European Journal of Oncology Nursing
Volume 14, Issue 4 , Pages 263-270, September 2010

A longitudinal study about the body image and psychosocial adjustment of breast cancer patients during the course of the disease

  • Helena Moreira

      Affiliations

    • Research line Relações, Desenvolvimento e Saúde, Faculty of Psychology and Educational Sciences, University of Coimbra, Rua do Colégio Novo, Apartado 6153, 3001-802 Coimbra, Portugal
    • Psychological Intervention Unit (UnIP) of Coimbra University Hospitals, Dr. Daniel de Matos Maternity, Rua Miguel Torga, 3030-165 Coimbra, Portugal
    • Corresponding Author InformationCorresponding author. Linha de Investigação Relações, Desenvolvimento e Saúde, Instituto de Psicologia Cognitiva e Desenvolvimento Vocacional e Social (IPCDVS), Faculdade de Psicologia e Ciências da Educação, Universidade de Coimbra, Rua do Colégio Novo, Apartado 6153, 3001-802 Coimbra, Portugal.
    • ,
  • Maria Cristina Canavarro

      Affiliations

    • Research line Relações, Desenvolvimento e Saúde, Faculty of Psychology and Educational Sciences, University of Coimbra, Rua do Colégio Novo, Apartado 6153, 3001-802 Coimbra, Portugal
    • Psychological Intervention Unit (UnIP) of Coimbra University Hospitals, Dr. Daniel de Matos Maternity, Rua Miguel Torga, 3030-165 Coimbra, Portugal

published online 24 May 2010.

Article Outline

Abstract 

Purpose

The research of body image among breast cancer patients is characterized by some limitations, such as the lack of longitudinal studies or the absence of a multidimensional perspective of body image. This study intends to overcome these limitations, by examining the evolution of body image dimensions (investment, emotions and evaluations) from the period of surgery (T1) to 6-months after the treatment’s ending (T2). It also aims to explore the predictors of body image at T2 and, simultaneously, the predictive role of initial body image to psychosocial adjustment at T2.

Methods

A total of 56 breast cancer patients participated in both assessments and completed a battery of instruments that included measures of body image dimensions (appearance investment, self-consciousness of appearance, shame and appearance satisfaction) and psychosocial adjustment (quality of life and emotional distress).

Results

Within the dimensions of body image, only shame increased over time. In general, initial levels of investment predicted subsequent body image dimensions and having a mastectomy done was associated with higher shame and lower appearance satisfaction at T2. Initial body image did not predict later adjustment, with the exception of depression, where appearance investment played a relevant role.

Conclusions

Our findings contributed to the advance of knowledge in this area, providing relevant data about the evolution of body image dimensions, its predictors and its predictive role on psychosocial adjustment among breast cancer patients. This study also suggested some clinical implications that can assist health professionals to implement strategies focused on body image throughout the disease.

Keywords: Breast cancer, Body image, Psychosocial adjustment, Quality of life, Emotional distress, Appearance investment

 

Back to Article Outline

Introduction 

The process of adjustment to the diagnosis and treatment of breast cancer is complex (Brennan, 2001) and influenced by several factors (Wenzel et al., 1999). One of the most distressing aspects of this disease, which seems to play a significant role on a patient’s adjustment are the body image changes (DeFrank et al., 2007, Helms et al., 2007, Hopwood et al., 2001), such as breast amputation or chemotherapy-induced alopecia (Batchelor, 2001), often considered by patients to be more difficult to cope with than other secondary symptoms of treatments (White, 2000).

Although so far there is no comprehensive definition of body image in the field of psycho-oncology (White, 2000) and this concept has been defined in many different ways (Moyer, 1997), the multidimensional perspective that has emerged outside this specific field, in psychology (e.g. Cash and Pruzinsky, 2002), can also be applied to the understanding of cancer experiences. For example, the cognitive-behavioral model of body image among cancer patients developed by White, 2000, White, 2002, incorporates several interrelated dimensions, such as body image schemas, whereby the content determines the degree of appearance investment (that is, the importance placed on appearance) and the subsequent process of appearance-related information, influencing, for example, automatic thoughts, appearance assumptions, evaluations (satisfaction or dissatisfaction with appearance), compensatory behaviors (e.g. avoidance, concealing) and emotions (e.g. shame, self-consciousness). The emphasis given to the appearance investment, an extremely important body image research variable which has so far been neglected, is an important strength of this model.

In the last years, increasing attention has been given to the subject of body image among breast cancer patients. Nevertheless, this research topic has mainly been addressed in the context of broader studies about the patient’s quality of life (QoL), particularly in studies about the differences between surgical procedures (e.g. Ganz et al., 1992, Parker et al., 2007, Yurek et al., 2000). Few studies have examined the role of body image as a predictor of QoL or emotional distress, as most consider it to be a component of QoL and not as a specific factor that can also account for individual differences in breast cancer adjustment. However, despite the absence of these studies, the literature has provided some evidence about the significant role that body image can play in the psychological functioning of breast cancer patients (Carver et al., 1998, Moyer, 1997, White, 2000). For instance, Pikler and Winterowd (2003) showed that those patients who felt better about their bodies had a stronger belief in their ability to cope with the disease and its treatments.

Additionally, research on factors that can influence the body image of breast cancer patients continues to be insufficient (Moyer and Salovey, 1996). In this study, we focused on the role of surgery (mastectomy and conserving surgery), length of time since surgery and appearance investment. The type of surgery is certainly an important factor and various researchers have explored its impact on women’s adjustment, reporting few or no differences between breast conserving surgery and mastectomy, but consistently showing a decline in body image among patients treated with mastectomy (e.g. Curran et al., 1998, Ganz et al., 1992, Hartl et al., 2003, Moreira et al., 2009b, Schain et al., 1994).

Another relevant factor is the length of time since diagnosis or surgery. However, research on this matter is not consistent and the dearth of longitudinal studies does not allow for a deeper understanding of this question. Some studies indicate a better body image soon after diagnosis and a deterioration over time (Bloom et al., 1998, Hartl et al., 2003), others report a significant improvement in body image over time (Arora et al., 2001, Ganz et al., 1992) and others demonstrate a stability on body image along the disease (DeFrank et al., 2007). These inconsistent results may be due to the different instruments that were used, as well as to the different definitions of body image and assessment points that were chosen.

Appearance investment has been less investigated (Cash, 2002, White, 2000), but a few studies (Carver et al., 1998, Moreira et al., 2009a, Petronis et al., 2003) have provided some empirical evidence of its role in the adjustment among breast cancer patients, generally demonstrating that those who place greater importance on their appearance are more vulnerable to a poor adjustment when facing appearance changes resulting from cancer treatment. For instance, Carver et al. (1998) demonstrated that initial investment in appearance predicted emotional distress during the postsurgical year. According to Cash (2002), this concept encompasses two facets: (1) the self-evaluative salience (SES; the importance an individual places on physical appearance for their definition of self-worth and self-concept) and (2) the motivational salience (MS; the individual’s efforts to engage in appearance management behaviors in order to maintain or improve their attractiveness). Some studies, mainly outside the field of psycho-oncology, have demonstrated fundamental differences between these two facets (Cash et al., 2004, Ip and Jarry, 2008, Melnyk et al., 2004, Moreira et al., 2009a, Rudiger et al., 2007), consistently showing an association between MS and better adjustment outcomes and SES and more dysfunctional outcomes. For example, in a study with breast cancer patients (Moreira et al., 2009a), SES was associated with poorer social and psychological QoL and higher levels of depression and of fear of negative evaluations, while MS was associated with a better QoL in those domains and lower levels of depression.

In spite of these findings, research on body image among breast cancer patients has been characterized by several limitations, for instance, the lack of longitudinal studies specifically designed to evaluate its evolution along the course of the disease. Moreover, most researches did not conceptualize body image as a multidimensional construct, only focusing on one of its specific dimension’s and frequently using subscales or a few items of QoL cancer global measures to assess body image, which does not allow for a comprehensive understanding of this construct. Additionally, studies that explore its predictive role on psychosocial adjustment are still insufficient, as many studies have examined body image as a component of QoL and not as a separate element which may also have some influence on adjustment.

This study was conducted with the purpose of overcoming the aforementioned limitations. Therefore, it aims to longitudinally examine the body image of breast cancer patients and its relation to psychosocial adjustment, from the period close to surgery (T1) to 6-months after the treatment’s ending (T2). The criterion used for selecting these points was the phase of the disease (to more accurately evaluate the impact of idiosyncratic events of the initial and recovery phases) and not the regular time intervals that are usually chosen for longitudinal studies, regardless of the phase of the disease (Deshields et al., 2005). Body image was conceptualized as a multidimensional construct (Cash, 2002, White, 2000) and, thus, it was operationalized in terms of appearance satisfaction (evaluative dimension), body shame and self-consciousness of appearance (emotional dimension) and appearance investment (schematic dimension). Psychosocial adjustment was operationalized in terms of QoL and emotional distress, both relevant outcomes in psycho-oncology research.

The first goal of this study was to examine changes across the disease’s trajectory in body image dimensions. A worse body image, that is, higher levels of shame, self-consciousness and dissatisfaction with appearance, was expected at T2. Although literature concerning this subject is inconsistent, there is some evidence that body image issues are not a main source of concern at the initial phases of the disease, becoming more central later on (e.g. Bloom et al., 1998). Considering that appearance investment is a trait-level construct (Cash, 2002), no differences were expected.

The second goal was to examine the predictors of the emotional and evaluative dimensions of body image at T2. Although appearance investment is also a dimension of body image, it has been examined as a predictor variable, because of its structural and schematic nature. Based on past research (e.g. Ganz et al., 1992, Ip and Jarry, 2008), it was expected that having had mastectomy done and presenting higher initial levels of SES would predict poor body image results at T2 and that higher initial levels of MS would predict better results.

Finally, this study also aimed to investigate the relationship between body image dimensions at T1 and adjustment indicators at T2. It was expected that, above and beyond control variables and baseline outcome levels, a better body image at T1 would predict better adjustment outcomes at T2. Also, based on previous studies (e.g. Moreira et al., 2009a), it was hypothesized that higher initial levels of SES would be associated with poor adjustment results and that higher initial levels of MS would be associated with better adjustment results.

Back to Article Outline

Method 

Participants and procedure 

The sample was collected in the Gynaecologic department of Coimbra University Hospitals, EPE (CUH), a main public hospital in the centre region of Portugal. Ethical approval for conducting this study was obtained from the CUH Research Ethics Committee. Women were recruited if they fulfilled the following criteria: (1) primary diagnosis of breast cancer; (2) no execution of neo-adjuvant treatment prior to surgery; (3) no evidence of metastasis or local recurrence of the disease during the study time; (4) no evidence of a current psychiatric disorder; (5) 18 years of age or older.

Data were collected at two time points: following primary surgery (T1), and 6-months after adjuvant treatments had ended (T2). Patients were invited to participate in the study during their hospitalization for primary breast surgery. Those who agreed to participate completed the questionnaires after the surgery (2–4 days), during the hospitalization period (T1). A detailed explanation of research objectives were given to all participants and informed consent was obtained. Patients were contacted again 6-months after completing their adjuvant treatments (T2). About two weeks prior a phone call was made to remind them of this last assessment. The questionnaires were then sent by mail with a postage-paid, pre-addressed envelope that had to be posted back to the researchers within the next 15 days. If after this period the patients had not yet returned the questionnaires, a phone call was made requesting their return.

A total of 87 breast cancer patients were initially contacted and accepted to participate in the study. Of these, 3 were excluded due to incomplete questionnaire forms. Of the remaining 84, 56 participated in both assessment points (participation rate of 66.7%). No patient presented a disease recurrence and/or died within the study time. There were no significant differences between patients who completed the assessment at T2 (n=56) and those who withdrew from the study (n=28) on sociodemographic or clinical characteristics, on initial body image, QoL and emotional distress. The only exception being the level of education, χ2(1, N=84)=6.89, p=.009, as women that discontinued the study proved to have completed fewer years of schooling (only 28.6% had completed high school or more) when compared with those who had participated in the entire study (58.9% had completed high school or more).

The sample’s sociodemographic and clinical characteristics are presented in Table 1. The mean time from T1 to T2 was 10.39 months (standard deviation=2.74; range=6–15). Although the time criterion was specified at 6-months after the conclusion of treatment, the length of time of adjuvant treatment varied somewhat according to each case, depending on the number of cycles of chemotherapy and/or sessions of radiotherapy.

Table 1. Participant’s sociodemographic and clinical characteristics.
Participants, N=56
n (%)
Age
Mean (SD); range52.39 (7.77); 37–68

Marital status
Married/living with someone46 (82.1)
Divorced/single/widowed10 (17.9)

Education
Lower than high school23 (41.1)
High school or more33 (58.9)

Type of cancer
Invasive ductal carcinoma41 (73.2)
Ductal carcinoma in situ9 (16.1)
Invasive lobular carcinoma3 (5.4)
Lobular carcinoma in situ3 (5.4)

Surgery
Conservative33 (58.9)
Mastectomy23 (41.1)

Adjuvant treatment
None10 (17.9)
Chemotherapy5 (9.1)
Radiotherapy20 (35.7)
Chemotherapy and radiotherapy21 (37.3)

Axillary node dissection
Yes14 (25.9)
No40 (74.1)

Time since surgery until 6-months follow-up
Mean (SD); range10.32 (2.88); 6–15

Instruments 

Self-consciousness of appearance 

The Derriford Appearance Scale 24 [DAS24] (Carr et al., 2005), was used to assess the self-consciousness of appearance or more generally, the discomfort and inhibition felt because of appearance. This includes 10 items which are rated from 1 (lowest distress) to 4 (highest distress) and 14 items which are rated from 0 (not applicable) to 4 (highest distress). The final score can range from 10 to 96, with higher scores indicating more appearance discomfort. The preliminary psychometric properties of the Portuguese version (Moreira and Canavarro, unpublished results), as well as the ones of the original version, are both adequate.

Body shame 

The body shame subscale of the Experience of Shame Scale [ESS] (Andrews et al., 2002; Portuguese version translated by: Moreira and Canavarro, 2008) was used. It has four items, rated on a 4-point scale ranging from 1 (nothing) to 4 (very much), and requires the respondent to select the option that best expresses the intensity to which they experienced each item in the last 3 months (e.g. Have you avoided looking at yourself in the mirror?). The total score ranges between 1 and 4, with higher scores indicating higher levels of shame.

Appearance satisfaction 

Appearance satisfaction was assessed through a single item (How satisfied are you with your physical appearance?), developed specifically for this study whereby women were requested to rate their appearance satisfaction on a 10-point scale, from 1 (extremely dissatisfied) to 10 (extremely satisfied). The total score ranges from 1 to 10.

Appearance investment 

The Appearance Schemas Inventory – Revised [ASI-R] (Cash et al., 2004) was used to assess appearance investment. It has 20 items, uses a 5-point scale (1=strongly disagree; 5=strongly agree) and has two factors: (1) the motivational salience (MS) subscale measures the individual’s efforts to be or feel attractive; (2) the self-evaluative salience (SES) subscale assesses the individual’s belief on how their appearance influences their self-worth and self-concept. The total score for each subscale ranges from 1 to 5, with higher scores indicating higher levels of appearance investment. The original version, as does the Portuguese version (Nazaré et al., in press) of ASI-R both present adequate psychometric characteristics.

Quality of Life 

To measure the individual’s subjective perception of QoL, the World Health Organization Quality of Life-bref [WHOQOL-Bref] (WHOQOL-Group, 1998) was used. This instrument was validated for the Portuguese population according to the guidelines of the Whoqol-Group and presented good reliability and validity (Vaz Serra et al., 2006). It is comprised of 26 items providing scores for four domains: Physical (e.g. pain/discomfort), Psychological (e.g. positive feelings), Social Relationships (e.g. social support) and Environment (e.g. physical environment), including a facet of the overall QoL (general QoL and general health). It uses a 5-point scale, with higher scores indicating higher QoL. The environment domain and the overall QoL were not analyzed in this study.

Emotional adjustment 

The Hospital Anxiety and Depression Scale [HADS] (Zigmond and Snaith, 1983) is a 14-item scale measuring current levels of depression and anxiety. The Portuguese version (Pais-Ribeiro et al., 2007) has good psychometric qualities and comprises two subscales: depression and anxiety, both with seven items. It uses a 4-point scale (0–3) and the total score for each subscale ranges from 0 to 21, with higher scores indicating more symptomatology.

Statistical analysis 

All statistical analyses were conducted with SPSS, version 17.0.

Cronbach’s coefficient alpha measured the internal consistency of the instruments. Descriptive statistics were computed for all variables. Differences between participants and non-participants were analyzed by chi-square tests, univariate analysis of variance (ANOVA) and multivariate analysis of variance (MANOVA).

Four repeated-measures MANOVA’s on body image dimensions, investment facets, QoL domains and emotional adjustment were performed. When a multivariate effect was found, we proceeded with univariate ANOVAs, one per dependent variable. To control alpha inflation due to multiple testing, we performed a Bonferroni adjustment to our alpha level and evaluated the univariate F tests against the corrected alpha (Tabachnick and Fidell, 2007).

Pearson’s and point-biserial correlations were computed between study, clinical and sociodemographic variables at T1 and T2, in order to select the appropriate variables for introduction into the regression models. Several hierarchical regression models were developed for body image and adjustment outcomes at T2. If significantly correlated with the outcome, variables were introduced into the regression analyses of body image in the following order: sociodemographic/clinical variables, baseline scores of the dependent variable and initial levels of investment. The order for adjustment regressions analyses was: sociodemographic/clinical variables, baseline scores of the dependent variable, initial levels of body image evaluative and emotional dimensions and initial levels of investment. Multicollinearity was analyzed through Tolerance and Variance Inflation Factor (VIF) statistics and was considered to be present when tolerance<.10 and VIF>10 (Meyers et al., 2006).

Partial eta squared (ηp2) was used as an estimate of the effect size. According to Cohen (1988), values of .01, .06 and .14 were considered as small, medium and large effect sizes, respectively. Post-hoc power calculations were performed for all analyses (G*Power; Faul et al., 2007).

Back to Article Outline

Results 

Body image and psychosocial adjustment: patterns of change over time 

Table 2 presents mean scores and standard deviations on women’s body image and psychosocial adjustment at Times 1 and 2.

Table 2. Body image, QoL and emotional adjustment at T1 and T2.
T1T2Univariate FPartial eta squared
M (SD)M (SD)
Body image
Satisfaction6.08 (1.69)5.71 (2.56)1.36.03
Body shame1.32 (0.56)1.75 (0.79)11.19**.19
Self-consciousness31.54 (10.27)32.48 (13.90)0.28.01

Quality of life
Social76.06 (16.44)71.82 (14.21)5.43*.09
Psychological69.02 (14.85)68.94 (15.48)0.00.00
Physical66.43 (15.30)64.09 (20.44)0.80.02

Emotional distress
Anxiety9.70 (5.28)6.84 (3.98)21.05***.30
Depression5.28 (4.06)4.58 (4.11)1.60.03

Appearance investment
Self-evaluative salience2.92 (0.51)2.95 (0.50)0.23.00
Motivational salience3.32 (0.62)3.37 (0.50)0.48.01

*p<.05, **p<.01, ***p<.001.

Body image dimensions (self-consciousness, shame and appearance satisfaction) were analyzed together in a MANOVA. A significant multivariate effect was found, [Pillai’s Trace=.24, F(3,45)=4.82, p=.005, ηp2=.24]. The following univariate analyses, with an alpha level set at .017, showed that only body shame significantly increased over time, [F(1,47)=11.19, p=.002, ηp2=.19].1

The pattern of change for both facets of appearance investment was also analyzed in a MANOVA. The multivariate effect of time was not significant [Pillai’s Trace=.10, F(2,54)=0.26, p=.775, ηp2=.10], therefore the mean scores on MS and SES facets did not differ significantly over time.

With respect to QoL (psychological, social and physical domains), the MANOVA did not yield a significant multivariate effect [Pillai’s Trace=.10, F(3,52)=1.92, p=.13, ηp2=.10]. Nevertheless, the subsequent univariate analysis revealed a significant effect for social QoL [F(1,54)=5.43, p=.024, ηp2=.09], although superior to the corrected alpha level (.017).

Regarding emotional adjustment (depression and anxiety), a significant multivariate effect was found [Pillai’s Trace=.38, F(2,48)=14.73, p<.001, ηp2=.38]. The univariate analyses, with an alpha level set at .025, revealed that only anxiety decreased over time [F(1,49)=21.05, p<.001, ηp2=.30] (see Table 2).

Considering the sample size, post-hoc power calculations demonstrated that the power was sufficient (.80) to detect medium to large effects in each repeated-measures MANOVA (f=.27–.29, p<.05, power=.80).

Predictors of body image and psychosocial adjustment 

Before conducting the regression analyses, bivariate associations between study variables were explored (see Table 3). Additionally, correlations between sociodemographic/clinical variables and dependent variables at T2 showed significant associations between: age and self-consciousness of appearance (r=−.29, p=.03); education and social QoL (r=.30, p=.03); marital status and self-consciousness of appearance (r=.32, p=.02); surgery and body shame (r=.34, p=.01) and appearance satisfaction (r=−.33, p=.01). These variables were introduced into the regression models when appropriate.

Table 3. Correlations among variables.
Measureα1234567891011121314
1. Self-consciousness_T1.86–
2. Self-consciousness _T2.92.52**–
3. Shame_T1.86.39**.25–
4. Shame_T2.84.36**.82**.20–
5. App. Satisfaction_T1–−.29*−.40**−.28*−.39**–
6. App. Satisfaction_T2–−.42**−.56**−.32*−.52**.51**–
7. SES_T1.75.49**.36**.10.33*−.27*−.14–
8. SES_T2.77.37**.55**.18.57**−.26†−.33*.54**–
9. MS_T1.78.00−.25†.03−.30*.11.32*.39**−.07–
10. MS_T2.70−.15−.33*−.08−.31*.43**.37**−.03−.09.69**–
11. Anxiety_T2.84.35*.67**.19.60**−.41**−.47**.23.52**−.21−.24†–
12. Depression_T2.85.33*.73**.19.67**−.41**−.50**.28*.48**−.28*−.41**.73**–
13. S QoL_T2.60−.24−.44**−.27*−.46**.38**.27*−.26†−.48**.21.32*−.50**−.49**–
14. Ps QoL_T2.88−.36**−.65**−.25−.69**.49**.57**−.20−.51**.38**.50**−.66**−.78**.72**–
15. P QoL_T2.91−.21−.51**−.17−.38**.37**.44**−.09−.34*.23†.32*−.72**−.75**.53**.70**

Note. App. Satisfaction=appearance satisfaction; SES=self-evaluative salience; MS=motivational salience; S QoL=Social QoL; Ps QoL=Psychological QoL; P QoL=Physical QoL.

†p<.10, *p<.05, **p<.01.

Body image 

The regression models for body image dimensions at T2 were significant (see Table 4). No evidence of multicollinearity was detected. Regarding self-consciousness of appearance, the final model accounted for 47% of the overall variance. The first step explained 15% of variance and only marital status was a marginally significant predictor of this emotion (p=.07). The baseline level of self-consciousness explained the greater amount of variance (23%) and initial levels of investment added 9% to the last step, but only MS was predictive of this emotion (p=.01).

Table 4. Hierarchical multiple regression analyses for body image at T2.
Dependent variables T2
Self-consciousnessBody shameAppearance satisfaction
ΔR2Final βΔR2Final βΔR2Final β
Step1.15* .12** .12**
Age −.20 – –
Marital statusa .21† – –
Surgeryb – .35** −.34**

Step 2.23*** .02 .22***
Baseline score .35* .06 .38**

Step 3.09* .28*** .07*
SES_T1 .21 .45*** –
MS_T1 −.32* −.51*** .29*

Total R2.47*** .42*** .41*
Adjusted R2.41*** .37*** .38*
F(df)F(5,45)=7.86*** F(4,48)=8.48*** F(3,51)=11.84***

SES=Self-evaluative salience; MS=motivational salience. The variables marked with a dash were not introduced into the model because they were not significantly correlated with the dependent variable.

†p<.10, *p<.05, **p<.01, ***p<.001.

aMarital status (0=married/living with someone; 1=separated/divorced/widowed).

bSurgery (0=breast conserving surgery; 1=mastectomy).

The regression model for body shame accounted for 42% of total variance, with step one explaining 12% thereof and step three 28%. Baseline levels of shame did not significantly increase the amount of total variance. Having had a mastectomy done predicted higher levels of body shame at T2 (p=.004), as well as initial levels of SES (p<.001); MS (p<.001) predicted lower levels of this emotion.

Finally, the regression model for appearance satisfaction accounted for 41% of total variance. The first step explained 12% of this variance, the baseline scores of satisfaction explained 22% and, finally, initial levels of investment added 7% to the last step. Had breast conserving surgery done (p=.004), presenting higher appearance satisfaction at T1 (p=.002) and higher levels of MS at T1 (p=.016), predicted greater appearance satisfaction at T2.

Psychosocial adjustment 

In the regression models for psychosocial adjustment, no evidence of multicollinearity was detected. The model for depression was significant, explaining 38% of variance. Only baseline levels of depression (p=.01) and initial levels of SES (p=.03) and MS (p=.04) significantly predicted depression at T2. With regard to anxiety, the model accounted for 38% of overall variance, and only the initial levels of this emotion predicted anxiety at T2 (p<.001), explaining 35% of the initial variance.

The regression models for social, psychological and physical QoL were significant, accounting for 47%, 60% and 30% of total variance, respectively. The baseline levels of each domain of QoL were the only significant predictors of QoL at T2, explaining most of the initial variance. Body image dimensions did not significantly predict QoL at the recovery phase (see Table 5).

Table 5. Hierarchical multiple regression analyses for psychosocial adjustment at T2.
Adjustment variables T2
DepressionAnxietySocial QoLPsychological QoLPhysical QoL
ΔR2Final βΔR2Final βΔR2Final βΔR2Final βΔR2Final β
Step1– – .05 – –
Control variablesa .10

Step 2.27*** .35*** .37*** .53*** .22***
Baseline score .43** .51*** .56*** .50*** .42***

Step 3.01 .03*** .04 .04 .05*
Self-consciousness_T1 −.07 .03 – −.14 –
App. Satisfaction_T1 – −.17 .20 .18 .19
Shame_T1 −.04 – .04 – –

Step 4.10* – .01 .03 .03
SES_T1 .36* – −.09 – –
MS_T1 −.31* – – .17† .17

Total R2.38*** .38*** .47*** .60*** .30**
Adjusted R2.30*** .33*** .41*** .56*** .26**
F(df)F(5,40)=4.91*** F(3,43)=8.63*** F(5,45)=8.01*** F(4,46)=17.07*** F(3,50)=7.30***

SES=Self-evaluative salience; MS=Motivational salience. The variables marked with a dash were not introduced into the model because they were not significantly correlated with the dependent variable.

*p<.05, **p<.01, ***p<.001.

aControl variables only included education (0=lower than high school; 1=high school or more) for social QoL.

The regression analyses were rerun with body image predictors measured at T2 to explore concurrent associations. It was observed that self-consciousness was significantly associated with depression (β=.39, p=.04); appearance satisfaction with psychological (β=.29, p=.01) and physical (β=.32, p=.04) QoL; and body shame with physical QoL (β=.42, p=.05). SES was only significantly associated with psychological (β=−.25, p=.02) and social QoL (β=−.31, p=.02).

Back to Article Outline

Discussion 

The present study represents a significant contribution to the knowledge of body image issues among breast cancer patients. Its prospective and longitudinal nature and the adoption of a multidimensional perspective of body image allowed an in-depth examination of the three main goals of this study: analyzing the evolution of body image dimensions during the course of the disease, examining the predictors of emotional and evaluative body image dimensions and examining the predictive role of body image in breast cancer adjustment.

Concerning the first goal, the initial hypothesis of a lower body image at the 6-month follow-up was only partially confirmed, since only body shame increased with time. This result somehow reflects the inconsistency that characterizes literature about this subject and underlines the need of further research. Nevertheless, it suggests that body image issues are probably not a main source of concern at the initial phase of the disease when women are likely to be focusing more on their disease and ways of survival, only to concentrate on this at a later phase (Bloom et al., 1998, Spencer et al., 1999). Additionally, it is important to keep in mind that the first assessment was conducted a few days after surgery, during the hospitalization period, which could be too early to assess changes in body image satisfaction or self-consciousness of appearance. Future research should examine these variables after a longer period of time after surgery and during the treatment phase, when appearance changes are likely to be more salient.

We also analyzed the evolution of QoL domains and emotional adjustment over time. We found a significant decrease in anxiety levels, which is consistent with previous studies (e.g. Schwarz et al., 2008, Vahdaninia et al., 2010) showing that levels of psychological distress are higher at the beginning of the disease and that they tend to diminish over time. Moreover, the fact that these patients were hospitalized for breast surgery when they were initially assessed, may have functioned as a stressful experience, since the hospital was an unknown environment that may have confronted them with the reality of their disease and future treatment. It is also likely that the recent surgery had caused high levels of anxiety related to the success of this procedure and the anticipated changes in body image.

Regarding the predictors of emotional and evaluative dimensions of body image, our findings showed that, in general, initial levels of investment and having a mastectomy done predicted subsequent self-consciousness of appearance, body shame and appearance satisfaction. These analyses controlled initial levels of body image dimensions, which allowed a rigorous test of the impact of initial investment on body image dimensions. As expected, higher initial levels of MS predicted less self-consciousness and body shame and higher satisfaction at T2. This indicates that making efforts to maintain an attractive appearance can function as a protective factor, helping women to feel better with their physical appearance at the reentry phase. The initial levels of SES only predicted appearance satisfaction at T2, revealing that a high investment in appearance as a source of self-worth has a negative impact on appearance satisfaction at the recovery phase, but not on self-consciousness and shame. These results are in accordance with previous studies, although these exist mainly outside the field of psycho-oncology (e.g. Cash et al., 2004). To date, no study has examined the impact of investment on these dimensions of body image among breast cancer patients.

Another main finding of this study was the predictive role of the type of surgery on body image. Specifically, and as expected, having a mastectomy done proved to be associated with higher body shame and less satisfaction with appearance at T2, when compared with breast conserving surgery. These results are in line with our exploratory analyses on the evolution of body image and the effects of surgery, which showed that patients treated with mastectomy were more dissatisfied and felt more ashamed of their appearance than those who underwent conserving surgery. Overall, these results are consistent with the large majority of studies (e.g. Curran et al., 1998, Hartl et al., 2003, Moreira et al., 2009b, Schain et al., 1994) demonstrating that breast conserving surgery is associated with a better body image, because of their less mutilating effects. According to Moyer (1997), the benefits of breast conserving surgery for body image “is already a firmly established finding” (p. 290), although this concept has been conceptualized and measured in diverse ways. Even though it was not an objective of this study, it is also interesting to note that the type of surgery was not associated with any indicator of adjustment, a result that has been repeatedly shown in literature.

Concerning the last main goal of this study, in general, our hypotheses were not confirmed. In fact, initial body image did not predict later adjustment, as expected. The exception being for depression, since higher levels of SES and lower levels of MS predicted a higher level of depression at the recovery phase. As suggested in a previous study (Moreira et al., 2009a), MS functioned as a protective factor and SES as a vulnerability factor. Trying to implement efforts to improve physical attractiveness seems to provide women with a sense of control over their appearance changes, which in turn can lead to lower levels of depression later on. Contrarily, relying on appearance for definition of self-worth and self-esteem seems to leave women more vulnerable, not only to higher body shame but also to higher levels of depression. This negative effect was also found in the Carver et al.’s study (1998), which showed that women who invested more in their appearance reported higher emotional distress before surgery and during the following year.

Nevertheless, when the body image variables measured at T2 were tested in regression models, the SES also showed to be associated with psychological and social QoL in an opposing direction, as observed in the Moreira et al. (2009a) study. Although these associations do not allow for causal deductions, they evidence an important link at the recovery period that should be considered. Similarly, higher levels of satisfaction were associated with better physical and psychological QoL and higher levels of shame with lower physical QoL. These results support, in some way, the hypothesized relationship between body image and adjustment, but not its predictive role.

In summary, this study tried to overcome several limitations that characterize the research in the field of body image among breast cancer patients. Its prospective and longitudinal nature, the conceptualization of body image as a multidimensional construct, the emphasis on appearance investment and the examination of body image, both as a predictor of adjustment and also as an outcome variable, are strengths of this study that contribute to the advance of knowledge in this area. However, some limitations should be indicated. Firstly, the small size of our sample determined that only medium to large effects could be detected, and although typical for psychological studies (Cohen, 1992), this means that smaller effects may have been ignored. Secondly, the internal consistency of the social domain of QoL was below the values considered adequate, determining some caution in the interpretation of the results obtained with this subscale. Thirdly, the behavioral dimension of body image was not assessed, which prevented a complete evaluation of all the dimensions of this multidimensional concept. Future studies should include a measure of appearance-related behaviors. Fourthly, the absence of a baseline (the assessment of all the study variables before surgery) limits our ability to fully understand the evolution of body image, as well as QoL and emotional adjustment, over time. Finally, the first assessment point (2–4 days after surgery) may not have been the ideal time to accurately assess body image predictors of later adjustment. On one hand, most patients had not yet observed the bodily changes resulting from surgery; on the other hand, as they had already had this procedure done, their answers may have been influenced by the anticipation of these changes. Future studies should evaluate body image dimensions not only before surgery (baseline), as well as after a longer period of time after surgery (a week or more).

This study also has important implications for clinical practice. Health care professionals that work with cancer patients, especially nurses due to their regular contact with patients, should carefully consider body image issues during the course of the disease and especially at the initial phase, since appearance investment proved to have an important role in subsequent levels of depression and body image emotions and evaluations. As such, it seems particularly important to stimulate the patients at an early phase to implement behaviors aimed at improving their sense of physical attractiveness. Moreover, as mastectomy showed to have a detrimental effect on body image, it is essential to discuss surgical options with patients, including the possibility of doing reconstructive surgery, when mastectomy is the only possible procedure.

Back to Article Outline

Conflicts of interest statement 

None declared.

Back to Article Outline

References 

  1. Andrews B, Qian M, Valentine J. Predicting depressive symptoms with a new measure of shame: the experience of shame scale. British Journal of Clinical Psychology. 2002;41:29–42
  2. Arora N, Gustafson D, Hawkins R, McTavish F, Cella D, Pingree S, et al. Impact of surgery and chemotherapy on the quality of life of younger women with breast carcinoma: a prospective study. Cancer. 2001;92:1288–1298
  3. Batchelor D. Hair and cancer chemotherapy: consequences and nursing care – a literature review. European Journal of Cancer Care. 2001;10:147–163
  4. Bloom J, Stewart S, Johnston M, Banks P. Intrusiveness of illness and quality of life in young woman with breast cancer. Psycho-Oncology. 1998;7:89–100
  5. Brennan J. Adjustment to cancer – coping or personal transition?. Psycho-Oncology. 2001;10:1–18
  6. Carr T, Moss T, Harris D. The DAS24: a short form of the Derriford Appearance Scale DAS59 to measure individual responses to living with problems of appearance. British Journal of Health Psychology. 2005;10:285–298
  7. Carver C, Pozo-Kaderman C, Price A, Noriega V, Harris S, Derhagopian R, et al. Concern about aspects of body image and adjustment to early stage breast cancer. Psychosomatic Medicine. 1998;60:168–174
  8. Cash T. Cognitive-behavioral perspectives on body image. In:  Cash T,  Pruzinsky T editor. Body Image: A Handbook of Theory, Research, and Clinical Practice. New York: The Guilford Press; 2002;p. 38–46
  9. Cash T, Melnyk S, Hrabosky J. The assessment of body image investment: an extensive revision of the appearance schemas inventory. International Journal of Eating Disorders. 2004;35:305–316
  10. Cash T, Pruzinsky T. Body Image: A Handbook of Theory, Research, and Clinical Practice. New York: Guilford Press; 2002;
  11. Cohen J. Statistical Power Analysis for the Behavioural Sciences. second ed.. Hillsdale: Erlbaum; 1988;
  12. Cohen J. A power prime. Psychological Bulletin. 1992;112(1):155–159
  13. Curran D, van Dongen JP, Aaronson N, Kiebert G, Fentiman IS, Mignolet F, et al. Quality of life of early-stage breast cancer patients treated with radical mastectomy or breast-conserving procedures: results of EORTC trial 10801. European Journal of Cancer. 1998;34(3):307–314
  14. DeFrank J, Mehta C, Stein K, Baker F. Body image dissatisfaction in cancer survivors. Oncology Nursing Forum. 2007;34(3):625–631
  15. Deshields T, Tibbs T, Fan M-Y, Bayer L, Taylor M, Fisher E. Ending treatment: the course of emotional adjustment and quality of life among breast cancer survivors immediately following radiation therapy. Supportive Care Cancer. 2005;13:1018–1026
  16. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods. 2007;39:175–191
  17. Ganz P, Coscarelli A, Lee J, Polinsky M, Tan S. Breast conservation versus mastectomy: is there a difference in psychological adjustment or quality of life in the year after surgery?. Cancer. 1992;69:1729–1738
  18. Hartl K, Janni W, Kastner R, Sommer H, Strobl B, Rack B, et al. Impact of medical and demographic factor on long-term quality of life and body image of breast cancer patients. Annals of Oncology. 2003;14:1064–1071
  19. Helms R, O’Hea E, Corso M. Body image issues in women with breast cancer. Psychology, Health & Medicine. 2007;13(3):313–325
  20. Hopwood P, Fletcher I, Lee A, Al Ghazal S. A body image scale for use with cancer patients. European Journal of Cancer. 2001;37:189–197
  21. Ip K, Jarry J. Investment in body image for self-definition results in greater vulnerability to the thin media than does investment in appearance management. Body Image. 2008;5:59–69
  22. Melnyk S, Cash T, Janda L. Body image ups and downs: Prediction of intra-individual level and variability of women’s daily body image experiences. Body Image. 2004;1:225–235
  23. Meyers L, Gamst G, Guarino A. Applied Multivariate Research: Design and Interpretation. Thousand Oaks: Sage Publications; 2006;
  24. Moreira, H., Canavarro, M.C., unpublished results. The Portuguese version of the Derriford Appearance Scale – 24: psychometric properties in a sample of breast cancer patients.
  25. Moreira M, Silva S, Canavarro MC. The role of appearance investment in the adjustment of women with breast cancer. Psycho-Oncology. 2009;
  26. Moreira M, Silva S, Marques A, Canavarro MC. The Portuguese version of the Body Image Scale (BIS) – psychometric properties in a sample of breast cancer patients. European Journal of Oncology Nursing. 2009;
  27. Moyer A. Psychosocial outcomes of breast-conserving surgery versus mastectomy: a meta-analytic review. Health Psychology. 1997;16(3):284–298
  28. Moyer A, Salovey P. Psychosocial sequelae of breast cancer and its treatment. Annals of Behavioral Medicine. 1996;18(2):110–125
  29. Nazaré, B., Moreira, M., Canavarro, M.C. Uma perspectiva cognitivo-comportamental sobre o investimento esquemático na aparência: Estudos psicométricos do Questionário de Crenças sobre a Aparência (ASI-R) [A cognitive-behavioral perspective on schematic investment in appearance: Psychometric studies of The Beliefs About Appearance Questionnaire (ASI-R)]. Laboratório de Psicologia, in press.
  30. Pais-Ribeiro J, Silva I, Ferreira T, Martins A, Meneses R, Baltar M. Validation study of a Portuguese version of the hospital anxiety and depression scale. Psychology, Health & Medicine. 2007;12(2):225–237
  31. Parker P, Youssef A, Walker S, Basen-Engquist K, Cohen L, Gritz E, et al. Short-term and long-term psychosocial adjustment and quality of life in women undergoing different surgical procedures for breast cancer. Annals of Surgical Oncology. 2007;14(11):3078–3089
  32. Petronis V, Carver C, Antoni M, Weiss S. Investment in body image and psychosocial well-being among women treated for early stage breast cancer: partial replication and extension. Psychology and Health. 2003;18(1):1–13
  33. Pikler V, Winterowd C. Racial and body image differences in coping for women diagnosed with breast cancer. Health Psychology. 2003;22(6):632–637
  34. Rudiger J, Cash T, Roehrig M, Thompson J. Day-to-day body image states: prospective predictors of intraindividual level and variability. Body Image. 2007;4:1–9
  35. Schain W, D’Angelo T, Dunn M, Lichter A, Pierce L. Mastectomy versus conservative surgery and radiation therapy: psychosocial consequences. Cancer. 1994;73:1221–1228
  36. Spencer S, Lehman J, Wynings C, Arena P, Carver C, Antoni M, et al. Concerns about breast cancer and relations to psychosocial well-being in a multiethnic sample of early-stage patients. Health Psychology. 1999;18(2):159–168
  37. Schwarz R, Krauss O, Hockel M, Meyer A, Zenger M. The course of anxiety and depression in patients with breast cancer and gynecological cancer. Breast Care. 2008;3:417–422
  38. Tabachnick B, Fidell L. Using Multivariate Statistics. Boston: Allyn and Bacon; 2007;
  39. Vahdaninia M, Omidvari S, Montazeri A. What do predict anxiety and depression in breast cancer patients? A follow-up study. Social Psychiatry and Psychiatric Epidemiology. 2010;45(3):355–361
  40. Vaz Serra A, Canavarro MC, Simões MR, Pereira M, Quartilho M, Rijo D, et al. Estudos psicométricos do instrumento de avaliação da qualidade de vida da Organização Mundial de Saúde (WHOQOL-Bref) para Português de Portugal (Psychometric studies of the World Health Organization Quality of Life Assessment (WHOQOL-Bref) for Portuguese from Portugal), Psiquiatria Clínica. 2006;27(2):41–49
  41. Wenzel L, Fairclough D, Brady M, Cella D, Garrett K, Kluhsman B, et al. Age-related differences in the quality of life of breast carcinoma patients after treatment. Cancer. 1999;86:1768–1774
  42. White C. Body image dimensions and cancer: a heuristic cognitive behavioural model. Psycho-Oncology. 2000;9:183–192
  43. White C. Body images in oncology. In:  Cash T,  Pruzinsky T editor. Body Image: A Handbook of Theory, Research, and Clinical Practice. New York: The Guilford Press; 2002;p. 379–386
  44. WHOQOL-Group . The World Health Organization Quality of Life Assessment (WHOQOL): Development and general psychometric properties. Social Science & Medicine. 1998;46(12):1569–1585
  45. Yurek D, Farrar W, Anderson B. Breast cancer surgery: comparing surgical groups and determining individual differences in postoperative sexuality and body change stress. Journal of Consulting and Clinical Psychology. 2000;68(4):697–709
  46. Zigmond AP, Snaith RP. The hospital and depression scale. Acta Psychiatrica Scandinavica. 1983;67:361–370
  • 1 This analysis was repeated to examine whether the body image dimensions change differently over time, according to the type of surgery. As such, a mixed MANOVA on body image dimensions was performed with type of surgery (mastectomy, conserving surgery) as the between-subjects factor, and time (T1, T2) as the within-subjects factor. However, the power analysis showed that, with our sample size, we were only able to detect large effects (f=.40, p<.05, power=.80). This analysis was, therefore, merely exploratory. A significant multivariate effect was found for time [Pillai’s Trace=.26, F(3,44)=5.16 p=.004, ηp2=.26] and a marginally significant multivariate effect was found for surgery [Pillai’s Trace=.16, F(3,44)=2.71, p=.057, ηp2=.16]. The multivariate effect for the interaction surgery by time was not significant. The univariate analyses revealed that only shame increased over time and that patients who underwent mastectomy felt more ashamed and were more dissatisfied with their appearance.

PII: S1462-3889(10)00062-1

doi:10.1016/j.ejon.2010.04.001

European Journal of Oncology Nursing
Volume 14, Issue 4 , Pages 263-270, September 2010

Article Tools