- Email: [email protected]
Attempted versus successful avoidance: associations with distress, symptoms, and strategies for mental control
Personality and Individual Differences 33 (2002) 897–907 www.elsevier.com/locate/paid
Attempted versus successful avoidance: associations with distress, symptoms, and strategies for mental control Leanne Andrewsa,*, Nicholas Troopb, Stephen Josephc, Syd Hiskeya, Iain Coyned a
Department of Psychology, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK b Department of Psychology, London Guildhall University, London E1 7NT, UK c Department of Psychology, University of Warwick, Coventry CV4 7AL, UK d Department of Psychology, University of Hull, East Yorkshire HU6 7RX, UK
Received 13 July 2001; received in revised form 12 October 2001; accepted 12 November 2001
Abstract The most widely used measures of avoidance strategies following stressful or traumatic experiences are unidimensional. However, recent work has emphasised the multifactorial nature of avoidance. One intriguing and as yet unexplored distinction is that between attempts at avoidance and successful avoidance. Two studies are reported with the aim of investigating the differential relationships between attempted and successful avoidance and measures of distress and thought control strategies. In the first study 207 participants completed measures of attempted and successful avoidance along with a measure of distress. The results indicated that distress was associated with attempted avoidance but not successful avoidance. In the second study, 143 participants completed measures of attempted and successful avoidance along with measures of thought control strategies and distress. The results of study two replicated those of study one and also found that attempted avoidance was associated with ‘punishment’ and ‘worry’ thought control strategies, whereas, successful avoidance was associated with ‘social control’ strategies. It was concluded that attempts at avoidance that are not successful are maladaptive following stressful or traumatic experiences but successful avoidance is not. These data highlight the importance of identifying and making explicit the distinction between attempted and successful avoidance in future operational definitions. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Avoidance; Control strategies; Trauma; Distress; PTSD
1. Introduction The evidence regarding the question of whether avoidance following traumatic or other stressful life events is adaptive or maladaptive is mixed. For example, Perry, Difede, Musngi, Frances, and Jacobsberg (1992) found that severity of avoidant thoughts at 6 months post trauma * Corresponding author. Fax: +44-1206-873590. E-mail address: [email protected] (L. Andrews). 0191-8869/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0191-8869(01)00200-8
898
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
predicted post-traumatic stress disorder (PTSD) at 12 months post trauma. In contrast, Joseph, Yule, and Williams (1994) found that avoidance at 7 months post trauma did not predict distress at 19 months post trauma. These and other similar studies (e.g. Dalgleish, Joseph, Thrasher, Tranah & Yule, 1996; McFarlane, 1988; Rothbaum, Foa, Riggs, Murdock, & Walsh, 1992) have all used the Impact of Event Scale (IES; Horowitz, Wilner, & Alvarez, 1979) to assess avoidance. The Impact of Event Scale (IES) is a self-report measure which yields a single unidimensional score for avoidance. However, in a recent review of the psychometric properties of the IES, Joseph (2000) emphasised the need for investigation into the multidimensionality of avoidance in order to explain these contradictory findings. Evidence for the multidimensionality of avoidance has been found previously. For example, Williams, Joseph, and Yule (1994), in their study with survivors of the Herald of Free Enterprise, identified three different types of avoidance strategies. These were cognitive (e.g. trying not to think about the event), behavioural (e.g. refusing to talk about the event) and emotional avoidance (e.g. emotional numbing). More recently, Dalgleish, Mathews, and Wood (1999) have put forward a four factor taxonomy of avoidance encompassing automatic and controlled avoidance, and avoidance of the event and of emotions relating to the event. However, as the most widely used measures of avoidance, such as the IES, yield unidimensional avoidance scores, relationships between specific types of avoidance and psychological distress are confounded. Whilst attention has been directed towards measuring the different facets of avoidance there is one aspect of avoidance that has been neglected. This is the notion of whether attempting to avoid trauma related stimuli has the same relationship with subsequent psychological distress as successfully avoiding trauma related stimuli. One indication that individuals who attempt to avoid trauma related stimuli do not actually achieve successful avoidance is the association between intrusion and avoidance that is often reported in studies of stressful or traumatic life events. Generally correlations between intrusion and avoidance have been found to range from moderate to strong (e.g. Creamer, Burgess, & Pattison, 1992; Hodgkinson & Joseph, 1995; Horowitz et al., 1979). The presence of both avoidance and intrusion would indicate that attempts to avoid are not actually successful. Again, unidimensional measures of avoidance are not able to investigate attempts to avoid and successful avoidance in order to explore their subsequent relationships with psychopathology. The two studies described below have used items from the widely used IES avoidance subscale in order to (1) investigate whether it is possible to measure attempted and successful avoidance separately, and (2) to investigate the relationships between attempted and successful avoidance and psychological distress.
2. Study one 2.1. Method 2.1.1. Participants Respondents for this study participated in on-line data collection (at http://privatewww.essex. ac.uk/shiske/form.html). Links were provided to this web site from various stress and trauma support sites available on the world wide web. The purpose of the web site was to investigate
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
899
international responses to traumatic and extremely stressful life events. This sample consisted of 207 respondents. There were 147 female and 60 male participants. The age range was 15–62 years, with a mean age of 38 years (S.D.=11 years). One hundred and ninety-five respondents were white, and nine belonged to other ethnic groups. Three respondents failed to complete this question. The level of education for the sample was: up to high/secondary school=34; vocational, technical school or college=32; university=75; postgraduate or professional degree=29; other=24 (13 missing). 2.1.2. Materials 2.1.2.1. IES (Horowitz et al., 1979). The IES is one of the most popular and widely used measures to assess subjective distress following stressful or traumatic events. The 15-item measure yields two scores representing the frequency of intrusive experiences and avoidance strategies during the preceding 7 days. Each item is scored on a four-point scale (0, 1, 3 and 5), anchored by ‘‘Not at all’’ and ‘‘Often’’. Inspection of the items on the IES avoidance subscale indicates that some items are asking how frequently the individual ‘avoided’ something while other items ask how frequently the individual ‘tried’ to avoid something. For example, item 3, asks how frequently the individual ‘tried’ to avoid doing something [I tried to remove it from memory]. Whereas item 2 asks how frequently the individual ‘actually’ avoided doing something [I avoided letting myself get upset when I thought about it or was reminded of it]. Summing all the avoidance items regardless of whether the individual has actually succeeded in avoiding or simply attempted to avoid assesses the level of avoidance a person has used. Therefore, an individual who ‘tries not to talk about it’ but can’t stop himself/herself being drawn into conversations relating to the event is assumed to be equivalent to the individual who ‘tries not to talk about it’ and succeeds in doing this. In order to measure attempts to avoid and successful avoidance additional items were created based on a rewording of the original items. Thus the three ‘attempted’ avoidance items had equivalent ‘successful’ avoidance counterparts (and vice versa). These additional modified items are given in Table 1. As has been noted previously by Joseph and colleagues (Hodgkinson & Joseph, 1995; Joseph, 2000; Yule, Ten Bruggencate, & Joseph, 1994) item 8 [I felt as if it hadn’t happened or wasn’t real] and item 15 [My feelings about it were kind of numb] from the original IES were not considered Table 1 Attempted and successful items from the original IES avoidance subscale with reworded counterparts Original item
New item
2. I avoided letting myself get upset when I thought about it or was reminded of it. 3. I tried to remove it from memory. 7. I stayed away from reminders of it. 9. I tried not to talk about it. 12. I was aware that I still had a lot of feelings about it, but I didn’t deal with them. 13. I tried not to think about it.
I tried to avoid letting myself get upset when I thought about it or was reminded of it. I removed it from memory. I tried to stay away from reminders of it. I did not talk about it. I was aware that I still had a lot of feelings about it, but I tried not to deal with them. I did not think about it.
Items in italics refer to attempted avoidance, all other items refer to successful avoidance.
900
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
to be asking about aspects of avoidance that could consciously be attempted. In addition, it was apparent that translating these items to represent both attempted and successful avoidance would have been problematic. Therefore, these items were left unchanged and did not contribute to the new subscales of attempted and successful avoidance (although they are still included in the scoring for the original IES avoidance subscale). Thus the modified IES was increased from a 15item scale to a 21-item scale. Scores can be computed for the subscales of intrusion and avoidance (as in the original) as well as attempted and successful avoidance. Alphas for the intrusion and avoidance subscales, as well as attempted and successful avoidance subscales are displayed in Table 2. 2.1.2.2. PTSD Symptom Scale (PSS: Foa, Riggs, Dancu, & Rothbaum, 1993). The PSS is a 17item self-report questionnaire, scored on a four-point scale (0, 1, 2 and 3). The questions are designed to assess PTSD as characterised by the three types of symptoms described in DSM-III-R (APA, 1987): intrusion (range of 0–12); avoidance (range of 0–21) and arousal (range of 0–18). The questions are asked in relation to how the individual has felt in the last 2 weeks in relation to a traumatic event. Alphas for intrusion, avoidance and arousal subscales are displayed in Table 2. 2.1.3. Procedure The participants in this study accessed a battery of questionnaires on-line via links from various stress and trauma web sites. The battery comprised the modified IES, the PSS along with other pilot measures that were under psychometric development. These pilot measures are not reported here. An information sheet was provided detailing the nature and requirements of the study. Table 2 Means, standard deviations and alphas for all subscales Subscale
Mean
Standard deviation
Alpha
Study one Attempted Avoid Successful Avoid IES Avoidance IES Intrusion PSS Avoidance PSS Intrusion PSS Arousal
14.42 11.88 18.39 19.85 9.59 5.33 9.09
9.12 6.48 10.30 10.42 6.22 3.56 5.66
0.86 0.63 0.82 0.90 0.87 0.84 0.88
Study two Attempted Avoid Success Avoid IES Avoidance IES Intrusion TCQ Punishment TCQ Reappraisal TCQ Social Control TCQ Worry TCQ Distraction
9.91 9.14 13.76 12.93 9.71 13.17 13.41 9.97 14.66
8.66 6.90 10.94 9.98 2.95 5.65 4.42 3.09 3.25
0.86 0.69 0.86 0.90 0.73 0.51 0.87 0.83 0.74
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
901
Participants were asked to describe their most stressful or traumatic experience and state in what year this event happened. They then filled out the IES followed by the PSS. Additional comments could be added before submitting the form. 2.2. Results Approximately 2% of the responses received were ‘spoiled’ responses and subsequently discarded. These were comprised largely of incomplete or multiple submissions (in the form of identical responses in a short space of time from participants with the same demographic details). Missing values were excluded listwise. Means and standard deviations of each subscale from both measures are included in Table 2. Bivariate correlations for both attempted and successful avoidance as well as the original IES avoidance subscales are shown in Table 3. Partial correlations were calculated for the two types of avoidance (attempted and successful) with all other subscales, these are also displayed in Table 3.
Table 3 Bivariate correlations for modified IES avoidance subscales with all other subscales and partial correlations for attempted and successful avoidance Subscale
IES Avoidance
Bivariate
Partial
Attempted Avoidance
Successful Avoidance
Attempted Avoidance
Successful Avoidance
Study one Attempted Avoidance Successful Avoidance IES Avoidance IES Intrusion PSS Avoidance PSS Intrusion PSS Arousal
0.93*** 0.80*** – 0.59*** 0.67*** 0.53*** 0.54***
– 0.74*** 0.93*** 0.52*** 0.60*** 0.49*** 0.49***
0.74*** – 0.80*** 0.34*** 0.47*** 0.31*** 0.35***
– – 0.82*** 0.39*** 0.38*** 0.37*** 0.35***
– – 0.44*** 0.07 0.11 0.04 0.00
Study two Attempted Avoidance Successful Avoidance IES Avoidance IES Intrusion TCQ Punishment TCQ Reappraisal TCQ Social Control TCQ Worry TCQ Distraction
0.94*** 0.83*** – 0.68*** 0.32*** 0.03 0.27** 0.16 0.26**
– 0.79*** 0.94*** 0.65*** 0.30*** 0.03 0.25** 0.13 0.20*
0.79*** – 0.83*** 0.47*** 0.24** 0.05 0.34*** 0.03 0.27**
– – 0.82*** 0.56*** 0.19* 0.02 0.02 0.18* 0.04
– – 0.37*** 0.17 0.02 0.03 0.26** 0.11 0.15
* P< 0.05 level (2-tailed). ** P< 0.01 level (2-tailed). *** P< 0.001 level (2-tailed).
902
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
Table 3 illustrates a large (and highly significant) bivariate correlation between attempted and successful avoidance. Significant correlations were also found for both attempted and successful avoidance and all other subscales. However, when controlling for attempted avoidance successful avoidance only correlated with IES avoidance, whereas, when controlling for successful avoidance attempted avoidance remained significantly correlated with all other subscales. 2.3. Discussion This study supports the idea that attempted and successful avoidance can be measured separately. The results indicate that attempted avoidance is more closely related to measures of distress than successful avoidance. This is clearly illustrated by the associations between the PSS and the attempted and successful avoidance subscales. All three subscales of the PSS correlated with attempted avoidance when the effects of successful avoidance were partialled out. This indicates that increases in attempted but not successful avoidance are related to increases in the types of distress measured by the subscales of the PSS (intrusions, avoidance and arousal). Successful avoidance, however, did not correlate with any of the PSS subscales when attempted avoidance had been partialled out. An interesting finding was the associations between the original IES avoidance subscale and the attempted and successful avoidance subscales. The results of the bivariate correlations indicated strong positive relationships between the three avoidance subscales. These associations may initially appear to indicate that all three avoidance subscales are measuring more or less the same concept. However, the partial correlations do not support this notion. When successful avoidance was controlled for attempted avoidance still correlated very highly with the original IES avoidance subscale and moderately highly with IES intrusion whereas, when controlling for attempted avoidance, successful avoidance correlated with IES avoidance to a lesser extent and did not correlate with intrusion at all. These results suggest two things. Firstly, the original IES avoidance subscale seems more closely to reflect attempts at avoidance than actual successful avoidance and secondly, that attempted avoidance is related to intrusions but successful avoidance is not. As this was a correlational study we cannot conclude as to whether more intrusions lead to more attempts at avoidance or whether attempts to avoid trauma related stimuli leads to an increase in the frequency of intrusions.
3. Study two This study aimed to replicate the findings from study one whilst testing further hypotheses relating to the various strategies individuals use in order to control unpleasant and unwanted thoughts. Literature relating to the mental control of thoughts has illustrated the ironic effects of thought suppression whereby the mere act of trying to avoid thinking about something results in increasing the frequency of that thought. For example, Wegner and colleagues (e.g. Wegner, 1989; Wegner, Schneider, Carter, & White, 1987) found that attempts at thought suppression actually produced a preoccupation with the ‘suppressed’ thought. Wegner and colleagues propose that as the task of suppressing a thought is so difficult, it actually results in the thought
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
903
being held in consciousness while effort is made to eliminate it. In order to understand the processes that lead to the ‘suppressed’ thought being reactivated with increased frequency (rebound effect) they introduced a distractor task and found that although this didn’t facilitate the suppression task it did serve the purpose of reducing the rebound effect. Further evidence of the use of a distracting task as a means by which the process of suppression can be facilitated has since been reported (e.g. Clark, Ball, & Pape, 1991; Clark, Winton, & Thyne, 1993; Salkovskis & Campbell, 1994). These findings have implications for self-report measures of avoidance. It could be hypothesised that if a person is reporting the types of avoidance they are consciously aware of having used then the stimuli they were trying to avoid must have been held in consciousness for a period of time in order for an attempt at avoidance to be made. Therefore, the use of a distracting task as a means by which to avoid the relevant stimuli may prove to be a useful strategy with which to achieve successful avoidance. Wells and Davies (1994) using their Thought Control Questionnaire (TCQ), found that distraction in the form of doing something enjoyable or keeping busy with work, etc., had no relationship to measures of stress vulnerability or psychopathology. They also failed to find associations between social control and measures of stress vulnerability. However, Reynolds and Wells (1999) using two patient samples (one PTSD and one depressed) reported negative correlations between distraction and anxiety and depression, indicating that the more an individual used distraction as a strategy for controlling unwanted thoughts the less anxious or depressed they were. They also found a negative correlation between social control and avoidance, indicating that the more they talked about the unwanted thought the less avoidance they used. This led them to support Wells and Davies’ (1994) conclusions that these two subscales may buffer against stress vulnerability, which suggests that both distraction and social control are adaptive thought control strategies. However, using the same measure they found that punishment and worry were associated with anxiety in the depressed patients, and punishment was also associated with depression. From this it can be seen that different strategies of mental control have differing effects on distress levels. The question of whether attempted avoidance and successful avoidance relate to thought control strategies and measures of psychological distress differentially will be investigated. It is predicted that attempted avoidance will be associated with measures of distress and to dysfunctional thought control strategies, such as ‘worry’ and ‘punishment’, whereas successful avoidance will be related to more beneficial thought control strategies such as ‘distraction’ and ‘social control’. 3.1. Method 3.1.1. Participants This study consisted of 143 respondents. Eighty-nine were undergraduate students at the University of Hull, 33 were undergraduates at the University of Essex, and 21 were non-students from outside the university. There were 90 females and 51 males (N=141, two missing data). The age range was 18–63 years, with a mean age of 22.25 years (S.D.=7 years). The level of education for the sample was: up to secondary school=6; further education=125; and higher education=11 (one missing datum).
904
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
3.1.2. Measures The IES was modified as described in Section 2.1.2.1. 3.1.2.1. TCQ (Wells & Davies, 1994). The TCQ is a 30-item self report questionnaire, scored on a four-point scale (1, 2, 3, and 4). The questions were designed to assess strategies for controlling unpleasant and unwanted thoughts. There are five subscales relating to different strategies ‘Distraction’, ‘Punishment’ ‘Reappraisal’, ‘Worry’ and ‘Social Control’. Each subscale has a score range of 6–24. The questions all relate to the techniques an individual ‘generally’ uses to control unpleasant thoughts. Alphas for all TCQ subscales are displayed in Table 2. 3.1.3. Procedure The participants in this study were given a battery of questionnaires comprising the modified IES, the TCQ and other measures that were included as part of a different study. The pack also included a covering letter explaining the nature and requirements of the study. Participants were asked to think of their most stressful or traumatic experience and describe this in the space provided, also indicating what year this took place. They were then required to respond to each questionnaire with this incident in mind. Students returned their questionnaires to one of the authors while non-students were provided with pre paid envelopes in order to return their completed questionnaires individually. 3.2. Results Missing values were excluded listwise. Means and standard deviations for each subscale from all measures are included in Table 2. Bivariate and partial correlations were calculated for all measures, these are presented in Table 3. A significant bivariate correlation between attempted and successful avoidance was found (similar in magnitude to that in study one). Significant bivariate correlations were also found for both attempted and successful avoidance with the following subscales: IES avoidance, IES intrusion, TCQ punishment, TCQ social control and TCQ distraction. The IES avoidance subscale also correlated with all these subscales. However, when controlling for successful avoidance, attempted avoidance only correlated with IES avoidance, IES intrusion, TCQ punishment and TCQ worry subscales. When controlling for attempted avoidance, successful avoidance only correlated with IES avoidance and the TCQ social control subscale, with a trend towards significance with (positively) TCQ distraction (r=0.15, P<0.10) and (negatively) IES intrusion (r= 0.17, P=0.06). 3.3. Discussion The results of this study lend support to the findings of study one, as attempted avoidance was again more strongly associated with measures of distress than successful avoidance. Also there was support for the prediction that attempted avoidance would be associated with ‘maladaptive’ thought control strategies such as ‘punishment’ and ‘worry’, whereas successful avoidance would be associated with the more ‘adaptive’ thought control strategies such as ‘social control’ and ‘distraction’.
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
905
As in study one, the bivariate correlations indicate a significant association between attempted and successful avoidance but once again, when partialling out the effects of successful avoidance attempted avoidance correlated highly with the original IES avoidance and moderately with IES intrusion whereas, when partialling out the effects of attempted avoidance successful avoidance only correlated with IES avoidance. One interesting point is that a negative trend (although not quite significant at the 5% level) was evident between successful avoidance and IES intrusion indicating that as successful avoidance increases intrusive experiences decrease. As well as replicating the findings from the first study, one of the aims of this study was to investigate the relationships between the thought control strategies measured by the TCQ and attempted and successful avoidance. The reason for this was to examine Wells and Davies’ (1994) suggestion that some thought control strategies (e.g. distraction and social control) may be related to positive mental health whilst others (e.g. worry and punishment) may be related to psychopathology. The results of the partial correlations indicated that attempted avoidance was associated with punishment and worry. Attempted avoidance was also found to be correlated with intrusions, this is consistent with the findings from the thought suppression literature which suggests that attempting to suppress a thought results in increasing activation of the thought. Successful avoidance correlated significantly only with social control (negatively). This could either be interpreted in the same manner in which Reynolds and Wells (1999) did by suggesting that the more the individual talked about the thought the less avoidance they used or it could indicate that the more successful they were at avoiding the less they discussed the thought with friends. This is an important distinction, as disclosure has been found to have beneficial long-term health effects (Pennebaker, 1993) whereas, reducing levels of social support by relying increasingly on avoidance strategies, whilst initially producing a reduction in distress levels, appears to be maladaptive in the long term. Social support is often cited as an extremely important factor in the prevention of psychopathology (e.g. Brown, Andrews, Harris, Adler & Bridge, 1986) and also as part of the recovery process from psychological disorders (e.g. Stallard & Law, 1993). Therefore, it may be that while successful avoidance is unrelated to pathology concurrently there may be serious consequences if relied upon in the long term. The lack of a significant association between distraction and successful avoidance doesn’t support the findings from either the thought suppression literature or Wells and Davies (1994) study. However, the partial correlation between distraction and successful avoidance did at least indicate a trend towards an association whereas the correlation with attempted avoidance was at zero. It would be interesting to replicate in a clinical group to examine this relationship further in a group with higher distress levels. Overall these findings provide partial support for Wells and Davies’ (1994) notion of adaptive and maladaptive thought control strategies with worry and punishment being related to attempted avoidance which is also related to measures of distress.
4. General discussion The results of these studies, although only partially supporting the predictions, are interesting as they do indicate differential relationships between thought control strategies, measures of distress (IES and PSS) and attempted and successful avoidance processes.
906
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
The main findings from these studies indicate that attempted avoidance is related to measures of psychological distress (PTSD symptoms) and to mental control strategies of punishment and worry (Wells & Davies, 1994). Successful avoidance on the other hand doesn’t appear to relate to measures of psychological distress but does relate to those mental control strategies that are thought to promote positive well being (Wells & Davies, 1994). The inverse relationship with social control is interesting as it may be the case that successful avoidance, while being unrelated to pathology in the short term, may not actually promote well being in the long term. Therefore, a replication employing longitudinal methodology would be necessary in order to understand whether it is possible to avoid successfully for a long period of time and also to investigate the psychological effects of such long term avoidance. One of the limitations with employing a self report methodology to the study of avoidance is that there may be instances where an individual doesn’t feel the need to avoid (presumably because the trauma has been fully processed and the outcome is satisfactory adjustment). In this instance when asked whether they ‘did not talk about it’ an affirmative response would yield a higher score representing more avoidance use when in actual fact the individual might be indicating that they did not feel the need to talk about the event during the preceding 7 days. However, this limitation is one that is apparent in all self report measures of avoidance and isn’t limited to this study only. Employing an experimental methodology to the differentiation of attempted and successful avoidance may resolve this issue. Finally, it must be noted that this study is not putting forward a modified IES for general use. The intention was to use the items in the IES to illustrate a conceptual issue. The rationale for the study was to highlight the distinction between attempted and successful avoidance in order for this to be taken into account when new measures of avoidance are developed. Whilst this study has focussed explicitly on the role of attempted and successful avoidance following traumatic experiences we are also aware that the results of this study have implications for the wider coping literature. Therefore, it is important that researchers and professionals working within the field of health care are made aware of this distinction particularly when using measures that sum all avoidance items to yield a single score.
Acknowledgements We would like to thank to all the anonymous participants of the study. We would also like to express our gratitude to Elaine Fox and Ray Meddis.
References American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: American Psychiatric Association. Brown, G. W., Andrews, B., Harris, T., Adler, Z., & Bridge, L. (1986). Social support, self-esteem and depression. Psychological Medicine, 16, 813–831. Clark, D. M., Ball, S., & Pape, D. (1991). An experimental investigation of thought suppression. Behaviour Research and Therapy, 29, 253–257.
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
907
Clark, D. M., Winton, E., & Thyne, L. (1993). A further experimental investigation of thought suppression. Behaviour Research and Therapy, 31, 207–210. Creamer, M., Burgess, P., & Pattison, P. (1992). Reaction to trauma: a cognitive processing model. Journal of Abnormal Psychology, 101(3), 452–459. Dalgleish, T., Joseph, S., Thrasher, S., Tranah, T., & Yule, W. (1996). Crisis support following the herald of free enterprise disaster: a longitudinal perspective. Journal of Traumatic Stress, 9, 833–845. Dalgleish, T., Mathews, A., & Wood, J. (1999). Inhibition processes in cognition and emotion: a special case? In T. Dalgleish, & M. Power (Eds.), The handbook of cognition and emotion. Chichester, UK: Wiley. Foa, E. B., Riggs, D. S., Dancu, C. V., & Rothbaum, B. O. (1993). Reliability and validity of a brief instrument for assessing post-traumatic stress disorder. Journal of Traumatic Stress, 6(4), 459–473. Hodgkinson, P., & Joseph, S. (1995). Factor analysis of the Impact of Event Scale with female bank staff following an armed raid. Personality and Individual Differences, 19( 5), 773–775. Horowitz, M., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: a measure of subjective stress. Psychosomatic Medicine, 41( 3), 209–218. Joseph, S. (2000). Psychometric evaluation of Horowitz’s Impact of Event Scale: a review. Journal of Traumatic Stress, 13, 101–113. Joseph, S., Yule, W., & Williams, R. (1994). The Herald of Free Enterprise disaster: the relationship of intrusion and avoidance to subsequent depression and anxiety. Behaviour Research and Therapy, 32(1), 115–117. McFarlane, A. C. (1988). The aetiology of post-traumatic stress disorders following disaster. British Journal of Psychiatry, 152, 116–121. Pennebaker, J. (1993). Putting stress into words: health, linguistic, and therapeutic implications. Behaviour Research and Therapy, 31, 539–548. Perry, S., Difede, J., Musngi, G., Frances, A. J., & Jacobsberg, L. (1992). Predictors of posttraumatic stress disorder after burn injury. American Journal of Psychiatry, 149, 931–935. Reynolds, M., & Wells, A. (1999). The Thought Control Questionnaire—psychometric properties in a clinical sample, and relationships with PTSD and depression. Psychological Medicine, 29, 1089–1099. Rothbaum, B. O., Foa, E. B., Riggs, D. S., Murdock, T., & Walsh, W. (1992). A prospective examination of posttraumatic stress disorder in rape victims. Journal of Traumatic Stress, 5, 455–476. Salkovskis, P., & Campbell, P. (1994). Thought suppression induces intrusion in naturally occurring negative intrusive thoughts. Behaviour Research and Therapy, 32(1), 1–8. Stallard, P., & Law, F. (1993). Screening and psychological debriefing of adolescent survivors of life threatening events. British Journal of Psychiatry, 163, 660–665. Wegner, D. M. 1989. White bears and other unwanted thoughts: suppression, obsession, and the psychology of mental control. New York: Viking Press. Wegner, D. M., Schneider, D. J., Carter, S. R., & White, T. L. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53, 5–13. Wells, A., & Davies, M. I. (1994). The thought control questionnaire: a measure of individual differences in the control of unwanted thoughts. Behaviour Research and Therapy, 32(8), 871–878. Williams, R., Joseph, S., & Yule, W. (1994). The role of avoidance in coping with disasters: a study of survivors of the capsize of the ‘Herald of Free Enterprise’. Clinical Psychology and Psychotherapy, 1(2), 87–94. Yule, W., Ten Bruggencate, S., & Joseph, S. (1994). Principal components analysis of the Impact of Event Scale in adolescents who survived a shipping disaster. Personality and Individual Differences, 16, 685–691.
Attempted versus successful avoidance: associations with distress, symptoms, and strategies for mental control Leanne Andrewsa,*, Nicholas Troopb, Stephen Josephc, Syd Hiskeya, Iain Coyned a
Department of Psychology, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK b Department of Psychology, London Guildhall University, London E1 7NT, UK c Department of Psychology, University of Warwick, Coventry CV4 7AL, UK d Department of Psychology, University of Hull, East Yorkshire HU6 7RX, UK
Received 13 July 2001; received in revised form 12 October 2001; accepted 12 November 2001
Abstract The most widely used measures of avoidance strategies following stressful or traumatic experiences are unidimensional. However, recent work has emphasised the multifactorial nature of avoidance. One intriguing and as yet unexplored distinction is that between attempts at avoidance and successful avoidance. Two studies are reported with the aim of investigating the differential relationships between attempted and successful avoidance and measures of distress and thought control strategies. In the first study 207 participants completed measures of attempted and successful avoidance along with a measure of distress. The results indicated that distress was associated with attempted avoidance but not successful avoidance. In the second study, 143 participants completed measures of attempted and successful avoidance along with measures of thought control strategies and distress. The results of study two replicated those of study one and also found that attempted avoidance was associated with ‘punishment’ and ‘worry’ thought control strategies, whereas, successful avoidance was associated with ‘social control’ strategies. It was concluded that attempts at avoidance that are not successful are maladaptive following stressful or traumatic experiences but successful avoidance is not. These data highlight the importance of identifying and making explicit the distinction between attempted and successful avoidance in future operational definitions. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Avoidance; Control strategies; Trauma; Distress; PTSD
1. Introduction The evidence regarding the question of whether avoidance following traumatic or other stressful life events is adaptive or maladaptive is mixed. For example, Perry, Difede, Musngi, Frances, and Jacobsberg (1992) found that severity of avoidant thoughts at 6 months post trauma * Corresponding author. Fax: +44-1206-873590. E-mail address: [email protected] (L. Andrews). 0191-8869/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0191-8869(01)00200-8
898
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
predicted post-traumatic stress disorder (PTSD) at 12 months post trauma. In contrast, Joseph, Yule, and Williams (1994) found that avoidance at 7 months post trauma did not predict distress at 19 months post trauma. These and other similar studies (e.g. Dalgleish, Joseph, Thrasher, Tranah & Yule, 1996; McFarlane, 1988; Rothbaum, Foa, Riggs, Murdock, & Walsh, 1992) have all used the Impact of Event Scale (IES; Horowitz, Wilner, & Alvarez, 1979) to assess avoidance. The Impact of Event Scale (IES) is a self-report measure which yields a single unidimensional score for avoidance. However, in a recent review of the psychometric properties of the IES, Joseph (2000) emphasised the need for investigation into the multidimensionality of avoidance in order to explain these contradictory findings. Evidence for the multidimensionality of avoidance has been found previously. For example, Williams, Joseph, and Yule (1994), in their study with survivors of the Herald of Free Enterprise, identified three different types of avoidance strategies. These were cognitive (e.g. trying not to think about the event), behavioural (e.g. refusing to talk about the event) and emotional avoidance (e.g. emotional numbing). More recently, Dalgleish, Mathews, and Wood (1999) have put forward a four factor taxonomy of avoidance encompassing automatic and controlled avoidance, and avoidance of the event and of emotions relating to the event. However, as the most widely used measures of avoidance, such as the IES, yield unidimensional avoidance scores, relationships between specific types of avoidance and psychological distress are confounded. Whilst attention has been directed towards measuring the different facets of avoidance there is one aspect of avoidance that has been neglected. This is the notion of whether attempting to avoid trauma related stimuli has the same relationship with subsequent psychological distress as successfully avoiding trauma related stimuli. One indication that individuals who attempt to avoid trauma related stimuli do not actually achieve successful avoidance is the association between intrusion and avoidance that is often reported in studies of stressful or traumatic life events. Generally correlations between intrusion and avoidance have been found to range from moderate to strong (e.g. Creamer, Burgess, & Pattison, 1992; Hodgkinson & Joseph, 1995; Horowitz et al., 1979). The presence of both avoidance and intrusion would indicate that attempts to avoid are not actually successful. Again, unidimensional measures of avoidance are not able to investigate attempts to avoid and successful avoidance in order to explore their subsequent relationships with psychopathology. The two studies described below have used items from the widely used IES avoidance subscale in order to (1) investigate whether it is possible to measure attempted and successful avoidance separately, and (2) to investigate the relationships between attempted and successful avoidance and psychological distress.
2. Study one 2.1. Method 2.1.1. Participants Respondents for this study participated in on-line data collection (at http://privatewww.essex. ac.uk/shiske/form.html). Links were provided to this web site from various stress and trauma support sites available on the world wide web. The purpose of the web site was to investigate
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
899
international responses to traumatic and extremely stressful life events. This sample consisted of 207 respondents. There were 147 female and 60 male participants. The age range was 15–62 years, with a mean age of 38 years (S.D.=11 years). One hundred and ninety-five respondents were white, and nine belonged to other ethnic groups. Three respondents failed to complete this question. The level of education for the sample was: up to high/secondary school=34; vocational, technical school or college=32; university=75; postgraduate or professional degree=29; other=24 (13 missing). 2.1.2. Materials 2.1.2.1. IES (Horowitz et al., 1979). The IES is one of the most popular and widely used measures to assess subjective distress following stressful or traumatic events. The 15-item measure yields two scores representing the frequency of intrusive experiences and avoidance strategies during the preceding 7 days. Each item is scored on a four-point scale (0, 1, 3 and 5), anchored by ‘‘Not at all’’ and ‘‘Often’’. Inspection of the items on the IES avoidance subscale indicates that some items are asking how frequently the individual ‘avoided’ something while other items ask how frequently the individual ‘tried’ to avoid something. For example, item 3, asks how frequently the individual ‘tried’ to avoid doing something [I tried to remove it from memory]. Whereas item 2 asks how frequently the individual ‘actually’ avoided doing something [I avoided letting myself get upset when I thought about it or was reminded of it]. Summing all the avoidance items regardless of whether the individual has actually succeeded in avoiding or simply attempted to avoid assesses the level of avoidance a person has used. Therefore, an individual who ‘tries not to talk about it’ but can’t stop himself/herself being drawn into conversations relating to the event is assumed to be equivalent to the individual who ‘tries not to talk about it’ and succeeds in doing this. In order to measure attempts to avoid and successful avoidance additional items were created based on a rewording of the original items. Thus the three ‘attempted’ avoidance items had equivalent ‘successful’ avoidance counterparts (and vice versa). These additional modified items are given in Table 1. As has been noted previously by Joseph and colleagues (Hodgkinson & Joseph, 1995; Joseph, 2000; Yule, Ten Bruggencate, & Joseph, 1994) item 8 [I felt as if it hadn’t happened or wasn’t real] and item 15 [My feelings about it were kind of numb] from the original IES were not considered Table 1 Attempted and successful items from the original IES avoidance subscale with reworded counterparts Original item
New item
2. I avoided letting myself get upset when I thought about it or was reminded of it. 3. I tried to remove it from memory. 7. I stayed away from reminders of it. 9. I tried not to talk about it. 12. I was aware that I still had a lot of feelings about it, but I didn’t deal with them. 13. I tried not to think about it.
I tried to avoid letting myself get upset when I thought about it or was reminded of it. I removed it from memory. I tried to stay away from reminders of it. I did not talk about it. I was aware that I still had a lot of feelings about it, but I tried not to deal with them. I did not think about it.
Items in italics refer to attempted avoidance, all other items refer to successful avoidance.
900
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
to be asking about aspects of avoidance that could consciously be attempted. In addition, it was apparent that translating these items to represent both attempted and successful avoidance would have been problematic. Therefore, these items were left unchanged and did not contribute to the new subscales of attempted and successful avoidance (although they are still included in the scoring for the original IES avoidance subscale). Thus the modified IES was increased from a 15item scale to a 21-item scale. Scores can be computed for the subscales of intrusion and avoidance (as in the original) as well as attempted and successful avoidance. Alphas for the intrusion and avoidance subscales, as well as attempted and successful avoidance subscales are displayed in Table 2. 2.1.2.2. PTSD Symptom Scale (PSS: Foa, Riggs, Dancu, & Rothbaum, 1993). The PSS is a 17item self-report questionnaire, scored on a four-point scale (0, 1, 2 and 3). The questions are designed to assess PTSD as characterised by the three types of symptoms described in DSM-III-R (APA, 1987): intrusion (range of 0–12); avoidance (range of 0–21) and arousal (range of 0–18). The questions are asked in relation to how the individual has felt in the last 2 weeks in relation to a traumatic event. Alphas for intrusion, avoidance and arousal subscales are displayed in Table 2. 2.1.3. Procedure The participants in this study accessed a battery of questionnaires on-line via links from various stress and trauma web sites. The battery comprised the modified IES, the PSS along with other pilot measures that were under psychometric development. These pilot measures are not reported here. An information sheet was provided detailing the nature and requirements of the study. Table 2 Means, standard deviations and alphas for all subscales Subscale
Mean
Standard deviation
Alpha
Study one Attempted Avoid Successful Avoid IES Avoidance IES Intrusion PSS Avoidance PSS Intrusion PSS Arousal
14.42 11.88 18.39 19.85 9.59 5.33 9.09
9.12 6.48 10.30 10.42 6.22 3.56 5.66
0.86 0.63 0.82 0.90 0.87 0.84 0.88
Study two Attempted Avoid Success Avoid IES Avoidance IES Intrusion TCQ Punishment TCQ Reappraisal TCQ Social Control TCQ Worry TCQ Distraction
9.91 9.14 13.76 12.93 9.71 13.17 13.41 9.97 14.66
8.66 6.90 10.94 9.98 2.95 5.65 4.42 3.09 3.25
0.86 0.69 0.86 0.90 0.73 0.51 0.87 0.83 0.74
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
901
Participants were asked to describe their most stressful or traumatic experience and state in what year this event happened. They then filled out the IES followed by the PSS. Additional comments could be added before submitting the form. 2.2. Results Approximately 2% of the responses received were ‘spoiled’ responses and subsequently discarded. These were comprised largely of incomplete or multiple submissions (in the form of identical responses in a short space of time from participants with the same demographic details). Missing values were excluded listwise. Means and standard deviations of each subscale from both measures are included in Table 2. Bivariate correlations for both attempted and successful avoidance as well as the original IES avoidance subscales are shown in Table 3. Partial correlations were calculated for the two types of avoidance (attempted and successful) with all other subscales, these are also displayed in Table 3.
Table 3 Bivariate correlations for modified IES avoidance subscales with all other subscales and partial correlations for attempted and successful avoidance Subscale
IES Avoidance
Bivariate
Partial
Attempted Avoidance
Successful Avoidance
Attempted Avoidance
Successful Avoidance
Study one Attempted Avoidance Successful Avoidance IES Avoidance IES Intrusion PSS Avoidance PSS Intrusion PSS Arousal
0.93*** 0.80*** – 0.59*** 0.67*** 0.53*** 0.54***
– 0.74*** 0.93*** 0.52*** 0.60*** 0.49*** 0.49***
0.74*** – 0.80*** 0.34*** 0.47*** 0.31*** 0.35***
– – 0.82*** 0.39*** 0.38*** 0.37*** 0.35***
– – 0.44*** 0.07 0.11 0.04 0.00
Study two Attempted Avoidance Successful Avoidance IES Avoidance IES Intrusion TCQ Punishment TCQ Reappraisal TCQ Social Control TCQ Worry TCQ Distraction
0.94*** 0.83*** – 0.68*** 0.32*** 0.03 0.27** 0.16 0.26**
– 0.79*** 0.94*** 0.65*** 0.30*** 0.03 0.25** 0.13 0.20*
0.79*** – 0.83*** 0.47*** 0.24** 0.05 0.34*** 0.03 0.27**
– – 0.82*** 0.56*** 0.19* 0.02 0.02 0.18* 0.04
– – 0.37*** 0.17 0.02 0.03 0.26** 0.11 0.15
* P< 0.05 level (2-tailed). ** P< 0.01 level (2-tailed). *** P< 0.001 level (2-tailed).
902
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
Table 3 illustrates a large (and highly significant) bivariate correlation between attempted and successful avoidance. Significant correlations were also found for both attempted and successful avoidance and all other subscales. However, when controlling for attempted avoidance successful avoidance only correlated with IES avoidance, whereas, when controlling for successful avoidance attempted avoidance remained significantly correlated with all other subscales. 2.3. Discussion This study supports the idea that attempted and successful avoidance can be measured separately. The results indicate that attempted avoidance is more closely related to measures of distress than successful avoidance. This is clearly illustrated by the associations between the PSS and the attempted and successful avoidance subscales. All three subscales of the PSS correlated with attempted avoidance when the effects of successful avoidance were partialled out. This indicates that increases in attempted but not successful avoidance are related to increases in the types of distress measured by the subscales of the PSS (intrusions, avoidance and arousal). Successful avoidance, however, did not correlate with any of the PSS subscales when attempted avoidance had been partialled out. An interesting finding was the associations between the original IES avoidance subscale and the attempted and successful avoidance subscales. The results of the bivariate correlations indicated strong positive relationships between the three avoidance subscales. These associations may initially appear to indicate that all three avoidance subscales are measuring more or less the same concept. However, the partial correlations do not support this notion. When successful avoidance was controlled for attempted avoidance still correlated very highly with the original IES avoidance subscale and moderately highly with IES intrusion whereas, when controlling for attempted avoidance, successful avoidance correlated with IES avoidance to a lesser extent and did not correlate with intrusion at all. These results suggest two things. Firstly, the original IES avoidance subscale seems more closely to reflect attempts at avoidance than actual successful avoidance and secondly, that attempted avoidance is related to intrusions but successful avoidance is not. As this was a correlational study we cannot conclude as to whether more intrusions lead to more attempts at avoidance or whether attempts to avoid trauma related stimuli leads to an increase in the frequency of intrusions.
3. Study two This study aimed to replicate the findings from study one whilst testing further hypotheses relating to the various strategies individuals use in order to control unpleasant and unwanted thoughts. Literature relating to the mental control of thoughts has illustrated the ironic effects of thought suppression whereby the mere act of trying to avoid thinking about something results in increasing the frequency of that thought. For example, Wegner and colleagues (e.g. Wegner, 1989; Wegner, Schneider, Carter, & White, 1987) found that attempts at thought suppression actually produced a preoccupation with the ‘suppressed’ thought. Wegner and colleagues propose that as the task of suppressing a thought is so difficult, it actually results in the thought
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
903
being held in consciousness while effort is made to eliminate it. In order to understand the processes that lead to the ‘suppressed’ thought being reactivated with increased frequency (rebound effect) they introduced a distractor task and found that although this didn’t facilitate the suppression task it did serve the purpose of reducing the rebound effect. Further evidence of the use of a distracting task as a means by which the process of suppression can be facilitated has since been reported (e.g. Clark, Ball, & Pape, 1991; Clark, Winton, & Thyne, 1993; Salkovskis & Campbell, 1994). These findings have implications for self-report measures of avoidance. It could be hypothesised that if a person is reporting the types of avoidance they are consciously aware of having used then the stimuli they were trying to avoid must have been held in consciousness for a period of time in order for an attempt at avoidance to be made. Therefore, the use of a distracting task as a means by which to avoid the relevant stimuli may prove to be a useful strategy with which to achieve successful avoidance. Wells and Davies (1994) using their Thought Control Questionnaire (TCQ), found that distraction in the form of doing something enjoyable or keeping busy with work, etc., had no relationship to measures of stress vulnerability or psychopathology. They also failed to find associations between social control and measures of stress vulnerability. However, Reynolds and Wells (1999) using two patient samples (one PTSD and one depressed) reported negative correlations between distraction and anxiety and depression, indicating that the more an individual used distraction as a strategy for controlling unwanted thoughts the less anxious or depressed they were. They also found a negative correlation between social control and avoidance, indicating that the more they talked about the unwanted thought the less avoidance they used. This led them to support Wells and Davies’ (1994) conclusions that these two subscales may buffer against stress vulnerability, which suggests that both distraction and social control are adaptive thought control strategies. However, using the same measure they found that punishment and worry were associated with anxiety in the depressed patients, and punishment was also associated with depression. From this it can be seen that different strategies of mental control have differing effects on distress levels. The question of whether attempted avoidance and successful avoidance relate to thought control strategies and measures of psychological distress differentially will be investigated. It is predicted that attempted avoidance will be associated with measures of distress and to dysfunctional thought control strategies, such as ‘worry’ and ‘punishment’, whereas successful avoidance will be related to more beneficial thought control strategies such as ‘distraction’ and ‘social control’. 3.1. Method 3.1.1. Participants This study consisted of 143 respondents. Eighty-nine were undergraduate students at the University of Hull, 33 were undergraduates at the University of Essex, and 21 were non-students from outside the university. There were 90 females and 51 males (N=141, two missing data). The age range was 18–63 years, with a mean age of 22.25 years (S.D.=7 years). The level of education for the sample was: up to secondary school=6; further education=125; and higher education=11 (one missing datum).
904
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
3.1.2. Measures The IES was modified as described in Section 2.1.2.1. 3.1.2.1. TCQ (Wells & Davies, 1994). The TCQ is a 30-item self report questionnaire, scored on a four-point scale (1, 2, 3, and 4). The questions were designed to assess strategies for controlling unpleasant and unwanted thoughts. There are five subscales relating to different strategies ‘Distraction’, ‘Punishment’ ‘Reappraisal’, ‘Worry’ and ‘Social Control’. Each subscale has a score range of 6–24. The questions all relate to the techniques an individual ‘generally’ uses to control unpleasant thoughts. Alphas for all TCQ subscales are displayed in Table 2. 3.1.3. Procedure The participants in this study were given a battery of questionnaires comprising the modified IES, the TCQ and other measures that were included as part of a different study. The pack also included a covering letter explaining the nature and requirements of the study. Participants were asked to think of their most stressful or traumatic experience and describe this in the space provided, also indicating what year this took place. They were then required to respond to each questionnaire with this incident in mind. Students returned their questionnaires to one of the authors while non-students were provided with pre paid envelopes in order to return their completed questionnaires individually. 3.2. Results Missing values were excluded listwise. Means and standard deviations for each subscale from all measures are included in Table 2. Bivariate and partial correlations were calculated for all measures, these are presented in Table 3. A significant bivariate correlation between attempted and successful avoidance was found (similar in magnitude to that in study one). Significant bivariate correlations were also found for both attempted and successful avoidance with the following subscales: IES avoidance, IES intrusion, TCQ punishment, TCQ social control and TCQ distraction. The IES avoidance subscale also correlated with all these subscales. However, when controlling for successful avoidance, attempted avoidance only correlated with IES avoidance, IES intrusion, TCQ punishment and TCQ worry subscales. When controlling for attempted avoidance, successful avoidance only correlated with IES avoidance and the TCQ social control subscale, with a trend towards significance with (positively) TCQ distraction (r=0.15, P<0.10) and (negatively) IES intrusion (r= 0.17, P=0.06). 3.3. Discussion The results of this study lend support to the findings of study one, as attempted avoidance was again more strongly associated with measures of distress than successful avoidance. Also there was support for the prediction that attempted avoidance would be associated with ‘maladaptive’ thought control strategies such as ‘punishment’ and ‘worry’, whereas successful avoidance would be associated with the more ‘adaptive’ thought control strategies such as ‘social control’ and ‘distraction’.
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
905
As in study one, the bivariate correlations indicate a significant association between attempted and successful avoidance but once again, when partialling out the effects of successful avoidance attempted avoidance correlated highly with the original IES avoidance and moderately with IES intrusion whereas, when partialling out the effects of attempted avoidance successful avoidance only correlated with IES avoidance. One interesting point is that a negative trend (although not quite significant at the 5% level) was evident between successful avoidance and IES intrusion indicating that as successful avoidance increases intrusive experiences decrease. As well as replicating the findings from the first study, one of the aims of this study was to investigate the relationships between the thought control strategies measured by the TCQ and attempted and successful avoidance. The reason for this was to examine Wells and Davies’ (1994) suggestion that some thought control strategies (e.g. distraction and social control) may be related to positive mental health whilst others (e.g. worry and punishment) may be related to psychopathology. The results of the partial correlations indicated that attempted avoidance was associated with punishment and worry. Attempted avoidance was also found to be correlated with intrusions, this is consistent with the findings from the thought suppression literature which suggests that attempting to suppress a thought results in increasing activation of the thought. Successful avoidance correlated significantly only with social control (negatively). This could either be interpreted in the same manner in which Reynolds and Wells (1999) did by suggesting that the more the individual talked about the thought the less avoidance they used or it could indicate that the more successful they were at avoiding the less they discussed the thought with friends. This is an important distinction, as disclosure has been found to have beneficial long-term health effects (Pennebaker, 1993) whereas, reducing levels of social support by relying increasingly on avoidance strategies, whilst initially producing a reduction in distress levels, appears to be maladaptive in the long term. Social support is often cited as an extremely important factor in the prevention of psychopathology (e.g. Brown, Andrews, Harris, Adler & Bridge, 1986) and also as part of the recovery process from psychological disorders (e.g. Stallard & Law, 1993). Therefore, it may be that while successful avoidance is unrelated to pathology concurrently there may be serious consequences if relied upon in the long term. The lack of a significant association between distraction and successful avoidance doesn’t support the findings from either the thought suppression literature or Wells and Davies (1994) study. However, the partial correlation between distraction and successful avoidance did at least indicate a trend towards an association whereas the correlation with attempted avoidance was at zero. It would be interesting to replicate in a clinical group to examine this relationship further in a group with higher distress levels. Overall these findings provide partial support for Wells and Davies’ (1994) notion of adaptive and maladaptive thought control strategies with worry and punishment being related to attempted avoidance which is also related to measures of distress.
4. General discussion The results of these studies, although only partially supporting the predictions, are interesting as they do indicate differential relationships between thought control strategies, measures of distress (IES and PSS) and attempted and successful avoidance processes.
906
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
The main findings from these studies indicate that attempted avoidance is related to measures of psychological distress (PTSD symptoms) and to mental control strategies of punishment and worry (Wells & Davies, 1994). Successful avoidance on the other hand doesn’t appear to relate to measures of psychological distress but does relate to those mental control strategies that are thought to promote positive well being (Wells & Davies, 1994). The inverse relationship with social control is interesting as it may be the case that successful avoidance, while being unrelated to pathology in the short term, may not actually promote well being in the long term. Therefore, a replication employing longitudinal methodology would be necessary in order to understand whether it is possible to avoid successfully for a long period of time and also to investigate the psychological effects of such long term avoidance. One of the limitations with employing a self report methodology to the study of avoidance is that there may be instances where an individual doesn’t feel the need to avoid (presumably because the trauma has been fully processed and the outcome is satisfactory adjustment). In this instance when asked whether they ‘did not talk about it’ an affirmative response would yield a higher score representing more avoidance use when in actual fact the individual might be indicating that they did not feel the need to talk about the event during the preceding 7 days. However, this limitation is one that is apparent in all self report measures of avoidance and isn’t limited to this study only. Employing an experimental methodology to the differentiation of attempted and successful avoidance may resolve this issue. Finally, it must be noted that this study is not putting forward a modified IES for general use. The intention was to use the items in the IES to illustrate a conceptual issue. The rationale for the study was to highlight the distinction between attempted and successful avoidance in order for this to be taken into account when new measures of avoidance are developed. Whilst this study has focussed explicitly on the role of attempted and successful avoidance following traumatic experiences we are also aware that the results of this study have implications for the wider coping literature. Therefore, it is important that researchers and professionals working within the field of health care are made aware of this distinction particularly when using measures that sum all avoidance items to yield a single score.
Acknowledgements We would like to thank to all the anonymous participants of the study. We would also like to express our gratitude to Elaine Fox and Ray Meddis.
References American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: American Psychiatric Association. Brown, G. W., Andrews, B., Harris, T., Adler, Z., & Bridge, L. (1986). Social support, self-esteem and depression. Psychological Medicine, 16, 813–831. Clark, D. M., Ball, S., & Pape, D. (1991). An experimental investigation of thought suppression. Behaviour Research and Therapy, 29, 253–257.
L. Andrews et al. / Personality and Individual Differences 33 (2002) 897–907
907
Clark, D. M., Winton, E., & Thyne, L. (1993). A further experimental investigation of thought suppression. Behaviour Research and Therapy, 31, 207–210. Creamer, M., Burgess, P., & Pattison, P. (1992). Reaction to trauma: a cognitive processing model. Journal of Abnormal Psychology, 101(3), 452–459. Dalgleish, T., Joseph, S., Thrasher, S., Tranah, T., & Yule, W. (1996). Crisis support following the herald of free enterprise disaster: a longitudinal perspective. Journal of Traumatic Stress, 9, 833–845. Dalgleish, T., Mathews, A., & Wood, J. (1999). Inhibition processes in cognition and emotion: a special case? In T. Dalgleish, & M. Power (Eds.), The handbook of cognition and emotion. Chichester, UK: Wiley. Foa, E. B., Riggs, D. S., Dancu, C. V., & Rothbaum, B. O. (1993). Reliability and validity of a brief instrument for assessing post-traumatic stress disorder. Journal of Traumatic Stress, 6(4), 459–473. Hodgkinson, P., & Joseph, S. (1995). Factor analysis of the Impact of Event Scale with female bank staff following an armed raid. Personality and Individual Differences, 19( 5), 773–775. Horowitz, M., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: a measure of subjective stress. Psychosomatic Medicine, 41( 3), 209–218. Joseph, S. (2000). Psychometric evaluation of Horowitz’s Impact of Event Scale: a review. Journal of Traumatic Stress, 13, 101–113. Joseph, S., Yule, W., & Williams, R. (1994). The Herald of Free Enterprise disaster: the relationship of intrusion and avoidance to subsequent depression and anxiety. Behaviour Research and Therapy, 32(1), 115–117. McFarlane, A. C. (1988). The aetiology of post-traumatic stress disorders following disaster. British Journal of Psychiatry, 152, 116–121. Pennebaker, J. (1993). Putting stress into words: health, linguistic, and therapeutic implications. Behaviour Research and Therapy, 31, 539–548. Perry, S., Difede, J., Musngi, G., Frances, A. J., & Jacobsberg, L. (1992). Predictors of posttraumatic stress disorder after burn injury. American Journal of Psychiatry, 149, 931–935. Reynolds, M., & Wells, A. (1999). The Thought Control Questionnaire—psychometric properties in a clinical sample, and relationships with PTSD and depression. Psychological Medicine, 29, 1089–1099. Rothbaum, B. O., Foa, E. B., Riggs, D. S., Murdock, T., & Walsh, W. (1992). A prospective examination of posttraumatic stress disorder in rape victims. Journal of Traumatic Stress, 5, 455–476. Salkovskis, P., & Campbell, P. (1994). Thought suppression induces intrusion in naturally occurring negative intrusive thoughts. Behaviour Research and Therapy, 32(1), 1–8. Stallard, P., & Law, F. (1993). Screening and psychological debriefing of adolescent survivors of life threatening events. British Journal of Psychiatry, 163, 660–665. Wegner, D. M. 1989. White bears and other unwanted thoughts: suppression, obsession, and the psychology of mental control. New York: Viking Press. Wegner, D. M., Schneider, D. J., Carter, S. R., & White, T. L. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53, 5–13. Wells, A., & Davies, M. I. (1994). The thought control questionnaire: a measure of individual differences in the control of unwanted thoughts. Behaviour Research and Therapy, 32(8), 871–878. Williams, R., Joseph, S., & Yule, W. (1994). The role of avoidance in coping with disasters: a study of survivors of the capsize of the ‘Herald of Free Enterprise’. Clinical Psychology and Psychotherapy, 1(2), 87–94. Yule, W., Ten Bruggencate, S., & Joseph, S. (1994). Principal components analysis of the Impact of Event Scale in adolescents who survived a shipping disaster. Personality and Individual Differences, 16, 685–691.