COVID-19 clinical trial participation and awareness in Texas

ABSTRACT

The COVID-19 pandemic required the rapid development of COVID-19 vaccines and treatments, necessitating quick yet representative clinical trial enrollment to evaluate these preventive measures. However, misinformation around the COVID-19 pandemic and general concerns about clinical trial participation in the U.S. hindered clinical trial enrollment. This study assessed awareness of, willingness to participate in, and enrollment in COVID-19 vaccine and treatment clinical trials in Texas. A quota sample of 1,089 Texas residents was collected online from June – July 2022. Respondents were asked if they were aware of, willing to participate in, and had enrolled in clinical trials for COVID-19 vaccines or treatments. Overall, 45.8% of respondents reported being aware of clinical trials for COVID-19 treatments or vaccines, but only 21.7% knew how to enroll and only 13.2% had enrolled in a COVID-19 clinical trial. Respondents with bachelor’s or graduate degrees were more likely to be aware of clinical trials, more likely to have enrolled in trials, and more willing to participate in treatment trials. Women were less willing to participate and less likely to have enrolled in COVID-19 clinical trials than men. Respondents aged 55 years and older were more willing to participate, but less likely to have enrolled in COVID-19 clinical trials than 18-to-24-year-olds. Common reasons given for not participating in clinical trials included concerns that COVID-19 treatments may not be safe, government distrust, and uncertainty about what clinical trial participation would entail. Substantial progress is needed to build community awareness and increase enrollment in clinical trials.

KEYWORDS:

• COVID-19
• clinical trials
• vaccines
• COVID-19 treatments
• public health
• health communication

This article is related to:

The seroepidemiology of immunoglobulin G antibodies against pertussis toxin and filamentous hemagglutinin in the east of China during the COVID-19 pandemic

Introduction

The COVID-19 pandemic required the rapid development of COVID-19 prevention strategies, including vaccines and treatments. As such, quick and population-representative clinical trial enrollment was essential to prevention success. Unfortunately, historical atrocities (e.g., the Tuskegee Experiment) have led to concerns about enrolling in clinical trials among historically marginalized populations, such as Black, Indigenous, and People of Color (BIPOC) communities. For example, at one point in the pandemic, Black people represented only 3% of enrollees in vaccine trials, while comprising 13% of the U.S. population.¹ Moreover, clinical trial enrollment concerns centered on the swift development of the COVID-19 vaccines in the United States and uncertainty of the safety of participating in a vaccine trial.² Because of these public concerns regarding clinical trials and the expedient need for scientists to enroll participants in these trials, the Community Engagement Alliance (CEAL) Against COVID-19 was formed by the National Institutes of Health (NIH). CEAL partnered with states to implement community-engaged research and outreach with a goal of increasing inclusive participation in clinical trials. The Texas CEAL Consortium consists of community stakeholders and academic partners representing six diverse counties in Texas.³

Several studies have assessed COVID-19 clinical trial enrollment intentions and attitudes. Studies from other countries have found a range of willingness to participate in COVID-19 vaccine trials early in 2020 during the pandemic, spanning 20% among a sample in Nigeria to 64% among a sample in China.^4–8 Within the United States, a survey among adults in Michigan found a low willingness to participate in clinical trials at the end of 2020.⁹ Moreover, racial/ethnic group-based medical mistrust was a contributing factor to low willingness to participate among underrepresented minority participants. Another contributing factor to difficulty enrolling clinical trial participants may have been complicated and lengthy informed consent forms,¹⁰ which can be challenging for people with low literacy. Interviews with vaccine clinical trial participants in Iowa reported their motivations for participating in the trial were related to protecting themselves and others, promoting vaccination, career motivations (e.g., return to work), a sense of duty, and supporting the scientific process. Yet, participants still had concerns related to side effects and the speed of vaccine development.¹¹ Broadly speaking, the United States lacks an established national infrastructure for launching rigorous clinical trials, including established recruitment networks through public hospitals and clinics.¹² For example, the United Kingdom tapped into public healthcare networks to enroll more than 12,000 hospitalized patients across 176 hospitals into the World Health Organization-sponsored RECOVERY trial less than two weeks after the trial protocol was approved.¹³ Because clinical trials research in the United States is decentralized and driven by individual investigators,¹² modes of recruitment may be inconsistently applied, and medical providers themselves may not be aware of or motivated to offer clinical trial enrollment options for their patients.

Most of the research on COVID-19 clinical trials was conducted early in the pandemic at the time of vaccine development and with much of the public discourse focused on COVID-19 prevention. Yet, clinical trials are still ongoing for COVID-19 prevention activities, which requires motivated and willing people to participate. Texas CEAL has continued to assess the public’s willingness to participate in clinical trials. The purpose of this study was to assess awareness of, willingness to participate in, and enrollment in COVID-19 clinical trials in Texas in mid-2022. This study also examined demographic differences in trust of different informational sources around COVID-19 clinical trials. Findings from this study can inform future clinical trial enrollment techniques both related and unrelated to the COVID-19 pandemic.

Methods

Sample

This paper is part of a larger study that examines attitudes and behaviors around COVID-19 vaccination and clinical trial enrollment across Texas. A quota sample was recruited in June-July 2022 based on the racial/ethnic distribution of Texas using a Qualtrics Panel. Potential respondents were identified from a double-opt-in market research panel, and respondents were randomly selected among those deemed likely to qualify for eligibility. Eligibility criteria necessitated that participants be residents of Texas, of age 18 years or older, and able to read and respond to a survey written in English. The total sample size consisted of 1,213 respondents, with 1,089 of those being completed and verified responses (i.e., met the minimum threshold for time to complete, and a quality check item was answered correctly). This resulted in a 90% completion rate of administered surveys. Additionally, some individuals were excluded due to missing data on the outcome questions or key demographic variables. The total sample analyzed for this study was N = 985. Study protocols were approved by the North Texas Regional IRB (Project ID 2022-020), and respondents had to complete a consent form to be eligible for participation.

Measures

Measures used in this study were obtained from the survey developed by the NIH-wide CEAL Assessment and Evaluation Workgroup. The primary outcomes for this study were 1) participant awareness of existing COVID-19 clinical trials (“Are you aware of COVID-19 clinical trials that are being done?” and “do you know what to do to sign up for a trial in your area?”); 2) the willingness to participate in a treatment trial (”If you get COVID-19, how willing would you be to sign up for a clinical trial for a COVID-19 treatment?”); and 3) prior participation in COVID-19 clinical trials (“Have you ever enrolled in a COVID-19 clinical trial?”). For the awareness and past enrollment questions, participants were given the option to select from “yes – for treatment,” “yes – for vaccine,” or “no” (including “not sure”). Due to the low number of responses for “yes – for treatment,” responses were combined into two categories for the analysis models, with “yes – for treatment” and “yes – for vaccine” as “yes.” Willingness to participate in a trial was rated on a scale from 1 to 7, with 1 being “not at all willing” and 7 being “very willing.”

Participants were asked how much they trust seven different sources to provide correct information pertaining to COVID-19 clinical trials. The sources included: NIH, your doctor or healthcare provider, your local healthcare clinic or hospital, companies that make drugs for medical use, people who do research, friends/family, and community leaders. Response options were “a great deal,” “a fair amount,” “not very much,” “not at all” or “no opinion.” Based on the low response rate of some categories such as “no opinion,” responses were dichotomized as “lack of trust” by combining “not very much,” “not at all,” and no opinion“; and “yes” for affirming trust by combining “a fair amount” and “a great deal.” Because “no opinion” was included in the negative trust, this variable is an indicator of a respondent affirming trust in the information source vs. not affirming trust in this source, as opposed to a strict dichotomy of trust vs. distrust in the given information source. To identify why participants were not willing to enroll in a clinical trial, they were asked, “if you were to get COVID-19, why would you NOT take part in a clinical trial for a COVID-19 treatment?” Respondents could select more than one option for this question, as well as an “Other” response. Additionally, self-reported data on race and ethnicity, age, gender, and education level were also collected. During analysis, age was categorized into five groups: 18–24, 25–34, 35–44, 45–54, and 55+ years old.

Analysis

Descriptive statistics captured the proportion of respondents who were aware of trials and had signed up for trials, both overall and stratified by demographic groupings. Binomial regression was used to determine which variables and demographic categories were significantly associated with trial awareness and trial enrollment, and odds ratios were estimated to quantify the magnitude of the associations. A generalized linear model with a log link was used to model associations with willingness to participate in trials, as willingness was rated as an integer value between 1 and 7. Stratified bar charts examined descriptive differences between trust levels of information sources and awareness of or past enrollment in COVID-19 clinical trials. Multivariable binomial regressions were also used as a follow-up analysis to examine any demographic variables that were associated with trust or distrust of the seven sources of clinical trial information. Finally, reasons given for not participating in clinical trials were summarized, with the number of endorsements for each reason represented in a bar chart. All analyzes were conducted in R,¹⁴ and data management and visualization were executed using packages from the tidyverse.¹⁵

Results

Descriptive results

Table 1 provides representation of each category for the categorical outcomes and demographic variables. Overall, 451 individuals (45.8%) reported being aware of clinical trials around either treatments or vaccines for COVID-19, and 214 (21.7%) were aware of how to sign up for a clinical trial. However, only 130 respondents (13.2%) had signed up for a COVID-19 clinical trial. Of these participants, the majority (N = 95) had enrolled in a vaccine trial, 31 enrolled in a treatment trial, and 4 participants had enrolled in both types of trials.

Table 1. Univariate counts and percentages for outcomes and demographic variables included in the analyzes. Totals are based on final sample size of N = 985 after filtering missing data and small group sizes. HS, high school.

Variable	Category	N (%)
Aware of COVID-19 Trials	Yes	451 (45.8)
	No	534 (54.2)
Know how to sign up for clinical trial	Yes	214 (21.7)
	No	771 (78.3)
Have signed up for COVID-19 clinical trial	Yes	130 (13.2)
	No	855 (86.8)
Age	18–24 years	157 (15.9)
	25–34	234 (23.8)
	35–44	201 (20.4)
	45–54	126 (12.8)
	55+	267 (27.1)
Race/Ethnicity	White non-Hispanic	402 (40.8)
	Black/African American	116 (11.8)
	Hispanic	389 (39.5)
	Asian	49 (5.0)
	Other	29 (3.0)
Gender	Man	405 (41.1)
	Woman	580 (58.9)
Education Level	No HS Degree	28 (2.8)
	HS Graduate	254 (25.8)
	Some College	252 (25.6)
	Associate’s Degree	121 (12.3)
	Bachelor’s Degree	214 (21.7)
	Graduate Degree	116 (11.8)

Trial awareness

Overall, 7.51% of respondents reported awareness of COVID-19 treatment trials, 25.6% reported awareness of vaccine trials, and 13.2% reported awareness of both types of trials. 53.7% were unaware of either type of trial. Binomial regression was used to examine differences in awareness by demographics. Regression results are presented in Table 2, excepting the analysis of knowing how to sign up. Across demographic variables, Asian Americans were more likely to be aware of trials than White non-Hispanic respondents (OR = 2.04; 95% CI 1.06, 3.91), and those with higher education were more likely that individuals with a High School degree to be aware (bachelor’s OR = 1.48; 95% CI 1.01, 2.17; graduate degree OR = 2.27; 95% CI 1.41, 3.64). There were no significant differences between men and women, between racial/ethnic groups compared to non-Hispanic whites, or across age in terms of trial awareness (Table 2). In terms of knowing how to sign up, the only significant finding was those with a bachelor’s degree being more aware of how to sign up compared to high school graduates (OR = 1.71, 95% CI 1.10, 2.65), and the oldest individuals were somewhat less aware of how to sign up (OR = 0.55, 95% CI 0.31, 0.91). All other demographic comparisons were non-significant.

Table 2. Regression model estimates of covariate associations with awareness of, willingness to participate in, and enrollment in COVID-19 vaccination or treatment trials. Awareness and having signed up were modeled with logistic regression, and willingness to participate was modeled with log-linear regression.

		Awareness of COVID-19 Trials			Willing to Participate in Treatment Trial			Have signed up for COVID-19 Clinical Trial
Variable	Category	Proportion aware within category	Odds Ratios	95% CI	Median willingness in category (Q1, Q3)	Incidence Rate Ratios	95% CI	Proportion enrolled within category	Odds Ratios	95% CI
Age	18–24 years	0.48	1.00	reference	3 (1, 4)	1.00	reference	0.22	1.00	reference
	25–34	0.47	0.85	0.56–1.28	3 (1, 5)	0.98	0.87–1.10	0.17	0.61	0.36–1.04
	35–44	0.44	0.66	0.42–1.02	4 (1, 6)	1.07	0.95–1.21	0.19	0.61	0.34–1.07
	45–54	0.48	0.83	0.51–1.37	3 (1, 5)	1.06	0.93–1.21	0.05	0.14	0.05–0.34
	55+	0.43	0.65	0.41–1.01	4 (1, 6)	1.14	1.01–1.28	0.04	0.10	0.05–0.22
Race/Ethnicity	White non-Hispanic	0.47	1.00	reference	4 (1, 6)	1.00	reference	0.10	1.00	reference
	Black/African American	0.36	0.67	0.42–1.04	3 (1, 5)	1.00	0.89–1.12	0.13	1.00	0.49–1.97
	Hispanic	0.46	0.96	0.70–1.32	4 (1, 6)	1.07	0.99–1.16	0.16	1.09	0.67–1.78
	Asian	0.67	1.96	1.04–3.85	3 (1, 4.75)	0.85	0.71–1.01	0.18	1.17	0.48–2.63
	Other	0.31	0.54	0.22–1.20	3 (1, 4.25)	0.92	0.74–1.14	0.14	1.33	0.35–4.04
Gender	Man	0.49	1.00	reference	4 (1, 6)	1.00	reference	0.18	1.00	reference
	Woman	0.43	0.79	0.61–1.03	3 (1, 5)	0.86	0.80–0.92	0.10	0.46	0.31–0.69
Education Level	No HS Degree	0.43	0.99	0.44–2.20	3 (1, 4.5)	0.97	0.76–1.21	0.25	2.11	0.75–5.41
	HS Graduate	0.42	1.00	reference	3 (1, 5)	1.00	reference	0.11	1.00	reference
	Some College	0.42	1.07	0.74–1.54	3 (1, 5)	1.03	0.93–1.14	0.09	1.10	0.60–2.01
	Associate’s Degree	0.36	0.86	0.54–1.36	4 (1, 5)	1.09	0.97–1.23	0.09	1.17	0.53–2.45
	Bachelor’s Degree	0.53		1.03–2.22	4 (2, 6)	1.25	1.13–1.38	0.19	2.50	1.43–4.42
	Graduate Degree	0.62	2.43	1.51–3.94	5 (2, 6)	1.27	1.13–1.44	0.17	2.82	1.39–5.68

Bolded results indicate statistical significance at $\mathit{p}<.05$ level; CI, confidence interval. HS, High School.

Treatment trial willingness

On the 1 to 7 willingness to participate in a trial scale, the median response was a 4 (mean = 3.56, 1^st quartile = 1, 3^rd quartile = 5). Overall, 31.8% of respondents were “not at all willing” to participate, while 17.4% were “very willing.” Log-linear regression was used to assess trial willingness because it was scored on a seven-point scale with a skewed distribution (many 1’s and 7’s). Women were less willing than men to sign up for a treatment trial (incidence rate ratio [IRR] = 0.86; 95% CI 0.8, 0.92). Those with a bachelor’s or graduate degree were somewhat more willing to sign up (IRR = 1.24 and 1.28, respectively; see Table 2). There were no racial or ethnic differences in willingness to participate. Compared to the youngest group 18–24 years old, the oldest group (55+) said they were significantly more willing to participate in trials, IRR = 1.14 (95% CI 1.01, 1.28).

Trial participation

There were no significant racial differences in the odds of having signed up for a trial. Women were much less likely to have signed up for a trial than men (OR = 0.46; 95% CI 0.31, 0.69). Those with a bachelor’s or graduate degree were much more likely to have signed up for a trial than those with a high school degree (OR = 2.5 (95% CI 1.43, 4.42), and OR = 2.82 (95% CI 1.39, 5.68), respectively; see Table 2). Those who were 45–54 years of age or 55+ years of age were significantly less likely to have signed up for any trial compared to the youngest group (OR = 0.14, 95% CI 0.05, 0.34; OR = 0.10, 95% CI 0.05, 0.22, respectively).

Sources of trusted information regarding clinical trials

First, we examined levels of trust stratified by those who were or were not aware of trials and those who had or had not signed up for clinical trials, presented in Figure 1 (awareness) 2 (enrollment). Next, we added trust of the sources of information about COVID-19 trials to the binomial model for trial participation, with “trust” scored dichotomously. Individuals who trusted the NIH were marginally more likely to have signed up for trials (OR = 1.74, 95% CI 1.00, 3.10, p = .053). Interestingly, individuals who trusted their doctor and local hospital had lower odds of participating in trials, OR = 0.4 (95% CI 0.20, 0.79) and OR = 0.41 (95% CI 0.20, 0.80), respectively. Reporting trust in the companies that make drug treatments was also associated with higher odds of participation (OR = 2.59; 95% CI 1.56, 4.41). Additionally, those who trusted their social contacts (“family, friends, or community leaders”) had higher odds of participation (OR = 1.92; 95% CI 1.20, 3.11). Trust or distrust of university hospitals and researchers did not have a significant impact on the rates of trial participation. However, in terms of reported willingness to participate in a treatment trial, those who trusted the NIH (IRR-1.31, 95% CI 1.05, 1.34) and university hospitals (IRR = 1.14, 95% CI 1.02, 1.28) reported significantly higher levels of willingness. Trusting or distrusting other information sources did not have a significant impact on reported willingness.

Figure 1. Proportions of trust ratings (“no,” “no opinion,” or “yes”) for respondents that were and were not aware of ongoing COVID-19 clinical trials.

Comm., community.

Figure 2. Proportions of trust ratings (“no,” “no opinion,” or “yes”) for respondents that had and had not signed up for COVID-19 clinical trials.

Comm., community.

Finally, we examined any demographic differences between trust and distrust for each of the sources of information about COVID-19 trials. Binomial logistic models were fitted with probability of reporting trust in the seven sources as the outcome and age level, education level, race/ethnicity, and gender included in the model. Only education level resulted in differences in trust of NIH (adjusting for other variables), with those with a bachelor’s (OR = 1.95, 95%CI 1.32, 2.9) or graduate degree (OR = 2.18, 95%CI 1.34, 3.59) more likely to trust the NIH. The oldest individuals were more likely to trust their own doctor compared to the youngest group (OR = 1.93, 95% CI 1.11, 3.37), whereas Asians were less like to trust their doctor compared to non-Hispanic White respondents (OR = 0.45, 95% CI 0.22, 0.92), and women were less likely to trust their doctors compared to men (OR = 0.71, 95% CI 0.51, 0.99). Those with a graduate degree were also much more likely to trust their doctors than high school graduates (OR = 2.36, 95% CI 1.21, 4.92). Only those with a graduate degree reported any difference in trust of hospitals or clinics, with those individuals more likely to trust these sources for information on COVID-19 trials (OR = 2.72, 95% CI 1.42, 5.59). Hispanic individuals were more likely to trust drug companies than non-Hispanic White individuals (OR = 1.42, 95% CI 1.03, 1.94), controlling for the other demographic variables. In the same model, those with bachelor’s (OR = 1.53, 95% CI 1.04, 2.24) or graduate degrees (OR = 2.39, 95% CI 1.49, 3.85) compared to high school graduates were more likely to trust drug companies. There were no demographic differences in terms of trusting one’s social contacts when it came to COVID-19 trials.

Reasons for non-participation

Figure 3 presents the raw number of times that reasons for nonparticipation in COVID-19 clinical trials were given. The most common reason given was fear that the treatment might not be safe (n = 337). Individuals also cited distrust of the government (n = 245) and researchers (n = 161) as reasons not to join a clinical trial, as well as uncertainty about what would happen if they chose to participate (n = 211). Finally, respondents worried that clinical trial participation would cost them time (n = 175) and money (n = 102). Interestingly, only 40 respondents said that they “don’t believe clinical trials are important,” meaning that a general dismissal of the usefulness of clinical trials was the least endorsed reason for nonparticipation.

Figure 3. Reasons given for being unwilling to participate in a COVID-19 treatment clinical trial. Respondents were asked “if you were to get COVID-19, why would you NOT take part in a clinical trial for a COVID-19 treatment?,” and more than one reason could be selected. A total of 950 participants provided at least one valid response, and 35 individuals in the “other” response wrote in that they would be willing to participate in a trial. The number of endorsements for each response are printed at the top of each bar.

Discussion

This study examined demographic characteristics associated with awareness, willingness to participate, and actual enrollment in COVID-19 clinical trials in Texas. Furthermore, this study highlighted reasons for nonparticipation and the influence of sources of trusted information about COVID-19 clinical trials. The association of higher educational level with increased awareness and participation in COVID-19 clinical trials may indicate that individuals with higher education have more access to and understanding of clinical trials. Our findings are consistent with previous studies that report individuals with a bachelor’s degree and higher were more knowledgeable about clinical trials and more likely to enroll in clinical trials.^16–18 Although our findings show that the oldest age group, aged 55 years and above, were more willing to participate in COVID-19 treatment trials, individuals aged 45 years and above were less likely to have signed up for a clinical trial. This may be due to the oldest age group reporting less awareness of how to sign up. Healthcare providers may support their older patients in navigating clinic trial recruitment and participation. A systematic review highlights lack of knowledge about clinical trials as one of the reasons for nonparticipation.¹⁹ Older adults may require assistance in signing up for clinical trials, which highlights the need for personalized resources. Information concerning clinical trials should be easily comprehensible by individuals of all education levels and age groups. Proper understanding of clinical trial procedures may motivate interest and enrollment as participants are able to make informed decisions. The lack of recommendations from healthcare providers on clinical trial participation can impede enrollment among older adults. In a previous study among oncologists, academic oncologists reported that advanced age and presence of multiple health conditions deterred physicians from suggesting clinical trials to their older patients.²⁰ In this same study, the academic and community oncologists highlighted the possible risk of adverse reactions as obstacles to recruitment. Furthermore, the diminished likelihood to participate in clinical trial among older adults can be attributed to transportation concerns^19,21 as this limits access to attend the clinic trials.

Furthermore, messages concerning clinical trials should be tailored to meet the needs of the participants. A randomized controlled trial found that using a video to educate participants about clinical trials did not improve perception about clinical trials, and these participants were less likely to enroll.²² Discussions about clinical trials during clinic interactions can help address misconceptions and concerns about the safety of trials, ensuring that messages are personalized. Another study found that collaboration with community leaders and members to frame messages around COVID-19 clinical trials encouraged trial participation and increased diversity.²³

We found that women were less likely to enroll in COVID-19 clinical trials. For decades, women of reproductive age have been excluded from clinical trials due to fertility-related concerns.^24–26 A study that examined the participation of women in COVID-19 vaccine clinical trials using the US National Library of Medicine discovered that about 97.8% and 81.1% of the trials did not include pregnant and breastfeeding women, respectively. Also, more than 50% of the trials mandated that women of childbearing age should be on contraception.²⁷ Lack of representation of women in clinical trials may contribute to insufficient information or scientific evidence on the efficacy and appropriate treatment regimen of the proposed treatment in women. This may also increase the risk of adverse health consequences. Our study findings highlight the need to engage more women to participate in clinical trials. Incentives should be considered for enrollment as women from minority populations may face additional barriers, such as transportation, childcare, and inability to take time off work.^23,28

Survey respondents also reported their level of trust in numerous sources of information regarding COVID-19 clinical trials. Participants who reported the NIH as their most trusted source were marginally more likely to have signed up for clinical trials, and having a bachelor’s or graduate degree was associated with more trust the NIH than those with other education levels. This conclusion is in line with prior literature that explains the positive effect of college degree attainment on health behavior outcomes.²⁹ Similarly, participants who trusted the companies that make drug treatments were more likely to participate in clinical trials, and Hispanic individuals and those with bachelor’s or graduate degrees were more likely to report trust in these companies. Individuals of Hispanic ethnicity reporting trust in drug companies (even after accounting for education level) was surprising, as prior literature has shown that minority populations in the U.S. were less likely to trust pharmaceutical companies, which then contributed to less favorable attitudes toward vaccination and lower vaccine uptake.³⁰ Also, analysis of a national sample found that Hispanic Americans were less willing to participate in clinical trials than White non-Hispanic individuals.³¹ Higher levels of clinical trial participation were also seen in those who most trusted their social contacts, which constituted family, friends, or community leaders. This is a common finding as prior studies have shown that trust in social contacts positively impact health-seeking behaviors like vaccination and clinical trial participation, particularly if one highly values the opinion of their social networks.³² In this study, there were no demographic differences in terms of trusting one’s social contacts, despite there being individualistic or collectivistic cultural differences across racial/ethnic lines that can dramatically shape social connections and ultimately influence participation in clinical trials.³³

Interestingly, individuals who reported their doctors and local hospitals as their most trusted source were less likely to participate in trials. This finding is contradictory to numerous literature findings, as a stronger and more trusting patient-physician relationship generally led to increased participation in clinical trials.^34–38 However, this could be partially explained because of the low rate of clinical trial participation in our sample. In our sample, most individuals reported trust in their doctor and local hospital provider (N = 775 and 728, respectively), yet only 130 had enrolled in a trial. For example, the percentage of non-enrolled participants who trusted their doctor (79.5%) was slightly higher than the percentage of trial enrolled individuals who trusted their doctor (73.1%), but levels of trust in doctors were high for both groups. Individuals who trust their providers may be more willing to consider clinical trial participation. Sood et al.³⁹ reported that more than half of the patients within a clinic setting anticipated information about clinical trials from their doctors. This emphasizes the responsibility of healthcare providers to provide adequate information, including addressing misconceptions about clinical trial participation. However, studies show that providers’ perceptions or bias about clinical trials can influence their willingness to recommend participation to their older patients.^19,40 Trust or distrust of university hospitals and researchers did not significantly impact participants’ rates of trial participation.

The most common reason for nonparticipation in COVID-19 clinical trials was fear that the treatment given at the trial would not be safe. Another commonly reported reason was uncertainty about what would happen to individuals that chose to participate. Lack of certainty about the safety of a given treatment is a commonly cited fear that hinders motivation for clinical trial participation.^41,42 Such fears and safety concerns can be remedied by providing prospective trial participants with a thorough and detailed explanation of what the proposed treatment entails.⁴³ Additionally, we found that distrust of the government and researchers was also a common reason for nonparticipation in clinical trials. Prior literature has shown that mistrust in government, doctors, and researchers is especially prominent in minority and underrepresented groups, due in part to infamous experiments such as the Tuskegee Syphilis Study, which violated numerous aspects of human rights.⁴⁴ However, we did not find evidence of elevated distrust of these information sources among racial/ethnic minorities. Finally, respondents worried that participation would cost them time and money. Numerous studies assessing facilitators and barriers to clinical trial participation have found that cost of time and money was consistently ranked as an important factor in individuals’ decision to participate.^45–47 One commonly proposed solution to remedy such barriers is for insurance programs to guarantee coverage of routine-care costs associated with clinical trials for their beneficiaries.⁴⁷ Another recommendation is for providers to reimburse clinical-trial services for their patients.⁴⁷ Understanding these barriers to clinical trial participation and developing recommendations for solutions could dramatically increase participation levels. The fact that dismissal of the importance of clinical trials was the least endorsed reason for unwillingness to participate indicates that more people might get involved in COVID-19 clinical trials, and clinical trials in general, if they are confident in the safety of the treatment and comfortable with the processes of clinical trial involvement.

This study has limitations that should be considered. First, the sample is from an online panel, so a sampling frame cannot be established. Moreover, only English speakers were included the sample, which may not reflect the full diversity of the population in Texas. Additionally, these data were self-reported, and we are unable to confirm enrollment in COVID-19 clinical trials, which may lead to misclassification. We also did not gather information on the location or type of community that respondents lived in, and access to COVID-19 clinical trial information and enrollment may differ between urban, suburban, and rural regions of Texas. Finally, the study is a cross-sectional design, and we cannot establish temporality between variables (e.g., trust and participation).

Conclusion

Awareness of, participation willingness, and actual participation in COVID-19 clinical trials are important public health challenges. The Texas CEAL Consortium, and CEAL more broadly, has been charged with promoting awareness and participation in clinical trials and promoting trust in the scientific process involved in vaccine and treatment development. This study demonstrates that substantial progress is needed to build community trust and increase enrollment in clinical trials and scientific research. Future work should focus on science and health literacy, straightforward messaging around clinical trials, and utilizing trusted messengers to disseminate clinical trial enrollment information.

Authors’ contributions

ELT, JML: conception and design; JML, ELT: analysis and interpretation of the data; JKV, ELT: resources and data curation; JML, ELT, INA, SA, TT: original drafting of the paper; all authors: critical review and revision; all authors: final review. All authors agree to be accountable for the work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, JML. The data are not publicly available due to privacy concerns.

Additional information

Funding

This study was funded by the National Institutes of Health (NIH) Agreement [1OT2HL156812] as part of the NIH Community Engagement Alliance (CEAL), PI: Vishwanatha. The funding agency had no role in the design of the study or in the writing of the manuscript.