Journal of Psychosomatic Research, Vol. 40, No. 5, pp.

495-501, 1996
Copyright © 1996 Elsevier Science Inc.
All rights reserved.
ELSEVIER 0022-3999/96 $15.00 + .00

0022-3999(95)00639-7

PREDICTING V A S O V A G A L REACTIONS
IN VOLUNTEER BLOOD DONORS

M I C H E L L E A. M E A D E , * C H R I S T O P H E R R. F R A N C E , *
and L I S A M. P E T E R S O N t

(Received 15 May 1995; accepted 24 November 1995)

Abstract- Measures of blood and injury sensitivity, pain sensitivity, and anxiety were examined as potential
predictors of syncope and related reactions in 215 volunteer blood donors. The Blood and Injury Fears
subscale of the Fear Survey Schedule was the best predictor for first-time donors (r[84] = 0.43, p<0.001),
whereas the Mutilation Questionnaire was the best predictor for experienced donors (r[127] = 0.3 l,p<0.01).
In contrast with previous studies, few significant predictions were observed for demographic variables
(e.g., age, body mass index, or number of previous donations), suggesting that psychological measures
may be better suited to the task of identifying high risk donors.

Keywords: Blood donation; Vasovagal syncope; Prediction.

Medical interventions that rely on transfusion o f whole blood and its c o m p o n e n t s

create an incessant d e m a n d for healthy blood donors. A l t h o u g h attempting to attract
new donors is one m e t h o d o f meeting this d e m a n d , retention o f previous d o n o r s is
significantly m o r e efficient. For example, it has been estimated that experienced donors
exist in sufficient numbers to meet d e m a n d 5-fold in most regions [1], and that these
d o n o r s are less likely to be denied the o p p o r t u n i t y to donate for health reasons [2].
Unfortunately, even a m o n g repeat d o n o r s the average n u m b e r o f lifetime blood
donations is less than 5 [2].
One factor that m a y limit d o n o r retention is the experience o f negative physical
reactions such as syncope. The incidence o f syncope, or fainting, in response to blood
d o n a t i o n has been estimated at approximately 3°7o [3], while associated reactions such
as weakness, dizziness, lightheadedness, and nausea are observed in a significantly
larger p r o p o r t i o n o f b l o o d donors [4, 5]. For individuals w h o have never donated,
concern over possible physical reactions is offered as a m a j o r deterrent [2, 5]. A m o n g
experienced donors, negative physical reactions are associated with decreased inten-
tions to d o n a t e in the future [6, 7] and fewer actual repeat donations [2]. Therefore,
one m e t h o d o f retaining donors m a y be to identify individuals at increased risk o f
reacting so that p r e d o n a t i o n instruction in distraction, applied tension, and other
coping strategies that reduce the risk o f syncopal reactions can be offered [8-10].

* Ohio University, Athens, Ohio, USA.

tOhio State University, Columbus, Ohio, USA.
Address correspondence to: Christopher France, Ph.D., Department of Psychology, 245 Porter Hall,
Ohio University, Athens OH USA 45701-2979.

495
496 M.A. MEADE et al.

Research on the prediction of syncope and related reactions to blood donation first
emerged in the early 1940s, when large supplies o f blood were required to meet wartime
needs [3, 11-13]. These studies indicated that adverse reactions were more likely
among donors who were young, female, donating for the first time, and had a history of
fainting. Subsequent investigations identified several additional predictors, including
lower body weight [14, 15], lower blood pressure [14-16], and decreased caffeine
intake on the day of donation [14]. One important limitation shared by many of
the existing studies is an assessment of syncope as a dichotomous, unidimensional
phenomenon. As a result, less extreme forms of disturbance of consciousness (e.g.,
lightheadedness) and associated aversive reactions have been either ignored or incorpo-
rated into the definition of syncope. Better understanding of the relationship between
negative physical reactions and donor retention will require a more complete assess-
ment of syncopal reactions.
Anticipatory anxiety has also been identified as a significant precursor of syncope
[4, 7], yet relatively little empirical attention has been devoted to psychological pre-
dictors of syncopal reactions to blood donation. Thus, it remains to be determined
how psychological variables compare with demographic information in the prediction
of blood donation reactions.
Findings from several studies concerning fainting in individuals with blood and
injury fears suggest that existing psychological measures may prove useful in this
regard. Kleinknecht and colleagues have demonstrated that individuals with a history
of fainting in response to blood and injury cues score significantly higher than nonfaint-
ers on the Mutilation Questionnaire [17-19], the Anxiety Sensitivity Index [18], and
a fear of human blood item on the Fear Survey Schedule [17]. Similarly, individuals
with blood phobia, as defined by extreme Mutilation Questionnaire scores, report
significantly higher scores on the anxiety sensitivity index [20], and such fears have
been hypothesized to be related to elevated levels of trait anxiety [21]. In sum, there
is considerable evidence to support the potential utility of psychological indices in
the prediction of syncopal reactions.
In the present study, a prospective design was used to assess psychological measures
of blood and injury sensitivity, pain sensitivity, and anxiety as potential predictors
of syncopal reactions in volunteer blood donors. To provide a comparative assess-
ment of predictive efficacy, established personal and demographic predictors of blood
donation reactions were also obtained.

METHOD
Participants
An undergraduate subject pool was screened for students who were willing to donate blood at campus
blood drives. A total of 311 individuals volunteered to donate blood and participate in the present study.
The volunteers were primarily female (64%), had a mean age of 19.6years (range= 17-40; SD = 2.9), and
had previouslydonated bloodan averageof 2.2 times(range = 0-30; SD = 3.1). Completedata wereobtained
from 215 donors. The remaining volunteers were either deferred by Red Cross personnel for reasons of
health and safety (e.g., anemia, recent infectious illness) or failed to complete all of the questionnaires.

Questionnaires
The Mutilation Questionnaire was designed to assess the verbal-cognitive component of blood and
injury fear. The scale consists of 30 true/false items and has acceptable levels of internal consistency
(a = 0.75-0.85). Criterion validity is supported by enhanced avoidance of blood and injury stimuli [22].
The Blood and Injury Fears Subscale was derived from a factor analysis of the Fear SurveySchedule
 Blood donation reactions 497

llI [23]. This subscale assesses fear and discomfort in response to situations and stimuli relating to bodily
injury, death, and illness. The respondent indicates the degree to which they are disturbed by each of 12
stimuli on a scale of 1 n o t at all to 5 very m u c h .
The Anxiety Sensitivity Index is a 16-item scale that assesses beliefs about anxiety and alertness to
fear-invoking stimuli [24]. This scale has good test-retest reliability (r= 0.75), and validity is supported
by elevated ratings among panic patients [25].
The Spielberger Trait Anxiety Inventory [26] consists of 20 statements concerning anxiety as a stable
characteristic. Respondents rate each item on a 4-point scale, from 1 n o t at all to 4 very m u c h so. The
trait anxiety scale has demonstrated acceptable internal consistency (a = 0.86-0.92), test-retest reliability
(0.73-0.86), and construct validity.
The Pain Sensitivity Index [27] assesses fear of pain as a personality characteristic. Responses to the
10-item scale are provided on an 8-point rating scale, from 0 very little to 7 very m u c h . This scale has
demonstrated acceptable internal consistency (Q = 0.89). Concurrent and criterion validity are supported
by significant correlations with other blood/injury fear scales and pain ratings during injection and den-
tal drilling.
The Blood Donation Reactions Inventory was developed for the present study. This inventory requires
respondents to rate 11 subjective physiological reactions associated with blood donation, including faint-
ness, dizziness, weakness, facial flushing, visual disturbance, difficulty hearing, lightheadedness, rapid or
pounding heartbeat, sweating, rapid or difficult breathing, and nausea or upset stomach. Each reaction
is endorsed on a 0 n o t at all to 5 to an e x t r e m e degree scale, and individual ratings are summed to create
a total score. Initial testing of this instrument at two prior blood drives revealed that this scale has a high
level of internal consistency (ct = 0.93) and was significantly correlated with phlebotomist ratings of donor
reactions (r[595] = 0.46, p<0.001).
The Caffeine Intake Survey [28] was used to assess the amount of caffeine consumed by participants
on the day of donation. A brief personal history questionnaire was given to obtain demographic information
as well as prior history of fainting and blood donation experience.

Procedure
One week before a scheduled campus blood drive, participants attended a group testing session and
completed the Mutilation Questionnaire, the Blood and Injury Subscale of the Fear Survey Schedule III,
the Anxiety Sensitivity Index, the Spielberger Trait Anxiety Inventory, the Pain Sensitivity Index, and a
personal information questionnaire. After completing the questionnaires, participants received verbal and
written reminders of the time and location o f the campus blood drive.
On the day of donation, participants completed the standard American Red Cross blood donation
protocol, which includes registration, health screening, donation of 450 mL of blood, and a brief postdona-
tion recovery period. While seated in the recovery area, participants completed the Blood Donation
Reactions Inventory and the Caffeine Intake Survey. The entire donation procedure, from entry to the
clinic until departure, lasted from 1 to 3 hours.

RESULTS

Scores on the Blood Donation Reactions Inventory were transformed to correct

for a significant positive skew in the raw data. ~Correlational analyses were then used
to examine the relationship between personal and psychological measures and total
score on the Blood Donation Reactions Inventory. Since prior donation experience
may influence the relationship between these variables, separate correlations were
conducted for first-time donors (n = 86) and individuals with one or more previous
donations (n = 129).

Personal and demographic variables

Few significant correlations were observed for the personal and demographic vari-
ables. Blood Donation Reaction Inventory scores were not significantly related to

Scores on the Blood Donation Reactions Inventory were submitted to a modified inverse transformation
[ - ( I / x + 1)]. Specifically, prior to inverse calculations, a constant of one was added to all raw data to
permit an inverse transformation of zero scores. The resulting values were then multiplied by - 1 to restore
the original relative order of the data.
498 M.A. MEADE et aL
Table I.-Correlations between Blood Donation Reactions Inventory and personal,
demographic, and psychological variables
Blood Donation Reactions Inventory
Experienced donors First-time donors
Measure Units (n = 129) (n = 86)
Age years - 0.12 0.10
Body Mass Index Ibs/in2 × 1000 - 0.05 - 0.15
Prior donations number - 0.16 N/A
Caffeine intake mg - 0.04 0.07
History of fainting yes/no 0.19' 0.10
Mutilation questionnaire score 0.31"* 0.36**
Blood and injury fears score 0.27** 0.43**
Anxiety Sensitivity Index score 0.22* 0.25*
Trait Anxiety Inventory score 0.26** 0.04
Pain Sensitivity Index score 0.27** 0.41"*
**p < 0.01; *p < 0.05; two-tailed.

age, body mass index, or caffeine intake on the day of donation (Table I). Likewise,
number of prior donations was not significantly correlated with reaction scores for
experienced donors. However, first-time donors did report significantly higher reaction
scores than experienced donors (F[1, 213] = 6.29, .o<0.05). A previous history of
fainting in response to blood and injury stimuli was positively correlated with donation
reaction ratings for experienced donors only ( r = 0.19, p<0,05). Finally, an analysis
of variance indicated that novice women donors reported significantly higher reaction
scores than men (F[ 1,84] = 4.04,p<0.05]. However, this effect was no longer significant
when body mass index was included in the analysis as a covariate. A gender effect
was not observed for experienced donors.

Psychological measures
Significant positive correlations were observed for each of the psychological mea-
sures (see Table I). With the exception of trait anxiety, which was significant for
experienced donors only, the psychological measures were slightly better predictors
of donation reaction scores among first-time donors. The single best predictor of
donation reactions a m o n g experienced donors was the Mutilation Questionnaire
(r=0.31, p<0.01), while the Blood and Injury Fears Subscale of the Fear Survey
Schedule was the best predictor for first-time donors ( r = 0.43,/9<0.01).
To examine the combined and unique contribution of the psychological measures
to the prediction of donation reactions, separate stepwise multiple regression analyses
were conducted for experienced and first-time donors. For experienced donors the
multiple regression was completed in 2 steps: the Mutilation Questionnaire was entered
on Step 1 (Fchange= 13.83, p<0.01; R z= 0.10), and trait anxiety was entered on Step
2 (F~hange= 4.30, ,O<0.05; R2= 0.13). With both variables in the equation, the partial
correlation coefficients were r = 0.27 for the Mutilation Questionnaire and r = 0.18
for trait anxiety. For first-time donors, the multiple regression was also completed
in 2 steps: the Blood and Injury Fears Subscale of the Fear Survey Schedule was
entered on Step 1 (Fchang~= 19.42, p<0.01; R 2= 0.19), and the Pain Sensitivity Index
entered on Step 2 ( F c h a n g e = 4.38, p<0.05; R z = 0.23]. With both variables in the equa-
 Blood donationreactions 499

tion, the partial correlation coefficients were r= 0.27 for the Fear Survey Schedule
and r = 0.22 for the Pain Sensitivity Index.

Deferred versus nondeferred volunteers

Because a significant proportion of the participants in this study were deferred by
the Red Cross, volunteers who completed the donation process were compared with
those who did not to assess potential factors associated with deferral. There were no
significant differences between the groups on any of the psychological variables,
however, significant group differences were observed for body mass index
(t[278] = 2.73, p<0.01) and gender ()(,2[1] = 6.74, p<0.01). These differences indicated
that deferred volunteers were smaller and that a greater proportion were female as
compared with nondeferred donors. This finding is not unexpected, given the increased
likelihood of anemia and low body weight among female volunteers.

DISCUSSION
The present findings indicate that psychological measures of blood and injury
sensitivity are significant predictors of syncope and related reactions in volunteer
blood donors. These results are consistent with previous suggestions that psychological
factors mediate syncopal reactions [4, 7] and highlight the advantage of a more precise
assessment of fear and anxiety among prospective donors. Although the Mutilation
Questionnaire and Fear Survey Schedule yielded the best prediction for experienced
and first-time donors, respectively, the high correlation between these instruments
(r= 0.80, p<0.001) suggests that a common focus on blood and injury stimuli was
central to their predictive power. This notion is supported by the results of regression
analyses conducted for both experienced and first-time donors. Specifically, only one
of these instruments was retained in the final regression equation for each subsample,
indicating a significant overlap in the proportion of the variance accounted for by
each measure. These analyses also revealed that trait anxiety predicted unique variance
beyond the Mutilation Questionnaire among experienced donors, while the Pain
Sensitivity Index explained unique variance beyond the Fear Survey Schedule for
first-time donors. One reason why the Pain Sensitivity Index explained unique variance
only among novice donors may be that more experienced donors tend to rate venipunc-
ture as less painful [29], possibly due to habituation or attrition of those most sensitive
to venipuncture pain.
Previous studies have implicated donor demographic and personal characteristics
as significant contributors to blood donation reactions [14-16]. In contrast, the present
findings fail to support a significant relationship between negative reaction ratings
and donor characteristics of age, body mass index, number of previous donations,
or caffeine intake. There are several potential explanations for this inconsistency.
First, results from previous studies are based on data collected from several thousand
donors, hence small effects were more likely to reach statistical levels of significance.
Second, because previous studies often defined syncope as a dichotomous variable,
variability in demographic factors may significantly predict extreme reactions, but
not the milder syncopal reactions assessed in this investigation. Third, the lack of
significant correlations for such variables as age and body mass index may be due to
the relatively homogeneous sample of young donors used in this study. This homogene-
500 M.A. MEADE et al.

ity may have constricted the range for many demographic and personal variables,
thereby limiting the possibility of observing larger correlation coefficients.
Consistent with previous studies, a higher incidence of negative donation reactions
was observed for women. This effect was eliminated after controlling for body mass
index, suggesting that gender differences were partially attributable to the smaller
size of the average female donor. Because the American Red Cross routinely collects
450 mL of blood from all donors, the proportion of total blood volume collected is
inversely related to body mass. As a result, smaller donors may experience a greater
decrease in blood pressure and an increased incidence of syncopal reactions.
In sum, the present findings suggest that psychological measures of blood and
injury sensitivity are significant predictors of syncope and related reactions to blood
donation. As a result, these scales may prove useful in identifying high-risk donors
who should benefit most from instruction in simple strategies to reduce syncopal
reactions [8-10]. If effective, these interventions would promote donor safety, increase
the likelihood of a positive donation experience, and ultimately enhance donor re-
tention.

Acknowledgments-The authors would like to thank Susan Pepper, Saundra Gentry, and Dr. Ambrose
Ng of the American Red Cross for their assistance with this project. This research was supported by an
Academic Challenge grant from the State of Ohio.

REFERENCES

1. WALLACE EL, PEGELS CC. Analysis and design of a model regional blood management system.
American Red Cross, Selected Readings in Donor Motivation and Recruitment 1974; 2: 71-77.
2. PILIAV1N JA. Why do they give the gift of life? A review of research on blood donors since 1977.
Transfusion 1990; 30: 444-459.
3. POLES FC, BOYCOTT M. Syncope in blood donors. Lancet 1942; 2: 531-535.
4. GRAHAM DT. Prediction of fainting in blood donors. Circulation 1961; 23: 901-906.
5. OSWALT RM. A review of blood donor motivation and recruitment. Transfusion 1977; 17: 123-135.
6. CALLERO PL, PILIAVIN JA. Developing a commitment to blood donation: the impact of one's
first experience. J A p p l Soc Psychol 1983; 13: 1-16.
7. KALOUPEK DG, SCOTT JR, KHATAMI V. Assessmentof coping strategies associated with syncope
in blood donors. JPsychosom Res 1985; 29: 207-214.
8. COHN CK, KRON RE, BRADY JP. A case of blood-illness-injury phobia treated behaviorally. J
Nerv Mental Dis 1976; 162: 65-68.
9. KALOUPEK DG, STOUPAKIS T. Coping with a stressful medical procedure: further investigation
with volunteer blood donors. J Behav Med 1985; 8:131-148.
10. OST LG, STERNER U. Applied tension: a specific behavioral method for treatment of blood phobia.
Behav Res Ther 1987; 25: 25-29.
I 1. BROWN H, McCORMACK P. An analysis of vasomotor phenomena occurring in blood donors. Br
Med J 1942; 1: 1-5.
12. GREENBURY CL. An analysis of the incidence of "fainting" in 5,897 unselected blood donors. Br
Med J 1942; 1: 253-255.
13. WILLIAMS GEO. Syncopal reactions in blood donors: an investigation of 222 cases. B r M e d J 1942;
1: 783-786.
14. KASPRISIN DO, GLYNN SH, TAYLOR F, MILLER KA. Moderate and severe reactions in blood
donors. Transfusion 1992; 32: 23-26.
15. OGATA H, IINUMA N, NAGASHIMA K, AKABANE T. Vasovagal reactions in blood donors.
Transfusion 1980; 20: 679-683.
16. CALLAHAN R, EDELMAN EB, SMITH MS, SMITH JJ. Study of the incidence and characteristics
of blood donor reactors. Transfusion 1963; 3: 76-82.
17. KLEINKNECHT RA. Vasovagal syncope and blood/injury fear. Behav Res Ther 1987; 25: 175-178.
18. KLEINKNECHT RA. Specificity and psychosocial correlates of blood/injury fear and fainting. Behav
Res Ther 1988; 26: 303-309.
 Blood donation reactions 501

19. KLE1NKNECHT RA, THORNDIKE RM. The mutilation questionnaire as a predictor of blood/
injury fear and fainting. Behav Res Ther 1990; 28: 429-437,
20. LUMLEY MA, MELAMED BG. Blood phobics and nonphobics: psychological differences and affect
during exposure. Behav Res Ther 1992; 30: 425-434.
21. PAGE AC. Blood-injury phobia. Clin Psychol Rev 1994; 14: 443-461.
22. KLORMAN R, HASTINGS J, WEERTS T, MELAMED B, LANG P. Psychometric description of
some specific-fear questionnaires. Behav Ther 1974; 5: 401-409.
23. ARRINDEL WA. Dimensional structure and psychopathology correlates of the fear survey schedule
(FSS-III) in the phobic population. Behav Res Ther 1980; 18: 229-242.
24. REISS S, PETERSON RA, GURSKY DM, McNALLY RJ. Anxiety sensitivity, anxiety frequency
and the prediction of fearfulness. Behav Res Ther 1986; 24: 1-8.
25. TELCH M J, SHERMIS MD, LUCAS JA. Anxiety sensitivity: unitary personality trait or domain-
specific appraisals? J Anxiety Disord 1989; 3: 25-32.
26. SPIELBERGER CD, GORSUCH RL, LUSHENE RE. Manual for the State-Trait Anxiety Inventory.
Palo Alto, CA: Consulting Psychologists Press, 1970.
27. GROSS PR. Is pain sensitivity associated with dental avoidance? Behav Res Ther 1992; 30: 7-13.
28. FRANCE C, DITTO B. Cardiovascular responses to the combination of caffeine and mental arithmetic,
cold pressor, and static exercise stressors. Psychophysiology 1992; 29: 272-282.
29. FRANCE C, ADLER PSJ, FRANCE J, DITTO B. Family history of hypertension and pain during
blood donation. Psychosom Med 1994; 56: 52-60.