記住我
At the turn of the 21st century, the Institute of Medicine released the article Crossing the Quality Chasm, which emphasized the need for an effective, equitable, and patient-centered healthcare delivery system (Agency for Healthcare Research and Quality [AHRQ], 2019). The need for improved healthcare derived from the long-standing systemic health inequities experienced by racial and ethnic minority groups. More than two decades later, amidst a global pandemic, the United States continues to seek care solutions as minority populations persistently face devastating health disparities. To provide effective and equitable care, improved healthcare research is necessary (Beattie et al., 2012).
Healthcare research is a general term that includes a variety of research methods that ultimately develop or provide knowledge regarding disease, risk factors, outcomes of treatment, public health interventions, functional abilities, patterns of care, and healthcare usage (Beattie et al., 2012). Precision health research has been touted as a novel and person-centered method of healthcare research. Precision health approaches disease treatment and prevention by accounting for the individual’s genetic variability, lifestyle, and social determinants for providers to care for patients most effectively and equitably.
Although research is the necessary foundation for healthcare advancement and understanding, significant challenges exist with recruiting and engaging underrepresented populations in healthcare research. The National Institutes of Health recognize African Americans (AAs), American Indians (AIs), Alaskan Natives, Hispanics, Native Hawaiians, and other Pacific Islanders as underrepresented populations in research studies (National Institutes of Health, n.d.). In 2017, the U.S. Food and Drug Administration reported 81% of clinical trial participants as White, 14% as AA, 2.7% as Asian, and the remaining 2.3% as Hispanic, Pacific Islander, AI, Alaskan Native, or Native Hawaiian (U.S. Food and Drug Administration, 2017). Although minority populations are underrepresented in clinical research, they face the greatest health disparities (AHRQ, 2019).
The 2018 National Healthcare Quality and Disparities Report documented that minority populations receive poorer quality of care and face greater barriers in accessing care compared to White populations (AHRQ, 2019). In addition, AAs, AIs, and Hispanics have greater rates of preventable hospitalizations and higher mortality rates compared to Whites (AHRQ, 2019). Incidence of specific diseases, cancers, and reactions to medications and treatments differ between races and ethnicities. For example, AA men have a greater incidence of prostate cancer compared to White men. AA women have the same incidence of breast cancer as White women, but they have higher mortality rates (Reifenstein, 2018). Amid the ongoing pandemic, AAs account for one third of COVID-19 cases and are twice as likely to die from the virus (Garg et al., 2020). Hispanics and AAs have the greatest prevalence of diabetes, adolescent obesity, and asthma compared to Whites (National Center for Health Statistics, 2019). Similarly, almost half of all AA adults have some form of cardiovascular disease compared to about one third of White adults (Reifenstein, 2018). It is necessary to recruit and engage diverse populations to create equitable health systems. However, engaging and recruiting participants to accurately represent the diversity of the population is a challenging process (Erves et al., 2017).
Participation barriers in research and variables that can influence an individual’s willingness to participate have been identified in the literature. Barriers identified include the participant’s level of trust, access to research information, fear of the unknown or adverse effects, inconvenience, and reputation of researchers and research institutions.
Level of TrustThe concept of trust is examined in healthcare literature, specifically because of relationship dynamics between healthcare professionals and patients (Hall et al., 2001). Trust is defined as the degree to which the patient relies and depends on and is confident about the provider (Hall et al., 2001). Trust is present in situations of risk, uncertainty, vulnerability, or unequal status where there is a level of dependence on another individual creating a relationship of vulnerability (Hall et al., 2001). Variables identified throughout literature that act as barriers to participant trust in clinical research include inadequate information regarding research studies, unethical behavior by the research team, and safety concerns (Ceballos et al., 2014; Cortés et al., 2017; Erves et al., 2017; Scharff et al., 2010). Many individuals believe that collected samples like blood, urine, saliva, or stool are unethically disposed of or used after the research study without permission (Ceballos et al., 2014; Kraft et al., 2018). In addition, study participants expressed fear of taking medications that would cause adverse effects, receiving unnecessary surgery, experiencing unintended consequences of the study, having personal information used against them, and being treated like “guinea pigs” or “lab rats” (Cortés et al., 2017; Durant et al., 2011; Erves et al., 2017; Scharff et al., 2010).
Race and EthnicityRace and ethnicity are variables that not only influence patient participation in research but also influence trust. Because of historic and recent events of segregation, racism, and unequal civil rights, AAs report less willingness to participate in research compared to Whites (Dunlop et al., 2011; Durant et al., 2011; Kraft et al., 2018; Ma et al., 2013). A variety of studies reference the Tuskegee syphilis study that was conducted from the 1930s to the 1970s that left the AA community fearful and distrusting of research (Alsan & Wanamaker, 2018; Durant et al., 2011; Ma et al., 2013; Scharff et al., 2010). The Tuskegee study permitted hundreds of adult AA men with syphilis to go untreated despite the availability of an effective treatment: penicillin (Alsan & Wanamaker, 2018). In addition, the treatment of Henrietta Lacks and her family in the 1950s continues to alter the perspectives of AAs toward healthcare institutions and American society (Kraft et al., 2018). Henrietta Lacks was an AA woman whose cells were collected from a cervical cancer biopsy and later used to develop HeLa cells, which were commercialized and highly profitable. The Lacks family did not gain any profit, however, from Ms. Lacks cells (Lee et al., 2019).
Hispanic individuals also face specific cultural and racial variables that influence participation in clinical research (Ceballos et al., 2014; Kraft et al., 2018; Ulrich et al., 2013). Some Hispanic study participants have expressed their willingness to participate in research but have limited understanding of the healthcare system because of immigration to the United States later in their lives (Ceballos et al., 2014; Kraft et al., 2018; Ulrich et al., 2013). In addition, Hispanic persons have expressed fear of racial discrimination and misunderstanding because of language barriers (Ceballos et al., 2014; Ulrich et al., 2013).
EducationAn individual’s education level may affect an individual’s literacy and understanding (Asare et al., 2017), thus affecting what a participant knows and understands about research. In one study measuring recruitment and participation in clinical research, individuals with increased levels of education, particularly college graduates, were more likely to participate (Baquet et al., 2006). The results were consistent with another study where 97% of participants were college educated and reported favorable views of research and willingness to participate in clinical trials (Brewer et al., 2014). In a study measuring AAs’ willingness to participate in research before and after a preconsent education session, individuals with a high school level of education or less were more likely to participate in a clinical trial after receiving preconsent education (Dunlop et al., 2011). Although past researchers have explored the relationship between education level and an individual’s participation in research, convincing evidence of the influence on education and research participation is lacking.
Although many researchers have identified numerous barriers and facilitators to participation in research, few have examined specific correlations to an individual’s attitude toward participating in research. In addition, geographic variation exists in participation barriers throughout the United States. Individuals living in urban areas report greater distrust of research compared to those living in rural areas, yet rural participants report lack of interest in participating in clinical trials compared to those living in urban areas (Baquet et al., 2006; Friedman et al., 2013).
In 2014, the National Patient-Centered Clinical Research Network (PCORnet) was established by the Patient-Centered Outcomes Research Institute with the goal of transforming the culture of clinical research through patient-centered engagement and recruitment (Unertl et al., 2018). Using the multiple healthcare facilities and millions of patients in the network, the Stakeholders, Technology, and Research Clinical Research Network (STAR-CRN), formally known as the Mid-South Clinical Data Research Network (CDRN)—a subunit of the PCORnet—aims to increase the number of research participants through their diverse patient network. To effectively engage patients in the diverse STAR-CRN network, it is necessary to identify the barriers that specific patients encounter during the research process. The purpose of this analysis was to determine how stakeholders' race, trust, and level of education influence participation barriers in clinical research.
The social cognitive theory (SCT) by Bandura (1971) provided the theoretical framework for the study. The theory illustrates how individuals learn and maintain behaviors in the social context in which they live (see Figure 1). The SCT considers the continual interaction between cognitive, environmental, and behavioral factors to ultimately determine human behavior. Cognitive factors include an individual’s knowledge, expectations, and attitudes. Environmental factors include societal and cultural norms, community access and resources, and the influence of others. Behavioral factors include skills, practice, and an individual’s self-efficacy. The triadic reciprocal relationship between cognitive factors, the environment, and human behavior explains the theorized relationship between variables in the study.
FIGURE 1: Core factors of the social cognitive theory with variables used in study.
METHODSThe study used secondary, cross-sectional survey data that were collected between 2014 and 2016 through the former CDRN, currently known as the STAR-CRN. The surveys were distributed throughout clinics within the former Mid-South CDRN after receiving approval from the Vanderbilt University Medical Center Institutional Review Board. The Belmont University Institutional Review Board approved the project as exempt in April 2019.
Clinical SettingAlthough the former Mid-South CDRN is expansive throughout the Southeast, the survey was specifically distributed to patients visiting Vanderbilt University Medical Center or Nashville General Hospital clinic.
Project SampleThe research participants were adults (18 years old and older) living in the Southeastern United States who received care at least one time from a provider at one of the aforementioned clinical sites. There were no further inclusion or exclusion criteria.
Data Collection InstrumentsBetween 2014 and 2016, approximately 5,000 patients in the CDRN were surveyed to identify barriers that impede patient involvement in research. Two parallel surveys were administered using a random process (Erves et al., 2017). The surveys differed by tools measuring the concept of trust. One survey included the tool Hall–Trust in Medical Research (Hall et al., 2006), whereas the other included the Mainous–Trust in Medical Research (Mainous et al., 2006). The current analysis only used data collected from the survey containing the Trust in Medical Research by Hall et al. (2006). All surveys were administered using research electronic data capture (Harris et al., 2019).
Race, ethnicity, and level of education were collected in the demographic portion of the survey. The Trust in Medical Research by Hall et al. (2006) was used to measure the respondent’s level of trust in medical research. The trust tool was developed initially through a pilot study with a 25-item questionnaire. It was then simplified by following an item reduction procedure to develop the current 12-item tool (Hall et al., 2006). The Cronbach's alpha coefficient was .87, and the response pattern was normally distributed (Hall et al., 2006). Questions to assess barriers to participation in medical research were taken from a previous study by Mouton et al. (1997) using a 5-point Likert scale for each statement, ranging from strongly agree to strongly disagree. The specific questions were created from a literature review of barriers to participation in research. A panel of four experts reviewed the 12-question instrument for clarity, content validity, and cultural sensitivity. The content validity and cultural sensitivity both scored as 1.0 (Millon-Underwood et al., 1993).
Data Collection ProcessParticipants were recruited in person at participating clinics. Prior to receiving a survey, participants were informed of the purpose, time commitment, risks and benefits, and compensation, including a $25 gift card. Survey results were stored in a data set through the Meharry–Vanderbilt Alliance.
IBM SPSS Statistics (Version 26) was used for the analysis. Descriptive statistics were performed on the variables of level of education, trust level, race, and each attitude statement in the barriers to participation scale. A Spearman rank correlation was performed between level of education, level of trust, and attitude statement for each racial category.
RESULTSA total of 2,190 survey responses were used in the analysis. Sociodemographic characteristics of survey respondents are shown in Table 1. The mean age of respondents was 52 years (SD = 15.65), with majority being female (68.3%, n = 1,496), White (77.4%, n = 1,696), insured (73.5%, n = 1,610), and working full time (50.4%, n = 1,103). The mean trust score was 39.85 (SD = 6.7). Trust scores by racial grouping are shown in the Supplemental Digital Content (http://links.lww.com/NRES/A386). Middle Easterners reported the least amount of trust (M = 36.11, SD = 5.8) compared to other groups. Most of the respondents had at least 2 years of college education (85.8%, n = 1,880). Education levels are separated by racial groupings in Table 2. Very few respondents in each racial grouping had less than an eighth-grade education.
TABLE 1 - Sociodemographic Characteristics of Survey Respondents Characteristic n % Gender Male 640 29.2 Female 1496 68.3 Other 54 2.5 Race White 1696 77.4 African American 336 15.3 Hispanic 56 2.6 Native American 39 1.8 Asian 23 1.05 Prefer not to answer 27 1.23 Middle Eastern 9 0.4 Native Hawaiian 4 0.2 Education 8th grade or less 17 0.8 Some high school (did not graduate) 58 2.6 High school graduate or GED 219 10.1 Some college or 2-year degree 561 25.6 College degree 638 29.1 More than a college degree 681 31.1 Prefer not to answer 16 0.7 Employment Full time 1103 50.4 Part time (<32 hours) 193 8.8 Unemployed 108 4.9 Volunteer 22 1.0 Stay-at-home parents 87 4.0 Retired 376 17.0 Receiving disability 158 7.2 Other 143 6.5 Insurance Insured 1610 73.5 Uninsured 68 3.1 Medicaid 73 3.3 Self-pay 318 14.5 Other 121 5.5 Household Income <$10,000 142 6.5 $10,000–$14,999 72 3.3 $15,000–$24,999 136 6.2 $25,000–$34,999 197 8.9 $35,000–$49,999 233 10.6 $50,000–$74,999 356 16.3 $75,000–$99,999 288 13.2 $100,000–$149,999 260 11.9 $150,000 or more 218 10.0 Other 288 13.2Overall, the respondents had favorable attitudes toward research participation. Percentage of respondent agreement toward attitude statements are displayed in Table 3. Most of the participants agreed that research benefits society, participation in research means better care, and research in the United States is ethical. Attitudes toward researchers were generally positive in that only a few agreed that “researchers do not care about me” (5.2%, n = 114) and “scientists cannot be trusted” (2.2%, n = 48).
TABLE 3 - Percentage of Survey Respondents Reporting Agreement With Each Attitude Statement Agreement (N = 2,190) Research participation attitude statements n % Participation in research benefits society 1,934 88.3 Participation in research will mean better care 1,637 74.7 Participation in research is risky 526 24 Researcher do not care about me 114 5.2 Participation in research is enjoyable 727 33.2 Participation in research allows me to socialize 372 17 Participation in research is against my religion 41 1.9 Participation in research is morally wrong 31 1.4 Transportation is a problem for research participants 351 16 Scientists cannot be trusted 48 2.2 Research conducted in the United States is ethical 1,523 69.5 It is better to be treated by doctors who are researchers 677 30.9Note. Agreement combines the responses agree and strongly agree.
Spearman correlations were performed using the racial groupings of White, AA, Hispanic, Native American, Asian, and Middle Eastern. Correlation results are displayed in Table 4. Correlations were not performed for the Native Hawaiian group because of a small sample size (n < 5) and the “prefer not to answer” grouping. White, AA, Hispanic, and Native American respondents displayed positive associations between trust and agreement toward “participation in research will mean better care.” Trust was strongly correlated with agreement for each attitude statement for White respondents, except for “participation is against my religion” (correlation coefficient [CC] = −.005, p = .844) and “participation in research is morally wrong” (CC = −.016, p = .509); however, association with education level was variable. Trust level was negatively associated with agreement toward the statement “researchers do not care about me” in White (CC = −.192, p = .000) and Native American (CC = −.371, p = .020) respondents. Trust level was correlated with specific attitude statements for Native American respondents, but there was less evidence of associations involving their education level. Conversely, in Asian respondents, education level was positively correlated with the statements “participation in research is morally wrong” (CC = .540, p = .008), “scientists cannot be trusted” (CC = .568, p = .005), “research conducted in the United States is ethical” (CC = .453, p = .030), and “it is better to be treated by doctors who are researchers” (CC = .418, p = .047).
TABLE 4 - Spearman Correlations Between Attitude Statements, Trust Score, and Education by Race Statement White
留言 (0)