Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: J Behav Med. 2021 Sep 4;45(1):103–114. doi: 10.1007/s10865-021-00253-7

The Role of Future-Oriented Affect in Engagement with Genomic Testing Results

Arielle S Gillman 1, Irina A Iles 2, William M P Klein 2, Barbara B Biesecker 3, Katie L Lewis 4, Leslie G Biesecker 4, Rebecca A Ferrer 1
PMCID: PMC8821111  NIHMSID: NIHMS1741596  PMID: 34480685

Abstract

Background.

Future-oriented emotions such as anticipatory affect (i.e., current affect experienced regarding a potential future outcome) and anticipated affect (i.e., expectations about potential future affect), are uniquely associated with health decision-making (e.g., electing to receive results of genomic testing).

Purpose.

This study investigated the degree to which negative anticipated and anticipatory emotions predict health decision making over time, and whether such emotions predict social, emotional, and behavioral responses to anticipated information (e.g., genomic testing results).

Methods.

461 participants (M age=63.9, SD=5.61, 46% female) in a genomic sequencing cohort who elected to receive genomic sequencing (carrier) results were included in the current study. Anticipated and anticipatory affect about sequencing results were assessed at baseline. Psychological and behavioral responses to sequencing results, including participants’ reported anxiety, decisional conflict, and distress about sequencing results, whether they shared results with family members, and their intentions to continue learning results in the future, were collected immediately, one month, and/or six months after receiving results.

Results.

More negative anticipated and anticipatory affect at baseline was significantly and independently associated with lower intentions to continue learning results in the future, as well as higher levels of anxiety and uncertainty at multiple time points after receiving results. Anticipated negative affect was also associated with greater decisional conflict, and anticipatory negative affect was also associated with greater distress after receiving results.

Conclusions.

Future-oriented emotions may play an important role in decisions that unfold over time, with implications for genomic testing, behavioral medicine, and health decision-making broadly.

Keywords: Future-oriented emotions, genomic testing results, affect, health decision-making, genomic sequencing

Rapid advances in genomic sequencing technology have made it increasingly possible and relatively inexpensive for individuals to be tested for genetic predispositions to a variety of heritable disease states (Frieser et al., 2018). Genomic testing results can contain not only information relevant to one’s own personal risk, but also to one’s family members (e.g., siblings, offspring; B. B. Biesecker et al., 2018). Obtaining sequencing results is a potentially valuable avenue through which individuals can learn about, and subsequently take medical or behavioral steps to manage, their personal risk (Frieser et al., 2018) and the risk to their relatives (after communication of results; B. B. Biesecker et al., 2018). Thus, examining factors that are associated with decisions to learn this information – initially and over time – is a pertinent avenue for research.

Behavioral science has increasingly recognized the importance of affective influences on decision making, including health decision-making (Ferrer et al., 2014; E. Peters et al., 2006). In the current context, although sequencing results may contain useful information for managing one’s risk, such results may also be affectively-laden; for example, learning that one is at increased risk for a serious disease may evoke fear or other negative affective states (Ferrer, Green, et al., 2015). Moreover, sequencing results are often laden with uncertainty and ambiguity, which can also lead to negative affective states (Han et al., 2006). The expectation that health-related information may provoke negative affect in the future can lead individuals to perceive this information as threatening (Howell & Shepperd, 2012). This may be true even for sequencing results that do not pertain directly to the self (i.e., carrier results); learning that one may have passed on a negative health outcome to one’s offspring may also be perceived as threatening, because such results may cause feelings of guilt or other negative affective states (Persky et al., 2015). Such expectations can influence decisions to receive such information, and may lead to responses such as information avoidance (Ferrer, Taber, et al., 2015).

Because people might predict that genetic testing results will increase negative affect, futureoriented emotions may be consequential contributors to decision making in this context.

Future-oriented emotions fall into two distinct categories (Baumgartner et al., 2008; Loewenstein & Lerner, 2003). The first category is anticipatory emotions, which are emotions experienced in the present when thinking about an outcome that may occur in the future (Baumgartner et al., 2008) (these have also been described as “current” or “immediate” emotions; Ferrer, Taber, et al., 2015; Loewenstein & Lerner, 2003). Experiencing worry in the present in anticipation of the potential to receive negative health information in the future is an example of an anticipatory emotion. The second category is anticipated emotions, which are not experienced in the present, but rather, are expected to occur in the future if a given (positive or negative) outcome occurs. For example, people might expect that they will be unable to cope if they were to learn negative health information in the future, irrespective of whether they experience worry in the present.

Anticipated and anticipatory emotions each explain independent variance in a variety of decisions (Baumgartner et al., 2008), including in the context of health (Chapman & Coups, 2006; Ferrer, Taber, et al., 2015; Schlösser et al., 2013). A recent meta-analysis examining the role of anticipated versus anticipatory emotions on health decisions found that across studies, anticipated emotions had moderate-sized associations with health intentions (r=0.38) and behavior (r=0.48), whereas anticipatory emotions had smaller associations with intentions (r=.25) and behavior (r=0.18; Xu & Guo, 2019). Consistent with this literature on future-oriented emotions in health decision-making broadly, people use future-oriented affective information to guide (at least initial) decisions to receive sequencing results specifically (Ferrer, Taber, et al., 2015). Ferrer and colleagues demonstrated, using cross-sectional data from the baseline wave of the current longitudinal study, that anticipated negative affect was associated with lower intentions to obtain and share genomic sequencing results at the beginning of a genomic sequencing trial, whereas the association of anticipatory affect (worry) and these outcomes only trended in the expected direction (Ferrer, Taber, et al., 2015).

Although relying on future-oriented emotions to make decisions is common, it may lead to unnecessary avoidance of potentially useful health information, including genetic testing results. Learning that one is not at higher risk for disease can alleviate worry and provide relief. Further, people may underestimate the fact that experiencing uncertainty (i.e., through avoidance of health information) is a negative affective experience in itself (Han et al., 2009), and learning even negative health information can alleviate uncertainty-related negative affect (S. A. Peters et al., 2014). As such, electing not to learn such information may have negative consequences for health and well-being. Perhaps even more importantly, when information conveys increased risk, this information can help individuals to change behaviors that may exacerbate or alleviate health risk (Frieser et al., 2018), or better plan for an uncertain future. For example, learning that one’s lifespan may be shortened can allow individuals to reprioritize goals to be more in line with their values (Fung & Carstensen, 2006). Further, learning negative information may not be as dire as predicted. Research on affective forecasting demonstrates that people often overestimate the duration and intensity of negative emotions from unfortunate life events (Wilson & Gilbert, 2003). Importantly, there is inherent uncertainty and ambiguity associated with sequencing results (i.e., even if one is at genetic risk, developing the condition remains uncertain; S. A. Peters et al., 2014), and learning genetic risk information is likely an unfamiliar experience for most people. Both of these factors are associated with a greater likelihood of inaccurate affective forecasting (Reyna et al., 2015). This literature suggests that the use of future-oriented emotions to guide intentions to receive sequencing results may reflect a biased decision-making process, as anticipated and anticipatory affect may not reflect actual responses to learning such information.

Importantly, because the consequences of receiving health information can be mixed, including both positive (e.g., ability to prevent negative outcomes or shift goals) and negative (e.g., need to cope with threatening information that is alarming), reactions to health information can produce decisional conflict (Rini et al., 2009), which is associated with negative emotions (Knops et al., 2013). Anticipating such conflict may further complicate the ability to correctly anticipate future affect or calibrate current affect to potential future consequences. At the same time, despite ample evidence of affective forecasting errors in the psychological literature, their prevalence might be exaggerated. Using affective information to make decisions is not always the “bias” it is often deemed to be (Clore et al., 2001). Rather, individuals’ intuitions about their own ability to cope in challenging circumstances is often at least somewhat accurate, even if affective forecasts are not perfect (Doré et al., 2016; Kwong et al., 2013; Levine et al., 2012). However, there is limited research on affective forecasting accuracy in the health context (Ferrer, Green, et al., 2015). Thus, is unclear based on the affective forecasting literature whether anticipated or anticipatory affect would be associated with the actual affective consequences of a decision to receive genomic sequencing results— that is, one’s later decisional conflict, as well as one’s social, emotional, and behavioral responses to such decisions. Finally, it is possible that, gaining experience with receiving genetic testing results, and thus, experience with one’s actual affective responses to such results and familiarity with the type of information such results provide, may reduce affective forecasting errors for future genetic testing results (Reyna et al., 2015) – that is, decrease the strength of the relationship between anticipated and anticipatory emotions and decisions to receive such results again in the future. Considering how these factors may influence decisions to receive genetic testing results over time is important because advances in genetic testing mean that receiving multiple (and potentially conflicting) results over an extended time period may become more common. Situations where this pattern of testing may occur include when genetic testing reports are re-issued as previously-identified variants are reclassified (e.g., Mersch et al., 2018), when individuals receive new testing as new variants are identified, when larger genetic testing panels become available, or when individuals receive clinical genetic counseling after receiving direct-to-consumer genome sequencing results (Koeller et al., 2017; Marzulla et al., 2021).

At this time, it is unclear whether future-oriented emotions a) are associated with long-term psychological responses to genomic testing results and the decision to receive them; b) are associated longitudinally with intentions to receive genomic testing results when results are available at multiple points in time; and c) are more or less associated with intentions to receive future results than individuals’ actual psychological responses to a previous set of results. The current study sought to address these gaps in the literature in the context of a genomic sequencing trial, involving participants who chose to receive carrier results specifically (i.e., variants they can pass on to offspring but that will not affect their own health or their already adult children’s health, but pose a very low risk to (future) grandchildren).

The trial took place within a larger cohort study, the purpose of which was to advance methods for identifying genetic variants associated with disease risk. As such, although participants knew they might have the opportunity to learn sequencing results as they became available during the course of the study, delivering such results was not the primary purpose of the parent cohort study and was an optional element of participation. Anticipated and anticipatory affect were assessed after enrollment in the parent cohort study, but prior to enrollment in the carrier results trial. Later, a subset of participants volunteered to take part (on average, two years after enrollment in the parent study) in a secondary intervention study within the parent cohort study, the purpose of which was to deliver carrier results to participants through one of several randomly-assigned modes of health communication. After they learned the carrier results as part of the intervention study, participants remained enrolled in the parent cohort study, where they would have future opportunities to learn additional genetic testing results, if they were willing. Participants’ affective (anxiety, decisional conflict, distress, uncertainty, and positive emotions) and social-behavioral (communication of results with others) responses to receiving results were assessed at varying time points after learning results. Intentions to continue learning actionable and nonactionable personally-relevant sequencing results, as well as future carrier results were assessed six-months after they learned results through the intervention study. We examined longitudinal relationships between baseline anticipatory and anticipated affect and reactions to receiving carrier results, as well as intentions to receive additional (e.g., personally relevant) genetic testing results after the trial.

Several relevant hypotheses were tested. Hypothesis 1 was that both anticipatory and anticipated affect would be significantly and independently associated with both immediate and long-term affective, decisional and behavioral responses to receipt of results. Hypothesis 2 was that anticipatory and anticipated negative affect at baseline would be significantly and independently associated with lower intentions to receive future results at follow-up (in accordance with previous cross-sectional findings in these participants at baseline; see Ferrer, Taber, et al., 2015). Finally, Hypothesis 3 was that actual affective reactions and decisional conflict in response to receiving results would explain a greater proportion of variance in intentions to receive future results than baseline future-oriented emotions (in the context of a model that included both types of affective predictors).

Methods

Design and Procedures

This prospective study analyzed data from a subset of the first cohort of the ClinSeq® study, a large-scale medical sequencing clinical research pilot study (L. G. Biesecker et al., 2009). Between 4 months and 4 years after completing survey measures1 and providing DNA samples at study enrollment/baseline, some participants in ClinSeq® were given the opportunity to learn carrier sequencing results, in the context of a genetic counseling non-inferiority trial (B. B. Biesecker et al., 2018). Participants who chose to receive carrier genetic sequencing results and to participate in the communication intervention trial completed follow-up surveys immediately after receiving results and one and six months later. Importantly, the carrier results that participants received were not those that would be included in a standard clinical genetic testing panel (e.g., cystic fibrosis, BRCA mutations). Moreover, although participants in the trial were those who opted to receive this set of carrier results, they were not obligated to receive additional carrier results or personally relevant results available in the future as part of the parent cohort study. After the intervention trial and its follow-up period ended, participants remained enrolled in the larger cohort study, with the understanding that they may have further opportunity to learn additional genetic testing results over time.

Participants

Participants were recruited from the greater Bethesda, MD community. The parent trial was approved by the National Human Genome Research Institute’s IRB. The initial cohort (n=969; 48% female, M age=63.9, SD=5.61) primarily identified as white (90.5%), and the majority had at least a college degree (86.5%) (Lewis et al., 2015). A subset of the initial cohort (n=462, 45% female, M age = 63.1), specifically, participants who 1) completed a baseline survey; 2) had at least one carrier gene variant confirmed by their sequencing results; 3) had not received prior genetic test results from ClinSeq, participated in the return of results intervention trial from which the follow-up data that informs the current paper were collected. Of the participants in the intervention, 92.9% identified as white, 86.2% had at least a college degree, 75% were married, and 71.9% reported a household income of more than $100,000 per year (see B. B. Biesecker et al., 2018 for a detailed characterization of the sample and description of the trial and its findings).

Participants provided informed consent before initial enrollment at baseline and again before participating in the counseling intervention study. At baseline, participants were informed that there may not be personal benefit to participating, and that although gene variants important to their own or relatives’ health may be identified, participants might not learn anything from their sequencing results.

Measures

Figure 1 depicts the time point(s) during which each measure of interest was collected.

Figure 1. Study design and measures of interest collected at each study time point.

Figure 1

Open in a new tab

Note. Baseline survey was completed at study enrollment for the initial cohort. Post results measures occurred after the genetic counseling intervention study, which took place between 4 months and 4 years after the baseline visit, followed by the 1- and 6-month follow-ups. a(Marteau & Bekker, 1992), b(O’Connor, 1995), c(Cella et al., 2002). *Only one item from the anticipatory affect scale (assessing worry about carrier results) was included in the 6-month follow-up survey.

Anticipatory and Anticipated Affect

Anticipatory and anticipated affective responses were measured in ways consistent with the decision science literature (Schlösser et al., 2013). Discriminant validity of these measures has been demonstrated in previous work (Ferrer, Taber, et al., 2015). Anticipatory, or current, affect was measured at study enrollment using three items: “On a scale of 1–7 how worried are you about the following outcomes?: (1) that your genes put you at increased risk for developing a common chronic disease, like cancer or heart disease; (2) that your relatives could be affected with a genetic condition you have passed on; (3) that you already have a health condition that was caused primarily by your genes”. These three items were averaged to create a mean baseline anticipatory affect score (α=.80). Worry is one of the most commonly-studied anticipated emotions, including in the context of health (Magnan et al., 2017). At the six-month follow-up, anticipatory affect was measured using only the second item on the scale referenced above.

Anticipated affect was measured at enrollment and at the six-month follow-up by asking participants to respond to two items rated on a 7-point Likert Scale (1= Strongly Disagree, 7 = Strongly Agree): “Please rate how strongly you agree with each statement as it describes you:

  1. If I found out that my genes put me at high risk for a fatal disease, I would be devastated”;

  2. “I don’t think I would be able to cope with finding out that my genes put me at high risk for a fatal disease.” These two items were averaged to create a mean anticipated affect score (r=.47).

For the purposes of the current study, that is, predicting participants’ reactions after receiving carrier results, this operationalization of anticipated affect is considered a measure of how participants would generally respond to negative health news in the future, and was used as a proxy for how participants might believe they would feel if they learned their genes conferred risk to their grandchildren. Although participants were not directly asked about their anticipated affect regarding carrier results, anticipated affect about results suggesting increased risk for fatal disease (for oneself) is a plausible proxy measure for examining the potential influence of anticipated affect on reactions to carrier results that are not relevant to one’s personal health, but may be relevant to the health of someone one is close to and whose health could still affect one’s own well-being (e.g., feeling guilty for having passed on a genetic variant conferring increased risk for a health condition; Persky et al., 2015). Affective forecasts, even across very different domains, tend to be consistent across individuals (Marroquín et al., 2016) as is true of the accuracy of those predictions (Dunn et al., 2007; Ellis et al., 2018). Therefore, an individual’s forecast regarding the affective impact of receiving a personally relevant, fatal genomic testing result is likely to be correlated with that same individual’s forecast about a carrier genomic testing result.

State Anxiety

Anxiety immediately after return of sequencing results was measured using the short-form State Trait Anxiety Inventory (STAI) (Marteau & Bekker, 1992) which consists of six items (e.g., “I am worried”) on a 4-point response scale. Scores are calculated by averaging responses and multiplying by twenty to facilitate comparison to earlier versions of the STAI; larger values on this scale represent higher levels of anxiety (α=.80).

Decisional Conflict

Decisional conflict was measured immediately after return of results, and at one and six months follow-up using a version of the Decisional Conflict Scale, a 15-item (e.g., “I am satisfied with my decision”) scale adapted for genetic sequencing results (O’Connor, 1995) which was summed to create a total score (α=.93). Higher values indicate greater levels of conflict about the decision to learn one’s sequencing results.

Multidimensional Impact of Cancer Risk Assessment (MICRA)

The MICRA (Cella et al., 2002) was adapted for genetic sequencing results to assess test-specific affective reactions at 1 and 6 months follow-up. Participants were prompted with, “The next questions are about some specific responses you may have had after receiving your genetic test result(s). Please indicate whether you have experienced each item in the past two weeks”. Six items on the MICRA scale assess distress (e.g., “Feeling anxious or nervous about your result(s)”); nine items assess uncertainty (e.g., “Being uncertain about what your result means for your child(ren) and/or your family’s future health”); four items assess positive experiences (e.g., “Feeling happy about your result”). The items for each of these subscales are summed to create total scores where higher values indicate higher levels of each type of test-specific affect (α=.75–.84).

Communication of Results with Relatives

To assess a relevant behavioral response to receiving carrier testing results, participants were asked to indicate how many biological sons, daughters, brothers, and sisters they had and with how many of each they had shared their results. Responses were dichotomized into having told at least one child or sibling or not by the 6-month time point.

Intentions to Continue Learning ClinSeq® Results

Intentions for receiving sequencing results in the future were assessed at the 6-month followup. Intentions were assessed separately for medically actionable, nonactionable, and carrier results. Participants responded to items preceded by the introductions, “By participating in the ClinSeq® study and having your genome sequenced you could learn about a gene variant that predisposes you to a disease that…can be prevented or treated/cannot be prevented or treated/does not affect your health, but may be important to the health of other relatives, such as your children”. For each type of intentions, participants responded to the item, “How likely is it that you will choose to learn about such a result?”, rated on a 7-point Likert scale (Extremely Unlikely-Extremely Likely).

Intervention Condition

As part of a 2×2 non-inferiority design, participants were randomly assigned to receive information about one’s carrier results via either 1) a web-based platform or a genetic counselor, and either 2) genetic counseling or none (education only). As hypothesized, return of results using the online platform was found to be non-inferior to usual care on outcomes including knowledge, genomic testing-related distress, and decisional conflict (see B. B. Biesecker et al., 2018). Intervention condition was nevertheless included as a covariate in all analyses.

Results Returned

Participants received carrier results that were deemed “pathogenic,” “likely pathogenic,” or “variant of uncertain significance” (VUS) according to criteria from the ClinGen Sequence Variant Interpretation Working Group (Richards et al., 2015). The number of pathogenic or likely pathogenic carrier results varied among participants (range: 0–6, M=1.61, SD=1.10). Because receiving more concerning results might influence affective and decisional responses to receiving results, the number of pathogenic/likely pathogenic results was included as a covariate in all analyses2; the number of VUS results was not included as a covariate, because these results should be less likely to influence such responses.

Analysis Strategy

All analyses were conducted while controlling for sociodemographic variables that were significantly associated with primary outcomes (age, race, and gender), along with the number of results returned and intervention arm.

To test Hypothesis 1 which predicted associations between anticipated and anticipatory affect and responses to receipt of results, analytic methods differed based on the specific outcome variable in question. Outcomes measured at a single timepoint (i.e., STAI, communication of results with family members) were each regressed on anticipated affect and anticipatory affect simultaneously, controlling for the covariates listed above. To reduce multiple testing concerns for outcome variables assessed at multiple time-points (i.e., decisional conflict, distress, uncertainty, positive experiences), we examined these outcomes using hierarchical linear modeling, using a random intercept model with repeated observations of the outcome variable nested within participants as the dependent variable and anticipated and anticipatory affect as independent variables, controlling for the covariates listed above. These models allow for the examination of the effects of anticipated and anticipatory affect on these outcomes, on average across time. As each of our analyses tested a separate null hypothesis (e.g., relationships between anticipated and anticipatory affect and separate, distinct outcomes of interest), we did not adjust our alpha level, following recommendations from Rubin (Rubin, 2017).

To test Hypothesis 2, regarding whether anticipatory and anticipated affect at baseline would be associated with lower intentions to receive future results after the intervention, intentions were regressed on anticipated affect and anticipatory affect simultaneously, controlling for the covariates listed above.

Finally, to test Hypothesis 3, examining the relative influence of anticipated/anticipatory affect compared to posttest psychological responses (i.e., anxiety) on intentions, intentions were regressed on anticipated and anticipatory affect and posttest affective responses and decisional conflict (i.e., comparable to a hierarchical regression with affective responses and decisional conflict added to the model used to test Hypothesis 2). For affective reaction measures taken at multiple time points (decisional conflict, distress, uncertainty, and positive experiences), participants’ within-person average on these measures across time points was computed to be used as the independent variable (scales were correlated between r=.37–.58 across time). Once again, these models controlled for the above-listed covariates.

Results

Descriptive Statistics

On average, anticipated affect for participants was below the midpoint of the 7-point scale, M=2.81, SD=1.29, Range 1–7. Anticipatory affect was also below the midpoint of the scale, M=2.44, SD=1.14, Range 1–7. Anticipated and anticipatory affect were moderately correlated with one another, r=.34, p<.001. Additionally, we examined both the consistency and stability of anticipated and anticipatory affect from enrollment to six-months after receiving genomic testing results. Anticipated affect increased modestly though significantly over time (baseline M=2.78, SD=1.28, 6-month M=2.96, SD=1.31, paired t(244)=11.29, p<.001), and was highly correlated across individuals, r=.59, 95% CI [.50, .66], p<.001. As noted above, anticipatory affect was measured at the six-month follow-up only with regard to being a carrier (i.e., one of the items that made up the 3-item measure at baseline); carrier worry significantly decreased over time (baseline M=2.57, SD=1.51, 6-month M=1.96, SD=1.11, paired t(379)=−7.91, p<.001), and was moderately correlated across individuals, r=.35, 95% CI [.26, .44], p<.001.

On average, across the outcome variables of interest, negative affective responses to receiving results were also low and did not meet clinically significant levels of negative affect (B. B. Biesecker et al., 2018). In terms of results disclosure, 80% of participants with living biological sons, daughters, sisters, or brothers reported informing at least one of these relatives of their carrier results by 6 months. Intentions to receive further results after the 6-month follow-up to the intervention study were high, all above 6 on the 7-point scale (actionable results M=6.60, SD=0.97, nonactionable results M=6.23, SD=1.19, carrier results M=6.58, SD=0.95), which was consistent with mean intentions in this sample at enrollment in ClinSeq®, on average two years prior (Ferrer, Taber, et al., 2015).

Hypothesis 1: Associations Between Future-Oriented Emotions and Reactions to Sequencing Results

Table 1 presents summary of the analyses to test Hypotheses 1, 2, and 3.

Table 1.

Summary of results

Outcome Anticipated affect Anticipatory affect
Coef [95% CI] sig. Coef [95% CI] sig.
State Trait Anxiety Inventory (STAI) b= 0.80 [0.07, 1.53] * b= 1.48 [0.66, 2.30] ***
Decisional conflict b= 1.15 [0.69, 1.61] *** b= 0.45 [−0.06, 0.97] ns
Distress b= 0.16 [−0.03, 0.35] ns b= 0.41 [0.20, 0.61] ***
Uncertainty b= 0.36 [0.08, 0.64] * b= 0.71 [0.40, 1.01] ***
Positive experiences b= −0.04 [−0.46, 0.39] ns b= 0.19 [−0.26, 0.66] ns
Communicated results with family OR= 0.81 [0.65, 1.00] ns OR= 1.12 [0.81, 1.45] ns
Intentions: actionable results b= −0.11 [−0.21, −0.01] * b= −0.07 [−0.17, 0.04] ns
Intentions: nonactionable results b= −0.23 [−0.35, −0.11] *** b= −0.13 [−0.26, −0.01] *
Intentions: carrier results b= −0.18 [−0.28, −0.09] *** b= 0.00 [−0.10, 0.10] ns
Intentions: actionable results (controlling for responses) b= −0.13 [−0.23, −0.02] * b= −0.06 [−0.17, 0.05] ns
Intentions: nonactionable results (controlling for responses) b= −0.20 [−0.32, −0.08] ** b= −0.10 [−0.23, 0.03] ns
Intentions: carrier results (controlling for responses) b= −0.15 [−0.25, −0.05] ** b= 0.02 [−0.08, 0.13] ns
Open in a new tab

Note. Coefficient estimates from multiple regression, hierarchical linear modeling, and logistic regression analyses controlling for number of results returned, intervention condition, age, race, and gender. Unstandardized beta coefficients/odds ratios with 95% confidence intervals.

*

p<.05.

**

p<.01.

***

p<.001.

Post-Results Anxiety

In the regression examining STAI scores post-intervention as a function of anticipated and anticipatory affect, anticipated affect was significantly associated with post-results anxiety, b=0.80, 95% C.I. [0.068, 1.53], ηp2=.011, p=.032, as was anticipatory affect, b=1.48, 95%. C.I. [0.66, 2.20], ηp2=.029, p<.001.

Decisional Conflict

In the hierarchical linear model examining decisional conflict scores across time as a function of anticipated and anticipatory affect, anticipated affect was significantly associated with higher levels of decisional conflict, on average, across the post-test, 1-month, and 6-month follow-ups, b=1.15, 95% C.I. [0.69, 1.61], t(429.8)=4.82, p<.001, although anticipatory affect was not, b=0.45, 95% C.I. [−0.06, 0.97], t(415.6)=1.71, p=.09.

Test-Specific Affective Reactions

In the hierarchical linear model examining test-specific distress across time as a function of anticipated and anticipatory affect, anticipated affect was not significantly associated with higher levels of distress, on average, across the 1-and 6-month follow-ups, b=0.16, 95% C.I. [0.03, 0.35], t(409.1)=1.61, p=.11, but anticipatory affect was, b=0.41, 95% C.I. [0.20, 0.61], t(404.6)=3.78, p<.001.

Anticipated affect was significantly associated with higher levels of test-specific uncertainty, on average, across the 1-and 6-month follow-ups, b=0.36, 95% C.I. [0.08, 0.64], t(408.0)=2.49, p=.012, as was anticipatory affect, b=0.71, 95% C.I. [0.40, 1.01], t(401.9)=4.45, p<.001. Finally, neither anticipated affect, b=−0.04, 95% C.I. [−0.46, 0.38], p=.86, nor anticipatory affect, b=0.19, 95% C.I. [−0.26, 0.66], p=.40, were significantly associated with positive experiences, on average, across the 1 and 6-month follow-ups.

Communication of Results with Family

In the logistic regression examining whether participants reported telling at least one biological child or sibling about their results by the six-month follow-up as a function of anticipated and anticipatory affect, there was a trend such that anticipated affect was associated with results disclosure, OR=0.81, 95% C.I. [.65, 1.00], p=.057; as baseline anticipated affect increased by one point, participants were 19% less likely to have reported telling their family members about their carrier results. Anticipatory affect was not significantly associated with results disclosure, OR=1.12, 95% C.I. [0.88, 1.45], p=.35.

Hypothesis 2: Associations Between Future-Oriented Emotions and Later Intentions to Learn Results

First, associations between anticipated and anticipatory affect at enrollment and intentions to continue learning results six-months following return of carrier results were examined. Anticipated affect, b=−0.111, 95% C.I. [−0.2, −0.01], ηp2=.021, p=.027, was associated with lower intentions to learn actionable results in the future, whereas anticipatory affect was not, b=−0.067, 95% C.I. [−0.17, 0.03], ηp2=.007, p=.20. For nonactionable results, both anticipated affect, b=−0.23, 95% C.I. [−0.35, −0.11], ηp2=.061, p<.001, and anticipatory affect, b=−0.13, 95% C.I. [−0.26, 0.01], ηp2=.020, p=.036, were associated with lower intentions. Finally, with regard to future carrier results, anticipated affect was negatively associated with intentions, b=−0.182, 95% C.I. [−0.28, −0.09], ηp2=.058, p<.001, but anticipatory affect was not, b=0.002, 95% C.I. [−0.10, 0.10], ηp2=.000, p=.967. These findings are particularly notable given that all participants had already expressed an interest in getting their results as a precondition of the intervention trial.

Hypothesis 3: Relative Explanatory Power of Future-Oriented Emotions vs. Actual Affective Reactions to Results on Intentions to Receive Future Results

Next, the question of whether follow-up intentions were more strongly associated with anticipated and anticipatory affect at baseline or actual self-reported affective reactions after receiving carrier results was examined. Multiple regression analyses examined each intentions measure as a function of anticipated and anticipatory affect, post-results anxiety, average decisional conflict, average distress, average uncertainty, and average positive experiences (controlling for the number of carrier results received, intervention condition, age, race, and gender, as above).

With regard to intentions for learning actionable results, baseline anticipated affect, b=−0.125, 95% C.I. [−0.23, −0.02], ηp2=.026, p=.017, was associated with lower intentions to learn results, whereas baseline anticipatory affect was not, b=−0.057, 95% C.I. [−0.17, 0.05], ηp2=.005, p=.30. Additionally, in this model, none of the post-results affective reactions measures were significantly associated with 6-month intentions to receive actionable results (all p’s >.22), and this model did not explain significantly more variance than the model without the post-results affective measures (ΔR2=.013, F(5, 217)=0.68, p=.64). For nonactionable results, a similar pattern was observed, such that anticipated affect, b=−0.20, 95% C.I. [−0.32, −0.07], ηp2=.045, p=.002, was negatively associated with intentions, but anticipatory affect was not, b=−0.095, 95% C.I. [−0.23, 0.03], ηp2=.010, p=.15. Of the post-results affective reactions measures in the model, only average post-results decisional conflict was a significant predictor of intentions for nonactionable results, b=−0.032, 95% C.I. [−0.06, −0.00], ηp2=.023, p=.03 (all other p’s >.69). Again, the post-results affective predictors did not explain significantly more variance in intentions than the previous model (ΔR2=.030, F(5, 217)=1.75, p=.12). Last, the same pattern was observed for carrier results, where once again baseline anticipated affect was negatively associated with intentions, b=−0.148, 95% C.I. [−0.25, −0.05], ηp2=.037, p=.004, but baseline anticipatory affect was not, b=0.023, 95% C.I. [−0.08, 0.12], ηp2=.001, p=.67. Average post-results decisional conflict trended towards significance as a predictor of intentions to receive carrier results, such that higher decisional conflict was associated with lower intentions to receive carrier results at the 6month follow-up, b=−0.022, 95% C.I. [−0.06, 0.00], ηp2=.016, p=.059 (all other affective reaction p’s >.25). For carrier results, the post-results affective predictors also did not explain significantly more variance in intentions than the previous model, (ΔR2=.04, F(5, 217)=2.12, p=.06)

Discussion

The current study examined prospective relationships between future-oriented emotions and affective, decisional, and behavioral engagement with genomic testing results. Overall, these findings supported consistent prospective relationships between future-oriented affect at study enrollment and the various post-intervention outcomes of interest. Importantly, negative psychological outcomes to receiving results were low (i.e., sub-clinical) in this study (B. B. Biesecker et al., 2018). Still, anticipated affect, or participants’ baseline expectations of their responses to future testing results (i.e., being devastated or unable to cope with a potential fatal testing result) was significantly associated with relatively higher immediate post-results anxiety, decisional conflict, and test-specific uncertainty (in support of Hypothesis 1); and relatively lower intentions to learn actionable, nonactionable, and carrier results at the 6month follow-up (Hypothesis 2) (even when participants’ actual self-reported reactions to receiving sequencing feedback were included in the model—a finding that was contrary to Hypothesis 3). There was also a trend such that higher anticipated affect was associated with lower odds of having communicated one’s carrier results to one’s relatives during follow-up, though with p=.057, this relationship did not reach a p<.05 threshold for significance when controlling for sociodemographic variables (Hypothesis 1).

Anticipatory affect, or one’s reported levels of worry about testing results at baseline was significantly associated with immediate post-results anxiety, test-specific distress and uncertainty after receiving carrier results (in support of Hypothesis 1), and intentions to learn nonactionable results at follow-up (in partial support of Hypothesis 2). However, anticipatory affect was no longer associated with intentions to receive future nonactionable results when accounting for affective and decisional responses to results, whereas decisional conflict was associated with intentions (in partial support of Hypothesis 3). This finding suggests that once participants had experienced their own psychological response to receiving carrier results, their baseline levels of worry were less important to informing their future decisions than baseline anticipated affect and experienced levels of decisional conflict.

Anticipated and anticipatory negative affect were not significantly associated with positive experiences related to receiving carrier results. This finding is consistent with literature suggesting that positive and negative affect are not opposite poles on the same scale (Barrett & Bliss-Moreau, 2009), therefore, it would not necessarily be expected that anticipated or anticipatory negative affect would be predictive of later positive affective experiences. While it was not a primary aim of our paper to examine how anticipated and anticipatory affect about genetic testing may change over time, we found that anticipated affect about personallyrelevant results increased from baseline to post-intervention, whereas anticipatory affect regarding carrier results decreased over time. It is logical that participants’ level of worry about whether they may have passed on a carrier variant to a relative decreased after participating in this intervention study, since all participants received carrier results in this study, and because these results were relatively mild in nature. As for why anticipated affect may have increased over time, it is possible that this finding could reflect a phenomenon similar to “bracing”—that is, individuals become less optimistic and “brace” for bad news as the opportunity to receive results gets closer in time (Sweeny & Shepperd, 2007) –though this explanation is somewhat speculative.

These results expand on prior work (Ferrer, Taber, et al., 2015) which found baseline crosssectional associations of future-oriented emotions, particularly anticipated affect, with intentions to receive sequencing results among a sample that included participants from the overall ClinSeq cohort. The current work demonstrates that anticipated affect measured up to years previously is associated with later intentions to receive future actionable, nonactionable, and carrier results, even after participants gain experience with receiving a type of genomic testing results (i.e., carrier results) and with their own affective reactions to those results. Additionally, anticipated and anticipatory affect at baseline were both associated with participants’ reactions to receiving carrier results. These findings suggest that future-oriented emotions may provide some accurate information about how an individual will react to receiving sequencing results, consistent with previous work which demonstrates that people can be quite accurate in their affective forecasting abilities (Doré et al., 2016; Kwong et al., 2013; Levine et al., 2012), but contrasting with more classic perspectives on affective forecasting, which typically deem it an inaccurate “bias” (S. A. Peters et al., 2014). Indeed, these findings suggest it is possible that the effects observed here would be stronger if individuals received more serious carrier findings (i.e., were informed that the health risks for children or grandchildren were higher and/or for more severe threats), given that examining reactions to relatively benign carrier information is a conservative test of the predictive power of anticipated affect for receiving serious future results.

This work contributes to existing literatures on affective science, genome sequencing, and health decision-making. One implication is derived from the fact that in the current study, participants’ anticipated negative affect about finding out negative information regarding their own health (i.e., results suggesting increased risk for fatal disease), was robustly prospectively associated with their own affective (and decisional) reactions to a different type of negative information relevant to their grandchildren (i.e., receiving carrier results). That affective forecasts about receiving genomic testing results about fatal disease risk predict actual reactions to receiving information about a less severe set of genomic testing results supports previous work demonstrating that anticipated affect about one scenario may generalize across similar decisions (Marroquín et al., 2016), and that individuals’ responses to negative information are likely related within each individual. Such a pattern is consistent with findings that related affective and appraisal tendencies are also somewhat stable within-person, including coping dispositions (Carver & Scheier, 1994), optimism and pessimism (Scheier et al., 2004), and neuroticism (Canli, 2004). In sum, these findings suggest specific contexts for which anticipated affect may generalize across similar decisions within a given domain.

Genomic information has not only increased in accessibility over time but may also be provided at various moments during the life course. These findings may inform how genetic counselors approach individuals who have received previous results about their own risks. The current results suggest that affective influences on decisions relevant to genetic counseling, and their psychological outcomes, can change over time and with experience. Continuing to understand relationships between affective factors and these outcomes over time is clearly important. As genetic counselors help patients and their relatives parse what genomic testing information may be of greatest value in identifying potential health threats, patient affective biases should be taken into consideration. The current results suggest that perhaps if individuals have very high anticipated negative affect about their genomic testing results, they may also have negative reactions to receiving results. This finding may be relevant to patients and counselors alike when considering what types of reactions people may be anticipating and how that may be affecting their decisions to receive certain types of results.

The study reported here does have some important limitations. First, our sample was comprised of older adults who were primarily white and highly educated; although the demographic makeup of the sample is relatively representative of the geographic region from which they were recruited, it is possible that these results would not generalize to other populations. Additionally, all participants in the study had expressed an interest in receiving their carrier results as a precondition to participating in the intervention trial, which may also limit generalizability. The sequencing results returned did not pose actual health risks to participants’ adult children, and participants demonstrated low levels (sub-clinical) of negative affect and decisional conflict in response to receiving carrier results. Thus, although anticipated and anticipatory affect were significantly associated with greater levels of negative affect in response to these results, it is unclear how truly meaningful the difference is between participants who reported low levels of negative affect in response to results versus no negative affect in response to results. Importantly, the findings may have been even stronger in the context of results that posed health risks. Another potential limitation is the seeming mismatch between the measurement of anticipated affect, which examined participants’ expected reactions to receiving a genomic testing result that indicated they themselves were “at risk for a fatal disease”, to some of the outcomes of interest in this study, that is, participants’ affective, decisional, and behavioral reactions to receiving (relatively innocuous) genomic testing results that indicated that they might be a carrier for a disease they themselves and their children do not have. Although this may be a limitation, the findings may speak to the ability for anticipated affect to generalize across types of similar decisions. That is, anticipated affect regarding a particular outcome may provide valuable insight into overall ability to manage threatening results. Although this measurement issue existed for the responses examining participants’ reactions to the carrier results, it is notable that anticipated affect predicted intentions to receive future personal actionable and nonactionable sequencing results, so this issue only pertains to a subset of the results. Finally, we tested relationships between our independent variables of interest and multiple outcome variables; given the number of tests, it is possible that some of our results may reflect type I errors. However, several of our findings were significant at p<.001 or p<.01 thresholds (see Table 1), which makes this less concerning. These limitations are balanced by a number of strengths. The use of a cohort of participants in the context of an ongoing genomic sequencing study allowed us to examine how these affective factors may be related to long-term decision making in a consequential health context. Our results contribute to a growing body of evidence on the role of future-oriented emotions in genomic testing, behavioral medicine, and health decision-making more broadly. Future work should continue to examine the impact of anticipated and anticipatory emotions in behavioral medicine contexts, including defining how the effects of specific future-oriented emotions (e.g., anticipated regret vs. anticipated upset) differ, and identifying whether interventions targeting change in future-oriented emotions may have clinically meaningful effects in the context of genetic counseling.

Funding:

The ClinSeq study was supported by the Intramural Research Program of the National Human Genome Research Institute, Grant HG200359.

Footnotes

Publisher's Disclaimer: This AM is a PDF file of the manuscript accepted for publication after peer review, when applicable, but does not reflect post-acceptance improvements, or any corrections. Use of this AM is subject to the publisher’s embargo period and AM terms of use. Under no circumstances may this AM be shared or distributed under a Creative Commons or other form of open access license, nor may it be reformatted or enhanced, whether by the Author or third parties. See here for Springer Nature’s terms of use for AM versions of subscription articles: https://www.springernature.com/gp/open-research/policies/accepted-manuscript-term

Conflicts of interest:

ASG, IAI, WMPK, KLL, BBB, and RAF have no conflicts of interest to declare. LGB has the following disclosures: Unpaid member of Illumina Ethics Advisory Board, in kind research support from ArQule Incorporated, now wholly owned by Merck, Inc., Deputy Editor Cold Spring Harbor Press, Molecular Case Reports.

Ethics approval: The ClinSeq study was approved by the National Institutes of Health Internal Review Board, Protocol # 7-HG-0002.

Consent to participate: Informed consent was collected from all participants.

Code availability: Data and reproducible code from the current study are available to individual investigators by official request, as required by regulations for data collected by a federal employee.

1

Including the amount of time between the baseline assessment and enrollment in the intervention study as a covariate did not meaningfully change our results and was not a significant moderator of the effects of interest.

2

Arguably, this construct could also be represented with a categorical variable (no pathogenic/likely pathogenic results vs. one or more pathogenic/likely pathogenic results). Treating this covariate as a categorical variable did not meaningfully change our results.

Availability of data and material:

Data and reproducible code from the current study are available to individual investigators by official request, as required by regulations for data collected by a federal employee.

References

  1. Barrett LF, & Bliss-Moreau E (2009). Affect as a Psychological Primitive. In Advances in Experimental Social Psychology (Vol. 41, pp. 167–218). Academic Press. 10.1016/S0065-2601(08)00404-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baumgartner H, Pieters R, & Bagozzi RP (2008). Future-oriented emotions: conceptualization and behavioral effects. European Journal of Social Psychology, 38(4), 685–696. 10.1002/ejsp.467 [DOI] [Google Scholar]
  3. Biesecker BB, Lewis KL, Umstead KL, Johnston JJ, Turbitt E, Fishler KP, Patton JH, Miller IM, Heidlebaugh AR, & Biesecker LG (2018). Web platform vs inperson genetic counselor for return of carrier results from exome sequencing a randomized clinical trial. JAMA Internal Medicine, 178(3), 338–346. 10.1001/jamainternmed.2017.8049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Biesecker LG, Mullikin JC, Facio FM, Turner C, Cherukuri PF, Blakesley RW, Bouffard GG, Chines PS, Cruz P, Hansen NF, Teer JK, Maskeri B, Young AC, Manolio TA, Wilson AF, Finkel T, Hwang P, Arai A, Remaley AT, … Green ED (2009). The ClinSeq Project: Piloting large-scale genome sequencing for research in genomic medicine. Genome Research, 19(9), 1665–1674. 10.1101/gr.092841.109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Canli T (2004). Functioned brain mapping of extroversion and neuroticism: Learning from individual differences in emotion processing. Journal of Personality, 72(6), 1105–1132. 10.1111/j.1467-6494.2004.00292.x [DOI] [PubMed] [Google Scholar]
  6. Carver CS, & Scheier MF (1994). Situational Coping and Coping Dispositions in a Stressful Transaction. Journal of Personality and Social Psychology, 66(1), 184–195. 10.1037/0022-3514.66.1.184 [DOI] [PubMed] [Google Scholar]
  7. Cella D, Hughes C, Peterman A, Chang C-H, Peshkin BN, Schwartz MD, Wenzel L, Lemke A, Marcus AC, & Lerman C (2002). A brief assessment of concerns associated with genetic testing for cancer: the Multidimensional Impact of Cancer Risk Assessment (MICRA) questionnaire. Health Psychology, 21(6), 564–572. http://www.ncbi.nlm.nih.gov/pubmed/12433008 [PubMed] [Google Scholar]
  8. Chapman GB, & Coups EJ (2006). Emotions and preventive health behavior: Worry, regret, and influenza vaccination. Health Psychology, 25(1), 82–90. 10.1037/02786133.25.1.82 [DOI] [PubMed] [Google Scholar]
  9. Clore GL, Gasper K, & Garvin E (2001). Affect as information. In Handbook of Affect and Social Cognition. Lawrence Erlbaum Associates. [Google Scholar]
  10. Doré BP, Meksin R, Mather M, Hirst W, & Ochsner KN (2016). Highly Accurate Prediction of Emotions Surrounding the Attacks of September 11, 2001 Over 1-, 2-, and 7Year Prediction Intervals. Journal of Experimental Psychology: General, 145(6), 788–795. 10.1037/xge0000168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dunn EW, Brackett MA, Ashton-James C, Schneiderman E, & Salovey P (2007). On emotionally intelligent time travel: individual differences in affective forecasting ability. Personality & Social Psychology Bulletin, 33(1), 85–93. 10.1177/0146167206294201 [DOI] [PubMed] [Google Scholar]
  12. Ellis EM, Elwyn G, Nelson WL, Scalia P, Kobrin SC, & Ferrer RA (2018). Interventions to Engage Affective Forecasting in Health-Related Decision Making: A MetaAnalysis. Annals of Behavioral Medicine, 52(2), 157–174. 10.1093/abm/kax024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ferrer RA, Green PA, & Barrett LF (2015). Affective Science Perspectives on Cancer Control: Strategically Crafting a Mutually Beneficial Research Agenda. Perspectives on Psychological Science. 10.1177/1745691615576755 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ferrer RA, Klein W, Lerner JS, Reyna V, & Keltner D (2014). Emotions and health decision making: Extending the Appraisal Tendency Framework to improve health and health care. In Roberto CA & Kawachi I (Eds.), Behavioral Economics and Public Health (Vol. 10). 10.1093/med/9780199398331.003.0004 [DOI] [Google Scholar]
  15. Ferrer RA, Taber JM, Klein WMP, Harris PR, Lewis KL, & Biesecker LG (2015). The role of current affect, anticipated affect and spontaneous self-affirmation in decisions to receive self-threatening genetic risk information. Cognition and Emotion, 29(8), 1456–1465. 10.1080/02699931.2014.985188 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Frieser MJ, Wilson S, & Vrieze S (2018). Behavioral impact of return of genetic test results for complex disease: Systematic review and meta-analysis. Health Psychology, 37(12), 1134–1144. 10.1037/hea0000683 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fung HH, & Carstensen LL (2006). Goals change when life’s fragility is primed: Lessons learned from older adults, the September 11 attacks and SARS. Social Cognition, 24(3), 248–278. 10.1521/soco.2006.24.3.248 [DOI] [Google Scholar]
  18. Han PKJ, Moser RP, & Klein WMP (2006). Perceived Ambiguity About Cancer Prevention Recommendations: Relationship to Perceptions of Cancer Preventability, Risk, and Worry. Journal of Health Communication, 11(sup001), 51–69. 10.1080/10810730600637541 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Han PKJ, Reeve BB, Moser RP, & Klein WMP (2009). Aversion to Ambiguity Regarding Medical Tests and Treatments: Measurement, Prevalence, and Relationship to Sociodemographic Factors. Journal of Health Communication, 14(6), 556–572. 10.1080/10810730903089630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Howell JL, & Shepperd JA (2012). Reducing information avoidance through affirmation. Psychological Science, 23(2), 141–145. 10.1177/0956797611424164 [DOI] [PubMed] [Google Scholar]
  21. Knops AM, Goossens A, Ubbink DT, Legemate DA, Stalpers LJ, & Bossuyt PM (2013). Interpreting patient decisional conflict scores: behavior and emotions in decisions about treatment. Medical Decision Making : An International Journal of the Society for Medical Decision Making, 33(1), 78–84. 10.1177/0272989X12453500 [DOI] [PubMed] [Google Scholar]
  22. Koeller DR, Uhlmann WR, Carere DA, Green RC, & Roberts JS (2017). Utilization of Genetic Counseling after Direct-to-Consumer Genetic Testing: Findings from the Impact of Personal Genomics (PGen) Study. Journal of Genetic Counseling 2017 26:6, 26(6), 1270–1279. 10.1007/S10897-017-0106-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kwong JYY, Wong KFE, & Tang SKY (2013). Comparing predicted and actual affective responses to process versus outcome: An emotion-as-feedback perspective. Cognition, 129(1), 42–50. 10.1016/j.cognition.2013.05.012 [DOI] [PubMed] [Google Scholar]
  24. Levine LJ, Lench HC, Kaplan RL, & Safer MA (2012). Accuracy and artifact: Reexamining the intensity bias in affective forecasting. Journal of Personality and Social Psychology, 103(4), 584–605. 10.1037/a0029544 [DOI] [PubMed] [Google Scholar]
  25. Lewis KL, Han PKJ, Hooker GW, Klein WMP, Biesecker LG, & Biesecker BB (2015). Characterizing participants in the clinSeq genome sequencing cohort as early adopters of a new health technology. PLoS ONE, 10(7). 10.1371/journal.pone.0132690 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Loewenstein G, & Lerner JS (2003). The Role of Affect in Decision Making. In Davidson R, Goldsmith H, & Scherer K (Eds.), Handbook of Affective Science (pp. 619–642). Oxford University Press. [Google Scholar]
  27. Magnan RE, Shorey Fennell BR, & Brady JM (2017). Health decision making and behavior: The role of affect-laden constructs. Social and Personality Psychology Compass, 11(8), e12333. 10.1111/spc3.12333 [DOI] [Google Scholar]
  28. Marroquín B, Boyle CC, Nolen-Hoeksema S, & Stanton AL (2016). Using emotion as information in future-oriented cognition: Individual differences in the context of state negative affect. Personality and Individual Differences, 95, 121–126. 10.1016/j.paid.2016.02.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Marteau TM, & Bekker H (1992). The development of a six‐item short‐form of the state scale of the Spielberger State—Trait Anxiety Inventory (STAI). British Journal of Clinical Psychology. 10.1111/j.2044-8260.1992.tb00997.x [DOI] [PubMed] [Google Scholar]
  30. Marzulla T, Roberts JS, DeVries R, Koeller DR, Green RC, & Uhlmann WR (2021). Genetic counseling following direct-to consumer genetic testing: Consumer perspectives. Journal of Genetic Counseling, 30(1), 329–334. 10.1002/JGC4.1309 [DOI] [PubMed] [Google Scholar]
  31. Mersch J, Brown N, Pirzadeh-Miller S, Mundt E, Cox HC, Brown K, Aston M, Esterling L, Manley S, & Ross T (2018). Prevalence of Variant Reclassification Following Hereditary Cancer Genetic Testing. JAMA, 320(12), 1266–1274. 10.1001/JAMA.2018.13152 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. O’Connor AM (1995). Validation of a Decisional Conflict Scale. Medical Decision Making, 15(1), 25–30. 10.1177/0272989X9501500105 [DOI] [PubMed] [Google Scholar]
  33. Persky S, McBride CM, Faith MS, Wagner LK, & Ward DS (2015). Mothers’ guilt responses to children’s obesity risk feedback. Journal of Health Psychology, 20(5), 649–658. 10.1177/1359105315576608 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Peters E, Västfjäll D, Gärling T, & Slovic P (2006). Affect and decision making: a “hot” topic. Journal of Behavioral Decision Making, 19(2), 79–85. 10.1002/bdm.528 [DOI] [Google Scholar]
  35. Peters SA, Laham SM, Pachter N, & Winship IM (2014). The future in clinical genetics: affective forecasting biases in patient and clinician decision making. Clinical Genetics, 85(4), 312–317. 10.1111/cge.12255 [DOI] [PubMed] [Google Scholar]
  36. Reyna VF, Nelson WL, Han PK, & Pignone MP (2015). Decision making and cancer. American Psychologist, 70(2), 105–118. 10.1037/a0036834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, & Rehm HL (2015). Standards and guidelines for the interpretation of sequence variants: A joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in Medicine, 17(5), 405–424. 10.1038/gim.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rini C, O’Neill SC, Valdimarsdottir H, Goldsmith RE, Jandorf L, Brown K, DeMarco TA, Peshkin BN, & Schwartz MD (2009). Cognitive and emotional factors predicting decisional conflict among high-risk breast cancer survivors who receive uninformative BRCA1/2 results. Health Psychology, 28(5), 569–578. 10.1037/a0015205 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rubin M (2017). Do p Values Lose Their Meaning in Exploratory Analyses? It Depends How You Define the Familywise Error Rate. Review of General Psychology, 21(3), 269–275. 10.1037/gpr0000123 [DOI] [Google Scholar]
  40. Scheier MF, Carver CS, & Bridges MW (2004). Optimism, pessimism, and psychological well-being. In Optimism & pessimism: Implications for theory, research, and practice, (pp. 189–216). American Psychological Association, 10.1037/10385-009 [DOI] [Google Scholar]
  41. Schlösser T, Dunning D, & Fetchenhauer D (2013). What a Feeling: The Role of Immediate and Anticipated Emotions in Risky Decisions. Journal of Behavioral Decision Making, 26(1), 13–30. 10.1002/bdm.757 [DOI] [Google Scholar]
  42. Sweeny K, & Shepperd JA (2007). Do People Brace Sensibly? Risk Judgments and Event Likelihood: Http://Dx.Doi.Org/10.1177/0146167207301024, 33(8), 1064–1075. 10.1177/0146167207301024 [DOI] [PubMed] [Google Scholar]
  43. Wilson TD, & Gilbert DT (2003). Affective Forecasting. In Zanna MP (Ed.), Advances in Experimental Social Psychology (Vol. 35, pp. 345–411). Elsevier Academic Press. 10.1016/S0065-2601(03)01006-2 [DOI] [Google Scholar]
  44. Xu Z, & Guo H (2019). Advantages of Anticipated Emotions over Anticipatory Emotions and Cognitions in Health Decisions: A Meta-Analysis. Health Communication, 34(7), 774–781. 10.1080/10410236.2018.1434738 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data and reproducible code from the current study are available to individual investigators by official request, as required by regulations for data collected by a federal employee.

RESOURCES