Setting and participants
Breast cancer patients were enrolled in this prospective multicenter study between October 2004 and February 2007. Thirteen hospital locations participated in this study, with six offering scalp cooling. If patients in the scalp-cooling hospitals did not chose for scalp cooling, they were not included in this study.
Specialized oncology nurses informed patients about the study. Patients who decided to participate received a set of questionnaires before the start of chemotherapy, 3 weeks after the last cycle of chemotherapy and 6 months after chemotherapy. If a questionnaire was not returned, the patient received a reminder. Approval for this study was obtained from the Medical Ethics Committees of all participating hospitals. All study participants provided written informed consent.
Inclusion and exclusion criteria
Inclusion criteria were treatment for breast cancer with one of the following intravenous chemotherapies: four or six adriamycine (60 mg/m2) and cyclophosphamide (600 mg/m2) treatments; five or six 5-fluorouracil (500 mg/m2), epirubicine (90 mg/m2), and cyclophosphamide (500 mg/m2) treatments; five or six 5-fluorouracil (500 mg/m2), adriamycine (50 mg/m2), and cyclophosphamide (500 mg/m2) treatments; and five or six docetaxel (75 mg/m2), adriamycine (50 mg/m2), and cyclophosphamide (500 mg/m2) treatments. These chemotherapies had to be given in the adjuvant setting in a 21-day cycle. Patients treated with intravenous trastuzumab for a year following chemotherapy were excluded because of the possible influence of long-lasting intensive contact with oncology nurses and other cancer patients on the measures. Patients were excluded if they lacked basic proficiency in Dutch, if they were unable to understand the patient information folder, or if they suffered from alopecia before the onset of chemotherapy.
Measures
Patient and tumor characteristics
The measured patient and tumor characteristics were date of birth, marital status, educational level, type of surgery, and lymph node dissection.
Severity of hair loss
The severity and importance of hair loss was measured by asking patients if they felt the need to wear a wig or other head covering. We defined the success of cooling on the basis of whether the patient reported the use of a wig or head covering. Furthermore, the severity of hair loss was reported by patients on the four-point scale for alopecia of the World Health Organization (WHO) with grade 0 for no hair loss, grade 1 for mild hair loss, grade 2 for pronounced hair loss, and grade 3 for total hair loss [40]. In addition, a visual analog scale (VAS) was applied ranging from 0 for no hair loss to 100 for total baldness. The severity of hair loss was measured 3 weeks after the last cycle of chemotherapy.
Burden of hair loss
A newly developed questionnaire assessed the impact of hair loss. Part of the items were selected from questionnaires with respect to alopecia androgenetica [6, 11, 15]; other items were self-defined based on discussions with female cancer patients. This measure consisted of 40 statements that could be rated on a four-point Likert scale ranging from “not at all” to “very much.” Higher scores indicated higher burden of hair loss. The answers on the four-point Likert scale were divided into two groups, namely, patients who did or did not agree with a particular statement. Patients who did not receive scalp cooling as well as scalp-cooled patients who reported hair loss (defined as WHO score grades 1, 2, or 3) completed the questionnaire 3 weeks and 6 months following chemotherapy.
Burden of scalp cooling
The burden of scalp cooling was evaluated by nine self-defined items which were based on complaints reported by patients who were treated with scalp cooling in the past. The items concerned the psychological burden, physical effects, and the influence of uncertainty about the final result of cooling. Response format was a six-point Likert scale ranging from “not at all” to “a lot.” Higher scores thus indicate a higher burden of scalp cooling. The answers on the six-point Likert scale were divided into two groups, namely, patients who did or did not agree with a particular statement. Furthermore, patients were asked whether or not they had taken pain killers during scalp cooling.
This questionnaire was completed by 51 out of 98 scalp-cooled patients. This was due to the fact that this questionnaire was added to our set of questionnaires at a later stage and only in some of the hospitals which offered scalp cooling. Since patients were asked to fill out this questionnaire after each cooling, we received a total of 153 completed questionnaires.
Wig use
Patients were asked what kind of head covering they used and if they wore the head covering inside the house, only outside the house, or both.
A newly developed questionnaire additionally evaluated the satisfaction with wigs and head coverings and was based on consultations with an expert panel (nurses and patients). This measure consisted of 18 statements rated on a four-point Likert scale ranging from “not at all” to “very much.” Higher scores indicated higher levels of satisfaction with wigs or head coverings. Patients completed the questionnaire 3 weeks and 6 months following chemotherapy.
Hair regrowth
Satisfaction with hair regrowth was measured by a five-item questionnaire addressing the rapidity of regrowth, length and thickness, color, and hair style (curly or straight). The answer categories were rated on a four-point Likert scale ranging from “not at all” to “very much.” Higher scores indicated higher levels of satisfaction with hair regrowth. The answers on the four-point Likert scale were divided into two groups, namely, patients who were or were not satisfied with hair regrowth. The questionnaire was filled out 6 months after completing chemotherapy by patients who reported a score between 1 and 3 on the WHO scale for hair loss.
Body image
Body image was assessed with the revised version of the body image scale (BIS), consisting of ten items rated on a four-point Likert scale ranging from “not at all” to “very much.” The sum score of the ten items ranged from 0 to 30 with higher scores representing increased symptoms or distress with regard to body image. The BIS has high reliability (Cronbach’s alpha 0.93) and validity [18].
Concern over body image was assessed by the Measure of Body Apperception (MBA) questionnaire [5]. Scores were calculated for two subscales, namely, “concern about physical appearance” and “concern about body integrity.” These subscales both consisted of four items and measured personal investment in both aspects of body image, rather than assessing the body image the person currently holds [5]. Two self-defined items were added to the MBA to measure the importance of hair for a person’s body image. Response options of the subscales and self-defined items ranged from “strongly agree” to “strongly disagree.” The scores were rated on a scale from 1 to 5, with higher scores meaning more concerns or higher importance. This questionnaire was completed before chemotherapy and 3 weeks after the completion of chemotherapy.
Statistical analyses
Statistical analyses were performed using SAS (version 9.1 for Windows, SAS institute Inc., Cary, NC, USA). Patient and tumor characteristics were compared between scalp-cooled patients and not scalp-cooled patients by chi-square tests for categorical variables.
From the questionnaire on the burden of hair loss, ten statements were selected that caused the most burden to patients. We reported the number and percentage of questionnaires on which the statements were answered affirmatively. The same procedure was followed for all statements on the questionnaires: “burden of scalp cooling,” “hair regrowth,” and “wig use.”
One week prior to chemotherapy, body image was compared between those who were going to be treated with scalp cooling and those who were not. Three weeks after chemotherapy, body image was compared between patients that were successfully cooled, not successfully cooled, and patients that did not receive scalp cooling. All group comparisons were made with analyses of variance followed by Tukey’s test for multiple group comparison.