The role of cognitions, trait anxiety and disgust sensitivity in generating faintness around blood–injury phobic stimuli

ARTICLE IN PRESS

Journal of Behavior Therapy

and Experimental Psychiatry 37 (2006) 41–52

0005-7916/$ -

doi:10.1016/j

�CorrespoE-mail ad

www.elsevier.com/locate/jbtep

The role of cognitions, trait anxiety and disgustsensitivity in generating faintness around

blood–injury phobic stimuli

Holly A. Exeter-Kent, Andrew C. Page�

School of Psychology, University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia

Abstract

The effects on blood–injury fear and fainting of scripts concerning pain, nausea, and anger and

individual differences in trait anxiety and disgust sensitivity were investigated. Eighteen participants

were high in disgust sensitivity and trait anxiety, 11 were low in disgust sensitivity but high in trait

anxiety, 10 were high in disgust sensitivity but low in trait anxiety, and 16 were low in disgust

sensitivity and trait anxiety. Participants were exposed to pain, nausea, and anger scripts during

presentation of blood–injury slides. The ability of the scripts to increase symptoms of fear and faint-

ness, on a state version of the Blood–Injection Symptom Scale (BISS; Page, A. C., Bennett, K. S.,

Carter, O., Smith, J., & Woodmore, K. (1997). Blood–Injection Symptom Scale (BISS): Assessing the

structure of phobic symptomatology elicited by blood and injections. Behaviour Research and

Therapy, 35, 457–464) were examined. Analyses indicated that individual differences in trait anxiety

and disgust sensitivity interact to generate symptoms of faintness when the pain script was read. That

is, disgust sensitive and trait anxious participants reported greater faintness relative to other

conditions. The implications for theory and treatment of blood–injury–injection phobia are

discussed.

r 2005 Elsevier Ltd. All rights reserved.

Keywords: Blood–injury phobia; Disgust; Cognitions; Fainting; Trait anxiety

see front matter r 2005 Elsevier Ltd. All rights reserved.

.jbtep.2005.09.004

nding author. Tel.:+61 9 6488 3577; fax: +619 6488 2655.

dress: [email protected] (A.C. Page).

www.elsevier.com/locate/jbtep

ARTICLE IN PRESSH.A. Exeter-Kent, A.C. Page / J. Behav. Ther. & Exp. Psychiat. 37 (2006) 41–5242

1. Introduction

Blood–injury phobia is a persistent, intense and irrational fear of situations involvingblood, injuries, wounds, and mutilation (Marks, 1988). Blood–injury phobia may lead toavoidance of life-threatening medical procedures (Kleinknecht & Lenz, 1989). Prevalenceestimates of blood–injury phobia range between 3.1% and 4.5% (Agras, Sylvester, &Oliveau, 1969; Bienvenu & Eaton, 1998; Costello, 1982; Fredrikson, Annas, Fischer,& Wik, 1996). It is the second most frequent specific phobia for which people seek help(Kleinknecht & Thorndike, 1990). Blood–injury phobia is an atypical phobia as themajority of sufferers faint in phobic situations (Connolly, Hallam, & Marks, 1976; Ost,Sterner, & Lindahl, 1984; Thyer & Curtis, 1985) and a component of this response appearsto be heritable (Page & Martin, 1998).Blood–injury phobia has in common with other specific phobias that fear involves

activation of the sympathetic nervous system (SNS). This activation is associated with thefight or flight response (Cannon, 1927), and co-ordinates the body’s responses to stresses(Rhoades & Pflanzer, 1992). Unlike other specific phobias, blood–injury phobia involves asecond autonomic nervous system response that entails activation of the parasympatheticnervous system (American Psychiatric Association, 1994; Ost, Lindahl, Sterner, &Jerremalm, 1984) that involves an abrupt and severe drop in blood pressure and heart rate,increases in blood glucose, cortisol, and human growth hormone and decreases ofnoradrenaline (Vingerhoets, 1984). As blood pressure drops, a fainting episode (orvasovagal syncope; Lewis, 1932) may occur. Thus, blood–injury phobia differs from otherspecific phobias in that it involves a diphasic autonomic nervous system response (Curtis &Thyer, 1983; Engel, 1978; Graham, 1961). Recent empirical support has been found forthis diphasic response in blood–injury phobics (Page, 2003).Although fear and fainting may co-exist within the same individuals (e.g., Ost, Sterner

et al., 1984; Page, 1996; Thyer & Curtis, 1985), Kleinknecht (1987) illustrated that thesetwo responses are partially independent. Kleinknecht (1987) found that some participantsreported being fearful around blood–injury, but did not report fainting whereas others,who fainted, did not report fear. Interestingly, Kleinknecht and Lenz (1989) found thatearly episodes of fainting were more predictive of present fear, than early episodes of fearwere predictive of present fainting. They suggested that fear might evolve out of severalfainting episodes (rather than vice versa). This study suggested that the onset of faintingpreceded the onset of fear. It may be postulated that the factors that lead to thedevelopment of fainting may interact with those that lead to fear, and in turn begs thequestion, what caused the initial fainting?In response to this question, Kleinknecht and Lenz (1989) proposed that blood–injury

fainters may be predisposed to react with vagal mediated responses to blood–injurystimuli, independent of fear, which is not seen in normal individuals. Others, such asDavey and colleagues have described a model that introduced disgust, in addition to fear inthe presentation of phobias (Davey, 1994; Matchett & Davey, 1991; Webb & Davey, 1992).Matchett and Davey examined the relationships between fear and disgust sensitivity todifferent categories of animals. Disgust sensitivity measured by the Disgust SensitivityScale (DSS; Rozin, Fallon, & Mandell, 1984) was related to fear-evoking animals but notto those that were associated with revulsion (e.g. maggot, snail, and slug). Webb andDavey (1992) extended this research to show that introducing violent video materialresulted in increased ratings of fear for animals in a high fear and high predatory category

ARTICLE IN PRESSH.A. Exeter-Kent, A.C. Page / J. Behav. Ther. & Exp. Psychiat. 37 (2006) 41–52 43

(e.g., lion). While exposure to a gory (i.e.. hospital operation) video resulted in increases infear ratings to animals classified as high fear and low predatory (e.g., rat) and to those inthe high revulsion category (e.g., slug). Thus, the categories of animals related to thepotential spread of disease, dirt, and contamination were rated as more fearful followingthe presentation of gory video material.

Disgust has also been discussed in relation to blood–injury phobia (Hepburn & Page, 1999;Page, 1994, 2003; Penner & Kleinknecht, 1994; Woody & Teachman, 2000). Penner andKleinknecht (1994) reported that participants with a history of vasovagal syncope reportedgreater fear and disgust to the relevant stimuli than did non-fainters. Thus a fainting historywas associated with greater fear. The authors proposed that vasovagal fainting might be elicitedby an interaction between a predisposition to experience fear and disgust around blood–injurystimuli and situational factors. In a similar vein, Page (1994, 2003) states that the tendency to beafraid might be associated with individual differences in trait anxiety (e.g., Andrews et al.,2003), whereas, the tendency to faint may be associated with individual differences in disgustsensitivity interacting with trait anxiety. Since disgust involves parasympathetic mechanisms itmay be associated with decreases in blood pressure. Thus, the elicitation of disgust concurrentlywith the homeostatic parasympathetic response recruited to attenuate the fight or flightresponse will produce a sufficient reduction in blood pressure to produce fainting.

Kleinknecht, Kleinknecht, and Thorndike (1997) used a correlational methodology toobserve relationships among individual differences in sensitivity to disgust. They foundthat their index of disgust sensitivity was negatively correlated with fainting anddisappeared when fear was taken into account. They concluded that the relationshipbetween disgust and fainting is illusory and is mediated by the covariation between disgustand fear. On the basis of these data one would predict that disgust would protect a personfrom fainting. This prediction contrasts with Page’s (1994) prediction which suggested thatthose most likely to report symptoms of faintness would be high in disgust sensitivity andhigh in trait anxiety in response to blood stimuli.

Another consideration is that the way a person processes information regarding a fearedobject may modify the form of the response. If participants were instructed to contemplateeither pain or disgust in response to the same stimulus, it is likely that the former would bemore likely to experience fear (Vlaeyen & Linton, 2000), and those instructed tocontemplate disgust would be more likely to experience faintness (Hepburn & Page, 1999;Page, 2003). Vlaeyen and Linton (2000) refer to a fear-avoidance model for chronic painthat involves fear of pain resulting in avoidance of activities that might involve pain andtherefore no opportunity to discover disconfirming evidence and thus, greater functionalimpairment resulting because of the avoidance. Clients with blood–injury phobia asked tocontemplate pain (e.g. through a script) would be thus more likely to experience fear inresponse to the stimuli, while those asked to contemplate disgust (e.g., nausea) may bemore likely to experience fainting (or feelings of faintness).

Bearing on this issue Hepburn and Page (1999) investigated whether scripts describingpain, nausea or disgust differentially interfered with the reduction in responding duringhabituation trials to bloody slides for a group of analog participants. They found thatwhen participants reported on their contents of consciousness, the scripts elicitedproportionally more mental images than thoughts. More importantly, when participantsheard the pain script and saw the bloody slide, they reported increases in faintness as wellas fear. Following the nausea script and blood slide, participants reported increases infaintness but not fear. Overall, disgust evoking images did not appear to retard habituation


or extinction of fear to the bloody slides. The findings of Hepburn and Page (1999)contrasts with the findings of Kleinknecht et al. (1997). One possible explanation of thedifferences between these studies is that Kleinknecht et al. only examined the additivecontributions of trait anxiety and disgust sensitivity, they did not examine the interaction(i.e., the multiplicative function). Thus, it remains to be determined if a different pattern ofresults were observed when the interaction is examined.Assuming that thoughts and images of pain will be associated with the emotions of

anxiety and fear, whereas thoughts and images of nausea will be associated with emotionsof disgust, it is apparent that a revision of existing models is needed to incorporate a moreintegrative model such as Page (1994). First, it is clear that disgust plays a stronger role ineliciting faintness than it does fear to a bloody stimulus. This is consistent with Page (1999,2003), and is not predicted by any theory that tries to explain faintness solely in terms offear. Second, it also appears clear that faintness is not solely caused by disgust. When fearis elicited, faintness also increased (Hepburn & Page, 1999).These accounts imply that variables predictive of disgust response will be associated with

fainting. Therefore, one hypothesis is that an individual difference measure of disgustsensitivity will be predictive of fainting. That is, people who are sensitive to disgust mayexhibit greater fainting responses in the presence of blood and injury. Alternatively, disgustsensitivity may interact with trait anxiety to increase symptoms of faintness triggered byblood and injury. For instance, the homeostatic processes that enable heart rate and bloodpressure to return to normal following fear-related increases may combine in amultiplicative manner with the processes involved in disgust that decrease heart rate.This would result in fear-related increases in sympathetic activation, followed by aparasympathetic overshoot. Consequently, the combination would be stronger than eitherprocess alone. Page (2003) supported the diphasic blood pressure response pattern inanalog participants reporting distress of blood and/or injections. Further, this diphasicpattern was most evident in those high in disgust. However, the study did not examine theeffects or interactions of disgust sensitivity and trait anxiety.To investigate the separate and combined effects of disgust sensitivity and trait anxiety,

and the effects of eliciting images associated with pain or nausea, reported fear andfaintness to slides depicting bloody injuries were examined. Participants were divided intoone of four groups on the basis of high and low trait anxiety, and high and low disgustsensitivity. An explanation of fainting around blood and injury in terms of individualdifferences pertaining to fear would predict that fainting scores would be higher forparticipants with elevated trait anxiety. An explanation of fainting around blood andinjury in terms of individual differences pertaining to disgust would predict that faintingscores would be higher for participants susceptible to disgust. Finally, an explanationpostulating an interaction between these variables, would expect that participants high intrait anxiety, who were also sensitive to disgust, would be most likely to experiencesymptoms of faintness around blood and injury.

2. Method

2.1. Design

The study used a 2� 2� (3) design. The between subjects factors designated whether thesubject was high or low in trait anxiety and high or low in disgust sensitivity. The within


subjects factor consisted of the three script conditions, with themes concerning pain,nausea, and anger. The dependent variables were change in self-reported fear and faintingmeasured on three separate occasions.

2.2. Participants

Fifty-five first year psychology university students who reported having concerns aroundblood and/or injury were divided into four groups on the basis of median splits onmeasures of trait anxiety and disgust sensitivity. Eighteen participants were high in disgustsensitivity and high on trait anxiety, 11 were low in disgust sensitivity but high on traitanxiety, 10 were high in disgust sensitivity but low on trait anxiety, and 16 were low indisgust sensitivity and low on trait anxiety (see Table 1 for the subject description). Of thefinal 55 participants; 16 were male and 40 female. All participants received course credit forparticipation.

2.3. Materials

Spielberger State-Trait Anxiety Inventory. The trait form of the State-Trait AnxietyInventory (STAI; Spielberger, Gorsuch, Lushene, Vogg, & Jacobs, 1983) consists of 20 selfreport items measuring relatively stable individual differences in proneness to anxiety. Thetest–retest reliabilities were .73 (male), and .77 (female), and the Cronbach alphacoefficient was .91 (male), and .93 (female; Spielberger et al., 1983). The mean score of theSTAI for working adults aged 19–39 for men is 36.54 and 36.17 for women (Spielbergeret al., 1983), which are comparable to values reported in Australian samples (e.g.,M ¼ 35:38; SD ¼ 8.65; Ray, 1984).

Disgust Sensitivity Scale. The Disgust Sensitivity Scale (DSS; Rozin et al., 1984) contains24 items that require ratings on a 9-point scale from dislike extremely to like extremely.Internal consistency on the total subject pool of introductory psychology students fromwhich the present participants were selected was high (coefficient a ¼ .96). The mean scoreof the DSS on a sample of university students (not pre-selected for any tendency to befearful or faint around blood–injury stimuli) was 109.6 (SD ¼ 34.8; Muris et al., 2000).This version of the scale was chosen in favor of the more recent version (Haidt, McCauley,& Rozin, 1994), because it specifically concerns food-related disgust alone. Therefore, there

Table 1

Descriptive statistics of the participants separated into two groups based on a median split of STAI and DSS

STAI

Low High

N M SD N M SD

Disgust sensitive STAI 10 32.5 4.0 18 46.4 6.5

DSS 10 79.0 13.6 18 70.8 14.6

Not disgust sensitive STAI 16 33.3 3.9 11 48.8 9.6

DSS 16 126.9 25.7 11 136.5 28.6

Note: The higher the DSS score the lower the disgust sensitivity.


is no obvious potential confound with blood fears. The more recent version contains itemsrelating to ‘‘body envelope violations,’’ that are related to blood–injury situations.Therefore, a high score on the more recent version could also indicate concerns related toblood–injury fears.

The Blood Injection Symptom Scale (BISS). The Blood Injection Symptom Scale (BISS;Page, Bennett, Carter, Smith, & Woodmore, 1997) is a 17-item scale, measuring symptomsof fear (i.e., sum of anxiety and tension subscales) and faintness around situationsinvolving blood and injections. Coefficient alphas were .86 and .70 for the faintness andfear subscales, respectively. The instructions were modified to permit an assessment ofsymptoms that occurred during critical phases of the experiment. Therefore, rather thanasking about the occurrence of symptoms during the worst experience with blood andinjections, the instructions were changed to read, ‘‘These questions, ask about sensationsyou may have experienced. For each sensation, circle ‘‘yes’’ if you noticed the sensation inthe last 5min (while viewing the slide) and circle ‘‘no’’ if you did not notice the sensation inthe last five minutes.’’ Hepburn and Page (1999) found such a modified version of the BISShad satisfactory psychometric properties in a sample of 91 participants. Based on the totalscale scores, the coefficient alpha was higher during the immediate reporting of symptomsduring the presentation of the slides (a ¼ :71) than in the period prior to presentation ofthe slides (a ¼ :61), indicating satisfactory internal consistency. They also reported aconservative estimate of test–retest reliability of .62.

Absorption assessment. Participants completed an 11-point Likert-type scale to assesstheir involvement in each experimental condition. The scale required the subject to ratehow absorbed they felt in the script, ranging from not at all absorbed, to extremelyabsorbed.

2.4. Stimuli

The three experimental scripts, descriptive of pain, nausea and anger, were adapted fromHepburn and Page (1999). Each 5-min script centered on a scenario set at a cinema andwas recorded onto a cassette. Words with direct reference to blood, injury, and injectionswere not included. The pain script is provided as an example below.

You are watching a movie. It is action packed. I would like you to imagine thesounds of the cinema. You can hear the sound track. It is fast and loud. It is a carcase scene. The police sirens are screaming, the cars collide together and you hear thecrunch of metal. The gunman scrambles from the wreck and grabs a pedestrian fromthe footpath.

Now you notice he is twisting the woman’s arm, pinning it back behind her. She isscreaming, struggling. The man puts a gun to her head. She can’t do anything. Younotice he has pulled her arm further back. You hear the woman scream and thebones in her arm snap. You hold your breath and notice a pain in your own arm.You can feel the pain in your own arm. You feel it pinned behind your back. Now itis throbbing. You see the man yank the woman’s arm. You gasp and feel an acutepain shoot up your own arm and into your shoulder and explode into your head.You hear a deafening crack as the woman’s shoulder is ripped from its socket. It isdislocated. You gasp and grimace. You notice you are sweating. You are gettinghotter and hotter. You can’t escape the roaring pain in your shoulder. Your arm is


burning. You feel the excruciating pain shooting down your back. Your vision blursand the screams begin to swim.

Now the music is too loud. You notice a pulsing pain in your head. You recognize amigraine. Your head is pounding. Boom. Boom. Still you can feel the agony in yourarm. Every movement sends lightning bolts of pain ripping through your body. Yougrit your teeth. You are sweating profusely. The pain is intense. You feel your heartbeating faster and faster as the pain becomes more and more intense. You wish toleave the cinema, but you can’t move. You can’t stand the pain any more. You take adeep breath and tears well up in your eyes.

Four images were used: a scenic (control) slide and three (experimental) slides depictingbloody injuries. The experimental slides depicted open wounds; two of the leg and one atthe ankle. Each 1� 1.5m image was projected using a slide projector on a 2� 3m whitescreen.

To check that all three scripts were equally absorbing, an additional ten independentjudges rated the scripts in terms of how absorbing the narratives were in the absence ofblood and injury stimuli. The absorption measure was a visual analogue scale ranging from0 representing ‘‘not at all absorbing’’ to 10 ‘‘extremely absorbing’’. These ratings wereanalyzed by a one-way ANOVA and found to be equally absorbing (F (2,18) ¼ 0.88;p ¼ ns). The judges also rated the degree to which each slide (in the absence of a script)could induce feelings of fear and faintness on separate 10-point visual analogue scales.One-way ANOVAs of the ratings indicated that there were no apparent differencesbetween slides in terms of their ability to induce fear (F (2,18) ¼ 0.78; p ¼ ns) or faintness(F (2,18) ¼ 1.13; p ¼ ns).

2.5. Procedure

Participants sat the STAI and DSS prior to the beginning of the experimentalmanipulation. Participants were selected on the basis of reported fainting or fear historyaround blood and injury. Participants attended (singly or in pairs) three sessions onseparate days within the same week. The participants sat two meters from the screen, withthe projector and cassette deck positioned behind them. The room was dimly litthroughout, but an overhead light was turned on during completion of the questionnaires.

In the first session, the procedure was outlined, and informed consent obtained. Thesession continued with the presentation of a 2-min focusing script. This script encouragedparticipants to relax their bodies and directed their attention to what followed. The scenic(control) slide was presented for the duration of the focusing script and completion of theinitial BISS questionnaire, which followed termination of the control slide. Immediatelyfollowing completion of the questionnaire, the blood–injury slide was presented togetherwith one of the 5-min experimental scripts. Upon completion of the script, the subjectcompleted the BISS for the second time. Finally, the 11-point rating scale, which indicatedtheir involvement in the experimental script, was administered. This procedure wasrepeated for each of the three sessions. One session involved presentation of the pain script,another the nausea script, and the third session involved the anger script. The order ofpresentation of the experimental scripts was counterbalanced across participants. Likewisethe order of slides was counterbalanced across participants and independently of scriptcontent.


Following completion of participation in the study the STAI and DSS questionnaireswere scored and participants were divided into high and low anxiety groups, and high andlow disgust sensitivity groups, based on their STAI and DSS scores, respectively, using amedian split process. Four groups resulted, high anxiety and high disgust sensitivity, highanxiety and low disgust, low anxiety and high disgust, and low anxiety and low disgust.

3. Results

3.1. Validity of experimental scripts

A rating scale was administered to each subject to examine any differences in the scripts’strengths that may confound interpretation of differences in reported symptomatology.There was no significant group effect (F (1,58) ¼ 0.16; p4:05), nor was there any group byscript interaction (F (2,116) ¼ 1.75; p4:05). However, there was a significant main effect ofscript (F (2,116) ¼ 6.13; po:005). This result occurred because the nausea script(M ¼ 7:13; SD ¼ 1.37) was rated as more absorbing than the pain (M ¼ 6:38; SD ¼1.64) or anger (M ¼ 6:43; SD ¼ 1.78) scripts. Thus, it is important in subsequent analysesto interpret cautiously any effects involving the nausea script.

3.2. Analyses

The procedure itself was effective in generating moderate levels of symptoms. The BISSgenerates a mean (for men and women) of 4.9 among students describing their worstexperience with blood and injections (Page et al., 1997) and the overall BISS score in thepresent investigation was 3.1. This placed the scores in the 40th percentile, consistent witha moderate level of symptom activation. For each of the three trials, participantscompleted the BISS on two occasions, once after the control slide and script and once afterthe experimental slide and script. To simplify presentation of the data, the BISS subscalesof fear and fainting after the control slide were subtracted from the equivalent ratings afterthe experimental slide. Two univariate repeated measures ANOVAs were conducted, onefor the fear subscale and one for the fainting subscale. Thus, the repeated measure waseither fear or faintness on the BISS on each assessment occasion. Separate ANOVAs wereconducted as hypotheses related to the about changes in fear and fainting separately (andnot the interaction of the two dependent measures). Planned contrasts tested the two groupmain effects (high versus low trait anxiety and high versus low disgust sensitivity), tworepeat main effects (anger versus pain script and anger versus nausea script), and theresulting two and three-way interactions.

3.3. BISS fear scores

The fear ratings elicited by the pain and anger scripts were equivalent (F (1,51) ¼ 1.19;p ¼ ns) while the fear elicited by the nausea script was significantly lower than that elicitedby the control (anger) script (F (1,51) ¼ 6.66; po:05; Fig. 1). No other group or repeatmain effects or interactions were significant.

ARTICLE IN PRESS

0.0

0.5

1.0

1.5

2.0

2.5

Cha

nge

in B

ISS

Fea

r

Anger NauseaScript

Pain

Fig. 1. Changes in BISS symptoms of fear (anxiety plus tension) while watching a bloody slide and hearing a

script relating to pain, nausea, and anger.

0.4

0.6

0.8

1.0

1.2

1.4

1.6

BIS

S F

aint

ness

Anger Nausea PainScript

Low Disgust; Low Anxiety Low Disgust; High Anxiety High Disgust; Low Anxiety High Disgust; High Anxiety

Fig. 2. BISS symptoms of faintness while watching a bloody slide and hearing a script relating to pain, nausea,

and anger for participants high and low in trait anxiety and disgust sensitivity.

H.A. Exeter-Kent, A.C. Page / J. Behav. Ther. & Exp. Psychiat. 37 (2006) 41–52 49

3.4. BISS faintness scores

The three-way interaction reached significance, revealing that participants high in traitanxiety and sensitive to disgust were more likely to experience symptoms of faintness whilehearing the pain script (F (1.51) ¼ 4.67; po:05; see Fig. 2) relative to other scriptconditions. The nausea condition was not-significantly different to anger (F (1,51) ¼ .54;p ¼ ns). The interaction between disgust sensitivity and trait anxiety failed to reachsignificance (F (1,51) ¼ 2.53; p ¼ ns). Neither the pain script (F (1,51) ¼ .11; p ¼ ns), northe nausea script (F (1,51) ¼ .94; p ¼ ns) elicited a different amount of faintness than theanger script, and nor did these effects interact significantly with disgust sensitivity or traitanxiety. Likewise, participants high and low in disgust (F (1,51) ¼ 2.00; p ¼ ns) and highand low in trait anxiety (F (1,51) ¼ .27; p ¼ ns) were not different in terms of the degree offaintness elicited by the stimuli.


4. Discussion

Considering the self-reported fear data first, it was apparent that both the anger andpain scripts produced equivalent increases in fear, and this was greater than the fearelicited by the nausea script in spite of the finding that the participants found the nauseascript more absorbing. However, symptoms of fear were unaffected by the individualdifference variables of trait anxiety and disgust sensitivity. One possible interpretation ofthis pattern of data is that cognitions concerning pain and anger both recruit the SNS (andthe associated fight and flight response) to a greater extent that does cognitions concerningnausea. As such, these data point to the importance for the production of symptoms of theinterpretation that people place upon stimuli involving blood and injury.Moving to consider the symptoms of faintness, it is clear that these contrasted with the

symptoms of fear. Symptoms of faintness resulted from an interaction of three variables. Ablood stimulus produced the greatest increase in faintness when participants who werehigh in both trait anxiety and disgust sensitivity were exposed to the pain script. Neithertrait anxiety nor disgust sensitivity was important in their own right, and the interaction ofthe two failed to produce the same strong pattern of responses when the participants werelistening to the nausea script. This is even though the participants had rated the nauseascript as the most absorbing. Thus, it may be that studies such as that reported byKleinknecht et al. (1997) may fail to identify a positive relationship between disgustsensitivity and faintness if they do not consider the particular stimuli in addition to theinteraction of individual difference variables, such as trait anxiety and disgust sensitivity.While the strength of this study was the use of change scores to control for pre-

experimental differences, several limitations of the current study must also be addressed.The first potential limitation is the under-powered nature of the study. It is difficult tomake statements about the absence of an interaction in predicting fear ratings with such asmall sample size. Perhaps a greater powered study will be more successful at finding thepredicted relationships. Secondly, the independent group of judges rated the level ofabsorption in the scripts to ensure that they were comparable. In the present study thenausea script was considered to be more absorbing that the other two scripts, thus futureresearch should use manipulations that are equally absorbing.Thirdly, the choice of instrument to assess disgust sensitivity may be a limitation. The

DSS (Rozin et al., 1984) was chosen to minimize a potential confound with blood fears, asthis scale specifically concerns food-related disgust alone. However, others may assert thatas a consequence that it is not perhaps the best estimate of disgust. In response to this itwould be interesting to see if the same pattern of results were replicated with disgustmeasured differently, as for example through a set of behavioral avoidance tests (e.g., Tsao& McKay, 2003). Fourthly, the present study used a median split for a continuousvariable; this was in part due to having a restricted range in the sample, as the participantswere university students and not phobic clients. Further, a question remaining to beanswered is whether the relationships in a clinical sample would be the same as those foundin this analog sample. If the relationships were qualitatively similar, then a multipleregression would be the better technique to use. Future research should aim to replicatethis pattern of findings on a clinical sample to examine if the same qualitative patternexists.Extrapolating from the present data to understanding fear and fainting around blood

and injury three points are clear. First, changes in the patterns of fear and fainting cannot


be explained by the same variables. Fear seemed to be influenced principally by direct fear-eliciting effects of the scripts in the present study, whereas faintness also interacted withtrait anxiety and disgust sensitivity. Second, changes in faintness cannot be understoodsolely with reference to pain alone or in terms of trait anxiety or disgust sensitivity alone.Third, since trait anxiety and disgust sensitivity interact, treatments that reduce either ofthese tendencies should be effective in reducing tendencies towards fainting, but treatmentsthat reduce both tendencies should be maximally effective. Thus, the present findings andtheir implications, suggest a model of blood–injury fear and fainting that contrasts withthe accounts described earlier. It appears that cognitions concerning pain play a principalrole in the development of both fear (Vlaeyen & Linton, 2000) and fainting (Hepburn &Page, 1999; Page, 2003). However, cognitions concerning pain exert a direct effect uponsymptoms of fear and an indirect effect on faintness. That is, thoughts of pain elicit feelingsof faintness only among people who are both sensitive to disgust and high in trait anxiety.At present, these extrapolations to clinical reactions around blood and injury must remainspeculative. Although they point to the variables that may be important targets intreatment trials, the participants in the present investigation were non-clinical individuals.A clinical sample may simply be quantitatively difference and thus have greater severity,which would suggest that the results of this study might be more profound in a clinicalsample. Alternatively, the clinical sample could be qualitatively different and the findingsnot be applicable. More research is needed to assess the generalizability of these findings.

References

Agras, S. W., Sylvester, D., & Oliveau, D. (1969). The epidemiology of common fears and phobias. Comprehensive

Psychiatry, 10, 439–447.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed).

Washington, DC: Author.

Andrews, G., Creamer, M., Crino, R., Hunt, C., Lampe, L., & Page, A. (2003). The treatment of anxiety disorders:

Clinician guides and patient manuals (2nd ed). New York: Cambridge University Press.

Bienvenu, O. J., & Eaton, W. W. (1998). The epidemiology of blood–injection–injury phobia. Psychological

Medicine, 28, 1129–1136.

Cannon, W. B. (1927). The James–Lange theory of emotion: A critical examination and an alternative theory.

American Journal of Psychology, 39, 106–124.

Connolly, J., Hallam, R. S., & Marks, I. M. (1976). Selective association of fainting with blood–injury–illness fear.

Behavior Therapy, 7, 8–13.

Costello, C. G. (1982). Fears and phobias in women: A community study. Journal of Abnormal Psychology, 91,

280–286.

Curtis, G. C., & Thyer, B. (1983). Fainting on exposure to phobic stimuli. American Journal of Psychiatry, 140,

771–774.

Davey, G. C. L. (1994). Disgust. In V. S. Ramachandran (Ed.), Encyclopedia of Human Behavior, Vol. 2.

New York: Academic Press.

Engel, G. L. (1978). Psychologic stress, vasodepressor (vasovagal) syncope, and sudden death. Annals of Internal

Medicine, 89, 403–412.

Fredrikson, M., Annas, P., Fischer, H., & Wik, G. (1996). Gender and age differences in the prevalence of specific

fears and phobias. Behaviour Research and Therapy, 34, 33–39.

Graham, D. T. (1961). Prediction of fainting in blood donors. Circulation, 23, 901–906.

Haidt, J., McCauley, C., & Rozin, P. (1994). Individual differences in sensitivity to disgust: A scale sampling seven

domains of disgust elicitors. Personality and Individual Differences, 16, 701–713.

Hepburn, T., & Page, A. C. (1999). Effects of images about fear and disgust upon responses to blood–injury

phobic stimuli. Behavior Therapy, 30, 63–77.


Kleinknecht, R. A. (1987). Vasovagal syncope and blood/injury fear. Behaviour Research and Therapy, 25,

175–178.

Kleinknecht, R. A., Kleinknecht, E. E., & Thorndike, R. M. (1997). The role of disgust and fear in blood and

injection-related fainting symptoms: A structural equation model. Behaviour Research and Therapy, 35,

1075–1087.

Kleinknecht, R. A., & Lenz, J. (1989). Blood/injury fear, fainting and avoidance of medically related situations:

A family correspondence study. Behaviour Research and Therapy, 27, 537–547.

Kleinknecht, R. A., & Thorndike, R. M. (1990). The mutilation questionnaire as a predictor of blood/injury fear

and fainting. Behaviour Research and Therapy, 28, 429–437.

Lewis, T. (1932). Vasovagal syncope and the carotid sinus mechanism. British Medical Journal, 1, 873–876.

Marks, I. M. (1988). Blood–injury phobia: A review. American Journal of Psychiatry, 145, 1207–1213.

Matchett, G., & Davey, G. C. L. (1991). A test of a disease-avoidance model of animal phobias. Behaviour

Research and Therapy, 29, 91–94.

Muris, P., Merckelbach, H., Nederkoorn, S., Rassin, E., Candel, I., & Horselenberg, R. (2000). Disgust and

psychopathological symptoms in a non clinical sample. Personality and Individual Differences, 29, 1163–1167.

Ost, L.-G., Lindahl, I., Sterner, U., & Jerremalm, A. (1984). Exposure in vivo vs applied relaxation in treatment of

the blood phobia. Behaviour Research and Therapy, 22, 205–216.

Ost, L.-G., Sterner, U., & Lindahl, I. (1984). Physiological responses in blood phobics. Behaviour Research and

Therapy, 22, 109–117.

Page, A. C. (1994). Blood–injury phobia. Clinical Psychology Review, 14, 443–461.

Page, A. C. (1996). Blood–injury–injection fears in medical practice. Medical Journal of Australia, 164, 189.

Page, A. C. (1999). Effects of images on the renewal of blood–injury fears. Behaviour Change, 16, 105–110.

Page, A. C. (2003). The role of disgust in faintness elicited by blood and injection stimuli. Journal of Anxiety

Disorders, 17, 45–58.

Page, A. C., Bennett, K. S., Carter, O., Smith, J., & Woodmore, K. (1997). Blood–Injection Symptom Scale

(BISS): Assessing the structure of phobic symptomatology elicited by blood and injections. Behaviour Research

and Therapy, 35, 457–464.

Page, A. C., & Martin, N. G. (1998). Testing a genetic structure of blood–injury–injection fears. American Journal

of Medical Genetics (Neuropsychiatric Genetics), 81, 377–384.

Penner, C.G., & Kleinknecht, R.A. (1994). The role of fear and disgust in emotional fainting to blood/injury

stimuli. Poster Presented at the 28th Annual Convention of the Association for Advancement of Behavior

Therapy, San Diego.

Ray, J. J. (1984). Measuring trait anxiety in general population samples. Journal of Social Psychology, 123,

189–193.

Rhoades, R., & Pflanzer, R. (1992). Human physiology (2nd ed). New York: Harcourt Brace.

Rozin, P., Fallon, A. E., & Mandell, R. (1984). Family Resemblance in attitudes to food. Developmental

Psychology, 20, 309–314.

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vogg, V. R., & Jacobs, G. A. (1983). State trait anxiety inventory

for adults. California: Consulting Psychologists Press Inc.

Thyer, B. A., & Curtis, G. C. (1985). On the diphasic nature of vasovagal of vasovagal fainting associated with

blood–injury–illness phobia. Pavlovian Journal of Biological Science, 20, 84–87.

Tsao, S. D., & McKay, D. (2003). Behavioural avoidance tests and disgust in contamination fears: Distinctions

from trait anxiety. Behaviour Research and Therapy, 42, 207–216.

Vingerhoets, A. J. J. M. (1984). Biochemical changes in two subjects succumbing to syncope. Psychosomatic

Medicine, 46, 95–103.

Vlaeyen, J. W. S., & Linton, S. J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal pain:

A state of the art. Pain, 85, 317–332.

Webb, K., & Davey, G. C. L. (1992). Disgust sensitivity and fear of animals: Effect of exposure to violent or

repulsive animals. Anxiety, Stress and Coping, 5, 329–335.

Woody, S. R., & Teachman, B. A. (2000). Intersection of disgust and fear: Normative and pathological views.

Clinical Psychology: Science and Practice, 7, 291–311.

Documents

The role of cognitions, trait anxiety and disgust sensitivity in generating faintness around blood–injury phobic stimuli