|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 15 (August 1), 2004: pp. 2997-3002 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.10.073
Managing Accrual in Cooperative Group Clinical Trials
1 Roswell Park Cancer Institute, Buffalo, NY Designing a multicenter clinical trial is a process affected by time, economic, and political constraints. Unfortunately, this can result in overlooking needs or concerns of the potential participants of the study. Given the magnitude of resources spent on developing and implementing clinical trials, we have examined the value of systematic accrual management at various stages of protocol development, activation, delivery, and enrollment. Magnitude of the Problem The National Cancer Institute funds multiple large adult cooperative oncology groups (see www.ctsu.org) for the important missions of preventing cancer and optimizing the treatments for and quality of life of cancer patients. Their combined budgets total over $160,000,000 annually.1 One such organization, the Cancer and Leukemia Group B (CALGB) administers 100 active protocols while it develops new studies and analyzes completed research concurrently. Over a 9-year retrospective tabulation (1989 to 1997), 14 (9.4%) of 149 CALGB protocols ended because of failure to meet accrual goals. These studies generated too few data to analyze, and so there was no information about treating or preventing cancer to share with the medical and scientific community. These were the most extreme cases of a larger group lagging behind accrual goals. Approximately 20% of the currently active CALGB treatment studies are lagging behind the expected accrual postulated in the study protocol. Since other cooperative cancer groups are organized similarly, there are impressive opportunities for improved efficiency by developing methods of accrual enhancement. Smaller or noncancer study groups could benefit as well. Low accrual impacts individual institutions, given the considerable resources used to generate and maintain regulatory documents. During a recent review at one center (Roswell Park Cancer Institute, Buffalo, NY with an average of 160 active therapeutic protocols annually), 43 cooperative group and other therapeutic studies were closed/terminated after the audit showed prolonged low accrual. The following year, another 23 therapeutic studies were closed for similar reasons. Measurement Dilemmas—Why Collect Data? Health care providers naturally embrace new systems developed using scientific methods, but show aversion to applying business strategies to health care matters.2-5 Peter Drucker (an expert in modern management) used quantifying scientific methods in business systems and is credited with such quotes as, "If you can't measure it, you can't manage it."6 Instruments currently exist to measure opinions and interests, such as those used to rate the quality of speakers for continuing education courses. As frequently occurs in committees, time constraints prevent expressions of concerns. Also, individuals avoid presenting opposing viewpoints to prevent divisive or uncooperative appearances that could reduce their standing in the organization or dampen others' enthusiasm for a study. Investigators could, however, document their concerns in writing anonymously while others speak about the protocol. Similarly, by measuring the concerns of potential participants, improved accrual techniques could overcome these barriers. The Business Analogy—Why Have Implementation Plans? At present, individual investigators who offer studies to patients are responsible for devising their own methods of presenting information, such as risks and benefits, about the protocol to potential participants. This is often done in an untested way, without sufficient understanding of the psychosocial issues of the study population or with any consistency of approach. Individuals may support the proposed research ideologically, but may have difficulty avoiding some personal bias when discussing the option of going on a study with patients. Businesses, on the other hand, prefer and promote clear messages regarding their products. They outline key advantages regarding the product and formulate methods to convince consumers of the value of that product. While it is important to avoid coercion or the "selling" of a clinical trial, important new research should not fail as a result of poor investigator/participant education. Uniform Grant Proposal Analogy—Why Predict Potential Problems? In basic science laboratory research, careful peer review is applied to the section where the investigator anticipates problems with key experimental methods that might prevent the research from going forward (fatal flaws). Most successful grant applications delineate alternate ways to successfully complete the research for as many problems as the investigator can anticipate. Multicenter cooperative group protocols often require more preparation than grants for single-center clinical or laboratory research. This sort of strategic planning for recruitment in a clinical trial is uncommon, possibly because of difficulties assessing potential participant enthusiasm or because there is no specified expectation for such planning. Cost and Efficiency Cooperative groups and other organizations that support multicenter clinical trials have embarked on an important yet costly activity. In addition to finances, resources such as time, energy, and passion support active clinical trials. Improving the recruitment rates of trials would: (1) allow studies to finish faster, (2) achieve the statistical goals of the study, (3) improve patient care because results could be published more quickly and new studies that answer more questions could be developed faster, and (4) improve the cost per study or efficiency rate, freeing resources for more studies to be completed. If clear accrual measurements and contingency plans existed, fatally flawed protocols would not be opened, or would be closed sooner, and thus valuable resources could be reallocated for new activities and other studies. Ethics and Informed Consent Individuals participating in clinical research understand by the informed consent process that outcomes are uncertain and that there may not be benefits to themselves or others.7 However, participants may not understand that poor accrual could close the study prematurely, thus preventing the gaining of any meaningful knowledge. The ethical concern/question then becomes: would patients have agreed to participate in a study if they knew their participation could be for naught? The following recommendations have been developed based on pilots run with the CALGB Committee on Advocacy, Research Communications, and Ethics. Recommendation 1: Study Concept Stage—Harness the Full Potential of the Protocol Development Team by Offering Anonymous Written and Electronic Portals of Communication Figure 1 depicts the protocol evaluation form used in select CALGB committees; collected data are analyzed from successive meetings. Such anonymous communication allows full participation by those unwilling to comment openly in group meetings or who are unable to comment because of time limitations or absence. E-mail and web-based applications can also improve communications for investigators (as well as for clinical trial staff) not able to attend meetings, and these applications are currently being developed for several CALGB protocols. Recently, investigators' comments from these forms helped plan the active study CALGB 30102 (Phase III Comparison of Catheter Based Therapy of Pleural Effusions in Cancer Patients). Written criticisms were addressed directly at meetings. Similar data have quantified the need to support or abandon other proposed research.
Creating a list of perceived challenges to accrual, and potential action steps that may solve them, is useful to consider in the early stages of protocol development. Additionally, consultations with patients, physicians, advocates, or individuals outside the development team may provide fresh insights on the issues of participant interest, randomization concerns, and investigator equipoise. Recommendation 2: Protocol Development Phase—Develop a Plan for Accrual Management Using Methods Pertinent to the Study, and Document the Plan in the Protocol Before protocol development is completed, a well-considered experimental design can be communicated clearly to others outside the core planning group. Figure 2 depicts a template designed into a phase III protocol recently activated, (CALGB 30102) and includes tools mentioned below.
Conceptual designs are normally presented to specific disease and modality committees which are made up of cooperative group investigators. At this time, a protocol is measured by written feedback using the Protocol Evaluation Form (Fig 1), which helps identify major problems with a study's design and assesses interest of potential contributing investigators. Specific goals for accrual and rates of study activation can be prepared with remedies for each contingency. Rather than delegating the education process to clinical practices without guidance, talking points that show the merits of the research or the equipoise inherent in a randomized trial can be provided. An institutional review board (IRB) -approved discussion guide included within the protocol document can reduce accrual failures caused by such interview variability. The most likely referral patterns for a proposed study can be identified; focused meetings can be developed to update and involve physicians (both inside and outside the cooperative group), other health care providers, and patient advocacy groups. Research nurses, clinical research associates, and similarly trained individuals are an important sounding board because they help implement the study and attract participants. They often identify weaknesses in study design and offer helpful accrual strategies. Within the accrual management protocol template, a distribution plan for the study can also be provided. Presentation materials for physician conferences (like grand rounds) may initiate such a plan. Another strategy is to provide a recruitment hotline to standardize the enrollment process at multiple centers.8 Methods to detect, analyze, and possibly correct future problems that might cause termination of the study can also be included in this section. The accrual management plan can also anticipate possible IRB concerns and include steps that may help resolve them. Recommendation 3: Capture Concerns With the Study Design and Address Them There are several indicators of interest to track as the study is activated. First, the rate at which studies are submitted to local IRBs can be an indirect indicator of the relative priority that is given to trials at the participating centers. Problems with insufficient staffing of data managers and research nurses at certain centers may be detected at this point.9 Once submitted, IRB concerns that delay approval can be documented and compiled by the lead investigator. Principal investigators should present the research proposal to health care providers well-positioned to refer patients, and a variation of Figure 2 could document their concerns. The CALGB has done such surveys with regard to special patient populations.10 Given the money spent in the development process and the potential for more timely intervention, funding local meetings, such as Grand Rounds, to discover grassroots reactions to a new study seems logical. For sites with low participation rates, documenting patient screening or nonaccrual may provide areas for improvement to increase accrual.11,12 Unfortunately, there is no such thing as a partial accrual credit for those interviewed, but not entered, onto a trial, so ways to reward centers for compiling screening data need to be explored. Once the results from these steps are available (preferably within 6 to 12 months after study activation), discussions based on data, rather than anecdotes, can occur at cooperative group meetings. This process may help to identify actions to increase accrual rates or may identify severe flaws in study design requiring study revision or termination. With accrual management planning, as described in this article, study closures should be infrequent; although they may occur occasionally due to the subsequent development of major new treatment options or unexpected adverse events.13 A directed response to ameliorate the concerns of patients, key physicians, or research staff can be developed and implemented. Focus groups of physicians, research staff, patients, or patient advocates should be consulted. Specific interventions for special groups, such as women, children, and minorities, should be developed when these groups are under-represented in clinical trials.14,15 IRB-approved patient or physician pamphlets, updated slide presentations, and video or multimedia presentations may be used. The ongoing success of cooperative clinical trials has, like the practice of medicine in general, led to high expectations by patients, funding agencies, and health care providers. Management methods to speed protocol activation and accrual should enhance an already successful system. Like many other management initiatives, the recommendations cited herein focus on communication enhancement tools. While quantifying the benefits of these tools may be challenging, it is possible by comparing protocols that embrace these methods with similar studies that employ traditional ones. We found that the rates of accrual correlated with favorable survey data on activated protocols. Further analyses may show independently predictive parameters that can reduce survey complexity. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest.
REFERENCES 1. National Cancer Institute: FY 2001 Appropriation. Bethesda, MD, NCI, 2001. http://www3.cancer.gov/admin/fmb/2001appropriation.pdf, 2003 2. Anonymous: Are they patients or customers? Physician Relat Update 7:78-79, 1998[Medline] 3. Mulhall KJ, Ahmed A, Masterson E: The "doctor-customer" relationship: Hippocrates in the modern marketplace. Int J Health Care Qual Assur Inc Leadersh Health Serv 15:9-10, 2002[Medline] 4. Piper LE: Customer vs. patient: A sublime difference in hospitals. Hosp Health Serv Adm 31:126-132, 1986[Medline] 5. Ramdass MJ, Naraynsingh V, Maharaj D, et al: Question of ‘patients’ versus ‘clients’. J Qual Clin Pract 21:14-15, 2001[Medline] 6. Drucker PF: Practice of Management. New York, Harper Business, 1991, p 344 7. Schron EB, Wassertheil-Smoller S, Pressel S: Clinical trial participant satisfaction: Survey of SHEP enrollees. SHEP Cooperative Research Group. Systolic Hypertension in the Elderly Program. J Am Geriatr Soc 45:934-938, 1997[Medline] 8. Sutton-Tyrrell K, Crow S, Hankin B, et al: Communication during the recruitment phase of a multicenter trial: The recruitment hotline. Control Clin Trials 17:415-422, 1996[Medline] 9. Street A, Strong J, Karp S: Improving patient recruitment to multicentre clinical trials: The case for employing a data manager in a district general hospital-based oncology centre. Clin Oncol (R Coll Radiol) 13:38-43, 2001 10. Kornblith AB, Kemeny M, Peterson BL, et al: Survey of oncologists' perceptions of barriers to accrual of older patients with breast carcinoma to clinical trials. Cancer 95:989-996, 2002[CrossRef][Medline] 11. Foster D, Cook D, Granton J, et al: Use of a screen log to audit patient recruitment into multiple randomized trials in the intensive care unit. Canadian Critical Care Trials Group. Crit Care Med 28:867-871, 2000[Medline] 12. Gennari JH, Reddy M: Participatory design and an eligibility screening tool. Proc AMIA Symp:290-294, 2000 13. Plaisier PW, Berger MY, van der Hul RL, et al: Unexpected difficulties in randomizing patients in a surgical trial: A prospective study comparing extracorporeal shock wave lithotripsy with open cholecystectomy. World J Surg 18:769-772, 1994[Medline] 14. Bleyer WA: Cancer in older adolescents and young adults: Epidemiology, diagnosis, treatment, survival, and importance of clinical trials. Med Pediatr Oncol 38:1-10, 2002[CrossRef][Medline]
15. Hutchins LF, Unger JM, Crowley JJ, et al: Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med 341:2061-2067, 1999
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
