Abstract
Research participants should be drawn as fairly as possible from the potential volunteer population. Underlying personality traits are underexplored as factors influencing research decision-making. Dispositional optimism, known to affect coping, physical health, and psychological well-being, has been minimally studied with respect to research-related attitudes. We conducted an exploratory, online survey with 151 individuals (self-reported mental illness [n = 50], physical illness [n = 51], neither [n = 50]) recruited via MTurk. We evaluated associations between dispositional optimism (assessed with Life Orientation Test – Revised) and general research attitudes, perceived protectiveness of five research safeguards, and willingness to participate in research utilizing safeguards. Strongly optimistic respondents expressed more positive research attitudes and perceived four safeguards as more positively influencing willingness to participate. Optimism was positively associated with expressed willingness to participate in clinical research. Our findings add to a limited literature on the influence of individual traits on ethically-salient research perspectives.
Keywords: Safeguards, Optimism, Research, Perspectives, Willingness
INTRODUCTION
Much of the existing literature on ethical aspects of human research has scrutinized what happens after recruitment to research—in particular, during the informed consent process itself.1 For instance, conceptual work has articulated models for assessing decision-making capacity, voluntarism, and the therapeutic misconception.2 Empirical studies have also examined predictors of decision-making capacity for research, methods of improving the informed consent process, and participants’ perceptions of research risks and benefits, among other issues.3
Another crucial, yet less studied, set of research ethics issues encompasses those activities that occur prior to informed consent. For example, upholding the ethical principle of justice in research (i.e., the fair distribution of the benefits and burdens of research) requires that research participation be offered not only to those individuals who are the most willing or the most conveniently accessible to researchers, but also to those who may be harder to reach or less interested in or amenable to research. An ongoing challenge in health research has been the persistence of disparities in research participation rates across various sociodemographic groups.4 The ethical principles of respect for persons and beneficence are supported by the implementation of a range of research safeguards (e.g., institutional review boards; investigator training in the ethical conduct of research; procedures for alternate decision-making when individuals lack capacity).
Fulfilling these fundamental ethical principles necessitates understanding the myriad factors that may initially interest people in research as well as those that may influence participation decisions.5 Such factors might include individual disease- or treatment-related factors (e.g., having the illness being studied; lack of other treatment options; having a family member with a serious illness); external factors (e.g., compensation, physician referral or recommendation, views of family members); protocol-related factors and the individual’s assessment of these factors (e.g., risk assessment, awareness and perceptions of research safeguards), and individual-level factors, such as personality traits, that are not necessarily illness- or protocol-related (e.g., hope, trust, altruism, optimism).
In terms of protocol-related factors, relatively little empirical work has examined potential research participants’ views of research safeguards.6 This gap is notable because the ethical caliber of research relies on the effective implementation of a range of safeguards. Moreover, the informed consent process is predicated, in part, on the assumption (rarely tested) that potential research participants appreciate being informed about the array of safeguards that are in place to protect them. Empirical work is needed that examines, for example, whether participants themselves actually view these various safeguards as protective, and, in turn, whether the perceived protectiveness of safeguards is associated with potential participants’ willingness to participate.
In prior work, members of our team have examined the association of research safeguards with willingness to participate in research across varied stakeholder groups, including among people with schizophrenia (compared to their psychiatrists’ predictions of their willingness), as well as among people with mental illness, family members of people with mental illness, and community controls.7 While these studies found that potential research participants viewed safeguards as both generally protective and expressed greater willingness to participate in research studies with safeguards in place, only one prior study adjusted for individual traits as potential confounding variables within this relationship.8
Underlying individual personality traits have been relatively underexplored as factors influencing research perceptions and decision-making, whereas a large body of research has documented the association of personality traits with many health-related behaviors, risk factors, and health outcomes, which are often treated as either inclusion or exclusion criteria, or primary outcomes of research studies.9 Moreover, the salience of personality traits as predictors of research participation is supported by recent work that examined whether personality could predict non-response or the decision not to participate in a research study.10
Optimism, in particular, has theoretical links to ethically-salient perspectives on medical research and willingness to participate in research.11 Research from early phase cancer trials, for example, revealed that patients with serious illness were optimistic about participating and receiving personal benefit from clinical studies.12 In early pilot work related to the present study, we found anecdotal evidence for higher levels of optimism among prospective research participants who expressed high willingness to participate in innovative research.
At first blush, it may not be obvious why optimism should be of concern in the context of research ethics. First, it is important to distinguish between “unrealistic” optimism and dispositional optimism. In their research examining potential associations between optimism and therapeutic misconception (i.e., poor appreciation of crucial differences between clinical research and clinical care, leading to incorrect interpretation of research as having therapeutic intent), Jansen and colleagues utilized the following definition of unrealistic optimism: “a bias in which a person mistakenly believes that she is more likely to experience positive outcomes with respect to a specific event compared to similar others facing the same event.”13 In contrast, dispositional optimism is an enduring personality trait, defined as an individual’s predisposition toward positive expectations about the future, or, as defined by Scheier and Carver, the “tendency to expect good outcomes across important life domains.”14
In a study of 171 patients enrolled in early phase cancer clinical trials, Jansen and colleagues examined the relationship of dispositional optimism with unrealistic optimism, as well as the association between these forms of optimism and expectations of personal therapeutic benefit as well as therapeutic misconception.15 Interestingly, in multivariate analyses, unrealistic optimism was associated with both the therapeutic misconception and expectations of personal benefit, while dispositional optimism was associated independently only with expectations of personal benefit but not with the therapeutic misconception. These findings underscore the importance of carefully defining and measuring different forms of optimism.
Jansen and colleagues’ important work laid the groundwork for examining the association of different forms of optimism with expectations of benefit and therapeutic misconception among patients already enrolled in early-phase cancer clinical trials. However, in the broader landscape of clinical research, including research that enrolls healthy individuals, no prior studies have examined potential associations between optimism and other key research attitudes and perceptions, despite a robust literature supporting relationships between dispositional optimism and physical health.16 In particular, there is a need to examine how individual traits, such as optimism, are associated with ethically-relevant dimensions of research participation that go beyond risks and benefit assessments, such as attitudes toward research and perceptions of research safeguards.
In a sample of individuals living with mental illness, physical illness, or neither, we conducted an exploratory pilot study that evaluated: 1) whether dispositional optimism was associated with attitudes toward medical research; 2) whether dispositional optimism was associated with perceived protectiveness of various research safeguard procedures; 3) whether dispositional optimism was associated with willingness to participate in research that included these safeguards; and 4) whether dispositional optimism was associated with willingness to participate in research that included these safeguards, after adjusting for safeguard type, perceived protectiveness, and other potential confounders. As this was a pilot study, we did not specify the directionality of the above hypotheses.
METHODS
Design.
This study was approved by the IRB at Stanford University and was funded by the National Institutes of Health (R01MH114856). This pilot phase involved evaluation of the feasibility of using the Amazon MTurk Platform (as described below) to recruit individuals who self-identify as either healthy, mentally ill, or physically ill in order to assess stakeholder views of ethical issues in neuroscience research. This pilot phase was completed after a pre-pilot phase in order to “fine tune” the survey instrument for a later, larger scale project.
Study recruitment and population.
Respondents were recruited through Amazon Mechanical Turk (MTurk), an online crowdsourcing platform that enables data collection.17 MTurk workers were recruited through an advertisement on the MTurk platform to complete a Human Intelligence Task (HIT) entitled “Health Information Survey.” This task was a screening questionnaire completed by MTurk workers who met the following criteria: (1) reported living in the United States; (2) were at least 18 years of age; and (3) had an MTurk approval rating of at least 90%. These default criteria for participant selection are recommended in the Amazon Mechanical Turk Requester UI Guide. The screening survey was available on the MTurk Platform for eight days in May 2018. Participants received $0.25 for completion of the initial screening. 1299 workers completed the screening survey.
Following the initial screening, respondents were invited to participate in a subsequent survey if they reported: (a) living with mental illness (i.e., self-reported history of any mental illness and/or substance use disorder and no self-reported history of any physical disorders); (b) living with physical illness (i.e., self-reported history of any physical illness disorder and no self-reported history of any mental and/or substance use disorders);and (c) living in good health (i.e., rating their overall health at least a 4 on a 5 point scale, with no self-reported history of mental or physical disorders). Individuals who reported both physical illness and mental illness were excluded. MTurk workers who met these study inclusion criteria (n=854) were eligible to participate in our study.
Using the same procedure as described above, an advertisement for the follow-up survey was posted to MTurk in May 2018 and was visible only to workers who met the inclusion criteria. Our target recruitment was 50 individuals per group, based on findings from a pilot test of the feasibility of recruiting populations with self-reported mental and physical illness using MTurk. Workers who clicked on the advertisement for the follow-up survey were given a brief description of the survey and a weblink to the 175-item survey hosted on Qualtrics. Electronic consent was obtained before a participant began the survey. Four attention check questions were included, and no participants failed the attention check questions. This survey took an average of 21 minutes, 46 seconds to complete (SD = 13 minutes, 48 seconds). Respondents were paid $8 for completing the survey. 210 HIT assignments were posted in 7 installments onto MTurk. A total of 151 of these assignments were completed, resulting in a 71.9% “completion” rate. The final participant group consisted of 50 individuals in good health, 50 individuals living with mental illness and/or substance use disorder, and 51 individuals living with physical illness.
Procedures.
The main survey was developed based on prior work to comprehensively understand various ethical issues related to research.18 The instrument was composed of multiple questionnaires to address a range of topics including (but not limited to) general health, optimism/pessimism, safeguards in research, and general research attitudes. In addition, the survey included four attention checks; no respondent failed the attention check questions.
Research safeguard procedures: perceived protectiveness and safeguard-specific willingness to participate.
Each of five research safeguard procedures was described by a brief paragraph (see Appendix 1). After the description of each safeguard, respondents were asked to rate the protectiveness of each safeguard, i.e., “How much does [safeguard procedure X] protect people who participate in research?” Responses were rated on a 5-point Likert scale (i.e., 1 = “Does not protect at all”; 5 = “Highly protects”).
Respondents were then asked to rate how the presence of each safeguard would influence their willingness to participate in research, i.e., “If you knew that a project [had safeguard procedure X], how would it influence your willingness to participate in the project?” Responses were rated on a five-point scale (i.e., 1 = “Does not influence my willingness to participate at all”; 5= “I’d be very much more willing to participate”).
Optimism.
Respondents completed the 10-item Life Orientation Test-Revised (LOT-R), in which three items assess optimism, three measure pessimism, and four serve as fillers.19 Each item is rated on a five-point Likert scale (i.e., 1 = “strongly disagree” to 5 = “strongly agree”). For this analysis, we examined the three items assessing optimism, as those items were most salient to our hypotheses. Thus, the total score on these three items could range from 3 to 15. We evaluated the distribution of the continuous optimism subscale and chose a score of 12 as a cut off for two reasons: 1) a score of 12 was the cut off for the 75th percentile (3rd quartile score), which we felt was an appropriate threshold for distinguishing the very optimistic individuals from the less optimistic individuals; 2) a score of 12 also indicated that an individual, on average, selected a score of 4 (out of 5) or higher for each of the 3 optimism items.
Attitudes toward medical research.
The Research Attitudes Questionnaire (RAQ-7) was used to assess respondents’ attitudes toward biomedical research.20 Respondents rated each of the seven items on a five-point Likert scale and a total score was calculated (possible range: 7 to 35; higher scores reflect more positive attitudes toward research).
Data analysis.
Descriptive statistics were calculated for sociodemographic characteristics for each optimism group (i.e., weak to moderate optimism vs. strong optimism), using means and standard deviations. One-way ANOVA, chi square tests, and Fisher exact tests were used to evaluate differences between the two optimism groups on these variables.
Differences between the two optimism groups in mean responses to survey items of interest (i.e., research attitudes, perceived protectiveness of research safeguards, and willingness to participate given the presence of each safeguard) were analyzed using paired t tests. Finally, we evaluated the influence of optimism on participation willingness adjusting for safeguard type, perceived protectiveness, and other potential confounders using generalized estimating equations. We took responses of multiple items (participation willingness for each of the 5 safeguards) as a vector outcome of the GEE model. Each individual contributed 5 outcomes in this multivariate model. The variables in the model included safeguard type (i.e., a vector of binary variables indicated whether the safeguard was a mandated safeguard and always present, versus not always present), perceived protectiveness, health status group, and optimism level as covariates. We report coefficient estimates and confidence intervals drawn from the final analytic model. GEE was used with a Gaussian link and unstructured correlation structure. Statistical analyses were performed in R software (R Version 3.0.0, GNU project).
RESULTS
Characteristics of study population.
As shown in Table 1, respondents had a mean age of 38.3 years (SD 13.6). Slightly more than half (57%) were male. By self-report, the majority of respondents were not Hispanic or Latino (92%), and most were white (69%). There was a significant association between optimism level and self-reported health status group (P value = .01).
Table 1.
Background characteristics of participants, by level of optimism
| Weak to Moderate Optimism1 (n = 89) |
Strong Optimism (n = 62) |
Overall (N = 151) |
P value2 | |
|---|---|---|---|---|
| Mean (SD) | Mean (SD) | Mean (SD) | ||
| Age (years) | 38.7 (13.4) | 37.6 (13.9) | 38.3 (13.6) | 0.62 |
| % (n) | % (n) | % (n) | ||
| Gender | ||||
| Male | 56.2 (50) | 58.1 (36) | 57.0 (86) | 0.89 |
| Female | 41.6 (37) | 38.7 (24) | 40.4 (61) | |
| Ethnicity | ||||
| Not Hispanic or Latino | 94.4 (84) | 88.7 (55) | 92.1 (139) | 0.23 |
| Hispanic or Latino | 5.6 (5) | 11.3 (7) | 7.9 (12) | |
| Race | ||||
| White | 69.7 (62) | 67.7 (42) | 68.9 (104) | 0.86 |
| Other | 29.2 (26) | 32.3 (20) | 30.5 (46) | |
| Education level | ||||
| Bachelor’s degree or higher | 70.8 (63) | 56.5 (35) | 58.3 (88) | 0.1 |
| < Bachelor’s degree | 40.4 (36) | 43.5 (27) | 41.7 (63) | |
| Employed in the health field | ||||
| No | 89.9 (80) | 83.9 (52) | 87.4 (132) | 0.29 |
| Yes | 9.1 (8) | 16.1 (10) | 11.9 (18) | |
| Illness group | ||||
| Mental illness and/or substance use disorder | 42.7 (38) | 19.4 (12) | 33.1 (50) | 0.01 |
| Physical Illness | 30.3 (27) | 38.7 (24) | 33.8 (51) | |
| Healthy | 27.0 (24) | 41.9 (26) | 33.1 (50) |
Weak-to-moderate optimism: LOT-R optimism items < 12; Strong optimism: LOT-R optimism items ≥ 12
Test statistics: t test (age); chi-square for all other characteristics
Research attitudes by optimism level.
As shown in Table 2, the overall sample expressed moderate to strong agreement on each of the RAQ-7 items (range of means = [3.1-4.1]). For example, respondents, on average, strongly agreed that “society needs to devote more resources to medical research” (overall mean 4.1; SD 0.9). On average, respondents expressed neutral endorsement of the item “We all have some responsibility to help others by volunteering for medical research” (overall mean 3.1; SD 1.1).
Table 2.
Participants’ responses on Research Attitudes Questionnaire items, by optimism level
| Weak to Moderate Optimism (n = 89) |
Strong Optimism (n = 62) |
Overall (N = 151) |
P value1 | |
|---|---|---|---|---|
| Mean (SD) | Mean (SD) | Mean (SD) | ||
| Responsibility to help others by volunteering for medical research | 2.8 ( 1.1 ) | 3.4 ( 1.0 ) | 3.1 ( 1.1 ) | 0.001 |
| Personal information kept private and confidential | 3.4 ( 1.0 ) | 3.8 ( 1.0 ) | 3.6 ( 1.0 ) | 0.005 |
| Participating in medical research is generally safe | 3.5 ( 0.8 ) | 4.0 ( 0.8 ) | 3.7 ( 0.8 ) | 0.001 |
| Medical research can be trusted to protect the interests of participants | 3.6 ( 0.8 ) | 4.0 ( 0.9 ) | 3.8 ( 0.9 ) | 0.003 |
| Positive view about medical research in general | 3.7 ( 0.8 ) | 4.1 ( 0.8 ) | 3.9 ( 0.8 ) | 0.007 |
| Medical research will find cures for many major diseases | 3.7 ( 0.9 ) | 4.1 ( 0.7 ) | 3.9 ( 0.8 ) | 0.003 |
| More resources devoted to medical research | 4.0 ( 0.9 ) | 4.3 ( 0.8 ) | 4.1 ( 0.9 ) | 0.04 |
| RAQ Total Score (Actual Score Range: [13,35]) | 24.8 (4.1) | 27.8 (4.4) | 26.0 (4.5) | <0.001 |
Notes:
Test statistic: t test.
RAQ = Research Attitudes Questionnaire. Participants were asked to report their attitudes towards biomedical research. Each item was rated on a 1 to 5 scale (1 = “strongly disagree”, 2 = “agree”, 3 = “neutral”, 4 = “agree”, 5 = “strongly agree”). Total possible score range: [5,35].
Compared to individuals with weak-to-moderate optimism, individuals with strong optimism agreed more strongly with each RAQ-7 item (all P values statistically significant, including RAQ-7 total score).
Perceived protectiveness of safeguard procedures by optimism level.
As shown in Table 3, respondents, overall, perceived all five of the research safeguard procedures as “somewhat” to “quite a bit” protective (range of means = 3.2 to 4.0). Respondents, on average, perceived Institutional Review Board (IRB) review (overall mean = 4.0; SD 0.8) and Data Safety Monitoring Board (DSMB) oversight (overall mean = 4.0; SD 0.9) as “quite a bit” protective. On average, respondents perceived alternative decision makers as “somewhat” protective (overall mean = 3.2; SD 1.0). A similar pattern was found within each of the optimism groups.
Table 3.
Participants’ perceptions of protectiveness of research safeguards, by optimism level
| Weak to Moderate Optimism (n = 89) |
Strong Optimism (n = 62) |
Overall N = 151 |
P value1 | |
|---|---|---|---|---|
| Mean (SD) | Mean (SD) | Mean (SD) | ||
| Informed Consent | 3.3 ( 1.1 ) | 3.8 ( 1.1 ) | 3.5 ( 1.1) | 0.001 |
| Confidential Coding of Data | 3.6 ( 1.0 ) | 4.0 ( 1.0 ) | 3.7 ( 1.0 ) | 0.02 |
| Institutional Review Board | 3.9 ( 0.8 ) | 4.2 ( 0.8 ) | 4.0 ( 0.8) | 0.08 |
| Alternative Decision Maker | 3.1 ( 0.9 ) | 3.3 ( 1.2 ) | 3.2 ( 1.0 ) | 0.15 |
| Data Safety Monitoring Board | 3.9 ( 0.9 ) | 4.0 ( 0.9 ) | 4.0 ( 0.9 ) | 0.51 |
Notes:
Test statistic: t test.
Participants were asked, “How much does this [safeguard procedure X] protect people who participate in research?”. Each item was rated on a 1 to 5 scale (i.e., 1 = “does not protect at all”; 2 = “only protects a little bit”; 3 = “somewhat protects”; 4 = “protects quite a bit”; 5 = “highly protects”).
Compared to individuals with weak-to-moderate optimism, individuals with strong optimism perceived informed consent as more protective (mean = 3.3 vs. 3.8; P value = 0.001). Similarly, compared to individuals with weak-to-moderate optimism, individuals with strong optimism perceived confidential coding as more protective (mean = 3.6 vs. 4.0; P value = 0.02).
Influence of research safeguards on willingness to participate in research, by optimism level.
As shown in Table 4, respondents, overall, perceived all five of the research safeguard procedures to be “somewhat” to “quite a bit” influential on their willingness to participate (range of means = 2.9 to 3.9). Respondents, on average, perceived IRB review (overall mean = 3.9; SD 1.0) or DSMB oversight (overall mean = 3.8; SD 1.1) as “quite a bit” influential on their willingness to participate. On average, respondents perceived alternative decision makers as “somewhat” influential (overall mean = 2.9; SD 1.3). A similar pattern was found within each of the optimism groups.
Table 4.
Influence of research safeguards on willingness to participate in research, by optimism level
| Weak to Moderate Optimism (n = 89) |
Strong Optimism (n = 62) |
Overall (N = 151) |
P value1 | |
|---|---|---|---|---|
| Mean (SD) | Mean (SD) | Mean (SD) | ||
| Informed Consent | 3.0 ( 1.2 ) | 3.8 ( 1.2 ) | 3.3 ( 1.3 ) | <0.001 |
| Alternative Decision Maker | 2.6 ( 1.2 ) | 3.3 ( 1.3 ) | 2.9 ( 1.3 ) | <0.001 |
| Confidential Coding of Data | 3.3 ( 1.2 ) | 3.9 ( 1.1 ) | 3.5 ( 1.2 ) | 0.002 |
| Institutional Review Board | 3.7 ( 1.1 ) | 4.1 ( 0.9 ) | 3.9 ( 1.0 ) | 0.03 |
| Data Safety Monitoring Board | 3.7 ( 1.1 ) | 4.0 ( 1.1 ) | 3.8 ( 1.1 ) | 0.19 |
Notes:
Test statistic: t test.
Participants were asked, “If you knew that a project [had safeguard procedure X], how would it influence your willingness to participate in the project?”. Each item was rated on a 1 to 5 scale (i.e., 1 = “does not influence my willingness to participate at all;” 2 = “I’d be a little bit more willing to participate;” 3 = “I”d be somewhat more willing to participate;” 4 = “I’d be quite a bit more willing to participate;” 5= “I’d be very much more willing to participate.”)
Compared to individuals with weak-to-moderate optimism, individuals with strong optimism perceived informed consent, alternative decision makers, confidential coding, and IRB review as more positively influential on their willingness to participate (all four P values statistically significant).
Results of GEE analysis.
Optimism effect:
Perceived protectiveness was significantly positively associated with mean participation willingness (Figure 1), controlling for health status group and safeguard type. In particular, adjusting for confounders, participants with strong optimism consistently had a higher mean overall participation willingness rating as compared to participants with weak-to-moderate optimism (B = 0.28, P value = 0.002).
Figure 1. Willingness to participate is positively associated with perceived protectiveness of safeguard and differs by self-reported optimism ratings.
Figure 1 presents a graphical depiction of the association between expressed willingness to participate in research and the perceived protectiveness of safeguards. The X-axis refers to ratings of perceived protectiveness of safeguards. The red line represents the relationship for participants who expressed strong optimism, and the black line represents the relationship for participants who expressed weak-to-moderate optimism.
Perceived protectiveness effect:
Controlling for confounders, for every 1 point increase in perceived protectiveness, there was a 0.73 point increase in mean overall participation willingness (B = 0.73, P value = <0.001).
Safeguard effect:
Controlling for confounders, participants had a 0.36 point lower mean overall participation willingness in the context of the alternate decision makers safeguard as compared to IRB review (B = −0.36, P value = 0.001). There were no other significant safeguard effects on overall mean participation willingness.
Demographic effects:
Controlling for confounders, there was no significant health status group effect on mean participation willingness.
DISCUSSION
Utilizing an online sample of respondents of differing self-reported health status, this pilot study explored associations between dispositional optimism and attitudes toward medical research, perceptions of the protectiveness of five core research safeguards, and willingness to participate in research involving these safeguards. The present findings add needed data to the limited literature on the factors that may initially bring people to the research situation and influence their research perspectives and willingness to participate.
A number of key findings emerged. First, respondents in the mental illness group were more likely to report weak-to-moderate optimism, while participants in good health were more likely to report being strongly optimistic. This finding, while not surprising, is consistent with an extensive literature on the association between optimism and various health outcomes. Optimism has demonstrated significant associations with numerous health outcomes, including physical and psychological well-being and health-related quality of life.21 However, this literature is not entirely consistent.22 The topic of to what degree, and why, optimism—as well as the absence of pessimism—is associated with health outcomes, has been the focus of a recent meta-analysis, which found that better health outcomes were more robustly associated with the absence of pessimism than with optimism.23 Moreover, the potential behavioral and biological mechanisms that underlie these associations continue to be the focus of numerous investigations.24
Second, optimistic respondents held more positive overall research attitudes, which to our knowledge is a novel finding. While one prior study examined the association of dispositional optimism with expectations of therapeutic benefit in cancer clinical trials, the individuals in that study were already enrolled in oncology trials, and their general attitudes toward medical research were not assessed.25 Overall, individuals with higher optimism expect good outcomes across multiple life domains, which is consistent with the present finding.26 In inspecting the individual items on the RAQ-7, it also makes sense that optimistic individuals would endorse items conveying the need to engage actively in research, as optimism has been shown to be positively associated with various types of engagement and problem-focused coping.27 In other words, research engagement may be related to specific modes of coping, which are more likely to be utilized by individuals with higher levels of optimism.
Third, while respondents overall perceived the described safeguard procedures as “somewhat” to “quite a bit” protective, some differences between optimism groups were found for specific safeguards. In particular, strongly optimistic respondents saw informed consent and confidential coding as more protective than did respondents with weak-to-moderate optimism. In both the total sample population and the population with weak to moderate optimism, respondents viewed the informed consent safeguard as less than somewhat protective. For this safeguard, an endorsement of strong optimism appeared to be the factor that “tipped” respondents into perceiving the safeguard as more than somewhat protective. Optimists’ overall positive expectancies may be reflected in these views of protectiveness; however, further research in larger and more diverse samples is needed to examine whether specific safeguards, in the context of specific types of research, are viewed differently by individuals with different underlying traits. One implication of this result is how researchers may engage and support individuals in sharing information about safeguards. The principle of respect for persons in clinical research requires, at a minimum, that researchers ensure an adequate informed consent process. In addition to this minimum standard, a broader view of ethical recruitment and consent practices could involve research teams providing potential participants with a more comprehensive picture of safeguards, including researchers’ intentions.
Fourth, respondents with strong optimism perceived four of the five safeguard procedures (all except DSMB oversight) as more positively influential on their willingness to participate than did respondents with weak-to-moderate optimism. One explanation of positive future expectancies among optimists relates to the belief that “other people (will) look out for you”, which may offer a plausible explanation for the positive influence of IRB review and alternative decision makers on willingness to participate.28 Whether the dampened enthusiasm for DSMB oversight is related to opinions regarding the function and role of the DSMBs or whether it was related to level of understanding (i.e., in terms of how this safeguard was described) was not examined, but could be further probed.
Finally, even after adjusting for self-reported health status, type of research safeguard, and perceived safeguard protectiveness, optimism was significantly positively associated with expressed willingness to participate in clinical research. This novel finding suggests that strongly optimistic individuals may be more willing to participate in research than those with low-to-moderate levels of optimism. We found that strong dispositional optimism even, in one case, made the difference between a weak and a strong view of safeguard protectiveness. Given that perceptions have been shown to positively influence hypothetical willingness to participate in research, this represents an opportunity to engage with prospective research volunteers in ensuring their support, and to consider further how to strengthen engagement and clarify intentions for those individuals who may not perceive research as very protective.
The ethical implications of lower versus higher levels of dispositional optimism in clinical research, however, warrant further exploration. Our findings indicate that level of dispositional optimism may be associated with multiple factors that bring people to the research situation and influence their research perspectives and willingness to participate (i.e., attitudes toward research, perceptions regarding research safeguards, and willingness to participate in clinical research). If level of dispositional optimism is a “hidden” influence on research participants’ overall assessments and appreciation of research, then it would be important to understand whether individuals with lower levels of optimism (including, for example, individuals with mental illness) are less inclined to participate in clinical research. Given that the present study did not evaluate for other personality factors that may be associated with research perspectives, the present finding should be consider preliminary. From a methodological perspective, whether personality and other psychological traits influence perceptions of research could affect recruitment and potentially bias samples in ways that may go unrecognized, possibly affecting study findings. From an ethics perspective, research participants should be drawn as fairly as possible from the potential volunteer population; both under- and overrepresentation may be ethically problematic.
Limitations of this pilot study include the use of self-reported health status, as well as the relatively homogeneous sample in terms of race and ethnicity. In addition, the scale used to assess willingness to participate in the context of each safeguard procedure did not inquire about lower likelihood of participation, possibly biasing these findings toward greater participation willingness. Our use of a sample of MTurk workers may limit the generalizability of our findings, as MTurk workers are not representative of the US population. Other limitations with nonprobability samples, and online workers in particular, have been addressed.29 While not without its limitations, the MTurk platform enabled the recruitment of a diverse sample in a short time frame for this study.
Another methodological limitation is the use of the dichotomous variable (i.e., weak-to-moderate vs. strong) when assessing optimism. While other research teams have dichotomized different versions of this scale, there remains active debate regarding how best to utilize and analyze this scale in the assessment of optimism and pessimism.30 While our cut off is based on exploration of our data, this approach would lead to a different threshold for a different dataset, and thus further work should be done in order to determine a useful and interpretable threshold value that could be widely used.
Our novel findings may help flesh out a relatively underexplored dimension of research engagement and enrollment—namely, the association of underlying, individual traits with ethically-salient perceptions of research, including perceptions of the protectiveness of research safeguards and willingness to participate. Given the importance of individual traits to coping and multiple aspects of well-being, these personal characteristics have significance in how individuals engage—or do not engage—with clinical research relevant to their own or others’ health and well-being. While the ethical conduct of clinical research depends upon the effective implementation of a range of research safeguards, designed, adopted, and refined over time, it is important to note the interindividual heterogeneity in how such safeguards may be perceived, as well as how these perceptions may, in turn, influence participants’ research engagement.31
Further research into how individual traits relate to perceptions of research, as well as to crucial decisions such as whether or not to participate, may illuminate vulnerabilities and strengths that have previously gone unexamined. Further, we recognize that it is insufficient to examine optimism in isolation as a factor in research volunteer decisions, as humans’ decision-making processes are complex and involve multiple factors considered jointly.32 Multivariable analytic strategies, such as the approach we undertook in the current study, are needed to take into account the various potential factors that influence research decision-making. With many potential factors under consideration, analytic strategies focused on reducing the dimensionality of the factors being studied are also needed.
Supplementary Material
Acknowledgements:
The authors acknowledge the contributions of Max Kasun for his valuable assistance in the production of this article.
Funding:
This research study was supported by grant NIMH R01 MH114856 from the National Institute of Mental Health.
Footnotes
Disclosures: Dr. Roberts serves as the Editor-in-Chief of the journal Academic Medicine. The other authors have no conflicts of interest.
Human Subjects Protection Statement: This study was approved by the IRB at Stanford University.
REFERENCES
- 1.Dunn LB, Candilis PJ, Roberts LW, “Emerging Empirical Evidence on the Ethics of Schizophrenia Research,” Schizophrenia Bulletin, 32, no. 1 (2006): 47–68; [DOI] [PMC free article] [PubMed] [Google Scholar]; Kim SY, “Evidence-Based Ethics for Neurology and Psychiatry Research,” NeuroRx, 1, no. 3 (2004): 372–377; [DOI] [PMC free article] [PubMed] [Google Scholar]; Roberts L, Roberts B, “Psychiatric Research Ethics: An Overview of Evolving Guidelines and Current Ethical Dilemmas in the Study of Mental Illness,” Biological Psychiatry, 46 (1999): 1025–1038; [DOI] [PubMed] [Google Scholar]; Roberts LW, “Informed Consent and the Capacity for Voluntarism,” American Journal of Psychiatry, 159 (2002): 705–712. [DOI] [PubMed] [Google Scholar]
- 2.Roberts LW, “Informed Consent and the Capacity for Voluntarism.”; [DOI] [PubMed] [Google Scholar]; Appelbaum PS, Grisso T, MacCAT-CR: MacArthur Competence Assessment Tool for Clinical Research (Sarasota, FL: Professional Resource Press, 2001); [Google Scholar]; Lidz CW, Appelbaum PS, “The Therapeutic Misconception: Problems and Solutions,” Medical Care, 40 (2002): V55–63. [DOI] [PubMed] [Google Scholar]
- 3.Dunn LB, Candilis PJ, Roberts LW, “Emerging Empirical Evidence on the Ethics of Schizophrenia Research.”; [DOI] [PMC free article] [PubMed] [Google Scholar]; Marson DC, et al. , “Cognitive Models that Predict Physician Judgments of Capacity to Consent in Mild Alzheimer’s Disease,” Journal of the American Geriatrics Society, 45 (1997): 458–464; [DOI] [PubMed] [Google Scholar]; Dunn LB, Jeste DV, “Enhancing Informed Consent for Research and Treatment,” Neuropsychopharmacology, 24, no. 6 (2001): 595–607; [DOI] [PubMed] [Google Scholar]; Flory J, Emanuel E, “Interventions to Improve Research Participants’ Understanding in Informed Consent for Research: A Systematic Review,” JAMA, 292 (2004): 1593–1601; [DOI] [PubMed] [Google Scholar]; Leykin Y, et al. , “Participants’ Perceptions of Deep Brain Stimulation Research for Treatment-Resistant Depression: Risks, Benefits, and Therapeutic Misconception,” AJOB Primary Research, 2 (2001): 33–41; [DOI] [PMC free article] [PubMed] [Google Scholar]; Roberts LW, Kim JP, “Healthy Individuals’ Perspectives on Clinical Research Protocols and Influences on Enrollment Decisions,” AJOB Empirical Bioethics, 8 (2017): 89–98; [DOI] [PubMed] [Google Scholar]; Weinfurt KP, et al. , “The Correlation Between Patient Characteristics and Expectations of Benefit from Phase I Clinical Trials,” Cancer, 98 (2003): 166–175. [DOI] [PubMed] [Google Scholar]
- 4.Murthy VH, Krumholz HM, Gross CP, “Participation in Cancer Clinical Trials: Race-, Sex-, and Age-Based Disparities,” JAMA, 291 (2004): 2720–6. [DOI] [PubMed] [Google Scholar]
- 5.Trauth JM, et al. , “Public Attitudes Regarding Willingness to Participate in Medical Research Studies,” Journal of Health and Social Policy, 12 (2000): 23–43. [DOI] [PubMed] [Google Scholar]
- 6.Roberts LW, et al. , “Do human Subject Safeguards Matter to Potential Participants in Psychiatric Genetic Research?” Journal of Psychiatric Research, 116 (2019): 95–103; [DOI] [PMC free article] [PubMed] [Google Scholar]; Roberts LW, et al. , “Influence of Ethical Safeguards on Research Participation: Comparison of Perspectives of People with Schizophrenia and Psychiatrists,” American Journal of Psychiatry, 161 (2004): 2309–2311; [DOI] [PubMed] [Google Scholar]; Roberts LW, et al. , “Schizophrenia Research Participants’ Responses to Protocol Safeguards: Recruitment, Consent, and Debriefing,” Schizophrenia Research, 67 (2004): 283–291. [DOI] [PubMed] [Google Scholar]
- 7.Roberts LW, et al. , “Influence of Ethical Safeguards on Research Participation: Comparison of Perspectives of People with Schizophrenia and Psychiatrists,” American Journal of Psychiatry, 161 (2004): 2309–2311; [DOI] [PubMed] [Google Scholar]; Roberts LW, et al. , “Do human Subject Safeguards Matter to Potential Participants in Psychiatric Genetic Research?” Journal of Psychiatric Research, 116 (2019): 95–103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Roberts LW, et al. , “Do Human Subject Safeguards Matter to Potential Participants in Psychiatric Genetic Research?” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward,” American Psychologist, 2018;73(9):1082–1094; [DOI] [PMC free article] [PubMed] [Google Scholar]; Carver CS, Connor-Smith J, “Personality and Coping,” Annual Review of Psychology, 61 (2010): 679–704; [DOI] [PubMed] [Google Scholar]; Scheier MF, Carver CS, “Optimism, Coping, and Health: Assessment and Implications of Generalized Outcome Expectancies,” Health Psychology, 4 (1985) 219–247. [DOI] [PubMed] [Google Scholar]
- 10.Jansen LA, et al. , “Variations in Unrealistic Optimism Between Acceptors and Decliners of Early Phase Cancer Trials,” Journal of Empirical Research on Human Research Ethics, 12, no. 4 (2017): 280–288; [DOI] [PMC free article] [PubMed] [Google Scholar]; Cheng A, Zamarro G, Orriens B, “Personality as a Predictor of Unit Nonresponse in an Internet Panel,” Sociological Methods & Research, 49, no. 3 (2020): 672–698. [Google Scholar]
- 11.Carver CS, Connor-Smith J, “Personality and Coping.”; [DOI] [PubMed] [Google Scholar]; Scheier MF, Carver CS, “Optimism, Coping, and Health: Assessment and Implications of Generalized Outcome Expectancies.” [DOI] [PubMed] [Google Scholar]
- 12.Jansen LA, et al. , “Dispositional Optimism and Therapeutic Expectations in Early-Phase Oncology Trials,” Cancer, 122 (2016): 1238–1246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lidz CW, et al. , “Therapeutic Misconception and the Appreciation of Risks in Clinical Trials,” Social Science and Medicine, 58 (2004): 1689–1697; [DOI] [PubMed] [Google Scholar]; Jansen LA, et al. , “Dispositional Optimism and Therapeutic Expectations in Early-Phase Oncology Trials.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Carver CS, Scheier MF, “Dispositional Optimism,” Trends in Cognitive Sciences, 18, no. 6 (2014): 293–299; [DOI] [PMC free article] [PubMed] [Google Scholar]; Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jansen LA, et al. , “Dispositional Optimism and Therapeutic Expectations in Early-Phase Oncology Trials” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gillan CM, Daw ND, “Taking Psychiatry Research Online,” Neuron, 91, no. 1 (2016): 19–23. [DOI] [PubMed] [Google Scholar]
- 18.Roberts LW, Kim JP, “Healthy Individuals’ Perspectives on Clinical Research Protocols and Influences on Enrollment Decisions.”; [DOI] [PubMed] [Google Scholar]; Roberts LW, et al. , “Influence of Ethical Safeguards on Research Participation: Comparison of Perspectives of People with Schizophrenia and Psychiatrists.”; [DOI] [PubMed] [Google Scholar]; Roberts LW, Kim JP, “Giving Voice to Study Volunteers: Comparing Views of Mentally Ill, Physically Ill, and Healthy Protocol Participants on Ethical Aspects of Clinical Research,” Journal of Psychiatric Research, 56 (2014): 90–97; [DOI] [PMC free article] [PubMed] [Google Scholar]; Roberts LW, et al. , “Patient and Psychiatrist Ratings of Hypothetical Schizophrenia Research Protocols: Assessment of Harm Potential and Factors Influencing Participation Decisions,” American Journal of Psychiatry, 159 (2002): 573–584. [DOI] [PubMed] [Google Scholar]
- 19.Scheier MF, Carver CS, Bridges MW, “Distinguishing Optimism from Neuroticism (and Trait Anxiety, Self-mastery, and Self-esteem): A Reevaluation of the Life Orientation Test,” Journal of Personality and Social Psychology, 67 (1994): 1063–1078. [DOI] [PubMed] [Google Scholar]
- 20.Rubright JD, et al. , “Measuring How People View Biomedical Research: Reliability and Validity Analysis of the Research Attitudes Questionnaire,” Journal of Empirical Research on Human Research Ethics, 6 (2011): 63–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward.”; [DOI] [PMC free article] [PubMed] [Google Scholar]; Carver CS, Scheier MF, “Dispositional Optimism.”; [DOI] [PMC free article] [PubMed] [Google Scholar]; Roberts BW, et al. , “The Power of Personality: The Comparative Validity of Personality Traits, Socioeconomic Status, and Cognitive Ability for Predicting Important Life Outcomes,” Perspectives on Psychological Science, 2, no. 4 (2007): 313–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Scheier MF, et al. , “Optimism Versus Pessimism as Predictors of Physical Health: A Comprehensive Reanalysis of Dispositional Optimism Research.” American Psychologist, 76, no. 3 (2021): 529–548. [DOI] [PubMed] [Google Scholar]
- 24.Scheier MF, Carver CS, “Dispositional Optimism and Physical Health: A Long Look Back, A Quick Look Forward.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Jansen LA, et al. , “Dispositional Optimism and Therapeutic Expectations in Early-Phase Oncology Trials.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Carver CS, Connor-Smith J, “Personality and Coping.” [DOI] [PubMed] [Google Scholar]
- 27.Carver CS, Connor-Smith J, “Personality and Coping.” [DOI] [PubMed] [Google Scholar]
- 28.Carver CS, Scheier MF, “Dispositional Optimism.” [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Chandler J, Shapiro D., “Conducting Clinical Research Using Crowdsourced Convenience Samples,” Annual Review of Clinical Psychology, 12 (2016): 53–81. [DOI] [PubMed] [Google Scholar]
- 30.Lynch JW, Kaplan GA, Shema SJ, “Cumulative Impact of Sustained Economic Hardship on Physical, Cognitive, Psychological, and Social Functioning,” New England Journal of Medicine, 337, no. 26 (1997): 1889–1895; [DOI] [PubMed] [Google Scholar]; Heinonen E, et al. , “Dispositional Optimism as Predictor of Outcome in Short- and Long-Term Psychotherapy,” Psychology and Psychotherapy: Theory, Research and Practice, 90, no. 3 (2017): 279–298; [DOI] [PubMed] [Google Scholar]; Scheier MF, et al. , “Optimism Versus Pessimism as Predictors of Physical Health: A Comprehensive Reanalysis of Dispositional Optimism Research.” [DOI] [PubMed] [Google Scholar]
- 31.Sugarman J “Examining Provisions Related to Consent in the Revised Common Rule,” American Journal of Bioethics, 17 (2017): 22–26. [DOI] [PubMed] [Google Scholar]
- 32.Kahneman D, Tversky A, “Prospect Theory: An Analysis of Decision under Risk,” Econometrica, 47, no. 2 (1979): 263; [Google Scholar]; Tversky A, Kahneman D, “Rational Choice and the Framing of Decisions,” The Journal of Business, 59, no. 4 (1986): 251–278. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.