Robert L. Schmidt, MD, PhD, MBA, Jeanne Panlener, MT(ASCP), and Jerry W. Hussong, DDS, MS, MD

Transcription

1 An Analysis of Clinical Consultation Activities in Clinical Pathology Who Requests Help and Why Robert L. Schmidt, MD, PhD, MBA, Jeanne Panlener, MT(ASCP), and Jerry W. Hussong, DDS, MS, MD From the Department of Pathology, University of Utah School of Medicine and ARUP Laboratories, Salt Lake City, UT. Key Words: Consultation; Clinical pathology; Service design Am J Clin Pathol September 2014;142: ABSTRACT Objectives: To examine the distribution of callers who made consultation requests and to identify associations between caller categories and consultation topics. Methods: Review of prospectively collected database of consultations. Results: Direct care personnel made more consultation requests than non direct care personnel. Consultation topics varied by caller type. Direct care personnel requested more consultations on test interpretation and few consultations on test selection than laboratory personnel. Differences in consultation requests by primary care physicians and specialists were significant. Conclusions: At our laboratory, consultation requests primarily originate from primary care physicians. Consultation requests vary by caller type. The role of pathologists is evolving and the College of American Pathologists has initiated a Transformation Program to explore the future role of pathologists. 1 One part of that initiative has been to identify, model, and analyze the impact and economics of these new roles for pathologists. 2 Previous discussions on the roles of pathologists have focused on attributes of successful pathologists 3,4 and training needs for residents Consensus is developing on the need for pathologists to advance and expand their roles as consultants and active contributors to the patient s medical team. 19,20 Clinical consultation has been identified as a critical activity of pathology. Although considerable attention has been placed on the opportunities to expand pathologists roles through consultation, little is known about the characteristics of requested pathology consultations. Consultation is a service offering and, as with any service, it is critical to identify the customers and to understand their needs to define and improve the service. Several studies have described the nature of client questions (client needs) but, to our knowledge, no studies have described the customers or determined whether different customers have different needs As a large reference laboratory, we are in a unique position to study this issue. We provide a wide range of tests, our test volume is high, and tests are sent from a wide range of client types with a national distribution. We provide a consultation service called Path on Call (POC) to support our testing program, and we keep a record of all consultations that go through this service. We conducted a retrospective review and content analysis of our consultation records to characterize the customers who ask for consultations and to determine whether 286 Am J Clin Pathol 2014;142: Downloaded 286 from

2 consultative topics differ by customer type. Our objective was to collect information about our consultation customers and their questions so that we could better meet their needs. Because of the national reach of our laboratory, the results of our study may be applicable to consultations in a wide range of laboratories. Materials and Methods Case Selection Triage cases of the POC Service at ARUP Laboratories (Salt Lake City, UT) seen between July 1 and November 15, 2013, were eligible for inclusion. Cases were included in this study if the caller type and consultation topic could be classified. Customer requests for assistance are initially received by client services. All requests are received by phone. Calls regarding nonmedical issues (sample requirements, tracking, logistics, etc) are generally handled by client services or referred to laboratory personnel. Calls that require medical input (eg, questions regarding test selection, interpretation, or performance) are referred to POC. Calls to POC are received by a service coordinator (medical technologist) who evaluates whether the call is appropriate for POC and assigns calls accepted for triage to the appropriate specialists. The POC service handles approximately 300 calls per month. All calls were recorded prospectively as we routinely do through our POC service. The POC triage process at ARUP Laboratories has been described in detail elsewhere. 26 In brief, the POC service is staffed by a resident during a 3-month clinical chemistry rotation. The resident handles all calls during the 3-month period and is supported by staff pathologists. POC is a central component of the clinical chemistry rotation at our institution. Overall, the service involves a coordinator (1 fulltime equivalent [FTE]), the clinical chemistry resident (1 FTE), and supporting staff pathologists (approximately 0.1 FTE). Caller Classification Callers were initially classified into two categories: direct medical care and non direct medical care professionals Figure 1. Direct medical care professionals included physicians and nonphysicians (physician assistants, nurse practitioners, nurses). Direct medical care physicians were classified as primary care (family practice, pediatricians, and internists) or specialists. We did not collect information on specialty type. Non direct care personnel were classified as doctoral-level scientists (pathologists or PhD) or nondoctoral level. Nondoctoral non direct care personnel were classified as laboratory scientists (medical technologist/clinical laboratory scientist) or administrative personnel. Contact details of the callers (name, client) were collected by client services before being assigned to POC triage. Callers were asked for their position title during the consultation with POC. At our institution, most calls are handled by the resident on the clinical chemistry rotation. In some cases, position titles were determined retrospectively by internet query. Consultation Topic Classification Consultation topics were classified according to the scheme presented by Prak et al. 14 Case notes were recorded in a database (Salesforce, San Francisco, CA) immediately after each consultation. Each case was retrospectively summarized by the POC coordinator on the date that the case was closed. Topics were also classified according to medical subject (chemistry, microbiology, immunology, molecular, and other). Statistical Analysis All analyses were performed using Stata 13 software (StataCorp, College Station, TX). Cross classifications were analyzed using the Pearson c 2 test. Statistical tests were performed at an a priori defined significance level of 5%. Results Consultations were analyzed both by topic type and caller type. We included 1,384 calls that covered 1,694 consultation topics. Most calls (80%) involved one consultation topic; however, 18% of cases involved two topics, and 2% involved three or more topics. One hundred seventy-two calls were omitted because caller information was not collected. The distribution of caller types is presented in Figure 1. Sixtyseven percent of consultations (n = 929) were requested by direct care personnel (physicians, physician assistants, nurse practitioners, and nurses). Of the requests made by direct care personnel, 81% were made by physicians. Among the direct care physicians, 73% were primary care physicians and 27% were specialists. Of the requests made by non direct care personnel, 32% were made by doctoral-level personnel. Of the calls made by nondoctoral non direct care personnel, 70% were made by laboratory scientists and 30% were made by laboratory administrative personnel. Doctoral-level laboratory personnel accounted for 11% of the total callers. The distribution of consultation topics varied by caller type Figure 2. Direct care personnel requested more consultations on test interpretation and fewer consultations on test selection compared with non direct care personnel (c 2 (2) = 67.9; P <.001). Laboratory administration personnel requested more consultations on test selection and fewer consultations on test interpretation than laboratory technologists (c 2 (2) = 12.7; P =.002). No significant differences were seen in consultation request patterns for direct care physicians vs nonphysicians Downloaded from Am J Clin Pathol 2014;142:

3 Schmidt et al / Analysis of Clinical Pathology Consultations All cases (n = 1,384) (100%) 67% 33% Direct care personnel (n = 929) (67%) Nondirect care (n = 455) (33%) 81% 19% 68% 32% Physician (n = 750) (54%) Nonphysician (n = 179) (13%) Nondoctoral (n = 309) (22%) Doctoral (n = 146) (11%) 73% 27% 30% 70% Primary care physician (n = 547) (40%) Specialist (n = 203) (15%) Lab administrator (n = 92) (7%) Lab scientist (n = 217) (16%) Figure 1 Classification of callers. The first branch divides callers into direct care and other (non direct care personnel). Direct care personnel are divided into physicians and nonphysicians. Non direct care personnel are divided into doctoral and nondoctoral. Direct care physicians are divided into primary care and specialist physicians. Non direct care personnel are divided into physicians (pathologists) and nonphysicians (midlevels, nurses). Non direct care nonphysicians are divided into laboratory technicians and other laboratory administrators. The numbers on the arrows indicate the relative percentage of callers from the level above. For example, 81% of consultation requests from direct care personnel were from physicians. The percentages in the boxes are the number of cases relative to the total. For example, 67% of cases (929/1,384) were from direct care personnel. Percentages may not add up to 100 due to rounding. (c 2 (2) = 3.5; P =.17). or primary care physicians vs specialists (c 2 (2) = 0.4; P =.81). There were significant differences in consultation requests among physician categories (c 2 (2) = 17.3; P =.002). In general, doctoral level non direct care personnel tended to ask more questions about test performance and fewer questions about test selection than direct care physicians. The majority of consultations (84%) regarding test selection were concerned with selection of the best test. This pattern was consistent across caller types. Similarly, 96% of consultation requests regarding test interpretation involved discussing the meaning of the test results for a particular patient. This pattern was also consistent across caller types. Calls regarding test performance were mostly concerned with reference intervals (36%) and interferences (50%). The majority of consultations (51%) were concerned with chemistry analyses Figure 3. The remainder was divided among immunology (18%), microbiology (20%), molecular testing (9%), and other miscellaneous tests (2%). Direct care physicians ordered more molecular tests than nonphysicians (c 2 (2) = 11.3; P =.01). Discussion Our study is the first to define the customers for pathology consultations and to identify a relationship between caller types and consultation queries. We found that the rate of consultation requests was 2.6 times greater from primary care physicians than from specialists. The rate of consultation request is surprising because primary care physicians comprise only one third of physicians involved in direct patient care. 27 If the percentage of primary care physicians in our customer base is similar to the distribution in the United States, our data suggest that the rate of consultation requests by primary care physicians would be approximately five times the rate of consultation requests by specialists. This difference may be because specialists order a relatively restricted range of tests and have deeper knowledge of the tests they order. We found some differences between caller categories and consultation topic requests. It is not surprising that pathologists would be more concerned with aspects of test performance than test selection. Similarly, direct care 288 Am J Clin Pathol 2014;142: Downloaded 288 from

4 All topics (n = 1,694) Selection, Interpretation, Performance (n = 465 [27%], n = 1,043 [62%], n = 186 [11%]) 68% 32% Direct care personnel (n = 1,150) (n = 254 [22%], n = 784 [68%], n = 112 [10%]) Nondirect care (n = 544) (n = 211 [39%], n = 259 [48%], n = 74 [14%]) 81% 19% 66% 34% Physician (n = 932) (n = 216 [23%], n = 625 [67%], n = 91 [10%]) Nonphysician (n = 218) (n = 38 [17%], n = 159 [73%], n = 21 [10%]) Nondoctoral (n = 359) (n = 151 [42%], n = 160 [45%], n = 48 [13%]) Doctoral level (n = 185) (n = 60 [32%], n = 99 [54%], n = 26 [14%]) 72% 28% 28% 72% Primary care physician (n = 672) (n = 156 [23%], n = 453 [67%], n = 63 [9%]) Specialist (n = 260) (n = 60 [23%], n = 172 [66%], n = 28 [11%]) Lab administrator (n = 100) (n = 57 [57%], n = 33 [33%], n = 10 [10%]) Lab scientist (n = 259) (n = 94 [36%], n = 127 [49%], n = 38 [15%]) Figure 2 Distribution of queries by caller type. n designates the total number of queries (topics) for the caller type. The numbers in the first set of parentheses shows the distribution of call topics among test selection, test interpretation, and test performance. The second set of parentheses shows the relative percentage of topics for each caller type. For example, direct care physicians raised 932 queries: 216 for test selection, 625 on test interpretation, and 91 regarding test performance. Queries regarding test interpretation accounted for 23% (216/932) of direct care physician queries. Percentages may not add up to 100 due to rounding. personnel asked more questions about test interpretation and fewer questions on test selection than laboratory personnel. Consultation requests were not evenly distributed within categories. We found that consultation requests regarding test selection were mostly (84%) concerned with the initial test selection (selecting the best test) and relatively few were concerned with the sequence and timing of tests or suggestion of alternative tests. Similarly, we found that 96% of consultations regarding test interpretation were concerned with the interpretation of the result for a particular patient. There were very few requests for consultations regarding changes in results for a particular patient (2.7%) or for comparing results performed at different laboratories (1.6%). The requests regarding test performance were relatively infrequent (11% of all queries) but were almost evenly distributed between consultations on reference intervals (41%) and on interferences (56%). These results suggest that the categorization of consultations suggested by Prak et al 14 may need to be modified because the frequencies of some consultation categories are low. One of the limitations of our study is that we recorded the person who placed the call rather than the person who had the question. In our experience, many consultation requests are secondary. For example, a physician may ask a nurse or laboratory technologist to place a call to find the answer to a question. Thus, many of the queries posed by non direct care personnel may have been initiated by direct care personnel. Thus, our estimate of the percentage of queries initiated by physicians is probably low. Finally, our institution is a national reference laboratory and the distribution of call topics and caller types may not be representative of those seen in other contexts. In summary, we provide data on the requests for consultation to pathologists. Previous studies have provided data on the topics of such consultations. Our study expands on previous studies by identifying the requestor and showing that consultation topics vary by requestor type. This type of analysis could be used by pathologists to develop processes to meet the informational needs of particular groups of clients. Downloaded from Am J Clin Pathol 2014;142:

5 Schmidt et al / Analysis of Clinical Pathology Consultations All topics (n = 1,694) Chem, Immuno, Micro, Molecular, Other (51%, 18%, 20%, 9%, 2%) 68% 32% Direct care personnel (n = 1,150) (51%, 18%, 21%, 9%, 2%) Nondirect care (n = 544) (51%, 16%, 19%, 11%, 3%) 81% 19% 66% 34% Physician (n = 932) (49%, 19%, 20%, 10%, 2%) Nonphysician (n = 218) (56%, 15%, 24%, 4%, 1%) Nondoctoral (n = 359) (51%, 16%, 20%, 11%, 3%) Doctoral level (n = 185) (50%, 16%, 18%, 11%, 5%) 72% 28% 28% 72% Primary care physician (n = 672) (50%, 16%, 23%, 9%, 2%) Specialist (n = 260) (48%, 26%, 13%, 12%, 1%) Lab administrator (n = 100) (45%, 19%, 15%, 15%, 6%) Lab scientist (n = 259) (53%, 14%, 22%, 10%, 1%) Figure 3 Distribution of topics by caller type. n designates the total number of queries (topics) for the caller type. The numbers in parentheses show the relative percentage of topics for each caller type. The percentages are listed as shown in the top box (Chemistry [Chem], Immunology [Immuno], Microbiology [Micro], Molecular, Other). For example, direct care physicians raised 932 queries. Of these queries, 49% were related to chemistry tests and 19% were about immunology. Percentages may not add up to 100 due to rounding. Although specialists outnumber primary care physicians, our analysis suggests that there is a particular need for processes that support test ordering by primary care physicians. Our analysis shows the importance of data collection to support the design and improvement of consultation processes. Address reprint requests to Dr Schmidt: Dept of Pathology, University of Utah School of Medicine, 50 N Medical Dr E, Salt Lake City, UT 84112; robert.schmidt@hsc.utah.edu. References 1. College of American Pathologists. Transforming Pathologists. Northfield, IL: College of American Pathologists College of American Pathologists. The Case for Change. Northfield, IL: College of American Pathologists Horowitz RE. Expectations and essentials for the community practice of pathology. Hum Pathol. 2006;37: Horowitz RE. The successful community hospital pathologist: what it takes. Hum Pathol. 1998;29: Kass ME, Crawford JM, Bennett B, et al. Adequacy of pathology resident training for employment: a survey report from the Future of Pathology Task Group. Arch Pathol Lab Med. 2007;131: Talbert ML, Ashwood ER, Brownlee NA, et al. Resident preparation for practice: a white paper from the College of American Pathologists and Association of Pathology Chairs. Arch Pathol Lab Med. 2009;133: Zaino RJ, Abendroth CS. Preparing residents for practice: an opportunity to teach professionalism and communication skills as well as diagnostic criteria. Hum Pathol. 2010;41: Klatt EC. Resident preparation for practice. Arch Pathol Lab Med. 2010;134: Copeland AR. Adequacy of pathology resident training for employment: a survey report from the Future of Pathology Task Group [3]. Arch Pathol Lab Med. 2007;131: Rinder HM, Smith BR. Innovative methods in laboratory medicine resident teaching. Clin Lab Med. 2007;27: Pearson MJ. The clinical role of clinical biochemists. Ann Clin Biochem. 1999;36: Peerschke EIB, Agrawal Y, Alexander CB, et al. Proposed research training guidelines for residents in laboratory medicine. Clin Lab Med. 2007;27: Am J Clin Pathol 2014;142: Downloaded 290 from

6 13. Prak ETL, Park J, Yu G, et al. Point: developing a curriculum in clinical pathology. Clin Chem. 2006;52: Prak ETL, Young DS, Kamoun M, et al ACLPS panel discussion on resident education in clinical pathology. Am J Clin Pathol. 2009;131: Smith BR, Wells A, Alexander CB, et al. Curriculum content and evaluation of resident competency in clinical pathology (laboratory medicine): a proposal. Clin Chem. 2006;52: Scott MG, Bruns DE. Improving training in laboratory medicine. Clin Chem. 2006;52: Wells A, Smith B. Counterpoint: developing a clinical pathology curriculum to meet current and future needs. Clin Chem. 2006;52: Wells A, Smith BR, Sacks DB. The challenge of training pathologists in the 21st century. Hum Pathol. 2006;37: Schuerch C, Selna M, Jones J. Laboratory clinical effectiveness: pathologists improving clinical outcomes. Clin Lab Med. 2008;28: Hess JL. What are we anyway? the role of pathologists in the 21st century. Arch Pathol Lab Med. 2010;134: Buck TP, Connor IM, Horowitz GL, et al. Tracking resident calls to improve clinical utilization of pathology laboratories. Arch Pathol Lab Med. 2011;135: Do MC, Ben-Ezra J, McPherson RA. Call subject patterns among on-call clinical pathology residents in an academic institution: how can tracking changes in patterns over time benefit resident education? Arch Pathol Lab Med. 2008;132: Hoofnagle AN, Chou D, Astion ML. Online database for documenting clinical pathology resident education. Clin Chem. 2007;53: Hobbs GA, Jortani SA, Valdes R Jr. Implementation of a successful on-call system in clinical chemistry. Am J Clin Pathol. 1997;108: Pappas AA, Drew MJ, Flick J, et al. An introductory orientation to clinical pathology core and on-call responsibilities. Arch Pathol Lab Med. 1994;118: Schmidt RL, Garcia CA, Panlener J, et al. An analysis of clinical consultation activities in clinical chemistry: implications for transformation and resident training. Arch Pathol Lab Med. 2014;138: Agency for Healthcare Research and Quality. The Number of Practicing Primary Care Physicians in the United States pcwork1/index.html. Accessed December 26, Downloaded from Am J Clin Pathol 2014;142: