Information-seeking behavior and use of information resources by ...

Information-seeking behavior and use of information resources by clinical research coordinators By Charles B. Wessel, MLS [email protected] Coordinator of Affiliated Hospital Services Nancy H. Tannery, MLS [email protected] Associate Director for Information Services Barbara A. Epstein, MSLS, AHIP [email protected] Director Health Sciences Library System Falk Library of the Health Sciences 200 Scaife Hall 3550 Terrace Street University of Pittsburgh Pittsburgh, Pennsylvania 15261

Purpose: The study sought to understand the literature searching experiences and skills of clinical research coordinators at a large academic medical center. Setting/Participants/Resources: The Health Sciences Library System, University of Pittsburgh, conducted a survey of clinical research coordinators at the University of Pittsburgh and the University of Pittsburgh Medical Center to solicit their perceived use and knowledge of the library’s electronic resources. Brief Description: The University of Pittsburgh Institutional Review Board (IRB) is a ‘‘high volume IRB’’ that monitors human subject research at both the University of Pittsburgh and the University of Pittsburgh Medical Center. More than 3,500 human research studies and clinical trials are active at any given time. Many studies entail more than minimal risk to human subjects, with the majority evaluating or including a drug or medical device. Clinical research coordinators are involved in most of these studies or trials. Their roles and responsibilities focus on managing many aspects of the study or clinical trial. As a first step in understanding the literature searching experiences and skills of these research coordinators, baseline data were gathered from this group in November 2004. Results/Outcome: The data from this survey indicate that clinical research coordinators are a population who would benefit from training by academic medical center librarians in how to use electronic library resources and services. Evaluation Method: A Web-based survey solicited participants’ information (gender, education, job title) and role in the IRB process (job responsibilities, number studies they manage). The majority of the survey questions focused on the use of specific electronic library resources, the type of information wanted, and the types of problems encountered.

48

J Med Libr Assoc 94(1) January 2006

Clinical research coordinators

INTRODUCTION Clinical research coordinators (CRCs) play a crucial role in developing, organizing, and administering many aspects of human research studies and clinical trials taking place at academic medical centers [1–4]. Though the specific duties of CRCs vary, their role is to work closely with principal investigators to provide daily oversight of research studies. Most CRCs are nurses, but the group can also include administrators, data coordinators, nurse practitioners, individuals with philosophy and pharmacy doctorates, physicians, and social workers. Their responsibilities generally involve some or all of the following: n grant, budget, and program development; n preparation and submission of Institutional Review Board (IRB) applications and reports; n audits of IRB protocols; n data collection, analysis, and monitoring; n case management of patients; n recruitment and enrollment of human subjects; n auditing and reporting of adverse events; n maintenance of drug records; and n education of patients, family, and other health care professionals [2]. The events surrounding the death of a healthy volunteer in an asthma study funded by the National Institutes of Health demonstrate the importance of responsible and comprehensive literature searching to protect the safety of human subjects participating in research studies [5]. It has become evident after investigation of the incident that many researchers conduct literature searches on their own, with little guidance on what constitutes an appropriate or sufficient literature search to support human subject research [6]. The exponential growth of print and electronic information mandates that researchers be knowledgeable in how to identify and use a broad array of information tools. MEDLINE is just one of many tools used to identify primary scientific studies [7, 8]. Physicians’ information needs, information-seeking behavior, and use of online resources have been described in detail [9–12]. Less is known about the information needs and use of electronic library resources by researchers [13, 14]. The information-seeking behavior and need studies employ a variety of methods of investigation and data collection, including mail and telephone surveys [13]. With CRCs playing a significant role in human research studies and the IRB process, it has become evident that little is known about their information needs, information-seeking behavior, and use of electronic library resources and information tools.

of Pittsburgh Medical Center/UPMC. ‘‘High-volume IRBs handle thousands of actions per year, while lowvolume IRBs typically handle fewer than 125’’ [15, 16]. More than 3,500 human research studies and clinical trials are active at the university and/or UPMC at any given time. Many studies entail more than minimal risk to human subjects, and the majority involve evaluation or inclusion of a drug or medical device. CRCs are involved in most of these studies or trials, with the roles and responsibilities described above. The University of Pittsburgh Health Sciences Library System (HSLS) ,http://www.hsls.pitt.edu. meets the information needs of the CRCs by providing licensed electronic library resources as well as a comprehensive print collection. The electronic library resources offer onsite and remote access to approximately thirty Web-based databases, such as MEDLINE, Web of Science, EMBASE, drug and disease full-text databases, and evidence-based health care resources. The library’s electronic collection includes access to more than 3,000 electronic journals and more than 600 electronic books. In 2004, HSLS developed Responsible Literature Searching for Research (RLS): A Self-Paced Interactive Educational Program, available on the university’s Research Practice Fundamentals Website [17].* The objectives of the RLS program are to describe and teach generally accepted practices and principles associated with the biomedical literature search process and to promote awareness of the available electronic library resources and services that can provide assistance in responsible searching of the literature. In November 2004, before the introduction of RLS at the university, HSLS gathered baseline data to understand the CRCs’ literature searching experiences and skills. The intention of this baseline data is to assist in measuring the training impact of this program, because CRCs are a prime audience for its content. HSLS created a brief Web-based survey (Appendix) to solicit information about CRCs’ use of specific electronic library resources, types of information searched, and types of problems encountered. Information about demographics (gender, education level, job title) and role in the IRB process (job responsibilities, number studies they manage) was also collected. A review of the literature did not provide a validated survey instrument to meet the study’s needs. The survey was tailored specifically to measure use of locally provided electronic information resources. The Pitt Research Network ,http://www.rcco.pitt.edu/educ/ECOpPRN.htm., an umbrella group of individuals involved in research administration, sent a global email to its distribution list asking CRCs to respond to this Web-based survey.

SETTING/METHOD The IRB of the University of Pittsburgh ,http:// www.irb.pitt.edu. is a ‘‘high volume IRB’’ established to protect the rights and welfare of human subjects at both the University of Pittsburgh and the University J Med Libr Assoc 94(1) January 2006

* Development of Responsible Literature Searching for Research was sponsored by the Association of Academic Health Sciences Libraries and funded by the Association of American Medical Colleges and the Office of Research Integrity of the United States Department of Health and Human Services.

49

Wessel et al.

Table 1 Clinical research coordinators (CRCs) search the medical literature for the following types of information* (n 5 46) Type of information 1. Journal articles on a particular topic 2. Drug information 3. Disease-specific information 4. A specific journal article 5. The specific source of the information 6. Adverse event literature 7. Other information Never search the medical literature

% (number of responses) 24% 19% 17% 17% 11% 8% 1% 1%

(35) (27) (25) (25) (16) (11) (2) (2)

RESULTS The global email was sent to 150 participants in the distribution list. A follow-up email reminder was sent to get additional responses to the survey. Forty-nine percent of the recipients of the email message (74 individuals) linked to the survey from the email message, and 31% (46 individuals) finished the entire survey. Of the 46 people who completed the survey, 96% were female and 4% were male. These 46 individuals had a variety of job titles, including education and compliance coordinator, senior program coordinator, clinical research coordinator, clinical nurse specialist, research nurse coordinator, project manager, research coordinator, and other similar titles. The highest educational degree was a bachelor degree for 52% of the respondents, a master’s degree for 30% of the respondents, a doctorate for 9% respondents, a medical degree for 2% of the respondents, and some college coursework or a registered nurse diploma for 6% of the respondents. When asked about their role in the IRB submission process, 85% of respondents assisted with IRB applications, 83% completed the required paperwork, 56% assisted in preparing the initial protocol, 33% did the literature searching for the protocol, and 41% had other responsibilities. The other responsibilities included adverse event reporting, clinical nursing care, statistical and data management, recruitment, and oversight of study coordinators. On average, each of these 46 individuals managed 5 research studies. When asked how often in a average month, the CRCs consulted the medical literature to do their jobs, 50% responded that they consulted the literature 1 to 5 times a month, 20% responded 6 to 10 times a month, 15% responded 11 to 25 times a month, and 9% responded more than once a day. Only 6% responded that they never consulted the medical literature. The type of medical information typically sought is listed in Table 1. In decreasing order of frequency, these were: (1) journal articles on a particular topic, (2) drug information, (3) disease-specific information, (4) specific journal articles, (5) specific sources of information, (6) adverse event literature, and (7) other information. 50

Table 2 CRCs’ confidence levels searching Ovid MEDLINE and PubMed* (n 5 46) % (number of respondents) Confidence level Not confident Somewhat confident Confident Extremely confident

Ovid MEDLINE 15% 28% 28% 17%

(7) (13) (13) (8)

PubMed 22% 28% 24% 15%

(10) (13) (11) (7)

* 10% (5) of the respondents never search Ovid MEDLINE or PubMed.

When asked on average, how many hours a month they personally searched Ovid MEDLINE or PubMed, 39% of respondents spent 1 to 5 hours per month, 28% spent 6 to 15 hours, 4% spent 16 or more hours, and 28% spent 0 hours in this activity. The CRCs’ confidence in their ability to find medical literature was measured. Sixty-six percent of the respondents felt ‘‘confident’’ or ‘‘somewhat confident,’’ 26% were ‘‘extremely confident,’’ and 7% were ‘‘not confident’’ in their ability to find medical literature. The CRCs’ confidence level in their searching ability using Ovid MEDLINE and PubMed reflected a higher level of discomfort. When using Ovid MEDLINE, 28% were ‘‘somewhat confident’’ and 15% were ‘‘not confident.’’ When using PubMed, 28% were ‘‘somewhat confident’’ and 22% were ‘‘not confident.’’ Table 2 illustrates confidence levels searching Ovid MEDLINE and PubMed. The survey solicited their responses to a list of options to indicate their specific use of and encounters with Ovid MEDLINE and PubMed. A number of respondents (29%) were able to find the necessary article but spent more time than they wanted on the task, while 24% found articles in a timely manner. Table 3 illustrates their encounters with and use of Ovid MEDLINE and PubMed. When asked if there were times over the last month they were unable to find needed information from the medical literature to complete a job-related task, 39% responded ‘‘yes’’ and 61% responded ‘‘no.’’ The top 3 information tasks the CRCs were unable to complete were: (1) locate a specific journal article, (2) locate specific information on a topic in Ovid MEDLINE or PubMed, and (3) find drug information. Table 4 demTable 3 CRCs’ encounters and use of Ovid MEDLINE and PubMed (n 5 45) Encounter and use 1. Find the articles I need but it takes more time than I want to spend 2. Find the articles I need in a timely manner 3. Never have the time to use Ovid MEDLINE or PubMed 4. Usually get too many articles 5. Have trouble selecting the correct keywords 6. Have difficulty finding the articles I need 7. Other problems 8. Usually do not find enough articles 9. Never encounter any problems

% (number of responses)

29% 24% 10% 9% 9% 7% 7% 3% 1%

(20) (16) (7) (6) (6) (5) (5) (2) (1)



Table 4 Information tasks the CRCs were unable to complete* (n 5 46) Information task Locate a specific journal article Locate specific information on the topic from Ovid MEDLINE or PubMed Find drug information Other problems Due to time constraints, I never located the information A specific adverse event Locate the source of the information

% (number of responses) 23% (14) 15% 10% 7% 5% 3% 3%

(9) (7) (4) (3) (2) (2)

* 33% (20) indicate they never have any problems finding information.

onstrates the breakdown of the information problems encountered by the CRCs. CRCs were asked about their use of specific electronic library resources. Table 5 indicates which electronic library resources were ‘‘frequently used’’ and which were ‘‘infrequently used’’ by this population. Frequently used resources were those that at least 50% of respondents used ‘‘often’’ or ‘‘sometimes.’’ These included Ovid MEDLINE, electronic journals, and PubMed. Infrequently used resources were those that at least 50% of respondent said they ‘‘never use’’ or were ‘‘unfamiliar with.’’ These included electronic books, evidence-based health care resources, Micromedex, drug and pharmacy databases, Web of Science, Biological Abstracts, and EMBASE.com.

DISCUSSION Survey results indicate that 93% of CRCs consult the medical literature 1 or more times a month to do their job, and 72% personally search Ovid MEDLINE or PubMed more than 1 hour a month. Even with their high use of the medical literature, 39% feel only ‘‘somewhat confident’’ or ‘‘not confident’’ in their ability to locate medical literature. The information tasks listed in Table 4 indicate the difficulty perceived by CRCs in completing specific information tasks or finding relevant information. Their use of Ovid MEDLINE and PubMed, listed in Table 3, indicates problems in locating relevant medical literature. These problems include time factors, trouble selecting appropriate keywords for the literature search, and difficulties associated with balancing precision and recall [18]. Even though the ‘‘frequently used’’ electronic library resources (Ovid MEDLINE, PubMed, and electronic journals) are significant, many important resources such as Web of Science, EMBASE.com, Micromedex, and OLDMEDLINE are infrequently used by or unfamiliar to this population. As described by Lefebrve, MEDLINE is no longer the only tool researchers need to use to locate relevant studies [8]. Databases, such as EMBASE.com and Micromedex, and citation-searching databases, such as Web of Science, are extremely useful information tools that should be consulted by CRCs and other research professionals for comprehensive literature reviews to support human subject research.

Table 5 CRCs’ use of the Health Sciences Library System’s (HSLS) electronic resources (n 5 46)


51

Wessel et al.

CONCLUSION The data from this brief survey indicate opportunities and challenges for academic medical center libraries in training and marketing their services to this unique population. CRCs, as demonstrated, play a significant role in clinical research, with many helping to protect human subjects from harm. Given their responsibilities, this population may respond positively to initiatives such as librarian IRB programs [19], tailored library instruction, and marketing of the value of the librarian expert searcher [20, 21]. REFERENCES 1. ANONYMOUS. The role of a research coordinator. Dimens Crit Care Nurs 2003 Nov–Dec;22(6):270–1. 2. ROBERTSON C. Research coordinators—who are they and what do they do? Axone 2004 Jun;25(4):4–5. 3. PELKE S, EASA D. The role of the clinical research coordinator in multicenter clinical trials. J Obstet Gynecol Neonatal Nurs 1997 May–Jun;26(3):279–85. 4. RICO-VILLADEMOROS F, HERNANDO T, SANZ JL, LOPEZALONSO A, SALAMANCA O, CAMPS C, ROSELL R. The role of the clinical research coordinator—data manager—in oncology clinical trials. BMC Med Res Methodol 2004 Mar 25;4:6. 5. STEINBROOK R. Protecting research subjects—the crisis at Johns Hopkins. N Engl J Med 2002 Feb 28;346(9):716–20. 6. DANG CV, SCHAFFER DV. Letter to: Michael A. Carome, Office for Human Research Protections. 2001 Jul 13. [Web document]. Baltimore, MD: Johns Hopkins University. [13 Jul 2001; cited 5 Nov 2004]. ,http://www.hopkinsmedicine .org/press/2001/JULY/letterptopOHRP.htm.. 7. MCLELLAN F. 1966 and all that—when is a literature search done? Lancet 2001 Aug 25;358(9282):646. 8. LEFEBVRE C, LUSHER A, DICKERSIN K, MANHEIMER E. Literature searches. Lancet 2002 Mar 9;359(9309):896. 9. HAYNES RB, MCKIBBON KA, WALKER CJ, RYAN N, FITZGERALD D, RAMSDEN MF. Online access to MEDLINE in clinical settings: a study of use and usefulness. Ann Intern Med 1990 Jan 1;112(1):78–84. 10. OSHEROFF JA, FORSYTHE DE, BUCHANAN BG, BANKOWITZ RA, BLUMENFELD BH, MILLER RA. Physicians’ information needs: analysis of questions posed during clinical teaching. Ann Intern Med 1991 Apr 1;114(7):576–81. 11. D’ALESSANDRO DM, KREITER CD, PETERSON MW. An evaluation of information-seeking behaviors of general pediatricians. Pediatrics 2004 Jan;113(1 pt.1):64–9. 12. GORMAN PN, YAO P, SESHADRI V. Finding the answers in primary care: information seeking by rural and nonrural clinicians. Medinfo 2004;11(pt.2):1133–7. 13. DETLEFSEN EG. The information behaviors of life and health scientists and health care providers: characteristics of the research literature. Bull Med Libr Assoc 1998 Jul;86(3): 385–90. 14. MURRAY J, CAREY E, WALKER S. The information needs and information seeking behaviour of medical research staff. Health Libr Rev 1999 Mar;16(1):46–9. 15. BELL J, WHITON J, CONNELLY S, JAMES BELL & ASSOCIATES. Evaluation of NIH implementation of section 491 of the Public Health Service Act, mandating a program of protection for research subjects. [Web document]. The Office of Extramural Research, National Institutes of Health. [15 Jun 1998; cited 5 Nov 2004]. ,http://www.washingtonfax.com/ samples/docs/bioethics/patients/hsppfinalprpt.pdf..

52

16. WAGNER TH, BHANDARI A, CHADWICK GL, NELSON DK. The cost of operating institutional review boards (IRBs). Acad Med 2003 Jun;78(6):638–44. 17. EDUCATION AND CERTIFICATION PROGRAM IN RESEARCH AND PRACTICE FUNDAMENTALS. Module 11: responsible literature searching for research. [Web document]. University of Pittsburgh. Pittsburgh, PA: The University of Pittsburgh, 2005. [20 Jan 2005; cited 20 Jan 2005]. ,http://rpf.health .pitt.edu/rpf/.. 18. POYNTON MR. Recall to precision: retrieving clinical information with MEDLINE. Clin Nurse Spec 2003 Jul;17(4): 182–4. 19. ROBINSON JG, GEHLE JL. Medical research and the institutional review board: the librarian’s role in human subject testing. Ref Serv Rev 2005 Jan;3(1):20–4. 20. SUMMERSKILL W. Literature searchers: look before you leap. Lancet 2005 Jul 2;366(9479):13–4. 21. MEDICAL LIBRARY ASSOCIATION. Role of expert searching in health sciences libraries. J Med Libr Assoc 2005 Jan;93(1): 42–4.

Received February 2005; accepted July 2005 APPENDIX Survey of individuals conducting clinical research projects The purpose of this research study is to determine how individuals conducting clinical research projects look for information and their role in the Institutional Review Board (IRB) process. For this reason, we will be surveying individuals conducting clinical research projects and ask them to complete a five-to-ten minute survey. If you are willing to participate, the survey will ask questions about your background (age, gender, education, job title) as well as your role in the IRB process, how you look for information, and the types of resources you may use to locate information. There are no foreseeable risks associated with this project, nor are there any direct benefits to you. This is an entirely anonymous survey; no responses will be identifiable in any way. Your participation is voluntary, and you may withdraw from completing the survey at any time. Thank you for taking our survey. 1. Your highest educational degree? 2. Your gender □ Male □ Female 3. What is your job title? 4. What is your role in the IRB submission process? □ Preparation of initial protocol □ Literature searching for the protocol □ IRB applications and resubmissions □ Completion of paperwork □ Other responsibilities (please describe): 5. How many studies do you manage? J Med Libr Assoc 94(1) January 2006


6. In an average month, how often do you need to consult the medical literature to do your job? □ 0 times a month □ 1 to 5 times a month □ 6 to 10 times a month □ 11 to 25 times a month □ More than 1 time a day 7. When you are searching for medical literature what type of information are you typically looking for (check all that apply): □ Drug information □ A specific journal article □ Journal articles on a particular topic □ Adverse event literature □ The specific source of the information □ Disease-specific information □ I never search for medical literature □ Other information (please specify): 8. How confident do you feel about your ability to find medical literature? □ Not confident □ Somewhat confident □ Confident □ Extremely confident □ Never do this 9. Thinking back over the past work month, were there times when you needed information from the medical literature to accomplish a job-related task but were unable to find the information? □ Yes □ No If yes, how many times? 10. Thinking back, what were you unable to locate or do (check all that apply)? □ Find drug information □ Find a specific journal article □ Find a specific adverse event □ Locate the source of the information □ Locate specific information on the topic from Ovid MEDLINE or PubMed □ Due to time constraints, I never located the information □ I never have any problems finding information I need □ Other (please specify): 11. In a typical month, how many hours do you personally use Ovid MEDLINE or PubMed on average? □ 0 hours □ 1 to 5 hours □ 11 to 15 hours □ 16 or more hours 12. How confident are you in searching Ovid MEDLINE? □ Not confident □ Somewhat confident □ Confident □ Extremely confident □ Never do this J Med Libr Assoc 94(1) January 2006

13. How confident are you searching PubMed? □ Not confident □ Somewhat confident □ Confident □ Extremely confident □ Never do this 14. When using Ovid MEDLINE or PubMed, I usually (check all that apply): □ Find the articles I need in a timely manner □ Find the articles I need, but it takes more time than I want to spend □ Have difficulty finding the articles I need □ Usually get too many articles □ Usually do not find enough articles □ Have trouble selecting the correct keywords □ Never have the time to use Ovid MEDLINE or PubMed □ Never encounter any problems □ Other problems (please specify:) 15. To what extent do you use PubMed? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 16. To what extent do you use Ovid MEDLINE? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 17. To what extent do you use Biological Abstracts (BIOSIS Previews)? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 18. To what extent do you use Web of Science? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 19. To what extent do you use EMBASE.com? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 20. To what extent do you use drug and pharmacy databases? (IPA, TOXLINE) □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 21. To what extent do you use evidence-based health care resources? (ACP Journal Club, Cochrane databases, DARE) □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 53

Wessel et al.

22. To what extent do you use Micromedex? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 23. To what extent do you use Health Sciences Library System (HSLS) electronic journals? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 24. To what extent do you use HSLS electronic books? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature

54

25. To what extent do you use PubMed’s related article feature? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 26. To what extent do you use the Web of Science cited reference searching feature? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature 27. To what extent do you use OLDMEDLINE? □ Use often □ Use sometimes □ Never use □ Unfamiliar with resource or feature
