- Email: [email protected]
Body esteem as a moderator of cardiovascular stress responses in anatomy students viewing cadaver dissections
Journal of Psychosomatic Research 61 (2006) 501 – 506
Body esteem as a moderator of cardiovascular stress responses in anatomy students viewing cadaver dissections Brian M. Hughes a,4, Alexander Black b a
Department of Psychology, National University of Ireland, Galway, Ireland b Department of Anatomy, National University of Ireland, Galway, Ireland
Received 18 October 2005; received in revised form 11 April 2006; accepted 9 May 2006
Abstract Objective: Anatomy training requires students to take courses involving the dissection of human cadavers. It is suspected that such tuition is stressful for some students more than for others. Terror management theory suggests that the mortality salience presented by cadaver dissections would heighten stress among those students with positive self-images. The present study sought to investigate the association between body esteem (BE) and cardiovascular stress reactivity to cadaver dissections. Methods: The study involved the measurement of cardiovascular stress responses in 36 anatomy students during a dissection class.
Psychometric assessments of BE and other variables were taken. Results: Students with high BE had lower heart rate ( P=.011) and exhibited greater decreases in systolic blood pressure ( P=.008) and pulse pressure ( P=.003) throughout the session, suggestive of maladaptive stress response. Conclusions: As findings are consistent with theories linking BE with stress in occupations that require workers to view dead bodies, the consideration of BE may be an effective component in stress reduction interventions. D 2006 Elsevier Inc. All rights reserved.
Keywords: Anatomy training; Body esteem; Cardiovascular stress response; Terror management theory
Introduction For most people, seeing a dead body is an extremely distressing experience for cognitive, biological, and social reasons [1] and, as such, they will automatically avoid such exposure. However, in some occupations, exposure to dead bodies is a necessary facet of work. For example, emergency response personnel frequently encounter images of death in the course of their duties. A number of researchers have examined the impact of such exposure [2] and have considered variables that moderate the intensity of its stressfulness, such as age and perceived gruesomeness [3], and emotional attachment or identification [3,4]. In occupational settings, handling and viewing dead bodies have been implicated as a contributor to stress and anxiety disorders [2,4,5] and somatoform disorders [6]. 4 Corresponding author. Tel: +353 91 524411; fax: +353 91 521355. E-mail address: [email protected] (B.M. Hughes ). 0022-3999/06/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jpsychores.2006.05.004
Another group that customarily faces dead bodies comprises college students who are enrolled in programs where gross anatomy forms part of the curriculum (e.g., medical training). During their education, anatomy students participate in practical work involving the dissection of cadavers and cadaveric tissues. Whether exposure to dead bodies has deleterious effects on anatomy students has been the subject of a modest research literature. Questionnaire studies by faculty members have tended to report few lasting negative psychological effects of dissection (although initial trepidation among some students is acknowledged) [7–10]. It has been proposed, however, that due to the discomfort they themselves experience upon students’ reactions, anatomy faculty members are unlikely to fully acknowledge the stressfulness of dissections [11–13]. Research conducted by investigators other than faculty members has revealed a number of adverse consequences [1,11,14,15], as well as the finding that females exhibit more distress than males [16,17].
502
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
Given the negative effects of viewing dead bodies, as seen in other occupations, it stands to reason that anatomy dissection classes carry some risks. Stress may have a number of negative consequences for anatomy students. Firstly, stress may lead some students to terminate their studies [18]. Secondly, stress may elevate the risk of negative mental health in those who persist with training [18]. Thirdly, coping strategies such as depersonalization may damage the professional performance of students in their subsequent careers [16]. Fourthly, frequent and sustained stress responses are associated with an increased risk of ill physical health [19]. However, whether there are particular factors that predispose some students to adverse consequences has not been explored. One theoretical framework within which to examine the stress of exposure to death is terror management theory (TMT) [20 – 22]. TMT argues that people engage in various strategies (such as distraction or avoidance) in order to obviate the paralyzing terror that would otherwise result from contemplating the inevitability of death while possessing a self-preservation instinct. Proponents of TMT argue that maintaining a positive self-image goes some way to protect people from such existential anxieties. (Such argument is consistent with views that self-esteem is functionally adaptive and is not merely a bneedQ [23], that it protects against anxiety [21], and that it is largely subjective rather than based on objective cues [24]). A corollary is that reminders of death serve to make people’s self-images more salient. The conclusion, therefore, is that reminders of death are more threatening to individuals who rely heavily on selfimage as a stress buffer. One consequence of TMT is that anatomy students with highly positive self-images would be more likely to find dissections stressful. This would especially be the case in students with high esteem in terms of body image issues, as dissections present reminders of mortality that are specifically visual and body-related. Accordingly, high levels of body esteem (BE; the extent to which people value their bodies) might indicate a particular risk for stress in the anatomy laboratory. The present study sought to test this inference. Unlike previous research, the present study examined anatomy students’ stress objectively, assessing cardiovascular arousal patterns. Such patterns are widely employed as objective proxy measures of perceived stress because they are not subject to the problems of socially desirable responding that affect self-report stress indices.
Method Participants The participants were 36 anatomy students who were enrolled on the second year of a degree program in either medicine or science at a European university. The sample consisted of 29 women and 7 men, with a mean age of
19.67 years (S.D.=1.65 years). Participants had a mean height of 1.67 m (S.D.=.10 m) and a mean weight of 60.74 kg (S.D.=10.69 kg), yielding a mean body mass index of 21.69 kg/m2 (S.D.=2.80 kg/m2). Participants were enrolled on a volunteer basis and were not tangibly rewarded for participation. Psychometric instruments Four psychometric measures were used. The Body Esteem Scale (BES) [25] was used to assess BE. The BES is a 35-item instrument that measures positive and negative feelings toward different aspects of a respondent’s own body. Previous research has demonstrated the scale’s psychometric soundness [20]. The Rosenberg Self-Esteem Scale (RSES) [26] was used to assess self-esteem. The scale consists of 10 self-descriptive statements, for which agreement ratings are required. The scale has been identified as having sound psychometric qualities [27,28]. Trait anxiety was assessed using the trait form of the State–Trait Anxiety Inventory (STAI) [29], an inventory of 20 anxiety-related items requiring agreement ratings. It is widely utilized and is of repeatedly confirmed psychometric soundness. Finally, socially desirable responding was assessed using the Marlowe–Crowne Scale (MCS) [30]. The MCS is a 33-item scale requiring true–false judgments of self-descriptive statements, which is designed to establish whether the respondent has a tendency toward socially desirable responding in questionnaires. Procedure Participants were invited to take part following a lecture address in an anatomy class. Groups of six students were enrolled from each of six separate randomly generated groups, all of whom presented for anatomy practicals on the same week. Participants were invited to present at their subsequent dissection practicals 30 min prior to the scheduled start time. On each occasion, participants were met at the laboratory and brought to an adjoining room where they completed psychometric instruments, after which they had an opportunity to relax. The dissection practical chosen for the study took place halfway through a 12-week teaching semester and required students to witness the dissection of a human head by a faculty member. All students had previously been involved in hands-on dissection exercises on other parts of the body; this was the first opportunity for them to see the head and the face of the cadaver that they had been working with over the preceding weeks. Participants stood in close proximity to the cadaver throughout the dissection; during the practical, they were instructed to identify anatomical structures and were occasionally required to touch the cadaver on the surface of the face and inside the mouth. At the commencement of the practical (i.e., at 0 min), each participants’ blood pressure and heart rate (HR) were
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
503
Table 1 Psychometric characteristics of the sample (n=36) BEa Self-esteemb Trait anxietyc Social desirabilityd a b c d
Mean
S.D.
44.19 11.09 23.06 15.89
16.90 3.98 8.18 3.80
BES scores (possible range, 0 –140). RSES scores (possible range, 0 –30). STAI scores (possible range, 0 – 60). MCS scores (possible range, 0 – 33).
recorded by a research assistant using an Omron 637IT blood pressure monitor, which contains an automated sphygmomanometer that was thoroughly evaluated using international standardization protocols [31,32]. Subsequent readings of blood pressure and HR were taken after 30, 60, and 90 min of the practical.
Results Descriptive statistics and analytic strategy Psychometric features of the sample are presented in Table 1. Self-esteem was found to be correlated (Pearson) positively with BE (r 36=.61, Pb.001) and negatively with trait anxiety (r 36= .54, P=.001). Scores on the MCS did not correlate significantly with other scores, suggesting that BE, self-esteem, and trait anxiety were free from social desirability bias. However, whereas reliability analyses suggested satisfactory internal consistency for the RSES (a=.79), the BES (a=.91), and the STAI (a=.85), the MCS was found to have poor internal consistency (a=.46). As such, MCS scores were not included in subsequent analyses. Measures of systolic blood pressure (SBP), diastolic blood pressure (DBP), HR, and pulse pressure (PP) for each time point in the study are shown in Table 2. Participants were divided into low-BE, medium-BE, and high-BE groups (n=12 each) on the basis of a tertile split. A one-way analysis of variance (ANOVA) confirmed the validity of the split [ F(2,35)=78.59, Pb.001], with mean BE scores for each group being significantly different from the other two ( Pb.001 for each multiple comparison; Bonferroni comparisons used for all ANOVA post hoc tests). Further one-way ANOVAs revealed no significant between-group differences in age ( P=.08), body mass index
Fig. 1. SBP across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
( P=.42), or trait anxiety ( P=.40), implying that there was no need to statistically control for these variables in subsequent analyses involving BE. However, there were significant differences in self-esteem [ F(2,35)=7.90, P=.002], with the low-BE group having a lower self-esteem (M=1.82, S.D.=.37) than the high-BE group (M=2.36, S.D.=.33, P=.001). One-way ANOVAs revealed no between-group differences in initial SBP ( P=.13), DBP ( P=.65), HR ( P=.31), or PP ( P=.11). The seven male participants were distributed quite evenly across the groups (one in the lowBE group and three in each of the others). In order to assess the relationship between BE and cardiovascular functioning throughout the dissection, a series of two-way mixed ANOVAs was conducted. In each ANOVA, the repeated-measures variable, Time, had four levels (0, 30, 60, and 90 min), and the between-subjects variable, BE, had three levels (low, medium, and high). A separate ANOVA was conducted for each cardiovascular dependent variable. Given that there were between-subject differences in selfesteem, all analyses were followed up by equivalent analyses that control statistically for self-esteem. Throughout the study, the Mauchley test showed that the assumption of sphericity was not violated for repeated measures. SBP and DBP ANOVA for SBP revealed a significant interaction between BE and Time [ F(6,99)=2.58, P=.023], suggesting
Table 2 Mean cardiovascular measures (with standard deviations) taken at each time point 0 min SBP (mmHg) DBP (mmHg) HR (beats/min) PP (mmHg)
118.06 77.17 89.14 40.89
30 min (12.07) (8.60) (17.05) (8.09)
115.17 76.58 89.53 38.58
60 min (9.91) (8.53) (15.63) (6.16)
114.86 76.44 85.50 38.42
90 min (11.29) (9.45) (16.34) (6.97)
114.39 76.50 84.78 37.89
(10.92) (8.79) (15.38) (6.85)
504
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
Covariance analyses
Fig. 2. HR across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
that there were between-group differences in SBP changes during the practical (see Fig. 1). To explore this interaction, post hoc one-way repeated-measures ANOVAs were conducted for each group. For the high-BE group, a main effect for Time was revealed [ F(3,33)=4.17, P=.013], with SBP decreasing over the first three measures. SBP at 0 min was higher than that at 30 min ( P=.018) and that at 60 min ( P=.028). SBP at 90 min was not significantly different from that at 60 min. There were no such differences for the low-BE or the medium-BE group. Reconducting analyses while controlling for self-esteem revealed no new effects. Corresponding analyses revealed no significant relationships for DBP. Heart rate For HR, ANOVA revealed a main effect for Time [ F(3,99)=3.89, P=.011], with post hoc tests showing a decrease over time (see Fig. 2). HRs at 0 and 30 min were both significantly higher than that at 90 min ( P=.029 and .014, respectively). HR at 0 min was significantly higher than that at 60 min ( P=.029). There was neither a significant main effect for BE [although the effect approached significance; F(2,33)=2.94, P=.07] nor a significant TimeBE interaction. No significant main or interaction effect involving self-esteem was found when analyses were reconducted using a similar analysis of covariance (ANCOVA), as with SBP.
Given the fact that tertile splits transform continuous data into categories and thus result in loss of statistical power, intratrial cardiovascular functioning was further explored using covariance analyses. For each dependent variable, a one-way repeated-measures ANCOVA was conducted. Each analysis corresponded to the ANOVA conducted earlier, except that, instead of entering BE as a fixed-factor independent variable, raw BE scores were entered as a covariate in each case. However, given that covariance analysis requires a number of conditions that were not clearly met in the present data, these results should be interpreted conservatively. The ANCOVA for SBP confirmed the presence of a significant TimeBE interaction [ F(3,102)=4.13, P=.008]. Unlike ANOVA, ANCOVA revealed a significant main effect for Time [ F(3,102)=3.00, P=.034]. SBP at 0 min was significantly higher than that at 30 min ( P=.041) or that at 60 min ( P=.037) (see Fig. 1). The ANCOVA for DBP confirmed the absence of effects for that variable. The ANCOVA for HR confirmed the presence of a significant main effect for Time [ F(3,102)=3.01, P=.034]. However, unlike in ANOVA where the main effect for BE approached significance, in ANCOVA, the effect reached significance [ F(1,34)=4.64, P=.039]. This reflected a significant inverse correlation between BE and overall HR (r 36= .35, P=.039). Finally, the ANCOVA for PP failed to detect the main effect for BE observed in ANOVA. However, unlike ANOVA, ANCOVA revealed a significant main effect for Time [ F(3,102)=3.54, P=.017; see Fig. 3] and a significant TimeBE interaction [ F(3,102)=4.83, P=.003]. Post hoc tests showed PP at 0 min to be higher than that at 30 min ( P=.048). The TimeBE interaction reflected strong positive correlations between BE and PP at 0 min (r 36=.52, P=.001) and 30 min (r 36=.34, P=.042), but correlations
Pulse pressure For PP, ANOVA revealed a significant main effect for BE [ F(2,33)=3.85, P=.032]. Low-BE participants had lower overall PP than high-BE participants ( P=.038; see Fig. 3). Reconducting these analyses with self-esteem entered as a covariate revealed no additional effects involving self-esteem.
Fig. 3. PP across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
505
Fig. 4. Associations between PP at each time point and BE across the sample (n=36).
close to zero at 60 min (r 36= .01, P=.95) and 90 min (r 36= .02, P=.91) (see Fig. 4).
Discussion In summary, the data indicate a pattern of cardiovascular functioning throughout the dissection that implicates BE as a moderator of perceived stress. There was an overall tendency for cardiovascular activity to decrease over time (for SBP, HR, and PP); however, students identified as having high levels of BE exhibited greater decreases (particularly in SBP and PP). High-BE participants had lower HR throughout the entire practical. Finally, even though there were differences in self-esteem across the three BE groups, self-esteem was not found to have any bearing on observed associations. The overall decrease in cardiovascular functioning across time is consistent with the phenomenon of habituation and, as such, corroborates past self-report research [9]. Students began the dissection in a state of (anticipatory) arousal, which dissipated as they became familiar with the situation. However, the distinctly greater decreases in SBP and PP among the high-BE group warrant further explanation. These decreases were probably due to two factors. Firstly, high-BE participants may have demonstrated enhanced levels of initial arousal because of greater anticipatory anxiety. Secondly, the sharper reduction in blood pressure among these participants may suggest that they were at greater risk for vasovagal syncope. Both explanations would suggest that high-BE students adjusted relatively poorly to the stressful situation and were at relatively greater risk for adverse stress responses. This is further suggested by the inverse correlation between BE and overall HR. This implied that students with high levels of BE operated with reduced arousal, which again is a risk factor for syncope. Effects regarding PP are explicable in terms of those for SBP, given the absence of effects for DBP. BE was not associated with trait anxiety, and the role of BE in cardiovascular functioning was independent of selfesteem. This suggests that BE plays a unique role in the relationship between stress and arousal in the particular
context under scrutiny. In the present study, high levels of esteem (viz., BE) were found to be maladaptive in dealing with stress. Although high esteem is considered valuable in many contexts, there are disadvantages in situations where esteem is strongly threatened. The fact that high BE was associated with indicators of anxiety and fainting risk is consistent with the prediction that mortality salience will affect those for whom the body is a source of esteem. In terms of TMT, students whose bodies are a source of stressbuffering esteem suffered exaggerated stress in response to a stimulus that directly reminded them of the vulnerability of that source. One practical implication of these findings is that psychometric assessments of BE might be useful as screening tools to identify students who are likely to find dissections stressful. Correspondingly, interventions to ameliorate the stressfulness of dissections may benefit from incorporating BE-related information. Prospective students’ attention could be drawn toward the arbitrariness of valuing physical appearance over other personal attributes, in an attempt to reduce the negative impact of seeing a physical body being dissected. Such conclusions may have similar implications for other occupational contexts where dead bodies are viewed. Among the limitations of this type of research is that the intensive nature of data collection limits sample size. However, in the present study, the four-level repeatedmeasures design compensated for this with statistical power. Future research could seek to employ continuous, rather than periodic, cardiovascular measurement; further data from male students would enhance the generalizability of the findings. Future studies could explore the risk of syncope more directly, either by measuring HR variability or by examining behavioral indices such as dizziness or blushing. Further research could also examine whether the observed effects were contingent on the particular setting by, for example, recording responses to other dissection types and/ or didactic exercises. Finally, a separate assessment of resting cardiovascular functioning outside the laboratory would shed further light on the mechanisms underlying arousal and syncope. Given the human anxiety associated with compulsory exposure to cadavers during medical and scientific training, and the potential for developing effective stress-
506
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
reducing interventions, further investigation on the role of BE in this domain is warranted.
References [1] Sanner MA. Encountering the dead body: experiences of medical students in their anatomy and pathology training. Omega J Death Dying 1997;35:173 – 91. [2] Newhill CE, Sites EW. Identifying human remains following an air disaster: the role of social work. Soc Work Health Care 2000; 31:85 – 105. [3] McCarroll JE, Ursano RJ, Fullerton CS, Oates GL, Ventis WL, Friedman H, Shean GL, Wright KM. Gruesomeness, emotional attachment, and personal threat: dimensions of the anticipated stress of body recovery. J Trauma Stress 1995;8:343 – 9. [4] Ursano RJ, McCarroll JE. The nature of a traumatic stressor: handling dead bodies. J Nerv Ment Dis 1990;178:396 – 8. [5] Carson MA, Paulus LA, Lasko NB, Metzger LJ, Wolfe J, Orr SP, Pitman RK. Psychophysiologic assessment of posttraumatic stress disorder in Vietnam nurse veterans who witnessed injury or death. J Consult Clin Psychol 2000;68:890 – 7. [6] Labatte LA, Cardena E, Dimitreva J, Roy M, Engel CC. Psychiatric syndromes in Persian Gulf War veterans: an association of handling dead bodies with somatoform disorders. Psychother Psychosom 1998; 67:275 – 9. [7] Dinsmore CE, Daugherty S, Zeitz HJ. Student responses to the gross anatomy laboratory in a medical curriculum. Clin Anat 2001; 14:231 – 6. [8] McGarvey MA, Farrell T, Conroy RM, Kandiah S, Monkhouse WS. Dissection: a positive experience. Clin Anat 2001;14:227 – 30. [9] Arra´ez-Aybar L-A, Casado-Morales MI, Castan˜o-Collado G. Anxiety and dissection of the human cadaver: an unsolvable relationship. Anat Rec B New Anat 2004;279B:16 – 23. [10] Houwink AP, Kurup AN, Kollars JP, Kral Kollars CA, Carmichael SW, Pawlina W. Help of third-year medical students decreases first-year medical students’ negative psychological reactions on the first day of gross anatomy dissection. Clin Anat 2004;17:328 – 33. [11] William JL. Don’t discuss it: reconciling illness, dying, and death in a medical school anatomy laboratory. Fam Syst Med 1992;10:65 – 78. [12] Arra´ez-Aybar L-A, Castan˜o-Collado G, Casado-Morales MI. Dissection from the Spanish anatomist’s perspective: aims, attitudes, and related aspects. Anat Rec B New Anat 2004;281B:15 – 20. [13] Tschernig T, Schlaud M, Pabst R. Emotional reactions of medical students to dissecting human bodies: a conceptual approach and its evaluation. Anat Rec B 2000;261:11 – 3. [14] Horne DJ, Tiller JW, Eizenberg N, Tashevska M, Biddle N. Reactions of first year medical students to their initial encounter with a cadaver in the dissecting room. Acad Med 1990;65:645 – 6. [15] Tuohimaa P, Tamminen T, Fabrin V. Is it appropriate to speak of death during a dissection course? Teach Learn Med 1993;5:169 – 73.
[16] Charlton R, Dovey SM, Jones DG, Blunt A. Effects of cadaver dissection on the attitudes of medical students. Med Educ 1994; 28:290 – 5. [17] Eid J, Thayer JF, Johnsen BH. Measuring post-traumatic stress: a psychometric evaluation of symptom- and coping questionnaires based on a Norwegian sample. Scand J Psychol 1999;40:101 – 8. [18] Finkelstein P, Mathers L. Post-traumatic stress among medical students in the anatomy dissection laboratory. Clin Anat 1990;3: 219 – 26. [19] Lovallo WR, Gerin W. Psychophysiological reactivity: mechanisms and pathways to cardiovascular disease. Psychosom Med 2003;65: 36 – 45. [20] Goldenberg JL, McCoy SK, Pyszczynski T, Greenberg J, Solomon S. The body as a source of self-esteem: the effect of mortality salience on identification with one’s body, interest in sex, and appearance monitoring. J Pers Soc Psychol 2000;79:118 – 30. [21] Greenberg J, Pyszczynski T, Solomon S. The causes and consequences of the need for self-esteem: a terror management theory. In: Baumeister RF, editor. Public self and private self. New York7 Springer-Verlag, 1986. pp. 189 – 212. [22] Solomon S, Greenberg J, Pyszczynski T. A terror management theory of social behavior: the psychological functions of self-esteem and cultural worldviews. In: Zanna MP, editor. Advances in experimental social psychology. San Diego7 Academic Press, 1991. pp. 91 – 159. [23] Stevens A, Price J. Evolutionary psychiatry: a new beginning. London7 Routledge, 1996. [24] Pelham BW. Self-investment and self-esteem: evidence for a Jamesian model of self-worth. J Pers Soc Psychol 1995;69:1141 – 50. [25] Franzoi SL, Shields SA. The Body-Esteem Scale: multidimensional structure and sex differences in a college population. J Pers Assess 1984;48:173 – 8. [26] Rosenberg M. Society and the adolescent self-image. Princeton (NJ)7 Princeton University Press, 1965. [27] Gray-Little B, Williams VSL, Hancock TD. An item response theory analysis of the Rosenberg Self-Esteem Scale. Pers Soc Psychol Bull 1997;23:443 – 51. [28] Banos RM, Guillen V. Psychometric characteristics in normal and social phobic samples for a Spanish version of the Rosenberg SelfEsteem Scale. Psychol Rep 2000;87:269 – 74. [29] Spielberger CD, Gorusch RL, Lushene R. Manual for the State– Trait Anxiety Inventory. Palo Alto (Calif)7 Consulting Psychologists Press, 1970. [30] Crowne DP, Marlowe D. A new scale of social desirability independent of psychopathology. J Consult Psychol 1960;24:349 – 54. [31] Eckert S, Gleichmann U, Zagorski O, Klapp A. Validation of the Omron R3 blood pressure self-measuring device through simultaneous comparative invasive measurements according to protocol 58130 of the German Institute for Validation. Blood Press Monit 1997;2:189 – 92. [32] Watson S, Wenzel RR, di Matteo C, Meier B, Luscher TF. Accuracy of a new wrist cuff oscillometric blood pressure device: comparisons with intraarterial and mercury manometer measurements. Am J Hypertens 1998;11:1469 – 74.
Body esteem as a moderator of cardiovascular stress responses in anatomy students viewing cadaver dissections Brian M. Hughes a,4, Alexander Black b a
Department of Psychology, National University of Ireland, Galway, Ireland b Department of Anatomy, National University of Ireland, Galway, Ireland
Received 18 October 2005; received in revised form 11 April 2006; accepted 9 May 2006
Abstract Objective: Anatomy training requires students to take courses involving the dissection of human cadavers. It is suspected that such tuition is stressful for some students more than for others. Terror management theory suggests that the mortality salience presented by cadaver dissections would heighten stress among those students with positive self-images. The present study sought to investigate the association between body esteem (BE) and cardiovascular stress reactivity to cadaver dissections. Methods: The study involved the measurement of cardiovascular stress responses in 36 anatomy students during a dissection class.
Psychometric assessments of BE and other variables were taken. Results: Students with high BE had lower heart rate ( P=.011) and exhibited greater decreases in systolic blood pressure ( P=.008) and pulse pressure ( P=.003) throughout the session, suggestive of maladaptive stress response. Conclusions: As findings are consistent with theories linking BE with stress in occupations that require workers to view dead bodies, the consideration of BE may be an effective component in stress reduction interventions. D 2006 Elsevier Inc. All rights reserved.
Keywords: Anatomy training; Body esteem; Cardiovascular stress response; Terror management theory
Introduction For most people, seeing a dead body is an extremely distressing experience for cognitive, biological, and social reasons [1] and, as such, they will automatically avoid such exposure. However, in some occupations, exposure to dead bodies is a necessary facet of work. For example, emergency response personnel frequently encounter images of death in the course of their duties. A number of researchers have examined the impact of such exposure [2] and have considered variables that moderate the intensity of its stressfulness, such as age and perceived gruesomeness [3], and emotional attachment or identification [3,4]. In occupational settings, handling and viewing dead bodies have been implicated as a contributor to stress and anxiety disorders [2,4,5] and somatoform disorders [6]. 4 Corresponding author. Tel: +353 91 524411; fax: +353 91 521355. E-mail address: [email protected] (B.M. Hughes ). 0022-3999/06/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jpsychores.2006.05.004
Another group that customarily faces dead bodies comprises college students who are enrolled in programs where gross anatomy forms part of the curriculum (e.g., medical training). During their education, anatomy students participate in practical work involving the dissection of cadavers and cadaveric tissues. Whether exposure to dead bodies has deleterious effects on anatomy students has been the subject of a modest research literature. Questionnaire studies by faculty members have tended to report few lasting negative psychological effects of dissection (although initial trepidation among some students is acknowledged) [7–10]. It has been proposed, however, that due to the discomfort they themselves experience upon students’ reactions, anatomy faculty members are unlikely to fully acknowledge the stressfulness of dissections [11–13]. Research conducted by investigators other than faculty members has revealed a number of adverse consequences [1,11,14,15], as well as the finding that females exhibit more distress than males [16,17].
502
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
Given the negative effects of viewing dead bodies, as seen in other occupations, it stands to reason that anatomy dissection classes carry some risks. Stress may have a number of negative consequences for anatomy students. Firstly, stress may lead some students to terminate their studies [18]. Secondly, stress may elevate the risk of negative mental health in those who persist with training [18]. Thirdly, coping strategies such as depersonalization may damage the professional performance of students in their subsequent careers [16]. Fourthly, frequent and sustained stress responses are associated with an increased risk of ill physical health [19]. However, whether there are particular factors that predispose some students to adverse consequences has not been explored. One theoretical framework within which to examine the stress of exposure to death is terror management theory (TMT) [20 – 22]. TMT argues that people engage in various strategies (such as distraction or avoidance) in order to obviate the paralyzing terror that would otherwise result from contemplating the inevitability of death while possessing a self-preservation instinct. Proponents of TMT argue that maintaining a positive self-image goes some way to protect people from such existential anxieties. (Such argument is consistent with views that self-esteem is functionally adaptive and is not merely a bneedQ [23], that it protects against anxiety [21], and that it is largely subjective rather than based on objective cues [24]). A corollary is that reminders of death serve to make people’s self-images more salient. The conclusion, therefore, is that reminders of death are more threatening to individuals who rely heavily on selfimage as a stress buffer. One consequence of TMT is that anatomy students with highly positive self-images would be more likely to find dissections stressful. This would especially be the case in students with high esteem in terms of body image issues, as dissections present reminders of mortality that are specifically visual and body-related. Accordingly, high levels of body esteem (BE; the extent to which people value their bodies) might indicate a particular risk for stress in the anatomy laboratory. The present study sought to test this inference. Unlike previous research, the present study examined anatomy students’ stress objectively, assessing cardiovascular arousal patterns. Such patterns are widely employed as objective proxy measures of perceived stress because they are not subject to the problems of socially desirable responding that affect self-report stress indices.
Method Participants The participants were 36 anatomy students who were enrolled on the second year of a degree program in either medicine or science at a European university. The sample consisted of 29 women and 7 men, with a mean age of
19.67 years (S.D.=1.65 years). Participants had a mean height of 1.67 m (S.D.=.10 m) and a mean weight of 60.74 kg (S.D.=10.69 kg), yielding a mean body mass index of 21.69 kg/m2 (S.D.=2.80 kg/m2). Participants were enrolled on a volunteer basis and were not tangibly rewarded for participation. Psychometric instruments Four psychometric measures were used. The Body Esteem Scale (BES) [25] was used to assess BE. The BES is a 35-item instrument that measures positive and negative feelings toward different aspects of a respondent’s own body. Previous research has demonstrated the scale’s psychometric soundness [20]. The Rosenberg Self-Esteem Scale (RSES) [26] was used to assess self-esteem. The scale consists of 10 self-descriptive statements, for which agreement ratings are required. The scale has been identified as having sound psychometric qualities [27,28]. Trait anxiety was assessed using the trait form of the State–Trait Anxiety Inventory (STAI) [29], an inventory of 20 anxiety-related items requiring agreement ratings. It is widely utilized and is of repeatedly confirmed psychometric soundness. Finally, socially desirable responding was assessed using the Marlowe–Crowne Scale (MCS) [30]. The MCS is a 33-item scale requiring true–false judgments of self-descriptive statements, which is designed to establish whether the respondent has a tendency toward socially desirable responding in questionnaires. Procedure Participants were invited to take part following a lecture address in an anatomy class. Groups of six students were enrolled from each of six separate randomly generated groups, all of whom presented for anatomy practicals on the same week. Participants were invited to present at their subsequent dissection practicals 30 min prior to the scheduled start time. On each occasion, participants were met at the laboratory and brought to an adjoining room where they completed psychometric instruments, after which they had an opportunity to relax. The dissection practical chosen for the study took place halfway through a 12-week teaching semester and required students to witness the dissection of a human head by a faculty member. All students had previously been involved in hands-on dissection exercises on other parts of the body; this was the first opportunity for them to see the head and the face of the cadaver that they had been working with over the preceding weeks. Participants stood in close proximity to the cadaver throughout the dissection; during the practical, they were instructed to identify anatomical structures and were occasionally required to touch the cadaver on the surface of the face and inside the mouth. At the commencement of the practical (i.e., at 0 min), each participants’ blood pressure and heart rate (HR) were
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
503
Table 1 Psychometric characteristics of the sample (n=36) BEa Self-esteemb Trait anxietyc Social desirabilityd a b c d
Mean
S.D.
44.19 11.09 23.06 15.89
16.90 3.98 8.18 3.80
BES scores (possible range, 0 –140). RSES scores (possible range, 0 –30). STAI scores (possible range, 0 – 60). MCS scores (possible range, 0 – 33).
recorded by a research assistant using an Omron 637IT blood pressure monitor, which contains an automated sphygmomanometer that was thoroughly evaluated using international standardization protocols [31,32]. Subsequent readings of blood pressure and HR were taken after 30, 60, and 90 min of the practical.
Results Descriptive statistics and analytic strategy Psychometric features of the sample are presented in Table 1. Self-esteem was found to be correlated (Pearson) positively with BE (r 36=.61, Pb.001) and negatively with trait anxiety (r 36= .54, P=.001). Scores on the MCS did not correlate significantly with other scores, suggesting that BE, self-esteem, and trait anxiety were free from social desirability bias. However, whereas reliability analyses suggested satisfactory internal consistency for the RSES (a=.79), the BES (a=.91), and the STAI (a=.85), the MCS was found to have poor internal consistency (a=.46). As such, MCS scores were not included in subsequent analyses. Measures of systolic blood pressure (SBP), diastolic blood pressure (DBP), HR, and pulse pressure (PP) for each time point in the study are shown in Table 2. Participants were divided into low-BE, medium-BE, and high-BE groups (n=12 each) on the basis of a tertile split. A one-way analysis of variance (ANOVA) confirmed the validity of the split [ F(2,35)=78.59, Pb.001], with mean BE scores for each group being significantly different from the other two ( Pb.001 for each multiple comparison; Bonferroni comparisons used for all ANOVA post hoc tests). Further one-way ANOVAs revealed no significant between-group differences in age ( P=.08), body mass index
Fig. 1. SBP across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
( P=.42), or trait anxiety ( P=.40), implying that there was no need to statistically control for these variables in subsequent analyses involving BE. However, there were significant differences in self-esteem [ F(2,35)=7.90, P=.002], with the low-BE group having a lower self-esteem (M=1.82, S.D.=.37) than the high-BE group (M=2.36, S.D.=.33, P=.001). One-way ANOVAs revealed no between-group differences in initial SBP ( P=.13), DBP ( P=.65), HR ( P=.31), or PP ( P=.11). The seven male participants were distributed quite evenly across the groups (one in the lowBE group and three in each of the others). In order to assess the relationship between BE and cardiovascular functioning throughout the dissection, a series of two-way mixed ANOVAs was conducted. In each ANOVA, the repeated-measures variable, Time, had four levels (0, 30, 60, and 90 min), and the between-subjects variable, BE, had three levels (low, medium, and high). A separate ANOVA was conducted for each cardiovascular dependent variable. Given that there were between-subject differences in selfesteem, all analyses were followed up by equivalent analyses that control statistically for self-esteem. Throughout the study, the Mauchley test showed that the assumption of sphericity was not violated for repeated measures. SBP and DBP ANOVA for SBP revealed a significant interaction between BE and Time [ F(6,99)=2.58, P=.023], suggesting
Table 2 Mean cardiovascular measures (with standard deviations) taken at each time point 0 min SBP (mmHg) DBP (mmHg) HR (beats/min) PP (mmHg)
118.06 77.17 89.14 40.89
30 min (12.07) (8.60) (17.05) (8.09)
115.17 76.58 89.53 38.58
60 min (9.91) (8.53) (15.63) (6.16)
114.86 76.44 85.50 38.42
90 min (11.29) (9.45) (16.34) (6.97)
114.39 76.50 84.78 37.89
(10.92) (8.79) (15.38) (6.85)
504
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
Covariance analyses
Fig. 2. HR across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
that there were between-group differences in SBP changes during the practical (see Fig. 1). To explore this interaction, post hoc one-way repeated-measures ANOVAs were conducted for each group. For the high-BE group, a main effect for Time was revealed [ F(3,33)=4.17, P=.013], with SBP decreasing over the first three measures. SBP at 0 min was higher than that at 30 min ( P=.018) and that at 60 min ( P=.028). SBP at 90 min was not significantly different from that at 60 min. There were no such differences for the low-BE or the medium-BE group. Reconducting analyses while controlling for self-esteem revealed no new effects. Corresponding analyses revealed no significant relationships for DBP. Heart rate For HR, ANOVA revealed a main effect for Time [ F(3,99)=3.89, P=.011], with post hoc tests showing a decrease over time (see Fig. 2). HRs at 0 and 30 min were both significantly higher than that at 90 min ( P=.029 and .014, respectively). HR at 0 min was significantly higher than that at 60 min ( P=.029). There was neither a significant main effect for BE [although the effect approached significance; F(2,33)=2.94, P=.07] nor a significant TimeBE interaction. No significant main or interaction effect involving self-esteem was found when analyses were reconducted using a similar analysis of covariance (ANCOVA), as with SBP.
Given the fact that tertile splits transform continuous data into categories and thus result in loss of statistical power, intratrial cardiovascular functioning was further explored using covariance analyses. For each dependent variable, a one-way repeated-measures ANCOVA was conducted. Each analysis corresponded to the ANOVA conducted earlier, except that, instead of entering BE as a fixed-factor independent variable, raw BE scores were entered as a covariate in each case. However, given that covariance analysis requires a number of conditions that were not clearly met in the present data, these results should be interpreted conservatively. The ANCOVA for SBP confirmed the presence of a significant TimeBE interaction [ F(3,102)=4.13, P=.008]. Unlike ANOVA, ANCOVA revealed a significant main effect for Time [ F(3,102)=3.00, P=.034]. SBP at 0 min was significantly higher than that at 30 min ( P=.041) or that at 60 min ( P=.037) (see Fig. 1). The ANCOVA for DBP confirmed the absence of effects for that variable. The ANCOVA for HR confirmed the presence of a significant main effect for Time [ F(3,102)=3.01, P=.034]. However, unlike in ANOVA where the main effect for BE approached significance, in ANCOVA, the effect reached significance [ F(1,34)=4.64, P=.039]. This reflected a significant inverse correlation between BE and overall HR (r 36= .35, P=.039). Finally, the ANCOVA for PP failed to detect the main effect for BE observed in ANOVA. However, unlike ANOVA, ANCOVA revealed a significant main effect for Time [ F(3,102)=3.54, P=.017; see Fig. 3] and a significant TimeBE interaction [ F(3,102)=4.83, P=.003]. Post hoc tests showed PP at 0 min to be higher than that at 30 min ( P=.048). The TimeBE interaction reflected strong positive correlations between BE and PP at 0 min (r 36=.52, P=.001) and 30 min (r 36=.34, P=.042), but correlations
Pulse pressure For PP, ANOVA revealed a significant main effect for BE [ F(2,33)=3.85, P=.032]. Low-BE participants had lower overall PP than high-BE participants ( P=.038; see Fig. 3). Reconducting these analyses with self-esteem entered as a covariate revealed no additional effects involving self-esteem.
Fig. 3. PP across the four time points of the dissection class for each BE group (n=12 in each case) and overall (n=36).
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
505
Fig. 4. Associations between PP at each time point and BE across the sample (n=36).
close to zero at 60 min (r 36= .01, P=.95) and 90 min (r 36= .02, P=.91) (see Fig. 4).
Discussion In summary, the data indicate a pattern of cardiovascular functioning throughout the dissection that implicates BE as a moderator of perceived stress. There was an overall tendency for cardiovascular activity to decrease over time (for SBP, HR, and PP); however, students identified as having high levels of BE exhibited greater decreases (particularly in SBP and PP). High-BE participants had lower HR throughout the entire practical. Finally, even though there were differences in self-esteem across the three BE groups, self-esteem was not found to have any bearing on observed associations. The overall decrease in cardiovascular functioning across time is consistent with the phenomenon of habituation and, as such, corroborates past self-report research [9]. Students began the dissection in a state of (anticipatory) arousal, which dissipated as they became familiar with the situation. However, the distinctly greater decreases in SBP and PP among the high-BE group warrant further explanation. These decreases were probably due to two factors. Firstly, high-BE participants may have demonstrated enhanced levels of initial arousal because of greater anticipatory anxiety. Secondly, the sharper reduction in blood pressure among these participants may suggest that they were at greater risk for vasovagal syncope. Both explanations would suggest that high-BE students adjusted relatively poorly to the stressful situation and were at relatively greater risk for adverse stress responses. This is further suggested by the inverse correlation between BE and overall HR. This implied that students with high levels of BE operated with reduced arousal, which again is a risk factor for syncope. Effects regarding PP are explicable in terms of those for SBP, given the absence of effects for DBP. BE was not associated with trait anxiety, and the role of BE in cardiovascular functioning was independent of selfesteem. This suggests that BE plays a unique role in the relationship between stress and arousal in the particular
context under scrutiny. In the present study, high levels of esteem (viz., BE) were found to be maladaptive in dealing with stress. Although high esteem is considered valuable in many contexts, there are disadvantages in situations where esteem is strongly threatened. The fact that high BE was associated with indicators of anxiety and fainting risk is consistent with the prediction that mortality salience will affect those for whom the body is a source of esteem. In terms of TMT, students whose bodies are a source of stressbuffering esteem suffered exaggerated stress in response to a stimulus that directly reminded them of the vulnerability of that source. One practical implication of these findings is that psychometric assessments of BE might be useful as screening tools to identify students who are likely to find dissections stressful. Correspondingly, interventions to ameliorate the stressfulness of dissections may benefit from incorporating BE-related information. Prospective students’ attention could be drawn toward the arbitrariness of valuing physical appearance over other personal attributes, in an attempt to reduce the negative impact of seeing a physical body being dissected. Such conclusions may have similar implications for other occupational contexts where dead bodies are viewed. Among the limitations of this type of research is that the intensive nature of data collection limits sample size. However, in the present study, the four-level repeatedmeasures design compensated for this with statistical power. Future research could seek to employ continuous, rather than periodic, cardiovascular measurement; further data from male students would enhance the generalizability of the findings. Future studies could explore the risk of syncope more directly, either by measuring HR variability or by examining behavioral indices such as dizziness or blushing. Further research could also examine whether the observed effects were contingent on the particular setting by, for example, recording responses to other dissection types and/ or didactic exercises. Finally, a separate assessment of resting cardiovascular functioning outside the laboratory would shed further light on the mechanisms underlying arousal and syncope. Given the human anxiety associated with compulsory exposure to cadavers during medical and scientific training, and the potential for developing effective stress-
506
B.M. Hughes, A. Black / Journal of Psychosomatic Research 61 (2006) 501 – 506
reducing interventions, further investigation on the role of BE in this domain is warranted.
References [1] Sanner MA. Encountering the dead body: experiences of medical students in their anatomy and pathology training. Omega J Death Dying 1997;35:173 – 91. [2] Newhill CE, Sites EW. Identifying human remains following an air disaster: the role of social work. Soc Work Health Care 2000; 31:85 – 105. [3] McCarroll JE, Ursano RJ, Fullerton CS, Oates GL, Ventis WL, Friedman H, Shean GL, Wright KM. Gruesomeness, emotional attachment, and personal threat: dimensions of the anticipated stress of body recovery. J Trauma Stress 1995;8:343 – 9. [4] Ursano RJ, McCarroll JE. The nature of a traumatic stressor: handling dead bodies. J Nerv Ment Dis 1990;178:396 – 8. [5] Carson MA, Paulus LA, Lasko NB, Metzger LJ, Wolfe J, Orr SP, Pitman RK. Psychophysiologic assessment of posttraumatic stress disorder in Vietnam nurse veterans who witnessed injury or death. J Consult Clin Psychol 2000;68:890 – 7. [6] Labatte LA, Cardena E, Dimitreva J, Roy M, Engel CC. Psychiatric syndromes in Persian Gulf War veterans: an association of handling dead bodies with somatoform disorders. Psychother Psychosom 1998; 67:275 – 9. [7] Dinsmore CE, Daugherty S, Zeitz HJ. Student responses to the gross anatomy laboratory in a medical curriculum. Clin Anat 2001; 14:231 – 6. [8] McGarvey MA, Farrell T, Conroy RM, Kandiah S, Monkhouse WS. Dissection: a positive experience. Clin Anat 2001;14:227 – 30. [9] Arra´ez-Aybar L-A, Casado-Morales MI, Castan˜o-Collado G. Anxiety and dissection of the human cadaver: an unsolvable relationship. Anat Rec B New Anat 2004;279B:16 – 23. [10] Houwink AP, Kurup AN, Kollars JP, Kral Kollars CA, Carmichael SW, Pawlina W. Help of third-year medical students decreases first-year medical students’ negative psychological reactions on the first day of gross anatomy dissection. Clin Anat 2004;17:328 – 33. [11] William JL. Don’t discuss it: reconciling illness, dying, and death in a medical school anatomy laboratory. Fam Syst Med 1992;10:65 – 78. [12] Arra´ez-Aybar L-A, Castan˜o-Collado G, Casado-Morales MI. Dissection from the Spanish anatomist’s perspective: aims, attitudes, and related aspects. Anat Rec B New Anat 2004;281B:15 – 20. [13] Tschernig T, Schlaud M, Pabst R. Emotional reactions of medical students to dissecting human bodies: a conceptual approach and its evaluation. Anat Rec B 2000;261:11 – 3. [14] Horne DJ, Tiller JW, Eizenberg N, Tashevska M, Biddle N. Reactions of first year medical students to their initial encounter with a cadaver in the dissecting room. Acad Med 1990;65:645 – 6. [15] Tuohimaa P, Tamminen T, Fabrin V. Is it appropriate to speak of death during a dissection course? Teach Learn Med 1993;5:169 – 73.
[16] Charlton R, Dovey SM, Jones DG, Blunt A. Effects of cadaver dissection on the attitudes of medical students. Med Educ 1994; 28:290 – 5. [17] Eid J, Thayer JF, Johnsen BH. Measuring post-traumatic stress: a psychometric evaluation of symptom- and coping questionnaires based on a Norwegian sample. Scand J Psychol 1999;40:101 – 8. [18] Finkelstein P, Mathers L. Post-traumatic stress among medical students in the anatomy dissection laboratory. Clin Anat 1990;3: 219 – 26. [19] Lovallo WR, Gerin W. Psychophysiological reactivity: mechanisms and pathways to cardiovascular disease. Psychosom Med 2003;65: 36 – 45. [20] Goldenberg JL, McCoy SK, Pyszczynski T, Greenberg J, Solomon S. The body as a source of self-esteem: the effect of mortality salience on identification with one’s body, interest in sex, and appearance monitoring. J Pers Soc Psychol 2000;79:118 – 30. [21] Greenberg J, Pyszczynski T, Solomon S. The causes and consequences of the need for self-esteem: a terror management theory. In: Baumeister RF, editor. Public self and private self. New York7 Springer-Verlag, 1986. pp. 189 – 212. [22] Solomon S, Greenberg J, Pyszczynski T. A terror management theory of social behavior: the psychological functions of self-esteem and cultural worldviews. In: Zanna MP, editor. Advances in experimental social psychology. San Diego7 Academic Press, 1991. pp. 91 – 159. [23] Stevens A, Price J. Evolutionary psychiatry: a new beginning. London7 Routledge, 1996. [24] Pelham BW. Self-investment and self-esteem: evidence for a Jamesian model of self-worth. J Pers Soc Psychol 1995;69:1141 – 50. [25] Franzoi SL, Shields SA. The Body-Esteem Scale: multidimensional structure and sex differences in a college population. J Pers Assess 1984;48:173 – 8. [26] Rosenberg M. Society and the adolescent self-image. Princeton (NJ)7 Princeton University Press, 1965. [27] Gray-Little B, Williams VSL, Hancock TD. An item response theory analysis of the Rosenberg Self-Esteem Scale. Pers Soc Psychol Bull 1997;23:443 – 51. [28] Banos RM, Guillen V. Psychometric characteristics in normal and social phobic samples for a Spanish version of the Rosenberg SelfEsteem Scale. Psychol Rep 2000;87:269 – 74. [29] Spielberger CD, Gorusch RL, Lushene R. Manual for the State– Trait Anxiety Inventory. Palo Alto (Calif)7 Consulting Psychologists Press, 1970. [30] Crowne DP, Marlowe D. A new scale of social desirability independent of psychopathology. J Consult Psychol 1960;24:349 – 54. [31] Eckert S, Gleichmann U, Zagorski O, Klapp A. Validation of the Omron R3 blood pressure self-measuring device through simultaneous comparative invasive measurements according to protocol 58130 of the German Institute for Validation. Blood Press Monit 1997;2:189 – 92. [32] Watson S, Wenzel RR, di Matteo C, Meier B, Luscher TF. Accuracy of a new wrist cuff oscillometric blood pressure device: comparisons with intraarterial and mercury manometer measurements. Am J Hypertens 1998;11:1469 – 74.