Patient Navigation As a Model to Increase Participation of African Americans in Cancer Clinical Trials

Abstract

Purpose:

Less than 10% of patients enrolled in clinical trials are minorities. The patient navigation model has been used to improve access to medical care but has not been evaluated as a tool to increase the participation of minorities in clinical trials. The Increasing Minority Participation in Clinical Trials project used patient navigators (PNs) to enhance the recruitment of African Americans for and their retention in therapeutic cancer clinical trials in a National Cancer Institute–designated comprehensive cancer center.

Methods:

Lay individuals were hired and trained to serve as PNs for clinical trials. African American patients potentially eligible for clinical trials were identified through chart review or referrals by clinic nurses, physicians, and social workers. PNs provided two levels of services: education about clinical trials and tailored support for patients who enrolled in clinical trials.

Results:

Between 2007 and 2014, 424 African American patients with cancer were referred to the Increasing Minority Participation in Clinical Trials project. Of those eligible for a clinical trial (N = 378), 304 (80.4%) enrolled in a trial and 272 (72%) consented to receive patient navigation support. Of those receiving patient navigation support, 74.5% completed the trial, compared with 37.5% of those not receiving patient navigation support. The difference in retention rates between the two groups was statistically significant (P < .001). Participation of African Americans in therapeutic cancer clinical trials increased from 9% to 16%.

Conclusion:

Patient navigation for clinical trials successfully retained African Americans in therapeutic trials compared with non–patient navigation trial participation. The model holds promise as a strategy to reduce disparities in cancer clinical trial participation. Future studies should evaluate it with racial/ethnic minorities across cancer centers.

INTRODUCTION

The scientific value of cancer clinical trials has been established. Over the past several decades, remarkable progress in the treatment of cancer has been achieved through federally and industry-sponsored clinical trials. It also has been established that, for a number of scientific, methodologic, ethical, and social reasons, participation from all population groups is required for the success of clinical trials.^1-3 Although the enrollment and retention of patients in cancer clinical trials is difficult among all population groups,^4-7 it is especially challenging among racial/ethnic minorities.⁸ As a result, less than 10% of all patients with cancer enrolled in clinical trials are minorities.⁴

A number of barriers to the recruitment of minorities for, and their retention in, clinical trials have been described in the literature.^9-12 Although minorities are just as willing to be involved in research studies as the general population,^13,14 efforts to recruit and retain them in cancer clinical trials necessitate more labor-intensive approaches^15-19 and more personal contacts.^12,20,21 These findings speak to the importance of culturally appropriate strategies for recruitment of underrepresented minorities for, and their retention in, cancer clinical trials.

Patient navigation, or the use of lay community health workers to help patients overcome barriers in their communities and in the health care system, has been used to assist racial/ethnic minorities to obtain medical care.^22-33 The model has also been used to recruit rural and low-income, mostly African American, patients into research studies, and projects report overall positive results.^34-36 Despite the success of the patient navigation model in facilitating access to care, and evidence suggesting that patient navigators (PNs) can address barriers to clinical trial participation for ethnic/racial minorities,^37-39 the patient navigation model has not been evaluated systematically as a tool for increasing the participation of minorities in clinical trials, and outcome measures to assess its effectiveness are scarce.⁴⁰ The Increasing Minority Participation in Clinical Trials (IMPaCT) project used patient navigation to enhance the recruitment of African Americans for, and their retention in, therapeutic cancer clinical trials in one racially and socioeconomically diverse National Cancer Institute (NCI) –designated comprehensive cancer center. This article reports on the project’s impact over a period of 8 years, between 2007 and 2014.

METHODS

IMPaCT was implemented at the University of Alabama at Birmingham (UAB) Comprehensive Cancer Center beginning in 2007. The program’s goal was to educate African American patients with cancer about clinical trial participation and to assist with the recruitment of African American patients with cancer for, and their retention in, therapeutic cancer clinical trials. We hypothesized that recruitment for and retention in therapeutic cancer clinical trials would be higher among African American clinical trial participants who received patient navigation services compared with those who did not. The study was approved by the UAB Institutional Review Board for Human Use (Protocol X060406009).

Theoretical Model

The IMPaCT project was based on the Community Health Advisors Network (CHAN) model,^35,41,42 which enables community members to serve as educators, navigators, and agents of change. The CHAN model is based on a social network theory about the health-enhancing effects of social support and community organization.⁴³ The model reflects the fact that natural helpers or informal leaders in the community are recognized by their network (friends, families, and neighbors) as reliable sources of advice, help, and support.⁴⁴ Community health advisors (CHAs) have often been used to provide a linkage between community members and the health care system.⁴⁵ IMPaCT used the CHAN model to train community members as PNs who educate African American patients with cancer about clinical trials and help them overcome barriers to clinical trial participation.

Recruitment of PNs

Individuals who were already serving as CHAs for cancer prevention and control or embodied the qualities of CHAs were considered for these positions. The project team advertised the positions in the community, reviewed applications, and conducted a series of interviews. Two lay individuals matching the demographic characteristics of the patients (African American women from the Greater Birmingham area) were hired and trained to serve as full-time PNs.

Training of PNs

A training curriculum and training manual were developed. The training consisted of a background section and three modules. The background section educated PNs about the research team and the specific roles of team members and presented an overview of the research process. Module I offered an overview of the patient navigation as a concept, and training about cancer clinical trials using NCI Cancer Clinical Trials publications. Module II provided information about the navigation process, interacting with patients, patient interventions, and the clinic environment. In Module III, the PNs received case-management and data-management training, as well as training on how to explain the clinical trials consent process to potential participants. The training sessions were co-led by a diverse team of clinical trial research nurses, physicians, behavioral scientists, and health educators.

Program Implementation

Before the implementation of the program, a series of in-service presentations were conducted for clinical research nurses and clinical trial principal investigators to introduce them to the patient navigation initiative and to garner their support. A program implementation protocol, which included a PN work plan and a flowchart for PNs’ interaction with patients, was developed. Potential participants were identified in one of two ways: through reviews of electronic clinic schedules and patient charts by the IMPaCT nurse, or through referrals from nurses, physicians, and social workers.

Each day, the IMPaCT program manager received a list of patients scheduled to see a physician in oncology outpatient clinics within the next 3 days. The IMPaCT program manager identified the African American patients with cancer, reviewed the reasons for their clinic visit, and developed a list of patients likely to be offered participation in a clinical trial. This list of potential African American clinical trial participants was provided to each PN.

Before the scheduled clinic appointment, a PN contacted the patient by phone and asked if he or she would like to receive patient navigation support if offered to participate in clinical trial. If the patient agreed to patient navigation support, the PN met the patient in the waiting room before the scheduled clinic appointment to talk about clinical trials, obtain informed consent for participation in IMPaCT, and administer a needs assessment to determine what barriers the patient may have. If the time was not appropriate for needs assessment, the PN made arrangements with the patient to conduct the needs assessment at a different time.

Services Offered by PNs

Clinic-based education about clinical trials

PNs educated African American patients with cancer about clinical trials when they first contacted them by phone, when they met in the waiting room before the patient’s clinic appointment, or when the research nurse invited the PN to join her as she discussed with the patient the clinical trial and the consent form. When explaining clinical trials to potential participants, PNs used NCI Publication No. 97-2706: What Are Clinical Trials All About?⁴⁶

Support for patients who enroll in clinical trials

Potential clinical trial participants who opted to receive patient navigation support completed a needs assessment. On the basis of the needs assessment, the PN determined if there were barriers to the patient’s participation in a clinical trial. Together with the patient, the PN developed a plan to address these barriers. The IMPaCT program manager helped PNs identify community resources that can be used for support of clinical trial participants. Services provided by PNs included assistance with transportation and lodging, reminder calls for appointments, referrals to social workers when appropriate, and linking the patient with social and community services and resources. In addition, PNs provided culturally appropriate peer support: they accompanied patients to clinic visits, especially when clinical trials were discussed, called patients to offer social and emotional support, encouraged patients to report symptoms/concerns to their physicians, and so forth.

PNs communicated regularly with clinic nurses and social workers to receive status updates and to monitor the trial participants’ attendance of clinic visits. PNs supplemented the institutional services provided to clinical trial participants without interfering with the clinical trial protocol. For example, they contacted participants before clinic visits to remind them of their appointments and after the visits to obtain feedback and ensure satisfaction with the visit. The preappointment contact allowed the PNs to assist with barriers that may have prevented the patient from making the visit, and the postvisit contact allowed the PNs to resolve issues that may have influenced the patient’s return visit or compliance with treatment protocols. Ultimately, the goal of the PNs was to help patients navigate the health care system, take advantage of available resources, and follow the clinical trial protocol.

Data Management and Analysis

Participant data were entered by PNs into a database designed to help manage their patient caseload and document their activities. The database was compliant with Health Insurance Portability and Accountability Act guidelines and Institutional Review Board guidelines. To monitor progress, monthly summary reports of deidentified aggregate data were shared with members of the investigative team. As the IMPaCT project expanded and gained acceptability among clinicians, PNs were invited to attend the weekly meetings of the Comprehensive Cancer Center Clinical Trials Monitoring Committee and report on the patients they served.

Demographic information and clinical characteristics were summarized using descriptive summary statistics by IMPaCT participation status. Characteristics were compared using a χ² test. For the trial completion rate, percentages with corresponding 95% CIs were calculated. The likelihood of trial completion was described using odds ratios and corresponding 95% CIs.

RESULTS

Between 2007 and 2014, a total of 454 interventional cancer trials were available in the study institution. Of them, 72 were phase I, 39 were phase I/II, 167 were phase II, eight were phase II/III, 149 were phase III, and 16 were pilot trials. The majority of available trial protocols were in cancers of the brain and nervous system (n = 70), female breast (n = 61), lung (n = 34), lymphoid leukemia (n = 31), myeloid and monocytic leukemia (n = 30), ovary (n = 21), skin melanoma (n = 21), non-Hodgkin lymphoma (n = 18), prostate (n = 16), kidney (n = 13), and lip, oral cavity, and pharynx (n = 13).

During the study period (2007 to 2014), a total of 432 African American patients with cancer were referred to IMPaCT. Of them, 272 (63%) participated in the program. Patients who did not participate in IMPaCT (n = 160, 37%) were not enrolled for the following reasons: 42% were ineligible for a clinical trial, 22% declined to participate in a clinical trial, 22% did not need assistance, 6% could not participate for other reasons (clinical trial was full, insurance did not cover trial, moved out of state, and so on), 5% were lost to contact, and 3% declined to participate in IMPaCT.

The majority of patients in both groups were 40 to 64 years old (67% IMPaCT, 71% non-IMPaCT), female (71% IMPaCT, 64% non-IMPaCT), and lived with someone (42% IMPaCT, 40% non-IMPACT). In terms of cancer type, they had breast cancer (31% IMPaCT, 25% non-IMPACT), lung cancer (13% IMPaCT, 10% non-IMPaCT), cervical cancer (9% IMPaCT, 14% non-IMPaCT), lymphoma (9% IMPaCT, 5% non-IMPaCT), head and neck cancer (7% IMPaCT, 8% non-IMPaCT), leukemia (5% IMPaCT, 4% non-IMPACT), and other cancers (26% IMPaCT, 34% non-IMPaCT). None of the differences between the two groups, however, were statistically significant (P > .05; Appendix Table A1, online only).

Among patients enrolled in a clinical trial, available socioeconomic information (educational attainment, income, health insurance status, and employment status) between patients who accepted patient navigation and those who did not was compared. Conclusions may be limited because of the proportion of missing information (ranging from 32% for health insurance status to 57% for income). However, the available data showed no difference between the two groups in terms of these socioeconomic variables.

During the course of the project, PNs provided a range of services in response to identified barriers (Table 1). Social and emotional support services included appointment reminders and confirmation of plans, escort and provision of Guest Services, assistance with paperwork, calls to identify community resources, linking of patients with programs such as the American Cancer Society’s Look Good Feel Better and Reach to Recovery, calls after a clinic visit to inquire about outcome, provision of emotional support, and counseling. Transportation services included making arrangements for or providing assistance with purchase of bus ticket, cab fare, airfare, parking, gas voucher, private transportation, Medicaid NET, Special Needs/Disability, and meal vouchers. Lodging services included making arrangements for, or providing assistance with stays at UAB Townhouse, American Cancer Society Hope Lodge, and area hotels. Insurance services included provision of Charity Care.

Table 1.

No. of IMPaCT Services Provided by Clinic or Disease Site

Clinic or Disease Site	Transportation	Lodging	Insurance	Social and Emotional Support
Bone marrow transplantation	4	1	0	43
GI/genitourinary	62	3	0	348
Gynecologic oncology	137	15	4	913
Head and neck	27	7	0	223
Hematology-oncology	168	13	2	958
Invasive ductal breast carcinoma	394	10	1	1,463
Cooper Green Mercy Hospital	28	0	0	86
Lung	25	12	0	415
Neuro-oncology	25	5	1	191
Other	0	0	0	21
Radiation oncology	45	2	0	344
Solid tumors	12	3	0	147
Total	927	71	8	5,152

Abbreviation: IMPaCT, Increasing Minority Participation in Clinical Trials

As listed in Table 1, a total of 5,152 social support services, 927 transportation services, 71 lodging services, and eight insurance services were provided by PNs to program participants. On average, each PN had an active case load of 33 patients at any given time.

Over the duration of IMPaCT, referrals of African American patients from oncology clinics to cancer clinical trials increased more than three-fold, from 5.5% of all patients with cancer in 2007 to 16.6% of all patients with cancer in 2014 (Fig 1). Participation of African American patients in therapeutic clinical trials at the UAB Comprehensive Cancer Center increased from 9% to 16% between 2007 and 2014.

FIG 1.

Referral rate to IMPaCT (Increasing Minority Participation in Clinical Trials) of African American patients with cancer.

Of the African American patients referred to IMPaCT who were eligible for a clinical trial (n = 378), 304 (80.4%) enrolled in a trial and 272 (72%) enrolled in IMPaCT (ie, consented to receive patient navigation support); clinical trial enrollment and IMPaCT enrollment rates by year are presented in Figure 2.

FIG 2.

Clinical trial and IMPaCT (Increasing Minority Participation in Clinical Trials) enrollment rates by year.

Of patients receiving patient navigation support, 74.5% completed the trial, compared with 37.5% of those who chose not to receive patient navigation support. The corresponding 95% CIs for the trial completion rates were 69.2% to 79.8% and 23.8% to 51.2%, respectively. IMPaCT-enrolled participants were 4.88 times (95% CI, 2.56 to 9.31) more likely to complete the clinical trial (P < .001; χ² test). In both groups, patients withdrawn by physician because of toxicity, disease progression, or death were considered to have completed the trial. The reasons for not completing a clinical trial were not available for either group.

DISCUSSION

Patient navigation, which has been used extensively to increase access to medical care, has not been evaluated systematically as a strategy for increasing the participation of minorities in clinical trials,⁴⁰ although interest in this field has been increasing and initial successes have been reported.^47-52 IMPaCT used patient navigation to enhance the recruitment of African Americans for, and their retention in, therapeutic cancer clinical trials in an NCI-designated comprehensive cancer center. Between 2007 and 2014, the study reported a 72% average enrollment in therapeutic clinical trials of referred African American patients with cancer. These enrollment rates are consistent with the ones reported by other patient navigation studies.^52-54 To our knowledge, however, this is the first lay patient navigation study to report increased minority participation in clinical trials (from 9% to 16%), as well as the first study in which the primary outcome measure is clinical trial completion rate.⁴⁰ Importantly, the IMPaCT study doubled the clinical trial completion rate of African American trial participants (74.5% v 37.5%) by identifying and addressing potential barriers to compliance with treatment protocols and clinical trial completion.

The project was successful because it addressed both system- and patient-level barriers to participation of African Americans in cancer clinical trials. On the system level, the patient navigation model was integrated successfully into the clinical trial setting, and the PNs became part of the Comprehensive Cancer Center clinical trials team. As barriers to enrollment and retention were addressed appropriately by the trained PNs, the acceptability and the credibility of the program with clinicians and staff increased, resulting in improved referral rates by investigators and research nurses (from 5.5% to 16.6% between 2007 and 2014). This is one of the major reasons for the success of the program. As documented in the literature, patient enrollment in a clinical trial requires a gateway offer to participate, most frequently coming from a physician.^55-57 Only 20% of those eligible for a clinical trial are explicitly offered the opportunity to enroll,⁵⁸ and physicians tend to be selective about offering participation on the basis their perception of a patient’s likelihood of adhering to the protocol.⁵⁹ Addressing these systemic barriers to the participation of minorities in clinical trials is critical and can be accomplished by integration of trained lay PNs into the clinical trial teams.

On the patient level, the project addressed individual barriers to clinical trial participation. This resulted in a clinical trial completion rate among African American patients that was comparable to the completion rate of white clinical trial participants.

Minority ethnicity and low socioeconomic status have been reported previously to be factors associated with clinical trial enrollment.^4,5,8,60-62 The IMPaCT project demonstrated that patient navigation can address barriers to clinical trial participation among African American patients with cancer and significantly improve their retention in clinical trials. As the project made an impact on a systems level, PNs became legitimate members of the clinical and research teams. Over the project’s duration, both referral and enrollment rates in cancer clinical trials increased. Because of the study limitations of IMPaCT (nonrandomized design and lack of control group), it is not possible to estimate precisely what proportion of this increase is directly attributable to the patient navigation model. Further studies may consider pre/post or interrupted time series comparison of patient cohorts to determine if implementation of a patient navigation protocol results in higher rates of clinical trial enrollment and completion.

In addition, future research should address the cost-effectiveness of the patient navigation model for recruitment of racial/ethnic minorities for, and their retention in, clinical trials.^40,63 Although the economic impact of patient navigation for diagnostic resolution has been discussed,⁶⁴ such analyses should be performed in the context of clinical trial participation.⁶⁵ The NIH Revitalization Act of 1993 mandated the inclusion of women and minorities in clinical trials. Recruitment of minority participants involves additional costs for implementation of recruitment strategies, methods, and tools that have been proven effective.⁶⁶ If cancer clinical trials must include underrepresented minority populations, recruitment plans need to include adequate resources for their enrollment and retention.

Appendix

Table A1.

Characteristics of Patients Referred to IMPACT (N = 432)

Characteristic	Enrolled in IMPaCT	Did Not Enroll in IMPaCT
Total, No. (%)	272 (63)	160 (37)
Age group, years
< 40	12.13	7.50
40-64	66.91	71.25
≥ 65	20.96	21.25
Sex
Female	70.96	63.75
Male	29.04	36.25
Marital status
Lives with someone	42.23	40.31
Single	31.92	40.31
Separated or divorced	15.00	13.18
Widowed	8.85	6.20
Cancer type
Breast	31.25	25.00
Lung	12.87	10.00
Cervical	8.82	14.38
Lymphoma	8.82	5.00
Head and neck	6.62	7.50
Leukemia	5.15	3.75
Other	26.47	34.38

NOTE. Data are presented as % unless indicated otherwise.

Abbreviation: IMPaCT, Increasing Minority Participation in Clinical Trials.

AUTHOR CONTRIBUTIONS

Conception and design: Mona N. Fouad, Andres Forero, Michelle Y. Martin, Edward E. Partridge, Selwyn M. Vickers

Provision of study materials or patients: Andres Forero

Collection and assembly of data: Aras Acemgil, Andres Forero, Nedra Lisovicz

Data analysis and interpretation: Mona N. Fouad, Aras Acemgil, Sejong Bae, Andres Forero, Michelle Y. Martin, Gabriela R. Oates, Edward E. Partridge

Manuscript writing: All authors

Final approval of manuscript: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Patient Navigation As a Model to Increase Participation of African Americans in Cancer Clinical Trials

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.

Mona Fouad

No relationship to disclose

Aras Acemgil

No relationship to disclose

Sejong Bae

No relationship to disclose

Andres Forero

Research Funding: Pfizer (Inst), Novartis (Inst), Seattle Genetics (Inst), MedImmune (Inst), TRACON Pharma (Inst), Genentech (Inst), Abbvie (Inst), Daiichi Sankyo (Inst), Juno Therapeutics (Inst)

Nedra Lisovicz

No relationship to disclose

Michelle Y. Martin

No relationship to disclose

Gabriela R. Oates

No relationship to disclose

Edward E. Partridge

No relationship to disclose

Selwyn M. Vickers

No relationship to disclose