The Behavioral Immune System and Attitudes About Vaccines

Abstract

The present research utilized evolutionary theory to examine the relation between the behavioral immune system (i.e., disgust sensitivity) and attitudes about vaccines. The findings from the studies suggest that higher levels of dispositional disgust sensitivity is predictive of more negative attitudes toward vaccines. These findings are consistent with several recent publications and thus have broad implications for public health research associated with vaccines. In Study 1, participants reporting higher dispositional disgust sensitivity (specifically, contamination disgust) tended to report more negative attitudes about vaccines. Study 2 replicated this result in a nonstudent sample using additional measures of disgust sensitivity more closely associated with aversion to perceived sources of contamination. Study 2 also revealed that beliefs about the likelihood of contracting illness in the future were unrelated to vaccine attitudes. Implications for the observed relation between intuitive aversion to contamination and vaccine attitudes are discussed.

Keywords

attitudes evolutionary psychology health individual differences

Recently, there has been controversy regarding the safety and effectiveness of vaccinations. A recent Gallup poll indicated that although a majority (84%) of the U.S. population believes that vaccines are “very important” or “extremely important,” a smaller subset (9%) say that vaccines are more dangerous than the diseases they are intended to prevent (Newport, 2015). These negative views have garnered considerable media attention, and recent outbreaks of vaccine-preventable diseases such as the 2014 measles outbreak in California and several neighboring states affected disproportionally high numbers of unvaccinated people (Fernandez, 2015). Thus, the potential for negative public health consequences exists when even a small number of people forego recommended vaccination schedules.

Understanding the factors that give rise to such differences in attitudes about vaccines is critical for public health programs seeking to reduce the spread of dangerous diseases. As such, the primary aim of the present research was to investigate the factors associated with attitudes about vaccines. Specifically, the present studies examined vaccine attitudes from an evolutionary perspective and sought to understand the nature of the relation between evolved psychological disease-avoidance mechanisms and modern attitudes about vaccines.

Disease Avoidance as an Adaptive Challenge

Pathogens and parasites can inflict considerable fitness costs on their host organisms and have represented a significant selection pressure throughout evolutionary history (Dobson & Carper, 1996). Not surprisingly, multiple adaptive strategies for combating and avoiding disease have evolved. The most familiar strategy is the biological immune system, which combats infections after they have entered the body. However, humans lived in mobile hunter-gatherer societies throughout much of their evolutionary history and tribes of hunter-gatherers tended to be small (50–100 individuals on average). Dobson and Carper (1996) note that pathogens that infected these populations were likely to be highly communicable and probably did not convey lasting immunity in their hosts. Given this disease profile, selection pressures would have favored the evolution of behavioral mechanisms for successfully avoiding contact with disease altogether.

Evolved behavioral strategies for avoiding disease have been tabbed the behavioral immune system (BIS; Schaller, 2006). The BIS constitutes a suite of psychological adaptations that promote disease avoidance in order to reduce the time and energy spent fighting off infections. By reducing activation of the biological immune system, more resources can be devoted to fitness-enhancing activities (e.g., finding food, pursuing mates, caring for children). However, because pathogens are too small to be detected directly, the BIS must trigger avoidance strategies based on indirect cues of disease risk (e.g., the presence of unfamiliar people, foul-smelling substances, the risk of infection from body envelope violations). The BIS functions by eliciting the emotional experience of disgust upon encountering such cues, which motivates avoidance of the stimuli, thereby reducing the risk of infection. The experience of disgust is so central to the BIS that some have even argued that the BIS and disgust are simply two names for the same thing (Lieberman & Patrick, 2014).

This BIS is not a perfectly accurate system, and two kinds of errors can occur: (1) A disease threat can be inferred when no threat is present or (2) the system can fail to detect a true disease threat. Also, there is an asymmetry in the costs associated with errors that the system can make. Failing to avoid pathogens is typically more costly to reproductive fitness than mistakenly avoiding a nondisease threat (see Nesse, 2005, for an overview of biases in threat detection systems). Therefore, the BIS is likely biased toward false-positive errors (avoidance of harmless stimuli; Schaller & Park, 2011).

In addition, evolutionary pressures that shaped the BIS and selected for dispositional sensitivity to experience disgust have been in flux throughout history. Some areas of the world have experienced more disease than others, and the threat of infection from disease in a given geographic region varies over time. Because of this variability, there is unlikely to be an optimal level of disgust sensitivity. Thus, dispositional disgust sensitivity is expected to vary from individual to individual and across geographic regions. Prior research supports this perspective and suggests that variability in disgust relates to cultural differences such as individualism and collectivism (Fincher, Thornhill, Murray, & Schaller, 2008), which may be by-products of evolved individual-level psychological disease-avoidance mechanisms (Clay, Terrizzi, & Shook, 2012).

There is also evidence that individual differences in disgust sensitivity relate to a number of social attitudes. For example, heightened disgust sensitivity predicts more negative attitudes toward foreigners (Faulkner, Schaller, Park, & Duncan, 2004), a relation which is likely to stem from the fact that people from geographically distant populations represented an important route of disease transmission throughout history (Diamond, 1997). Until relatively recently, avoiding unfamiliar people and maintaining close ties to in-group members likely reduced the risk of contracting disease. Although modern health practices have reduced this risk considerably in contemporary first-world nations, evolved BIS strategies are still likely to promote avoidance of unfamiliar people and will be more readily activated in individuals with higher dispositional disgust sensitivity. More generally, BIS processes operating in the contemporary environment, where the true risk posed by disease is lower compared to the evolutionary past, might motivate suboptimal behavioral strategies (Schaller, Murray, & Bangerter, 2015), a perspective that is consistent with a Darwinian approach to modern medicine (Nesse & Williams, 1994).

Such a Darwinian approach might be useful in understanding attitudes about vaccines, as vaccines have only recently become a feature of the human environment. Further, vaccines can reduce the transmission rates of disease but are also typically administered via common routes of pathogenic transmission (i.e., injection, ingestion, or inhalation) and intentionally trigger an immune response (something the BIS aims to minimize). Thus, there are plausible reasons to predict two very different relations between the BIS and vaccine attitudes: (1) People higher in dispositional disgust sensitivity might hold more positive attitudes toward vaccines because of the knowledge that vaccines protect against the threat of disease and (2) People higher in disgust sensitivity might hold more negative attitudes about vaccines as a by-product of being motivated to avoid potential infection via common routes of disease transmission.

Limited prior research supports the latter perspective. The physiological response of blood–injury–injection (BII)-phobic patients to blood and needles is consistent with a disgust response (Page, 1994), and several studies have found additional evidence of a link between disgust sensitivity and BII phobia (Olatunji, Haidt, McKay, & David, 2008; Sawchuk, Lohr, Tolin, Lee, & Kleinknecht, 2000; Tolin, Lohr, Sawchuk, & Lee, 1997). A relation between heightened disgust sensitivity and negative attitudes about vaccines would also make sense given common arguments from vaccine opponents that the ingredients in vaccines are harmful (Goldstein, 2014) and that vaccines circumvent the body’s natural defenses (Mercola, 2011). Consistent with this view, one recent study found that disgust sensitivity was associated with greater skepticism about vaccines (Clifford & Wendell, 2015).

Overview of the Present Research

Two cross-sectional studies sought to develop further clarity about the relation between dispositional disgust sensitivity and vaccine attitudes. The first study used multiple measures of disgust sensitivity to test the rival hypotheses about the relation between disgust sensitivity and attitudes about vaccines (i.e., testing for a positive or negative relation). The second study was confirmatory, seeking to replicate and extend the findings from the first study. Robust evidence of a relation between disgust sensitivity and vaccine attitudes would provide clues about the manner in which evolved psychological mechanisms influence attitudes in the modern environment.

Study 1

The first study investigated the relation between dispositional disgust sensitivity and attitudes about vaccines in a student sample. Measures of disgust sensitivity and vaccine attitudes were embedded in a larger online survey investigating social attitudes.

Participants

Undergraduates (N = 471) self-selected into an online survey titled “Opinions about Society” and received course credit. Participants were excluded if they completed the entire survey in less than 10 mins or more than 2 hr (n = 50). After excluding these participants, the average completion time was 39.65 min (SD = 21.16, range = 10.02–117.40). Participants were also excluded if their responses contained more than 10% missing data (the imputation threshold; n = 12) or if responses were invariant across any of the measures in which some of the items were reverse scored (n = 34). The remaining sample (n = 375) was 62% female with an average age of 20.46 years (SD = 4.55, range = 18–55).¹ Participants reported their ethnicity as White/Caucasian (47%), Hispanic/Latino (24%), African American/Black (11%), and Asian (10%), with an additional 8% of participants not reporting an ethnic identification. No a priori target was set for sample size. Rather, as many responses as possible were collected given that exclusion patterns such as those reported above were known to be typical in this particular student population. Post hoc sensitivity analysis using GPower (Faul, Erdfelder, Buchner, & Lang, 2009) indicated that this sample provided 80% power to detect a significant bivariate correlation of ±.14 or stronger using a two-tailed test, the main statistic of interest.

Measures

The variables assessed in the present analysis were part of a larger online study, and several measures unrelated to the present investigation are not reported below.

Ideological self-identification

Given prior research that indicates higher levels of disgust sensitivity relate to more conservative political views (Smith, Oxley, Hibbing, Alford, & Hibbing, 2011), participants reported their ideological self-identification for use as a control variable. This was a single-item measure captured on a 7-point scale anchored at 1 (liberal) and 7 (conservative).

Disgust sensitivity

Two measures of disgust sensitivity were included in the study: The Disgust Sensitivity Scale–Revised (DS-R; Olatunji et al., 2009) is a 25-item measure of sensitivity to disgust, which contains three subscales. Core disgust assesses the level of disgust associated with real or perceived oral incorporation of pathogens (e.g. “It bothers me to hear someone clear a throat full of mucous”), animal reminder disgust assesses concern over the animal origins of humans (e.g., “I would go out of my way to avoid walking through a graveyard”), and contamination disgust assesses aversion toward potential direct or indirect sources of infection (e.g., “I never let any part of my body touch the toilet seat in public restrooms”). Participants indicated their level of agreement with each statement on a 7-point scale anchored at 1 (strongly disagree) and 7 (strongly agree). Composite indices were formed by computing the average of all items within each subscale.

The Three-Domain Disgust Scale assesses disgust across three separate domains (Tybur, Lieberman, & Griskevicius, 2009). Pathogen disgust measures the level of disgust associated with possible sources of pathogenic infection (e.g., “Shaking hands with a stranger who has sweaty palms”), sexual disgust measures the level of disgust associated with sexual acts (e.g., “Hearing two strangers having sex”), and moral disgust measures disgust associated with moral transgressions (e.g., “Stealing from a neighbor”). Participants indicated their level of disgust with each statement on a 7-point scale anchored at 0 (not disgusting at all) and 6 (extremely disgusting). Again, indices were created by averaging items within each subscale.

Vaccine attitudes

Attitudes toward vaccines were measured using a 12-item scale developed to assess themes reported in prior research investigating vaccine attitudes (see Brown et al., 2010, for an overview). The scale incorporated various statements expressing both pro- and anti-vaccine positions about the safety, effectiveness, and necessity of vaccines. Participants indicated their agreement with each item on a 9-point scale anchored at −4 (strongly disagree) and +4 (strongly agree). Items phrased as anti-vaccine were reverse scored after data collection, and the items were averaged into a single index so that higher scores indicated more positive attitudes toward vaccines (see Table 1 for the full list of items). Given prior opinion polling indicating that only a small proportion of the population harbors negative attitudes about vaccines (Newport, 2015), it is important to note that attitudes about vaccines were normally distributed in the sample (skewness = .61, kurtosis = .55; see Figure 1a).

Table 1.
Vaccine Attitudes Survey Items.

Vaccinations are an important part of a child’s health care.

Now that major illnesses have largely disappeared, we really don’t need vaccines anymore—R.

Some vaccines might cause lasting health problems—R.

It is likely that vaccines cause autism and other serious disorders—R.

In general, vaccines are safe and effective.

Vaccines can overload a child’s immune system—R.

It is a parent’s responsibility to vaccinate their child for the good of the whole community.

Scientific research has helped us to understand a lot about vaccine safety and effectiveness.

A person can build up natural immunities to disease that is just as effective as a vaccination—R.

Sometimes vaccines can cause the diseases that they are supposed to prevent—R.

Vaccines can contain preservatives that are dangerous—R.

It is best to wait until children are older before starting to give them vaccines—R.

Note. R indicates that an item was reverse scored after data collection.

Figure 1.
Distribution of vaccine attitude scores in Studies 1 and 2.

Procedure

Study procedures were approved by the college’s institutional review board (IRB). The measures were embedded in a larger questionnaire assessing relations among a host of social attitude constructs. Undergraduates enrolled in an introductory psychology course self-selected into the online study and completed the survey at their convenience. Participants were awarded course credit for their participation.

Results and Discussion

Prior to analyses, data missing at random were imputed using the “Amelia” package in R (Version 3.3.0; Honaker, King, & Blackwell, 2014). Initial bivariate correlations were assessed between all disgust sensitivity measures and vaccine attitudes. As prior disgust research has examined relations to the overall index of all items contained in the DS-R measure (e.g., Clay et al., 2012; Terrizzi et al., 2014) in addition to the separate subscales, all four indices were assessed in relation to vaccine attitudes. All bivariate correlations as well as means, standard deviations, and reliability statistics (Cronbach’s α) are presented in Table 2. The correlations indicated that attitudes about vaccines were negatively correlated with the overall DS-R composite (p < .001), core disgust (p = .01), animal reminder disgust (p = .003), contamination disgust (p < .001), pathogen disgust (p = .01), and sexual disgust (p = .02). Moral disgust had a nonsignificant positive correlation with vaccine attitudes (p = .24). The distinction of the Moral Disgust subscale relative to the other BIS measures is not surprising, as the Moral Disgust Scale is not theorized to assess tendencies toward disease avoidance. Rather, the scale is intended to measure aversion to violations of social norms (Tybur et al., 2009).

Table 2.
Bivariate Correlations, Means, Standard Deviations, and Reliability Coefficients (Cronbach’s α) for All Study 1 Variables.

n = 375 1 2 3 4 5 6 7 8 9 Mean SD

1. DS-R—Composite Score — NA NA NA .59 .45 .12 −.02 −.20 4.90 1.05

2. DS-R—Core — .66 .52 .58 .38 .10 −.03 −.13 5.21 1.12

3. DS-R—Animal Reminder — .45 .43 .37 .09 −.07 −.15 4.78 1.35

4. DS-R—Contamination — .48 .40 .13 .10 −.29 4.35 1.28

5. TDD—Pathogen — .31 .21 .07 −.13 4.20 1.14

6. TDD—Sex — .26 .08 −.12 3.10 1.58

7. TDD—Moral — .02 .06 4.24 1.13

8. ISI — −.15 4.00 2.00

9. Vaccine attitudes — 0.66 1.16

Cronbach’s α .89 .84 .81 .60 .78 .84 .78 NA 0.84

Note. DS-R = Disgust Sensitivity Scale–Revised; TDD = Three-Domain Disgust Scale; ISI = ideological self-identification (higher scores = more conservative self-identification); NA = not applicable.

Differences in vaccine attitudes were also assessed across a host of demographic variables to determine whether any of the observed relations might be better explained by an intervening third variable. Analyses indicated no differences in vaccine attitudes based on participant sex or ethnicity. However, a significant correlation was observed between vaccine attitudes and ideological self-identification, r(373) = −.15, p = .004, such that participants who self-identified as more conservative reported more negative attitudes about vaccines. As such, follow-up analyses controlled for ideological self-identification.

Hierarchical linear regression was used to examine core disgust, animal reminder disgust, contamination disgust, pathogen disgust, and sexual disgust as simultaneous predictors of vaccine attitudes after controlling for ideological self-identification. This analysis addressed two concerns. First, it controlled for any effect of ideological self-identification on the relation between dispositional disgust sensitivity and vaccine attitudes. Second, it revealed the unique variance predicted by each subscale and provided an assessment of the extent to which the observed associations with vaccine attitudes were based on variance held in common across the Disgust Sensitivity subscales. Results from the analyses are presented in Table 3. The residuals of the model were normally distributed (skewness = .38, kurtosis = .24), and all other assumptions of regression were met.

Table 3.
Results of the Multiple Regression Analysis Investigating Disgust Subscales as Predictors of Vaccine Attitudes in Study 1.

n = 378 R ² B SE (b) 95% CI β p

Step 1—Control variable .02

ISI −.11 .04 [−.18, –.04] −.15 .003

Step 2—Disgust Sensitivity .10

DS-R Core .05 .08 [−.11, .20] .05 .54

DS-R Animal Reminder −.07 .06 [−.19, .05] −.08 .24

DS-R Contamination −.25 .06 [−.36, −.14] −.27 <.001

TDD Pathogen .02 .06 [−.11, .14] .02 .80

TDD Sex .00 .04 [−.08, .09] .01 .92

Note. R² = proportion of variance in vaccine attitudes explained; b = unstandardized regression coefficient; β = standardized regression coefficient; ISI = ideological self-identification (higher scores = more conservative self-identification); DS-R = Disgust Sensitivity Scale–Revised; TDD = Three-Domain Disgust Scale; CI, confidence interval.

When all of the predictors were entered into the model simultaneously, only contamination disgust predicted a significant amount of unique variance in vaccine attitudes. The fact that none of the other subscales emerged as significant predictors indicates that their bivariate correlations with vaccine attitudes were due to variance held in common with contamination disgust.

Because of this finding, a structural equation model (using the “lavaan” package in R; Rosseel et al., 2016) was generated in which the five disgust subscales were modeled as indicators of a single latent construct (referred to as “contamination aversion”). Contamination aversion and political ideology were then modeled as predictors of vaccine attitudes (see Figure 2). The results of this analysis indicated that the model provided adequate fit to the data, χ²(14, n = 375) = 54.00, p < .001, comparative fit index [CFI] = .94, root mean square error of approximation [RMSEA] = .09. Most importantly, the factor loadings of all of the disgust subscales from the latent contamination aversion construct were statistically significant (all ps < .001), and the factor loading from contamination aversion to vaccine attitudes was also significant, even while controlling for ideological self-identification.

Figure 2.
Structural equation model with contamination aversion modeled as a latent variable from Study 1. ISI = ideological self-identification; DS-CONT, DS-AR, and DS-CORE = Contamination, Animal Reminder, and Core Disgust subscales of the Disgust Sensitivity measure, respectively; VAC = vaccine attitude composite. Factor loadings for contamination disgust, political ideology, and vaccine attitude indicators were fixed (not estimated). p < .01. p < .001.

These findings provide evidence of a relation between dispositional disgust sensitivity and vaccine attitudes and suggest that contamination disgust is a particularly important factor in this relation such that participants who were more averse to perceived contaminants tended to express more negative attitudes about vaccines. Given that the BIS is theorized to promote avoidance of potential sources of contamination as a means of reducing biological resources spent fighting off an infection, it makes sense that heightened disgust sensitivity might motivate negativity toward a foreign substance (vaccines), which is intentionally introduced into the body in order to provoke a biological immune response. Given these findings, a follow-up confirmatory study was conducted using additional measures of disgust sensitivity in a nonstudent sample.

Study 2

In Study 1, contamination disgust demonstrated the strongest relation to vaccine attitudes and was the only Disgust Sensitivity subscale that demonstrated significant unique predictive value. This suggests that aversion to perceived sources of contamination is an important factor associated with negative attitudes about vaccines and is consistent with the common argument (by vaccine opponents) that the ingredients contained in vaccines are dangerous.

To investigate this relation further, Study 2 incorporated additional measures of dispositional disgust sensitivity more specifically related to contamination aversion. This study was confirmatory, and it was hypothesized that higher levels of disgust sensitivity would predict more negative attitudes toward vaccines. Study 2 was also conducted using a nonstudent sample. Replication and extension of the findings from Study 1 would provide further evidence for the relation between heightened disgust sensitivity (specifically, aversion to perceived contaminants) and negative attitudes toward vaccines.

Participants

Two hundred and fifty-four participants were recruited for participation via Amazon’s Mechanical Turk™ website and were paid US$2.00 in exchange for completing the survey. As in Study 1, participants were excluded if they completed the study in less than 10 min or more than 2 hr (n = 11). The average completion time for the remaining sample (n = 243) was 30.15 min (SD = 15.77, range = 11–106 min). Post hoc sensitivity analysis using GPower indicated that this sample size resulted in 80% power to detect a significant bivariate correlation of −.16 or stronger using a one-tailed test.

The sample was evenly split in regard to participant sex (121 female vs. 120 male) with two participants not reporting their biological sex. The majority of the participants reported their ethnicity as White/Caucasian (82%), with the remainder of the sample identifying as Hispanic/Latino/Latina (6%), Asian (5%), African American/Black (5%), and Native American (1%), and three participants not reporting their ethnic identification. The average age of the sample was 35.32 years (SD = 11.24, range = 18–75; 11 did not report).

Measures

The following measures constituted the primary variables of interest as this was a confirmatory study. Several additional measures that are not reported were included in the online survey for the purpose of secondary analysis.

Ideological self-identification

Self-reported political identification was assessed using the same single-item measure as in Study 1.

Disgust sensitivity

In addition to the Contamination Disgust subscale of the DS-R that was used in Study 1, the Contamination Obsessions and Washing Compulsions subscale of the Padua Inventory (PI-COWC; Burns, Keortge, Formea, & Sternberger, 1996) and the Perceived Vulnerability to Disease Scale (PVD; Duncan, Schaller, & Park, 2009) were included to assess disgust sensitivity, with a specific focus on sensitivity toward perceived contaminants. The PI-COWC is a 10-item subscale that asks participants to report the frequency of thoughts or behaviors associated with coming into contact with potential contaminants (e.g., “I find it difficult to touch garbage or dirty things”). Responses range from 0 (not at all) to 4 (very much). The PVD contains two subscales: The Germ Aversion subscale contains 8 items that assess the extent to which participants avoid contact with potential contaminants (e.g., “I prefer to wash my hands pretty soon after shaking someone’s hand.”) and is theorized to represent an intuitive measure of disgust sensitivity, whereas the Perceived Infectability subscale contains 7 items that assess participants beliefs about their likelihood of contracting infectious diseases (e.g., “If an illness is ‘going around,’ I will get it”) and is theorized to represent a rational (consciously reasoned) measure of disgust sensitivity. Participants reported their agreement with each item on a scale from −3 (strongly disagree) to +3 (strongly agree). Items within each of the subscales were averaged to create composite indices.

Vaccine attitudes

The same measure of vaccine attitudes was used as in Study 1. Overall, the distribution of vaccine attitudes had higher concentrations at the positive end of the scale (see Figure 1b), but measures of skewness (−.51) and kurtosis (−.51) were well within the acceptable range.

Procedure

The study was conducted following approval by the college’s IRB. The survey was created and hosted using SurveyMonkey™, and a link to the survey was posted on Amazon’s Mechanical Turk website. Participants self-selected into the study and completed the ideological self-identification, disgust sensitivity measures, and measures of vaccine attitudes before completing demographics questions. With the exception of the demographics questions, individual items within each measure were randomized.

Results and Discussion

Data missing at random were imputed prior to analyses using the same procedure as in Study 1. Preliminary analysis indicated that vaccine attitudes did not differ significantly based on participant sex, ethnicity, or age. As in Study 1, a significant correlation was found between vaccine attitudes and ideological self-identification r(241) = −.26, p < .001, such that participants identifying as more conservative tended to report more negative attitudes about vaccines.

Bivariate correlations were used to assess relations between disgust sensitivity measures and the vaccine attitudes measure (see Table 4). Contamination disgust, PI-COWC, and germ aversion were all significantly negatively correlated with attitudes toward vaccines (all ps < .02; see scatterplots in Figure 3), whereas perceived infectability did not demonstrate a significant relation with vaccine attitudes (p = .41). This distinction is important and indicates that the relation between disgust sensitivity and attitudes about vaccines is more likely to be based on intuitive aversion to contaminants as opposed to being based on reasoned beliefs about the probability of contracting an illness.

Table 4.
Bivariate Correlations, Means, Standard Deviations, and Reliability Coefficients (Cronbach’s α) for All Study 2 Variables.

n = 243 1 2 3 4 5 6 Mean SD

1. Contamination Disgust — .60 .58 .17 .17 −.32 3.52 1.23

2. PI-COWC — .71 .25 .07 −.20 1.20 .89

3. Germ Aversion — .30 .06 −.15 −.15 1.17

4. Perceived Infectability — .03 −.05 −.93 1.35

5. ISI −.26 3.15 1.74

6. Vaccine Attitudes — 1.48 1.78

Cronbach’s α .64 .91 .79 .92 NA .94

Note. Contamination Disgust = Contamination subscale of the Disgust Sensitivity Scale–Revised; PI-COWC = Contamination Obsessions and Washing Compulsions subscale of the PADUA Inventory; ISI = ideological self-identification (higher scores = more conservative self-identification); NA = not available.

Figure 3.
Scatterplots of contamination disgust, PI-COWC, and germ aversion versus vaccine attitudes in Study 2. PI-COWC = PADUA Inventory–Contamination Obsessions and Washing Compulsions.

As in Study 1, hierarchical linear regression was used to examine contamination disgust, PI-COWC, and germ aversion as simultaneous predictors of vaccine attitudes after controlling for ideological self-identification. Results from the analyses are presented in Table 5. The residuals of the model were normally distributed (skewness = .38, kurtosis = .24), and all other assumptions of regression were met.

Table 5.
Results of the Multiple Regression Analysis Investigating Disgust Subscales as Predictors of Vaccine Attitudes in Study 2.

n = 243 R ² B SE (b) 95% CI β p

Step 1—Control variable .07

ISI −.27 .06 [−.39, −.14] −.26 <.001

Step 2—Disgust Sensitivity .15

DS-R Contamination −.42 .11 [−.65, −.20] −.29 <.001

PI-COWC −.12 .18 [−.47, .23] −.06 .51

PVD—Germ Aversion .11 .13 [−.16, .37] .07 .43

Note*. R ² = proportion of variance in vaccine attitudes explained; b = unstandardized regression coefficient; β = standardized regression coefficient; ISI = ideological self-identification (higher scores = more conservative self-identification); DS-R = Disgust Sensitivity Scale–Revised; PI-COWC = PADUA Inventory—Contamination Obsessions and Washing Compulsions subscale; PVD = perceived vulnerability to disease measure; CI, confidence interval.

These results replicated the findings from Study 1, indicating that higher levels of dispositional disgust sensitivity predicted more negative attitudes toward vaccines over and above any variance that was simply related to ideology. Further, when all three Disgust Sensitivity subscales were entered into the model simultaneously, only contamination disgust predicted unique variance in vaccine attitudes, indicating that the observed correlations between the remaining two disgust subscales (PI-COWC and Germ Aversion) and vaccine attitudes were based on variance held in common with contamination disgust.

Finally, as in Study 1, a structural equation model evaluated contamination disgust, PI-COWC, and germ aversion as indicators of a latent contamination aversion construct, which, along with political ideology, was modeled as a predictor of vaccine attitudes. As in Study 1, the model provided adequate fit to the data, χ²(4, N = 243) = 16.18, p = .003, CFI = .96, RMSEA = .11, and the paths from the contamination aversion latent variable to all three disgust measures were statistically significant as was the path from contamination aversion to vaccine attitudes, while controlling for ideological self-identification (see Figure 4).

Figure 4.
Structural equation model with contamination aversion modeled as a latent variable from Study 2. ISI = ideological self-identification, DS-CONT = Contamination Disgust subscale of the Disgust Sensitivity measure, PI-COWC = Padua Inventory—Contamination Obsessions and Washing Compulsions subscale, GA = Germ Aversion subscale of the Perceived Vulnerability to Disease measure. Factor loadings for contamination disgust, political ideology, and vaccine attitude indicators were fixed (not estimated). ***p < .001.

Ultimately, participants with higher levels of disgust sensitivity, specifically those reporting higher levels of aversion to sources of contamination, held more negative attitudes toward vaccines.

General Discussion

The present studies provide further evidence about the relation between BIS mechanisms (i.e., disgust sensitivity) and attitudes about vaccines. Study 1 established the clear existence of a negative relation such that participants higher in disgust sensitivity tended to report more negative attitudes about vaccines. Additionally, it was observed that the relations between vaccine attitudes and disgust sensitivity measures were primarily associated with contamination aversion. Study 2 replicated the findings between contamination aversion and vaccine attitudes and demonstrated that the relation generalized to two additional measures of contamination aversion (germ aversion and PI-COWC) but not to a third (perceived infectability). The results indicated that variance that was common to the contamination disgust, PI-COWC, and germ aversion measures was driving the relation with vaccine attitudes. Taken together, these findings illustrate an important relation between disgust sensitivity and vaccine attitudes. Specifically, heightened aversion to perceived sources of contamination was predictive of more negative attitudes toward vaccines. This is particularly interesting because it suggests that vaccines, which are intended to be a tool for protecting against the spread of infectious diseases, might be viewed negatively due to psychological processes, which also evolved to reduce the likelihood of contracting disease.

The present research represents an initial step in developing an understanding of the relations between disgust sensitivity and attitudes about vaccines, and there is a clear path forward to understanding this relation further. First, it is theorized that the present results are indicative of a causal relation and that heightened aversion to perceived contaminants promotes more negative attitudes toward vaccines. It is essential that this proposed relation be validated with experimental studies, in which a heightened state of contamination aversion is induced and participants’ subsequent attitudes toward vaccines are evaluated. Prior research along these lines that manipulated pathogen disgust failed to produce differences in skepticism about vaccines (Clifford & Wendell, 2015). However, given the present findings that the relation between vaccine attitudes and disgust is particularly robust for measures assessing contamination aversion, further experimental studies are warranted. It is also crucial that future research investigate the extent to which people who hold more negative attitudes toward vaccines indicate that they perceive vaccines to be a potential source of contamination.

Assuming that the proposed causal relation holds, there are also important implications for public health that should be investigated. Specifically, the finding that the relation between disgust sensitivity and vaccine attitudes did not extend to the perceived infectability measure, which is more based in deliberative reasoning, could have important implications for public health campaigns. Communications designed to increase adherence to recommended vaccine schedules might benefit from addressing concerns over vaccine safety rather than promoting the vaccine’s effectiveness, a notion that is consistent with prior research indicating that people who chose to vaccinate and people who chose not to vaccinate differ very little with regard to their knowledge of vaccine effectiveness (Okoronkwo, Sieswerda, Cooper, Binette, & Todd, 2012). Further, classic work on attitude change and persuasion has found that emotionally based attitudes are more readily changed by emotionally based arguments (Fabrigar & Petty, 1999). Thus, attempts to improve attitudes toward vaccines by educating the public about empirical research findings may be less effective than addressing emotional concerns (e.g., disgust, fear).

Although the present findings represent an important first step in understanding the relation between disgust sensitivity and vaccine attitudes, they are solely based on self-report measures. In order to increase confidence in the validity of the relation and to develop a more grounded assessment of the impact of this relation to public health, future studies should utilize behavioral measures (adherence to recommended immunization schedules) to determine whether heightened disgust sensitivity is associated with reduced adherence. Such an understanding will help to quantify the value of developing interventions based on the disgust sensitivity–vaccine attitude relation.

Conclusion

Across two studies, self-report measures of dispositional disgust sensitivity (specifically, aversion to perceived sources of contamination) predicted negative attitudes toward vaccines. This relation held in both student and nonstudent samples and across multiple measures of disgust sensitivity. This finding represents an initial step in the investigation of the relation between disgust sensitivity and vaccine attitudes and illustrates the promise of a Darwinian approach to understanding attitudes about vaccines.

n = 375	1	2	3	4	5	6	7	8	9	Mean	SD
1. DS-R—Composite Score	—	NA	NA	NA	.59	.45	.12	−.02	−.20	4.90	1.05
2. DS-R—Core		—	.66	.52	.58	.38	.10	−.03	−.13	5.21	1.12
3. DS-R—Animal Reminder			—	.45	.43	.37	.09	−.07	−.15	4.78	1.35
4. DS-R—Contamination				—	.48	.40	.13	.10	−.29	4.35	1.28
5. TDD—Pathogen					—	.31	.21	.07	−.13	4.20	1.14
6. TDD—Sex						—	.26	.08	−.12	3.10	1.58
7. TDD—Moral							—	.02	.06	4.24	1.13
8. ISI								—	−.15	4.00	2.00
9. Vaccine attitudes									—	0.66	1.16
Cronbach’s α	.89	.84	.81	.60	.78	.84	.78	NA	0.84

n = 378	R ²	B	SE (b)	95% CI	β	p
Step 1—Control variable	.02
ISI		−.11	.04	[−.18, –.04]	−.15	.003
Step 2—Disgust Sensitivity	.10
DS-R Core		.05	.08	[−.11, .20]	.05	.54
DS-R Animal Reminder		−.07	.06	[−.19, .05]	−.08	.24
DS-R Contamination		−.25	.06	[−.36, −.14]	−.27	<.001
TDD Pathogen		.02	.06	[−.11, .14]	.02	.80
TDD Sex		.00	.04	[−.08, .09]	.01	.92

n = 243	1	2	3	4	5	6	Mean	SD
1. Contamination Disgust	—	.60	.58	.17	.17	−.32	3.52	1.23
2. PI-COWC		—	.71	.25	.07	−.20	1.20	.89
3. Germ Aversion			—	.30	.06	−.15	−.15	1.17
4. Perceived Infectability				—	.03	−.05	−.93	1.35
5. ISI						−.26	3.15	1.74
6. Vaccine Attitudes						—	1.48	1.78
Cronbach’s α	.64	.91	.79	.92	NA	.94

n = 243	R ²	B	SE (b)	95% CI	β	p
Step 1—Control variable	.07
ISI		−.27	.06	[−.39, −.14]	−.26	<.001
Step 2—Disgust Sensitivity	.15
DS-R Contamination		−.42	.11	[−.65, −.20]	−.29	<.001
PI-COWC		−.12	.18	[−.47, .23]	−.06	.51
PVD—Germ Aversion		.11	.13	[−.16, .37]	.07	.43

Footnotes

Acknowledgments

Russ Clay would like to thank John A. Terrizzi, Jr, Natalie J. Shook, and Vivian M. Rodriguez Archilla for feedback on early versions of the article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Note

References

Brown

K. F.

Kroll

J. S.

Hudson

M. J.

Ramsay

Green

Long

S. J.

… Sevdalis

(2010). Factors underlying parental decisions about combination childhood vaccinations including MMR: A systematic review. Vaccine, 28, 4235–4248. Retrieved from http://doi.org/10.1016/j.vaccine.2010.04.052

Burns

L. G.

Keortge

S. G.

Formea

G. M.

Sternberger

L. G.

(1996). Revision of the Padua inventory of obsessive compulsive disorder symptoms: Distinctions between worry, obsessions, and compulsions. Behaviour Research and Therapy, 34, 163–173. Retrieved from http://doi.org/10.1016/0005-7967(90)90087-Y

Clay

Terrizzi

J. A.

Jr Shook

N. J.

(2012). Individual differences in the behavioral immune system and the emergence of cultural systems. Social Psychology, 43, 174–184. Retrieved from http://psycnet.apa.org/journals/zsp/43/4/174/

Clifford

Wendell

D. G.

(2015). How disgust influences health purity attitudes—Online appendix. Political Behavior, 38, 1–3. Retrived from http://doi.org/10.1007/s11109-015-9310-z

Diamond

(1997). Guns, germs, and steel: The fates of human societies. New Yoirk, NY: W. W. Norton.

Dobson

A. P.

Carper

E. R.

(1996). Infectious Diseases and Human population History: Establishment of disease has been a side effect of the growth of civilization. BioScience, 46, 115–126.

Duncan

L. A.

Schaller

Park

J. H.

(2009). Perceived vulnerability to disease: Development and validation of a 15-item self-report instrument. Personality and Individual Differences, 47, 541–546. Retrieved from http://doi.org/10.1016/j.paid.2009.05.001

Fabrigar

L. R.

Petty

R. E.

(1999). The role of the affective and cognitive bases of attitudes in susceptibility to affectively and cognitively based persuasion. Personality and Social Psychology Bulletin, 25, 363–381. Retrieved from http://doi.org/10.1177/0146167299025003008

Faul

Erdfelder

Buchner

Lang

A.-G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149–1160. Retrieved from http://doi.org/10.3758/BRM.41.4.1149

10.

Faulkner

Schaller

Park

J. H.

Duncan

L. A.

(2004). Evolved disease-avoidance mechanisms and contemporary xenophobic attitudes. Group Processes & Intergroup Relations, 7, 333–353. Retrieved from http://doi.org/10.1177/1368430204046142

11.

Fernandez

(2015). California measles outbreak that started in disneyland is over: Health dept. Retrieved November 20, 2015, from http://www.nbcbayarea.com/news/local/California-Measles-Outbreak-That-Started-in-Disneyland-is-Over-Health-Dept-300297641.html

12.

Fincher

C. L.

Thornhill

Murray

D. R.

Schaller

(2008). Pathogen prevalence predicts human cross-cultural variability in individualism/collectivism. Proceedings. Biological Sciences/The Royal Society, 275, 1279–1285. Retrieved from http://doi.org/10.1098/rspb.2008.0094

13.

Goldstein

(2014). 10 reasons not to vaccinate your child. Retrieved from http://vactruth.com/2014/12/12/10-reasons-not-to-vaccinate/

14.

Honaker

King

Blackwell

(2014). Package “Amelia.” Retrieved from https://cran.r-project.org/web/packages/Amelia/Amelia.pdf

15.

Lieberman

Patrick

(2014). Are the behavioral immune system and pathogen disgust identical? Evolutionary Behavioral Sciences, 8, 244–250. Retrived from http://doi.org/https://dx-doi-org.web.bisu.edu.cn/10.1037/ebs0000018

16.

Mercola

(2011). Vaaccine can wreck your immune system. Retrieved from http://articles.mercola.com/sites/articles/archive/2011/04/29/the-emergence-of-vaccine-induced-diseases.aspx

17.

Nesse

R. M.

(2005). Natural selection and the regulation of defenses. A signal detection analysis of the smoke detector principle. Evolution and Human Behavior, 26, 88–105. Retrieved from http://doi.org/10.1016/j.evolhumbehav.2004.08.002

18.

Nesse

R. M.

Williams

G. C.

(1994). Why we get sick: The new science of Darwinian medicine. New York, NY: Times Books.

19.

Newport

. (2015). In U.S., percentage saying vaccines are vital dips slightly. Retrieved from http://www.gallup.com/poll/181844/percentage-saying-vaccines-vital-dips-slightly.aspx

20.

Okoronkwo

Sieswerda

L. E.

Cooper

Binette

Todd

(2012). Parental consent to HPV vaccination for their daughters : The effects of knowledge and attitudes. The Canadian Journal of Human Sexuality, 21, 117–127.

21.

Olatunji

B. O.

Haidt

McKay

David

(2008). Core, animal reminder, and contamination disgust: Three kinds of disgust with distinct personality, behavioral, physiological, and clinical correlates. Journal of Research in Personality, 42, 1243–1259. Retrieved from http://doi.org/10.1016/j.jrp.2008.03.009

22.

Olatunji

B. O.

Moretz

M. W.

McKay

Bjorklund

de Jong

P. J.

Haidt

… Schienle

. (2009). Confirming the three-factor structure of the Disgust Scale–Revised in eight countries. Journal of Cross-Cultural Psychology, 40, 234–255. Retrieved from http://doi.org/10.1177/0022022108328918

23.

Page

A. C.

(1994). Blood-injury phobia. Clinical Psychology Review, 14, 443–461. Retrieved from http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L24288530

24.

Rosseel

Oberski

Byrnes

Vanbrabant

Savalei

Merkle

… Barendse

(2015). Package “lavaan”. Retrieved from https://cran.r-project.org/web/packages/lavaan/lavaan.pdf

25.

Sawchuk

C. N.

Lohr

J. M.

Tolin

D. F.

Lee

T. C.

Kleinknecht

R. A.

(2000). Disgust sensitivity and contamination fears in spider and blood-injection-injury phobias. Behaviour Research and Therapy, 38, 753–762. Retrieved from http://doi.org/10.1016/S0005-7967(99)00093-5

26.

Schaller

(2006). Parasites, behavioral defenses, and the social psychological mechanisms through which cultures are evoked. Psychological Inquiry, 17, 96–137.

27.

Schaller

Park

J. H.

(2011). The behavioral immune system (and why it matters). Current Directions in Psychological Science, 20, 99–103. Retrieved from http://doi.org/10.1177/0963721411402596

28.

Schaller

Murray

D. R.

Bangerter

Schaller

(2015). Implications of the behavioural immune system for social behaviour and human health in the modern world. Philosophical Transactions of the Royal Society B, 370, 20140105. Retrieved from http://doi.org/10.1098/rstb.2014.0105

29.

Smith

K. B.

Oxley

Hibbing

M. V.

Alford

J. R.

Hibbing

J. R.

(2011). Disgust sensitivity and the neurophysiology of left-right political orientations. PloS One, 6, e25552. Retrieved from http://doi.org/10.1371/journal.pone.0025552

30.

Terrizzi

J. A.

Jr Clay

Shook

N. J.

(2014). Does the behavioral immune system prepare females to be religiously conservative and collectivistic? Personality and Social Psychology Bulletin, 40, 189–202. Retrieved from http://doi.org/10.1177/0146167213508792

31.

Tolin

D. F.

Lohr

J. M.

Sawchuk

C. N.

Lee

T. C.

(1997). Disgust and disgust sensitivity in blood-injection-injury and spider phobia. Behaviour Research and Therapy, 35, 949–953. Retrieved from http://doi.org/10.1016/S0005-7967(97)00048-X

32.

Tybur

J. M.

Lieberman

Griskevicius

(2009). Microbes, mating, and morality: Individual differences in three functional domains of disgust. Journal of Personality and Social Psychology, 97, 103–122. Retrieved from http://doi.org/10.1037/a0015474