Gynecologic Oncology 138 (2015) 445–448
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
A novel clinical trial recruitment strategy for women's cancer B.J. Rimel ⁎, Jenny Lester, Leah Sabacan, Diane Park, Catherine Bresee, Catherine Dang, Beth Karlan Cedars-Sinai Medical Center, 8700 Beverly Blvd, Los Angeles, CA 90048, United States
H I G H L I G H T S • An online patient-facing registry improved accrual over a paper based registry. • Online registry participants matched to studies faster than paper based participants • Fifteen percent of women who participated in the registry enrolled in clinical trials
a r t i c l e
i n f o
Article history: Received 25 January 2015 Accepted 13 May 2015 Available online 20 May 2015 Keywords: Clinical trial Registry Enrollment
a b s t r a c t Objective. To address a deficiency in clinical trial and research enrollment in gynecologic cancer studies, we launched a paper based patient research registry. To improve registry enrollment, we transitioned to an online registry and trial matching mechanism to aid women in accessing open studies. Methods. Utilizing a validated verification platform, we designed a web-based registry and trial matching mechanism for women over age 18. Participants completed a questionnaire to provide information for trial matching. A focus group of registry participants was held 9 months after the start of the study to evaluate barriers to participation. Results. A total of 322 women were enrolled in the online registry over a 14 month period which was a 4.3 fold increase over the paper-based registry (p b 0.0001). Two hundred and sixty three (82%) women were matched to at least one study. Fifteen percent (39/263) of those eligible for studies went on to enroll. The online enrollment rate to studies was not different from that observed in the paper-based registry (26/172, p = 0.934), however, the web-based registry linked participants to subsequent studies 27% more rapidly (68 (+/− 98) days vs. 93 (+/−81) days for the paper-based registry, p = 0.017). Focus group participants identified areas for improvement. Conclusion. Web-based patient driven registry provides dramatic improvement in the number of participants enrolled and the time to trial linkage compared to a paper based registry at a single institution. Further studies of barriers to research participation are necessary to improve on this model. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Less than 2% of patients with cancer participate in a clinical trial in the United States [1]. Gynecologic oncology patients do not appear to participate in trials with any more frequently than other cancer types. While gains in progression free survival have continued to improve overall life span for women with advanced gynecologic cancers, more cures have not been realized. Lack of accrual to trials leads to early closure of studies and a waste of critical resources as well as extended periods of enrollment which can hinder the ability to interpret the results. Stensland et al. reported that 1 in 4 cancer clinical trials were ⁎ Corresponding author at: Cedars-Sinai Medical Center, 8700 Beverly Blvd Suite 290W, Los Angeles, CA 90048, United States. Tel.: +1 310 423 5800 (office); fax: +1 310 967 1142. E-mail address: [email protected] (B.J. Rimel).
http://dx.doi.org/10.1016/j.ygyno.2015.05.008 0090-8258/© 2015 Elsevier Inc. All rights reserved.
stopped early with 1 in 10 being stopped for poor accrual [2]. Data are somewhat limited, but a panel of experts convened by the NCI and ASCO to discuss barriers to clinical trial enrollment in 2013 [3] cited barriers in three areas as most significant: 1) patient/community – 2) physician/provider level – 3) site/organizational. Physician/provider level barriers include willingness to refer a patient for study, lack of knowledge about available clinical trials and concern regarding a patient's ability to participate [3–5]. Patient/community barriers have been noted to include being unaware of trial opportunities and complexity and stringency of the protocol [6] Both of these barriers could be addressed by providing a research registry to inform and match patients for study. Clinical trials, defined by the National Institute of Health as “a research study in which one or more human subjects are prospectively assigned to one or more interventions to evaluate the effects of those interventions on health related biomedical or behavioral outcomes”
446
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448
[7], are the primary focus for enrollment of most patients with an active malignancy. Clinical research studies such as tissue banks and longitudinal cohort studies provide invaluable data for researchers but require a consenting population and may be overlooked when the focus is on therapeutic intent. Non-interventional research studies related to screening, surveillance and survivorship offer opportunities for women without cancer to engage in research that may result in improved detection or risk assessment. However, these women are harder to find as they are not commonly present in the cancer centers where clinical trial recruitment and research participation are more the norm. Recent data from the Pew Internet Research Project suggests that 86% of women use the internet and 72% of those users are accessing the internet to find medical information [8]. Similarly, 76% of women who use the internet participate in social networking sites, making the online distribution of clinical trial and research study information and recruitment possible in this population [9]. Based on the barriers to enrollment and the common utilization of digital technology in our target population, we hypothesized that a patient-facing online registry would provide greater registry participation than a previous paper based consent and registry. We also hypothesized that a computer trial matching process would improve enrollment of registry participants to clinical trials and research studies. The goal of this study was to develop an online patient driven registry utilizing a valid online consent process to connect women interested in participating in clinical research. 2. Methods In 2010 a paper based research registry was designed to obtain epidemiologic information from female patients at a single institution. This study was approved by the institutional review board. Paper consent forms and a short demographic and medical history questionnaire were provided to female patients checking in at the imaging, breast and cancer centers. Various marketing methods were used to improve registry accrual, including posters placed in the centers, informational cards, and card-drop boxes for patients interested in being contacted in several locations and closed circuit TV ads in the hospital. Recruitment to the registry was low with an average of 5 participants enrolled per month. The paper questionnaires were then manually entered into the research database by study staff, and clinical trial matching was performed on the research database. After noting continued low accrual despite internal marketing methods, we designed a novel online consent process utilizing a validation method that meets the federal guidelines for digital consent process (21CFR). This study was approved by the institutional review board and was initiated on March 17, 2013. Using a web-based informed consent process with a validation screen, women could chose to participate through an electronic signature that verifies that they have read the consent and are attesting that they wish to participate. The vendor used to create the email signature validation was DocuSign. Participants were given clear instructions that data security will be upheld to the highest standard, but like all other online data, could be subject to unavoidable data breach. Upon completion of the online consent process, participants were asked to complete a questionnaire with demographic, geographic, cancer history and medical history questions. The data points chosen for the questionnaire represent major inclusion and exclusion criteria for a large number of clinical trials and research studies at our institution. Once the questionnaire is completed a query of available studies that match the participant's criteria is run. Patients identify how they wish to be notified of potential study matches (by email, US postal mail, or phone call) in the questionnaire. Once a match is made, the information regarding the nature of the study and the study coordinator contact information is delivered to the participant in her chosen format. The list of available studies and the correct contact information for the study coordinator is carefully updated to ensure that the appropriate information is given to participants.
To inform potential participants of the online registry, we continued to have posters in the centers and informational cards. We also distributed the website to local ovarian cancer support communities and utilized online media outlets, and social media (Facebook posts by cancer community organizations/twitter posts) to promote the website. Paper consents and questionnaires were still available in the clinics for women who chose not to enroll online. Differences in observed frequencies between the web-based and traditional paper-based registry were tested using the Chi-square test or Fisher's exact test (in the case of sparse data). Continuous data are presented as means +/− standard deviation and tested by way of a Student's t-test. All data were analyzed with SAS v9.2 statistical software. For all statistical tests the level of significance was set at p b 0.05. 3. Results A total of 507 women enrolled in the research registry from March 2010 to May 2014. Conversion to the online registry format significantly increased registry participation compared with the paper based system. The average monthly enrollment to the registry increased from 5.4 women per month with the paper based registry to 23.0 women per month with the web-based registry (p b 0.0001). In the first 14 months of the online registry a total of 322 women participated. We also noted demographic changes in the registry participants with the conversion to the online format. Online participants tended to be younger, were less likely to have been tested for BRCA mutation, and less likely to have a cancer. In addition, online participants were significantly more diverse, with more non-white participants (25% vs. 15%, p b 0.001). Demographic changes in the registry are detailed in Table 1. In the paper based registry, 172 women were eligible for at least one study (93%) while in the online cohort, only 263 (82%) were eligible for at least one study, (p b 0.001). Despite the decrease in overall eligibility there was no difference in the rate women went on to enroll in another study with 15% in both the paper-based system (26/172) and the online system (39/263; p = 0.935). In fact, there were more total enrollments in the online cohort, 39 vs. 26, but the improvement did not reach statistical significance (p = 0.528). The studies that patients participated in were a hereditary cancer cohort study, an ovarian cancer screening study, an imaging study and a tissue bank study, a disease specific biobank study, and a BRCA survey and translational study. No patients
Table 1 Comparison of demographics, genetic testing and cancer types between the paper based registry and the web based registry.
Age, y (mean +/− SD) Race White Asian Black Latino Other/Declined to state Religion Jewish Other/Declined to state BRCA tested Yes No Unknown Cancer history Breast Gynecological Other types None
All cases
Paper based
Web based
(n = 507)
(n = 185)
(n = 322)
P-value
50.4 +/− 14.0
55.8 +/− 13.1
47.2 +/− 13.6
400 (79%) 39 (8%) 19 (4%) 17 (3%) 32 (6%)
158 (85%) 11 (6%) 9(5%) 1 (1%) 6 (3%)
242 (75%) 28 (9%) 10 (3%) 16 (5%) 26 (8%)
0.0049
112 (22%) 244 (48%)
46 (25%) 139 (75%)
66 (24%) 256 (80%)
0.2538
143 (28%) 332 (65%) 32 (6%)
84 (45%) 87 (47%) 14 (8%)
59 (18%) 245 (76%) 18 (6%)
b0.0001
113 (22%) 85 (17%) 66 (13%) 289 (57%)
81 (44%) 17 (9%) 34 (18%) 69 (37%)
32 (10%) 68 (21%) 32 (10%) 220 (68%)
b0.0001 0.0005 0.0065 b0.0001
b0.0001
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448 Table 2 Comparison of eligibility and enrollment between paper based and web based registry participants.
Eligible for studies Hereditary breast and ovarian cancer cohort study Ovarian cancer screening study Tissue banking study Imaging study Irritable bowel disease biobank study BRCA survey and translational study Therapeutic clinical trials Eligibility count 0 1 2 3 4 5 Mean eligibility Enrolled in a study Among eligible cases
447
2500 2191
Paper based registry
Web based registry
P value
2000
91 (49%)
76 (24%)
b0.0001
1500
103 (56%) 12 (6%) 81 (44%) 61 (33%) 41 (22%) 0 (0%)
101 (31%) 13 (4%) 15 (5%) 169 (52%) 99 (31%) 0 (0%)
b0.0001 0.2201 b0.0001 b0.0001 0.0374
Unique Website Hits Count Envelopes Created Count Envelopes Viewed Count Envelopes Signed
1000 683 527 500
13 (7%) 48 (26%) 59 (32%) 39 (21%) 24 (13%) 2 (1%) 2.1 +/− 1.2 26 (14%) 26/172 (15%)
59 (18%) 99 (31%) 127 (39%) 30 (9%) 5 (2%) 2 (1%) 1.5 +/− 1.0 39 (12%) 39/263 (15%)
b0.0001
0
b0.0001 0.5289 0.9345
were identified from either cohort as eligible for a cancer therapeutic study in the first year (Table 2). There were significant differences in the types of studies that participants enrolled in between the two groups (Table 3). This was especially apparent in the ovarian cancer screening study where 17 women participated from the online cohort, compared to no eligible women from the paper-based cohort (p = b 0.0001). In the traditional paper cohort, it took an average of 93 +/− 81 days for a participant to be linked to another study compared to the online cohort's average time to linkage of 68 +/− 98 days (p = 0.017). This can be attributed to several factors: In the online system, staff time is not required to perform data entry and manage the consent paperwork resulting in significantly less time to determine eligibility and those who consent online seem to be quick to respond emails while the paperbased cohort more often were notified through US mail. Website traffic defined as the number of individual views of the website from separate Internet Protocol (IP) addresses did not consistently correlate with enrollment. Both web-based and in person marketing interventions were performed during the first year of the online registry. Web-based marketing resulted in more website traffic, but did not necessarily result in more enrollments to the registry during this study period. Analysis of the consent process by measuring both discreet website views as well as the validation steps required for consent completion demonstrated attrition at all steps (Fig. 1). Attrition from the website view to the consent process (Click to join) is expected as not all viewers Table 3 Comparison of study enrollment between paper based registry participants and web based registry participants. Paper based registry Study enrollment Hereditary breast and ovarian cancer 24 (13%) cohort study Ovarian cancer screening study 0 (0%) Tissue banking study 0 (0%) MRI study 0 (0%) Irritable bowel disease biobank study 1 (1%) BRCA survey and translational study 2 (1%) Number of studies enrolled by individual participant 0 159 (86%) 1 25 (14%) 2 1 (1%)
Web based registry
P value
10 (3%)
b0.0001
17 (5%) 2 (1%) 2(1%) 5 (2%) 10 (3%)
b0.0001 0.2828 0.2828 0.4233 0.1489
283 (88%) 32 (10%) 7 (2%)
306
0.1851
Fig. 1. Analysis of attrition from website view to completion of web based consent process.
will be eligible (e.g. male, less than 18 years old, etc.). A focus group of participants who consented online was held in January 2014 to discuss the consent process and how to better retain participants throughout the process. Focus group data suggested that more information about the research and clinical goals of the registry would better enable visitors to the website to continue on through the consent. Other themes from the focus group were that participants were confused by the need to toggle between windows to complete the email validation step to sign the online consent. Focus group participants also suggested that a video of the consent process with a description of how long it takes to complete (approximately 5–10 min) would also be helpful. 4. Discussion Enrollment to clinical trials and research studies remains a significant problem for cancer research. Registries of potential participants have been used previously with good success [10,11]. Several trial matching programs are available online but they rely on data from www.clinicaltrials.gov, which is frequently out of date. The availability of accurate and timely clinical trial data from a single local source and the ability to quickly connect with interested participants should improve this situation. In our evaluation, using a web-based patient facing consent process has expanded enrollment onto non therapeutic clinical trials and research studies at a single institution. The web based approach to registry participation demonstrated improvement in the number of women who enrolled in a study, but the proportion of eligible registry participants was unchanged from a traditional paper-based approach. Only 15% of those women participating in the registry ultimately enrolled in a research study. There may be several reasons for this. First, women deemed eligible based on questionnaire data may not have met all study requirements after further investigation. Second, women eligible for studies may have chosen not to participate because they did not feel the study they were offered interested them. Third, it may be that women who were notified they were eligible were interested in participation but did not follow through due to other factors such as distance from the medical center or time constraints. Unfortunately we do not have data at this time on why subjects chose not to enroll in a study to which they were linked. It is important to note that the similarity in rates of enrollment between the two methods shows that the online registry is not inferior to the paper based approach. Lastly, the digital approach was begun during a time of great transition for the cooperative group structure and severe decline in the number of available therapeutic studies in gynecologic cancer both nationally and at our institution. It is unclear how the online trial matching model would perform if more studies were available.
448
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448
Decreased time from registry participation to enrollment on a study for the web based group was also a significant finding. This result seems most relevant to the timely completion of clinical trials and the time sensitive nature of both the trial eligibility and the length of funding. This also appears to be a significant programmatic improvement and the lack of manual data entry certainly improves staff time and cost. However, there was no difference in the proportion of registry participants who ultimately enrolled in research study between the two groups. This may be due to the significantly higher proportion of women eligible in the paper group, representing a population that was already in the medical center for some reason related to their health. There was a significant improvement in the accrual to an interventional ovarian cancer screening trial in the web based registry. We hypothesize that this is due to a younger population in the online cohort as there was no difference in eligibility between the two groups. The use of web metrics to evaluate the most effective recruitment methods for the web based approach and the use of focus groups to evaluate the areas of attrition are strengths of this study. Unfortunately, despite the presence of a short survey that is activated when a potential participant exits the questionnaire without completion, there were very few responders to the survey. Lack of data about why participants choose not to complete the questionnaire remains a weakness of the study. Comparison of number of registry enrollments between paper based and web based approaches showed that the web-based approach had significantly more registry enrollment. However, it should be noted that because of the nature of the website being able to be accessed by women outside of the medical center some or all of this increase could be due to the larger pool of potential participants from which to draw. Luck et al. examined the characteristics of 236 women diagnosed with gynecologic cancer between 1997 and 2001 regarding their participation in clinical trials [12]. In this cohort, 54 (23%) women participated in an NCI sponsored clinical trial during the time period of the study. The study concluded that the most important factor in women choosing to enter a clinical trial was their familiarity with the physician–investigator who describes the trial. These results suggest that a trusted source of information is essential to encouraging participation in a clinical study. In an era where 72% of internet users are looking for health information [8], that trusted source may be less of a person and more of a domain
name. The web based registry may provide that source for clinical trial knowledge for internet users. In conclusion, the web based registry improved overall participation and significantly decreased time to enrollment. Future improvement of this model will require innovative methods to capture complex data from patients such as molecular tumor profiling to further decrease the time to enrollment for studies and provide more comprehensive pre-screening for patients. These critical data areas should aid in locating and engaging patients for time sensitive cancer therapeutic trials. Conflict of interest The authors have no conflict of interest to disclose.
References [1] Mannel RS, Moore K. Research: an event or an environment? Gynecol Oncol 2014; 134:441–2. [2] Stensland KD, McBride RB, Latif A, Wisnivesky J, Hendricks R, Roper N, Boffetta P, Hall SJ, Oh WK, Galsky MD. Adult cancer clinical trials that fail to complete: an epidemic? J Natl Cancer Inst 2014;106. [3] Denicoff AM, McCaskill-Stevens W, Grubbs SS, Bruinooge SS, Comis RL, Devine P, Dilts DM, Duff ME, Ford JG, Joffe S, Schapira L, Weinfurt KP, Michaels M, Raghavan D, Richmond ES, Zon R, Albrecht TL, Bookman MA, Dowlati A, Enos RA, Fouad MN, Good M, Hicks WJ, Loehrer Sr PJ, Lyss AP, Wolff SN, Wujcik DM, Meropol NJ. The National Cancer Institute-American Society of Clinical Oncology Cancer Trial Accrual Symposium: summary and recommendations. J Oncol Pract 2013;9:267–76. [4] Lara Jr PN, Higdon R, Lim N, Kwan K, Tanaka M, Lau DH, Wun T, Welborn J, Meyers FJ, Christensen S, O'Donnell R, Richman C, Scudder SA, Tuscano J, Gandara DR, Lam KS. Prospective evaluation of cancer clinical trial accrual patterns: identifying potential barriers to enrollment. J Clin Oncol 2001;19:1728–33. [5] Comis RL, Miller JD, Colaizzi DD, Kimmel LG. Physician-related factors involved in patient decisions to enroll onto cancer clinical trials. J Oncol Pract 2009;5:50–6. [6] Mills EJ, Seely D, Rachlis B, Griffith L, Wu P, Wilson K, Ellis P, Wright JR. Barriers to participation in clinical trials of cancer: a meta-analysis and systematic review of patient-reported factors. Lancet Oncol 2006;7:141–8. [7] Health NIo. Notice of revised NIH definition of “Clinical Trial”. http://grants.nih.gov/ grants/guide/notice-files/NOT-OD-15-015.html; 2014. [8] Project PIR. http://www.pewinternet.org/fact-sheets/health-fact-sheet/; 2014. [9] Project PIR. http://www.pewinternet.org/fact-sheets/social-networking-fact-sheet/; 2014. [10] Bristol-Gould S, Desjardins M, Woodruff TK. The Illinois Women's Health Registry: advancing women's health research and education in Illinois, USA. Women's Health (Lond Engl) 2010;6:183–96. [11] Mitchell MD, Mannino DM, Steinke DT, Kryscio RJ, Bush HM, Crofford LJ. Association of smoking and chronic pain syndromes in Kentucky women. J Pain 2011;12:892–9. [12] Luck MB, McIntire M, Blades N, Robinson W. Characteristics of women with gynecologic cancer who enter clinical trials. J Reprod Med 2005;50:481–5.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
A novel clinical trial recruitment strategy for women's cancer B.J. Rimel ⁎, Jenny Lester, Leah Sabacan, Diane Park, Catherine Bresee, Catherine Dang, Beth Karlan Cedars-Sinai Medical Center, 8700 Beverly Blvd, Los Angeles, CA 90048, United States
H I G H L I G H T S • An online patient-facing registry improved accrual over a paper based registry. • Online registry participants matched to studies faster than paper based participants • Fifteen percent of women who participated in the registry enrolled in clinical trials
a r t i c l e
i n f o
Article history: Received 25 January 2015 Accepted 13 May 2015 Available online 20 May 2015 Keywords: Clinical trial Registry Enrollment
a b s t r a c t Objective. To address a deficiency in clinical trial and research enrollment in gynecologic cancer studies, we launched a paper based patient research registry. To improve registry enrollment, we transitioned to an online registry and trial matching mechanism to aid women in accessing open studies. Methods. Utilizing a validated verification platform, we designed a web-based registry and trial matching mechanism for women over age 18. Participants completed a questionnaire to provide information for trial matching. A focus group of registry participants was held 9 months after the start of the study to evaluate barriers to participation. Results. A total of 322 women were enrolled in the online registry over a 14 month period which was a 4.3 fold increase over the paper-based registry (p b 0.0001). Two hundred and sixty three (82%) women were matched to at least one study. Fifteen percent (39/263) of those eligible for studies went on to enroll. The online enrollment rate to studies was not different from that observed in the paper-based registry (26/172, p = 0.934), however, the web-based registry linked participants to subsequent studies 27% more rapidly (68 (+/− 98) days vs. 93 (+/−81) days for the paper-based registry, p = 0.017). Focus group participants identified areas for improvement. Conclusion. Web-based patient driven registry provides dramatic improvement in the number of participants enrolled and the time to trial linkage compared to a paper based registry at a single institution. Further studies of barriers to research participation are necessary to improve on this model. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Less than 2% of patients with cancer participate in a clinical trial in the United States [1]. Gynecologic oncology patients do not appear to participate in trials with any more frequently than other cancer types. While gains in progression free survival have continued to improve overall life span for women with advanced gynecologic cancers, more cures have not been realized. Lack of accrual to trials leads to early closure of studies and a waste of critical resources as well as extended periods of enrollment which can hinder the ability to interpret the results. Stensland et al. reported that 1 in 4 cancer clinical trials were ⁎ Corresponding author at: Cedars-Sinai Medical Center, 8700 Beverly Blvd Suite 290W, Los Angeles, CA 90048, United States. Tel.: +1 310 423 5800 (office); fax: +1 310 967 1142. E-mail address: [email protected] (B.J. Rimel).
http://dx.doi.org/10.1016/j.ygyno.2015.05.008 0090-8258/© 2015 Elsevier Inc. All rights reserved.
stopped early with 1 in 10 being stopped for poor accrual [2]. Data are somewhat limited, but a panel of experts convened by the NCI and ASCO to discuss barriers to clinical trial enrollment in 2013 [3] cited barriers in three areas as most significant: 1) patient/community – 2) physician/provider level – 3) site/organizational. Physician/provider level barriers include willingness to refer a patient for study, lack of knowledge about available clinical trials and concern regarding a patient's ability to participate [3–5]. Patient/community barriers have been noted to include being unaware of trial opportunities and complexity and stringency of the protocol [6] Both of these barriers could be addressed by providing a research registry to inform and match patients for study. Clinical trials, defined by the National Institute of Health as “a research study in which one or more human subjects are prospectively assigned to one or more interventions to evaluate the effects of those interventions on health related biomedical or behavioral outcomes”
446
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448
[7], are the primary focus for enrollment of most patients with an active malignancy. Clinical research studies such as tissue banks and longitudinal cohort studies provide invaluable data for researchers but require a consenting population and may be overlooked when the focus is on therapeutic intent. Non-interventional research studies related to screening, surveillance and survivorship offer opportunities for women without cancer to engage in research that may result in improved detection or risk assessment. However, these women are harder to find as they are not commonly present in the cancer centers where clinical trial recruitment and research participation are more the norm. Recent data from the Pew Internet Research Project suggests that 86% of women use the internet and 72% of those users are accessing the internet to find medical information [8]. Similarly, 76% of women who use the internet participate in social networking sites, making the online distribution of clinical trial and research study information and recruitment possible in this population [9]. Based on the barriers to enrollment and the common utilization of digital technology in our target population, we hypothesized that a patient-facing online registry would provide greater registry participation than a previous paper based consent and registry. We also hypothesized that a computer trial matching process would improve enrollment of registry participants to clinical trials and research studies. The goal of this study was to develop an online patient driven registry utilizing a valid online consent process to connect women interested in participating in clinical research. 2. Methods In 2010 a paper based research registry was designed to obtain epidemiologic information from female patients at a single institution. This study was approved by the institutional review board. Paper consent forms and a short demographic and medical history questionnaire were provided to female patients checking in at the imaging, breast and cancer centers. Various marketing methods were used to improve registry accrual, including posters placed in the centers, informational cards, and card-drop boxes for patients interested in being contacted in several locations and closed circuit TV ads in the hospital. Recruitment to the registry was low with an average of 5 participants enrolled per month. The paper questionnaires were then manually entered into the research database by study staff, and clinical trial matching was performed on the research database. After noting continued low accrual despite internal marketing methods, we designed a novel online consent process utilizing a validation method that meets the federal guidelines for digital consent process (21CFR). This study was approved by the institutional review board and was initiated on March 17, 2013. Using a web-based informed consent process with a validation screen, women could chose to participate through an electronic signature that verifies that they have read the consent and are attesting that they wish to participate. The vendor used to create the email signature validation was DocuSign. Participants were given clear instructions that data security will be upheld to the highest standard, but like all other online data, could be subject to unavoidable data breach. Upon completion of the online consent process, participants were asked to complete a questionnaire with demographic, geographic, cancer history and medical history questions. The data points chosen for the questionnaire represent major inclusion and exclusion criteria for a large number of clinical trials and research studies at our institution. Once the questionnaire is completed a query of available studies that match the participant's criteria is run. Patients identify how they wish to be notified of potential study matches (by email, US postal mail, or phone call) in the questionnaire. Once a match is made, the information regarding the nature of the study and the study coordinator contact information is delivered to the participant in her chosen format. The list of available studies and the correct contact information for the study coordinator is carefully updated to ensure that the appropriate information is given to participants.
To inform potential participants of the online registry, we continued to have posters in the centers and informational cards. We also distributed the website to local ovarian cancer support communities and utilized online media outlets, and social media (Facebook posts by cancer community organizations/twitter posts) to promote the website. Paper consents and questionnaires were still available in the clinics for women who chose not to enroll online. Differences in observed frequencies between the web-based and traditional paper-based registry were tested using the Chi-square test or Fisher's exact test (in the case of sparse data). Continuous data are presented as means +/− standard deviation and tested by way of a Student's t-test. All data were analyzed with SAS v9.2 statistical software. For all statistical tests the level of significance was set at p b 0.05. 3. Results A total of 507 women enrolled in the research registry from March 2010 to May 2014. Conversion to the online registry format significantly increased registry participation compared with the paper based system. The average monthly enrollment to the registry increased from 5.4 women per month with the paper based registry to 23.0 women per month with the web-based registry (p b 0.0001). In the first 14 months of the online registry a total of 322 women participated. We also noted demographic changes in the registry participants with the conversion to the online format. Online participants tended to be younger, were less likely to have been tested for BRCA mutation, and less likely to have a cancer. In addition, online participants were significantly more diverse, with more non-white participants (25% vs. 15%, p b 0.001). Demographic changes in the registry are detailed in Table 1. In the paper based registry, 172 women were eligible for at least one study (93%) while in the online cohort, only 263 (82%) were eligible for at least one study, (p b 0.001). Despite the decrease in overall eligibility there was no difference in the rate women went on to enroll in another study with 15% in both the paper-based system (26/172) and the online system (39/263; p = 0.935). In fact, there were more total enrollments in the online cohort, 39 vs. 26, but the improvement did not reach statistical significance (p = 0.528). The studies that patients participated in were a hereditary cancer cohort study, an ovarian cancer screening study, an imaging study and a tissue bank study, a disease specific biobank study, and a BRCA survey and translational study. No patients
Table 1 Comparison of demographics, genetic testing and cancer types between the paper based registry and the web based registry.
Age, y (mean +/− SD) Race White Asian Black Latino Other/Declined to state Religion Jewish Other/Declined to state BRCA tested Yes No Unknown Cancer history Breast Gynecological Other types None
All cases
Paper based
Web based
(n = 507)
(n = 185)
(n = 322)
P-value
50.4 +/− 14.0
55.8 +/− 13.1
47.2 +/− 13.6
400 (79%) 39 (8%) 19 (4%) 17 (3%) 32 (6%)
158 (85%) 11 (6%) 9(5%) 1 (1%) 6 (3%)
242 (75%) 28 (9%) 10 (3%) 16 (5%) 26 (8%)
0.0049
112 (22%) 244 (48%)
46 (25%) 139 (75%)
66 (24%) 256 (80%)
0.2538
143 (28%) 332 (65%) 32 (6%)
84 (45%) 87 (47%) 14 (8%)
59 (18%) 245 (76%) 18 (6%)
b0.0001
113 (22%) 85 (17%) 66 (13%) 289 (57%)
81 (44%) 17 (9%) 34 (18%) 69 (37%)
32 (10%) 68 (21%) 32 (10%) 220 (68%)
b0.0001 0.0005 0.0065 b0.0001
b0.0001
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448 Table 2 Comparison of eligibility and enrollment between paper based and web based registry participants.
Eligible for studies Hereditary breast and ovarian cancer cohort study Ovarian cancer screening study Tissue banking study Imaging study Irritable bowel disease biobank study BRCA survey and translational study Therapeutic clinical trials Eligibility count 0 1 2 3 4 5 Mean eligibility Enrolled in a study Among eligible cases
447
2500 2191
Paper based registry
Web based registry
P value
2000
91 (49%)
76 (24%)
b0.0001
1500
103 (56%) 12 (6%) 81 (44%) 61 (33%) 41 (22%) 0 (0%)
101 (31%) 13 (4%) 15 (5%) 169 (52%) 99 (31%) 0 (0%)
b0.0001 0.2201 b0.0001 b0.0001 0.0374
Unique Website Hits Count Envelopes Created Count Envelopes Viewed Count Envelopes Signed
1000 683 527 500
13 (7%) 48 (26%) 59 (32%) 39 (21%) 24 (13%) 2 (1%) 2.1 +/− 1.2 26 (14%) 26/172 (15%)
59 (18%) 99 (31%) 127 (39%) 30 (9%) 5 (2%) 2 (1%) 1.5 +/− 1.0 39 (12%) 39/263 (15%)
b0.0001
0
b0.0001 0.5289 0.9345
were identified from either cohort as eligible for a cancer therapeutic study in the first year (Table 2). There were significant differences in the types of studies that participants enrolled in between the two groups (Table 3). This was especially apparent in the ovarian cancer screening study where 17 women participated from the online cohort, compared to no eligible women from the paper-based cohort (p = b 0.0001). In the traditional paper cohort, it took an average of 93 +/− 81 days for a participant to be linked to another study compared to the online cohort's average time to linkage of 68 +/− 98 days (p = 0.017). This can be attributed to several factors: In the online system, staff time is not required to perform data entry and manage the consent paperwork resulting in significantly less time to determine eligibility and those who consent online seem to be quick to respond emails while the paperbased cohort more often were notified through US mail. Website traffic defined as the number of individual views of the website from separate Internet Protocol (IP) addresses did not consistently correlate with enrollment. Both web-based and in person marketing interventions were performed during the first year of the online registry. Web-based marketing resulted in more website traffic, but did not necessarily result in more enrollments to the registry during this study period. Analysis of the consent process by measuring both discreet website views as well as the validation steps required for consent completion demonstrated attrition at all steps (Fig. 1). Attrition from the website view to the consent process (Click to join) is expected as not all viewers Table 3 Comparison of study enrollment between paper based registry participants and web based registry participants. Paper based registry Study enrollment Hereditary breast and ovarian cancer 24 (13%) cohort study Ovarian cancer screening study 0 (0%) Tissue banking study 0 (0%) MRI study 0 (0%) Irritable bowel disease biobank study 1 (1%) BRCA survey and translational study 2 (1%) Number of studies enrolled by individual participant 0 159 (86%) 1 25 (14%) 2 1 (1%)
Web based registry
P value
10 (3%)
b0.0001
17 (5%) 2 (1%) 2(1%) 5 (2%) 10 (3%)
b0.0001 0.2828 0.2828 0.4233 0.1489
283 (88%) 32 (10%) 7 (2%)
306
0.1851
Fig. 1. Analysis of attrition from website view to completion of web based consent process.
will be eligible (e.g. male, less than 18 years old, etc.). A focus group of participants who consented online was held in January 2014 to discuss the consent process and how to better retain participants throughout the process. Focus group data suggested that more information about the research and clinical goals of the registry would better enable visitors to the website to continue on through the consent. Other themes from the focus group were that participants were confused by the need to toggle between windows to complete the email validation step to sign the online consent. Focus group participants also suggested that a video of the consent process with a description of how long it takes to complete (approximately 5–10 min) would also be helpful. 4. Discussion Enrollment to clinical trials and research studies remains a significant problem for cancer research. Registries of potential participants have been used previously with good success [10,11]. Several trial matching programs are available online but they rely on data from www.clinicaltrials.gov, which is frequently out of date. The availability of accurate and timely clinical trial data from a single local source and the ability to quickly connect with interested participants should improve this situation. In our evaluation, using a web-based patient facing consent process has expanded enrollment onto non therapeutic clinical trials and research studies at a single institution. The web based approach to registry participation demonstrated improvement in the number of women who enrolled in a study, but the proportion of eligible registry participants was unchanged from a traditional paper-based approach. Only 15% of those women participating in the registry ultimately enrolled in a research study. There may be several reasons for this. First, women deemed eligible based on questionnaire data may not have met all study requirements after further investigation. Second, women eligible for studies may have chosen not to participate because they did not feel the study they were offered interested them. Third, it may be that women who were notified they were eligible were interested in participation but did not follow through due to other factors such as distance from the medical center or time constraints. Unfortunately we do not have data at this time on why subjects chose not to enroll in a study to which they were linked. It is important to note that the similarity in rates of enrollment between the two methods shows that the online registry is not inferior to the paper based approach. Lastly, the digital approach was begun during a time of great transition for the cooperative group structure and severe decline in the number of available therapeutic studies in gynecologic cancer both nationally and at our institution. It is unclear how the online trial matching model would perform if more studies were available.
448
B.J. Rimel et al. / Gynecologic Oncology 138 (2015) 445–448
Decreased time from registry participation to enrollment on a study for the web based group was also a significant finding. This result seems most relevant to the timely completion of clinical trials and the time sensitive nature of both the trial eligibility and the length of funding. This also appears to be a significant programmatic improvement and the lack of manual data entry certainly improves staff time and cost. However, there was no difference in the proportion of registry participants who ultimately enrolled in research study between the two groups. This may be due to the significantly higher proportion of women eligible in the paper group, representing a population that was already in the medical center for some reason related to their health. There was a significant improvement in the accrual to an interventional ovarian cancer screening trial in the web based registry. We hypothesize that this is due to a younger population in the online cohort as there was no difference in eligibility between the two groups. The use of web metrics to evaluate the most effective recruitment methods for the web based approach and the use of focus groups to evaluate the areas of attrition are strengths of this study. Unfortunately, despite the presence of a short survey that is activated when a potential participant exits the questionnaire without completion, there were very few responders to the survey. Lack of data about why participants choose not to complete the questionnaire remains a weakness of the study. Comparison of number of registry enrollments between paper based and web based approaches showed that the web-based approach had significantly more registry enrollment. However, it should be noted that because of the nature of the website being able to be accessed by women outside of the medical center some or all of this increase could be due to the larger pool of potential participants from which to draw. Luck et al. examined the characteristics of 236 women diagnosed with gynecologic cancer between 1997 and 2001 regarding their participation in clinical trials [12]. In this cohort, 54 (23%) women participated in an NCI sponsored clinical trial during the time period of the study. The study concluded that the most important factor in women choosing to enter a clinical trial was their familiarity with the physician–investigator who describes the trial. These results suggest that a trusted source of information is essential to encouraging participation in a clinical study. In an era where 72% of internet users are looking for health information [8], that trusted source may be less of a person and more of a domain
name. The web based registry may provide that source for clinical trial knowledge for internet users. In conclusion, the web based registry improved overall participation and significantly decreased time to enrollment. Future improvement of this model will require innovative methods to capture complex data from patients such as molecular tumor profiling to further decrease the time to enrollment for studies and provide more comprehensive pre-screening for patients. These critical data areas should aid in locating and engaging patients for time sensitive cancer therapeutic trials. Conflict of interest The authors have no conflict of interest to disclose.
References [1] Mannel RS, Moore K. Research: an event or an environment? Gynecol Oncol 2014; 134:441–2. [2] Stensland KD, McBride RB, Latif A, Wisnivesky J, Hendricks R, Roper N, Boffetta P, Hall SJ, Oh WK, Galsky MD. Adult cancer clinical trials that fail to complete: an epidemic? J Natl Cancer Inst 2014;106. [3] Denicoff AM, McCaskill-Stevens W, Grubbs SS, Bruinooge SS, Comis RL, Devine P, Dilts DM, Duff ME, Ford JG, Joffe S, Schapira L, Weinfurt KP, Michaels M, Raghavan D, Richmond ES, Zon R, Albrecht TL, Bookman MA, Dowlati A, Enos RA, Fouad MN, Good M, Hicks WJ, Loehrer Sr PJ, Lyss AP, Wolff SN, Wujcik DM, Meropol NJ. The National Cancer Institute-American Society of Clinical Oncology Cancer Trial Accrual Symposium: summary and recommendations. J Oncol Pract 2013;9:267–76. [4] Lara Jr PN, Higdon R, Lim N, Kwan K, Tanaka M, Lau DH, Wun T, Welborn J, Meyers FJ, Christensen S, O'Donnell R, Richman C, Scudder SA, Tuscano J, Gandara DR, Lam KS. Prospective evaluation of cancer clinical trial accrual patterns: identifying potential barriers to enrollment. J Clin Oncol 2001;19:1728–33. [5] Comis RL, Miller JD, Colaizzi DD, Kimmel LG. Physician-related factors involved in patient decisions to enroll onto cancer clinical trials. J Oncol Pract 2009;5:50–6. [6] Mills EJ, Seely D, Rachlis B, Griffith L, Wu P, Wilson K, Ellis P, Wright JR. Barriers to participation in clinical trials of cancer: a meta-analysis and systematic review of patient-reported factors. Lancet Oncol 2006;7:141–8. [7] Health NIo. Notice of revised NIH definition of “Clinical Trial”. http://grants.nih.gov/ grants/guide/notice-files/NOT-OD-15-015.html; 2014. [8] Project PIR. http://www.pewinternet.org/fact-sheets/health-fact-sheet/; 2014. [9] Project PIR. http://www.pewinternet.org/fact-sheets/social-networking-fact-sheet/; 2014. [10] Bristol-Gould S, Desjardins M, Woodruff TK. The Illinois Women's Health Registry: advancing women's health research and education in Illinois, USA. Women's Health (Lond Engl) 2010;6:183–96. [11] Mitchell MD, Mannino DM, Steinke DT, Kryscio RJ, Bush HM, Crofford LJ. Association of smoking and chronic pain syndromes in Kentucky women. J Pain 2011;12:892–9. [12] Luck MB, McIntire M, Blades N, Robinson W. Characteristics of women with gynecologic cancer who enter clinical trials. J Reprod Med 2005;50:481–5.
