2. [16, 20, 21]. Hair loss was also feared because of the anticipation
that it would affect social activities [22]. A change in the percep-
tion of the sense of self and self-esteem has also been reported
[19, 23]. The decrease in sensuality and sexuality was also felt to
be related to hair loss [19].
Overall, research suggests that chemotherapy-induced alope-
cia (CIA) significantly impacts QOL and psychosocial adjust-
ment among breast cancer patients. This side-effect emphasizes
the necessity of systematic evaluation of the presence of asso-
ciated distress and effects on other aspects of women’s lives.
However, to date, there is no disease-specific tool for measuring
distress due to CIA with valid, reliable, and responsive question-
naires on perceptions of hair loss, QOL, psychological distress,
sexuality, or social functioning. The need for such an instrument
was noted in some scalp-cooling intervention studies because
the general QOL surveys administered were not able to ad-
equately capture these important issues. This study aims to
develop and validate the Chemotherapy-induced Alopecia
Distress Scale (CADS) for women with breast cancer.
methods
participants
Study participants were recruited from 1 May to 30 August 2009 at breast
cancer advocacy events held at 16 different hospitals in Korea. Subjects were
eligible to participate in the study, if they were ≥18 years of age, had histolo-
gically confirmed and diagnosed with breast cancer (stage I–III), had no evi-
dence of recurrence, metastasis, and had no evidence of psychological
problems at the time of the survey. We excluded people who were unable to
read Korean.
We approached 457 breast cancer patients; 353 of them (77.2%) agreed
to participate in the study. Among them, people who had a psychological
problem (n = 10), recurrence of the disease (n = 37), or who were stage 4
(n = 1) were excluded from the study, considering that they might experi-
ence different levels of CIA distress or other psychological problems. The
final study population included 305 breast cancer patients (Table 1). For
testing the external validity of the CADS, we conducted a survey using
another prospective study of 428 women with breast cancer at our institu-
tion in 2010. For the validation set, the same inclusion and exclusion cri-
teria were applied, and the final sample for the validation set was 428. All
procedures were approved by the Institutional Review Board of Samsung
Medical Center, Korea, and all participants provided written informed
consent.
instrument development
Before developing a questionnaire to assess CIA distress, an expert group
consisted of two oncologists, three oncology nurses, one primary physician,
and one behavior scientist carried out an extensive literature review, and
confirmed that there was no existing instrument that specifically addressed
CIA. Although there were studies on distress from CIA from the 1990s, pre-
vious studies had limitation to describe details of the distress due to CIA and
its impact on QOL. Moreover, only a few studies included experiences of
Asian or Korean women with breast cancer. Therefore, our study team
decided to conduct a qualitative study to develop survey items; semi-struc-
tured in-depth interviews were conducted with 21 breast cancer patients to
find out their perceptions, attitudes, preparedness, distress, and experience
of CIA. Detailed methods and results of the qualitative study are described
in other paper [20]. All patients thought that appearance was important and
they had negative perceptions about alopecia; however, they were not well
prepared regarding alopecia, and they experienced substantial physical, psy-
chological, and social distress. Lack of information and limited social
support, combined with negative images of cancer, made it more difficult for
patients to overcome the trauma. At the end, it deterred patients’ daily activ-
ities resulting in poor QOL.
Based on the literature review and the qualitative study, 33 items for asses-
sing CIA were prepared. Six of the 33 items submitted to content experts in
the first round were discarded; two additional items were discarded in the
second round. Then, the survey (25 items) was pilot tested with 30 breast
cancer patients who completed the survey and underwent a brief interview
to answer questions similar to those used in the qualitative study. Feedback
from those patients indicated that the survey addressed issues of importance
to breast cancer patients in terms of CIA and that the instructions and
Table 1. Characteristics of participants (n = 305)
Characteristics N (%)
Age
Mean (SD), range 49.3 (7.9), 28–72
Marital status
Married 256 (84.2)
Single/divorced/separated, widowed 48 (15.8)
Education
Less than middle school 57 (18.9)
High school 180 (59.6)
More than college 65 (21.5)
Working status
Working 50 (16.5)
Leave of absence/retired 69 (22.8)
Housewife 184 (60.7)
Monthly family income
Less than $2000 93 (32.0)
$2000∼$4000 128 (44.0)
More than $4000 70 (24.1)
Living area
Metropolitan 209 (68.5)
Urban/rural 96 (31.5)
Religion
Christian 87 (28.5)
Catholic 92 (30.2)
Buddhist 44 (14.4)
No religion 76 (24.9)
Others 6 (2.0)
Time from diagnosis (month)
Mean (SD), range 28.6 (36.4), 2.5–263.2
Disease stage at diagnosis
Stage 1 107 (35.3)
Stage 2 136 (44.9)
Stage 3 49 (16.2)
Don’t know 11 (3.6)
Current active treatmenta
Yes 127 (41.6)
No 178 (58.4)
Breast surgery type
Lumpectomy 122 (42.2)
Mastectomy 158 (54.7)
a
Receiving chemotherapy or radiotherapy at the time of the survey.
| Cho et al.
original article Annals of Oncology
atUniversityofCalifornia,SanFranciscoonNovember27,2014http://annonc.oxfordjournals.org/Downloadedfrom
3. wording were clear. Patients also felt comfortable answering all questions
and were glad to have been asked about the issues in the survey. Based on
this pilot test, slight changes were made to improve clarity.
The resulting set of 25 items in five domains (physical (2), emotional (9),
daily activity (6), relationship (6), and treatment (2)) was administered to
the study participants. Respondents were instructed to indicate on a four-
point Likert scale on each statement (1 = not at all, 2 = a little, 3 = quite a bit,
and 4 = very much). Total scores were calculated by summing responses for
all items; higher scores meant more distress due to CIA.
In addition to the CADS, we assessed patients’ body image using EORTC
BR-23, QOL with EORTC C-30, and self-esteem using the Rosenberg
Self-Esteem Scale (RSE) to evaluate concurrent and discriminant validity
[24, 25]. Sociodemographics and clinical characteristics—age, marital status,
education, monthly family income, date of diagnosis, stage, current treat-
ment, and surgery type—were also asked of study participants.
statistical analysis
Descriptive statistics were used for reporting characteristics of participants
and mean and standard deviation (SD) of each item of the CADS. To evalu-
ate construct validity, exploratory factor analysis was carried out to deter-
mine the underlying structure of the CADS. A common factor model with
alpha factor extraction was used (Ferketich and Muller, 1990). Alpha extrac-
tion generates its factors by finding the item groupings with maximum in-
ternal consistency, which makes it an appropriate choice for instrument
development (Gorusch, 1983).
After extracting factor structure, we carried out confirmatory factor ana-
lysis (CFA) using the maximum likelihood with missing values to test
whether our factor structure fit the data. Several goodness-of-fit indices were
used to evaluate the model fit, including the goodness-of-fit index (GFI),
comparative-fit-index(CFI), and standardized root mean-squared residual
(SRMR). A GFI and CFI >0.9 and SRMR <0.08 indicate a good fit to the data
[26, 27]. Factor loadings were categorized as low (<0.30), midrange (0.30–
0.59), and high (≥0.60) [28].
With regard to concurrent and discriminant validity, we expect CADS
domain and total scores to at least moderately correlate with the body image
(0.30 ≤ r ≥ 0.70) and also weaker correlations with the patient’s overall QOL
in the past week, and expected non-significant correlations with self-esteem.
To test internal consistency and reliability of the CADS, we calculated
Cronbach’s alpha on both derivation and validation data. We expected a
value >0.70, which is the standard for defining acceptable reliability of an in-
strument. The significant level was at P < 0.05 (two-sided), and all statistical
analyses were done using STATA software package 12.1 (STATA Corp., 4905
Lakeway Drive, College Station, TX).
results
study participants
The mean age (SD) of participants was 49.3 (7.9) years; 84.2%
were married. In terms of education, 59.6% of participants had
high school education, and 21.5% had more than college educa-
tion. Of the total, 16.5% of the participants were currently
Table 2. Factor loadings of the chemotherapy-induced alopecia (CIA) distress items (n = 305)
Items Domain Factor
1 2 3 4
1. The area is itching.a
Physical – – – 0.828
2. The area is burning or prickling resulting pain.a
Physical – – – 0.804
3. I feel different from others.a
Emotional 0.703 0.437 – –
4. I feel uncomfortable and awkward to see myself in mirror.a
Emotional 0.778 0.401 – –
5. I am dissatisfied with my appearance.a
Emotional 0.785 0.384 – –
7. I lose confidence about future.a
Emotional 0.807 – – –
8. I easily irritated and stressed.a
Emotional 0.839 – – –
9. I feel depressed.a
Emotional 0.88 – – –
10. I feel anxious.a
Emotional 0.853 – – –
11. I feel lonely.a
Emotional 0.772 – – –
13. I have difficulties to do personal care such as bath and make-ups.a
Activity 0.391 0.532 – –
14. I have limitations/difficulties to choose clothes.a
Activity 0.36 0.638 – –
15. I experience limitations to do leisure activities.a
Activity 0.404 0.669 – –
16. I feel sicker because of the hair loss.a
Activity 0.456 0.724 – –
17. I do not like that people find that I have cancer because of hair loss.a
Activity 0.394 0.656 – –
18. I have difficulties to concentrate such as working and reading. Activity 0.427 0.44 0.38 –
19. I am worried about that people would avoid me. Relationship 0.438 – 0.584 –
20. I have problems to go out for shopping and restaurants.a
Activity – 0.641 0.372 –
21. I am talking to people about my hair loss. Relationship – 0.468 – 0.483
22. I always wear wigs or scarves to hide hair loss.a
Activity – 0.698 – –
23. I am worried about relationship with family and friends.a
Relationship – – 0.564 –
24. I am worried about relationship with spouse or partner.a
Relationship 0.359 – 0.787 –
25. I am worried about sexual relationship with spouse or partner.a
Relationship – – 0.77 –
26. I hesitate to receive chemotherapy because of alopecia Treatment – – 0.686 –
27. I feel burden of treatment/chemotherapy because of alopecia. Treatment – – 0.62 –
a
This resulted in a CADS with 20 items.
doi:10.1093/annonc/mdt476 |
Annals of Oncology original article
atUniversityofCalifornia,SanFranciscoonNovember27,2014http://annonc.oxfordjournals.org/Downloadedfrom
4. working and rest of them were on leave of absence, retired, or
housewives. There were 35.3% and 44.9% patients who had
stage I and II breast cancer, respectively, and 41.6% of them
were receiving either chemotherapy or radiotherapy at the time
of the survey.
construct validity: factor analysis
The exploratory factor analysis indicated a four-factor solution
with eigenvalue >1.0, while it was initially designed for a five-
factor solution. Two items (item 26 and 27) were initially designed
for ‘treatment’ domain loaded on ‘relationship’ domain (Table 2).
Alopecia1
Physical
Emotional
Activity
Relationship
Alopecia2
Alopecia3
Alopecia5
Alopecia7
Alopecia8
Alopecia9
Alopecia11
Alopecia13
Alopecia15
Alopecia16
Alopecia17
Alopecia20
Alopecia22
Alopecia23
Alopecia24
Alopecia25
CFI=0.925
SRMR=0.065
.023
.59
.47
.42
.33
.16
.46
.51
.3
.15
.13
.22
.28
.31
.36
.37
.79
.8
.83
.85
.89
.93
.92
.83
.7
.73
.92
.82
.76
.73
.64
.99
.87
.58
.6
.38
.79
.47
.5
.24e15
e14
e13
e12
e11
e10
e9
e8
e7
e6
e5
e4
e3
e2
e1
e11
e12
Figure 1. Path diagram 17 items of Chemotherapy-induced Alopecia Distress CFA model (n = 305).
| Cho et al.
original article Annals of Oncology
atUniversityofCalifornia,SanFranciscoonNovember27,2014http://annonc.oxfordjournals.org/Downloadedfrom
5. According to the qualitative interview, breast cancer patients were
least likely to concern these issues [20], the study group decided
to move these two items from the tool. In addition, we excluded
three additional items (item 18, 19, and 21) which did not load
significantly on an interpretable factor solution. The EFA resulted
in the CADS with 20 items. The factor loadings for the four
retained and varimax rotated factors are presented in Table 2. The
variance explained by the four-factor solution was 70.8%.
Further examinations of the factor structure of the 20-item
CADS were evaluated using the CFA; however, the model fit was
not good due to correlated errors, and 3 items (items 4, 10, and
14) were excluded considering high content overlap. Then, we
carried out another CFA, and it revealed high loadings (0.64–
0.99) in general. The fit indices for this model were good:
df = 113, χ2
= 441.58, P < 0.001; CFI = 0.925; SRMR = 0.065
(Figure 1). ‘Emotional” and ‘Activity’ highest correlation and
‘Physical’ and ‘Relationship’ showed the lowest correlation
among four domains. The final version of the CADS with 17
items is in supplementary appendix 1, available at Annals of
Oncology online.
Coefficient alpha was recalculated for the total CADS (0.95)
and the four subscales (Table 3). The Cronbach’s alpha coeffi-
cients of the CADS and its subdomains were also very high,
with the validation set confirming the external validity of the
CADS (Table 3).
convergent and discriminant validity
Table 4 shows the correlations among CADS, body image,
overall QOL, and self-esteem. The total CADS moderately corre-
lated with body image (r = −0.47, P < 0.001), weak correlated
with overall QOL (r = −0.28, P < 0.001), and nonsignificant
correlated with self-esteem (r = −0.07, P = 0.23).
discussion
In this study, CADS is a reliable and valid instrument for meas-
uring distress due to chemotherapy-induced alopecia among
Korean breast cancer patients. The four factors that emerged in
this analysis are reflected in the four subdomains: physical, emo-
tional, activity, and relationship. These subdomains reflect the
findings of the previous qualitative study that evaluated distress
of breast cancer patients because of hair loss during chemother-
apy [20]. The EFA and CFA confirmed the construct validity of
the tool. Concurrent and discriminant validity were demon-
strated by its varying degree of correlations with body image,
overall QOL, and self-esteem.
The CADS was easily administered and completed by breast
cancer patients. Reliability was established by excellent internal
consistency (Cronbach’s alpha coefficient = 0.95 for total; 0.77–
0.95 for subscales) and its validity was confirmed again with the
validation dataset reporting Cronbach’s alpha was 0.95 for total.
The EFA and CFA confirmed our hypothesis regarding the
underlying constructs of the CADS. The scale has four sub-
scales: physical, emotional, activity, and relationship. The
themes of the subscales were consistent with previous identified
problems due to chemotherapy-induced alopecia [20]. The
physical subscale presents physical pains and symptoms that
patients experience due to alopecia and it covered actual pro-
blems caused by alopecia. Emotional, activity, relationships
covered secondary problems caused by hair loss, including pro-
blems in intrapersonal and interpersonal relationship. Although
these distresses would not be directly influenced by alopecia,
they might be more difficult issues for breast cancer patients to
cope with or solve. Therefore, health professionals should con-
sider providing tailored physical and psychosocial support to
improve distress induced by alopecia, including information
provision, active patient empowerment education, and resource
support (e.g. cosmetics, scarves, or wigs).
The finding that the CADS was more strongly associated with
body image than with overall QOL seems appropriate, as the
structure of alopecia distress is conceptually more closely related
to body image than to QOL. Higher chemotherapy-induced dis-
tress would lead to worse body image; however, it is also true
that patients with lower body image would have more alopecia
distress like previous studies reported [4, 7, 29]. The absence of
correlation with self-esteem attests to the discriminant validity
of the CADS. Furthermore, this finding can be regarded as the
advantage of measuring the structure of alopecia distress, as it
indicated that the CADS rating is not influenced by the respon-
dent’s self-esteem.
One limitation must be mentioned: this study was carried out
at a community event, and people who participated in the event
and people who did not participate would be different. However,
the response rate was relatively high and characteristics of the
study population were fairly similar to those of Korean breast
cancer patients in general. In addition, as this was developed
based on Korean breast cancer patients, it would not be able to
include all issues raised by women in other countries. However,
according to the literature, women with breast cancer experience
similar limitations and distress due to CIA, and we believe that
the CADS could also be a reliable measure in other countries.
Finally, we were not able to test known-group comparisons and
Table 3. Reliability of the chemotherapy-induced alopecia distress
scale (CADS)
CADS subscales Cronbach’s alpha coefficient
Derivation set Validation set
Physical 0.77 0.66
Emotional 0.95 0.95
Activity 0.89 0.91
Relationship 0.86 0.90
CADS 0.95 0.95
Table 4. Correlations of CADS with body image, overall QOL, and
self-esteem
Body image Overall QOL Self-esteem
CADS subscales r P R P R P
Physical −0.23 <0.001 −0.15 <0.001 −0.06 0.26
Emotional −0.46 <0.001 −0.30 <0.001 −0.16 0.01
Activity −0.39 <0.001 −0.25 <0.001 −0.07 0.24
Relationship −0.38 <0.001 −0.13 0.02 −0.11 0.06
Total CADS −0.47 <0.001 −0.28 <0.001 −0.07 0.23
doi:10.1093/annonc/mdt476 |
Annals of Oncology original article
atUniversityofCalifornia,SanFranciscoonNovember27,2014http://annonc.oxfordjournals.org/Downloadedfrom
6. test-rested due to data limitation, and further validation studies
in different population and cultures would be necessary.
In conclusion, our study developed a reliable and valid scale
for measuring chemotherapy-induced alopecia distress. This
can be used not only to assess distress due to CIA but also to
test the effectiveness of interventions for CIA management such
as the cooling scarf.
funding
The research was accomplished by the support from
AMOREPACIFIC and Korea Breast Cancer Foundation, No.
PHO1115515.
disclosure
The authors have declared no conflicts of interest.
references
1. Fall-Dickson JM, Rose L. Caring for patients who experience chemotherapy-
induced side effects: the meaning for oncology nurses. Oncol Nurs Forum 1999;
26: 901–907.
2. Carelle N, Piotto E, Bellanger A et al. Changing patient perceptions of the side
effects of cancer chemotherapy. Cancer 2002; 95: 155–163.
3. Montazeri A, Vahdaninia M, Harirchi I et al Quality of life in patients with breast
cancer before and after diagnosis: an eighteen months follow-up study. BMC
Cancer 2008; 8: 330.
4. Baxley KO, Erdman LK, Henry EB et al. Alopecia: effect on cancer patients’ body
image. Cancer Nurs 1984; 7: 499–503.
5. Rosman S. Cancer and stigma: experience of patients with chemotherapy-induced
alopecia. Patient Educ Couns 2004; 52: 333–339.
6. Hunt N, McHale S. The psychological impact of alopecia. BMJ 2005; 331:
951–953.
7. Fobair P, Stewart SL, Chang S et al. Body image and sexual problems in young
women with breast cancer. Psycho-Oncology 2006; 15: 579–594.
8. Cartwright T, Endean N, Porter A. Illness perceptions, coping and quality of life in
patients with alopecia. Br J Dermatol 2009; 160: 1034–1039.
9. Auvinen PK, Mahonen UA, Soininen KM et al. The effectiveness of a scalp cooling
cap in preventing chemotherapy-induced alopecia. Tumori 2010; 96: 271–275.
10. Coates A, Abraham S, Kaye SB et al. On the receiving end—patient perception of
the side-effects of cancer chemotherapy. Eur J Cancer Clin Oncol 1983; 19:
203–208.
11. Lemieux J, Bordeleau L, Goodwin P. Medical and psychological issues in breast
cancer survivors. In Chang AE, Ganz PA, Hayes DF et al. (eds), Oncology, An
Evidence-Based Approach. 1st edn. Springer, 2006; 1836–1858.
12. Kissane DW, Clarke DM, Ikin J et al. Psychological morbidity and quality of life
in Australian women with early-stage breast cancer: a cross-sectional survey.
Med J Aust 1998; 169: 192–196.
13. Williams J, Wood C, Cunningham-Warburton P. A narrative study of
chemotherapy-induced alopecia. Oncol Nurs Forum 1999; 26: 1463–1468.
14. Richer MC, Ezer H. Living in it, living with it, and moving on: dimensions of
meaning during chemotherapy. Oncol Nurs Forum 2002; 29: 113–119.
15. Kissane DW, Grabsch B, Love A et al. Psychiatric disorder in women with early
stage and advanced breast cancer: a comparative analysis. Aust N Z J Psychiatry
2004; 38: 320–326.
16. Luoma ML, Hakamies-Blomqvist L. The meaning of quality of life in patients being
treated for advanced breast cancer: a qualitative study. Psychooncology 2004; 13:
729–739.
17. Boehmke MM, Dickerson SS. Symptom, symptom experiences, and symptom
distress encountered by women with breast cancer undergoing current treatment
modalities. Cancer Nurs 2005; 28: 382–389.
18. Browall M, Gaston-Johansson F, Danielson E. Postmenopausal women with breast
cancer: their experiences of the chemotherapy treatment period. Cancer Nurs
2006; 29: 34–42.
19. Freedman TG. Social and cultural dimensions of hair loss in women treated for
breast cancer. Cancer Nurs 1994; 17: 334–341.
20. Kim IR, Cho J, Choi EK et al. Perception, attitudes, preparedness and experience
of chemotherapy-induced alopecia among breast cancer patients: a qualitative
study. Asian Pac J Cancer Prev 2012; 13: 1383–1388.
21. Maunsell E, Brisson C, Dubois L et al. Work problems after breast cancer: an
exploratory qualitative study. Psychooncology 1999; 8: 467–473.
22. Cowley L, Heyman B, Stanton M et al. How women receiving adjuvant
chemotherapy for breast cancer cope with their treatment: a risk management
perspective. J Adv Nurs 2000; 31: 314–321.
23. Carpenter JS, Brockopp DY. Evaluation of self-esteem of women with cancer
receiving chemotherapy. Oncol Nurs Forum 1994; 21: 751–757.
24. Yun YH, Bae SH, Kang IO et al. Cross-cultural application of the Korean version of
the European Organization for Research and Treatment of Cancer (EORTC) Breast-
Cancer-Specific Quality of Life Questionnaire (EORTC QLQ-BR23). Support Care
Cancer 2004; 12: 441–445.
25. Rosenberg M. Society and the Adolescent Child. Princeton, NJ: Princeton
University Press, 1965.
26. Rubinshtein R, Halon DA, Gaspar T et al. Usefulness of 64-slice cardiac computed
tomographic angiography for diagnosing acute coronary syndromes and predicting
clinical outcome in emergency department patients with chest pain of uncertain
origin. Circulation 2007; 115: 1762–1768.
27. Ximenez C. Recovery of weak factor loadings in confirmatory factor analysis
under conditions of model misspecification. Behav Res Methods 2009; 41:
1038–1052.
28. Kamaruzzaman S, Ploubidis GB, Fletcher A et al. A reliable measure of frailty for a
community dwelling older population. Health Qual Life Outcomes 2010; 8: 123.
29. Lemieux J, Maunsell E, Provencher L. Chemotherapy-induced alopecia and effects
on quality of life among women with breast cancer: a literature review.
Psychooncology 2008; 17: 317–328.
| Cho et al.
original article Annals of Oncology
atUniversityofCalifornia,SanFranciscoonNovember27,2014http://annonc.oxfordjournals.org/Downloadedfrom