- Email: [email protected]
Adverse psychological effects in women attending a second-stage breast cancer screening
Journal of Psychosomatic Research 52 (2002) 303 – 309
Adverse psychological effects in women attending a second-stage breast cancer screening Bonifacio Sandina,*, Paloma Chorota, Rosa M. Valientea, Lourdes Lostaob, Miguel A. Santeda a
Facultad de Psicologı´a, Universidad Nacional de Educacio´n a Distancia (UNED), Ciudad Universitaria s/n, 28040 Madrid, Spain b Universidad Pu´blica de Navarra, Campus Arrosadı´a, Pamplona, Spain Received 28 June 2000; accepted 22 February 2001
Abstract Objective: The purpose of this study was to examine the emotional and psychopathological impact associated with a second-stage screening for breast cancer. Method: We used a short-term longitudinal design. Interviews were conducted with 1195 women of 45 – 65 years old in three temporal conditions (premammogram, postmammogram, and follow-up). Participants included women attending for regular breast cancer screening who were recalled for a further mammogram (i.e., second-stage breast cancer screening) and women who were not recalled. Affectivecognitive concerns about cancer (worry, fear, and perceived vulnerability) were rated using a 10-point Likert scale. Psychopathology was assessed using the Hopkins Symptom Check ListRevised (SCL-90-R). Results: Women attending the second-stage screening exhibited significantly higher levels of breast cancer
worries, fears, and beliefs than women attending for routine screening before obtaining the results of the mammogram. This affective-emotional impact disappeared quickly and was not relevant 2 months following the mammogram. Despite the fact that levels of psychopathological symptoms were higher in the premammogram condition, there were no differences between groups on these measures. Conclusion: These results provide support for the hypothesis that women recalled for further mammograms tend to experience high levels of affective-cognitive distress but not psychopathological symptoms. Moreover, results do not sustain the prediction that this psychological impact persists beyond receipt of a negative result. Some recommendations to reduce these psychological side effects are suggested. D 2002 Elsevier Science Inc. All rights reserved.
Keywords: Breast cancer screening; Fear; Mammography; Psychopathology; Stress; Worry
Introduction Screening programmes based on the periodic use of mammography are the most effective method for the early detection of breast cancer. The principal benefit of these programmes derives from the finding that they can allow early treatment and a reduction in breast cancer mortality. A number of studies have demonstrated substantial mortality reductions from breast cancer by about 4– 40% [1 –4]. After recommendations by public organizations (e.g., the U.S. National Cancer Institute) promoting breast cancer screening for women of 50 or over, the percentage of women who receive regular mammography has increased
* Corresponding author. E-mail address: [email protected] (B. Sandin).
progressively [5 – 7]. For example, as was stated by Rakowski et al. [8], ‘‘The percentage of women ages 50– 75 who obtain screening mammography has increased steadily since the middle 1980s. Data from the 1992 National Health Interview Survey of Topics Related to Cancer Control indicate that about 68% of women in this age group report ever having had a mammogram and approximately 55% report having had one in the past 2 years’’ (p. 433). It is, therefore, a well-known fact that mammogram acceptance and use has been increasing in North America [8,9] as well as in Europe [7], including Spain [10]. In line with this growth, an increasing number of women who, attending breast cancer screening, receive notification of abnormal mammogram have been reported. Screening tests (i.e., mammography) regularly produce inaccuracies and false positives. The proportion of mammograms reported as inconclusive, unsatisfactory, or abnormal in large-scale
0022-3999/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 2 - 3 9 9 9 ( 0 1 ) 0 0 2 2 7 - 6
304
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
screening programmes is about 10% [11,12]. This indicates that a remarkable percentage of women who attend routine screening programmes are recalled for further mammograms. The majority of these women who need to repeat mammograms or undergo biopsy are not considered to be at high risk for developing breast cancer. Even though breast cancer screening has been associated with health benefits, it could also produce some cost to women attending regularly, such as worktime lost, workload, fear of the radiation, pain of the mammography, etc. However, a cost that has scarcely been investigated is the possible psychological distress and morbidity associated with the screening process, particularly the psychological impact produced in women recalled for further exams where the initial mammogram was interpreted as inconclusive or abnormal [8,13]. As Warren [14] suggested, it is possible that a number of women who receive a recall letter temporarily experience the diagnosis of cancer. According to current data, these ‘‘false positives’’ represent a high percentage of women attending routine screening programmes [9]. In line with Wardle and Pope’s [11] suggestion that ‘‘false positive’’ results in breast cancer screening probably represent a major area of concern so far as psychological costs are concerned, some authors have reported that women who receive false positive results have been found to experience elevated anxiety after receiving a letter recalling them for further mammography [15] as well as a possible disturbed quality of life [16]. Similarly, Stewart-Brown and Farmer [17] have recently suggested that screening programmes could seriously damage the health. However, little research has been done to specifically investigate the psychological side effects associated with notification of an abnormal mammogram. Ellman et al. [18] found no significant differences in psychiatric morbidity between women in the false positive group and those attending for regular breast cancer screening. However, these authors reported significantly higher levels of anxiety in the false positive group, though they did not persist at a 3-month follow-up. Similarly, Clutton et al. [12] found that women who attended a routine breast cancer screening and were notified of the need to attend for further assessment of a mammographically detected abnormality showed low levels of anxiety and psychopathology 6 weeks after the second testing. Clutton et al., however, did not use a control group (e.g., a sample of women who were not recalled). In a recent study, Sandin et al. [19] found no significant differences in hypochondriacal attitudes between women attending a second-stage screening and women attending routine screening, neither just before the screening nor in the follow-up (2 months after the mammogram). Lerman et al. [9] analyzed the psychological side effects of receiving abnormal mammogram results on women’s anxiety and breast cancer worries. The authors assessed psychological variables on a sample of 308 women over 50 by means of telephone surveys conducted 3 months following a screening mammogram. Subjects included
women with normal mammograms (N), nonsuspicious abnormal mammograms (A2), and suspicious abnormal mammograms (A3). Lerman et al. found a positive relationship between the index of suspicion of the mammogram and the levels of anxiety and worry. They observed consistent upward trends in negative psychological impact as levels of mammogram suspicion increased from N to A2 to A3 and concluded saying that because ‘‘concerns persisted even after these women had received additional tests ruling out cancer suggests than women with abnormal mammograms perceive themselves to be at elevated risk of developing breast cancer in the future’’ (p. 265). So, in contrast to the Ellman et al.’s [18] report, Lerman’s group found that the psychological impact of receiving abnormal mammogram results persists beyond receipt of the negative result (i.e., after knowing that they did not have cancer). In a more recent study, Cockburn et al. [20] reported that women who attended a breast cancer screening and were recalled for a further mammogram exhibited higher levels of psychological dysfunction than women who were not recalled. This psychological impact, which was measured 1 week after obtaining notification that there was no sign of cancer, suggests that at least a short-term psychological side effect associated to the second-stage screening could be present beyond possible adverse psychological effects related with routine screening. In spite of some relevant and suggestive studies, existing research on the psychological impact associated to ‘‘false positive’’ results in breast cancer screening is not convincing. Problems with some of the published reports include the use of low samples of women and cross-sectional methodology. The present study was designed to examine the psychological impact of receiving abnormal mammogram results in women attending the Breast Cancer Detection Programme (BCDP) of Navarra (Spain). Consistent with the existing literature, we predicted that three affectivecognitive measures of psychological distress associated to breast cancer screening, i.e., worry about breast cancer, fear of breast cancer, and perceived vulnerability, would be higher in women recalled for further mammograms than in women attending regular screening. We also predicted greater levels of psychopathological symptoms in the first group than in women with normal mammograms (i.e., women attending routine screening). The study used a short-term longitudinal design with a large sample of women who had attended the breast cancer screening programme.
Method Subjects Participants in this study were women, 45– 65 years old, attending the BCDP of Navarra (Spain) for regular breast cancer screening or for a rescreening (a second-stage mammography). According to these two types of participants, we
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
created two different groups, i.e., a routine screening (RS) group and a second-stage screening (SSS) group. All of the subjects were consecutively recruited. The RS group consisted of 600 women attending for regular screening. These women were not recalled for further mammogram and were randomly assigned to this group from a previous sample of 1600 participants. Two women who were diagnosed as having breast cancer were excluded from the study. The SSS group consisted of 600 women recalled for rescreening. The women of this group received a telephone call asking them to attend again because the results of the previous (routine) mammogram were unsatisfactory, abnormal, or inconclusive. Only three rescreened women were diagnosed of breast cancer and were discharged from the study. About 21% of the final 597 subjects were found to have minor or benign breast problems (nodule, fibroma, cyst, etc.). The final sample was composed of 1195 women (i.e., there were 598 subjects in the RS group and 597 in the SSS group). About 76% of the participants ranged in age from 45 to 55, 59% had lower than a high school education, and 76.5% were married. Average age of the participants was 51.4 years (S.D. = 5.2) for the RS group and 52.6 years (S.D. = 5.4) for the SSS group. There were no significant differences between the two groups in terms of age, education, and marital status. All of the subjects voluntarily agree to participate in the study and no one refused to take part or dropped out. Measures Demographic variables Demographic variables included age, educational level, and marital status. They were assessed during the first interview. Cognitive-affective variables Emotional variables were assessed in a similar way to the method used by Lerman et al. [9]. The following three variables were measured: (a) worry about breast cancer, (b) fear of having breast cancer, and (c) perceived vulnerability (belief of having breast cancer). Women were asked about their current levels of worry (i.e., worry about getting breast cancer), fear (i.e., fear of getting a breast cancer), and belief of having breast cancer or perceived vulnerability (i.e., the extent to which a woman thinks that she could get breast cancer). The subjects rated their levels on a 10-point Likert scale with points labeled none/absent (1), moderate (5 – 6), and a lot/very severe (10). Psychopathological symptoms Psychopathology was measured with the Hopkins Symptom Check List-Revised (SCL-90-R; [21]; Spanish version validated by Gonza´lez de Rivera et al. [22] and Valiente et al. [23]). The SCL-90-R is a 90-item questionnaire, which
305
includes the following nine subscales: Somatization (12 items), Obsessive-compulsive (10 items), Interpersonal sensitivity (9 items), Depression (13 items), Anxiety (10 items), Anger/hostility (6), Phobic anxiety (7), Paranoia (6), and Psychoticism (10 items). Subjects respond to each of 90 statements on a five-point Likert scale ranging in severity from ‘‘not at all’’ (scored as 0) to ‘‘extremely’’ (scored as 4). The score of each subscale was the mean of the corresponding subscale items. The global severity index was the mean of the 90 severity ratings. Internal consistence (a coefficients) for the subscales of the Spanish SCL version has ranged from .70 (Paranoia) to .89 (Depression and Phobic anxiety) [23]. Procedure All of the subjects were selected from women attending the BCDP of Navarra (Spain). Every 2 years, women in this program who are 45 – 65 years old receive a letter inviting them for a free mammogram. The majority of women have normal mammograms (RS women). However, results of the mammogram are inconclusive or abnormal (suspicious abnormal mammogram) for some women (about 10%). These women (SSS women) are invited again (this time by a telephone call) for a new mammogram and must attend the screening 2 days after the referral call (this second-stage screening also includes, in some instances, an ultrasonic imaging procedure). Packets of questionnaires were administered in two testing sessions separated by 2 months. The first session (i.e., premammography condition) was performed just before the mammogram, and the second (i.e., follow-up condition) about 2 months after the mammogram. During these periods, sociodemographics (only in precondition), affective variables, and psychopathology were assessed. In addition, the women of the SSS group were also tested (only for affective variables) 2 days following the mammogram (i.e., postmammography condition). In premammography condition, questionnaires were individually completed in an interview format. In postmammography and follow-up conditions, questionnaires were completed by telephone interviews, which were previously contacted. All questionnaires were completed by a trained female psychologist. Statistical analysis The statistical methods used included: (1) Pearson’s c2 test with the Yates continuity correction for differences between the proportion of women with moderate to severe levels (score 5) on the affective-cognitive variables (i.e., worry, fear, and belief). (2) Two-way group (SSS vs. RS) Conditions (premammography vs. followup) repeated-measures analysis of variance (ANOVA); separate ANOVAs were computed on the psychopathological measures (subscales and global index of the SCL90). (3) Two-way group (moderate/severe vs. none/mild
306
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
concerns) Conditions (premammography vs. follow-up) repeated-measures ANOVA; separate ANOVAs were computed on the anxiety, depression, and global index of the SCL-90. (4) Two sets of stepwise regression analyses for each group of women (SSS and RS groups).
Results Cognitive-affective impact Table 1 shows the proportions of SSS and RS women who reported moderate to severe breast cancer worries, fears, and beliefs before the screening mammogram. The majority of women of both groups reported worry and fear related to breast cancer, but only a low percentage of women reported the belief of having breast cancer in the premammogram condition. The SSS subjects suffered more frequently from worry ( P < .001), fear ( P < .001), and negative beliefs ( P < .001) than did the RS women. However, at follow-up (i.e., 2 months after the mammogram), the percentages of subjects who reported moderate to severe breast cancer concerns (worries, fears, and beliefs) were very low (see Table 2). Again, the numbers of women reporting worry and fear were higher in the SSS group than in the RS group. No differences were found in the frequency of subjects reporting beliefs of having breast cancer during this condition.
Table 2 Numbers and percentages (in parentheses) of subjects who reported breast cancer concerns (worry, fear, and beliefs) during the follow-up period (2 months after the mammogram) for each group Group RS (n = 598)
c2
Worry about breast cancer Moderate to severe 19 (3.2) None to mild 578 (96.8)
5 (0.8) 593 (99.2)
7.2**
Fear of having breast cancer Moderate to severe 19 (3.2) None to mild 578 (96.8)
5 (0.8) 593 (99.2)
7.2**
Belief of having breast cancer Moderate to severe 8 (1.3) None to mild 589 (98.7)
3 (0.5) 595 (99.5)
1.5
Affective response
SSS (n = 597)
SSS = second stage screening; RS = routine screening. ** P < .001.
vs. follow-up). The group factor was not significant for any of the SCL variables. However, there were significant differences between the conditions for all variables ( P < .001): Somatization ( F = 80.7; df = 1,1193), Obsessive-compulsive ( F = 168.6; df = 1,1193), Interpersonal sensitivity ( F = 82.9; df = 1,1193), Depression ( F = 63.5; df = 1,1193), Anxiety ( F = 71.1; df = 1,1193), Anger/hostility ( F = 42.8; df = 1,1193), Phobic anxiety ( F = 60.4; df = 1,1193), Paranoia ( F = 65.1; df = 1,1193), Psychoticism ( F = 19.7; df = 1,1193), and Global severity index ( F = 148;
Psychopathological symptoms As shown in Table 3, mean scores on the SCL variables appear to be similar for both groups of women. Moreover, the scores tend to be lower during the follow-up period. Two-way repeated-measures ANOVAs (Group CondiConditions) were computed to test differences between groups (SSS vs. RS) and conditions (premammography Table 1 Numbers and percentages (in parentheses) of subjects who reported breast cancer concerns (worry, fear, and beliefs) during the premammogram period for each group Group RS (n = 598)
c2
Worry about breast cancer Moderate to severe 514 (86.1) None to mild 83 (13.9)
430 (71.9) 168 (28.1)
35.4***
Fear of having breast cancer Moderate to severe 506 (84.8) None to mild 91 (15.2)
432 (72.2) 166 (27.8)
26.9***
Belief of having breast cancer Moderate to severe 94 (15.7) None to mild 503 (84.3)
16 (2.7) 582 (97.3)
59.5***
Affective response
SSS (n = 597)
SSS = second stage screening; RS = routine screening. *** P < .001.
Table 3 Psychopathological measures (SCL-90) by condition (premammography vs. follow-up) and group category Premammography
Follow-up (2 months)
SCL-90 measures
Mean
S.D.
Mean
S.D.
SSS group (n = 597) Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Anger/hostility Phobic anxiety Paranoia Psychoticism Global severity index
0.49 0.43 0.31 0.43 0.47 0.17 0.23 0.26 0.05 0.35
0.41 0.40 0.25 0.38 0.38 0.25 0.29 0.36 0.13 0.24
0.37 0.20 0.19 0.32 0.30 0.11 0.14 0.18 0.03 0.22
0.37 0.26 0.21 0.37 0.33 0.18 0.23 0.29 0.08 0.21
RS group (n = 598) Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Anger/hostility Phobic anxiety Paranoia Psychoticism Global severity index
0.46 0.46 0.34 0.45 0.41 0.22 0.24 0.24 0.04 0.34
0.38 0.38 0.29 0.37 0.33 0.27 0.27 0.35 0.11 0.32
0.36 0.26 0.20 0.36 0.28 0.15 0.16 0.15 0.02 0.24
0.37 0.30 0.22 0.38 0.30 0.21 0.22 0.29 0.09 0.20
SSS = second stage screening; RS = routine screening.
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
df = 1,1190). No interaction effects were found between the two factors. Relationship between cognitive-affective impact and symptoms of anxiety and depression According to data presented in Tables 1 –3, we wanted to test the hypothesis that worry about breast cancer, fear of breast cancer, and belief of having breast cancer are related to anxiety and depression. Thus, we assessed whether women who reported moderate to severe concerns about breast cancer (worry, fear, or belief) during the premammography condition had higher SCL-90 anxiety and depression scores than women who reported none to mild concerns. Given in Table 4 are the means and S.D. on anxiety, depression, and global severity index (during the premammography and at follow-up) for women who reported moderate to severe vs. none to mild concerns about breast cancer during the premammogram. Repeated-measures ANOVA (Group Conditions) demonstrated (1) significant group differences (moderate/severe vs. none/mild worry) on anxiety ( F = 20.5; df = 1,1193; P < .001), depression ( F = 7.6; df = 1,1193; P < .01), and global severity index ( F = 8.5; df = 1,1190; P < .01); (2) significant group differences (moderate/severe vs. none/mild fear) on anxiety ( F = 18.3; df = 1,1193; P < .001), depression ( F = 9.5; df = 1,1193; P < .01), and global severity index ( F = 11.9; df = 1,1190; P < .01); and (3) significant group Table 4 SCL-90 anxiety and depression scores in women who reported moderate to severe vs. none to mild breast cancer concerns during the premammography period (N = 1195) Premammography
Follow-up (2 months)
Subgroups measure
Mean
S.D.
Mean
S.D.
SCL-90 anxiety Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.46 0.32 0.46 0.32 0.59 0.41
0.36 0.29 0.36 0.30 0.44 0.33
0.30 0.21 0.30 0.22 0.40 0.26
0.32 0.24 0.32 0.26 0.40 0.29
SCL-90 depression Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.45 0.36 0.45 0.35 0.54 0.42
0.34 0.27 0.34 0.26 0.40 0.32
0.38 0.32 0.38 0.32 0.43 0.36
0.37 0.35 0.38 0.34 0.43 0.36
SCL-90 global severity index Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.34 0.29 0.34 0.28 0.41 0.32
0.23 0.21 0.24 0.21 0.31 0.21
0.23 0.18 0.24 0.18 0.28 0.22
0.20 0.17 0.20 0.17 0.26 0.20
307
differences (moderate/severe vs. none/mild belief) on anxiety ( F = 42.2; df = 1,1193; P < .001), depression ( F = 11.5; df = 1,1193; P < .01), and global severity index ( F = 21.4; df = 1,1190; P < .001). There were also significant differences between the conditions (premammography vs. follow-up) for all variables ( P < .001). No interaction effects were found between the factors, except on anxiety (Worry Conditions: F = 3.9, P < .05; Fear Conditions: F = 4.7, P < .05; Belief Conditions: F = 4.5, P < .05). These results suggest that women who reported moderate to severe levels of cognitive-affective concerns about breast cancer had higher SCL-90 anxiety and depression scores than the remainder women. Separate multiple stepwise regression analyses were performed for the two groups of women (SSS and RS) to examine whether the cognitive-affective impact experienced during the premammogram period was able to predict the levels of anxiety and depression reported during the conditions of premammography (Time 1) and follow-up (Time 2). A first set of analyses was conducted to analyze the effects of the cognitive-affective variables on Time 1 anxiety and depression. Results showed a low predictive power and different patterns according to the two groups. For the SSS group, the variables belief and fear of having breast cancer together significantly predicted Time 1 anxiety [adjusted r2=.05; F(2,594) = 30.2, P < .001] and depression [adjusted r2=.04; F(2,594) = 11.4, P < .001]. For the RS group, only worry about breast cancer predicted Time 1 anxiety (adjusted r2=.03; F(1,596) = 14.2, P < .001]. The remainder independent variables did not contribute significantly to the prediction of Time 1 anxiety or depression. A second set of analyses was performed to test the effects of the cognitive-affective variables on Time 2 anxiety and depression. After controlling for Time 1 levels of anxiety or depression, the cognitive-affective variables failed to add significantly to the prediction of Time 2 anxiety or depression. However, Time 1 anxiety was a good predictor of Time 2 anxiety for SSS [adjusted r2=.51; F(1,595) = 600.7, P < .001] as well as for RS [adjusted r2=.42; F(1,596) = 430.8, P < .001] subjects. Similarly, Time 1 depression contributed significantly to the prediction of Time 2 levels of depression for both groups of women [SSS group: adjusted r2=.62; F(1,595) = 969.3, P < .001; RS group: adjusted r2=.50; F(1,596) = 615.1, P < .001].
Discussion In the present study, we examined differences between women attending a second-stage breast cancer screening and women attending routine screening on affective-cognitive distress and psychopathology. We also examined whether the psychological impact is only temporarily related to the period that precedes the mammography or persists beyond
308
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
receipt of the negative result (i.e., being informed that there was no sign of cancer). A major finding of this research is that, according to our prediction, women recalled for further mammograms suffered higher levels of affective-cognitive impact than women attending regular breast cancer screening. These differences were found on three separate measures: specific worry, fear, and thinking about getting breast cancer (perceived vulnerability). Even though both groups of women appear to be emotionally affected by attending the breast cancer screening program, a high proportion of affected women belong to the SSS group. These results are consistent with the study of Lerman et al. [9] who found that subjects with abnormal mammograms reported greater negative impact than subjects with normal mammograms. In addition, these authors indicated a positive relationship between the index of suspicion of mammograms and women’s subsequent self-reports of anxiety and worry. However, in contrast with findings of Lerman et al. [9], only a low percentage of women attending the second-stage screening reported at follow-up (2 months after the mammogram) from moderate to severe levels of worry (3.2%), fear (3.2%), or belief (1.3%) of getting breast cancer. Based on the results of the present study, we may suggest that affective-cognitive distress did not persist following notification of the negative result. Our results are also in line with Cockburn et al. [20] and Sandin et al. [24] who accounted for higher psychological side effects for recalled than for nonrecalled women. Another important finding of our study is that the affective-cognitive concerns detected during the premammogram condition in SSS women disappeared quickly. In fact, only 5.9% of women were worried about getting breast cancer, 5.4% had a fear of getting breast cancer, and 2.7% believed that they had breast cancer in the postmammogram condition. This result suggests that possible emotionalcognitive impact experienced by women who attend complementary mammograms (i.e., SSS women) is a phenomenon limited to the short period of time, which precedes mammography, i.e., from the recall to the notification of the mammogram results. These data contrast with findings of Lerman et al. [9] and are in accordance with research that has failed to find evidence of sustained adverse psychological effects after an initial suspicious result not confirmed by subsequent testing [12,18,24]. Our results do not provide support for the prediction that SSS women would exhibit higher levels of psychopathological symptoms than RS women, neither during the premammogram condition nor during the follow-up. Overall, mean values on SCL-90-R measures are (1) comparable to the general population [22,23], (2) similar for both groups of women (Table 3), and (3) higher before mammogram than that in the follow-up. These findings are in line with results of Ellman et al. [18] who reported no differences in psychiatric morbidity between ‘‘false positive’’ women and those attending for regular breast cancer screening. Our data are also
consistent with results reported by Dean et al. [25], Chen et al. [26], and Sandin et al. [19]. Dean et al. found no differences in psychiatric morbidity between women attending a routine breast cancer screening and a control group (matched nonattenders); attending women were assessed 6 months after the screening. Chen et al. documented that only around 20% of the women who attend a second-stage breast cancer screening showed General Health Questionnaire (GHQ) scores reaching the probable psychiatric morbidity level [17]. Sandin et al. reported no significant differences in hypochondriacal attitudes between women attending regular vs. second-stage breast cancer screening. Likewise, Wardle et al. [27] pointed out that women who attended an ovarian cancer screening did not have elevated scores on psychiatric morbidity 1 year after receiving a ‘‘false positive’’ notification. To our knowledge, this is one of the first published studies that (a) was specifically focused on the psychological impact of receiving abnormal or inconclusive mammogram results and (b) used a longitudinal design and a large sample of women attending a breast cancer screening programme. Despite the presence of a notably emotional-cognitive impact before the mammogram in women attending the second-stage screening, our data do not provide support for the hypothesis that such psychological impact persists after the mammogram is performed (in the present breast screening program, the negative results are provided just after the mammogram). In addition, neither the distress suffered by women attending the second-stage screening nor the impact experienced by women attending the regular screening can be said to persist 2 months following the mammogram. These data suggest that, in order to minimize the psychological impact, it could be very important to minimize the period of time between the recall phone call and the rescreening as well as to inform women about results of the mammogram as soon as possible. Moreover, this study also highlights the significance of reducing the period from the time of recall to the mammogram as much as possible. Such implications are in line with Marteau’s [13] suggestion that ‘‘Uncertainly about test results was associated with more concern about health than receiving a clear positive or negative result’’ (p. 4). The conclusions about the timing of the recall phone calls and time to appointments are legitimate and should be noted by those who organise the administration of the screening. Other possible useful ways to decrease the psychological impact associated with screening programmes include the provision of rational and reassuring information to women (e.g., to inform in the letter or telephone call that in general the number of abnormal mammograms is elevated and that it does not mean a high risk of developing breast cancer). It is also worth noting that the quality of reassurance by the physician who tells the patient that there is no significant finding may have certain impact on the extent of concern retained by the rescreened women. In fact, a benefit of breast screening programmes may be derived from reassurance for those women in whom no evidence of breast cancer is found.
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
There are some possible limitations to this work. First, the use of telephone interviews in postmammography and follow-up conditions could explain, to some extent, differences between the conditions. However, this is not likely because the interviews were conducted by the same psychologist and in a similar way to during the premammogram condition. In addition, there is evidence that telephone interviewing is generally an accurate and valid method, and as compared with mail, it appears to be a better option [28]. Second, in this report, we have not examined personality and other individual characteristics as possible predictors of the psychological impact associated with breast cancer screening. As has been reported in some studies [12,26,29,30], neuroticism and other individual characteristics could moderate the psychological distress of attending breast cancer screening as well as the impact of distress on mammography use. Acknowledgments This research was supported by the Grant No. 2204 resolucio´n 634/96 from the Departamento de Salud del Gobierno de Navarra (Spain). We thank Drs. Nieves Ascunce and Angel Del Moral for their help with this work and their invaluable suggestions and Olga Ferna´ndez Soto for her collaboration with data collection. References [1] Chamberlain J, Palli D. Evidence of the effectiveness of breast cancer screening. In: Gad A, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. pp. 15 – 23. [2] Strax P. Mass screening for control of breast cancer. Cancer 1984;53: 665 – 70. [3] Shapiro S, Venet W, Strax P, Venet L, Roeser R. Selection, follow-up, and analysis in the Health Insurance Plan Study: a randomized trial with breast cancer screening. Natl Cancer Inst Monogr 1985;67:65 – 74. [4] American Cancer Society. Cancer facts and figures — 1998. Atlanta, GA: American Cancer Society, 1998. [5] Howe HL. Repeat mammography among women over 50 years of age. Am J Prev Med 1992;8:182 – 5. [6] Breen N, Kessler L. Trends in cancer screening — United States, 1987 and 1992. Morb Mortal Wkly Rep 1996;45:57 – 61. [7] Gad A, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. [8] Rakowski W, Stoddard AM, Rimer BK, Fox A, Andersen MR, Urban N, Lane DS, Costanza ME. Confirmatory analysis of opinions regarding the pros and cons of mammography. Health Psychol 1997;16:433 – 41. [9] Lerman C, Trock B, Rimer BK, Jepson C, Brody D, Boyce A. Psychological side effects of breast cancer screening. Health Psychol 1991;10:259 – 67. [10] Ascunce N, Del Moral A. Compliance in breast cancer screening: the
[11] [12]
[13] [14] [15]
[16] [17] [18]
[19]
[20] [21] [22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
309
Spanish experience. In: Gad AM, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. pp. 47 – 53. Wardle J, Pope R. The psychological costs of screening for cancer. J Psychosom Res 1992;36:609 – 24. Clutton S, Pakenham KI, Buckley B. Predictors of emotional wellbeing following a ‘‘false positive’’ breast cancer screening result. Psychol Health 1999;14:263 – 75. Marteau TM. Psychology and screening: narrowing the gap between efficacy and effectiveness. Br J Clin Psychol 1994;33:1 – 10. Warren R. The debate over mass mammography in Britain. Br Med J 1998;297:969 – 72. Sutton S, Saidi G, Bickler G, Hunter J. Does routine screening for breast cancer raise anxiety? Results from a three wave prospective study in England. J Epidemiol Community Health 1995;49:413 – 8. Gram IT, Lund E, Slenker SE. Quality of life following a false positive mammogram. Br J Cancer 1990;62:1018 – 22. Stewart-Brown S, Farmer A. Screening could seriously damage your health. Br Med J 1997;314:533 (Editorial). Ellman R, Angelin N, Christians A, Moss S, Chamberlain J, Maguire P. Psychiatric morbidity associated with screening for breast cancer. Br J Cancer 1989;60:781 – 4. Sandin B, Chorot P, Valiente RM, Lostao L, Santed MA. Actitudes hipocondrı´acas en mujeres antes y despue´s de participar en pruebas complementarias de deteccio´n de ca´ncer de mama. Psiquis 2000;21: 232 – 40. Cockburn J, Staples M, Hurley S, De Luise T. Psychological consequences of screening mammography. J Med Screening 1994;1:7 – 12. Derogatis LR. SCL-90-R administration, scoring and procedures manual: II. Towson, MD: Clinical Psychometric Research, 1983. Gonza´lez de Rivera JL, de las Cuevas C, Garcı´a-Marco R, Monterrey AL, Rodrı´guez-Pulido F, Henry-Benı´tez M. Morbilidad psiquia´trica menor en la poblacio´n general de Tenerife. Psiquis 1990; 11:11 – 22. Valiente RM, Sandin B, Chorot P, Santed MA, Gonza´lez de Rivera JL. Sucesos vitales mayores y estre´s: efectos del cambio por migracio´n asociados a manifestaciones psicopatolo´gicas. Psiquis 1996;17:211 – 30. Sandin B, Chorot P, Valiente RM, Lostao L, Santed MA. Ansiedad anticipatoria asociada a las pruebas complementarias de screening de ca´ncer de mama. Psicothema 2001;13:248 – 54. Dean C, Roberts MM, French K, Robinson S. Psychiatric morbidity after screening for breast cancer. J Epidemiol Community Health 1986;40:71 – 5. Chen CC, David A, Thompson K, Smith C, Lea S, Fahy T. Coping strategies and psychiatric morbidity in women attending breast assessment clinics. J Psychosom Res 1996;40:265 – 70. Wardle J, Pernet A, Collins W, Bourne T. False positive results in ovarian cancer screening. Psychol Health 1994;10:33 – 40. Zapka JG, Bigelow C, Hurley T, Ford LD, Egelhofer J, Cloud WM, Sachsse E. Mammography use among sociodemographically diverse women: the accuracy of self-report. Am J Public Health 1996;86: 1016 – 21. Sandin B, Chorot P, Lostao L, Valiente RM, Santed MA. Predictores psicolo´gicos y sociodemogra´ficos de la ansiedad anticipatoria ante la participacio´n en ‘‘segundas pruebas’’ de deteccio´n de ca´ncer de mama. Rev Psicopatol Psicol Clin 2001;6:17 – 36. Schwartz MD, Taylor KL, Siegel JE, Lamdan RM, Moran K. Distress, personality, and mammography utilization among women with a family history of breast cancer. Health Psychol 1999;18:327 – 32.
Adverse psychological effects in women attending a second-stage breast cancer screening Bonifacio Sandina,*, Paloma Chorota, Rosa M. Valientea, Lourdes Lostaob, Miguel A. Santeda a
Facultad de Psicologı´a, Universidad Nacional de Educacio´n a Distancia (UNED), Ciudad Universitaria s/n, 28040 Madrid, Spain b Universidad Pu´blica de Navarra, Campus Arrosadı´a, Pamplona, Spain Received 28 June 2000; accepted 22 February 2001
Abstract Objective: The purpose of this study was to examine the emotional and psychopathological impact associated with a second-stage screening for breast cancer. Method: We used a short-term longitudinal design. Interviews were conducted with 1195 women of 45 – 65 years old in three temporal conditions (premammogram, postmammogram, and follow-up). Participants included women attending for regular breast cancer screening who were recalled for a further mammogram (i.e., second-stage breast cancer screening) and women who were not recalled. Affectivecognitive concerns about cancer (worry, fear, and perceived vulnerability) were rated using a 10-point Likert scale. Psychopathology was assessed using the Hopkins Symptom Check ListRevised (SCL-90-R). Results: Women attending the second-stage screening exhibited significantly higher levels of breast cancer
worries, fears, and beliefs than women attending for routine screening before obtaining the results of the mammogram. This affective-emotional impact disappeared quickly and was not relevant 2 months following the mammogram. Despite the fact that levels of psychopathological symptoms were higher in the premammogram condition, there were no differences between groups on these measures. Conclusion: These results provide support for the hypothesis that women recalled for further mammograms tend to experience high levels of affective-cognitive distress but not psychopathological symptoms. Moreover, results do not sustain the prediction that this psychological impact persists beyond receipt of a negative result. Some recommendations to reduce these psychological side effects are suggested. D 2002 Elsevier Science Inc. All rights reserved.
Keywords: Breast cancer screening; Fear; Mammography; Psychopathology; Stress; Worry
Introduction Screening programmes based on the periodic use of mammography are the most effective method for the early detection of breast cancer. The principal benefit of these programmes derives from the finding that they can allow early treatment and a reduction in breast cancer mortality. A number of studies have demonstrated substantial mortality reductions from breast cancer by about 4– 40% [1 –4]. After recommendations by public organizations (e.g., the U.S. National Cancer Institute) promoting breast cancer screening for women of 50 or over, the percentage of women who receive regular mammography has increased
* Corresponding author. E-mail address: [email protected] (B. Sandin).
progressively [5 – 7]. For example, as was stated by Rakowski et al. [8], ‘‘The percentage of women ages 50– 75 who obtain screening mammography has increased steadily since the middle 1980s. Data from the 1992 National Health Interview Survey of Topics Related to Cancer Control indicate that about 68% of women in this age group report ever having had a mammogram and approximately 55% report having had one in the past 2 years’’ (p. 433). It is, therefore, a well-known fact that mammogram acceptance and use has been increasing in North America [8,9] as well as in Europe [7], including Spain [10]. In line with this growth, an increasing number of women who, attending breast cancer screening, receive notification of abnormal mammogram have been reported. Screening tests (i.e., mammography) regularly produce inaccuracies and false positives. The proportion of mammograms reported as inconclusive, unsatisfactory, or abnormal in large-scale
0022-3999/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 2 - 3 9 9 9 ( 0 1 ) 0 0 2 2 7 - 6
304
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
screening programmes is about 10% [11,12]. This indicates that a remarkable percentage of women who attend routine screening programmes are recalled for further mammograms. The majority of these women who need to repeat mammograms or undergo biopsy are not considered to be at high risk for developing breast cancer. Even though breast cancer screening has been associated with health benefits, it could also produce some cost to women attending regularly, such as worktime lost, workload, fear of the radiation, pain of the mammography, etc. However, a cost that has scarcely been investigated is the possible psychological distress and morbidity associated with the screening process, particularly the psychological impact produced in women recalled for further exams where the initial mammogram was interpreted as inconclusive or abnormal [8,13]. As Warren [14] suggested, it is possible that a number of women who receive a recall letter temporarily experience the diagnosis of cancer. According to current data, these ‘‘false positives’’ represent a high percentage of women attending routine screening programmes [9]. In line with Wardle and Pope’s [11] suggestion that ‘‘false positive’’ results in breast cancer screening probably represent a major area of concern so far as psychological costs are concerned, some authors have reported that women who receive false positive results have been found to experience elevated anxiety after receiving a letter recalling them for further mammography [15] as well as a possible disturbed quality of life [16]. Similarly, Stewart-Brown and Farmer [17] have recently suggested that screening programmes could seriously damage the health. However, little research has been done to specifically investigate the psychological side effects associated with notification of an abnormal mammogram. Ellman et al. [18] found no significant differences in psychiatric morbidity between women in the false positive group and those attending for regular breast cancer screening. However, these authors reported significantly higher levels of anxiety in the false positive group, though they did not persist at a 3-month follow-up. Similarly, Clutton et al. [12] found that women who attended a routine breast cancer screening and were notified of the need to attend for further assessment of a mammographically detected abnormality showed low levels of anxiety and psychopathology 6 weeks after the second testing. Clutton et al., however, did not use a control group (e.g., a sample of women who were not recalled). In a recent study, Sandin et al. [19] found no significant differences in hypochondriacal attitudes between women attending a second-stage screening and women attending routine screening, neither just before the screening nor in the follow-up (2 months after the mammogram). Lerman et al. [9] analyzed the psychological side effects of receiving abnormal mammogram results on women’s anxiety and breast cancer worries. The authors assessed psychological variables on a sample of 308 women over 50 by means of telephone surveys conducted 3 months following a screening mammogram. Subjects included
women with normal mammograms (N), nonsuspicious abnormal mammograms (A2), and suspicious abnormal mammograms (A3). Lerman et al. found a positive relationship between the index of suspicion of the mammogram and the levels of anxiety and worry. They observed consistent upward trends in negative psychological impact as levels of mammogram suspicion increased from N to A2 to A3 and concluded saying that because ‘‘concerns persisted even after these women had received additional tests ruling out cancer suggests than women with abnormal mammograms perceive themselves to be at elevated risk of developing breast cancer in the future’’ (p. 265). So, in contrast to the Ellman et al.’s [18] report, Lerman’s group found that the psychological impact of receiving abnormal mammogram results persists beyond receipt of the negative result (i.e., after knowing that they did not have cancer). In a more recent study, Cockburn et al. [20] reported that women who attended a breast cancer screening and were recalled for a further mammogram exhibited higher levels of psychological dysfunction than women who were not recalled. This psychological impact, which was measured 1 week after obtaining notification that there was no sign of cancer, suggests that at least a short-term psychological side effect associated to the second-stage screening could be present beyond possible adverse psychological effects related with routine screening. In spite of some relevant and suggestive studies, existing research on the psychological impact associated to ‘‘false positive’’ results in breast cancer screening is not convincing. Problems with some of the published reports include the use of low samples of women and cross-sectional methodology. The present study was designed to examine the psychological impact of receiving abnormal mammogram results in women attending the Breast Cancer Detection Programme (BCDP) of Navarra (Spain). Consistent with the existing literature, we predicted that three affectivecognitive measures of psychological distress associated to breast cancer screening, i.e., worry about breast cancer, fear of breast cancer, and perceived vulnerability, would be higher in women recalled for further mammograms than in women attending regular screening. We also predicted greater levels of psychopathological symptoms in the first group than in women with normal mammograms (i.e., women attending routine screening). The study used a short-term longitudinal design with a large sample of women who had attended the breast cancer screening programme.
Method Subjects Participants in this study were women, 45– 65 years old, attending the BCDP of Navarra (Spain) for regular breast cancer screening or for a rescreening (a second-stage mammography). According to these two types of participants, we
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
created two different groups, i.e., a routine screening (RS) group and a second-stage screening (SSS) group. All of the subjects were consecutively recruited. The RS group consisted of 600 women attending for regular screening. These women were not recalled for further mammogram and were randomly assigned to this group from a previous sample of 1600 participants. Two women who were diagnosed as having breast cancer were excluded from the study. The SSS group consisted of 600 women recalled for rescreening. The women of this group received a telephone call asking them to attend again because the results of the previous (routine) mammogram were unsatisfactory, abnormal, or inconclusive. Only three rescreened women were diagnosed of breast cancer and were discharged from the study. About 21% of the final 597 subjects were found to have minor or benign breast problems (nodule, fibroma, cyst, etc.). The final sample was composed of 1195 women (i.e., there were 598 subjects in the RS group and 597 in the SSS group). About 76% of the participants ranged in age from 45 to 55, 59% had lower than a high school education, and 76.5% were married. Average age of the participants was 51.4 years (S.D. = 5.2) for the RS group and 52.6 years (S.D. = 5.4) for the SSS group. There were no significant differences between the two groups in terms of age, education, and marital status. All of the subjects voluntarily agree to participate in the study and no one refused to take part or dropped out. Measures Demographic variables Demographic variables included age, educational level, and marital status. They were assessed during the first interview. Cognitive-affective variables Emotional variables were assessed in a similar way to the method used by Lerman et al. [9]. The following three variables were measured: (a) worry about breast cancer, (b) fear of having breast cancer, and (c) perceived vulnerability (belief of having breast cancer). Women were asked about their current levels of worry (i.e., worry about getting breast cancer), fear (i.e., fear of getting a breast cancer), and belief of having breast cancer or perceived vulnerability (i.e., the extent to which a woman thinks that she could get breast cancer). The subjects rated their levels on a 10-point Likert scale with points labeled none/absent (1), moderate (5 – 6), and a lot/very severe (10). Psychopathological symptoms Psychopathology was measured with the Hopkins Symptom Check List-Revised (SCL-90-R; [21]; Spanish version validated by Gonza´lez de Rivera et al. [22] and Valiente et al. [23]). The SCL-90-R is a 90-item questionnaire, which
305
includes the following nine subscales: Somatization (12 items), Obsessive-compulsive (10 items), Interpersonal sensitivity (9 items), Depression (13 items), Anxiety (10 items), Anger/hostility (6), Phobic anxiety (7), Paranoia (6), and Psychoticism (10 items). Subjects respond to each of 90 statements on a five-point Likert scale ranging in severity from ‘‘not at all’’ (scored as 0) to ‘‘extremely’’ (scored as 4). The score of each subscale was the mean of the corresponding subscale items. The global severity index was the mean of the 90 severity ratings. Internal consistence (a coefficients) for the subscales of the Spanish SCL version has ranged from .70 (Paranoia) to .89 (Depression and Phobic anxiety) [23]. Procedure All of the subjects were selected from women attending the BCDP of Navarra (Spain). Every 2 years, women in this program who are 45 – 65 years old receive a letter inviting them for a free mammogram. The majority of women have normal mammograms (RS women). However, results of the mammogram are inconclusive or abnormal (suspicious abnormal mammogram) for some women (about 10%). These women (SSS women) are invited again (this time by a telephone call) for a new mammogram and must attend the screening 2 days after the referral call (this second-stage screening also includes, in some instances, an ultrasonic imaging procedure). Packets of questionnaires were administered in two testing sessions separated by 2 months. The first session (i.e., premammography condition) was performed just before the mammogram, and the second (i.e., follow-up condition) about 2 months after the mammogram. During these periods, sociodemographics (only in precondition), affective variables, and psychopathology were assessed. In addition, the women of the SSS group were also tested (only for affective variables) 2 days following the mammogram (i.e., postmammography condition). In premammography condition, questionnaires were individually completed in an interview format. In postmammography and follow-up conditions, questionnaires were completed by telephone interviews, which were previously contacted. All questionnaires were completed by a trained female psychologist. Statistical analysis The statistical methods used included: (1) Pearson’s c2 test with the Yates continuity correction for differences between the proportion of women with moderate to severe levels (score 5) on the affective-cognitive variables (i.e., worry, fear, and belief). (2) Two-way group (SSS vs. RS) Conditions (premammography vs. followup) repeated-measures analysis of variance (ANOVA); separate ANOVAs were computed on the psychopathological measures (subscales and global index of the SCL90). (3) Two-way group (moderate/severe vs. none/mild
306
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
concerns) Conditions (premammography vs. follow-up) repeated-measures ANOVA; separate ANOVAs were computed on the anxiety, depression, and global index of the SCL-90. (4) Two sets of stepwise regression analyses for each group of women (SSS and RS groups).
Results Cognitive-affective impact Table 1 shows the proportions of SSS and RS women who reported moderate to severe breast cancer worries, fears, and beliefs before the screening mammogram. The majority of women of both groups reported worry and fear related to breast cancer, but only a low percentage of women reported the belief of having breast cancer in the premammogram condition. The SSS subjects suffered more frequently from worry ( P < .001), fear ( P < .001), and negative beliefs ( P < .001) than did the RS women. However, at follow-up (i.e., 2 months after the mammogram), the percentages of subjects who reported moderate to severe breast cancer concerns (worries, fears, and beliefs) were very low (see Table 2). Again, the numbers of women reporting worry and fear were higher in the SSS group than in the RS group. No differences were found in the frequency of subjects reporting beliefs of having breast cancer during this condition.
Table 2 Numbers and percentages (in parentheses) of subjects who reported breast cancer concerns (worry, fear, and beliefs) during the follow-up period (2 months after the mammogram) for each group Group RS (n = 598)
c2
Worry about breast cancer Moderate to severe 19 (3.2) None to mild 578 (96.8)
5 (0.8) 593 (99.2)
7.2**
Fear of having breast cancer Moderate to severe 19 (3.2) None to mild 578 (96.8)
5 (0.8) 593 (99.2)
7.2**
Belief of having breast cancer Moderate to severe 8 (1.3) None to mild 589 (98.7)
3 (0.5) 595 (99.5)
1.5
Affective response
SSS (n = 597)
SSS = second stage screening; RS = routine screening. ** P < .001.
vs. follow-up). The group factor was not significant for any of the SCL variables. However, there were significant differences between the conditions for all variables ( P < .001): Somatization ( F = 80.7; df = 1,1193), Obsessive-compulsive ( F = 168.6; df = 1,1193), Interpersonal sensitivity ( F = 82.9; df = 1,1193), Depression ( F = 63.5; df = 1,1193), Anxiety ( F = 71.1; df = 1,1193), Anger/hostility ( F = 42.8; df = 1,1193), Phobic anxiety ( F = 60.4; df = 1,1193), Paranoia ( F = 65.1; df = 1,1193), Psychoticism ( F = 19.7; df = 1,1193), and Global severity index ( F = 148;
Psychopathological symptoms As shown in Table 3, mean scores on the SCL variables appear to be similar for both groups of women. Moreover, the scores tend to be lower during the follow-up period. Two-way repeated-measures ANOVAs (Group CondiConditions) were computed to test differences between groups (SSS vs. RS) and conditions (premammography Table 1 Numbers and percentages (in parentheses) of subjects who reported breast cancer concerns (worry, fear, and beliefs) during the premammogram period for each group Group RS (n = 598)
c2
Worry about breast cancer Moderate to severe 514 (86.1) None to mild 83 (13.9)
430 (71.9) 168 (28.1)
35.4***
Fear of having breast cancer Moderate to severe 506 (84.8) None to mild 91 (15.2)
432 (72.2) 166 (27.8)
26.9***
Belief of having breast cancer Moderate to severe 94 (15.7) None to mild 503 (84.3)
16 (2.7) 582 (97.3)
59.5***
Affective response
SSS (n = 597)
SSS = second stage screening; RS = routine screening. *** P < .001.
Table 3 Psychopathological measures (SCL-90) by condition (premammography vs. follow-up) and group category Premammography
Follow-up (2 months)
SCL-90 measures
Mean
S.D.
Mean
S.D.
SSS group (n = 597) Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Anger/hostility Phobic anxiety Paranoia Psychoticism Global severity index
0.49 0.43 0.31 0.43 0.47 0.17 0.23 0.26 0.05 0.35
0.41 0.40 0.25 0.38 0.38 0.25 0.29 0.36 0.13 0.24
0.37 0.20 0.19 0.32 0.30 0.11 0.14 0.18 0.03 0.22
0.37 0.26 0.21 0.37 0.33 0.18 0.23 0.29 0.08 0.21
RS group (n = 598) Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Anger/hostility Phobic anxiety Paranoia Psychoticism Global severity index
0.46 0.46 0.34 0.45 0.41 0.22 0.24 0.24 0.04 0.34
0.38 0.38 0.29 0.37 0.33 0.27 0.27 0.35 0.11 0.32
0.36 0.26 0.20 0.36 0.28 0.15 0.16 0.15 0.02 0.24
0.37 0.30 0.22 0.38 0.30 0.21 0.22 0.29 0.09 0.20
SSS = second stage screening; RS = routine screening.
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
df = 1,1190). No interaction effects were found between the two factors. Relationship between cognitive-affective impact and symptoms of anxiety and depression According to data presented in Tables 1 –3, we wanted to test the hypothesis that worry about breast cancer, fear of breast cancer, and belief of having breast cancer are related to anxiety and depression. Thus, we assessed whether women who reported moderate to severe concerns about breast cancer (worry, fear, or belief) during the premammography condition had higher SCL-90 anxiety and depression scores than women who reported none to mild concerns. Given in Table 4 are the means and S.D. on anxiety, depression, and global severity index (during the premammography and at follow-up) for women who reported moderate to severe vs. none to mild concerns about breast cancer during the premammogram. Repeated-measures ANOVA (Group Conditions) demonstrated (1) significant group differences (moderate/severe vs. none/mild worry) on anxiety ( F = 20.5; df = 1,1193; P < .001), depression ( F = 7.6; df = 1,1193; P < .01), and global severity index ( F = 8.5; df = 1,1190; P < .01); (2) significant group differences (moderate/severe vs. none/mild fear) on anxiety ( F = 18.3; df = 1,1193; P < .001), depression ( F = 9.5; df = 1,1193; P < .01), and global severity index ( F = 11.9; df = 1,1190; P < .01); and (3) significant group Table 4 SCL-90 anxiety and depression scores in women who reported moderate to severe vs. none to mild breast cancer concerns during the premammography period (N = 1195) Premammography
Follow-up (2 months)
Subgroups measure
Mean
S.D.
Mean
S.D.
SCL-90 anxiety Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.46 0.32 0.46 0.32 0.59 0.41
0.36 0.29 0.36 0.30 0.44 0.33
0.30 0.21 0.30 0.22 0.40 0.26
0.32 0.24 0.32 0.26 0.40 0.29
SCL-90 depression Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.45 0.36 0.45 0.35 0.54 0.42
0.34 0.27 0.34 0.26 0.40 0.32
0.38 0.32 0.38 0.32 0.43 0.36
0.37 0.35 0.38 0.34 0.43 0.36
SCL-90 global severity index Moderate to severe worry None to mild worry Moderate to severe fear None to mild fear Moderate to severe belief None to mild belief
0.34 0.29 0.34 0.28 0.41 0.32
0.23 0.21 0.24 0.21 0.31 0.21
0.23 0.18 0.24 0.18 0.28 0.22
0.20 0.17 0.20 0.17 0.26 0.20
307
differences (moderate/severe vs. none/mild belief) on anxiety ( F = 42.2; df = 1,1193; P < .001), depression ( F = 11.5; df = 1,1193; P < .01), and global severity index ( F = 21.4; df = 1,1190; P < .001). There were also significant differences between the conditions (premammography vs. follow-up) for all variables ( P < .001). No interaction effects were found between the factors, except on anxiety (Worry Conditions: F = 3.9, P < .05; Fear Conditions: F = 4.7, P < .05; Belief Conditions: F = 4.5, P < .05). These results suggest that women who reported moderate to severe levels of cognitive-affective concerns about breast cancer had higher SCL-90 anxiety and depression scores than the remainder women. Separate multiple stepwise regression analyses were performed for the two groups of women (SSS and RS) to examine whether the cognitive-affective impact experienced during the premammogram period was able to predict the levels of anxiety and depression reported during the conditions of premammography (Time 1) and follow-up (Time 2). A first set of analyses was conducted to analyze the effects of the cognitive-affective variables on Time 1 anxiety and depression. Results showed a low predictive power and different patterns according to the two groups. For the SSS group, the variables belief and fear of having breast cancer together significantly predicted Time 1 anxiety [adjusted r2=.05; F(2,594) = 30.2, P < .001] and depression [adjusted r2=.04; F(2,594) = 11.4, P < .001]. For the RS group, only worry about breast cancer predicted Time 1 anxiety (adjusted r2=.03; F(1,596) = 14.2, P < .001]. The remainder independent variables did not contribute significantly to the prediction of Time 1 anxiety or depression. A second set of analyses was performed to test the effects of the cognitive-affective variables on Time 2 anxiety and depression. After controlling for Time 1 levels of anxiety or depression, the cognitive-affective variables failed to add significantly to the prediction of Time 2 anxiety or depression. However, Time 1 anxiety was a good predictor of Time 2 anxiety for SSS [adjusted r2=.51; F(1,595) = 600.7, P < .001] as well as for RS [adjusted r2=.42; F(1,596) = 430.8, P < .001] subjects. Similarly, Time 1 depression contributed significantly to the prediction of Time 2 levels of depression for both groups of women [SSS group: adjusted r2=.62; F(1,595) = 969.3, P < .001; RS group: adjusted r2=.50; F(1,596) = 615.1, P < .001].
Discussion In the present study, we examined differences between women attending a second-stage breast cancer screening and women attending routine screening on affective-cognitive distress and psychopathology. We also examined whether the psychological impact is only temporarily related to the period that precedes the mammography or persists beyond
308
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
receipt of the negative result (i.e., being informed that there was no sign of cancer). A major finding of this research is that, according to our prediction, women recalled for further mammograms suffered higher levels of affective-cognitive impact than women attending regular breast cancer screening. These differences were found on three separate measures: specific worry, fear, and thinking about getting breast cancer (perceived vulnerability). Even though both groups of women appear to be emotionally affected by attending the breast cancer screening program, a high proportion of affected women belong to the SSS group. These results are consistent with the study of Lerman et al. [9] who found that subjects with abnormal mammograms reported greater negative impact than subjects with normal mammograms. In addition, these authors indicated a positive relationship between the index of suspicion of mammograms and women’s subsequent self-reports of anxiety and worry. However, in contrast with findings of Lerman et al. [9], only a low percentage of women attending the second-stage screening reported at follow-up (2 months after the mammogram) from moderate to severe levels of worry (3.2%), fear (3.2%), or belief (1.3%) of getting breast cancer. Based on the results of the present study, we may suggest that affective-cognitive distress did not persist following notification of the negative result. Our results are also in line with Cockburn et al. [20] and Sandin et al. [24] who accounted for higher psychological side effects for recalled than for nonrecalled women. Another important finding of our study is that the affective-cognitive concerns detected during the premammogram condition in SSS women disappeared quickly. In fact, only 5.9% of women were worried about getting breast cancer, 5.4% had a fear of getting breast cancer, and 2.7% believed that they had breast cancer in the postmammogram condition. This result suggests that possible emotionalcognitive impact experienced by women who attend complementary mammograms (i.e., SSS women) is a phenomenon limited to the short period of time, which precedes mammography, i.e., from the recall to the notification of the mammogram results. These data contrast with findings of Lerman et al. [9] and are in accordance with research that has failed to find evidence of sustained adverse psychological effects after an initial suspicious result not confirmed by subsequent testing [12,18,24]. Our results do not provide support for the prediction that SSS women would exhibit higher levels of psychopathological symptoms than RS women, neither during the premammogram condition nor during the follow-up. Overall, mean values on SCL-90-R measures are (1) comparable to the general population [22,23], (2) similar for both groups of women (Table 3), and (3) higher before mammogram than that in the follow-up. These findings are in line with results of Ellman et al. [18] who reported no differences in psychiatric morbidity between ‘‘false positive’’ women and those attending for regular breast cancer screening. Our data are also
consistent with results reported by Dean et al. [25], Chen et al. [26], and Sandin et al. [19]. Dean et al. found no differences in psychiatric morbidity between women attending a routine breast cancer screening and a control group (matched nonattenders); attending women were assessed 6 months after the screening. Chen et al. documented that only around 20% of the women who attend a second-stage breast cancer screening showed General Health Questionnaire (GHQ) scores reaching the probable psychiatric morbidity level [17]. Sandin et al. reported no significant differences in hypochondriacal attitudes between women attending regular vs. second-stage breast cancer screening. Likewise, Wardle et al. [27] pointed out that women who attended an ovarian cancer screening did not have elevated scores on psychiatric morbidity 1 year after receiving a ‘‘false positive’’ notification. To our knowledge, this is one of the first published studies that (a) was specifically focused on the psychological impact of receiving abnormal or inconclusive mammogram results and (b) used a longitudinal design and a large sample of women attending a breast cancer screening programme. Despite the presence of a notably emotional-cognitive impact before the mammogram in women attending the second-stage screening, our data do not provide support for the hypothesis that such psychological impact persists after the mammogram is performed (in the present breast screening program, the negative results are provided just after the mammogram). In addition, neither the distress suffered by women attending the second-stage screening nor the impact experienced by women attending the regular screening can be said to persist 2 months following the mammogram. These data suggest that, in order to minimize the psychological impact, it could be very important to minimize the period of time between the recall phone call and the rescreening as well as to inform women about results of the mammogram as soon as possible. Moreover, this study also highlights the significance of reducing the period from the time of recall to the mammogram as much as possible. Such implications are in line with Marteau’s [13] suggestion that ‘‘Uncertainly about test results was associated with more concern about health than receiving a clear positive or negative result’’ (p. 4). The conclusions about the timing of the recall phone calls and time to appointments are legitimate and should be noted by those who organise the administration of the screening. Other possible useful ways to decrease the psychological impact associated with screening programmes include the provision of rational and reassuring information to women (e.g., to inform in the letter or telephone call that in general the number of abnormal mammograms is elevated and that it does not mean a high risk of developing breast cancer). It is also worth noting that the quality of reassurance by the physician who tells the patient that there is no significant finding may have certain impact on the extent of concern retained by the rescreened women. In fact, a benefit of breast screening programmes may be derived from reassurance for those women in whom no evidence of breast cancer is found.
B. Sandin et al. / Journal of Psychosomatic Research 52 (2002) 303–309
There are some possible limitations to this work. First, the use of telephone interviews in postmammography and follow-up conditions could explain, to some extent, differences between the conditions. However, this is not likely because the interviews were conducted by the same psychologist and in a similar way to during the premammogram condition. In addition, there is evidence that telephone interviewing is generally an accurate and valid method, and as compared with mail, it appears to be a better option [28]. Second, in this report, we have not examined personality and other individual characteristics as possible predictors of the psychological impact associated with breast cancer screening. As has been reported in some studies [12,26,29,30], neuroticism and other individual characteristics could moderate the psychological distress of attending breast cancer screening as well as the impact of distress on mammography use. Acknowledgments This research was supported by the Grant No. 2204 resolucio´n 634/96 from the Departamento de Salud del Gobierno de Navarra (Spain). We thank Drs. Nieves Ascunce and Angel Del Moral for their help with this work and their invaluable suggestions and Olga Ferna´ndez Soto for her collaboration with data collection. References [1] Chamberlain J, Palli D. Evidence of the effectiveness of breast cancer screening. In: Gad A, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. pp. 15 – 23. [2] Strax P. Mass screening for control of breast cancer. Cancer 1984;53: 665 – 70. [3] Shapiro S, Venet W, Strax P, Venet L, Roeser R. Selection, follow-up, and analysis in the Health Insurance Plan Study: a randomized trial with breast cancer screening. Natl Cancer Inst Monogr 1985;67:65 – 74. [4] American Cancer Society. Cancer facts and figures — 1998. Atlanta, GA: American Cancer Society, 1998. [5] Howe HL. Repeat mammography among women over 50 years of age. Am J Prev Med 1992;8:182 – 5. [6] Breen N, Kessler L. Trends in cancer screening — United States, 1987 and 1992. Morb Mortal Wkly Rep 1996;45:57 – 61. [7] Gad A, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. [8] Rakowski W, Stoddard AM, Rimer BK, Fox A, Andersen MR, Urban N, Lane DS, Costanza ME. Confirmatory analysis of opinions regarding the pros and cons of mammography. Health Psychol 1997;16:433 – 41. [9] Lerman C, Trock B, Rimer BK, Jepson C, Brody D, Boyce A. Psychological side effects of breast cancer screening. Health Psychol 1991;10:259 – 67. [10] Ascunce N, Del Moral A. Compliance in breast cancer screening: the
[11] [12]
[13] [14] [15]
[16] [17] [18]
[19]
[20] [21] [22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
309
Spanish experience. In: Gad AM, Rosselli del Turco M, editors. Breast cancer screening in Europe. Berlin: Springer-Verlag, 1993. pp. 47 – 53. Wardle J, Pope R. The psychological costs of screening for cancer. J Psychosom Res 1992;36:609 – 24. Clutton S, Pakenham KI, Buckley B. Predictors of emotional wellbeing following a ‘‘false positive’’ breast cancer screening result. Psychol Health 1999;14:263 – 75. Marteau TM. Psychology and screening: narrowing the gap between efficacy and effectiveness. Br J Clin Psychol 1994;33:1 – 10. Warren R. The debate over mass mammography in Britain. Br Med J 1998;297:969 – 72. Sutton S, Saidi G, Bickler G, Hunter J. Does routine screening for breast cancer raise anxiety? Results from a three wave prospective study in England. J Epidemiol Community Health 1995;49:413 – 8. Gram IT, Lund E, Slenker SE. Quality of life following a false positive mammogram. Br J Cancer 1990;62:1018 – 22. Stewart-Brown S, Farmer A. Screening could seriously damage your health. Br Med J 1997;314:533 (Editorial). Ellman R, Angelin N, Christians A, Moss S, Chamberlain J, Maguire P. Psychiatric morbidity associated with screening for breast cancer. Br J Cancer 1989;60:781 – 4. Sandin B, Chorot P, Valiente RM, Lostao L, Santed MA. Actitudes hipocondrı´acas en mujeres antes y despue´s de participar en pruebas complementarias de deteccio´n de ca´ncer de mama. Psiquis 2000;21: 232 – 40. Cockburn J, Staples M, Hurley S, De Luise T. Psychological consequences of screening mammography. J Med Screening 1994;1:7 – 12. Derogatis LR. SCL-90-R administration, scoring and procedures manual: II. Towson, MD: Clinical Psychometric Research, 1983. Gonza´lez de Rivera JL, de las Cuevas C, Garcı´a-Marco R, Monterrey AL, Rodrı´guez-Pulido F, Henry-Benı´tez M. Morbilidad psiquia´trica menor en la poblacio´n general de Tenerife. Psiquis 1990; 11:11 – 22. Valiente RM, Sandin B, Chorot P, Santed MA, Gonza´lez de Rivera JL. Sucesos vitales mayores y estre´s: efectos del cambio por migracio´n asociados a manifestaciones psicopatolo´gicas. Psiquis 1996;17:211 – 30. Sandin B, Chorot P, Valiente RM, Lostao L, Santed MA. Ansiedad anticipatoria asociada a las pruebas complementarias de screening de ca´ncer de mama. Psicothema 2001;13:248 – 54. Dean C, Roberts MM, French K, Robinson S. Psychiatric morbidity after screening for breast cancer. J Epidemiol Community Health 1986;40:71 – 5. Chen CC, David A, Thompson K, Smith C, Lea S, Fahy T. Coping strategies and psychiatric morbidity in women attending breast assessment clinics. J Psychosom Res 1996;40:265 – 70. Wardle J, Pernet A, Collins W, Bourne T. False positive results in ovarian cancer screening. Psychol Health 1994;10:33 – 40. Zapka JG, Bigelow C, Hurley T, Ford LD, Egelhofer J, Cloud WM, Sachsse E. Mammography use among sociodemographically diverse women: the accuracy of self-report. Am J Public Health 1996;86: 1016 – 21. Sandin B, Chorot P, Lostao L, Valiente RM, Santed MA. Predictores psicolo´gicos y sociodemogra´ficos de la ansiedad anticipatoria ante la participacio´n en ‘‘segundas pruebas’’ de deteccio´n de ca´ncer de mama. Rev Psicopatol Psicol Clin 2001;6:17 – 36. Schwartz MD, Taylor KL, Siegel JE, Lamdan RM, Moran K. Distress, personality, and mammography utilization among women with a family history of breast cancer. Health Psychol 1999;18:327 – 32.