system; (b) the ability to monitor results; (c) the ability to monitor the instrument and its operation; (d) the ability to implement repeat testing decisions; (e) the ability to implement a reflex testing decision; (f) the ability to cancel tests; and (g) the ability to manage the workload of the entire laboratory operation based on the need for turnaround time (TAT), throughput, instrument utilization, and instrument uptime.
The ability to interface between the LIS and LASs has been enhanced by the implementation of Health Level 7 system-to-system interfaces. The NCCLS has issued a proposed standard (AUTO 3P) that specifies the Health Level 7 interface as the system-to-system communications methodology for connecting a LIS and a LAS.
The management of the instruments in a clinical LAS environment requires the implementation of an instrument control software module with instrument-specific rules. In concept, this module is simply replacing the intelligent operator with embedded rules in the current nonautomated environment (the medical technologist) with an automation control system with embedded rules that will allow a predefined level of uninterrupted or controlled operation before human intervention.
technology-automation payback
One of the most important questions in the implementation of
clinical laboratory automation is "how can automation decrease
operating costs?". Few if any articles or other publications
scientifically document the justification of capital expenditure
(payback) or return on investment with respect to implementation of
clinical laboratory automation.
The implementation of laboratory automation in Japan is well documented (11) and consists of >100 distinct operating sites installed over a 17- to 20-year time period. The functionality of those systems implemented in Japan is well documented; however, the cost-effectiveness, payback, or return on investment has not been well documented. The relative paucity of operating clinical laboratory automation sites in North America or Europe is one of the main reasons that we lack payback data.
To obtain statistically significant results, a large number of operating sites would be required as a basis for analysis. It is our estimation that to obtain statistically significant data with respect to automation efficiency and payback, we will need the following characteristics: (a) between 10 and 25 operating clinical laboratory automation sites for each system evaluated; (b) that these sites be in continuous operation (excluding down time) for 2–3 years; and (c) that they be similar in operating characteristics. We would then have predetermined characteristics, e.g., TAT, that we would periodically measure, including baseline measurements made before implementation of automation technology and operational measurements made at 1-, 2-, and 3-year intervals after the implementation of automation technology.
The implementations of automation in North America have utilized the introduction of automation as a mechanism to force or expedite laboratory redesign. In some of these automation installations, it is difficult to differentiate the effects of implementing LASs from the effects of redesign.
Aultman Hospital is one of the few sites, if not the only site with implemented clinical laboratory automation that has performed measurements before and after the implementation of automation systems. Aultman Hospital implemented a LAB-InterLink automation system in February 1997. Before the implementation of the system (1996), measurement of operating metrics, including full-time equivalents (FTE); supplies, including disposables; testing capacity; errors; and TAT.
The most important impact of the combined laboratory reorganization and
automation implementation was the reduction in FTE (See Table 1
). From 1996 to the second measurement in 1998, there had been a
reduction of 35 FTE, representing $1.2 million reduction in labor
expenses per year. The components of the labor reduction included the
following categories and FTE savings: STAT laboratory consolidation,
six technologists; consolidation of like processes, eight
technologists; implementation of pneumatic tube system, three
entry-level FTE; robotic processes, eight technologists; implementation
of outreach testing processes, three entry-level FTE; and management
efficiencies, four management FTE (see Fig. 2
).
|
|
The unit costs of producing a clinical laboratory result in chemistry
decreased from $2.25 per requisition in 1996 to $1.45 per requisition
in 1998. The cost of chemistry reagents decreased from $1.65 per test
in 1996 to $1.50 per test in 1998. The capacity of the laboratory
increased 40% in the same time period, e.g., 40% more volume could be
handled by the laboratory without the addition of personnel (Fig. 3
).
|
The average TAT for urea nitrogen determinations between the hours of
0500 and 0700 decreased from 62 min in 1996 to a uniform 40-min TAT in
1998 (see Fig. 4
). The medical staff at Aultman has learned that the TATs are
reliable and reproducible following the implementation of automation
and that STAT testing is not misused as a method to decrease TAT. The
changes in TAT are directly attributed to the process control and
process management attributes of the automation system.
|
The error rate of for chemistry and hematology was substantially
reduced in the period from 1996 to 1998 (see Fig. 5
). The reduction in errors may also be attributed to the
implementation of automation and the uniformity that is part of the
"standardization" of the process internal to the laboratory
operation.
|
The payback of the laboratory reorganization and automation implementation project at Aultman Hospital was anticipated to be 2.5 years. The payback (accumulated cost savings equaling the cost of the reorganization and automation implementation) was 2.5 years. Aultman management would have accepted a 5-year payback time. The process by which the 2.5-year payback occurred included beginning the attrition process before the implementation of the automation system.
|
Tactics |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
The LAS evolutionary pathway is, in many ways, similar to the history of LIS development. Many of the big companies that entered the market and were considered to be "giants" in their industry, such as IBM, Honeywell, 3 M, and others, made early entries into the LIS marketplace. These big companies, similar to the IVD giants of today, did not understand the fundamentals of the business that they were entering. In the case of the LIS, the issue was information flow in the laboratory and the disparate sections of the laboratory, e.g., chemistry, hematology, microbiology, and anatomic pathology. In the case of LAS, the issues centered around process control and the relationships of the patient, the specimen(s), the "real-time" availability of the instruments, and the in-flight control of the entire laboratory as a "system", a virtual symphony with the LAS software as the conductor.
Laboratory automation may parallel the pattern of the LIS industry, where the large firms are no longer in the LIS business and the small entrepreneurial and focused companies, including Sunquest, Cerner, and Antrim, are now the leaders in the LIS industry. And if the parallel to the LIS industry holds, small focused laboratory automation companies such as LAB-InterLink and LABOTIX may become the laboratory automation providers of the future.
The market for clinical laboratory automation products is primarily the
mid-sized hospital that provides laboratory service to both a hospital
and outpatients. The trends that are appearing on the horizon are
listed in Table 2
and include the movement from total laboratory automation
approaches to the implementation of modular automation technologies;
the movement from hardware-driven automation technologies to
software-driven process control systems; the movement from
one-of-a-kind systems to standardization of the technology; and the
movement from an IVD novelty to a substantial marketing tool and from
"toy" status to a laboratory tool.
|
a path forward
The path forward is becoming apparent as the industry begins to
embrace automation technology. The introduction of information
systems-based automation technology that parallels the manufacturing
industry’s automation paradigm for computer-integrated manufacturing
is becoming the standard. The NCCLS has produced five (5)
separate but interrelated proposed standards documents that center on
the information technology (AUTO 3) and the relationships between
information and the hardware components that operate within the system.
The NCCLS standards provide the framework for the development of
purposeful automation technology supported by relational database
technology with remote operations and single-user interfaces for the
operation of the system in either a local or remote mechanism.
The hardware technology is also evolving to include workcells with integrated information, smaller footprints, flexibility, expandability, and low maintenance requirements. NCCLS standards also provide the high-level framework for the development of technology and the interconnectability of a variety of platforms.
|
Conclusions |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
LASs are evolving and now have the ability to improve services to the patient, physicians, and other healthcare providers by improving TAT and providing predictable throughput. Information-based LASs can provide support for care models that include the ability to provide reflex testing, repeat testing, test cancellation, and reduce financial risk for healthcare organizations. The IVD manufacturers have and may continue to use laboratory automation as a marketing tool, whereas laboratorians will likely continue to use laboratory automation as a reorganization tool.
Aultman Hospital provided before- and after-automation implementation measurements of key metrics, including TAT, error rates, reagent and labor costs, and payback. All of the metrics show improvement in the laboratory operations at Aultman with a payback of 2.5 years.
In the future, we as laboratorians will need data defining the operational characteristics of LASs. At this point, the data are lacking only as a result of the lack of implemented automation systems in North America. We estimate that we will need 10–25 installed operational automation sites with at least 2 years of operating data to provide data that will have statistical significance. Until we have these data, laboratorians will be required to take a leap of faith in the implementation of LASs, perhaps based on the successful implementation of other systems in other environments such as nonmedical industries.
|
Footnotes |
|---|
1 Nonstandard abbreviations: IVD, in vitro diagnostic; LIS, laboratory information system; LAS, laboratory automation system; TAT, turnaround time; and FTE, full-time equivalent(s). 
|
References |
|---|
|
TopAbstractIntroductionTrajectoryTechnologyTacticsConclusionsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and
wages (
) before (1996) and after
(1997–1999) implementation of automation.
