Psychological Functioning of Daughters of Breast Cancer Patients Part I: Daughters and Comparison Subjects DAVID
WELLISCH, PH.D., ELLEN
R. GRITZ, PH.D.
WENDY SCHAIN, ED.D., HE-JING WANG, M.D., M.P.H. JESSIE SIAU, M.S.
Sixty daughters ofmothers with breast cancer were matchedfor age. educationalleve/. and race with 60 comparison subjects without a maternal history ofbreast cancer to assess the impact on psychological adjustment. coping. body image. sexual functioning. and health knowledge and practices ofhaving had a mother with breast cancer. Daughters ofhreast cancer patients showed significantly lessfrequent sexual intercourse. lower sexual satisfaction. and greater feelings ofvulnerahility to breast cancer. and they could identify a 1:reater numher ofsymptoms ofhreast cancer. No differences between groups were found in psychological symptoms. coping styles. breast self-examination practices. mammography practices. health knowledge. or body-image ratings. Contrary to clinical studies. women at riskfor hreast cancer showed good overall coping with few signs of significant dysfunctions in relation to comparison suhjects.
ccording to most recent estimates, I in 10 American women will develop breast cancer in their lifetime. I The risk increases with the
Received April 4. 1990: revised September 18. 1990: accepted October II. 1990. From the Depanmenl of Psychiatry and Biobehavior.11 Sciences. and the Depanment of Surgery. School of Medicine. University of California. Los Angeles; the Division of Cancer Control. and the BASE Unit. Jonsson Comprehensive Cancer Center. University of California. Los Angeles; and the Memorial Cancer Institute. Long Beach. California. Address reprint requests 10 Dr. Wellisch. University of California. Los Angeles. School of Medicine. Depanmenl of Psychiatry. 760 Westwood Plaza. Los Angeles. California 90024. Copyright © 1991 The Academy of Psychosomatic Medicine.
presence of a number of factors, such as menopausal status, parity, age at first pregnancy, age at menarche, and family history, with family history being the most well-established factor. 2-4 Numerous studies have shown an approximate twofold increase in risk for women with an affected first-degree relative (mother, sister) compared to women without a family history.2.3.5-9 With both an affected mother and sister, a relative risk of 14 has been observed9; Le. the likelihood of developing breast cancer is 14 times more than that of a daughter who does not have an affected mother and sister. There is also evidence indicating that premenopausal or bilateral breast cancer in relatives poses a substantially higher risk than postmenopausal or unilateral disease.2.3·9.,o PSYCHOSOMATICS
Wellisch et al.
Health Practices Family history of breast cancer has also been studied with regard to its impact upon health care beliefs and attitudes and screening behaviors. High-risk women (those with a family history) compared to low-risk women (those without) acknowledge significantly greater feelings of susceptibility to breast cancer, as well as a fear of performing breast self-examination (BSE), although they are more knowledgeable about the BSE technique. II The two groups do not differ in self-confidence, embarrassment, or frequency of performing BSE. In other studies, perceived vulnerability to breast cancer does not correlate with BSE training and/or practice. 12-15 Mammography While it is generally recommended that women with a family history of breast cancer begin regular mammography screening at a younger age than women not at such risk, there are a modest number of studies to document actual practice. 16 One study of nonparticipation in low-cost mammography showed that a higher percentage of women with a family history was reported among nonparticipants than participants (24% vs. 16%).17 In contrast, two recent studies have shown family history of breast cancer to have definitely affected participation in mammography screening. Family history of breast cancer was among the most influential variables in both of these studies. 18 .19 Psychological Factors Psychologically, a limited number of observations about women with a family history of breast cancer have been reported in clinical studies. Such women are more likely to overestimate, rather than underestimate or correctly assess, breast cancer risk. 20.21 Emotional reactions include I) fear of death or identification with a mutilated body image and unresolved grief/depression, usually in regard to the maternal figure; 2) guilt, secondary to a sense of devoting insufficient time and attention to the relative, or feelings VOLUME 32· NUMBER 3· SUMMER 1991
of responsibility for the cancer; and 3) lowered self-esteem when failing to learn or to carry out effective breast health practices as a result of immobilizing anxiety or guilt.21 Additionally, denial of risk among some family members, displacement of anger to the medical profession regarding latency in diagnosis and treatment outcomes, and unresolved grief as a barrier to screening and preventative behaviors have been described especially for families of breast cancer patients. 22 Female relatives of breast cancer patients demonstrate anxiety-laden identification more frequently than other patient/family groupS.B A study of adolescent daughters of mastectomy patients found no increased acting out of conflicts, but increased psychosomatic problems in these daughters. 24 The general concept of vulnerability appears central to the daughters' experience. Koocher and O'Malley,25 in a long-term follow-up of 120 pediatric cancer patients who were over age 12 at postdiagnosis, plus 173 parents and 101 siblings, coined the clinical term "Damocles syndrome" to describe the lingering experience of threat present for these patients and families. They found a number of variables to be associated with favorable psychological adjustment, which included I) an open style of family communication about the patient's diagnosis, 2) the lack of a "protective" approach toward family members (especially toward children) about details of the illness, and 3) an early informing by the patients about their diagnosis (within 1 year). Although the Koocher and O'Malley study focused on pediatric cancer survivors and their families, it has strong implications for the present study. The purpose of the present study is to systematically assess differences in a group of daughters of breast cancer patients in relation to a well-matched group of comparison subjects in several areas, including knowledge and attitudes about breast cancer, health behaviors, quality of mother-daughter relationship (in light of one group having dealt with the experience of breast cancer), sexuality and body image, and two areas of psychological functioning, including symptomatology and coping behaviors. The daughters
Daughters of Breast Cancer Patients
of breast cancer patients would be expected. in light of the existing literature cited. to be more apt to be screened. to feel greater conflict about self- and other screening procedures. to have greater psychological symptomatology. and to have lower body-image evaluations than other women without such a life experience. However. as will be evident in the demography of this study population for both daughter and comparison groups (see Table I). their levels of income. education. and ethnic representation probably make them a special population. With this in mind. the findings are not likely to be applicable to the population as a whole. METHODS Subjects Sereenin~
Criteria. Screening criteria for daughter subjects included I) having a biological mother with breast cancer. 2) having no personal history of breast cancer, 3) being between the ages of 18 and 65. 4) having completed high school. and 5) being Caucasian. Screening criteria for the comparison group included being between 18 and 65 years old. being Caucasian. being a high school graduate. and having no personal or maternal history of breast cancer. Proceduresfor Recruitin~. Daughters of breast cancer patients were solicited from a single newspaper announcement. which provided a telephone number at UCLA. The project manager explained the study to each caller. If the woman remained interested. she was asked a series of questions to determine her eligibility. We asked how old the woman was when her mother was diagnosed with breast cancer, whether her mother was currently living. and. if not. what had been the cause of her death. These latter questions were not used in determining eligibility. although we did exclude women whose mothers were very ill at the time of the screening call. Our concern was that such women would be too preoccupied to meaningfully participate in. and concentrate on. our interview and testing battery. Women who did not meet the screening criteria were so
informed and thanked. Those who did were told that an interviewer would be contacting them to schedule the interview. Sixty-five women were deemed eligible for the study, of whom 60 were selected to form the daughters' group. Two of these were unable to schedule an interview. and replacements were selected from among the 5 remaining eligible women. Of the 60 daughters of breast cancer patients. 30 were selected whose mothers were living and 30 whose mothers had died. the great majority from breast cancer. The 60 matched comparison subjects were recruited and selected in the following manner. Daughter subjects. during their interviews. were asked to provide the names and telephone numbers offour acquaintances (not best friends) who they thought might be interested in participating in the study and whose mothers had not had breast cancer. The project manager then called each woman. explained the study. and asked if she was interested. Those willing to participate were asked a series of screening questions. They were also asked if their mother was currently living and. if not. the cause of death. If she was living. we asked about her current health. Eligible women were told that if they were selected. an interviewer would contact them. The best match for each daughter subject was selected from among her possible comparison subjects on the basis of age and educational level. Some of the daughter subjects were unable to provide us with the names of women with whom they were well matched. We selected matches for these daughter subjects from the pool of additional comparison subjects. The daughter and comparison subject were interviewed as closely together in time as possible over a 6-month period. A single interviewer conducted both interviews with a matched pair of subjects. Study Procedures for Assessments The study design was cross-sectional; each subject participated in a single structured interview that averaged about 2.5 hours in length. Interviews were all performed in the subject's home by one of two experienced female project interviewers.
Wellisch et al.
TABLE 1. Demographics of 60 daughters of breast cancer patients and 60 comparison subjects Percentage (n) Daughters
Education High school graduale and/or some college College graduale and/or some graduate educarion Graduate degree
26.7 (16) 46.7 (28) 26.7 (16)
26.7 (16) 38.3 (23) 35.0(21)
Marital status Married and living together Never married Widowed and divorced
50(30) 25 (15) 25 (15)
50(30) 20 (12) 30 (18)
Employment status Full time Part lime NO! working
48.3 (29) 26.7 (16) 25.0(15)
61.7(37) 25.0(15) 13.3 (8)
13.3 (8) 26.7 (16) 60.0(36)
6.7(4) 26.7 (16) 66.7 (40)
Religion Protestant Catholic Jewish Other
21.7 (13) 6.7 (4) 51.7(31) 20.0(12)
25.0(15) 6.7 (4) 51.7(31) 16.7(10)
Income $20.000 $20.000-39.999
Race Caucasian Mother's survival status Alive Dead
100(60) 50(30) 50 (30)
81.7(49) 18.3 (II)
Note: The mean age of the daughters was 42.4± 11.0 (SD) years (range 22-63); Ihe mean age of Ihe comparison subjecls was 42.5±10.4 years (range 22-65). ·p
Written informed consent was obtained after the procedure had been fully explained and prior to the interview/testing. The structured interview contained a section of written self-administered psychological testing and several sections of structured oral questions presented to the subject by the interviewer. The interviews for the daughter subjects were constructed to elicit data in each of the following areas: I) general demographics; 2) psychological functioning, including current symptomatology, self-perception of body image, sexual satisfaction (by behaviors), sexual arousal, frequency of sexual intercourse, global sexual satisfaction rating, and ways of coping with stress; 3) developmental and emotional issues, including an evaluation of the quality of the mother-daughter relationship if the mother was VOLUME 32· NUMBER 3· SUMMER 1991
alive when the subject was interviewed; 4) health care attitudes and knowledge, which emphasized physician-directed health care behaviors and most of the objective knowledge questions from the 1979 National Cancer Institute survey "Breast Cancer: A Measure of Progress in Public Understanding,,26; 5) perceptions of the experience of her mother's breast cancer, including one track for daughters whose mothers survived breast cancer as well as one track for daughters whose mother died of breast cancer; and 6) behaviors in health care, which emphasized self-directed and managed health habits and behavior patterns. The interviews for the comparison group were exactly the same, with the exclusion of the section on perceptions of the experience of the mother's breast cancer. 327
Daughters of Breast Cancer Patients
perience), and a Global Sexual Satisfaction Index (measuring sexual satisfaction globally and not behaviorally)2s; 3) the Sexual Arousability Inventory (an inventory for the measurement of female sexual arousability)29; and 4) the Ways of Coping Checklist (coping strategies used to deal with stress).30 These particular measures were selected for four reasons; they provided I) a broad coverage of issues relevant to this population, 2) some useful history of both the daughters of the cancer patients and the females with no family history of cancer, 3) acceptable reliability
Several psychological tests were used in the protocol: I) the Brief Symptom Inventory (BSI), which measures nine psychological symptom dimensions, including somatization, obsessivecompulsiveness, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism (and which yields a Global Symptom Index [GSI])27; 2) the Derogatis Sexual Functioning Inventory with specific sections on body image, including sexual drive (frequency of intercourse), sexual satisfaction (several behavioral facets of the sexual exTABLE 2. Knowledge about breast cancer Factor
NCI 26 (n=I,580)
Percentage (n) Giving Correct Response What is the lifetime incidence for breast cancer? Risk factors for breast cancer? Whitelblack women Under age SO/over age 50 PiIVno pill First child prior to 3O/after 30 Breast feed! did not breast feed Mother or sister with breast cancer/none Estimated percentage of breast lumps proven malignant
35.0 (21) 65.0 (39) 40.0(24) 51.7 (31) 53.3 (32) 93.3 (56)
26.7 (16) 68.3(41) 33.3 (20) 48.3 (29) 41.7(25) 98.3 (59)
15 33 22 30 55 67
Percentage (n) Giving Response No. of symptoms of breast cancer identified' No. of B.C. screening methods identified
No. of breast lump diagnostic techniques identified
Note: NCI 'p=O.025
I 15 (9) >1 85 (51) o 3.3 (2) I 26.7 (16) > I 70 (42) o 1.7 (I) I 28.3(17) >1 70 (42)
35.0(21) 65.0(39) 0.0 (0) 26.7( 16) 73.3 (44) 0.0 (0) 43.3 (26) 56.7 (34)
=National Cancer Institute sample.
TABLE 3. Attitudes toward breast cancer Percentage(n) Factor
Chances of getting breast cancer'
Likely Unlikely Great deal Some lillie None
Amount of personal control in the prevention of breast cancer
(n=60) 79.7 (47) 20.3 (12) 15.0(9) 41.7 (25) 26.7 (16) 16.7 (10)
(n=60) 22.0 (13) 78.0(46) 11.7 (7) 40.0 (24) 30.0 (18) 18.3 (II)
Wellisch et al.
and validity studies, and 4) relatively modest respondent burden. Data Analysis For this matched case control study, the paired t test for continuous variables and McNemar's test for categorical variables 31 were employed to compare the differences between the daughter and comparison groups. Factor analysis was performed by using the principal component method for extracting factors with oblique rotation ofthe factors. 31 - 33
percentage of the daughter group able to identify two or more symptoms of breast cancer than the comparison group (p=O.025). Both groups showed a higher percentage of correct responses on seven of the eight items on which the two groups in the current study could be compared to the NCI sample. Attitudes. In regard to attitudes (Table 3), the daughter group perceived their chances ofgetting breast cancer to be significantly higher than the comparison group (p<0.000 I). No difference was found in the subjects' sense of personal control over the prevention of breast cancer.
RESULTS Health Behaviors Demographics Table I demonstrates the adequacy of matching between the two groups. There were no significant differences in any demographic variable except the mothers' survival status, where the daughter subjects had a significantly higher proportion of deceased mothers than the comparison subjects, which was built into the study by stratification (p~O.()()I). Demographically, this study sample reflects the West Los Angeles area population from which it was drawn: subjects were very well educated (70% or more of both groups were at least college graduates), which was partially determined by screening criteria; they were relatively affluent (at least 60% of both groups had family incomes of $40,000 or more per year); and a majority were Jewish (51.7% of both groups). Knowledge and Attitudes about Breast Cancer Knowledge. The daughter and comparison groups were compared on II knowledge-based items from the National Cancer Institute (NCI) survey "Breast Cancer: A Measure of Progress in Public Understanding...26 Both of these groups were then compared to the original NCI survey sample of 1,580 women (Table 2).26 The daughter and comparison groups were essentially equivalent. Only I of the II knowledge questions showed a significant difference, with a higher VOLUME 32· NUMBER 3· SUMMER 1991
Table 4 presents screening practices and self-managed health behaviors for the two groups, compared with the 1979 NCI sample and other more recent women's health surveys.26.34-36 In regard to screening practices, no significant differences were found between the study groups on mammography, Pap smear, or blood pressure checks. The self-managed behaviors also showed no significant differences between the two study groups. A very high percentage of both groups had done breast self-examination (BSE), with a similar percentage having had either "little or no confidence" in BSE (daughters=30% vs. comparisons=41 %). Frequency of BSE was divided into a tripartite distribution consisting of hypervigilant (too often), appropriate, or suboptimal (too infrequently). The percentage of the study groups in the two dysfunctional (too often or too infrequently) categories was very similar (daughters=59% vs. comparisons=64%). A majority of both groups watched diet and nutrition, considered themselves overweight, exercised regularly, and did not smoke. Evaluation of the MotherDaughter Relationship Four specific questions involved a general evaluation of the mother-daughter relationship for subjects with living mothers, none of which 329
Daughters of Breast Cancer Patients
showed a significant between-group difference. The responses that were given reflected strong positive relationships. The questions were as follows: I. What is the general quality of your relationship with your mother at present? Possible range: I (excellent) to 4 (poor); daughters: 1.86±O.97 (mean±SD) vs. comparisons: 1.8I±O.97. 2. How easy is it for you to talk with mother? TABLE 4.
Possible range: I (very easy) to 5 (extremely difficult); daughters: 2.46±1.25 vs. comparisons: 2.28±1.22. 3. How supportive is your mother of you? Possible range: I (extremely supportive) to 5 (not at all supportive); daughters: 2.20±1.29 vs. comparisons: 1.97±I.I2. 4. How supportive are you of your mother? Possible range: I (extremely supportive) to 5 (not at all supportive); daughters: 2.26±l.OI vs. comparisons: 2.08±O.90.
Health behaviors Percentage (n)
Comparison Subjects (n=60)
75 (45) 25 (15)
63 (38) 37(22)
Have Pap smear regularly?
Have blood pressure checked?
93 (56) 7 (4)
90 (54) 10(6)
85 (51) 15(9)
Much Some Little None
II (6) 59 (32) 21 (10) 9 (6)
22 (II) 37 (19) 24 (12) 17 (9)
Daily-weekly Monthly-bimonthly 3-4 times/yearnot in last year
II (7) 41 (25)
5 (3) 36(22)
43 (26) 15 (9)
Watch diet and nutrition
Feeling about current weight
Underweight Overweight About right
5 (3) 50(30) 45 (27)
0(0) 62 (37) 38 (23)
68(41) 32 (19)
58 (35) 42 (25)
15(9) 85 (51)
Variable Screening Practices Ever had a mammogram?
Self.Managed Health Practices Ever done BSE?
24 24 12 4
Never done BSE
Alcohol use No. of drinks per monlh (mean±SD) Range Note:
Confidence in BSE
Frequency of BSE Hypervigilant Appropriate Sub-Optimal
NCI 26 (n =1,580)
48 52 22.5 77.5
NCI=National Cancer Institute sample.
PSYCHOSOMATICS _ _ ,L
• • ..-....,
_..-.. _._t/ ..... ~ ..,...
Wellisch et af.
Sexuality and Body-Image Evaluation Table 5 depicts the two study groups compared to two samples of female gynecologic cancer patients. 37.38 On frequency of sexual intercourse. the comparison group had significantly more frequent intercourse (3-4 times per month) than the daughter group (2-3 times per month) (p=0.OO3). On sexual satisfaction. as measured by the Derogatis Sexual Functioning Inventory subtest evaluating several aspects of sexual behavior and communication about sex. the comparison group had a significantly higher score (mean=8.18. 54th percentile on the test vs. mean=7.38. 37th percentile on the test.p=0.05). No difference was found on the remainder of body-image or sexuality measures. Psychological Symptomatology Psychological symptoms, as measured by the Brief Symptom Inventory (see Figure I). showed the following: I) no significant betweengroup differences on any of the nine symptom
subscales nor on the global severity index. and 2) scores for both groups in the normal range on all nine symptom subscales and on the global severity index. 27 Coping Behaviors Factor analyses were conducted for the first 42 questions on the Ways of Coping Checklist for all 120 subjects. The subjects were asked to think of a recent stressor and to endorse the coping options in relation to this stressor. Compared to the five factors that were identified by Vitaliano et al.;'o we identified seven factors based on criteria of I) having eigenvalues greater than I. and 2) having a cumulative percentage explained by the factors greater than 50%. Comparison of the factor scores between the daughter and comparison groups showed no significant differences in use of any of the coping factors derived. Both of the groups described similar use of the same coping mechanisms. with problem-focused coping and the seeking of social support being the most frequent. accounting for 34.2% of the variance.
TABLE S. Sexual behavior and attitudes of daughters of breast cancer patients and comparison subjects compared with two samples of gynecologic patients MeanfSD Comparison Daughters Subjects (n=60) (n=60)
Normative Sample of Women with/without Sexual Dysfunction Without With (n=59) (n=lS4 )
Cancer Patients with Exenteration Vulvar35 PelvicJ6 (n=IS) (n=9)
Sexual inlercourse 28 (DSFI)'
2.92±1.55 2-3 times/mo.
3.73f1.5 I 3-4 times/mo.
2.13 1-2 limes/mo.
Global sexual satisfaclion28 (DSA)
5.17f2.10 range 0-8 (41)
4.98±2.36 range 0-8 (39)
Sexual salisfaclion28 (DSFI)
Body image 28 (DSFI)
Note: Values in parentheses are percenliles. NA=normative data not available; DSFI=Derogalis Sexual Functioning Invenlory; HOON=Sexual Arousabilily Inventory.
VOLUME 32· NUMBER 3· SUMMER 1991
Daughters of Breast Cancer Patients
DISCUSSION Potentially the most important finding in this study is the lack ofdifferences between daughters of breast cancer patients and well-matched comparisons on most variables. This is in contrast to previously reported clinical papers that reported observations highlighting the coping difficulties and anxieties of daughters of breast cancer patients. 2o-2.1 Two explanations may account for this study's results. First, this study did not assess women coming to clinics seeking help for their concerns. The observations in these previous papers were entirely based on clinical, help-seeking populations.I-IO.2~B Second, the demography of the current study subjects. especially their educational levels and relative affluence, may have
provided a "buffering effect" from some of the stress patterns described in the clinical populations. The present study subjects, therefore, may be different in this regard even from another nonclinical population of daughters who are less educated and affluent. We expected a "halo effect" in the daughters' perception of the mother-daughter relationship in the daughter group. This was not borne out and shows that such an illness did not, by definition, create an overestimation or a devaluation of the relationship for this group. The potential importance of this may be that the daughter does not appear, at least consciously, to inextricably link her relationship with her mother to any qualitative evaluation of breast cancer. We predicted, based on previous studies,
FIGURE J. Brief Symptom Inventory for daughters and cootrols
Wellisch et al.
that the daughter group would be more knowledgeable about breast cancer.II.D.IS This was not supported by the data. Many factors potentially account for this lack of difference, as well as the greater factual knowledge of both groups compared to the 1979 NCI sample. 26 Perhaps the most cogent explanation is the generalized explosion of public knowledge and media attention to breast cancer since 1979. Another probable factor in the greater factual knowledge is the high educational level ofthe study subjects. However, the daughter group's identification of a significantly greater number of breast cancer symptoms may reflect their latent, yet potent, sensitization to its symptoms. Our findings support those of Kelly in terms of the daughter group's estimation of likelihood of getting breast cancer. 21 These women overestimated this likelihood, as did 51 % of Kelly's "daughter" subjects. 21 It is noteworthy that 22% of our comparison subjects indicated a likely chance of getting breast cancer, a percentage actually double the probable figure. I Again, this may reflect the societal raising of consciousness by the media, making even women at normal risk feel more threatened by breast cancer. We predicted that the daughter group might perform BSE more frequently because of greater feelings of vulnerability, based on several studies utilizing the Health Belief Model. IW This was not borne out by the data. In fact, our findings agreed with those of a previous study of high-risk women in that BSE was rarely performed in a hypervigilant fashion. I I The current study also supports an earlier finding that the daughter and comparison groups did not differ in confidence levels in performing BSE. II It is noteworthy, however, that a sizable percentage of both groups expressed little or no confidence in performing BSE, also performing BSE suboptimally in terms of frequency. This has not improved since the 1979 NCI survey.26 Participation in mammography was equal among the groups and appeared age-appropriate. According to American Cancer Society guidelines for women in general, all women aged 40 and above should have had at least one screening mammogram. 34 For the present study this would include 58.3% of both groups. VOLUME 32· NUMBER 3· SUMMER 1991
Both study groups exceeded this figure, the comparison group by 4.7%, the daughter group by 16.7%. This does appearto represent a change from the 1979 NCI survey, in which 46% of the subjects were aged 45 and above and the rate of ever having had a mammogram was 19%. Thus, despite the hypersensitive or even fatalistic attitude of the daughter group about getting breast cancer, this did not appear to translate into phobic avoidance or neurotic hypervigilance in terms of screening practices or self-managed health behaviors compared to the comparison subjects. This is in contrast to other findings in the area. For example, two previously cited studies showed family, especially maternal, history of breast cancer to be a potent motivator for mammographic screening. 'K.'9 It is likely that the demographic factors of education, socioeconomic level, and religion
Daughters of Breast Cancer Patients
avoidance and rumination (unproductive thinking) further explains the very high utilization rates of mammography and Pap smears. as well as the percentages of those watching their diets. exercising. and restricting their tobacco use. Both of these groups appeared to be "doers" as opposed to "ruminators"; they seemed to try to reduce anxiety and gain increased control over their lives and health by taking action. The results in regard to sexuality and body image highlight substantial and psychologically complex differences between the study groups. The significant differences in frequency ofsexual intercourse and (behavioral) sexual satisfaction in these two very well matched groups are noteworthy. These findings might be best understood by the comment of one daughter in a clinical interview speaking about her body and sexuality. She stated: "Why should I get attached to my body and start enjoying sex when all of that will be totally destroyed when I get breast cancer like my mother did?" The area of sexuality may be the most susceptible link in the chain affected by the daughter subjects' sense of heightened vulnerability about getting breast cancer (compared to that of the comparison subjects). The daughter subjects' mean score in the sexual satisfaction subtest of the Derogatis Sexual Functioning Inventory actually placed them closer (on a percentile basis) to patients who had vulvar cancer than to the study comparison group. This may reflect the risk and psychological vulnerability they feel in this area, i.e.. their profound identification with the most damaging aspect of their mothers' breast cancer. In a review of the psychosexual sequelae of breast cancer, Schain established an incidence of frank sexual dysfunction postmastectomy at approximately 25%.·~o Given that 25% of women postmastectomy experience frank sexual dysfunction, the other 75% may experience discomfort along a spectrum short of actual dysfunction. Clinical work with this population has shown this to be true. Data from this study indicate that maternal sexual dysfunction was communicated to the daughters and internalized by them. Such communication may have been multileveled. either by verbal interaction or by the daughters' observations of
maternal discomfort in the areas of self-eonsciousness and body image. Identification is a powerful and influential fonn of learning. We speculate that identification with this aspect of the mother's experience reduced the daughter's frequency and reported pleasure in the sexual sphere. If the usual fears of the mastectomy patient are considered-including fear of rejection. fear of humiliation. tension in being nude with a partner. and fears of disrobing-it follows that daughters might identify with these feelings and behavioral problems. Daughters. especially those who were younger when a mother was diagnosed and treated for breast cancer. might not discriminate sexual dysfunction or trauma secondary to breast cancer from general role modeling and learning in the area of sexuality. Surprisingly. both study groups scored substantially lower in body-image satisfaction than would be expected of healthy nonnals. This may reflect the national climate of ever-increasing body consciousness and body-image self-criticism. which perhaps increased between the time the Derogatis Sexual Functioning Inventory was standardized in the mid-1970s and now. 41 As with the other data in this study. it is probable that the sexuality and body-image findings are affected by the subjects' educational. socioeconomic. and religious characteristics. It is possible. especially in regard to the body-image data. that this is a reflection of upper-socioeconomic preoccupations with body image. However. another potent variable may have been the relatively young age of this study population (mean. early 4Os). while other studies in the area have focused on older women (mean. late 50s to early 6Os)}8.19 Younger women can be expected to have more body consciousness and self-criticism. As stated previously. a methodological caveat is necessary for the study. Although the two study groups were very well matched. they were selected from the West Los Angeles area and do reflect the demography of this area. This creates constraints as to the generalizability of the results. Further studies of daughters of breast cancer patients with more variability in socioeconomic status and with additional racial dimensions are definitely needed.
_ . ..a-..lLa..M.
Wellisch et al.
This research was partially funded by a grant from the American Cancer Society. National Office (PBR-23) and from the National Cancer Institute, NIH (CA-16042).
Part II of this article, "Characterizing the Distressed Daughter of the Breast Cancer Patient," will appear in aforthcoming issue.
References I. American Cancer Society: Cancer Facts and Fixures. New York. American Cancer Society. 1989 2. Adami HO. Hansen J. Jung B. et al: Familiarity in breast cancer: a case control study in a Swedish populalion. Br J Cancer 42:651~58. 1980 3. Anderson DE: Genetic risks for human breast cancer. in Control of Breast Cancer ThrouXh Mass S("I"eeninx. Edited by Strax P. Linleton. MA. PSG. 1979 4. Leis HP: Breast cancer: patients at risk. in Comrol 0/ Breast Cancer Thrnuxh Mass S("I"l.'I.'ninx. Edited by Strax P. Linleton. MA. PSG. 1979 5. Austin OF: Breast cancer and ethnic heritage. JAMA 259:587-588. 1988 6. Bain C. Speizer FE. Rosner B. et al: Family history of breast cancer as a risk indicator for the disease. Am J Epidemiollll:301-308.198O 7. Helmrich SP. Shapiro S. Rosenberg L. et al: Risk factors for breast cancer. Am J Epidemiol 117:35--45. 1983 8. Kelsey JL: A review of the epidemiology of human breast cancer. Epidemiol ReI' 1:74-109. 1979 9. Sanin R. Rubin GL. Webster LA. et al: Family history and risk of breast cancer. JAMA 253: 1908-1913. 1985 10. Ottman R. Bike MC. King MC. et al: Practical guide for estimating risk for familial breast cancer. Lancet 2:556558. 1983 II. Alagna SW. Morokoff PJ. Beveu JM. et al: Performance of breast self-examination by women at high risk for breast cancer. Womm Health 12:29--46. 1987 12. Calnan MW. Moss S: The health belief model and compliance with education given at a class in breast self-examination. J Health Soc Behal' 25:198-210.1984 13. Calnan MW: The health belief model and panicipation in programmes for the early detection of breast cancer: a comparative analysis. Soc Sci Med 19:823-830. 1984 14. Grady K. Kegeles SS. Lund AK. et al: Who volunteers for a breast self-examination program? Evaluating the bases for self-selection. Health Educ Q 19:79-94. 1983 15. Massey V: Perceived susceptibility to breast cancer and practice of breast self examination. Nurs Res 35:183185.1986 16. Nightingale S: Use of mammograms to detect breast cancer. Am Fam Physician 37:415--416. 1988 17. Rutledge ON. Hanmann WHo Kinman PO. et al: Exploration of factors affecting mammography behaviors. PI"I.'I'Med 17:412--422.1988 18. Taplin S. Anderman C. Grothaus L: Breast cancer risk and panicipation in mammographic screening. Am J Public Health 79: 1494-1498. 1989 19. Zapka JG. Stoddard AM. Costanza ME. el al: Breast
VOLUME 32· NUMBER 3· SUMMER 1991 oM •• ...... &
cancer screening by mammography: utilizalion and associaled factors. Am J Public Health 79: 1499-1502. 1989 20. Kelly PT: Risk counseling for relations of cancer patienls: new information. new approaches. Journal of Psychosocial ancoloxy 5:65-79. 1987 21. Kelly PT: "High risk" women: breast cancer concerns and health practices. From Radial Ther Onco/17: 11-15. 1983 22. Jasten OM. Evans AM. Love RR: The cancer prevention clinic: a service program for cancer prone families. Journal o/Psychosocial Oncoloxy 3:5-20.1985 23. Hyland JM. Novotny ES. Coyne L. et al: Coping with difficult to treat cancer patients. Bull Menninxer Clin 48:329-341. 1984 24. Rosenfeld A. Caplon G. Yaroslavsky A. et al: Adaptation of children of parents suffering from cancer: a preliminary study of a new field for primary prevention research. Journal of Primary Prl.'l'emion 3:244-250. 1983 25. Koocher GP. O'Malley JE: The "Damo('/es Syndrome" : Psycho.wcial Consequences (If Surl'il'inx Childhood Cancer. New York. McGrdw Hill. 1981 26. National Cancer Inslilule: Breast Cancer: A Mea.mre of Proxress in Publit" Understandinx. NIH Publication No. 84-2291. Bethesda. MD. National Cancer Institute. 1979 27. Derogatis LR. Spencer MS: Brief Symptom Inl'emory IBS/): Administration. Scorinx and Procedures Manual I. Baltimore. MO. Clinical Psychometric Research (pub). 1982 28. Derogalis LR. Melisaratos N: The OSFI: a multidimensional measure of sexual functioning. J Sex Marital Ther 5:244-281. 1979 29. Hoon EF. Hoon PW. Wincze JP: An invenlory for the measurement of female sexual arousability: the SAL Arch Sex Behal' 5:291-300. 1976 30. VitalianoPP. RussoJ.Carr JE.etal: The Ways ofCoping Checklist: revision and psychometric propenies. Multi\'U/'iate Behal'ioral Research 20:3-26. 1985 31. Breslow NE. Oay NE: Statistical Ml.'1hod in Cancer Research. Vol I: The Analysis of Case-Control Studies. Lyon. Frdnce. International Agency for Research on Cancer. 1980 32. Afifi A. Clark V: Computer Aided Multimriate Analysis. Belmont. CA. Lifetime Learning Publications. 1984 33. SAS Instilute.lnc: .'lAS/STAT Guide for Personal Computer. Version 6 Edition. Cary. NC. SAS Institute. 1987 34. American Cancer Sociely: A study of women's awareness and use of mammograms. Sponsored by the American Cancer Society: inlerviewing by the Gallup
Daughters of Breast Cancer Patients
Organization; analyzed by Lieberman Research Inc. 1987 35. Kessler L: Brea.~t cancer screening in the US: results from the 1987 Cancer Control Supplement. Paper presented at the 1988 annual meeting of the American Public Health Association. Boston. 1988 36. Office on Smoking and Health; Current Population Survey. 1985 (state-specific unpublished data). Cited in Reducinllthe Health ConsequenCfs of Smokinll: A Report of the Surlleon General. DHHS Publication No (CDC) 89-8411. Rockville. MD. US Dept of Health and Human Services. 1989 37. Andersen BL. Hacker NF; Psychosexual adjustment
after vulvar surgery. Obstet Gynecol 62;457-462. 1983 38. Andersen BL. Hacker NF: Psychosexual adjusunent following pelvic exenteration. Obstet Gynecol 61 :331-338. 1983 39. Hayward RA. Shapiro MF. Freeman HE. et al: Who gets screened for breast cancer? Results from a new national survey. Arch Intern Med 148; \177-\181. \988 40. Schain WS: Breast cancer surgeries and psychosexual sequelae: implications for remediation. Seminars in OncolollY Nursinll 1:200-205. 1985 41. Yager J. Landsverk J. Edelstein C: A 20-month followup of 628 women with eating disorders. I: Course and severity. Am J Psychiatry 144: 1172-1177. 1987