|Home | About | Journals | Submit | Contact Us | Français|
In response to a 2005 NIH solicitation, sixteen investigators received funding to test interventions that would reduce the barriers that prevent cancer patients from receiving adequate and appropriate symptom management therapies. Since the awards have been issued, the investigators have met two times and have identified a number of challenges to implementing their respective studies. A survey was conducted that focused on their experiences with hiring and retaining study personnel, gaining Institutional Review Board (IRB) approval, incurring unexpected costs, challenges to accruing participants, and a listing of standard measures used in the study. The survey was completed online by the Principal Investigator (PI) for each project in late 2006 and the initial results were confirmed one year later by resending the initial survey and by a follow up phone call. All but one PI completed the survey. Obtaining Institutional Review Board (IRB) approval, hiring and recruiting research personnel, establishing subcontracts and accruing research subjects were the primary challenges experienced by the investigators. This palliative care solicitation achieved more than its original intent of stimulating research in overcoming barriers to delivering cancer symptom management, palliative and end-of-life care. From a survey on the challenges and issues that emerged from their projects, grantees were able to identify specific hurdles and their unique solutions that may help other investigators as they plan their program of research.
In 2005, the National Cancer Institute (NCI) and the National Institute of Nursing Research (NINR) issued a Request for Applications (RFA) entitled, “Reducing Barriers to the Delivery of Symptom Management and Palliative Care (CA-05-013)1.” The purpose of the RFA was to solicit grant applications for research directed at developing and testing interventions to reduce or overcome barriers of the delivery of evidence-based symptom management and palliative care to patients suffering from cancer and/or treatment-related sequelae. The impetus for this solicitation was based on the recurring problem that was identified from a series of meetings and related publications2,3,4,5 specifically, that because of a variety of patient, clinician, and health system barriers, cancer patients do not receive adequate, evidence-based symptom management or palliative care throughout the course of their disease. Applications for this RFA were expected to cover the entire cancer trajectory from diagnosis, through survivorship, to end of life and emphasize understudied areas and gaps in the area of symptom management and palliative care. This report describes the background that led to the solicitation and the results of the survey completed by the Principal Investigators who were funded through this announcement.
Seventy applications were submitted to this call and 16 projects were funded that focused on a variety of barriers, diverse populations, and understudied disease sites along the cancer continuum. Funding for all projects began in September 2005. Depending on the scope of the project and funding mechanism, support ranged from two years (8 projects) to five years (6 projects), with two projects funded for 3 and 4 years respectively. Table 1 and Table 2 list the grantees, titles of their projects, and selected key characteristics.
One of the requirements of the funded investigators is to attend a yearly meeting in which study plans, findings, and issues of common interest and concern are shared and discussed. At the time of this report, two meetings have been conducted in September 2006 and 2007. At both meetings, the vast majority of grantees were encountering difficulties in achieving their accrual goals, of which most of the difficulties were not related to lack of access to the study population
To better understand these difficulties one of the authors (BF) designed and conducted a survey of the grantees, which asked the respondents a series of forced-choice and open-ended questions about their experiences with hiring and retaining study personnel, gaining Institutional Review Board (IRB) approval, incurring unexpected costs, challenges to accruing participants, and a listing of standard measures used in the study. All attendees agreed to participate and noted this as a rare opportunity to collectively capture data on a cohort of investigators studying a similar problem and beginning their projects at the same time. Information from the survey was shared with the NCI official in order to understand the full range of issues and identify recommendations.
The survey was completed online by the Principal Investigator for each project in late 2006, approximately one year after initiation of the projects. These initial results were confirmed one year later by resending the initial survey and by a follow up phone call from one of the co-authors. This was done to capture additional experiences that may have occurred as the projects proceeded and to capture any individual issues at that site not reflected in the written survey. All but one Principal Investigator completed the survey.
Table 3 presents data regarding challenges in institutional approval of the project. Nine of the projects (60%) reported challenges in institutional approval. The issues reported reflect the nature of the projects as clinical research in various community settings and involving multiple sites. The most common issues faced involved consent procedures, (such as unfamiliarity of the IRB with psychosocial studies and concern about telephone consents), clinical sites unfamiliar with IRB procedures, delays in IRB meetings, and the complexity of multi-site approvals. These issues are important for the future of end-of-life care research which relies on community hospice programs and multi-site research to obtain sufficient numbers of subjects.
Table 3 presents findings related to personnel and the grantee institution on the Barriers projects. Ten of the projects (66%) experienced challenges in recruitment or hiring of personnel, a factor known in clinical research to delay the initiation of the study. Investigators were also asked if there had been changes in personnel as the projects progressed. Nine of the fifteen (60%) had such personnel changes.
Table 3 reports findings related to subcontracts. An interesting finding was how common multi-site research is in this field with 10 projects (66%) reporting use of subcontracts to other sites. Of these 10 projects, 70% reported difficulties or delays in establishing the subcontracts. Comments from the survey respondents indicated problems from both the grantee institution as well as subcontractor sites.
Table 3 presents data from survey items related to subject accrual and minority recruitment. Twelve of the projects (80%) reported problems in accrual of subjects. The nature of the problem included several factors and the most common was protectiveness of clinical staff in subject accrual, an important issue in palliative care research. The acuity of the patients and proximity to time of death were other common issues.
Investigators were also asked about experiences in recruitment of minority patients. Fifty-three percent (53%) or 8/15 investigators replied that they had been able to recruit minority subjects. The most common recruitment strategy was having research staff attend clinics/rounds.
The projects used a variety of methods in their research designs as illustrated in Table 3. As directed by the initial NCI/NINR RFA, all projects had to be testing interventions and 10 of the projects involved randomized trials. Patient interviews were common as was abstraction of chart data. Interestingly, eight projects (53%) reported monitoring the fidelity of a clinical intervention included in the study.
The final question asked the investigators the measurement tools were used in the study. The intent of this question was to determine if there was common use of established tools that might facilitate comparing data across studies. The response to this question was very informative as common use of a few established tools was not reported. A total of 80 different scales were reported as used across the fifteen projects and only nine tools were used in more than one site. Among the 80 scales, 12 had never been published or had been designed specifically for the project. This finding indicates that there is little uniformity in measures and significant use of project initiated scales.
The ability to follow a cohort of 16 investigators involved in similar research that began at the same time is a rare opportunity and has resulted in rich qualitative data that revealed multiple challenges experienced by the investigators. Some are well-known when conducting symptom and palliative care research and others unforeseen yet provide valuable lessons learned for investigators, the funding agency and grant reviewers for these issues of special importance in symptom and palliative care research. In addition, the varied and innovative methodological approaches implemented by this cohort will contribute to the growing science of symptom management and palliative and end-of-life care.
Obtaining Institutional Review Board (IRB) approval, hiring and recruiting research personnel, establishing subcontracts and accruing research subjects were the primary challenges experienced by the investigators. Though these challenges occur independently, they can also be interrelated. Subject accrual is a well-known challenge primarily due to the serious nature of the subjects’ illness with significant disease and/or symptom burden and close proximity to death. Moreover, once enrolled, there is high attrition due to declining medical status resulting in large amounts of missing data. In an effort to offset this challenge, many of the investigators conducted their research in multiple sites. Consequently, many encountered significant delays in IRB approval and establishing subcontracts. Though anticipated, very few investigators encountered delays or challenges due to the new HIPAA guidelines.
Herr involved 16 community-based hospices in three states in the Midwest. Overall it took 12 months to obtain IRB approval for the 16 sites. Issues encountered included hospices who never participated in federal research requiring application for a Federal Wide Assurance (FWA)6, local IRBs who met quarterly versus monthly and sites that lacked access to a local IRB. The complexity of involving multiple sites can be mitigated by obtaining as much information as possible about each site, such as their IRB process, previous involvement with federal research, and developing plans for meeting individual accrual goals. In many institutions investigators cannot obtain IRB approval until their funding has been secured, and it is important to determine this early in the process.
The dilemma of how to streamline IRB reviews in the context of multi-site research is becoming ever more critical. The beginning of the modern IRB system is relatively new, dating back to the mid-1960s7 and at that time, the role of IRBs was primarily focused on local investigator-initiated single-site studies. As data from this survey show, many IRBs as currently configured are not equipped to efficiently review multi-site research projects. Newer approaches including regional ethics organizations and web-based programs for cooperative IRB review are emerging as possible solutions8.
Because of the high symptom and disease burden, palliative care clinicians focus their attention on patient comfort and naturally protect their patients. Consequently, offering a clinical study to this population may be perceived as placing undue burden on the patient. In this survey, staff gate keeping of the patients was another challenge to subject accrual. Clinic or hospital staff would not recommend eligible patients for the study because of the perception that they were too ill or tired or participation would add time and burden to the patient’s care. In response to this challenge, clinical institutions are using patient navigators to minimize barriers and increase minority accrual to clinical trials. One example of this is an NCI funded project, “Cancer Disparities Research Partnership Program”. Dr. David Kahn, one of the investigators involved in the project, realized the patient navigators had biases against clinical trial participation similar to the patients they were trying to recruit.9 Educating the staff and patients about the nature of the study; time and effort involved; and associated risks and benefits, can ease the perceived burden of the research. Despite these challenges, this group of investigators did employ strategies to decrease burden to their subjects. These included avoiding long assessment periods, unnecessary documentation, and overlong project duration, all of which have been documented in the literature as lessons learned.10
Other factors can contribute to the difficulty in conducting supportive care and behavioral research: lack of familiarity with supportive and behavioral research by the sites and local IRBs, prioritization of research involving cancer treatment and a lack of understanding of the importance of the research to be conducted. Each of these was encountered by one or more of the investigators. In a study conducted by Crowley and Casarett to determine the feasibility of using screening questions to identify patients interested in participating in research, patients were less likely to believe symptom-related research would provide potential benefit (10%) compared to disease-modifying research (33%)11.
In addition to facing these challenges, the investigators were asked to consider vulnerable populations such as children, the elderly, racial and ethnic minorities and individuals of lower socio-economic status. These underrepresented populations face numerous barriers to participation in clinical trials including lack of education about clinical trials, lack of provider referral, perceived harms of clinical trial participation, time commitment and loss of income, transportation, mistrust of research and the medical system, and inadequate health insurance.12 The investigators developed innovative projects to include these populations and address the barriers to palliative care and symptom management in the context of the known barriers to participation in clinical trials.
The funded projects emphasized the research gaps in symptom management and palliative and end-of-life care. Despite increased clinician education, the development of standards of care, published guidelines and published patient, clinician and system barriers, pain and cancer/treatment related symptoms continue to be under or poorly treated. Several investigators addressed these issues.
To advance palliative care and symptom management, it is essential to not only develop evidence-based interventions but to ensure their dissemination into practice and see them utilized appropriately. This can be particularly challenging in home and hospice settings, which is where several of the investigators chose to conduct their research.
These investigators met many of these challenges, demonstrated flexibility and persevered. Becoming aware of many of the potential challenges will help investigators identify insurmountable problems and quickly move to another approach to prevent significant delays. In several situations, the solutions and options came from outside the institution. For example, because of unforeseeable delays in starting projects and accruing participants, two of the R01 investigators have negotiated with NIH to restructure their budgets and will have an additional year to complete their work.
The early findings from this one time solicitation can serve to assist funding agencies in their development of future initiatives for palliative care research, as well as to guide potential applicants when interacting with a funding agency. As learned from this cohort of investigators, the feasibility for conducting these types of studies is a significant challenge. Investigators should carefully consider funding mechanisms, such as the NIH Exploratory Grant Mechanism (R21)13, which are specifically designed to support feasibility and pilot studies, before embarking on a larger more long term project.
Although this RFA included a special review panel to evaluate the scientific merit of the applications, this is not always feasible. In the absence of a special review panel, it is imperative that the investigator request, at the time of submission, the expertise needed to review the application. As the field matures and more qualified reviewers are available to review applications, standard study sections will become populated with adequate numbers of reviewers who are knowledgeable of and sensitive to the issues facing palliative care researchers.
When applying for NIH support, a very important consideration is to contact the Program Director, particularly, when there is a specific solicitation. By taking the time to discuss the relevance of the particular project to the goals and objectives of the solicitation, the investigator will get a better sense of how to shape the application so that it is responsive to the solicitation.
Investigators seeking R01 funding for their palliative care research projects are not required to cite a specific solicitation on their applications. However, speaking with a Program Director remains critically important. Because palliative and end-of-life issues cut across the expertise and resources of several NCI Divisions, no single unit within NCI has been designated as the focal point for palliative care research and this remains true today. The NCI Web site can help investigators find information and the most appropriate Program Director. Using the term ‘palliative care’, a recent search of the NCI Website identified six “Best Bet” links and another 625 links to a variety of documents and reports14. Among the six “Best Bet” links, two are particularly informative for investigators. The link entitled “Funding Opportunities in Symptom Management and Palliative Care Research”15 will yield a listing of all NIH funding opportunities related to this area of research. The other link entitled, “Cancer Research Portfolio: Palliative Care Research Projects“16 lists all current NCI funded projects, the Principal Investigator(s), and the NCI Program Director assigned to the Project. A cursory review of the 168 funded projects yielded five different Program Directors in three different Divisions managing these grants.
Applications that were submitted in response to this solicitation (CA 05-013) were reviewed by a special panel of scientists with expertise in symptom management, palliative and end-of-life care, health services research, as well as statistical methodology. The standard criteria of significance, approach, innovation, investigator qualifications, and environment were used to evaluate the applications. Despite the expertise of the study section, nowhere in the review criteria would the reviewers have been able to predict the unique hurdles and challenges that these investigators would later encounter. It is imperative that study sections/reviewers are available to evaluate palliative care and symptom management research who consider its unique challenges as seen with this RFA.
Of the eight R21 investigators, several have indicated that their projects are not feasible for a larger (R01) study, with poor patient accrual being the major reason. This finding, while not optimistic, is the main point of R21 projects, to determine feasibility and obtain pilot data. At the last annual meeting, several investigators, whose projects include similar populations, are collaborating on a project for their next study.
This palliative care solicitation achieved more than its original intent of stimulating research in overcoming barriers to delivering cancer symptom management, palliative and end-of-life care. From a survey on the challenges and issues that emerged from their projects, grantees were able to identify specific hurdles and their unique solutions that may help junior investigators as they plan their program of research. In addition, other issues emerged from the survey, specificially, the need to include instrument development as evidenced by the considerable diversity in assessment tools used in this survey. Finally, when evaluating budgets, reviewers and funding agencies should take into account the additional time and staff required to address accrual difficulties and high rates of attrition.
Significant lessons have been learned by these investigators which will enhance their future research efforts. By having the opportunity to share challenges as well as accomplishments, the investigators learned a variety of lessons and engaged in a rare opportunity to be supported and mentored.
Ann M. O’Mara, National Cancer Institute, 6130 Executive Blvd., Bethesda, MD 20892.
Diane St. Germain, National Cancer Institute, 6130 Executive Blvd., Bethesda, MD 20892.
Betty Ferrell, City of Hope Comprehensive Cancer Center, 1500 East Duarte Road, Duarte, California 91010.
Tami Bornemann, City of Hope Comprehensive Cancer Center, 1500 East Duarte Road, Duarte, California 91010.