Cancer Therapy Vol 3, 461-470, 2005

Adequacy of anticipatory anxiety in women receiving chemotherapy for breast cancer

Research Article

Michael Trimmel^1,*, Christina Semrad¹, Ernst Kubista⁶, Günther Steger⁵, Christoph Zielinski^2,4,5

¹Institute of Environmental Health, Unit for Public Health at the Medical University of Vienna, Austria

²Chair for Medical Experimental Oncology, University Hospital for Internal Medicine I, Medical University of Vienna, Austria

³University Hospital for Internal Medicine I, Medical University of Vienna, Austria

⁴Ludwig Boltzmann Institute for Clinical Experimental Oncology, Vienna, Austria

⁵Clinical Division of Oncology, University Hospital for Internal Medicine I, Medical University of Vienna, Austria

⁶Department of Special Gynecology, Medical University of Vienna, Austria.

__________________________________________________________________________________

*Correspondence: Prof. Michael Trimmel, Institute of Environmental Health, Medical University of Vienna, Kinderspitalgasse 15, A 1095 Vienna, Austria. Tel.: ++43/ 1 4277-64701, Fax: ++43/ 1 4277-9647, E-mail: michael.trimmel@univie.ac.at

Key words: Sufficiency of information, occurrence of symptoms, Source of information, Education, Age, Expectations, Trait anxiety, ACE-19-scores, APT-scores, Anxiety scores,

Abbreviations: Adequacy of Pre-Treatment Anxieties, (APT); Anxiety of 19 Chemotherapeutic Side Effects, (ACE-19)

Received: 15 March 2005; Revised: 26 May 2005

Accepted: 16 August 2005; electronically published: August 2005

Summary

Anxiety of potential side effects prior to cancer chemotherapy as well as retrospective judgements of these assessments as being appropriate or inappropriate were investigated. Forty-two breast cancer patients were asked to complete self-report questionnaires on their first and last day of cytotoxic treatment respectively. Results from 31 patients who completed the study indicate that trait anxiety and expectations of occurrence of side effects contribute to pre-treatment anxieties. In the retrospective view, younger patients were more likely to regard anticipatory anxiety levels as exaggerated. Subjective and objective amount of knowledge and its source were mainly unrelated to prospective and retrospective anxiety appraisals. Post-treatment rating showed that anxiety from physical consequences was highly overestimated at the beginning of the treatment. These findings suggest that patients should be informed as desired to avoid inappropriate and overwhelming anxiety.

I. Introduction

Chemotherapeutic treatment for breast cancer is associated with several short term and long term side effects such as nausea and vomiting, fatigue and impairment of cognitive ability. Patients who face chemotherapy were reported to be emotionally distressed and many experience significant levels of anxiety especially before commencement of chemotherapy (Cella, 1989; Vant Spijker 1997). The patients’ understanding of the illness and treatment can influence their coping ability and adjustment to the treatment regime. Exaggerated levels of anxiety before the beginning of treatment might be maladaptive for the coping process and might even lead to postponed chemotherapy or refusal (Gilbar and De-Nour, 1989). Moreover, excessive anxiety could pose a problem by intensifying side effects (Palmer et al, 1980) or worsening the response to the medication respectively (Walker et al, 1999).

On the other hand, very low anxiety might not be optimal either. According to Janis (1958), such people tend to minimize the impact of the medical intervention and are therefore not adequately motivated to examine and prepare for the treatment in advance. In line with this, Andersen et al, (1984) found out that while gynecologic cancer patients with low pre-treatment anxiety significantly increased in state anxiety during an interactive radiation therapy, whereas the high pre-treatment anxiety group revealed a significant anxiety reduction. After completion of the radiation therapy, the two groups thereby did not differ in state anxiety any longer.

In contrast to this, patients who reported moderate anxiety prior to a medical intervention managed to adapt to the new situation more easily. By showing a medium level of anxiety, “work of worry” is initiated, which encourages the person to attend to information that is required for successful coping (Andersen, 1990).

Adequate preparation of the patients prior to chemotherapy seems crucial to enable patients to correctly anticipate the consequences of the treatment and thereby optimize their coping skills and adjustment process (Burishet al, 1991; Tierney et al, 1991). Although several authors argue that full information can include certain risks for the cancer patients (Simes et al, 1986; Olver et al, 1995), the majority of studies suggest that extensive knowledge benefits individuals undergoing chemotherapy through correcting misbelieves or reducing anxiety of unexpected events.

The present longitudinal study had two main objects. Firstly, to investigate the influence of pretreatment anxiety, expectations and information levels of breast cancer patients on the occurrence and evaluation of chemotherapy related side effects. Secondly, to identify aspects which influence a patients perception of adequacy of pretreatment anxiety levels. With reference to previous research it was expected that the source of information on chemotherapy affected whether pre-treatment anxiety was overestimated or underestimated, or adequately rated. In addition, it was expected that demographic variables, trait anxiety and expectations of side effects (and their occurrence) will affect prospective and retrospective anxiety ratings.

II. Materials and methods

A. Patients

Participants were women who were scheduled to receive neoadjuvant or adjuvant chemotherapy at the general hospital in Vienna. Eligibility criteria were: diagnosis of breast cancer, no previous administration of chemotherapy and fluency in German. Of the 50 patients asked to participate, 42 (84%) consented. Of these, 31 women (74%) could again be contacted after completion of their treatment. Two patients had dropped out of chemotherapy at their own request, one woman had developed metastases, and eight participants could not be attained for administrative reasons. The 31 patients who concluded the whole course of the study did not differ from the remaining 11 in terms of demographic and medical variables and state anxiety (T = ‑.771, df = 40, p = .445) or trait anxiety (T = -2.41, df = 40, p = .811). Demographic characteristics of the sample are presented in Table 1.

B. Procedure

Patients were recruited during the waiting period right before their first infusion. After having signed a consent, the women announced demographic information and completed the

Table 1. Demographic and medical characteristics of participants who entered the study (total) and of those who completed the whole course (sub-sample).

		Total (N = 42)		Sub-sample (N = 31)
		N	%	N	%
Age group	< 30 30-39 40-49 50-59 60-69 > 70	1 7 8 14 8 4	2.4% 16.7% 19.0% 33.3% 19.0% 9.5%	1 4 6 11 6 3	3.2% 12.9% 19.4% 35.5% 19.4% 9.7%
Marital status	single community of life married separated / divorced Widowed	3 3 23 6 7	7.1% 7.1% 54.8% 14.3% 16.7%	2 1 17 6 5	6.5% 3.2% 54.8% 19.4% 16.1%
Highest education	9-year elementary school vocational / business school secondary school university / college	25 4 9 4	59.5% 9.5% 21.4% 9.5%	19 4 4 4	61.3% 12.9% 12.9% 12.9%
Chemotherapy	neoadjuvant Adjuvant	6 36	14.3% 85.7%	3 28	9.7% 90.3%
Regimen	CMF ACMF ATCMF AC FAC	16 2 8 14 2	38.1% 4.8% 19.0% 33.3% 4.8%	11 2 8 10 0	35.5% 6.4% 25.8% 32.3% 0%
Menopausal status	non-menopausal menopausal	16 26	38.1% 61.9%	11 20	35.5% 64.5%

Abbreviations: C = cyclophosphamide; M = methotrexate; F = 5-fluorouracil; A = doxorubicin; T = taxotere.

Spielberger State-Trait Anxiety Scale (STAI) (Laux, Glanzmann, Schaffner and Spielberger, 1981). Moreover, several questionnaires designed for this study were filled out. The first contained a list of 15 physical side effects intended to be rated in regard of the expectations of their occurrence (on a scale from 1 - “I am certain I will not have this” - to 5 - ”I am certain I will have this”). Furthermore, the patients assessed their anxiety of several physical and psychosocial symptoms on a four-point scale by means of the questionnaire “Anxiety of 19 Chemotherapeutic Side Effects” (ACE-19). In addition, the women were asked to check off the respective source of information about the possible side effects on an eight-item list: “physician / nurses”, “informative pamphlet from the hospital”, “other informative pamphlets”, “cancer patients”, “experience from professional life”, “media (for example newspapers, television)”, “I do not know” or “I did not receive any information about this”. Finally, patients were asked to indicate whether they regarded the obtained knowledge to be sufficient, and – if not – what was their favorite source of information in order to make further facts available. Medical information was obtained from the clinical records.

Patients who completed the whole study once again completed the STAI state form on the day of their last cycle of chemotherapy. In addition, the questionnaire “Adequacy of Pre-Treatment Anxieties” (APT), which contained the ACE-19 scores appraised by the women themselves on their first day of treatment, was administered. Subsequently, patients were asked to assess if the pre-treatment anxiety was appropriate, considering their actual experiences with the cytostatic therapy, or whether more or less anxiety would have been appropriate. Finally, the same 15 physical side effects were rated in respect of the subjective probability of occurrence in order to state their incidence during chemotherapy on a scale from 0 (“not at all”) to 4 (“very severe”).

III. Results

A. Clusters of anxiety and appropriateness

In order to identify homogeneous groups of pre-treatment anxieties, a hierarchical cluster analysis of ACE-19 scores (“no anxiety” versus “low”, “moderate” or “high anxiety”) was performed. By using Ward’s method, four congruent clusters were identified: C1) anxiety of loss of independence and self-control (emotional), C2) anxiety of serious physical consequences (physical), C3) anxiety of physical damage with cognitive relevance (cognitive), C4) anxiety of crossing of life plans (existential). The comparison of the average intensities per cluster revealed that physical anxieties (C2) were more prevalent than emotional (p = .050), cognitive (p = .000) or existential ones (p = .001). Furthermore, emotional anxieties (C1) turned out to be more common than existential ones (p = .032) (Figure 1).

An analysis of the mean APT ratings per cluster showed that in this case, physical compared to existential anxieties were more often regarded as exaggerated (p = .013). Figure 2 illustrates the differences between these retrospective assessments.

B. Anxiety scores

Mean STAI scores for state anxiety were 45.48 (SD = 15.05) (sub-sample, N = 31, mean = 46.5, SD = 15.12) prior to the first and 40.81 (SD = 12.51) at the last cycle of chemotherapy, having significantly declined in the course of the cytotoxic treatment (T = 2.447, df = 30, p = .010).

Figure 1. Means and standard errors of means of anxiety scores per cluster (0 = “no anxiety”, 1 = “high anxiety”).

Figure 2. Means and 2 standard errors of rated appropriateness of anxiety (1 = “less would be more appropriate”, 2 = “appropriate”, 3 = “more would be justified”) in the 4 clusters.

In comparison to previous research, these scores were fairly high. For example, Cassileth et al. (1986) found a decrease from an average of 39.5 to 35.5 in breast cancer patients randomized to chemotherapy.

As shown in Table 2, the present study indicates that several potential side effects are associated with considerable distress. Especially possible loss of hair led to high anxiety in about half of the sample. In contrast, most of the non-physical attendant phenomena did not mean distress at all for the majority of women.

In conclusion, final anxiety levels considered as appropriate in retrospective view were computed by combining the intensity of pre-treatment anxieties with the opinions given after the experience. Judgements stating that anticipatory anxiety was not fitting contributed to suitable alternatives in equal parts then. Hence, if “high anxiety” was judged exaggerated in pre-treatment, for instance, one third was added to “no anxiety” as well as to “low anxiety” and “moderate anxiety”. “Appropriate” anxiety levels remained constant. Table 3 demonstrates the results of this transformation.

Accordingly, compared to pre-treatment anxiety levels (Table 2), a high percentage of women (45%) reckoned self-assessed anxiety of hair loss as overstated – although this side effect still caused most distress after all. Besides, 33% of patients thought that less anxiety of vomiting was better-suited, and between one fourth and one third of participants considered anxiety of the infusion, of nausea, pain and of impairment of everyday life exaggerated. On the whole, the majority of breast cancer patients felt their assessments to correspond with experience though. Only anxiety of fatigue was regarded as understated by more than one fifth of women (23%).

C. Associations with ACE-19-scores

1. Trait anxiety

Patients’ scores on the STAI trait form were examined for their associations with anxiety of concomitant phenomena due to chemotherapy. As shown in table 4, anxiety as a person’s attribute proved to be a predictor of pre-treatment anxiety of most of the symptoms (except anxiety of sterility, loss of employment and being a guinea-pig).

2. Expectations

Anticipating the occurrence of physical symptoms turned out to predict anxiety of these likewise. Merely anxiety of hair loss and weakness did not correlate with the subjective possibility of their incidence at the significance level of five per cent (Table 4).

3. Age

Kruskal-Wallis H / Mann-Whitney U tests were performed in order to explore differences in anticipatory anxiety between four age groups (<40, 40-49, 50-59, >59). Only anxiety of impairment of family life was significantly more intense in both groups under 50 years of age than in the older patients (c²= 12.212, df = 3, p = .007). Further analysis revealed that these differences existed solely in women without a partner (single, separated / divorced or widowed) (c²= 9.894, df = 3, p = .006), whereas in cancer patients who were married or lived in a cohabitation, the result again failed significance (c²= 4.689, df = 3, p = .203).

4. Source of information

A comparison between patients who could quote any source for their knowledge and those who did not receive information about the physical or psychological side effects showed that this matter had no relation to the extent of the respective anxiety. Within those women who had been informed by one of the six subjects listed, a Kruskal-Wallis analysis of variance along with a Mann-Whitney U test resulted in just one significant finding: those patients who had obtained their knowledge about

Table 2. ACE-19 scores: Pre-treatment anxiety of attendant symptoms (total and subsample)

		No anxiety	Low anxiety	Moderate anxiety	High anxiety
C#^1	Anxiety of ...	N = 42 (N = 31)	N = 42 (N = 31)	N = 42 (N = 31)	N = 42 (N = 31)
C1	Infusion	36% (27%)	23% (46%)	29% (27%)	13% (0%)
C2	Nausea	13% (20%)	30% (30%)	33% (30%)	23% (20%)
C2	Vomiting	16% ( 9%)	32% (36%)	26% (36%)	26% (18%)
C2	Hair loss	16% (36%)	16% (9%)	19% (0%)	48% (55%)
C2	Injury to immune system	16% (9%)	19% (27%)	45% (36%)	19% (27%)
C1	Weakness	26% (36%)	39% (9%)	32% (46%)	3% (9%)
C1	Fatigue	16% (55%)	55% (0%)	29% (36%)	0% (9%)
C2	Injury to blood counts	13% (18%)	23% (27%)	42% (27%)	23% (27%)
C2	Pain	36% (55%)	32% (9%)	29% (18%)	3% (18%)
C3	Loss of appetite / loss of weight	61% (46%)	26% (27%)	7% (27%)	7% (0%)
C3	Heart problems	45% (46%)	29% (27%)	19% (27%)	7% (0%)
C3	Memory disturbances	55% (64%)	23% (9%)	19% (9%)	3% (18%)
C4	Sterility	82% (60%)	9% (20%)	0% (20%)	9% (0%)
C1	Impairment of everyday life	29% (36%)	23% (18%)	45% (18%)	3% (27%)
C1	Impairment of family life	57% (82%)	13% (9%)	30% (0%)	0% (9%)
C3	Loss of employment	80% (86%)	10% (0%)	5% (14%)	5% (0%)
C1	Being helpless	58% (55%)	13% (36%)	13% (0%)	16% (9%)
C1	Being a guinea-pig	77% (73%)	16% (18%)	3% (9%)	3% (0%)
C2	Ineffectiveness of treatment	48% (27%)	13% (46%)	29% (18%)	10% (9%)

¹Assignment of items to four clusters (C1, C2, C3, C4)

Table 3. Appropriate anxiety of attendant symptoms¹ (N = 31)

C#^1	Anxiety of ...	No anxiety	Low anxiety	Moderate anxiety	High anxiety
C1	Infusion	52%	29%	15%	5%
C2	Nausea	27%	23%	30%	20%
C2	Vomiting	36%	23%	18%	23%
C2	Hair loss	33%	22%	17%	28%
C2	Injury to immune system	23%	21%	31%	24%
C1	Weakness	29%	35%	30%	7%
C1	Fatigue	26%	36%	24%	15%
C2	Injury to blood counts	19%	20%	37%	24%
C2	Pain	56%	25%	15%	4%
C3	Loss of appetite / loss of weight	64%	18%	10%	9%
C3	Heart problems	51%	23%	20%	6%
C3	Memory disturbances	62%	16%	15%	7%
C4	Sterility	82%	9%	0%	9%
C1	Impairment of everyday life	39%	22%	28%	12%
C1	Impairment of family life	60%	18%	16%	6%
C3	Loss of employment	88%	3%	3%	8%
C1	Being helpless	60%	23%	9%	9%
C1	Being a guinea-pig	77%	13%	5%	4%
C2	Ineffectiveness of treatment	43%	17%	33%	8%

¹Computed by combining pre-treatment anxiety levels (ACE-19) and retrospective judgements (APT) (comments see text). ² Assignment of items to four clusters (C1, C2, C3, C4)

Table 4. Results of correlational analysis between self-assessed anxiety and STAI trait score / expectation of side effects

			Trait anxiety	Expectations
C#^2	Anxiety of...		r	r
C1		Infusion	.374^**	^1
C2		Nausea	.510^***	.601^***
C2		Vomiting	.516^***	.612^***
C2		Hair loss	.490^***	.217
C2		Injury to immune system	.337^*	.542^***
C1		Weakness	.392^**	.162
C1		Fatigue	.440^**	.350^*
C2		Injury to blood counts	.379^**	.363^**
C2		Pain	.404^**	.740^***
C3		Loss of appetite / loss of weight	.259^*	.405^**
C3		Heart problems	.383^**	.550^***
C3		Memory disturbances	.437^**	.389^**
C4		Sterility	.210	^1
C1		Impairment of everyday life	.656^***	^1
C1		Impairment of family life	.434^**	^1
C3		Loss of employment	.322	^1
C1		Being helpless	.341^*	^1
C1		Being a guinea-pig	.204	^1
C2		Ineffectiveness of treatment	.318^*	^1

¹Could not be computed because independent variable has not been collected.

²Assignment of items to four clusters (C1, C2, C3, C4)

^*p .05. ^** p .01. ^*** p .001.

nausea from the physician or nurses displayed less intense anxiety of this symptom than those who had been informed by other cancer patients (c²=11.291, df=5, p=0.046).

5. Sufficiency of information

The examination for associations between the request for further information and the intensity of anticipatory anxiety revealed no significant results. This suggested that anxiety levels of patients who were content with the amount of facts given did in no case differ from those of women who announced shortage of information.

D. Associations with APT-scores

1. Trait anxiety

By means of chi-square tests, possible connections between low or high scores on the STAI trait form and the individual APT ratings were studied. The results indicated that anxiety of memory disturbances was more often considered inappropriate by patients high in trait anxiety (² = 9.31, df = 2, p = .010). However, no significant tendency towards over- or understated pre-treatment anxiety assessments was observed.

2. Age

In order to explore differences between the four age groups in considering the extent of anticipatory anxieties to be appropriate or over- / understated, Mantel-Haenszel’s chi-square tests were performed. Thus, the hypothesis (derived from previous investigations) that younger patients are more likely to reckon their anxieties as exaggerated has been confirmed in regard to anxiety of vomiting (c² = 5.35, p = .011), impairment of everyday life (c² = 4.89, p = .014), nausea (c² = 4.16, p = .021), ineffectiveness of treatment (c² = 3.52, p = .031), pain (c² = 3.47, p = .032), impairment of family life (c² = 2.82, p = .047) and weakness (c² = 2.73, p = .050) – items of the clusters C1 (emotional) and C2 (physical) without exception. A partial correlation controlling for the occurrence of the physical symptoms led the associations concerning weakness to miss the one-tailed probability of five per cent (r = .27, p = .074), whereas those respecting nausea, vomiting and pain still remained significant.

3. Education

Having concluded school successfully turned out to have no connection with feeling anticipatory anxieties to correspond with experiences more frequently in retrospective view. Accordingly, women without school-leaving examinations did not prove to be more likely to regard their pre-treatment anxieties as over- or under-stated than graduated ones.

4. Source of information

The results of Yates’ chi-square tests indicated that breast cancer patients who did not receive information about the respective attendant circumstances of chemotherapy did not differ from informed participants at the dichotomized judgements of their anxieties as appropriate or inadequate (i. e. over- or understated). In the sub-sample of women who mentioned any source of information, a statistically significant difference was found only with reference to a potential impairment of everyday life (c² = 9.47, df = 4, p = .050). As Figure 3 illustrates, in that case, other cancer patients were able to prepare for the actual consequences on workaday routine best, whereas the informative pamphlet from the hospital as well as media caused unfitting anxiety levels without exception. - On the other hand, an analysis of the ordinal (three-point) judgements again did not produce significant results, so that no source of information differed from the others in evoking inappropriately low or high anxiety.

5. Sufficiency of information

Being content with the quantity of information given turned out to have no effect on anticipatory anxiety. Anxiety was not related to the source of information about possible side effects either.

6. Expectations and occurrence of symptoms

An exploration of the retrospective anxiety ratings of patients with or without correspondence of anticipation and occurrence of physical side effects revealed that expectations that came true mostly did not result in considering pre-treatment anxiety to be more suitable. Accordingly, only patients who did not suffer from nausea or heart problems regarded their pre-treatment anxieties more often as appropriate if they had not expected these symptoms (Table 5; p = .023 in both cases).

E. Associations with “appropriate” anxiety levels

1. Occurrence of side effects

After having determined the final intensity of the respective anxiety which patients considered appropriate after having undergone chemotherapy (computed as explained before by combining pre-treatment anxiety levels with retrospective judgements), correlation analyses were performed in order to explore associations with the incidence of the respective physical symptom. As shown in Table 6, “appropriate” anxiety of injury to blood counts (p = .120), hair loss (p = .457), injury to immune system (p = .811) and sterility (p = .423) did not statistically significant correlate with the actual occurrence of these side effects – all non-significant items belong to the clusters of physical or existential anxieties. Concerning anxiety of injury to immune system and in particular of sterility, even negative (non-significant) correlational coefficients emerged. As it was hypothesized that for the latter, age acted as a co-variable, a partial correlation was conducted. Although still non-significant, this analysis resulted in a positive Spearman coefficient then (r = .104, p = .403).

IV. Discussion

This study aimed at identifying variables that contribute to an appropriate anticipation of physical and psychosocial symptoms due to cytotoxic treatment. 19 side effects were explored and assigned to one of the following clusters: C1) anxiety of loss of independence and self-control (emotional), C2) anxiety of serious physical consequences (physical), C3) anxiety of physical damage with cognitive relevance (cognitive), C4) anxiety of crossing of life plans (existential). The results indicate that younger breast cancer patients show greater tendencies of experiencing exaggerated pre-treatment anxiety of physical and everyday lifestyle consequences than the elder patients. Since the initial intensities of the mentioned anxieties did not differ from those of the elder women, it is supposed that younger patients are able to cope with these side effects in a more effective way than do people advanced in years. In previous researches it has been

Figure 3. Retrospective ratings of anxiety of impairment of everyday life by source of information

Table 5. Judgements of pre-treatment anxiety levels as being appropriate or not by symptom expectations¹ and actual occurrence of side effects² (N = 31)

Variable	Not experienced			Experienced
		Not appropriate	Appropriate		Not appropriate	Appropriate
Nausea	not expected	0	3	not expected	0	3
	expected	2	0	expected	5	5
Vomiting	not expected	1	4	not expected	1	3
	expected	3	1	expected	0	3
Hair loss	not expected	1	0	not expected	2	4
	expected	2	0	expected	6	6
Injury to immune system	not expected	1	5	not expected	4	8
	expected	1	2	expected	0	4
Weakness	not expected	0	0	not expected	1	7
	expected	1	0	expected	4	5
Fatigue	not expected	0	0	not expected	1	4
	expected	0	0	expected	8	7
Injury to blood counts	not expected	0	4	not expected	3	3
	expected	0	2	expected	2	3
Pain	not expected	4	8	not expected	1	6
	expected	1	0	expected	0	1
Loss of appetite	not expected	0	4	not expected	1	6
	expected	0	1	expected	2	4
Heart problems	not expected	2	12	not expected	1	5
	expected	1	0	expected	0	1
Memory disturbances	not expected	0	10	not expected	6	5
	expected	0	0	expected	2	0

¹ “not expected” = 1 or 2 on expectation scale; “expected” = 4 or 5 on expectation scale

² “not experienced” = 0 on experience scale; “experienced” = 1-4 on experience scale

Table 6. Results of correlational analysis between appropriate anxiety levels¹ and occurrence of side effects

		Occurrence of side effect
#C	Appropriate anxiety of...	r
C2	Nausea	.667^***
C2	Vomiting	.546^**
C2	Hair loss	.144
C2	Injury to immune system	-.045
C1	Weakness	.522^**
C1	Fatigue	.389^*
C2	Injury to blood counts	.290
C2	Pain	.603^***
C3	Loss of appetite / loss of weight	.559^**
C3	Heart problems	.507^**
C3	Memory disturbances	.411^*
C4	Sterility	-.306

¹Computed by combining pre-treatment anxiety levels and retrospective judgements.

^* p .05. ^** p .01. ^*** p .001.

argued that younger patients tend to more pessimistic views by expressing increased perceived vulnerability and dramatizing chemotherapeutic treatment consequences and outcomes (Jacobsen, Bovbjerg and Redd, 1993; Mor, Allen and Malin, 1994). Our results stand in contradiction to this notion, since they suggest that during chemotherapy preparation, younger individuals were more tolerant of the previously mentioned concomitant phenomena than previously described. In light of this, unnecessarily high levels of anxiety of symptoms could perhaps be prevented.

It was also found that pre-treatment anxiety of impairment of family life was overestimated in young women. In that case anxiety turned out to be already more intense prior to the first infusion. Such findings imply that perhaps professionals should pay more attention to family background of young patients, and show ability in identifying high level of anxiety, thereby strengthening confidence in family acceptance.

Trait anxiety was found to predict anxiety levels of most of the side effects. The only exceptions were anxiety of sterility, loss of employment and being a guinea-pig. In agreement with previous investigations (Jacobsen et al, 1993), high trait anxiety could therefore be seen as a characteristic feature that represents propensity to feeling anxious in view of cytotoxic interventions. Since retrospective judgements except those concerning memory disturbances did not differ between patients with low or high scores on the STAI trait form, it can be concluded that for the latter, increased anxiety levels basically commensurate with experiences more truly, so that in this case, successful adaptation to treatment is thereby not impeded. On the other hand, patients high in trait anxiety felt anxiety of memory disturbances to be inadequate more frequently. As no significant trend towards the direction of the inadequacy appeared, these results indicate that high trait anxiety is likely to prevent appropriate anticipation of a cognitive side effect, whereas the remaining symptoms are affected to a smaller extent.

An examination of the relation between expectations and anxiety assessments revealed that anticipation of the majority of side effects entailed heightened distress. Neither Cassileth et al. (1985) nor Jacobsen et al. (1993) could find an association between symptom expectation and anticipatory anxiety. However, in the present investigation, anxiety correlated with patients’ conviction. Previous studies suggest that, while weakness barely caused considerable concerns it led to a lot of distress (Hasenbring et al, 1993; Nerenz et al, 1982). On the other hand, possible hair loss burdened even patients who did not expect to experience this side effect, which indicates that this symptom is subject to substantial anxiety in any case.

An amazing result emerged when retrospective ratings of anxiety levels were analyzed with reference to the point whether speculations about experiencing physical symptoms were fulfilled or not. Contrary to expectations, patients who turned out to comprise correspondence between anticipation and occurrence of secondary effects did hardly ever consider pre-treatment anxiety appropriate more often. With respect to this finding, it is suggested that patients’ outlook is very important. An optimistic approach would lead to low expectations that rather reflect hope than conviction, a phenomenon that had already been observed by Andrykowski and Gregg (1992), whereas adjustment processes in the case of symptom occurrence are not influenced. The possible assumption that one had the fortune to be unaffected by a certain secondary result by way of exception could explain the incoherence between absence of anticipated side effects and anxiety judgements.

According to earlier investigations, it is argued that education influences subsequent ratings of anxiety levels by modifying the understanding and retaining of provided information (Muss et al, 1979; Olver et al, 1995) and, through that, the use of coping mechanisms. In the present study, however, patients’ educational level had no effect on appropriateness of anxiety. A possible explanation is that knowledge does not entail the ability to anticipate appropriate anxiety more accurately. This notion is supported by the comparison of patients who had obtained knowledge about the concomitant symptoms and those who did not receive any information. Although several preceding studies indicate that extensive knowledge facilitates adjustment to cancer treatment, it reduces the discrepancy between what patients anticipate and what actually occurs (Buick, 1997; Langer et al, 1989). No significant differences concerning retrospective judgements of self-assessed anxiety were found in this study and also acquired information to be satisfactory or insufficient did not affect anxiety ratings. While contradictory results have been found in past studies, including both intensification and diminution of anxiety in patients who had been informed completely, the present findings suggest that neither the subjective nor the objective state of knowledge influences anxiety levels. Furthermore, the individual sources of information mostly did not entail significantly different pro- or retrospective anxiety assessments – contrary to expectations which were based upon previous studies that ascribed for instance to the mass media contents which are often outdated or dramatized (Knobf et al, 1998). Applied to clinical practice, this means that medical staff should comply with cancer patients’ demand for detailed information. According to the women’s statements about their favorite source of information, especially the physicians and nurses should undertake this function and, in addition, to some extent also informative pamphlets. Similarly, the request for not receiving any information should also be respected.

Finally, it is intended that with the help of the present study, a contribution is made to improve the understanding for the psychical situation of breast cancer patients, so that their needs could be dealt with to a better extent. Additional research, in particular dealing with coping strategies of people undergoing chemotherapy, could discover supplemental findings that are necessary to sustain quality of life during cytotoxic treatment.

Acknowledgement

The authors want to thank Karin Trimmel for helping to prepare the manuscript.

References

Andersen B (1990) Psychologische Aspekte der Bewältigung einer Krebserkrankung. In U. Koch and F. Potreck-Rose (Eds.) Krebsrehabilitation und Psychoonkologie (pp. 22–31) Berlin: Springer.

Andersen B, Karlsson J, Anderson B and Tewfik H (1984) Anxiety and cancer treatment: response to stressful radiotherapy. Health Psychol 3, 535–551.

Andrykowski M and Gregg M (1992) The role of psychological variables in post-chemotherapy nausea: anxiety and expectation. Psychosom Med 54, 48–58.

Beisecker A, Cook M, Ashworth J, Hayes J, Brecheisen M, Helmig L, Hyland S and Selenke D (1997) Side effects of adjuvant chemotherapy: perceptions of node-negative breast cancer patients. Psychooncology 6, 85–93.

Boakes R, Tarrier N, Barnes B and Tattersall M (1993) Prevalence of anticipatory nausea and other side-effects in cancer patients receiving chemotherapy. Eur J Cancer 29A, 866–870.

Buick D (1997) Illness representations and breast cancer: coping with radiation and chemotherapy. In K. Petrie and J. Weinman (Eds.) Perceptions of health and illness: current research and applications (pp. 379–409) Amsterdam: Overseas Publishers Association.

Burish T, Snyder S and Jenkins R (1991) Preparing patients for cancer chemotherapy: effect of coping preparation and relaxation interventions. J Consult Clin Psychol 59, 518–525.

Cassileth B, Lusk E, Bodenheimer B, Farber J, Jochimsen P and Morrin-Taylor B (1985) Chemotherapeutic toxicity – the relationship between patients’ expectations and posttreatment results. Am J Clin Oncol 8, 419–425.

Cassileth B, Knuiman M, Abeloff M, Falkson G, Ezdinli E and Mehta C (1986) Anxiety levels in patients randomized to adjuvant therapy versus observation for early breast cancer. J Clin Oncol 4, 972–974.

Gilbar O and De-Nour A (1989) Adjustment to illness and dropout of chemotherapy. J Psychosom Res 33, 1–5.

Griffin A, Butow P, Coates A, Childs A, Ellis P, Dunn S and Tattersall M (1996) On the receiving end V: patient perceptions of the side effects of cancer chemotherapy in 1993. Ann Oncol 7, 189–195.

Hasenbring M, Gassmann W, Arp K, Kollenbaum V and Schlegelberger T (1993) Belastungen Ressourcen und Krankheitsverarbeitung im Verlauf einer Polychemotherapie: Ergebnisse einer prospektiven Längsschnittstudie. In F. Muthny and G. Haag (Eds.) Onkologie im psychosozialen Kontext. Spektrum psychoonkologischer Forschung zentrale Ergebnisse und klinische Bedeutung (pp. 99–115) Heidelberg: Asanger.

Holland J and Lesko L (1990) Chemotherapy endocrine therapy and immunotherapy. In J. Holland and J. Rowland (Eds.) Handbook of psychooncology: psychological care of the patient with cancer (pp. 146–162) Oxford: Oxford University Press.

Jacobsen P, Bovbjerg D and Redd W (1993) Anticipatory anxiety in women receiving chemotherapy for breast cancer. Health Psychol 12, 469–475.

Janis I (1958) Psychological stress: psychoanalytic and behavioral studies of surgical patients. New York: Wiley and Sons.

Knobf M. T, Pasacreta J, Valentine A and McCorkle R (1998) Chemotherapy hormonal therapy and immunotherapy. In J. Holland (Ed.) Psychooncology (pp. 277–287) New York: Oxford University Press.

Krohn M and Trams G (1992) Subjektive Belastung einer adjuvanten Chemotherapie bei Mammakarzinom-Patientinnen. Geburtshilfe Frauenheilkd 52, 693–697.

Langer M, Flores-Genger H, Sevelda P, Fiegl J, Ringler M and Kubista E (1989) Nebenwirkungen adjuvanter Therapien beim Mammakarzinom: Ihre Bedeutung für die Patientin. Wien Klin Wochenschr 101, 134–138.

Laux L, Glanzmann P, Schaffner P and Spielberger C (1981) State-Trait-Angst-Inventar (STAI) Weinheim: Beltz.

Love R, Leventhal H, Easterling D and Nerenz D (1989) Side effects and emotional distress during cancer chemotherapy. Cancer 63, 604–612.

Macquart-Moulin G, Viens P, Bouscary ML, Genre D, Resbeut M, Gravis G, Camerlo J, Maraninchi D and Moatti JP (1997) Discordance between physicians' estimations and breast cancer patients' self-assessment of side-effects of chemotherapy: an issue for quality of care. Br J Cancer 76, 1640–1645.

Mor V, Allen S and Malin M (1994) The psychological impact of cancer on older versus younger patients and their families. Cancer 74(Suppl), 2118–2127.

Muss H, White D, Michielutte R, Richards F, Cooper M, Williams S, Stuart J and Spurr C (1979) Written informed consent in patients with breast cancer. Cancer 43, 1549–1556.

Nerenz D, Leventhal H and Love R (1982) Factors contributing to emotional distress during cancer chemotherapy. Cancer 50, 1020–1027.

Olver I, Turrell S, Olszewski N and Willson K (1995) Impact of an information and consent form on patients having chemotherapy. Med J Aust 162, 82–83.

Palmer B, Walsh G, McKinna J and Greening W (1980) Adjuvant chemotherapy for breast cancer: side effects and quality of life. Br Med J 281, 1594–1597.

Simes R, Tattersall M, Coates A, Raghavan D, Solomon H and Smartt H (1986) Randomised comparison of procedures for obtaining informed consent in clinical trials of treatment for cancer. Br Med J (Clin Res Ed) 25, 1065–1068.

Tierney A, Leonard R, Taylor J, Closs S, Chetty U and Rodger A (1991) Side effects expected and experienced by women receiving chemotherapy for breast cancer. BMJ 302, 272.

Walker H, Heys S, Walker M, Ogston K, Miller I, Hutcheon A, Sarkar T, Ah-See A and Eremin O (1999) Psychological factors can predict the response to primary chemotherapy in patients with locally advanced breast cancer. Eur J Cancer 35, 1783–1788.

Warner N, King M, Langlands A, Kenny P and Boyages J (1999) Symptoms 2 weeks 3 months and 12 months after treatment of early breast cancer: the patients' perspectives. Breast 8, 273–277.