guest commentary - Journal of Clinical Microbiology - American ...

2 downloads 0 Views 78KB Size Report
A recent survey by the American. Society for Microbiology determined that Ph.D. scientists' sal- aries range from $55,000 to $110,000. Not surprisingly, full-.
JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 1999, p. 883–889 0095-1137/99/$04.0010 Copyright © 1999, American Society for Microbiology. All Rights Reserved.

Vol. 37, No. 4

GUEST COMMENTARY Controversies Affecting the Future Practice of Clinical Microbiology ANN ROBINSON,1* MARIO MARCON,2 JOEL E. MORTENSEN,3 YVETTE S. MCCARTER,1 MARK LAROCCO,4 LANCE R. PETERSON,5 AND RICHARD B. THOMSON, JR.6 Division of Microbiology, Department of Pathology and Laboratory Medicine, Hartford Hospital, Hartford, Connecticut 06102-50371; Department of Laboratory Medicine, Children’s Hospital, Columbus, Ohio 432052; Department of Laboratory Medicine, St. Christopher’s Hospital for Children, Philadelphia, Pennsylvania 191343; Department of Pathology, St. Luke’s Episcopal Hospital, Houston, Texas 77225-02694; Department of Pathology, Northwestern Memorial Hospital, Chicago, Illinois 606115; and Department of Pathology and Laboratory Medicine, Evanston Hospital, Evanston, Illinois 60201-17836 tumor markers, microbial antigens, and antibodies; (ii) random-access instruments with a high-throughput capacity and computer linkage for specimen tracking and data reporting; (iii) an open-space design which encourages communication and the sharing of responsibilities among technologists; (iv) technologists who are cross-trained among disciplines and instruments; and (v) a 24-h-a-day, 7-day-a-week operation that emphasizes rapid turnaround, with batch testing minimized or eliminated whenever possible. The large commercial laboratories represent the largest, most automated laboratory businesses in the country (35). Although most have combined the high-volume, highly automated sections of hematology and chemistry into one analyzer-driven section, each of these companies has maintained microbiology as a separate section or laboratory. With proper design and operation, the integrated laboratory incorporates a significant portion of the total laboratory testing volume with a particular emphasis on basic, high-volume tests that require a “stat” or at least a same-day response. For this reason, the integrated laboratory is often referred to as the “core” or rapid-response laboratory, as shown in Fig. 1. In this example, the core laboratory consists of an analyzer for a large number of tests that would have been processed traditionally in chemistry, hematology, and serology. Separate but related laboratory sections include transfusion medicine, which encompasses both the donor center and the blood bank; client services, with informatics, phlebotomy, and information resource personnel; infectious diseases, which includes bacteriology, parasitology, mycology, mycobacteriology, virology, and nonautomated viral and bacteriological serology assays; and special testing, which is used for specialized coagulation studies and highly complex chemistry assays. In virtually every laboratory reorganized around this or similar models, automated sections are integrated and microbiology is a separate section or laboratory. Sometimes urinalysis and other hands-on or highly complex procedures are added to microbiology. Although the testing instruments are the focus of the integrated laboratory, pre- and postanalytical activities are also streamlined with the use of computers, bar coding, and robotics to speed up testing and throughput. The resulting laboratory presumably operates at a lower cost through the efficient use of personnel and automation while it provides rapid, quality testing (10). Further cost reduction and improved medical outcome may occur as a result of the rapid information reporting that contributes to earlier intervention and improved patient management via an integrated information network. It is unfortunate that general laboratory integration has of-

SHOULD AN INTEGRATED LABORATORY REPLACE THE CLINICAL MICROBIOLOGY LABORATORY? Health care delivery systems in the United States are undergoing dramatic changes in response to continued pressure to limit the utilization and cost of medical services. A major force in this process has been the shift in reimbursement to providers from fee-for-service arrangements to managed care contracts (13). Among the most visible changes are (i) the consolidation of hospitals, via mergers, acquisitions, and alliances, and the formation of regional networks (13), (ii) the emergence of national, for-profit hospital corporations (26), and (iii) the continued reduction of hospital length of stay with a shift to outpatient medical care. These changes have had an impact on the operation of clinical laboratories in general and hospitalbased laboratories in particular (26). Hospital consolidations often result in the elimination of many or all redundant laboratory services performed on site, with the exception of tests that require rapid turnaround times. Specimens for tests not requiring rapid turnaround times are transported to a central laboratory, where an economy of scale is realized. For those hospital-based laboratories not undergoing consolidation, the achievement of cost reductions with the simultaneous maintenance of quality and customer satisfaction is increasingly difficult (27). Many laboratories have reorganized or reengineered the work flow and division of labor on the basis of the required test turnaround time, i.e., rapid-response versus non-rapid-response tests, rather than the traditional work flow organization based on subspecialty testing such as chemistry, hematology, and microbiology (8, 38). Reengineering is a concept adapted from manufacturing. The integration of testing disciplines may improve the efficiency of personnel and may thus decrease overall operating costs (11, 14). Data that support the hypothesis that a massive reorganization will have a significant favorable impact on cost are minimal to nonexistent. Integrated clinical laboratories are organized primarily on the basis of processes and technologies rather than the traditional scientific disciplines (42); thus, they have the following operational features: (i) highly automated instrumentation capable of testing multiple classes of analytes, including therapeutic drugs and drugs of abuse, hormones, serum proteins, * Corresponding author. Mailing address: Microbiology, Department of Laboratory Medicine, Sacred Heart Medical Center, 101 West Eighth Ave., Spokane, WA 99220-2555. Phone: (509) 626-4426. Fax: (509) 455-2052. 883

884

GUEST COMMENTARY

FIG. 1. Model of a reengineered laboratory service.

ten proceeded in the absence of significant participation by the traditional clinical microbiology service. To be effective, reengineering must span traditional administrative lines (36). Microbiology should be a part of the integration effort whenever possible. However, not all microbiological testing is easily amendable to integration. Those tests that are performed manually and that have long turnaround times, such as cultures, fall into this category. However, some microbiology tests or test components could be successfully integrated. More complete integration of microbiology services is contingent on future advances in technology and automation. The advent of probeand amplification-based methods suggests that the future is close at hand. At present, blood cultures may be performed with an instrument that continuously incubates and monitors bottles for growth. Activities performed in the integrated laboratory might include accessioning and loading of bottles, performance of Gram stain smears and subcultures of positive bottles, calling of preliminary smear results, and setting up of rapid identification and susceptibility tests. Ideally, these activities should occur 24 h a day. Further isolate workup would be performed “off-line” in the traditional microbiology laboratory setting. Such an effort requires a significant commitment to the training and monitoring of technologists who may not be highly skilled in smear interpretation. Specimens for urine culture could be appropriately screened with a combination of automated urinalysis and bacterial load test instruments. Only the urine specimens that met the criteria for culture would be processed, and the workup of positive cultures would be performed off-line. The predictive value of urine screening is method dependent. Also, urine screening is somewhat controversial and can add to the laboratory’s costs. As an alternative, traditional plating of urine specimens for culture might be performed in the integrated laboratory. Throat swabs for detection of Streptococcus pyogenes could be screened by a rapid antigen test, and only the specimens that tested negative would be cultured and worked up off-line. Alternatively, all specimens could be tested by a commercially available, highly sensitive, nucleic acid probe-based test, and culture could be eliminated altogether. Appropriate specimens from the urogenital tract could be tested for Neisseria gonorrhoeae and Chlamydia trachomatis by a commercially available DNA amplification test. Specimens need to be batch tested at present, but improvements in instrumentation may permit real-time testing. Stool specimens for ovum and parasite examination could be

J. CLIN. MICROBIOL.

screened in the integrated laboratory by an antigen test for common pathogens such as Giardia lamblia and Cryptosporidium parvum. A microscopic examination would be performed in the traditional laboratory only when the clinical presentation or history warrants such a test. A variety of specimen types submitted for viral antigen tests, including blood, respiratory secretions, and stool, could be tested by automated or nonautomated methods. Many tests for antibodies to microbial agents could be performed with serum by using presently available instruments. For mycology, mycobacteriology, and other testing that cannot be readily integrated for the reasons given above or because of safety issues, a traditional clinical microbiology laboratory must remain on site or arrangements must be made to transport specimens for such testing to a reference laboratory. Integration of the clinical microbiology laboratory will be an evolutionary process and will be driven not only by technology and automation but also by the creativity and resourcefulness of professional microbiologists. The setup and processing of specimens and the interpretation of results are fundamentally different in the microbiology laboratory than in the automated core laboratory and represent significantly more interpretive tasks. Recognizing rather than ignoring these differences is important to a successful integration process. An important potential adverse consequence of integrating a microbiology laboratory into a core laboratory is a loss of quality as a result of a decrease in the proficiency of the resultant large group of generalist technicians and technologists. The maintenance of a reasonable level of proficiency would require a significant level of training and ongoing education for a group of nonmicrobiology specialist technologists. Practical decisions about what and how to integrate must be made not only with cost reductions in mind but also with the goal of improving the process of delivering timely, accurate laboratory reports to improve patient management. To achieve this objective, it may be preferable to redirect rather than reduce some personnel. Integration of microbiology testing may augur the end of the traditional, hospital-based, clinical microbiology laboratory as a physical entity; however, it will not eliminate the need for medical technologists with speciality training in microbiology and clinical microbiologists. Such professionals must participate in the decision-making process when questions regarding the integration of services are considered. In addition, the traditional role of microbiologists in teaching, test and instrument evaluation and implementation, and consultation on test selection, utilization, and interpretation will be even more essential. The integration of clinical microbiology services will be successful only with the full participation of professional microbiologists. IS POINT-OF-CARE TESTING A COST-EFFECTIVE, CLINICALLY RELEVANT ALTERNATIVE TO CENTRALIZED MICROBIOLOGY LABORATORY TESTING? The current limited scope of point-of-care testing in most health care facilities creates a tendency to discount its impact on the clinical microbiology laboratory. Although the complexity of microbiological testing is incompatible with the current decentralized testing environments, it would be shortsighted to believe that the evolution of point-of-care testing in clinical microbiology will remain stunted by technological limitations. Advances in non-culture-dependent methods of microorganism identification, coupled with automated instrumentation, may permit more sophisticated testing to be performed at sites other than the clinical microbiology laboratory. An assessment

VOL. 37, 1999

of the future of point-of-care testing in clinical microbiology requires a critical appraisal of the present status of point-ofcare testing in laboratory medicine. In an attempt to clarify terminology, Handorf (23) described point-of-care testing as one of several overlapping domains of decentralized laboratory testing. The movement toward decentralized testing began with a perception, mostly on the part of physicians, that decreasing the turnaround times associated with laboratory testing would improve the quality of patient care. The availability of more real-time laboratory data would more readily contribute to clinical decision making. The movement accelerated in the 1980s largely because of the anticipated repercussions of diagnosis-related groups on laboratory services. Fixed reimbursement by diagnosis would transform the hospital laboratory from a revenue-generating enterprise to a major cost center. The expected outcome was that less inpatient testing would be performed and that much of this service would be transferred to physician office laboratories and other outpatient venues. The Clinical Laboratory Improvement Amendments of 1988, however, slowed the growth of physician office laboratory testing, since there was a reluctance to adhere to regulatory requirements that were previously nonexistent. With technological strategies in danger of losing their position in the marketplace, vendors began to target hospital sites outside of the central laboratory for the placement of compact portable or transportable instrumentation, including operating rooms, intensive care units, emergency departments, and other critical care areas. Although an explosion in point-of-care testing has occurred in recent years, this type of testing is not new. The clinical ward laboratory was a standard not that long ago. However, increasing test complexity, automation, and quality control considerations relocated testing into centralized laboratories. Recently, technical advances in instrumentation and the desire for a more rapid turnaround time has shifted some testing back to the point of care. In order to understand why point-of-care testing might provide a cost-effective, clinically relevant alternative to centralized laboratory testing, the differences between the test life cycle in a centralized laboratory versus that in a point-of-care system must be recognized (44). With a laboratory-based testing system, the test is ordered; the request is processed; the specimen is collected, transported to the laboratory, processed, and analyzed; and the results are reviewed by the laboratory staff and reported to the clinician. In contrast, in a point-of-care system, the test is ordered, the specimen is collected and analyzed, and the clinician acts on the result. Thus, the life cycle is greatly truncated, resulting in a decreased turnaround time that allows the clinician to act on the test result more quickly. The decreased turnaround time is the result of the elimination of specimen processing and transport. Specimen transport is a major factor in test turnaround time (32). The greater the length of transport time is then the longer the overall test turnaround time will be. Salem and coworkers (39) demonstrated that specimen transport and processing represented from 73 to 85% of the total test turnaround time for routine chemistry and hematology tests. Although a centralized laboratory can implement changes to reduce specimen transport and processing times, such as a pneumatic tube system or a stat laboratory, the laboratory response from specimen receipt to analysis can theoretically approach, but never equal, that of point-of-care testing (39). Decreased test turnaround time can potentially provide important clinical benefits. More timely test results that more closely approximate the patient’s current condition permit clinicians to make evidenced-based medical decisions in real time (15, 31). This optimizes therapeutic decisions and provides for

GUEST COMMENTARY

885

the immediate treatment of patients with abnormal results and the patient’s disposition (29). Decreased turnaround time also minimizes additional, unnecessary tests and contributes to a concomitant reduction in unnecessary empiric medications administered to patients while they are awaiting laboratory test results (30). This testing also may provide benefits deemed important from the patient’s perspective, including reduced waiting time, greater convenience, and immediate treatment if necessary. The diagnosis of group A streptococcal pharyngitis is the best example of the potential benefits associated with point-of-care microbiology tests. Several studies have demonstrated the cost and clinical benefits associated with this type of point-of-care testing in a variety of venues. Wiedermann and colleagues (47) studied the impact of point-of-care testing in a pediatric setting. More than 2,400 patients with suspected streptococcal pharyngitis were evaluated by a rapid latex test at the point of care versus laboratory culture. They concluded that the rapid test results were available while the patient was still on site, usually in less than 20 min. This permitted physicians to provide same-day treatment for 90% of the patients with streptococcal pharyngitis, thus reducing the duration of symptoms and preventing post-streptococcal pharyngitis sequelae. DuBois and colleagues (18) evaluated the impact of point-of-care testing in an emergency department setting where patient follow-up after throat culture is often inadequate and patients may be treated inappropriately (18). The clinical judgment of physicians was compared to rapid testing for the diagnosis of pharyngitis, and it was determined that the rapid test was significantly superior to clinical impression alone in determining the presence of disease. In addition, point-of-care testing in this situation reduced the problems and costs associated with empiric therapy and patient compliance with follow-up care. The availability of rapid test results can also affect physician prescription patterns. True and colleagues (46) reported that prescription patterns changed when a rapid test for group A streptococcal pharyngitis was used at the point of care. Physicians were more likely to initiate therapy in response to a positive rapid test result and to wait for culture results before initiating therapy as a result of a negative test result. Thus, inappropriate antibiotic usage, unnecessary cost, and potentially harmful antibiotic exposure were reduced. Assessment of the cost-effectiveness of point-of-care testing requires an examination beyond test cost comparisons. A total economics perspective must be used to determine the costeffectiveness of point-of-care testing in the context of the total cost of patient care (7). Cost benefits associated with point-ofcare testing are the result of decreased test turnaround time. The improved turnaround time may lead to a reduction in duplicate test orders, additional supplementary test requests, and the premature or unnecessary consumption of other expensive ancillary services and pharmaceuticals which may be averted or postponed to a more appropriate stage of care. Additional cost benefits may be derived from the decreased length of stay associated with point-of-care testing. Despite the intuitive appeal of patient-centered care, the arguments for point-of-care testing that stress rapid turnaround time and improved clinical effectiveness have been minimally validated by experience. Goodwin (20) reported that, after implementing point-of-care testing in a postanesthesia care unit, test turnaround time was reduced from an average of 26 min to 2 min and patient length of stay was decreased by an average of 18 min. In a unit that bills patients in intervals of 15 min, this resulted in an average documented cost savings of $64 to either the patient or the third-party payer. In addition, Zaloga (49) has reported on point-of-care blood glucose level monitoring for diabetic ketoacidosis in the intensive care unit. Patients

886

GUEST COMMENTARY

receiving bedside testing had a reduced hospitalization cost of approximately $1,600 compared to the cost for patients with access only to laboratory-based glucose testing. These savings were a direct result of point-of-care testing. In general, there is a paucity of evidence that measures the impact of point-of-care testing on patient length of stay or the costs-benefits to the institution. A 1993 survey by Bickford (12) attempted to quantify various aspects of decentralized testing in 241 U.S. hospitals. The majority of respondents from institutions that performed decentralized testing were from hospitals with more than 200 beds, and 51% of all institutions reported an increase in decentralized testing since 1990. Despite this trend, only 2% of the respondents reported that their hospitals had performed quantitative studies to determine the impact of decentralized testing on patient length of stay, and only 7% were aware that any cost studies had been performed by the hospital to justify the purchase, quality control, calibration, and general maintenance of decentralized testing devices. The survey revealed that the demand for decentralized testing was strongest from physicians. As expected, perceptions regarding the advantages of decentralized testing included decreased turnaround time, increased patient satisfaction, decreased length of stay, and lower costs. When queried about the perceived disadvantages of decentralized testing, respondents highlighted test accuracy, precision, quality control and calibration, training, and personnel requirements. Although the performance of decentralized testing was favorably rated by most respondents, 1 of every 10 reported “not good” or “poor” performance, and up to 20% of the respondents chose not to respond to this query. The results of this survey underscore the need for careful scrutiny of the rationale for point-of-care testing before moving diagnostic microbiological testing out of the central laboratory. Bachner (9) has proposed five general criteria for the validation of decentralized testing: achievement of significantly reduced turnaround time and total cost compared to those for centralized testing, adequate accuracy and precision for the intended clinical purposes, definable operational or clinical advantage compared to centralized testing, administrative commitment for adequate training and the provision of needed resources, and preservation of the integrity of the database and charge capture. Although proponents of point-of-care testing advocate decreased turnaround times and lower test costs as reasons for decentralizing laboratory testing, there is a paucity of substantiating data. The potential to decrease laboratory fixed costs is dependent on systematic changes, such as with reengineering, of laboratory services to minimize centralized testing, yet much of the test base in laboratory medicine is not yet amenable to point-of-care testing solutions. Piecemeal implementation of point-of-care testing can cause duplication of equipment, inefficient use of personnel, and loss of economies of scale (9). Issues of quality also need to be addressed. While point-of-care testing has the potential to decrease preanalytical and postanalytical errors by bringing the analyst and patient closer together, the complexities associated with personnel training and test performance can have an adverse impact on the accuracy of the reported results. A constructive alternative to point-of-care testing in clinical microbiology is to optimize the organization, management, and processes of the centralized laboratory. Pre- and postanalytical delays that contribute to prolonged turnaround times can be addressed with appropriate engineering controls, such as rapid specimen delivery systems, including robotics, and a well-integrated laboratory information system. The goal is to limit nonanalytical sources of error ascribable to the physical separation from the patient. To achieve this goal, a useful

J. CLIN. MICROBIOL.

laboratory management approach is the cultivation of continuous quality improvement strategies as a mechanism for the provision of clinically relevant, cost-effective laboratory services (33, 40). The laboratory should participate with other medical disciplines in the design and implementation of critical pathways for patient care, in which issues regarding the selection and interpretation of laboratory tests, as they relate to patient outcomes, can be appropriately addressed. In summary, when considering the future of point-of-care testing for clinical microbiology, technological development should not dictate clinical need. The benefits of point-of-care testing in laboratory medicine have not been adequately validated by experience. Decentralized testing that offers rapid turnaround times but that lacks accuracy and reliability due to operator or instrument limitations will not meet current and future health care needs. In these times of increasing fiscal constraints on hospitals and a national demand for the control of rising health care costs, it is incumbent upon physicians, laboratorians, and administrators to carefully evaluate the benefits of point-of-care testing in order to justify its acquisition. WHO SHOULD DIRECT THE CLINICAL MICROBIOLOGY LABORATORY? According to the American Hospital Association, there are 6,000 hospital laboratories among the 12,000 regulated laboratories in the United States. Approximately 1,100 of these hospitals have more than 300 beds (4). The American Board of Medical Microbiology lists 261 active diplomates who are certified in clinical microbiology and public health, and 75% of these are Ph.D. scientists (5). The American Board of Pathology (ABP) reports 201 active, board-certified microbiologists, who are all physicians. Therefore, most hospital-based clinical microbiology laboratories are directed by non-board-certified individuals. Most doctoral-level laboratory directors are pathologists, followed by Ph.D. scientists and infectious disease specialists. A College of American Pathologists 1994 practice characteristics survey listed $164,000 to $239,000 as the range for pathologists’ salaries (16). A recent survey by the American Society for Microbiology determined that Ph.D. scientists’ salaries range from $55,000 to $110,000. Not surprisingly, fulltime physician laboratory directors are more expensive than Ph.D. directors. Several educational options are available for those interested in becoming a microbiology laboratory director. Directors certified by the ABP in clinical and anatomic pathology, the largest group of microbiology directors today, are required to spend 18 months training in clinical pathology, with specific training in microbiology for approximately 4 months (3). Those individuals qualifying for the ABP specialty microbiology certification must, in addition to their medical or pathology training, train for 1 year in a medical microbiology program approved by the Accreditation Council for Graduate Medical Education or complete 2 years of exceptional expertise by an alternate route (3). Certification by the American Board of Medical Microbiology requires 2 years of approved postdoctoral fellowship training plus 1 to 2 years of postdoctoral experience. If a fellowship program is not completed, 6 years of postdoctoral experience is required (2). The essentials required for an American Board of Medical Microbiology-approved program include training in microbiology, serology, quality management, laboratory safety, laboratory management and regulation, infectious diseases and pertinent clinical medicine, epidemiology and public health, molecular methods, and research methodology. Dieter Groschel, a microbiology director and physician who has trained many clinical microbi-

VOL. 37, 1999

GUEST COMMENTARY

TABLE 1. Quality index of laboratory by education and training of directora Director training

No. of individuals

Mean index

Ph.D. microbiologist M.D. pathologist, specialty training M.D. infectious disease specialist No doctoral-level director M.D. pathologist, no specialty training

177 78 64 43 93

22.8 22.4 22.1 18.1 17.4

a The quality index represents a sum of ratings (poor 5 1, excellent 5 5) for each of six microbiology laboratory specialty areas. The maximum combined score was 30. Data are from reference 37.

ologists who have become laboratory directors, has said, “From personal experience with postdoctoral trainees in clinical microbiology, I would like to state that they can learn the appreciation of the clinical situation in just as short a time as the internists and pathologists need to learn microbiology” (1). On the basis of the available programs, Ph.D. scientists not only qualify to direct clinical microbiology laboratories but, as a group, have access to comprehensive training which focuses on the skills needed by working directors. In spite of their qualifications, until recently, Ph.D. scientists were excluded by the College of American Pathologists and the Joint Commission for the Accreditation of Health Care Organizations from laboratory director positions. Neither organization would certify a laboratory without a physician director. In addition, there is no mechanism to provide a fee-for-service reimbursement for Ph.D.-trained laboratory directors. The fee-for-service system provides an incentive for pathologists, who can bill for their surgical pathology work, to act as parttime microbiology directors. However, the current laboratory climate is providing opportunities for Ph.D. scientists. Rules enforcing the Clinical Laboratory Improvement Act of 1988 changed the long-held bias against Ph.D. scientists as laboratory directors by allowing all doctoral-level individuals with appropriate training to direct clinical laboratories (6). Managed care reimbursement is eliminating fee-for-service arrangements. Lump sum or capitated payment programs that place an emphasis on quality and efficiency are emerging. This change promotes the selection of a director on the basis of a variety of skills rather than simply the ability to bill for services. Individuals who aspire to a doctoral-level directorship have access to postdoctoral training programs that are mandated to provide management training. Overall, there are fewer reasons for Ph.D. scientists to perceive a preferential exclusion from the clinical laboratory. A recent survey by Peddecord et al. (37) queried clinician members of the Infectious Disease Society of America concerning their satisfaction with the level of consulting in microbiology laboratories in their hospitals. Laboratory directors were divided into five categories and were rated according to a quality index (Table 1). Microbiologists with Ph.D. degrees were rated as high as pathologist and infectious disease directors who had specialty training in clinical microbiology regarding the quality of laboratory services that they provided. These three groups were all rated as providing higher-quality service than pathologists with no specialty training or laboratory directors without a doctoral degree. A recent assessment of the job market concluded that there is still ample opportunity for microbiologists; however, the most limited area of growth and employment was in the clinical-medical sector (19). This seems inappropriate at a time when health care consumes a large part of our national re-

887

sources and political discussions and when deaths directly related to infectious diseases in the United States rose 58% between 1982 and 1992. While the treatment of individual patients will continue to depend upon the accurate diagnosis of microbial infections, in the future the management of emerging and reemerging infectious diseases will depend heavily on epidemiology, antimicrobial use controls, and innovative therapeutic measures (28, 41). A survey of clinical microbiology laboratory directors by Thomson (45) in 1995 reflects the fact that many aspects of the role have already changed as a result of the nature of requested services in the work environment. The survey reported that laboratory management, clinical consultation and service, formal and informal teaching, and basic or applied research were the major responsibilities of microbiology directors (45). Work at the bench by microbiologists is less necessary, while consultation and innovation are more essential. A review of job descriptions by Gray and Baron (21) adds the evaluation of new methods and technical expertise to the list of skills required by directors. Additional skills, as summarized by Isenberg (24, 25) and D’Amato and Isenberg (17), include some less tangible traits that are needed by microbiology directors. These include political savvy for interactions with the medical staff, the art of clinical microbiology, and special personality traits. The art of clinical microbiology highlights a director’s olfactory acuity with a culture plate, interpretive skill with a Gram stain, and ability to construct a convincing diagnostic scenario, which combines medical and clinical microbiology with clinical medicine, to assist clinicians in the management of their patients. All of this must be performed in relative obscurity and without the glamour afforded other medical specialties. These new roles require a broad expertise in the pathophysiology of infectious diseases and can be provided only by a well-trained and highly skilled professional. Unfortunately, this need appears at a time when there exist forces that seek to eliminate the role of centralized laboratories and the skilled practitioners associated with them (48). From a patient’s perspective, although a patient is generally oblivious to the function of the microbiology laboratory, the patient seeks medical care to be cured of an infection. The patient also wants rapid laboratory results, particularly when the patient is in the physician’s office or the emergency department. Most importantly, the patient needs accurate results from a reliable laboratory. The accurate determination of whether or not an infection is present is as important as the accurate identification of potential pathogens. Rapid results are important, but speed must not sacrifice accuracy because this does not help the patient to be cured of infection. Detection of microbial drug resistance is pivotal to the patient’s goal to be cured of infection. Education is an important component of cost-efficient clinical microbiology practice. Microbiologists must actively teach physicians and patients about laboratory testing. One example is the potential utility of rapid screening tests for some populations but not for others. For instance, rapid screening for group A streptococcal infection in a child can be medically useful; however, when pharyngitis is suspected in an adult, traditional culture is a better diagnostic test. Education helps patients to better understand and appreciate their medical service. It builds confidence in and compliance with their health care provider and helps patients to attain their main goal of successful therapy. Physicians caring for patients are more aware of the laboratory than the patients are, and they want several services from it. Foremost, they need accurate results, including the accurate detection of drug resistance in microbial pathogens. They also need a rapid response from the laboratory. As the practice of

888

GUEST COMMENTARY

medicine becomes more complex and laboratories provide more diverse services that incorporate new technology, an increasing number of physician requests require an expanded interpretation of test results. The expertise of the clinical microbiologist in the interpretation of certain test results is critically important. For instance, one way to facilitate the delivery of medical care is to include additional interpretation for test results that otherwise would be unfamiliar to most practitioners. Some examples might include interpretation of the following: the pathogenic potential of organisms when unusual microbes are recovered from a clinical specimen, the likelihood that a peripheral stem cell harvest for transplantation that contains a single bacterial colony on an agar plate may be infected and inappropriate for reinfusion, and fungal or mycobacterial susceptibility tests, if no approved National Committee for Clinical Laboratory Standards interpretation standard exists. These are only a few of the clinical decisions that a microbiologist should routinely make. Patient care physicians also need support, and perhaps even leadership, from the director of microbiology for antimicrobial agent formulary decisions and for infection control policies and practices. For this, the microbiologist needs expertise in the pathophysiology of infectious diseases, pharmacokinetics, epidemiology, and infection control, as well as in traditional laboratory medical microbiology encompassing the detection and identification of microbes. To remain competitive, future training programs need to teach expertise in all of these areas. Microbiology staff want the medical director to be a technical consultant and an intermediary with patient care physicians. What is needed to optimize the benefits of the laboratory requires considerably more effort and direct involvement within the laboratory. Rather than acting solely as a technical consultant, the microbiologist as a medical consultant ensures that testing performed by the laboratory is medically relevant and that the results are reported in a manner understood by practicing physicians. Clinical training for the microbiologist is required in order to better relate to the clinician, judge the clinical relevance of testing, and provide information that improves patient care. Acting as a medical manager, the microbiology laboratory director must know how to select relevant tests for various infectious diseases, recognizing that such selection often must be population directed, for example, on the basis of age, immunosuppression, or human immunodeficiency virus infection. This selection process requires creative problem solving, such as the discontinuation of tests when they are no longer relevant and the introduction or development of new procedures as the problems in the diagnosis and treatment of infectious diseases evolve. An often overlooked, but necessary, role of the director is as an intermediary with an institution’s administration. In this case, the director needs to act as a resource defender. In this role the director must understand financial and resource accounting and respond appropriately when resources are targeted for reduction. The microbiology director must be recognized by the administration as capable of justifying expenditures, maintaining high-quality service in a cost-efficient manner, and envisioning and planning for the future. At times, this may even require increased expenditures by the laboratory (22). In the future, microbiology must avoid simply being cast as a cost center but rather must be viewed as a critical partner in the management of overall hospital expenditures. Microbiology can serve as a true revenue center through maximization of cost-efficient practice with expertise in diagnostic testing, hospital epidemiology, and institutional antimicrobial agent management. Many administrators appear to want a no-cost, no-error,

J. CLIN. MICROBIOL.

high-quality laboratory. A common practice, designed to lower costs, is to assign the role of microbiology laboratory director to someone with other responsibilities within the department of pathology or laboratory medicine. However, with the complex infectious disease environment of the future, this cannot be condoned as acceptable. To fully serve the needs of institutional administration, a dedicated microbiology director needs to be appointed. This director can provide a reasonable assessment of necessary microbiology services that includes alternative testing choices to maintain needed service and quality. The director must lead a laboratory engaged in planning for the future of medical microbiology, with a focus on bacteriology since the most common pathogens are bacteria. A dedicated director is equipped to provide leadership and service in appropriate consultative interpretation, antimicrobial agent formulary decisions, the implementation of infection control policies and practices, the development of clinical pathways for infectious diseases, medical education in microbiology, and a partnership with administration and practicing physicians in a cost-efficient medical practice. Such a practice has led to measurable benefits, ranging from an improved patient outcome in the treatment of an emerging pathogen, vancomycin-resistant Enterococcus faecium (34, 43), to a dramatic reduction in costs associated with overall nosocomial infections (22). The microbiology laboratory must function in such a way that the relevant information needed to make medical decisions is provided to clinicians. This implies that the laboratory director understands infectious disease needs, the availability of resources, and how to medically prioritize testing when resources are limited. Communication between each patient’s physician and the laboratory professional must be open and frequent. To effect meaningful communication, the director of microbiology must understand infectious diseases as well as or better than the clinician understands the laboratory aspects of microbiology. The director must participate in medical practice, assist in establishing guidelines for the management and control of infectious diseases, and provide a sound rationale to practicing physicians for these decisions so that guidelines are followed. In summary, the success or failure of microbiology laboratories in the 21st century will depend largely on the quality of the medical director leadership. In order to be successful, serious planning by the profession is needed. Since the most common serious infections in the United States for the foreseeable future will be caused by bacteria, bacteriology should be a major strength of new trainees, with one or more strong minors in other subspecialties of microbiology. Medical microbiology training should emphasize expertise in the nontraditional areas of disease pathophysiology, antimicrobial agent pharmacology, and hospital epidemiology as part of the core or critical knowledge base. Most importantly, for both physicians and doctoral scientists, appropriate and comprehensive training and credentialing programs are the key to the future. Laboratories directed by the graduates of such innovative programs will have an enormous potential to provide better health care for patients with less cost to the provider, and an expanded job market for the unquestionably needed clinical medical microbiologist will evolve. REFERENCES 1. Abramson, J. 1988. Ph.D. recognition: Ph.D. perspective. ASM News 54: 464–465. 2. American Board of Medical Microbiology. 1993. Information on certification programs. American Academy of Microbiology, Washington, D.C. 3. American Board of Pathology. 1996. Booklet of information. American Board of Pathology, Tampa, Fla.

VOL. 37, 1999 4. American Hospital Association. 1996. Fact sheet from the AHA resource center. American Hospital Association, Chicago, Ill. 5. Anonymous. 1989. Special report: public affairs and clinical microbiology. ASM News 55:189–196. 6. Anonymous. 1991. New standards allow nonphysicians to direct laboratories. ASM News 57:638–639. 7. Anonymous. 1995. “Point-of-care” laboratory testing: is decentralized testing the best alternative for your hospital? Health Devices 24:173–207. 8. Auxter, S. 1996. Surviving re-engineering: lab directors tell their tales. Clin. Lab. News 22:17. 9. Bachner, P. 1995. Alternate site testing: the old and new paradigm or the past is prologue. Arch. Pathol. Lab. Med. 119:881–885. 10. Bauer, S., and C. Teplitz. 1995. Total laboratory automation: a view from the 21st century. Med. Lab. Observ. 27:22–25. 11. Bergman, R. 1994. Reengineering health care. Hosp. Health Net. 68:28–36. 12. Bickford, G. R. 1994. Decentralized testing in the 1990’s. A survey of United States Hospitals. Clin. Lab. Med. 14:623–645. 13. Boyce, N. 1996. The healthcare system’s changing climate. Clin. Lab. News 22:10. 14. Castaneda-Menedez, K. 1994. Re-engineering: is it right for you. Adv. Admin. Lab. 3:16–22. 15. Castro, H. J., J. M. Oropello, and N. Halpern. 1995. Point-of-care testing in the intensive care unit: the intensive care physician’s perspective. Am. J. Clin. Pathol. 104(Suppl. 1):S95–S99. 16. College of American Pathologists. 1994. Practice characteristics survey report. College of American Pathologists, Northbrook, Ill. 17. D’Amato, R. F., and H. D. Isenberg. 1991. Practical and fiscally responsible application of clinical microbiology to patient care. ASM News 57:22–26. 18. DuBois, D., V. G. Ray, B. Nelson, and J. B. Peacock. 1986. Rapid diagnosis of group A strep pharyngitis in the emergency department. Ann. Emerg. Med. 15:157–159. 19. Fox, J. L. 1996. Employment prospects for U.S. Microbiologists are sound. ASM News 62:181–184. 20. Goodwin, S. A. 1994. Point-of-care testing in a post anesthia care unit. Med. Lab. Observ. 26(Suppl. 9S):15–18. 21. Gray, L. D., and E. J. Baron. Unpublished data. 22. Hacek, D. M., T. Suriano, G. A. Noskin, J. Kruszynski, B. Reisberg, and L. R. Peterson. Medical and economic benefit of a comprehensive infection control program that includes routine determination of microbial clonality. Am. J. Clin. Pathol., in press. 23. Handorf, C. R. 1994. Background: setting the stage for alternate site laboratory testing. Clin. Lab. Med. 14:451–459. 24. Isenberg, H. D. 1982. The role of clinical microbiology in health care. ASM News 48:101–105. 25. Isenberg, H. D. 1989. Role of the clinical microbiologist in patient care. ASM News 55:544–548. 26. Jahn, M. 1995. The managed care era strikes the lab. Med. Lab. Observ. 27:38–43. 27. Jandreski, M. 1996. What’s what in laboratory finance. Clin. Lab. News 22:10–11. 28. Jarvis, W. R. 1996. Preventing the emergence of multidrug-resistant microorganisms through antimicrobial use controls: the complexity of the problem. Infect. Control Hosp. Epidemiol. 17:490–495. 29. Keffer, J. H. 1995. Economic considerations of point of care testing. Am. J.

GUEST COMMENTARY

889

Clin. Path. 104(Suppl. 1):S107–S110. 30. Kost, G. J. 1992. The hybrid laboratory: shifting the focus to the point-ofcare. Med. Lab. Observ. 24(Suppl. 9):17–21, 24–28. 31. Kurec, A. S. 1993. Implementing point-of-care testing. Clin. Lab. Sci. 6:225– 227. 32. Lamb, L. S., R. S. Parrish, S. F. Goran, and M. H. Biel. 1995. Current nursing practice of point-of-care laboratory diagnostic testing in critical care units. Am. J. Crit. Care 4:429–434. 33. LaRocco, M. T. 1995. Quality and productivity in the clinical microbiology laboratory: continuous quality improvement. Clin. Microbiol. News 17:129– 131. 34. Mekonen, E. T., G. A. Noskin, D. M. Hacek, and L. R. Peterson. 1995. Successful treatment of persistent bacteremia due to vancomycin-resistant, ampicillin-resistant Enterococcus faecium. Microb. Drug Resist. 1:249–253. 35. Mortensen, J. E. 1994. Is one laboratory in town enough? Adv. Exp. Med. Biol. 349:131–134. 36. Mortensen, J. E. 1995. Quality and productivity in the microbiology laboratory: reengineering. Clin. Microbiol. News 17:131–133. 37. Peddecord, K. M., E. J. Baron, D. Francis, and J. A. Drew. 1996. Quality perceptions of microbiology services. A survey of infectious diseases specialists. Am. J. Clin. Pathol. 105:58–64. 38. Roseman, E. 1995. Reengineering the laboratory. Med. Lab. Observ. 27:36– 38. 39. Salem, M., B. Chernow, R. Burke, J. Stacey, M. Slogoff, and S. Sood. 1991. Bedside diagnostic testing: its accuracy, rapidity, and utility in blood conservation. JAMA 266:382–389. 40. Saxena, S., and E. T. Wong. 1993. Does the emergency department need a dedicated stat laboratory? Continuous quality improvement as a management tool for the clinical laboratory. Am. J. Clin. Pathol. 100:606–610. 41. Simmons, B. P., M. F. Parry, M. Williams, and R. A. Weinstein. 1996. The new era of hospital epidemiology: what you need to succeed. Clin. Infect. Dis. 22:550–553. 42. Stentz, F., M. V. Stevens, A. Rhodes, and S. Starnes. 1996. Redefining the laboratory in an era of cost constraints. Am. Clin. Lab. 14:6–8. 43. Stosor, V., L. R. Peterson, M. Postelnick, and G. A. Noskin. 1996. Therapy of vancomycin resistant Enterococcus faecium sepsis: a novel strategy, abstr. 170. In Program and abstracts of the 34th Annual Meeting of the Infectious Diseases Society of America. Infectious Diseases Society of America, Washington, D.C. 44. Thiebe, L., K. Vinci, and J. Gardner. 1993. Point-of-care testing: Improving Day-Stay services. Nurs. Manage. 24:54, 56. 45. Thomson, R. B. 1995. The changing role of the clinical microbiology laboratory director. Diagn. Microbiol. Infect. Dis. 23:45–51. 46. True, B. L., B. L. Carter, C. E. Driscoll, and J. D. House. 1986. Effect of a rapid diagnostic method on prescribing patterns and ordering of throat cultures for streptococcal pharyngitis. J. Fam. Pract. 23:215–219. 47. Wiedermann, B. L., R. H. Schwartz, and P. McCoy. 1991. Experience with rapid latex agglutination testing for group A streptococcal pharyngitis in a pediatric group office laboratory. J. Am. Board Fam. Pract. 4:79–82. 48. Zacharia, M. (ed). 1996. Measuring the cost savings of point-of-care testing. Med. Lab. Observ. 29(Suppl. 9S):2–16. 49. Zaloga, G. P. 1991. Monitoring versus testing technologies: present and future. Med. Lab. Observ. 23(Suppl. 9S):20–25, 28–31.

The views expressed in this Commentary do not necessarily reflect the views of the journal or of ASM.