J Canc Educ (2014) 29:74–79 DOI 10.1007/s13187-013-0548-z
Patient Enrollment onto Clinical Trials: the Role of Physician Knowledge Justin R. Gregg & Leora Horn & Mario A. Davidson & Jill Gilbert
Published online: 28 October 2013 # Springer Science+Business Media New York 2013
Abstract Sixty-six attending physicians at academic medical centers completed a 43-question self-assessment evaluating communication skills, comfort with clinical trial enrollment, and knowledge of patient-related barriers to enrollment on clinical trials. Responses and demographic information were analyzed for trends and for association with estimated trial enrollment. Physician-described enrollment of patients onto trials varied widely, with estimated enrollment varying from less than 5 patients to well over 125 enrolled during the previous year. Participants perceived themselves to have excellent communication skills and were comfortable with the trial enrollment process, though did not consistently identify patient-related barriers to enrollment. Physician knowledge of clinical trials currently enrolling within their field was associated with increased patient enrollment on study (p =0.03). Academic physicians expressed confidence in their skills related to clinical trial enrollment despite less than ideal reported enrollment. Knowledge of clinical trials currently enrolling within a physician’s specialty was associated with estimated patient enrollment, and may represent a correctable barrier to trial enrollment. Electronic supplementary material The online version of this article (doi:10.1007/s13187-013-0548-z) contains supplementary material, which is available to authorized users. J. R. Gregg (*) Department of Urologic Surgery, Vanderbilt University Medical Center, A – 1302 Medical Center North, Nashville TN 37232-2765, USA e-mail: [email protected] L. Horn : J. Gilbert Department of Medicine, Hematology and Oncology Division, Vanderbilt University Medical Center, Nashville, TN, USA M. A. Davidson Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA
Keywords Clinical Trials . Barriers . Self-Assessment . Knowledge . Enrollment
Introduction Enrollment of patients onto clinical trials is essential to advance the field of medicine; however, many barriers exist related to entry of adult patients onto clinical trials and successful completion of studies [1, 2]. Low patient enrollment into clinical trials leads to delays in trial completion and treatment advances as well as premature trial termination and wasted resources [3, 4]. Getz and colleagues estimated that over 90 % of trials are delayed due to problems with enrollment [5]. Known barriers to clinical trial enrollment include trial availability for individual patients or disease processes, patient-related factors, and physician-related factors [2, 6]. While trial availability relates to a lack of opportunity at a systemic level, patient-related factors are broad and include concerns about trial costs, transportation issues, and fears related to receiving a placebo or suffering from unknown side effects [7, 8]. Notably, underrepresented minorities have particularly low enrollment onto clinical trials and studies have shown that these patients may be particularly affected by a distrust of research and fear of harms related to trials, though not all of these fears have been consistently shown to result in quantitative decreases in enrollment [2]. Additionally, many fears faced by patients, particularly those involving cost, mistreatment, and the likelihood of receiving a placebo, are based on misconceptions [7, 9–11]. Physicians are in a unique position to allay patient fears, though it has also been shown that physicians do not consistently recognize patient-related barriers and concerns with regard to clinical trial enrollment [9, 12]. For this reason, provider perceptions and attitudes toward trials may affect whether a patient is offered a trial [13]. Physician surveys
J Canc Educ (2014) 29:74–79
have also indicated that limited time with a patient, a lack of dedicated staff, and other clinic resources may detract from their ability to enroll patients into trials [10]. When physicians do decide to offer a patient a trial, communication skills and clinical trial knowledge become important factors in presenting a trial to a patient. Good communication skills are a pre-requisite for the successful exchange of information between physicians and patients [14]. Communication is thought to provide the “means” [15] by which barriers associated with patient, physician, and research protocol factors impact an individual patient’s decision making process. It is known that physicians differ in the methods through which they explain clinical trials and treatment options [15]. These differences can impact clinical trial enrollment, as patients’ intentions to enroll into a clinical trial can be affected by whether the provider presents the trial in a positive, negative, or ambiguous manner [16]. Communication skills, while traditionally believed to be inherent to individuals, are increasingly viewed as skills that are best taught through established techniques such as experiential learning and directed feedback [17]. In fact, communication skills have recently been outlined as a core competency of residency training as outlined by the Accreditation Council for Graduate Medical Education [18]. Despite this emphasis, it has been shown that explicit communication skills training were rare in a group of surgical subspecialty residents [17]. To our knowledge, few studies have examined physicianrelated barriers to trial enrollment in academic physicians, including provider knowledge, perceptions, and communication skills. Our aim was to explore academic physician-related barriers to clinical trial enrollment. We also aim to determine factors associated with estimated patient enrollment in clinical trials.
Methods Survey Development We designed a questionnaire utilizing questions related to patient enrollment into clinical trials using two previously designed instruments [11, 17]. The survey included demographic data (eight questions) and questions related to enrollment into clinical trials that were broken down into four categories: Communication Skills (nine items), Clinical Trials—Personal Focus (nine items), Clinical Trials— General Focus (nine items), and Clinical Trials—Enrollment (eight items). Item generation was carried out by three of the authors (JRG, LH, and JG). All questions were rated on a 1–5point Likert scale, ranging from “strongly disagree” to “strongly agree.” A pilot test of the questionnaire was done with a group of five clinical investigators from Vanderbilt’s
75
Division of Hematology/Oncology to evaluate for item redundancy, readability, and content. The full questionnaire minus demographic questions can be reviewed in Appendix 1. Survey Administration The questionnaire was distributed between August and December 2009 to a sample of attending physicians in three different departments with a history of strong involvement in clinical trials in academic medical institutions. Hematology/ Oncology departments that were asked to participate were Vanderbilt University, Memorial-Sloan Kettering, and Harvard-affiliated hospitals. Vanderbilt University’s Division of Cardiology and Division of Radiation Oncology were also asked to participate. Study data were collected and managed using Research Electronic Data Capture (REDCap) hosted at Vanderbilt University [19]. Surveys were distributed both electronically and through the use of paper copies that were later entered into REDCap by a single author (JRG). All study participants signed an informed consent form approved by the Vanderbilt University Institution Review board. Statistical Analysis Data were analyzed using descriptive statistics such as means, median, percentiles, and frequencies on N =66 physicians. Not all physicians completed the questionnaire in its entirety; therefore, the above analysis was completed based on the number of physicians who answered each individual questionnaire item. Chi-squared, Kruskal–Wallis, and Wilcoxon analyses were done to determine associations between physician-estimated patient enrollment and demographic information and survey responses. Patient enrollment was chosen for analysis in order to determine if any physician characteristic or response was related to a lower quantitative estimate of patient enrollment. All statistical tests were twotailed at an alpha level of 0.05. All analysis was done using R version 2.10.1 [20].
Results Overall Study Participant Characteristics Sixty six physicians out of 115 potential participants completed the self-assessment questionnaire (57 % response rate). Physicians from each eligible department completed questionnaires. Demographic data collected in the study are summarized in Table 1. The average age of participants was 47.7 years, 68 % of them were male and 71 % were Caucasian. The majority (81 %) were US medical school graduates. In total, there were 36 hematologists/oncologists
76
J Canc Educ (2014) 29:74–79
Table 1 Demographics of academic physician survey responders
Clinical Trials—Personal Focus
a
Characteristic
Question response
Age, mean (SD) Sex, n (%) Male Female Ethnicity, n (%) White African American Asian/Pacific Islander Nationality, n (%) US medical graduate International graduate Physician degree, n (%) MD/DO Ph.D. MPH Clinical research involvement, n (%) Any previous involvement
47.7 (9.94) 44 (68 %) 21 (32 %) 46 (71 %) 5 (8 %) 14 (22 %) 53 (82 %) 12 (18 %) 62 (100 %) 6 (10 %) 8 (13 %) 57 (93 %)
No prior involvement 4 (7 %) Time dedicated to clinical activities, n (%) Less than 20 % 10 (16 %) Between 20 and 50 % 10 (16 %) Approximately 50 % 18 (29 %) Greater than 50 % 24 (39 %) Estimated clinical trial patient enrollment in past 12 months, n (%) 0–5 21 (34 %) 6–10 13 (21 %) 11–25 12 (20 %) >25 15 (25 %) a
The total number in each category does not equal 66 because complete data were not available from all survey responders
(55 % of study participants), 24 cardiologists (36 %), and 6 radiation oncologists (9 %). Thirty-four percent of the population reported that they enrolled 0–5 patients during the previous 12 months, while 25 % enrolled over 25 patients. Communication Skills Physicians are confident in their communication skills, as 51/ 64 (79.7 %) respondents either agreed or strongly agreed that they had excellent communication skills. Respondents also believed that good communication can improve patient outcomes (59/66 [89.3 %]) and enrollment into clinical trials (60/ 66 [90.9 %]). Physicians differed in their method of learning communication skills during training, with 26/66 (39.4 %) disagreeing and 27/66 (40.9 %) agreeing that they received explicit instruction during residency and fellowship training on how to improve communication skills (Appendix 1).
Participants are also confident and comfortable with clinical trials and the enrollment process. Sixty two of 65 (95.4 %) physicians agreed or strongly agreed that they have an understanding of the informed consent process and are comfortable discussing clinical trial results and consenting patients. When presented with the statement “I have adequate knowledge of clinical trials that are currently enrolling within my specialty,” 10/64 (15.6 %) disagreed or strongly disagreed while 43/63 (67.2 %) agreed or strongly agreed (Appendix 1). Clinical Trials—General Focus Participants agree on the importance of clinical trials, with 61/ 66 (92.4 %) agreeing that trials are the essential means of discovering new therapies. Physicians disagreed, though, on the reasons patients do not enroll in clinical trials, with 12/66 (18.1 %) disagreeing that patients fear potential side effects and 33/66 (50 %) disagreeing that patients fear financial liability, among others (Appendix 1). Clinical Trials—Enrollment Physicians admit to varying levels of bias in terms of which patients they consider for enrollment onto clinical trials, with 46/64 (71.9 %) agreeing or strongly agreeing that they are more likely to attempt to enroll patients in clinical trials if they suffer from an incurable illness (Appendix 1). Statistical Analysis Physician-estimated patient enrollment was then evaluated for association with demographic information and questionnaire item responses (Table 2). Estimated knowledge of currently enrolling clinical trials was significantly associated with increased estimated patient enrollment onto clinical trials during the past year (p =0.03) (Fig. 1). The items “Complicated clinical trial protocols deter me from attempting to enroll patients in clinical trials,” “Patients don’t participate in clinical trials because they fear going through too many tests,” and “I have sufficient time in clinic to explain clinical trials to patients” and whether the respondent held a Ph.D. were significantly associated with estimated patient enrollment (p =0.04, 0.02, 0.03, and 0.03, respectively). All other associations between physician-estimated patient enrollment and questionnaire items did not reach statistical significance.
Discussion The enrollment of patients onto clinical trials is often inadequate, leading to delayed trial completion [21] and thus delayed
J Canc Educ (2014) 29:74–79 Table 2 Physician characteristics associated with Estimated Patient Enrollment, n (%)
a
Pearson Test
77
Estimated Patient Enrollment
p Value
I have adequate knowledge of clinical trials that are currently enrolling participants within my specialty, n (%) Strongly disagree
0–5
5–10
3 (15)
0 (0)
1 (8)
0 (0)
Disagree Neutral Agree Strongly agree Complicated clinical trial protocols deter me from attempting to enroll patients in clinical trials, n (%) Strongly disagree Disagree Neutral Agree Strongly agree Patients don’t participate in clinical trials because they fear going through too many tests, n (%) Strongly disagree Disagree Neutral Agree Strongly agree I have sufficient time in clinic to explain clinical trials to patients, n (%) Strongly disagree Disagree Neutral Agree Strongly agree Degree, n (%) Ph.D. No Ph.D.
2 (10) 7 (35) 4 (20) 4 (20)
2 (15) 1 (8) 5 (38) 5 (38)
0 (0) 1 (8) 3 (25) 7 (58)
0 (0) 2 (13) 1 (7) 12 (80)
medical progress. Our group recently showed that 14 % of patients seen by thoracic medical oncologists at the Vanderbilt Ingram Cancer Center are entered onto a clinical trial; however, 50 % of eligible patients did not enroll on study [22]. Additionally, many physicians in our cohort conceded poor patient enrollment onto clinical trials, with a third of survey participants having enrolled 0–5 patients during the previous year—well below the NIH recommendations of 15 % of patient enrollment onto a clinical protocol [23]. Known barriers to clinical trial enrollment include patient-related and physician-related factors. This study aimed to evaluate academic physician-related barriers to enrollment of patients into clinical trials. The first finding evident in our study is that physicians may not recognize that trial enrollment is a nationwide problem. Those who participated in our study did not agree that patients were reluctant to join clinical trials, as 35 % disagreed or
10–25
>25
0.03a
0.04 2 (10) 5 (24) 7 (33) 7 (33) 0 (0)
1 (8) 3 (23) 4 (31) 3 (23) 2 (15)
7 (58) 3 (25) 1 (8) 0 (0) 1 (8)
3 (20) 5 (33) 3 (20) 4 (27) 0 (0) 0.02
0 (0) 4 (19) 9 (43) 8 (38) 0 (0)
0 (0) 5 (38) 0 (0) 7 (54) 1 (8)
0 (0) 6 (50) 4 (33) 2 (17) 0 (0)
0 (0) 1 (7) 8 (53) 4(27) 2 (13) 0.03
4 (19) 3 (14) 9 (14) 4 (43) 1 (5)
0 (0) 2 (15) 2 (15) 9 (69) 0 (0)
0 (0) 2 (17) 4 (33) 4 (33) 2 (17)
0 (0) 3 (20) 9 (60) 2 (13) 1 (7) 0.03
2 (10) 19 (90)
4 (31) 9 (69)
0 (0) 12 (100)
0 (0) 15 (100)
strongly disagreed and 39 % were “neutral” with respect to this idea. Secondly, academic physicians may not be aware of the concerns patients have about clinical trials. Similar to a prior study [23], several known barriers to trial enrollment, including protocol-related factors such as randomization, side effects, and additional required tests, were not identified by participants as potential concerns of perspective trial enrollees. Since physician perceptions and attitudes can affect their decision to discuss a clinical trial [13], a lack of understanding of known barriers to trial enrollment represents a possible physician-related barrier to enrollment. Academic physician knowledge of trials that were currently enrolling was disparate in our study, with 15.6 % of respondents disagreeing or strongly disagreeing that they had adequate knowledge of trials currently open at their institution, while 51.6 % agreed or strongly agreed that their knowledge
78
J Canc Educ (2014) 29:74–79
Fig. 1 Association of survey item “I have adequate knowledge of clinical trials that are currently enrolling participants within my specialty” (with “1” being strongly disagree and “5” being strongly agree) and estimated patient enrollment. Solid dots represent average responses, white dots represent individual survey responses, and horizontal, bolded lines represent the
median. The lines forming the upper and lower borders of the boxes delineate the first and third quartiles, respectively. Horizontal lines above and below the box represent the respective minimum and maximum values, excluding outliers
was adequate. Increasing physician-reported knowledge was significantly associated with estimated trial enrollment on univariate analysis. This is a new finding, and represents a potentially correctable barrier to clinical trial enrollment. A study by Davis et al. has shown that an educational program administered to practicing physicians can immediately improve comfort and confidence with trial introductions, though physician knowledge and long-term enrollment outcomes were not addressed in the study [24]. Faculty development initiatives aimed at addressing trial knowledge have the potential to improve enrollment of patients onto study. Our group is currently conducting a prospective, randomized trial to determine the most effective method to teach trainees how to introduce clinical trials. This study therefore has the potential to impact the way in which physician education is offered in order to optimize trial enrollment. Physician responses as to whether complicated trial protocols deterred them from attempting to enroll patients was also significantly associated with estimated enrollment, suggesting physicians may be less willing to introduce a clinical trial to patients when it is complicated and potentially timeconsuming. Time allotted to patient encounters is an oftcited barrier to physician enrollment onto trials [10]. In fact, nearly a third of physicians in our study disagreed or strongly disagreed with the statement that they had adequate time in the clinic to enroll patients onto trials. Communication skills are thought to be a means of mitigating this barrier, and previous work has shown that superior communication skills can actually save time in brief clinical settings [21]. Approximately 79 % of academic physicians in our study stated that they had excellent communication skills, which may explain why there
were not more physicians who felt limited by the amount of time they spent with patients. Of note, it is interesting that numerous physicians had not received explicit communication training, suggesting that there remains room for improvement in this area. This study is limited in that it was a self-assessment of physician knowledge and abilities, an area in which physicians are known to be poor [24, 25]. To our knowledge, no studies have investigated the accuracy of physician estimates of trial enrollment. Estimates of patient enrollment onto clinical trials reported in our study may therefore be subject to inaccurate estimation; however, it is reassuring that many physicians were likely honest in admitting very low patient enrollment given that they practice in fields in which high value is traditionally placed on clinical trials and enrollment. Despite these limitations, this study demonstrated barriers faced by academic physicians to enroll patients onto clinical trials including a lack of knowledge of clinical trial enrollment difficulties and patient concerns. Importantly, physician knowledge is a potentially modifiable factor that was significantly associated with decreased patient enrollment. Prospective studies are needed to objectively explore the relationship between physician knowledge and clinical trial enrollment and to determine the efficacy of initiatives designed to improve physician knowledge, communication skills, and awareness of barriers to trial enrollment. Acknowledgments We would like to thank Dr. David Johnson for his help with the initial abstract of this project and the Vanderbilt University School of Medicine emphasis program for providing funding for this research.
J Canc Educ (2014) 29:74–79
References 1. Christian MC, Trimble EL (2003) Increasing participation of physicians and patients from underrepresented racial and ethnic groups in national cancer institute-sponsored clinical trials. Cancer Epidemiol Biomarkers Prev 12:277–283 2. Ford JG, Howerton MW, Lai GY et al (2008) Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review. Cancer 112:228–242 3. Sateren WB, Trimble EL, Abrams J et al (2002) How sociodemographics, presence of oncology specialists, and hospital cancer programs affect accrual to cancer treatment trials. J Clin Oncol 20:2109–2117 4. Ehrlich PF, Newman KD, Haase GM et al (2002) Lessons learned from a failed multi-institutional randomized controlled study. J Pediatr Surg 37:431–436 5. Getz K (2004) Benchmarking patient recruitment and retention in clinical trials. In: Anderson D (ed) A guide to patient recruitment and retention. CenterWatch, Boston, pp 25–28 6. Mills EJ, Seely D, Raclis B et al (2006) Barriers to participation in clinical trials of cancer: a meta-analysis and systematic review of patient-reported factors. Lancet Oncol 7:141–148 7. Daugerty C (1999) Impact of therapeutic research on informed consent and the ethics of clinical trials: a medical oncology perspective. J Clin Oncol 17:1601–1617 8. Bennet CL, Adams JR, Knox KS et al (2001) Clinical trials: are they a good buy? J Clin Oncol 19:4330–4339 9. Lara PN, Paterniti DA, Chiechi C et al (2005) Evaluation of factors affecting awareness of and willingness to participate in cancer clinical trials. J Clin Oncol 23:9282–9289 10. Unger JM et al (2006) Impact of the year 2000 Medicare policy change on older patient enrollment to cancer clinical trials. J Clin Oncol 24:1–4 11. Comis RL, Aldig CR, Stovall EL, et al. (2000) Quantitative survey of public attitudes towards cancer clinical trials 12. Meropol NJ, Buzaglo JS, Millard J et al (2007) Barriers to clinical trial participation as perceived by oncologists and patients. J Natl Compr Canc Netw 5:753–762
79 13. Howerton MW, Gibbons MC, Baffi CR et al (2007) Provider roles in the recruitment of underrepresented populations to cancer clinical trials. Cancer 109:465–476 14. Albrecht TL, Blanchard C, Ruckdeschel JC et al (1999) Strategic physician communication and oncology clinical trials. J Clin Oncol 17:3324–3332 15. Fetting JH, Siminoff LA, Piantadosi S et al (1990) Effect of patients’ expectations of benefit on participation in a randomized clinical trial: a clinical vignette study. J Clin Oncol 8:1476–1482 16. Albrecht TL, Ruckdeschel JC, Riddle DL et al (2003) Communication and consumer decision making about cancer clinical trials. Patient Educ Couns 50:39–42 17. Lundine K, Buckley R, Hutchinson C et al (2008) Communication skills training in orthopaedics. J Bone Joint Surg 90:1393–1400 18. ACGME Outcome Project (2009) Competencies: new program requirements http://www.acgme-nas.org/ Accessed 8 Aug 2009 19. Harris PA, Taylor R, Thielke R et al (2009) Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 42(2):377–381 20. R Development Core Team (2009) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org. Accessed 8 Aug 2013 21. Getz KA, Wenger J, Campo RA et al (2008) Assessing the impact of protocol changes on clinical trial performance. Am J Ther 15:450– 457 22. Horn L, Keedy VL, Campbell N et al (2013) Identifying barriers associated with enrollment of patients with lung cancer into clinical trials. Clin Lung Cancer 14:14–18 23. Ross S, Grant A, Counsell C et al (1999) Barriers to participation in randomised controlled trials: a systematic review. J Clin Epidemiol 52:1143–1156 24. Davis DA, Mazmanian PE, Fordis M (2006) Accuracy of physician self-assessment compared with observed measures of competence: a systematic review. JAMA 296:1094–1102 25. Eva KW, Regehr G (2008) “I’ll never play professional football” and other fallacies of self-assessment. J Contin Educ Health Prof 28:9–14
Patient Enrollment onto Clinical Trials: the Role of Physician Knowledge Justin R. Gregg & Leora Horn & Mario A. Davidson & Jill Gilbert
Published online: 28 October 2013 # Springer Science+Business Media New York 2013
Abstract Sixty-six attending physicians at academic medical centers completed a 43-question self-assessment evaluating communication skills, comfort with clinical trial enrollment, and knowledge of patient-related barriers to enrollment on clinical trials. Responses and demographic information were analyzed for trends and for association with estimated trial enrollment. Physician-described enrollment of patients onto trials varied widely, with estimated enrollment varying from less than 5 patients to well over 125 enrolled during the previous year. Participants perceived themselves to have excellent communication skills and were comfortable with the trial enrollment process, though did not consistently identify patient-related barriers to enrollment. Physician knowledge of clinical trials currently enrolling within their field was associated with increased patient enrollment on study (p =0.03). Academic physicians expressed confidence in their skills related to clinical trial enrollment despite less than ideal reported enrollment. Knowledge of clinical trials currently enrolling within a physician’s specialty was associated with estimated patient enrollment, and may represent a correctable barrier to trial enrollment. Electronic supplementary material The online version of this article (doi:10.1007/s13187-013-0548-z) contains supplementary material, which is available to authorized users. J. R. Gregg (*) Department of Urologic Surgery, Vanderbilt University Medical Center, A – 1302 Medical Center North, Nashville TN 37232-2765, USA e-mail: [email protected] L. Horn : J. Gilbert Department of Medicine, Hematology and Oncology Division, Vanderbilt University Medical Center, Nashville, TN, USA M. A. Davidson Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA
Keywords Clinical Trials . Barriers . Self-Assessment . Knowledge . Enrollment
Introduction Enrollment of patients onto clinical trials is essential to advance the field of medicine; however, many barriers exist related to entry of adult patients onto clinical trials and successful completion of studies [1, 2]. Low patient enrollment into clinical trials leads to delays in trial completion and treatment advances as well as premature trial termination and wasted resources [3, 4]. Getz and colleagues estimated that over 90 % of trials are delayed due to problems with enrollment [5]. Known barriers to clinical trial enrollment include trial availability for individual patients or disease processes, patient-related factors, and physician-related factors [2, 6]. While trial availability relates to a lack of opportunity at a systemic level, patient-related factors are broad and include concerns about trial costs, transportation issues, and fears related to receiving a placebo or suffering from unknown side effects [7, 8]. Notably, underrepresented minorities have particularly low enrollment onto clinical trials and studies have shown that these patients may be particularly affected by a distrust of research and fear of harms related to trials, though not all of these fears have been consistently shown to result in quantitative decreases in enrollment [2]. Additionally, many fears faced by patients, particularly those involving cost, mistreatment, and the likelihood of receiving a placebo, are based on misconceptions [7, 9–11]. Physicians are in a unique position to allay patient fears, though it has also been shown that physicians do not consistently recognize patient-related barriers and concerns with regard to clinical trial enrollment [9, 12]. For this reason, provider perceptions and attitudes toward trials may affect whether a patient is offered a trial [13]. Physician surveys
J Canc Educ (2014) 29:74–79
have also indicated that limited time with a patient, a lack of dedicated staff, and other clinic resources may detract from their ability to enroll patients into trials [10]. When physicians do decide to offer a patient a trial, communication skills and clinical trial knowledge become important factors in presenting a trial to a patient. Good communication skills are a pre-requisite for the successful exchange of information between physicians and patients [14]. Communication is thought to provide the “means” [15] by which barriers associated with patient, physician, and research protocol factors impact an individual patient’s decision making process. It is known that physicians differ in the methods through which they explain clinical trials and treatment options [15]. These differences can impact clinical trial enrollment, as patients’ intentions to enroll into a clinical trial can be affected by whether the provider presents the trial in a positive, negative, or ambiguous manner [16]. Communication skills, while traditionally believed to be inherent to individuals, are increasingly viewed as skills that are best taught through established techniques such as experiential learning and directed feedback [17]. In fact, communication skills have recently been outlined as a core competency of residency training as outlined by the Accreditation Council for Graduate Medical Education [18]. Despite this emphasis, it has been shown that explicit communication skills training were rare in a group of surgical subspecialty residents [17]. To our knowledge, few studies have examined physicianrelated barriers to trial enrollment in academic physicians, including provider knowledge, perceptions, and communication skills. Our aim was to explore academic physician-related barriers to clinical trial enrollment. We also aim to determine factors associated with estimated patient enrollment in clinical trials.
Methods Survey Development We designed a questionnaire utilizing questions related to patient enrollment into clinical trials using two previously designed instruments [11, 17]. The survey included demographic data (eight questions) and questions related to enrollment into clinical trials that were broken down into four categories: Communication Skills (nine items), Clinical Trials—Personal Focus (nine items), Clinical Trials— General Focus (nine items), and Clinical Trials—Enrollment (eight items). Item generation was carried out by three of the authors (JRG, LH, and JG). All questions were rated on a 1–5point Likert scale, ranging from “strongly disagree” to “strongly agree.” A pilot test of the questionnaire was done with a group of five clinical investigators from Vanderbilt’s
75
Division of Hematology/Oncology to evaluate for item redundancy, readability, and content. The full questionnaire minus demographic questions can be reviewed in Appendix 1. Survey Administration The questionnaire was distributed between August and December 2009 to a sample of attending physicians in three different departments with a history of strong involvement in clinical trials in academic medical institutions. Hematology/ Oncology departments that were asked to participate were Vanderbilt University, Memorial-Sloan Kettering, and Harvard-affiliated hospitals. Vanderbilt University’s Division of Cardiology and Division of Radiation Oncology were also asked to participate. Study data were collected and managed using Research Electronic Data Capture (REDCap) hosted at Vanderbilt University [19]. Surveys were distributed both electronically and through the use of paper copies that were later entered into REDCap by a single author (JRG). All study participants signed an informed consent form approved by the Vanderbilt University Institution Review board. Statistical Analysis Data were analyzed using descriptive statistics such as means, median, percentiles, and frequencies on N =66 physicians. Not all physicians completed the questionnaire in its entirety; therefore, the above analysis was completed based on the number of physicians who answered each individual questionnaire item. Chi-squared, Kruskal–Wallis, and Wilcoxon analyses were done to determine associations between physician-estimated patient enrollment and demographic information and survey responses. Patient enrollment was chosen for analysis in order to determine if any physician characteristic or response was related to a lower quantitative estimate of patient enrollment. All statistical tests were twotailed at an alpha level of 0.05. All analysis was done using R version 2.10.1 [20].
Results Overall Study Participant Characteristics Sixty six physicians out of 115 potential participants completed the self-assessment questionnaire (57 % response rate). Physicians from each eligible department completed questionnaires. Demographic data collected in the study are summarized in Table 1. The average age of participants was 47.7 years, 68 % of them were male and 71 % were Caucasian. The majority (81 %) were US medical school graduates. In total, there were 36 hematologists/oncologists
76
J Canc Educ (2014) 29:74–79
Table 1 Demographics of academic physician survey responders
Clinical Trials—Personal Focus
a
Characteristic
Question response
Age, mean (SD) Sex, n (%) Male Female Ethnicity, n (%) White African American Asian/Pacific Islander Nationality, n (%) US medical graduate International graduate Physician degree, n (%) MD/DO Ph.D. MPH Clinical research involvement, n (%) Any previous involvement
47.7 (9.94) 44 (68 %) 21 (32 %) 46 (71 %) 5 (8 %) 14 (22 %) 53 (82 %) 12 (18 %) 62 (100 %) 6 (10 %) 8 (13 %) 57 (93 %)
No prior involvement 4 (7 %) Time dedicated to clinical activities, n (%) Less than 20 % 10 (16 %) Between 20 and 50 % 10 (16 %) Approximately 50 % 18 (29 %) Greater than 50 % 24 (39 %) Estimated clinical trial patient enrollment in past 12 months, n (%) 0–5 21 (34 %) 6–10 13 (21 %) 11–25 12 (20 %) >25 15 (25 %) a
The total number in each category does not equal 66 because complete data were not available from all survey responders
(55 % of study participants), 24 cardiologists (36 %), and 6 radiation oncologists (9 %). Thirty-four percent of the population reported that they enrolled 0–5 patients during the previous 12 months, while 25 % enrolled over 25 patients. Communication Skills Physicians are confident in their communication skills, as 51/ 64 (79.7 %) respondents either agreed or strongly agreed that they had excellent communication skills. Respondents also believed that good communication can improve patient outcomes (59/66 [89.3 %]) and enrollment into clinical trials (60/ 66 [90.9 %]). Physicians differed in their method of learning communication skills during training, with 26/66 (39.4 %) disagreeing and 27/66 (40.9 %) agreeing that they received explicit instruction during residency and fellowship training on how to improve communication skills (Appendix 1).
Participants are also confident and comfortable with clinical trials and the enrollment process. Sixty two of 65 (95.4 %) physicians agreed or strongly agreed that they have an understanding of the informed consent process and are comfortable discussing clinical trial results and consenting patients. When presented with the statement “I have adequate knowledge of clinical trials that are currently enrolling within my specialty,” 10/64 (15.6 %) disagreed or strongly disagreed while 43/63 (67.2 %) agreed or strongly agreed (Appendix 1). Clinical Trials—General Focus Participants agree on the importance of clinical trials, with 61/ 66 (92.4 %) agreeing that trials are the essential means of discovering new therapies. Physicians disagreed, though, on the reasons patients do not enroll in clinical trials, with 12/66 (18.1 %) disagreeing that patients fear potential side effects and 33/66 (50 %) disagreeing that patients fear financial liability, among others (Appendix 1). Clinical Trials—Enrollment Physicians admit to varying levels of bias in terms of which patients they consider for enrollment onto clinical trials, with 46/64 (71.9 %) agreeing or strongly agreeing that they are more likely to attempt to enroll patients in clinical trials if they suffer from an incurable illness (Appendix 1). Statistical Analysis Physician-estimated patient enrollment was then evaluated for association with demographic information and questionnaire item responses (Table 2). Estimated knowledge of currently enrolling clinical trials was significantly associated with increased estimated patient enrollment onto clinical trials during the past year (p =0.03) (Fig. 1). The items “Complicated clinical trial protocols deter me from attempting to enroll patients in clinical trials,” “Patients don’t participate in clinical trials because they fear going through too many tests,” and “I have sufficient time in clinic to explain clinical trials to patients” and whether the respondent held a Ph.D. were significantly associated with estimated patient enrollment (p =0.04, 0.02, 0.03, and 0.03, respectively). All other associations between physician-estimated patient enrollment and questionnaire items did not reach statistical significance.
Discussion The enrollment of patients onto clinical trials is often inadequate, leading to delayed trial completion [21] and thus delayed
J Canc Educ (2014) 29:74–79 Table 2 Physician characteristics associated with Estimated Patient Enrollment, n (%)
a
Pearson Test
77
Estimated Patient Enrollment
p Value
I have adequate knowledge of clinical trials that are currently enrolling participants within my specialty, n (%) Strongly disagree
0–5
5–10
3 (15)
0 (0)
1 (8)
0 (0)
Disagree Neutral Agree Strongly agree Complicated clinical trial protocols deter me from attempting to enroll patients in clinical trials, n (%) Strongly disagree Disagree Neutral Agree Strongly agree Patients don’t participate in clinical trials because they fear going through too many tests, n (%) Strongly disagree Disagree Neutral Agree Strongly agree I have sufficient time in clinic to explain clinical trials to patients, n (%) Strongly disagree Disagree Neutral Agree Strongly agree Degree, n (%) Ph.D. No Ph.D.
2 (10) 7 (35) 4 (20) 4 (20)
2 (15) 1 (8) 5 (38) 5 (38)
0 (0) 1 (8) 3 (25) 7 (58)
0 (0) 2 (13) 1 (7) 12 (80)
medical progress. Our group recently showed that 14 % of patients seen by thoracic medical oncologists at the Vanderbilt Ingram Cancer Center are entered onto a clinical trial; however, 50 % of eligible patients did not enroll on study [22]. Additionally, many physicians in our cohort conceded poor patient enrollment onto clinical trials, with a third of survey participants having enrolled 0–5 patients during the previous year—well below the NIH recommendations of 15 % of patient enrollment onto a clinical protocol [23]. Known barriers to clinical trial enrollment include patient-related and physician-related factors. This study aimed to evaluate academic physician-related barriers to enrollment of patients into clinical trials. The first finding evident in our study is that physicians may not recognize that trial enrollment is a nationwide problem. Those who participated in our study did not agree that patients were reluctant to join clinical trials, as 35 % disagreed or
10–25
>25
0.03a
0.04 2 (10) 5 (24) 7 (33) 7 (33) 0 (0)
1 (8) 3 (23) 4 (31) 3 (23) 2 (15)
7 (58) 3 (25) 1 (8) 0 (0) 1 (8)
3 (20) 5 (33) 3 (20) 4 (27) 0 (0) 0.02
0 (0) 4 (19) 9 (43) 8 (38) 0 (0)
0 (0) 5 (38) 0 (0) 7 (54) 1 (8)
0 (0) 6 (50) 4 (33) 2 (17) 0 (0)
0 (0) 1 (7) 8 (53) 4(27) 2 (13) 0.03
4 (19) 3 (14) 9 (14) 4 (43) 1 (5)
0 (0) 2 (15) 2 (15) 9 (69) 0 (0)
0 (0) 2 (17) 4 (33) 4 (33) 2 (17)
0 (0) 3 (20) 9 (60) 2 (13) 1 (7) 0.03
2 (10) 19 (90)
4 (31) 9 (69)
0 (0) 12 (100)
0 (0) 15 (100)
strongly disagreed and 39 % were “neutral” with respect to this idea. Secondly, academic physicians may not be aware of the concerns patients have about clinical trials. Similar to a prior study [23], several known barriers to trial enrollment, including protocol-related factors such as randomization, side effects, and additional required tests, were not identified by participants as potential concerns of perspective trial enrollees. Since physician perceptions and attitudes can affect their decision to discuss a clinical trial [13], a lack of understanding of known barriers to trial enrollment represents a possible physician-related barrier to enrollment. Academic physician knowledge of trials that were currently enrolling was disparate in our study, with 15.6 % of respondents disagreeing or strongly disagreeing that they had adequate knowledge of trials currently open at their institution, while 51.6 % agreed or strongly agreed that their knowledge
78
J Canc Educ (2014) 29:74–79
Fig. 1 Association of survey item “I have adequate knowledge of clinical trials that are currently enrolling participants within my specialty” (with “1” being strongly disagree and “5” being strongly agree) and estimated patient enrollment. Solid dots represent average responses, white dots represent individual survey responses, and horizontal, bolded lines represent the
median. The lines forming the upper and lower borders of the boxes delineate the first and third quartiles, respectively. Horizontal lines above and below the box represent the respective minimum and maximum values, excluding outliers
was adequate. Increasing physician-reported knowledge was significantly associated with estimated trial enrollment on univariate analysis. This is a new finding, and represents a potentially correctable barrier to clinical trial enrollment. A study by Davis et al. has shown that an educational program administered to practicing physicians can immediately improve comfort and confidence with trial introductions, though physician knowledge and long-term enrollment outcomes were not addressed in the study [24]. Faculty development initiatives aimed at addressing trial knowledge have the potential to improve enrollment of patients onto study. Our group is currently conducting a prospective, randomized trial to determine the most effective method to teach trainees how to introduce clinical trials. This study therefore has the potential to impact the way in which physician education is offered in order to optimize trial enrollment. Physician responses as to whether complicated trial protocols deterred them from attempting to enroll patients was also significantly associated with estimated enrollment, suggesting physicians may be less willing to introduce a clinical trial to patients when it is complicated and potentially timeconsuming. Time allotted to patient encounters is an oftcited barrier to physician enrollment onto trials [10]. In fact, nearly a third of physicians in our study disagreed or strongly disagreed with the statement that they had adequate time in the clinic to enroll patients onto trials. Communication skills are thought to be a means of mitigating this barrier, and previous work has shown that superior communication skills can actually save time in brief clinical settings [21]. Approximately 79 % of academic physicians in our study stated that they had excellent communication skills, which may explain why there
were not more physicians who felt limited by the amount of time they spent with patients. Of note, it is interesting that numerous physicians had not received explicit communication training, suggesting that there remains room for improvement in this area. This study is limited in that it was a self-assessment of physician knowledge and abilities, an area in which physicians are known to be poor [24, 25]. To our knowledge, no studies have investigated the accuracy of physician estimates of trial enrollment. Estimates of patient enrollment onto clinical trials reported in our study may therefore be subject to inaccurate estimation; however, it is reassuring that many physicians were likely honest in admitting very low patient enrollment given that they practice in fields in which high value is traditionally placed on clinical trials and enrollment. Despite these limitations, this study demonstrated barriers faced by academic physicians to enroll patients onto clinical trials including a lack of knowledge of clinical trial enrollment difficulties and patient concerns. Importantly, physician knowledge is a potentially modifiable factor that was significantly associated with decreased patient enrollment. Prospective studies are needed to objectively explore the relationship between physician knowledge and clinical trial enrollment and to determine the efficacy of initiatives designed to improve physician knowledge, communication skills, and awareness of barriers to trial enrollment. Acknowledgments We would like to thank Dr. David Johnson for his help with the initial abstract of this project and the Vanderbilt University School of Medicine emphasis program for providing funding for this research.
J Canc Educ (2014) 29:74–79
References 1. Christian MC, Trimble EL (2003) Increasing participation of physicians and patients from underrepresented racial and ethnic groups in national cancer institute-sponsored clinical trials. Cancer Epidemiol Biomarkers Prev 12:277–283 2. Ford JG, Howerton MW, Lai GY et al (2008) Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review. Cancer 112:228–242 3. Sateren WB, Trimble EL, Abrams J et al (2002) How sociodemographics, presence of oncology specialists, and hospital cancer programs affect accrual to cancer treatment trials. J Clin Oncol 20:2109–2117 4. Ehrlich PF, Newman KD, Haase GM et al (2002) Lessons learned from a failed multi-institutional randomized controlled study. J Pediatr Surg 37:431–436 5. Getz K (2004) Benchmarking patient recruitment and retention in clinical trials. In: Anderson D (ed) A guide to patient recruitment and retention. CenterWatch, Boston, pp 25–28 6. Mills EJ, Seely D, Raclis B et al (2006) Barriers to participation in clinical trials of cancer: a meta-analysis and systematic review of patient-reported factors. Lancet Oncol 7:141–148 7. Daugerty C (1999) Impact of therapeutic research on informed consent and the ethics of clinical trials: a medical oncology perspective. J Clin Oncol 17:1601–1617 8. Bennet CL, Adams JR, Knox KS et al (2001) Clinical trials: are they a good buy? J Clin Oncol 19:4330–4339 9. Lara PN, Paterniti DA, Chiechi C et al (2005) Evaluation of factors affecting awareness of and willingness to participate in cancer clinical trials. J Clin Oncol 23:9282–9289 10. Unger JM et al (2006) Impact of the year 2000 Medicare policy change on older patient enrollment to cancer clinical trials. J Clin Oncol 24:1–4 11. Comis RL, Aldig CR, Stovall EL, et al. (2000) Quantitative survey of public attitudes towards cancer clinical trials 12. Meropol NJ, Buzaglo JS, Millard J et al (2007) Barriers to clinical trial participation as perceived by oncologists and patients. J Natl Compr Canc Netw 5:753–762
79 13. Howerton MW, Gibbons MC, Baffi CR et al (2007) Provider roles in the recruitment of underrepresented populations to cancer clinical trials. Cancer 109:465–476 14. Albrecht TL, Blanchard C, Ruckdeschel JC et al (1999) Strategic physician communication and oncology clinical trials. J Clin Oncol 17:3324–3332 15. Fetting JH, Siminoff LA, Piantadosi S et al (1990) Effect of patients’ expectations of benefit on participation in a randomized clinical trial: a clinical vignette study. J Clin Oncol 8:1476–1482 16. Albrecht TL, Ruckdeschel JC, Riddle DL et al (2003) Communication and consumer decision making about cancer clinical trials. Patient Educ Couns 50:39–42 17. Lundine K, Buckley R, Hutchinson C et al (2008) Communication skills training in orthopaedics. J Bone Joint Surg 90:1393–1400 18. ACGME Outcome Project (2009) Competencies: new program requirements http://www.acgme-nas.org/ Accessed 8 Aug 2009 19. Harris PA, Taylor R, Thielke R et al (2009) Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 42(2):377–381 20. R Development Core Team (2009) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org. Accessed 8 Aug 2013 21. Getz KA, Wenger J, Campo RA et al (2008) Assessing the impact of protocol changes on clinical trial performance. Am J Ther 15:450– 457 22. Horn L, Keedy VL, Campbell N et al (2013) Identifying barriers associated with enrollment of patients with lung cancer into clinical trials. Clin Lung Cancer 14:14–18 23. Ross S, Grant A, Counsell C et al (1999) Barriers to participation in randomised controlled trials: a systematic review. J Clin Epidemiol 52:1143–1156 24. Davis DA, Mazmanian PE, Fordis M (2006) Accuracy of physician self-assessment compared with observed measures of competence: a systematic review. JAMA 296:1094–1102 25. Eva KW, Regehr G (2008) “I’ll never play professional football” and other fallacies of self-assessment. J Contin Educ Health Prof 28:9–14
