JOURNAL OF CLINICAL MICROBIOLOGY, Mar. 1996, p. 496–500 0095-1137/96/$04.0010 Copyright q 1996, American Society for Microbiology
Vol. 34, No. 3
GUEST COMMENTARY Infectious Disease Physicians Rate Microbiology Services and Practices ELLEN JO BARON,1* DIANE FRANCIS,2
AND
K. MICHAEL PEDDECORD2
Department of Medicine, University of California, Los Angeles, California 90024,1 and Graduate School of Public Health, College of Health and Human Services, San Diego State University, San Diego, California 92182-41622 Recent years have seen increasing emphasis on cost containment and quality improvement in clinical laboratory activities. Modifying those activities to enhance clinical relevance is one strategy that should be satisfying to both laboratory scientists and administrators. This guest commentary describes one approach to quality improvement—the use of user surveys to identify areas for improvement. As an initial attempt to define such areas in clinical diagnostic microbiology, infectious disease specialists, targeted for their particular interest and expertise in microbiology laboratory results, were polled and their responses were analyzed. Some of these data have been presented previously (E. J. Baron, D. P. Francis, and K. M. Peddecord, abstr. C-170, p. 520, in Abstracts of the 94th General Meeting of the American Society for Microbiology, 1994; K. M. Peddecord, E. J. Baron, D. P. Francis, and A. S. Benenson, abstr. C-172, p. 520, in Abstracts of the 94th General Meeting of the American Society for Microbiology, 1994; K. M. Peddecord, E. J. Baron, D. P. Francis, and J. A. Drew, Am. J. Clin. Pathol. 105:58–64, 1996). The discussion includes our recommendations for the use of these survey responses, and their limitations, as stimuli to initiate reexamination of certain microbiology laboratory practices in the interest of developing more cost-effective and clinically relevant protocols. based on their input. Survey responses confirmed, as has been established by survey researchers, that key elements in successful survey design include content of interest to potential respondents, consistency and convenience of response pattern, simplicity and clarity of language, minimal essential length, a targeted respondent population, use of a respected colleague to provide an endorsement, and professional image of the questionnaire. An initial mailing of 1,264 surveys was sent in September 1993. A second mailing was sent to 955 nonrespondents in November 1993. Responses were accepted until the end of January 1994.
DEVELOPMENT OF A SURVEY TO ASSESS PHYSICIAN SATISFACTION WITH MICROBIOLOGY SERVICES Few surveys of physicians’ impressions of the quality of the laboratories that they use have been published (3, 16–18). A recent survey attempted to ascertain the opinions of a specific group of physicians regarding their satisfaction with the microbiologists of their primary laboratories, their preferences regarding certain technical procedures, and their views on the overall quality of the microbiology laboratories that they use most often. Established survey research methods were used to poll 1,264 members and fellows of the Infectious Diseases Society of America (IDSA) who identified themselves as seeing patients, the group of physicians identified as having potentially the most informed opinions about microbiology laboratories. The broad sample of respondents and the wide-ranging topics for which opinions were sought precluded the survey’s ability to document a cohesive standard of practice in many areas. However, the results did converge regarding certain laboratory practices on which improvement activities may be focused. The questionnaire elicited responses in three basic areas: (i) the characteristics of respondents and their primary laboratories, (ii) opinions and problems related to laboratory service and interaction, and (iii) preferences regarding a limited number of technical issues, including culture processing and reporting of results. With input from several microbiologists, an eight-page draft questionnaire entitled, ‘‘Microbiology Service and Quality: Is Your Laboratory Meeting Your Needs?’’ was designed and sent to 16 infectious disease practitioners to preview. The final questionnaire format and content were
SURVEY RESPONSES Greater than 500 responses were received from more than 40 states and various practice settings. Approximately 34% of respondents worked primarily in a medical school or university hospital environment. Another 26% worked in private practice, 14% worked in a government health system, 11% worked in private hospitals, 9% worked in research, and the remaining 6% worked in either a health maintenance organization or other health care setting. While the respondents appeared to be representative of all IDSA members, the attitudes, needs, and opinions of IDSA members may be quite different from those of the majority of practitioners using community hospitals. Given the characteristics of the respondents, this survey’s conclusions would probably be most applicable to teaching and larger community hospitals. The limitations of survey research confined to a single interest group were apparent in this study. The overall response rate was 40%, but an important factor is the nature of membership in the IDSA. IDSA members have in common their interest in infectious diseases. It is assumed that they are more knowledgeable customers of microbiology laboratory services, but they may also be more critical and have higher
* Corresponding author. Mailing address: 756 Haverford Avenue, Pacific Palisades, CA 90272. Phone: (310) 454-0418. Fax: (310) 4543176. Electronic mail address:
[email protected]. 496
VOL. 34, 1996
expectations than physicians in other specialties. Because the survey included only physicians with an abiding interest in infectious diseases, the results should not be generalized to other physician groups. Microbiology ordering practices. With regard to test volume, respondents reported that, in general, they ordered greater than 20 (31% of respondents), 11 to 20 (31%), 4 to 10 (28%), or 1 to 3 (10%) cultures per week. Most medical school-associated respondents sent the majority of their culture requests to the microbiology laboratory affiliated with their institution, but diverse types of laboratories were utilized by the others. As the primary laboratory, private hospital laboratories received specimens for microbiological studies from 18% of the respondents, commercial laboratories received them from 12%, infectious disease research laboratories received them from 13%, physician office and other laboratories received them from 7%, group practice laboratories received them from 6%, and public health laboratories received them from 5%. Respondents were more likely to practice in a larger health care facility, as one might expect with this specialty. Multivariate analyses failed to reveal any response trends based on geographical location or type of practice. Laboratory service quality issues. Respondents were asked to rate the overall quality of six microbiology specialty areas for their primary laboratory by using numerical scores ranging from 1 (poor) to 5 (excellent). Infectious disease specialists were most satisfied with the quality of the bacteriology sections of their laboratories, with 86% rating this area excellent or very good. Mycobacteriology, mycology, and virology were rated less highly; approximately 48 to 66% rated the quality in these areas as very good to excellent. Less than 20% rated their immunoserology or their parasitology laboratories as excellent. A general laboratory ‘‘quality index’’ was developed by adding the numerical scores assigned by respondents to each of the six specialty areas. The resulting scores ranged from a low of 6 to a high of 30, with a mean quality index of 21.1 and a standard deviation of 5.6. A one-way analysis of variance was used to compare quality indexes among laboratory types and among laboratory director qualification characteristics. Respondents were asked to indicate the importance of specific microbiology laboratory services (from a list provided) to the perceived quality of any microbiology laboratory and then to indicate the level of service (from poor to excellent) available currently from their primary laboratory. Activities from a list provided in the survey thought to be very important for providing quality service were, in order of preference, ‘‘reasonably fast turnaround time for results,’’ ‘‘willingness to accommodate special requests,’’ ‘‘reporting selected results by telephone,’’ ‘‘range of services available from off-site reference laboratories,’’ ‘‘provision of collecting materials and instructions,’’ ‘‘reliable courier service,’’ ‘‘computerized reporting system,’’ ‘‘availability of immunoserological tests on site,’’ and ‘‘availability of full-service mycobacteriology on site.’’ Virology services on site and services provided on nights and weekends were rated of equal importance and less important than the previously listed services. Considered least important were interpretations and explanations of results, either included with test results or published separately, and the availability of esoteric tests on site. Of the activities (excluding consultation) that infectious disease specialists were asked to rate regarding the current level of service in their own primary laboratories, only two activities received a rating of excellent by 50% or more of the respondents. These service areas were ‘‘selected results reported by telephone’’ (53% of respondents) and ‘‘range of services available from reference (off-site) laboratories’’ (50% of respondents). The physicians were less satisfied with their current
GUEST COMMENTARY
497
laboratories’ range of esoteric tests available on site, availability of virus cultures on site, provision of a separate compendium of report interpretations or explanations, and number and types of services provided on nights and weekends. Respondents were asked to identify the incidence of problems from a list with space for a write-in answer and to rate the potential seriousness. Fortunately, the most serious type of problem, ‘‘failure to call a critical result,’’ was reported to occur least often, with 29% of respondents reporting that such problems occurred only one to three times during the previous 12 months and only 8% experiencing such problems more than four times during the previous year. One reason for the low numbers of this type of potentially serious mistake is that infectious disease specialists may often solicit results from the laboratory before the laboratory has begun to evaluate the culture results during the normal work flow, particularly when critically ill patients are involved. Forty-eight percent of respondents reported that culture mishandling occurred approximately one to three times in the past year in the microbiology laboratory, and 11% experienced this problem four or more times. Although 47% reported one to three incidents of culture loss in transit, 26% experienced this problem four or more times. The most common problems reported to occur 10 or more times during the previous year were failure of the laboratory report to be charted or reported in a timely fashion (12%) and loss of specimens in transit to the laboratory (6%). Additional problems reported to have occurred four or more times during the past year included inconsistent results between technologists (12% of respondents) and reports that needed a telephone call for explanation (18% of respondents). Respondents indicated that the three potentially most serious problems with regard to patient care issues were failure to call results of critical tests (considered to be the most serious problem by 61% of respondents), culture loss in transit, and culture mishandling in the microbiology laboratory (each of the latter problems was considered to be most serious by 48% of infectious disease physicians). Of note, however, 12 physicians indicated that various other miscellaneous unlisted problems, some of which were thought to be of major seriousness, occurred .10 times during the past year. Unfortunately, most respondents did not describe the nature of these problems. Statistical evaluation of these responses and the influence of problems on the overall quality ratings of laboratories by physicians is addressed in another publication (15). Briefly, when several variables were combined to create a statistical measurement of quality for microbiology laboratories, it appeared that physicians are willing to overlook problems, even those they consider to be serious, that occur in laboratories whose overall quality they have rated as high. The availability of consultation services was seen as important to the overall quality of a microbiology laboratory by most respondents; 78% felt that consultative services by a Ph.D. microbiology director were very important or essential; 80% reported a similar desire for consultative services by a nondoctoral microbiology supervisor. Thirty-two percent felt that such services from a pathologist were extremely important or essential. It should be noted, however, that only a very small number of laboratories with microbiology-trained pathologists were included in the survey. Despite a general lack of perceived importance of pathology consultation, 38% of those surveyed rated their pathologists’ consultative services as very good to excellent. In addition to user satisfaction, another potential measure of a laboratory’s quality is proficiency survey performance. Howanitz reviewed five such studies and found conflicting results regarding the presence of a connection between laboratory
498
GUEST COMMENTARY
FIG. 1. Infectious disease physicians’ ratings of the quality of the bacteriology specialty areas of their primary microbiology laboratories with respect to director training.
director educational qualifications and proficiency test results (9). He concluded that such measurements may be inappropriate as the sole measure of laboratory performance. Infectious disease clinician satisfaction ratings, as a potential measure of laboratory quality, were also not related to the educational qualifications of the directors. Bacteriology section quality in laboratories without doctoral level directors (only 9% of all laboratories represented) and those directed by pathologists without specialty training in microbiology (20% of laboratories) was rated ‘‘excellent’’ by 23 and 28% of respondents, respectively (Fig. 1). When analyzed with respect to the microbiology laboratory director’s qualifications, infectious disease physicians’ ratings of quality were related more to specialty training in microbiology than to degree (M.D. versus Ph.D.). Direction of the laboratory by a diplomate of the American Board of Medical Microbiology, however, showed a significant association with higher quality ratings in all specialty areas (bacteriology, mycology, parasitology, mycobacteriology, virology, and immunoserology) compared with the quality ratings of laboratories directed by nondiplomates. Microbiology laboratories whose directors were American Board of Medical Microbiology diplomates had higher overall quality index scores (mean, 23.3) than did laboratories with directors of unknown American Board of Medical Microbiology status (mean, 21.7) or those with directors reported as not being American Board of Medical Microbiology diplomates (mean, 19.7). These mean differences were statistically highly significant (P , 0.0001). SURVEY RESULTS RELATED TO TECHNICAL ISSUES Results of this survey exposed at least one pitfall of surveying users as to their preferences, particularly regarding technical aspects of laboratory practices. The responses may differ from approaches advocated in the published literature and may conflict with the laboratory’s preferred protocols. Results of this type, as demonstrated in several areas represented by the present survey, could prevent a laboratory from making appropriate methodological changes. The lesson to be learned is not to solicit opinions for those aspects of laboratory practice for which protocols should be established by laboratory professionals or for those that are not negotiable. Blood cultures. When asked if a laboratory should offer the Isolator lysis-centrifugation blood culture system regardless of the other blood culture systems available, 60% of respondents answered yes, 19% answered no, and 21% did not have an opinion. Although it would have been valuable to ascertain the situations for which respondents recommended the lysis-cen-
J. CLIN. MICROBIOL.
trifugation system or to expand on their reasons for favoring its availability, such information was not ascertained. Seventy-two percent of those surveyed reported that written recommendations for blood culture collection were provided by the laboratory, 14% reported that their laboratories had no such recommendations, and 14% did not know. The respondents’ opinions regarding their laboratories’ collection recommendations were not sought. Half of the infectious disease physicians reported that their laboratories had not established a limit on the number of blood cultures obtained per patient within 24 h, while 36% reported that there was a limit and 14% did not know. A number of publications have suggested that a limit of four blood culture sets be enforced per patient per 24-h period, although a metaanalysis stated that two or three blood culture sets almost always suffice to establish bacteremia (1, 19, 24). On the basis of these results reported by infectious disease physicians, many laboratories have not taken the opportunity to limit unnecessary and expensive blood culture orders. With regard to reporting of blood culture results, 97% of the 504 infectious disease physicians polled wished to be informed of a positive blood culture as soon as possible, before a presumptive identification is available. Even if only one bottle of a set shows growth of a potential pathogen, 84% of the respondents preferred a direct telephone call to the physician and the nursing unit. If the blood culture is suspected by the laboratory to reflect contamination (not further defined), however, the portion of respondents who desired such immediate notification dropped to 46%. In fact, 31% would be happy to be informed of a suspected contaminant (growth in one bottle only) sometime during the work shift and 20% did not feel that a report separate from the routine chart report is necessary. Although most laboratories have a policy that requires telephone reporting of all positive blood cultures, these survey results indicate that the need for such rapid reporting of suspected procurement-associated skin organisms by a laboratory could be reexamined with input from interested clinicians serving the facility. Lower respiratory tract specimens. Respondents were asked how seven types of lower respiratory tract specimens should be evaluated for quality (defined as ‘‘acceptability for culture based on absence of contaminating oropharyngeal flora’’). Although 70% of respondents chose ‘‘screen for epithelial cells and reject if greater than 25 per low-power field’’ with respect to expectorated sputum specimens, there was no other type of lower respiratory tract specimen for which more than 14% of those surveyed were willing to accept rejection criteria. Other choices for specimen evaluation included ‘‘report numbers of epithelial cells but culture specimen and process normally,’’ ‘‘report numbers of epithelial cells, culture specimen, and perform workup based on Gram stain results,’’ and ‘‘do not screen; set up all specimens received and process normally.’’ The largest numbers of physicians chose ‘‘do not screen . . .’’ for any specimen for Pneumocystis stain (59%), any specimen for acid-fast bacilli (56%), any specimen for fungal cultures (45%), and bronchoalveolar lavages (44%). Less than 10% of respondents advocated screening of endotracheal aspirates, in contrast to findings from at least one study. Morris and colleagues initiated screening and rejection of endotracheal aspirates that showed either greater than 10 squamous epithelial cells per low-power field or no organisms on Gram stain (13). This strategy, based on preliminary studies that showed that such specimens yielded no interpretable information on culture, was found to save approximately $66,000 per year without adversely affecting patient care (13). In this case, at least, microbiologists cannot use the infectious disease
VOL. 34, 1996
physicians’ opinions as a basis for support in cutting laboratory costs by reducing unnecessary culture processing. Although quantity criteria for culture workup of endotracheal aspirates or bronchoalveolar lavages, as advocated by numerous studies on ventilated patients (4–6, 11, 12), were presented as a choice, only 7% of respondents preferred that format for reporting of results. Processing of bronchial washings was another area in which infectious disease physicians appear to disagree with at least some published literature. Forty-six percent of respondents favored reporting of numbers of epithelial cells but normal processing of such specimens, and 30% indicated that bronchial washings should not be screened. Baselski and others, in contrast, have stated that bronchial washings alone are not acceptable specimens for the diagnosis of pneumonia and that they should not be cultured (4). There is an opportunity for microbiologists to make changes in their handling of bronchial washings, but only if they are able to inform infectious disease physicians about the current recommendations. It is possible that the opinions recorded by a survey of pulmonary specialists would have differed from those received from infectious disease specialists regarding handling of respiratory specimens. When asked to indicate their preferences for final report formats of three types of lower respiratory tract specimens, 65% of respondents were satisfied with descriptive identification (defined as ‘‘colony morphology and sometimes Gram stain based’’) only of ‘‘normal oropharyngeal flora’’ recovered on sputum cultures. However, fewer than half of those surveyed agreed that such limited identifications of isolates were acceptable for bronchoscopic washing or endotracheal aspirates (46%) and bronchoalveolar lavages (43%). Only 13% of respondents supported the popular concept of limiting culture workup of lower respiratory tract secretions to those microorganisms implicated as clinically significant on the basis of Gram stain results (10). It seems that microbiologists need to work closely with their own pulmonary specialists and infectious disease physicians to implement many of the newer concepts in the limiting of laboratory tests for diagnosis of lower respiratory tract disease, even in the face of almost universal agreement that laboratory studies of such specimens are generally unsatisfactory (21). Stool specimens. Respondents were asked to choose from a limited list of nine etiological bacterial agents their preferences for ‘‘microbiological processing of routine stool specimens (NOT for patients who have been hospitalized longer than 3 days and develop new-onset diarrhea).’’ Most infectious disease physicians responded that Salmonella (97%), Shigella (97%), and Campylobacter (94%) species should be sought routinely. Routine processing for detection of other agents, however, was controversial. Forty-one percent of respondents felt that Aeromonas and Plesiomonas species should be sought routinely, 53% were in favor of routine tests for Yersinia enterocolitica, and 38% favored routine methods for detection of Escherichia coli O157:H7. Public health authorities have called universally for laboratories to routinely include special media for detection of verotoxin-producing E. coli. In this case, the infectious disease specialists’ preferences ran counter to public health interests. Microbiologists must weigh the public health and clinical consequences of failing to detect this organism early in an outbreak before deciding to exclude tests for E. coli O157:H7 from routine stool processing. Seventeen percent of respondents felt that Clostridium difficile toxin testing is warranted for all stools sent to a microbiology laboratory without special orders, although the majority advocated this test by request only. Even fewer respondents (5%) advocated routine culture for this organism. Microbiol-
GUEST COMMENTARY
499
ogists can use the very strong consensus among those surveyed to resist performing routine cultures for C. difficile as opposed to toxin tests. The consensus panel convened as part of the First North American Congress on Anaerobic Bacteria and Anaerobic Infections stated that cultures for C. difficile should be followed by toxin production tests, an even more stringent limitation than that implied by the infectious disease physicians (7). Although such a labor-intensive and time-consuming protocol may be the most reliable for diagnosis of this disease, infectious disease physicians seem to appreciate the practical limitations of following that sort of protocol routinely. A recent Q-Probe (a quality assurance consensus program) conducted by the College of American Pathologists regarding stool processing found that participating laboratories (representing 601 institutions) performed routine tests for detection of C. difficile on 53% of all stools submitted for bacteriological studies (23). In this case, the infectious disease physicians’ opinion was more conservative than current laboratory practices, at least among Q-Probe participants. Another limitation of this survey was that preferences regarding ovum and parasite testing were not determined. Questions regarding ovum and parasite testing could legitimately be addressed to local clinicians by laboratorians seeking to develop individualized consensus protocols, such as whether a Giardia lamblia antigen test should be part of a routine ovum and parasite test request and whether the determination of fecal leukocytes by a new, more sensitive latex agglutination system is desired (8, 20). Susceptibility testing studies. When asked their preferences for reporting of susceptibility results, the majority of those surveyed (86%) chose ‘‘report antimicrobial agents selectively on the basis of mutual agreement with pharmacy and infectious disease specialists.’’ Only 14% chose ‘‘report all antimicrobial agents tested.’’ Seventy percent of respondents also felt that the microbiologists were capable of determining which isolates to test and when to test them, although only 58% of respondents agreed that urine culture isolates should be tested regardless of whether susceptibility tests were ordered originally. The University of Michigan system, in which culture results must be assimilated by physicians before they are allowed to request susceptibility tests, was rejected by 88% of respondents for potentially significant isolates from any site. Implementation of this system at the University of Michigan hospital, in which susceptibility tests are performed on isolates from nonsterile sites only after specific orders are received, resulted in a 47.3% decrease in the number of tests performed (25). Unfortunately, outcome data were never collected. Infectious disease physicians were more supportive of the reporting of MICs with category interpretations based on the specimen source than reporting of MICs alone or the other choices for reporting of susceptibility results (Table 1). A maximum of only 24% of respondents preferred the Kirby-Bauer disk diffusion method, and that response was directed toward urine isolates. For all potentially significant isolates, only 14% of those surveyed favored the disk diffusion method. Several problems noted with automated systems’ abilities to detect new resistances, such as vancomycin-resistant enterococci and methicillin-resistant staphylococci, have necessitated that laboratories retain disk diffusion capabilities for routine use in selected circumstances (22). In recent years, susceptibility testing of additional organism-antibiotic pairs, including a number of gram-negative rods, has moved back to the disk diffusion method because of its greater accuracy.
500
GUEST COMMENTARY
J. CLIN. MICROBIOL.
TABLE 1. Susceptibility test result reporting preferences of infectious disease specialistsa % of respondents preferring report format on basis of source or nature of isolate Reporting preference
Report MICs Report MICs and category interpretation Report MICs and category interpretation based on source Perform disk diffusion and report category interpretation Perform disk diffusion and report category interpretation based on source Report both MICs and disk diffusion results a
Isolates Isolates Potentially from from significant systemic urine isolates infections cultures
15 17 40
20 24 42
9 13 36
9
5
16
14
5
24
5
4
2
The number of respondents was 425 to 453.
SUMMARY Despite the many limitations raised throughout this discussion, results of physician surveys such as this one can be a useful adjunct to improving performance and customer satisfaction for most clinical laboratories. It is unrealistic to expect to satisfy all clinicians all of the time, but soliciting input from motivated customers can help to uncover problems that can be corrected. In fact, the act of soliciting input by itself pays public relations dividends for the laboratory by implying a willingness to accommodate physicians’ needs. Even if the problems reported are only perceived as such by the infectious disease physicians, their perceptions will color their use of the laboratory and their comments to other clinicians regarding the laboratory’s quality. Nonspecialists will accept the opinions of infectious disease physicians about the quality of microbiology laboratories, which may effect their utilization. For example, demanding that more tests be sent to reference laboratories or repeating cultures unnecessarily because of lack of faith in the results are direct results of a physician’s perception of a poorquality laboratory. The results reported for this study provide a framework and an initial benchmark to allow microbiologists to compare their practices and may suggest ways in which to improve services in their laboratories. ACKNOWLEDGMENTS This study was made possible by an unrestricted grant from Marion Merrell Dow, Inc. We are indebted to Joan Sondag for facilitating the study and to Patrick R. Murray for encouragement and concept development. We also thank Joe Drew, David Scharf, and Anne Evosevich for help with fielding the survey and statistical analyses. Special thanks are due to the infectious disease specialists who assisted us in improving the pretest survey instrument. REFERENCES 1. Aronson, M. D., and D. H. Bor. 1987. Diagnostic decision: blood cultures. Ann. Intern. Med. 106:246–253. 2. Baron, E. J., D. P. Francis, and K. M. Peddecord. 1994. Infectious diseases specialists offer mixed opinions on many aspects of microbiology testing and reporting practices, abstr. C-170, p. 520. In Abstracts of the 94th General Meeting of the American Society for Microbiology 1994. American Society
for Microbiology, Washington, D.C. 3. Bartlett, R. C., M. Mazens-Sullivan, J. Z. Tetreault, S. Lobel, and J. Nivard. 1994. Evolving approaches to management of quality in clinical microbiology. Clin. Microbiol. Rev. 7:55–88. 4. Baselski, V. S., M. El-Torky, J. J. Coalson, and J. P. Griffin. 1992. The standardization of criteria for processing and interpreting laboratory specimens in patients with suspected ventilator-associated pneumonia. Chest 102(Suppl.):571S–579S. 5. Baselski, V. S., and R. G. Wunderink. 1994. Practical laboratory guidelines for performing respiratory cultures on patients with ventilator-associated pneumonia. Clin. Microbiol. Newsl. 16:65–69. 6. El-Ebiary, M., A. Torres, J. Gonzalez, H. Puig de la Bellacasa, C. Garcia, J. De Anta, M. Ferrer, and R. Rodriguez-Roisin. 1993. Quantitative cultures of endotracheal aspirates for the diagnosis of ventilator-associated pneumonia. Am. Rev. Respir. Dis. 148:1552–1557. 7. Gerding, D. N., and J. S. Brazier. 1993. Optimal methods for identifying Clostridium difficile infections. Clin. Infect. Dis. 16(Suppl 4):S439–S442. 8. Guerrant, R. L., V. Araujo, E. Soares, K. Kotloff, A. A. M. Lima, W. H. Cooper, and A. G. Lee. 1992. Measurement of fecal lactoferrin as a marker of fecal leukocytes. J. Clin. Microbiol. 30:1238–1242. 9. Howanitz, P. J. 1988. Use of proficiency test performance to determine clinical laboratory director qualifications. Arch. Pathol. Lab. Med. 112:349– 353. 10. James, L., and J. E. Hoppe-Bauer. 1992. Processing and interpretation of lower respiratory tract secretions, p. 1.15.1–1.15.8. In H. D. Isenberg (ed.), Clinical microbiology procedures handbook. American Society for Microbiology, Washington, D.C. 11. Marquette, C. H., H. Georges, F. Wallet, P. Ramon, F. Saulnier, R. Neviere, D. Mathieu, A. Rime, and A. B. Tonnel. 1993. Diagnostic efficiency of endotracheal aspirates with quantitative bacterial cultures in intubated patients with suspected pneumonia. Am. Rev. Respir. Dis. 148:138–144. 12. Meduri, G. U., and J. Chastre. 1992. The standardization of bronchoscopic techniques for ventilator-associated pneumonia. Chest 102(Suppl):557S– 564S. 13. Morris, A. J., D. C. Tanner, and L. B. Reller. 1993. Rejection criteria for endotracheal aspirates from adults. J. Clin. Microbiol. 31:1027–1029. 14. Peddecord, K. M., E. J. Baron, D. P. Francis, and A. S. Benenson. 1994. Microbiology laboratory quality requirements: how do infectious diseases specialists rate their laboratories?, abstr. C-172, p. 520. In Abstracts of the 94th General Meeting of the American Society for Microbiology 1994. American Society for Microbiology, Washington, D.C. 15. Peddecord, K. M., E. J. Baron, D. P. Francis, and J. A. Drew. 1996. Quality perceptions of microbiology services: a survey of infectious disease specialists. Am. J. Clin. Pathol. 105:58–64. 16. Peddecord, K. M., L. K. Hofherr, A. S. Benenson, D. P. Francis, J. Rau, D. B. Scharf, Q. Xie, and R. N. Taylor. 1994. Physician requirements for quality in T-lymphocyte subsets, abstr. 1, p. 11, session 1017. In Abstracts of the 122nd Annual Meeting of the American Public Health Association. American Public Health Association, Washington, D.C. 17. Peddecord, K. M., L. K. Hofherr, A. S. Benenson, R. S. Garfein, D. P. Francis, J. L. Cross, and J. Rau. 1993. Use of a physician survey to identify opportunities for quality improvement. Clin. Lab. Sci. 6:110–115. 18. Pedler, S. J., and A. J. Bint. 1991. Survey of users’ attitudes to their local microbiology laboratory. J. Clin. Pathol. 44:6–9. 19. Reller, L. B., P. R. Murray, and J. D. MacLowry. 1982. Cumitech 1A, Blood cultures. Coordinating ed., J. A. Washington II. American Society for Microbiology, Washington, D.C. 20. Scerpella, E. G., P. C. Okhuysen, J. J. Mathewson, R. L. Guerrant, E. Latimer, D. Lyerly, and C. D. Ericsson. 1994. Evaluation of a new latex agglutination test for fecal lactoferrin in travelers’ diarrhea. J. Travel Med. 1:68–71. 21. Schifman, R. B., and F. Meier. 1991. Q-Probes: sputum specimen adequacy; data analysis and critique. College of American Pathologists, Chicago. 22. Swenson, J. A., J. A. Hindler, and L. R. Peterson. 1995. Special tests for detecting antibacterial resistance, p. 1356–1367. In P. R. Murray, E. J. Baron, M. A. Pfaller, F. C. Tenover, and R. H. Yolken (ed.), Manual of clinical microbiology. ASM Press, Washington, D.C. 23. Valenstein, P. N., M. A. Pfaller, and M. Yungbluth. 1995. Q-Probe: stool microbiology; data analysis and critique. College of American Pathologists, Chicago. 24. Washington, J. A. 1993. Evolving concepts on the laboratory diagnosis of septicemia. Infect. Dis. Clin. Pract. 2:65–69. 25. Young, C., W. Hubbard, and K. D. McClatchey. 1992. Implementation and impact of new antimicrobic susceptibility order procedure for improved utilization of microbiology resources, abstr. 1609, p. 379. In Program and abstracts of the 32nd Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C.
