SYNTHESIS OF EVIDENCE SUBSTANTIATING POTENTIAL INTERVENTIONS

Barriers to Clinical Trial Enrollment in Racial and Ethnic Minority
Patients With Cancer

Lauren M. Hamel, PhD, Louis A. Penner, PhD, Terrance L. Albrecht, PhD, Elisabeth Heath,
MD, Clement K. Gwede, PhD, RN, and Susan Eggly, PhD
Department of Oncology (LMH, LAP, TLA, EH, SE) and Population Studies Disparities Research
Program (LMH, LAP, TLA, SE), Wayne State University, Karmanos Cancer Institute, Detroit,
Michigan, and the Department of Oncologic Sciences/Population Sciences (CKG), H. Lee Moffitt
Cancer Center & Research Institute, Tampa, Florida

Abstract

Background—Clinical trials that study cancer are essential for testing the safety and
effectiveness of promising treatments, but most people with cancer never enroll in a clinical trial

— a challenge exemplified in racial and ethnic minorities. Underenrollment of racial and ethnic

minorities reduces the generalizability of research findings and represents a disparity in access to

high-quality health care.

Methods—Using a multilevel model as a framework, potential barriers to trial enrollment of
racial and ethnic minorities were identified at system, individual, and interpersonal levels. Exactly

how each level directly or indirectly contributes to doctor–patient communication was also

reviewed. Selected examples of implemented interventions are included to help address these

barriers. We then propose our own evidence-based intervention addressing barriers at the

individual and interpersonal levels.

Results—Barriers to enrolling a diverse population of patients in clinical trials are complex and
multilevel. Interventions focused at each level have been relatively successful, but multilevel

interventions have the greatest potential for success.

Conclusion—To increase the enrollment of racial and ethnic minorities in clinical trials, future
interventions should address barriers at multiple levels.

Introduction

Clinical trials focused on cancer research are essential for testing the safety and effectiveness

of potential treatments and translating new knowledge into tangible benefits for patients;

they also represent options for novel therapy for cancer.1,2 However, approximately 2% to

3% of all patients with cancer ever enroll in a trial. 3,4 Estimates of the number of trials that

fail to meet scientific objectives because of insufficient accrual rates range from 22% to

Address correspondence to Lauren M. Hamel, PhD, Wayne State University, Karmanos Cancer Institute, Department of Oncology,
Population Studies Disparities Research Program, 4100 John R Street, MM03CB, Detroit, MI 48201. [email protected].

No significant relationships exist between the authors and the companies/organizations whose products or services may be referenced
in this article.

HHS Public Access
Author manuscript
Cancer Control. Author manuscript; available in PMC 2016 December 01.

Published in final edited form as:
Cancer Control. 2016 October ; 23(4): 327–337.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

50%.5,6 Low accrual rates jeopardize the ability of researchers to assess the safety and

effectiveness of new approaches to cancer care, wastes resources, precludes follow-up

studies, and reduces the ability of our clinical research system to translate research into

evidence-based practice.6–8

Underenrollment is an even greater challenge among racial and ethnic minorities —

particularly African Americans — despite a requirement by the National Institutes of Health

that members of minority populations be represented in clinical research.1,3,4,9–11 A

systematic review compared the proportion of underrepresented minority participants in

phase 3 cancer treatment and prevention clinical trials conducted between the periods 1990

to 2000 and 2001 to 2010.12 In the treatment studies conducted between 2001 and 2010 that

reported race/ethnicity, the reviewers found that 82.9% of participants were white, 6.2%

were African American, 3.3% were Asian, 2.2% were Hispanic, and 0.1% were Native

American.12 This is in contrast to studies conducted between 1990 and 2000, in which 89%

of participants were white, 10.5% were African American, 0.4% were Hispanic, and 0.04%

were Asian.12 In other words, even though the proportion of white participants decreased,

whites continued to comprise a large majority of participants in cancer treatment trials, and

the proportion of African American participants decreased between the periods 1990 to 2000

and 2001 to 2010. However, during those same periods, the proportion of African American

participants in clinical prevention trials increased (5.5% from 1990–2000, 11.6% from

2001–2010).12

Although several patient populations are underrepresented in clinical trials, including elderly

patients (≥65 years), residents of rural areas, and those with low socioeconomic status, the

current review focuses on racial and ethnic minority underenrollment for several

reasons.4,10,13–16 Racial and ethnic minorities — particularly African Americans — bear the

greatest cancer burden in the United States, so they should be adequately represented in

cancer research.17–19 Under-representation of racial and ethnic minorities also limits the

generalizability of research findings.20–22 The Institute of Medicine has recommended that

every individual with cancer have access to high-quality clinical trials, so we believe that

under-representation of racial and ethnic minorities represents a disparity in health care.2

Underenrollment of racial and ethnic minorities in clinical trials may therefore contribute to

preventable disparities in treatment outcomes and survival.1,17,23,24

Purpose

The purpose of this paper is to identify and describe potential barriers to the enrollment of

racial and ethnic minorities in clinical trials at the system, individual (health care

professional, patient, and family), and interpersonal levels (eg, doctor–patient relationship).

The paper also describes selected examples of evidence-based interventions already

implemented to address some of these barriers at the patient-, physician-, and doctor–patient

interpersonal communication levels. We take this approach because multilevel interventions,

as compared with single-level interventions, may have the greatest potential to achieve

substantial and sustained change and to produce additive — and possibly multiplicative —

effects.25,26

Hamel et al. Page 2

Cancer Control. Author manuscript; available in PMC 2016 December 01.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

We propose a multilevel model derived from general systems theory, which posits that

organizations are comprised of discrete but related levels (Fig 1).27 Our multilevel model

demonstrates how system- and individual-level barriers contribute to interpersonal-level

barriers to the enrollment of racial and ethnic minorities in clinical trials.

System-Level Barriers

Barriers at the level of health care systems and hospitals include the limited number of trials

nationally and regionally available; hospital infrastructures that lack the resources to support

trials; financial costs to hospitals and patients; restrictive study designs and eligibility

criteria; and lack of community engagement. Many of these barriers have a disproportionate

effect on minority enrollment because members of minority populations are more likely to

receive care at under-resourced hospital systems where few clinical trials are available, to be

underinsured, and to present with comorbidities that make them ineligible for trials that may

be available.28–30

Available clinical trials supported by a well-resourced and functional infrastructure are

critical to the enrollment of minority patients.31 For example, this infrastructure includes

staff dedicated to enrolling, managing, and tracking participants; data collection and

management capabilities; and an efficient Institutional Review Board. Without a functional

infrastructure, efforts at increasing the enrollment of minorities in other levels are less likely

to be successful.31–33 The financial costs of enrollment to institutions and individual patients

affect all patient populations, but they have a disproportionate impact on minorities

compared with nonminorities. Minorities are more likely to be underinsured, to seek care at

under-resourced hospitals, and to have concerns about the cost of participating in a clinical

trial.29,34 In addition, lack of or inadequate health insurance acts as a barrier to enrollment in

clinical trials for several under-represented populations, including racial and ethnic

minorities.29,35

As a way to enhance community engagement, the infrastructure of clinical trials should exist

within a medical institution that has an established, trusted relationship with the members of

the community it serves.36 Building trust within a community requires that an institution

conduct its research in an open and honest manner that involves a complete and accurate

description of the known and potential differences in the risks and benefits for different

racial and ethnic populations. Ways and means to facilitate and encourage open discussions

into research ethics, including the past abuse of minority patients in research, are needed, as

are efforts to create and sustain partnerships with the community to share ownership of the

research.37–40

In response to system-level barriers, several national, regional, and consortia efforts have

implemented programs to increase enrollment among racial and ethnic minority

populations.8,41,42 For example, the Minority-Based Community Clinical Oncology

Program of the National Cancer Institute was designed to increase minority participation in

cancer clinical trials by developing outreach efforts at health care institutions that serve large

numbers of minority patients with cancer.32 Mc-Caskill-Stevens et al32 reported that the

success of this program was largely due to system- and hospital-level factors, such as

Hamel et al. Page 3

Cancer Control. Author manuscript; available in PMC 2016 December 01.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

opening trials that matched the clinical characteristics of patient populations and developing

relationships with local physicians and cancer advocacy groups that increased their

willingness to enroll or refer minority patients to clinical trials. However, their success was

limited by lack of funding for minority outreach activities, lack of system support to provide

staff and mentoring for minority investigators, and lack of funding to provide protocol-

related drugs and related services for uninsured patients.32

The Community Oncology Research Program of the National Cancer Institute is a national

network of health care professionals, investigators, and organizations that conducts cancer-

related research across the United States.43 The program focuses on determining the reasons

for the existence of racial and ethnic disparities in clinical trial participation, as well as ways

to increase the participation of racial and ethnic minorities in clinical trials through

numerous interventions.43 Because its inauguration was in 2014, evaluation data have not

yet been published.

In recent years NRG Oncology convened a workshop on the challenges and opportunities of

clinical trial enrollment in an attempt to address the issue of minority underenrollment in

clinical trials.44 Experts in oncology, including members of the Minority-Based Community

Clinical Oncology Program of the National Cancer Institute, and a large patient advocacy

group were in attendance.44 Ten themes emerged regarding future plans and interventions to

increase minority enrollment in clinical trials; of these, 8 themes focused on the system

level44:

1. Target recruitment and emphasize the personalized nature of trials as a
way to improve minority enrollment to specialized studies

2. Define study populations at the molecular level rather than by traditional
and less-precise eligibility criteria

3. Make the delivery of cancer care research available at the community and
academic levels so studies are designed to better improve health outcomes

4. Improve local infrastructure by providing translation and navigation
services and research-friendly electronic medical records

5. Expand and standardize the collection of demographics and real-time
information about accruals for all patient populations

6. Strengthen the infrastructure of information technology

7. Improve recruitment, training, and mentorship of young investigators,
particularly those of racial and ethnic minority backgrounds and those

with an interest in cancer disparities

8. Continue and improve budgetary support from the US Congress for all
clinical cancer research within the National Cancer Institute

Hamel et al. Page 4

Cancer Control. Author manuscript; available in PMC 2016 December 01.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

Individual-Level Barriers

Our multilevel model emphasizes individual factors that may present a barrier to the

enrollment of racial and ethnic minorities in clinical trials. The term individuals is meant to
refer to persons directly involved in the clinical trial enrollment process, including health

care professionals, patients, and their family members. Barriers on a system level can

influence how individuals perceive the enrollment of minority patients in clinical trials and

how they communicate about trials. For example, an investigation of physician perceptions

of clinical trials and trial support available at their institution revealed that most physicians

held a favorable attitude about clinical trials, both as a source of high-quality patient care

and as a professional benefit to themselves.33 However, approximately two-thirds of

practicing oncologists perceived a lack of needed infrastructure to conduct trials.33 This

included a low number of trials open and lack of nonphysician staff to support patient

participation in clinical trials.33 Similarly, the health care system can also influence patient

concerns about the cost of participating in trials and, in turn, their decisions to enroll or not

enroll in a clinical trial.34 Thus, these 2 levels are linked in terms of whether and how health

care professionals and minority patients discuss clinical trials and whether minority patients

agree to enroll.

Health Care Professionals

Health care professionals play a critical role in facilitating or inhibiting the enrollment of a

diverse population of patients into clinical trials.33,45,46 Several factors are related to the

reluctance of health care professionals to enroll their patients in trials, regardless of whether

or not they are members of minority groups themselves.

A major barrier is the lack of awareness of available trials: Health care professionals must be

made aware of national, regional, and local trials currently taking patients and the eligibility

criteria for each of those trials to discuss such enrollment with their patients.45,46 In

addition, the attitudes of health care professionals about trials and discussing such trials with

patients and their families present a major barrier to trial enrollment, and they likely affect

the quality of communication during discussions of clinical trials.33,45 For example, health

care professionals may not fully agree with or understand the scientific value of trials in

general or the details of specific trials; they may feel they do not have adequate system

support; or they may have concerns about practical issues such as strict protocol designs,

patient inconvenience, and added work for the health care staff.33,45 Some clinicians find it

difficult to reconcile the roles of physician and researcher, or are concerned about unduly

influencing patient decisions about enrolling in a clinical trial.47

Some attitudinal factors are specific to the enrollment of minority patients, such as the

concern about harming the therapeutic relationship. For example, a focus-group study

involving community physicians found that some physicians were hesitant to discuss clinical

trials with their African American patients because of their perceptions that African

Americans are often mistrustful of physicians and medical institutions.48

An additional attitudinal factor that may deter health care professionals from discussing

trials with minority patients is implicit bias against members of minority groups. Prior

Hamel et al. Page 5

Cancer Control. Author manuscript; available in PMC 2016 December 01.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

research has shown that implicit bias (comprised of conscious and unconscious biases)

among physicians toward members of minority groups has an impact on clinical interactions

with minority patients.49–52 We are unaware of studies that specifically investigate whether

the implicit bias of health care professionals is a barrier to the enrollment of minorities in

clinical trials, but research in other contexts suggests that some physicians may have implicit

negative attitudes about African Americans that may lead them to believe they will be poor

candidates for clinical trials.53–56 For example, research has shown that primary care

physicians tend to trust minority patients less than they trust white patients.54 Research has

also shown that physicians high in implicit bias may view their minority patients as less

likely to comply with recommended treatment,57 and that physicians’ diagnoses and

treatment of African Americans can be influenced by racial stereotypes.55 This body of

research suggests that physicians may limit offers for trial enrollment to those patients they

perceive as good study candidates so that the studies will be conducted in a timely and

efficient manner.58 This may mean physicians are less likely to offer a clinical trial to

patients who are racial and ethnic minorities than to white patients, and research has shown

this to be the case.46,59

Recommendations have been made for evidence-based approaches to reduce the impact of

health care professional bias in clinical interactions.60 One approach is to train health care

professionals in the use of high-quality patient-centered communication, which requires the

development of a positive interpersonal relationship.61 Patient-centered communication

focuses on patient needs and perspectives, and it does not ignore the racial and ethnic

background of the patient, because doing so denies an important part of an individual.61

Results of the NRG Oncology workshop identified the commitment of physicians and study

investigators to the enrollment of minorities into clinical trials as critical to the future of

clinical cancer research.44 They also emphasized that physicians must be culturally sensitive

and aware of the impact appropriate communication and patient trust have on minority

enrollment in clinical trials.44

Several communication curricula are available, but they include little training on discussing

clinical trials with patients or with specific patient populations.62 One exception is an

Australian study, which reported on a 1-day intensive course for physicians to improve

aspects of their communication related to discussing clinical trials with patients.45 Results

showed that physicians improved in some shared decision-making behaviors and were more

likely to describe some of the key clinical or ethical aspects of the trial.63 However, we are

unaware of any specific training curriculum that emphasizes ways to communicate about

clinical trials with minority patients.

Patients and Families

Some major barriers to enrolling racial and ethnic minority patients in clinical trials are that

patients are often not aware they are eligible for an existing trial; they are often underinsured

and cannot afford the extra expenses related to trial participation; and they may not meet the

clinical criteria due to comorbid conditions or age restrictions.29,30,64 Similar to health care

professionals, patients hold attitudes and beliefs that may also affect their willingness to

participate and the way they communicate during clinic visits in which trials are

Hamel et al. Page 6

Cancer Control. Author manuscript; available in PMC 2016 December 01.

A
u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t
A

u
th

o
r M

a
n
u
scrip

t

discussed.65,66 Despite this, research shows that, overall, minorities are as likely as whites to

consent if they are offered a trial.30,59,67,68 For example, an assessment of more than 4,000

racially diverse patients with cancer found no significant association between race and

ethnicity and refusal to participate in a clinical trial or lack of desire to participate in

research.30 Similarly, Katz et al69 implemented the Tuskegee Legacy Project to address and

understand if and how the Tuskegee syphilis experiment impacted the trial recruitment and

retention of African Americans into biomedical studies. The researchers found no difference

in self-reported willingness among African Americans to participate in biomedical

research.69 A systematic review of the available research on the factors that influence

participation in clinical trials among African Americans concluded that the strongest

inhibitors of participation were low levels of knowledge and lack of awareness of clinical

trials.66 The most important facilitators of participation were social support and

recommendations from physicians, family members, and friends.66

Several patient-focused interventions designed to improve knowledge of, attitudes about,

and participation in clinical trials among racial and ethnic minority patients has been

developed and tested.70,71 A multicenter, randomized trial of a web-based, interactive,

educational tool was designed to increase knowledge, decrease attitudinal barriers, and

improve preparation for making decisions about clinical trial enrollment among minority

patients with cancer.70 At baseline, the level of knowledge about clinical trials, attitudinal

barriers to participation, and preparation for decision-making among the participants were

assessed.70 Prior to their initial visit with an oncologist, patients in the intervention arm

watched a set of brief, individually tailored videos addressing knowledge and attitudinal

barriers to clinical trial participation, whereas those assigned to the control arm received

text-based general information about clinical trials from the National Cancer Institute. The

study results showed that, compared with baseline levels, participants in the intervention and

control arms both showed improved knowledge (control arm: mean [standard deviation

{SD}] = 2.5 [3.1]; intervention arm: mean [SD] = 3.2 [3.1]; P < .001) and decreased attitudinal barriers (control arm: mean [SD] = −0.2 [0.4]; intervention arm: mean [SD] = −0.3 [0.5]; P < .001).70 However, patients in the intervention arm had a significantly greater increase in knowledge and decrease in attitudinal barriers compared with patients in the control arm (P < .001).71 All patients had a significant increase in their preparedness to consider participation in clinical trials (control arm: mean [SD] = 3.4 [13.5]; intervention arm: mean [SD] = 4.7 [12.8]; P < .001).70 Another intervention specifically addressed the negative perceptions and attitudes of patients about clinical trials by assessing the effectiveness of a multimedia, psychoeducational intervention compared with printed educational materials.71 The intervention was designed to improve patient attitudes and knowledge of clinical trials, their ability to effectively make decisions, their receptivity to receiving more information, and their willingness to participate in clinical trials.71 The researchers observed that study participants who received the intervention had more positive attitudes toward clinical trials (mean [SD] = 0.21 [0.01]; P = . 02), and, although the 2 interventions were not developed and tested with minority patients, the study results showed no significant difference between intervention by patient demographics.71 Thus, the intervention could potentially be adapted to specifically benefit racial and ethnic minority patient populations. Hamel et al. Page 7 Cancer Control. Author manuscript; available in PMC 2016 December 01. A u th o r M a n u scrip t A u th o r M a n u scrip t A u th o r M a n u scrip t A u th o r M a n u scrip t Although racial and ethnic minority patients are overall as likely as white patients to agree to enroll in a clinical trial, reasons for declining may differ by racial and ethnic background.30,59,66–68 Minority patients — particularly African Americans — may hold negative race-related attitudes and beliefs that could directly and indirectly influence interactions with their health care professional regarding trials and decisions about enrollment.38,40,66,72 These attitudes, which have been derived in large part from issues of racism, uninformed consent, and poor health care for minorities in the United States, include greater mistrust in medical institutions and health care professionals and a greater sense of having been the target of discrimination.20,23,29,38,69,72–78 A systematic review of the research assessing the multiple barriers to minority enrollment in clinical trials found that mistrust in research and the medical system was the most common barrier to patients’ willingness to participate in a clinical trial.29 The attitudes of African American patients toward medical research were assessed using focus groups of African American adult patients from a single urban public hospital.72 Results showed that most were in favor of medical research as long as they were not treated as “guinea pigs.”72 The authors also reported that many of the participants had a limited understanding of the informed consent process, and many patients assumed the consent form was protecting hospitals and doctors from any legal responsibility, not to protect patients.72 Another study that assessed and compared the racial differences in factors that influence patient willingness to enroll in clinical research studies found that African Americans were less willing to participate in medical research if they attributed a high importance to the race of the physician when seeking routine medical care and believed that minorities bear most of the risks of medical research.40 Interventions designed to increase the recruitment of minorities in clinical trials are generally focused on communities, rather than on individuals.32,79 One such intervention targeted minority recruitment using strategic planning, which included meetings and conferences with key stakeholders and minority organizations as a means to increase minority enrollment.79 The authors examined institutions that employed these strategies and those that did not, and they found a significant increase in the rate of minority accrual in institutions that implemented strategic planning compared with those that did not.79 This effort suggests that the mistrust of the medical system seen among minority patients might be addressed with culturally sensitive and open community engagement. Other identified patient-level barriers to the enrollment of racial and ethnic minorities in clinical trials include lack of transportation, inadequate insurance, lack of childcare, and poor access to health care.42,66 The NRG Oncology workshop also identified direct-to- patient communication and advertising as critical for the future of minority enrollment in clinical trials, emphasizing that this should be done with collaboration with key stakeholders such as community groups, survivor advocacy groups, and religious institutios.44 The intersection between the impact of patient attitudes and access to trial participation illustrates the complex and multilevel nature of this issue and the need for more comprehensive interventional strategies. Hamel et al. Page 8 Cancer Control. Author manuscript; available in PMC 2016 December 01. A u th o r M a n u scrip t A u th o r M a n u scrip t A u th o r M a n u scrip t A u th o r M a n u scrip t Interpersonal-Level Barriers Our model suggests that system- and individual-level barriers to minority enrollment in clinical trials affect whether and how trials are discussed during doctor–patient clinical interactions. This suggestion is based on our work and that of others showing that patient and physician attitudes affect the quality of communication during clinical interactions with African Americans.80–88 Our model also partly explains why communication during these interactions is often of a lower quality and may affect racial and ethnic minorities’ decisions about …