|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Quality of Life of Women With Recurrent Breast Cancer and Their Family MembersByFrom the University of Michigan School of Nursing, Ann Arbor; Wayne State University College of Nursing, University of Detroit Mercy College of the Health Professions, and Karmanos Cancer Institute, Detroit; and Beaumont Hospital, Royal Oak, MI; and Yale University Department of Epidemiology and Public Health, New Haven, CT. Address reprint requests to Laurel L. Northouse, PhD, RN, University of Michigan School of Nursing, 400 N Ingalls, Ann Arbor, MI 48109-0482; email: lnortho{at}umich.edu
PURPOSE: Little information is available about the effects of recurrent breast cancer on the quality of life of women and their family members. The present study assessed patients’ and family members’ quality of life within 1 month after recurrence, and effects of multiple factors on quality-of-life scores. PATIENTS AND METHODS: Patient/family member dyads (N = 189) participated in this study. A stress-appraisal model guided selection of person factors, social/family factors, illness-related factors, appraisal factors, and quality of life, measured with psychometrically sound instruments. Quality of life was measured with both generic (Medical Outcomes Study SF-36) and cancer-specific (Functional Assessment of Cancer Therapy) scales. RESULTS: Patients reported significant impairments in physical, functional, and emotional well-being. Family members reported significant impairments in their own emotional well-being. Structural equation modeling revealed that self-efficacy, social support, and family hardiness had positive effects on quality of life, whereas symptom distress, concerns, hopelessness, and negative appraisal of illness or caregiving had detrimental effects. Study variables accounted for a sizable amount of variance in patients’ and family members’ physical and mental dimensions of quality of life (72% to 81%). Contrary to findings observed in studies of newly diagnosed breast cancer patients and spouses, little relationship was found between recurrent patients’ and family members’ quality of life. CONCLUSION: Women with recurrent breast cancer are in need of programs to assist them with the severe effects of the disease on their quality of life. Programs need to include family members to help counteract the negative effects of the recurrent disease on their mental health, and to enable them to continue as effective caregivers.
THE RECURRENT phase of breast cancer is an extremely difficult time for women and their family members,1-3 yet only a handful of studies have assessed women’s quality of life or identified factors that may affect it during this period of time.1,4,5 We also lack empirical information on how the recurrence affects family members’ quality of life and the mutual influence that patients and family members may have on each other’s adaptation. Recent research suggests that it is essential to extend the quality-of-life assessments to key family members because they are the primary support to the patient,6,7 and because the adjustment of family members seems to have a significant effect on the adjustment of patients.8 The purpose of this study was to provide a comprehensive assessment of the quality of life of women with recurrent breast cancer and their family members within 1 month after learning that the breast cancer had recurred or progressed. The specific objectives were to (1) describe the multidimensional quality of life of women with recurrent breast cancer and their family members; (2) identify the influence of demographic, medical, and other psychosocial factors on women’s and family members’ quality of life; and (3) examine the interrelationship of women’s and family members’ quality of life during the recurrent phase of illness.
A number of treatment approaches have been used to manage recurrent breast cancer, but the gains in survival time have been modest.9 Factors such as axillary lymph node status, site of recurrence, positive estrogen receptor status, and a longer disease-free interval have all been independently associated with longer survival times after recurrence.10-13 Quality of life is an important issue for these women and their family members because of the limitations imposed on women’s quantity of life by the return of the disease. Although much attention has been directed to the medical management of breast cancer, little information is available about the psychosocial aspects of recurrent breast cancer. Findings from the few studies in this area indicate that cancer patients with recurrent disease report poorer physical functioning,14 poorer perceptions of their health,14 less hope,15 and more problems in their relationships with physicians and other health professionals than patients with newly diagnosed disease.2,14-16 The few studies that have examined family members’ responses to recurrent breast cancer indicate that they experience considerable emotional distress, have many adjustment problems, report high levels of uncertainty, and perceive little support from others during the recurrent phase of illness.4,5,17-19
Although the recurrence of breast cancer is associated with considerable distress, there are few empiric data that explain why some women and family members are able to maintain their quality of life after a recurrence, whereas others are not. In this study, a stress-appraisal model was used to identify factors that may account for variance in the quality of life of women and family members during the recurrent phase of illness (Fig 1). This model was adapted from the theoretical work of Lazarus,20,21 Lazarus et al,22 and Lazarus and Folkman23 for cancer patients and their family members.
According to the theoretical model shown in Fig 1, there are a number of antecedent variables (eg, person, social, and illness-related) and mediating variables (eg, appraisal of illness, uncertainty, and hopelessness) that may directly or indirectly affect patients’ and family members’ quality of life. The stress-appraisal model also specifies that there is a reciprocal relationship between the quality of life of patients and family members. In our prior research, this model accounted for a significant amount of variance in couples’ adjustment to a new diagnosis of breast cancer24 and also to a new diagnosis of colon cancer.8 We were interested in determining how much variance this model would account for in the quality of life of women and family members dealing with recurrent breast cancer, and which variables would have the strongest association with quality of life. Within the model, person factors are those characteristics of individuals that may influence their appraisal of the illness and ultimately their quality of life. This broad category includes specific variables such as demographics (ie, age, education), self-efficacy, and current concerns. In a number of studies, younger women with breast cancer report more emotional distress than older women,25-27 whereas older women report more physical limitations.28 In the present study, we hypothesized that younger age may be related to more stressful appraisal of the illness because of the threat that often accompanies recurrent cancer, especially for younger women who are concerned about surviving to raise young children. Education also was hypothesized to affect the appraisal variables and indirectly affect quality of life. However, the findings from other studies related to this hypothesis have been mixed.29,30 In the present study, we expected that education would be related to quality of life, but that its contribution would be small. Self-efficacy is defined as the amount of confidence that cancer patients have in their ability to manage the demands associated with their illness. Self-efficacy for family members refers to the confidence they have in their ability to provide care (emotional or physical) to their ill family member. Existing studies have not examined the role of self-efficacy in managing recurrent breast cancer, but in studies of other health-threatening conditions, higher self-efficacy has been associated with better outcomes for patients and their caregivers.31 Current concerns, or the presence of other concerns that people are trying to manage at the same time that they are managing a cancer diagnosis, have been related to patients’ and family members’ appraisal of illness and/or adjustment in several studies.17,27,32,33 It seems that as these other competing concerns increase, patients and family members are more likely to view the cancer as threatening and have more difficulty adjusting to it, possibly because their resources to deal with the cancer are being depleted by other competing concerns. In the present study, we hypothesized that more concerns would be directly related to a more negative appraisal of illness, and indirectly to a lower quality of life. As illustrated in Fig 1, there are also social/family factors that may affect quality of life, such as family hardiness and social support. Family hardiness is the family’s internal strength and collective ability to manage hardship and change.34,35 Family caregivers in one study who reported higher family hardiness were more likely to appraise the illness/caregiving as "challenging" or "beneficial" rather than as a negative experience.34 In this study, we hypothesized that family hardiness would be directly related to the appraisal variables and indirectly to higher quality of life. Higher social support has been associated with better adjustment among women and their family members in the newly diagnosed and in the recurrent phase of breast cancer,6,7,17,36,37 and to longer survival among women with advanced disease.38 Support seems to help individuals by reducing their negative appraisal of illness (ie, threat) and by lessening their stressful response to the illness (ie, less depressed mood).39 In this study, we hypothesized that higher social support would be directly related to more positive appraisal and indirectly related to higher quality of life. There are also a number of illness-related factors that may affect quality of life after recurrence: symptom distress, stage of disease, and length of disease-free interval between diagnosis and recurrence. Of these factors, higher symptom distress has been most closely related to lower quality of life among women with breast cancer40-43 and their family members.17,44 In this study, we also examined the amount of symptom distress reported by family caregivers because of their own health problems. We hypothesized that higher symptom distress (in either patients or caregivers) would have a direct effect on each person’s quality of life and an indirect effect on their quality of life through the appraisal variables (Fig 1). Stage of disease at the time of recurrence and the length of disease-free interval also may affect quality of life because of their significant relationships to survival.10 We hypothesized that these medical variables would have a direct effect on quality of life. According to the stress-appraisal model, there are three variables in the model that are considered appraisal variables: appraisal of illness/caregiving, uncertainty, and hopelessness. These variables are hypothesized to directly affect quality of life and to mediate the effects of some of the antecedent variables on quality of life. On the basis of previous studies, patients and family members who appraise the cancer as more threatening,34,45 who express more uncertainty about the illness,8 and who report more hopelessness about their futures17 have more problems adjusting to the illness and report a lower quality of life. The model also specifies that a significant relationship exists between the quality of life of patients and of their family members (Fig 1). This significant relationship has been reported in a number of studies (Mellon et al, under review)17,46-47; however, the strength of the correlation varies across studies. In one recent study, the level of adjustment reported by family members accounted for more variance in the adjustment of patients with cancer than did demographic or medical variables.8 This finding underscores the importance of examining the influence of patients’ and family members’ quality of life on one another in multivariate studies. Finally, we examined quality of life from a cancer-specific perspective (Functional Assessment of Cancer Therapy [FACT] scales) and a general health perspective (Medical Outcomes Study SF-36) to gain a more comprehensive understanding of how recurrent breast cancer affects patients’ and family members’ quality of life.
This study was part of a larger prospective study that examined the effects of a family intervention on the quality of life of women with recurrent breast cancer and their family members.48 In this portion of the study, we examined baseline data obtained from patients and family members just before their randomization to treatment and control groups.
Sample The names of 250 eligible women with an eligible family caregiver were obtained from clinic staff and, of these eligible dyads, 200 (80%) agreed to participate in the study. The actual number of patients who were not interested in learning about the study when approached by clinic staff is unknown. Even though women with recurrent breast cancer from all stages of disease were eligible, only a small number (5.5%) had early-stage disease. Thus, only the 189 women with more advanced disease (ie, stage III [11.5%] and stage IV [83%]) and their caregivers were included in the analyses to provide a more homogeneous sample of women with advanced breast cancer. The caregivers for these women consisted of husbands (59%), adult daughters (14%), adult sons (5%), sisters (8%), or other relatives or friends (14%). The average age of patients was 54 ± 11.2 years (range, 22 to 86 years) and the average age of their family members was 52 ± 13.4 years (range, 18 to 87 years). Seventy-three percent of the dyads were white, 23% were African-American, 2% were Asian, and the remaining 2% were either Hispanic or Native American. The ethnic composition of our sample was similar to the ethnic composition reported for women with breast cancer by the Surveillance, Epidemiology, and End-Results cancer registry (eg, 20.3% African-American) for our geographic area. However, the median age of our sample was younger than the age reported for the Surveillance, Epidemiology, and End-Results registry sample (53 years v 60.7 years), but similar in age to other published studies of women with breast cancer.49,50 Both women and family members in our study reported an average of 14 years of formal education and a median family income in the $30,000 to 50,000 range. Approximately 43% of the women and 72% of the family members were working outside of the home. Half of the women had a family history of breast cancer, and half did not. At the time of initial diagnosis, approximately one third of the women had had a lumpectomy and two thirds had had a mastectomy; 65% reported evidence of disease in adjacent lymph nodes at the time of their surgery. Approximately one fourth of the women had undergone breast reconstruction after their initial surgery. A little over half of the women (55%) reported this was their first recurrence, whereas others reported this as their second (28%), third (8%), or fourth or greater (9%) recurrence/progression. The average disease-free interval between initial diagnosis and first recurrence was 3.8 years; between first and second recurrence, 2.1 years; and between second and third recurrence, 1.1 years. Most of the women were receiving one or more of the following treatments: chemotherapy (73.1%), radiation therapy (24.2%), hormone therapy (19.9%), or bone marrow transplant (11.8%). Approximately half of the women (56%) reported having other health problems, such as hypertension or heart disease.
Concepts, Variables, and Instruments
Person factors. Demographic variables and current concerns were measured with the Omega Screening Questionnaire (OSQ), developed by Mood51 from the original clinical screening interview by Weisman and Worden.52,53 The OSQ is composed of four sections: demographic information, health history, inventory of current concerns, and symptoms scale. Two sections of the OSQ, demographics and current concerns, were used to measure person factors. The demographic section asks about variables such as age, education, and income. The Inventory of Current Concerns is a 40-item scale that asks respondents to rate the extent to which they are experiencing concerns related to work/finance, friends, religion, family, and so on. Higher scores indicate a greater number of concerns. Test-retest and internal consistency reliability, in addition to content, predictive, and construct validity of the OSQ, including the Inventory of Current Concerns have been reported previously.51 Self-efficacy was measured with an abbreviated 17-item version of the Lewis Cancer Self-Efficacy Scale54 that assesses respondents’ confidence in managing the effects of cancer. We adapted the wording to create a separate 17-item scale that measured family members’ confidence in their ability as caregivers to manage the cancer. Higher scores indicate higher levels of confidence. Evidence of content and criterion validity has been reported by Lewis.54 Social/family factors. Family hardiness was measured with the Family Hardiness Index (FHI),35,55 a 20-item scale that assesses the internal strengths and durability of the family unit. Higher scores on the FHI indicate more family hardiness. Adequate content and construct validity and internal consistency reliability of the scale have been reported.55 In this study, the FHI was administered only to family members to decrease some of the instrument burden on patients. Social support was measured with the 25-item Likert-type Personal Resource Questionnaire developed by Brandt and Weinert.56 Higher scores indicate more support. Evidence about the internal consistency reliability, and predictive and construct validity of the scale has been reported in the literature.56,57 Illness-related factors. Symptom distress was measured with the symptoms scale of the OSQ discussed earlier.51 Participants rated 13 symptoms (eg, fatigue, breathing problems, pain); higher scores indicated more symptom distress. Patients and family members each rated their own symptom distress. Evidence of concurrent validity, internal consistency, and test-retest reliability of the scale has been reported.51,58 A researcher-designed questionnaire was used to supplement health history and medical treatment data on participants (eg, time since diagnosis). Appraisal factors. The patients’ appraisal of illness was measured with the Appraisal of Illness Scale, a 27-item Likert-type instrument, with higher scores indicating more negative appraisal of the illness.59,60 Family members’ appraisal of caregiving was measured with the Appraisal of Caregiving Scale, a 27-item measure that assessed their appraisal in response to tasks and problems associated with caregiving.61 A higher score on this scale indicates a more negative appraisal of caregiving. Adequate construct validity and internal consistency reliability of the two scales has been reported.59-61 Uncertainty about the illness/treatment was measured with the 28-item community version of the Mishel Uncertainty in Illness Scale for patients and the 29-item version of that instrument for family members.62 One item from the family version that was different from the patient version was deleted to enable us to make comparisons between patient and family members’ scores. Adequate internal consistency and construct validity of the original scales has been reported for both patients and family members.62 Hopelessness was measured with the Beck Hopelessness Scale, a 20-item true/false scale, with higher scores indicating more hopelessness about the future. Evidence of concurrent and construct validity and internal consistency has been reported.63 The quality-of-life outcome variable was measured with two instruments: a cancer-specific quality-of-life instrument (FACT scale, version 3) and a generic health-related survey (Medical Outcomes Study SF-36). The FACT scales were developed by Cella et al,64 and they assess five dimensions of quality of life: physical, social/family, relationship with doctor, emotional, and functional well-being. The FACT-G (ie, general format), which contains 28 generic quality-of-life items, was used to measure family members’ general quality of life. Slight modification was made in the wording of the instrument so that family members reported on their own quality of life rather than the patients’. The FACT-B (ie, breast cancer–specific format) was used to measure patients’ quality of life.49 The FACT-B contains the same 28 general items as the FACT-G plus nine additional breast cancer–specific items. Evidence of the reliability and validity of both the FACT-G and FACT-B have been reported.49,64 Higher scores on the FACT scales indicate a higher quality of life. The Medical Outcomes Study SF-36 consists of 36 items that comprise eight health concepts, such as role limitations caused by physical or emotional concerns, bodily pain, vitality, social health, and so on. Higher scores on this scale indicate higher quality of life. This instrument has undergone extensive psychometric testing and has been found to have adequate internal consistency and validity.65 For the description of patients’ and family members’ quality of life, we used the subscales and total scale scores of the FACT (cancer-specific scale) and the subscales and summary scores of the SF-36 (health-related scale). However, for the model testing, we created two latent constructs, mental quality of life and physical quality of life, from the various subscales of the FACT and SF-36 (Table 3).
Before using the latent constructs in the structural equation modeling, we first assessed the constructs using a measurement model as recommended by Anderson and Gerbing.66 The four latent constructs assessed were (1) patient mental quality of life, (2) patient physical quality of life, (3) family member mental quality of life, and (4) family member physical quality of life. Overall, the measurement model showed an adequate fit,  2 (df = 248) = 622.10, 2/df = 2.51, comparative fit index (CFI) = 0.82, and the root mean square error of approximation (RMSEA) = 0.091 (90% confidence interval, 0.082 to 0.099). The CFI was somewhat low; however, standardized factor loadings for the latent variables were acceptable in the moderate to large range; none of the loadings were below 0.40 (Table 3). Although the three mediator variables (appraisal of illness, uncertainty, and hopelessness) are all "appraisal variables" (Fig 1), they were treated as separate manifest variables rather than as one latent variable in order to learn more about the specific paths leading to and from these appraisal variables. This information was deemed essential because each of these variables may have different ramifications for future interventions designed to increase patients’ and caregivers’ health and quality of life.
Procedures Dyads who were willing to learn more about the study were contacted by a member of the research team, who explained the study in greater detail. Dyads who agreed to participate were scheduled for a data collection session that took place in the home. Before data collection, participants signed a consent form approved by the university institutional review boards at the two study sites. Patients and caregivers completed instrument packets separately.
Statistical Analysis Structural equation modeling, using covariances and maximum likelihood estimation, was used to assess the stress-appraisal model. A single model including both patients and family members was tested with AMOS 4.0.67 The proposed theoretical model had a number of observed predictors (age, education, self-efficacy, current concerns, family hardiness, social support, symptom distress, disease-free interval, and stage of disease), three observed mediators for both patient and family members (appraisal of illness/caregiving, uncertainty, and hopelessness), and two quality-of-life latent variables (mental quality of life and physical quality of life). We assessed a model where patient predictors influence patient appraisal and quality of life and family predictors influence family appraisal and quality of life. In addition, we assessed the influence of selected patient variables (eg, stage of disease, patient symptom distress) on family outcomes and selected family variables (eg, family hardiness, family member symptom distress) on patient outcomes. Reciprocal paths were assessed from the patient and family member quality-of-life scores to determine the possible mutual influence that each member’s quality of life may have on the other member. The partial mediation model shown in Fig 1 was tested. Modification indices were examined to determine whether any additional direct effects from the person and social/family predictors to the outcomes would significantly improve the model fit. Only direct effects that significantly improved the model fit, and were consistent with the theoretical framework, were added to the model.
The model was evaluated in three ways. First, the model was evaluated for goodness of fit using several indices of fit: the
Description of Patient and Family Member Quality of Life Means and SDs for the indicators of quality of life are listed in Table 1. One-sample t tests were conducted to compare patients’ scores to other published samples. A Bonferroni alpha adjustment (P = .001) was used to avoid an increase in type I error resulting from multiple comparisons. On the FACT-B scale, women with recurrent breast cancer in this study reported significantly lower overall quality of life than a normative sample of primarily early-stage breast cancer patients, particularly in areas related to physical, emotional, and functional well-being (Table 1). They also reported significantly lower scores than cancer patients in general on the physical health summary scale of the SF-36. On the specific subscales of the SF-36, they reported poorer physical, social, and mental health; more pain; and more role limitations than a comparison group of seriously ill medical patients (Table 1). Family members’ mean score on the SF-36 physical summary scale (50.9) was similar to the mean (50.0) reported for the normal population,73 but their mental health summary score (48.4) was worse than the mean for the normal population (50.0, t[119] = -2.22, P = .03). A series of matched t tests was used to compare patients’ and family members’ scores using an adjusted alpha of .001 (Bonferroni). Patients reported significantly lower overall quality of life than family members did on the FACT-G and on subscales that assessed physical and functional well-being. Patients did not differ from family members, however, on the subscales that assessed emotional or social well-being, with both reporting similar impairments. On the relationship with the doctor subscale, family members’ scores were significantly lower than patients’ scores. On this subscale, family members reported less confidence in the physician than did patients, and less satisfaction with the physician’s availability to answer their questions. On the second quality-of-life instrument (SF-36), the findings obtained were similar to those observed with the FACT-G. Patients reported significantly lower scores on the SF-36 physical health summary scale than family members, and scored lower than family members on almost all of the subscales except mental health. On the mental health subscale, patients and family members’ scores were similar; both reported impairments in their mental health. In addition, both patients’ and family members’ mental health scores were below (worse than) the level reported for a sample of seriously ill medical patients (Table 1).
Description of Other Study Variables
Factors Related to Quality of Life
Structural equation modeling was used to examine the direct and/or indirect effects of the independent variables on the mediators and quality-of-life constructs for patients and family members. In order to create a parsimonious model, and because we hypothesized relationships between patient and family member variables, a single model was tested that included both patient and family member variables. However, in order to facilitate interpretation, the results are presented separately for patients and family members (Fig 2). Correlations among similar predictors for patients and family members (eg, patient self-efficacy and family member self-efficacy) were incorporated in the model to represent familial influences and to take into account the dependence of measures.74 Individual patient and family member models were also tested, and although the nature and magnitude of effects did not differ substantially, the fit was slightly less adequate for the individual models versus the single model that incorporated both patient and family member variables.
The significant paths for patients and family members are presented in Fig 2 and the coefficients for all the direct and indirect paths tested are listed in Tables 6 and 7. Overall, the model showed adequate fit, 2 = 1,617.04 (df = 810), 2/df = 1.99, CFI = 0.82, and RMSEA = 0.07 (90% confidence interval, 0.068 to 0.078). More specifically, the 2/df and RMSEA fit indices showed adequate/good fit of the proposed model but the CFI was slightly lower than what is usually considered to be an adequate fit. However, the examination of residuals and modification indices did not suggest the addition of any paths that would help the fit and make conceptual sense. The model accounted for a sizable amount of variance in both patients’ and family members’ mental (80.6% and 81.2%, respectively) and physical quality of life (71.5% and 71.7%, respectively) (Fig 2). To explore the generalizability of our model, we also assessed model fit depending on whether the family caregiver was a spouse (n = 112) or another type of family member such as sibling, adult child, and so on (n = 77). The results showed a good fit of the model for both groups of caregivers, with similar significant direct and indirect paths between variables. Thus, the results that follow are derived from the combined data for both spouse and nonspouse family caregivers. As predicted in the theoretical model, person factors, namely, self-efficacy and current concerns, were significantly related to most of the appraisal variables for both patients and family members. The appraisal variables, with the exception of uncertainty for patients, also mediated the effect of self-efficacy and current concerns on quality of life. Hence, patients and family members with higher self-efficacy and fewer current concerns had less negative appraisals of the illness and/or caregiving, and subsequently a higher quality of life. In addition to these indirect effects, self-efficacy also had a direct, positive effect on patients’ mental health (Fig 2A), and increased current concerns had a direct, negative effect on family member’ physical health (Fig 2B). Education made no significant contribution to the patient model, but was indirectly related to family members’ mental health through negative appraisal of caregiving and hopelessness (Fig 2B). In other words, family caregivers who were more educated reported a more negative appraisal of their caregiving role and more hopelessness than less educated caregivers that subsequently related to lower mental quality of life. Of the social/family factors, social support was indirectly related to patients’ quality of life through hopelessness. Patients with more support reported less hopelessness and higher quality of life (mental and physical). Social support was not related directly or indirectly to family members’ quality of life. Family hardiness had direct and indirect effects on family members’ quality of life (through hopelessness and uncertainty), but no effect on patients’ quality of life. Thus, family members who perceived more hardiness reported less hopelessness and less uncertainty after the recurrence, and experienced better mental and physical quality of life. Because only family members completed the family hardiness scale, this may explain the significant effects on family members’ but not patients’ quality-of-life scores. Of the illness-related factors, patients’ symptom distress had a direct negative effect on patients’ own mental and physical quality of life and on their family members’ appraisal of caregiving. For the family members, their own symptom distress had direct negative effects on their mental and physical quality of life. Family members’ symptoms also had a negative effect on their appraisal of caregiving. The more symptom distress family members reported, the more negatively they evaluated the caregiving experience, and in turn, the more negatively they rated their own mental quality of life. Family members’ physical symptoms also were indirectly related to patients’ mental and physical quality of life through their influence on patient hopelessness and uncertainty. Stage of disease had a significant effect on all three appraisal variables for patients: patients with more advanced disease reported more negative appraisal of the illness, more hopelessness, and more uncertainty. Stage of disease had no significant effect on family member variables. Length of the disease-free interval did not make a significant contribution to either the patient or family member component of the model. For the most part, the appraisal variables had direct negative effects on patients’ and family members’ quality of life. The only exception to this pattern was uncertainty, which did not have a significant direct path to quality of life in the patient component of the model (Fig 2A). In general, however, appraisal variables exerted a negative effect on patients’ and family members’ quality of life, and these effects were greater on their mental compared with physical health (Tables 6 and 7). Finally, in our model testing, we examined the possible reciprocal influences between patients’ and family members’ quality of life. Patients’ and family members’ quality-of-life scores did not have a significant direct effect on one another in the structural equation modeling. Further indication of the lack of relationship between patients’ and family members’ quality-of-life scores was evident in the zero-order correlations among the quality-of-life indicators listed in Table 3. Only four of 12 pairs of indicators were significantly correlated (magnitudes ranged from 0.001 to 0.224), indicating the magnitude of the relationship among patient and family member indicators of quality of life was relatively weak. In the model testing undertaken, most of the significant relationships found were either within-patient variables or within–family member variables, rather than between patient and family member variables. However, symptom distress, either in the patient or family caregiver, consistently affected the other partners’ appraisal and/or quality of life.
One of the important findings from this study was the extent to which recurrent breast cancer compromised the quality of life of patients and their family members. On two well-established quality-of-life instruments, patients reported significantly poorer quality of life than a sample of breast cancer patients in general and a sample of seriously ill medical patients. Although it is not surprising that recurrent breast cancer seems to affect the quality of patients’ lives, it is alarming to observe the number of areas of their quality of life that are affected. Compared with other samples of patients, women with recurrent breast cancer had poorer physical, functional, and emotional health; more pain, concerns, role limitations, and symptom distress; and a more negative appraisal of their illness. Clearly, these women are in need of care programs that will help them to manage these severe effects of the recurrent cancer on their quality of life. Family members’ quality of life also was affected, but primarily related to their mental health. Family members’ mental health was significantly worse than the normal population, and their emotional well-being scores were slightly worse than even patients’ scores themselves. In almost all respects, the emotional profile of family members was similar to that of a seriously ill sample of medical patients on the SF-36, even though family members’ physical health was far superior. The findings of this study indicate that recurrent breast cancer seems to take a sizable emotional toll on family members. Family members are expected to be the primary providers of physical and emotional support to the patient, yet they perceive little support from others, either professional or personal, in carrying out this essential role. Family members in this study reported less support, less satisfaction with health professionals, and more uncertainty about the illness than even the patients do, findings that are consistent with other studies.4,15,75 These findings indicate that supportive services need to be extended to family members, so that they can maintain their own mental health and continue to be effective caregivers to the patient. The stress-appraisal model that we tested (Fig 1) explained a large amount of variance in both patients’ and family members’ mental (81%) and physical (72%) quality of life. This suggests the model is relevant to the recurrent phase of breast cancer, and that it contains several key variables related to the quality of life of both patients and family members. Contrary to conventional thinking, certain medical variables, such as disease-free interval, accounted for little or no unique variance in participants’ quality of life. Rather, other psychosocial variables contained within the stress-appraisal model were important factors related to participants’ quality of life. These included (1) characteristics of the person (self-efficacy, degree of current concerns), (2) their social or family situation (social support, family hardiness), (3) their illness experience (symptom distress), and (4) their appraisal of the illness or caregiving. Clinicians providing care to women with recurrent breast cancer and their family members may want to assess these four areas (personal, social/family, illness, and appraisal) because of the significant effects they have on patients’ and family members’ quality of life. These four areas also may be appropriate targets for intervention in further research with this population. Using structural equation techniques, we examined the paths through which these multiple factors affected quality of life. We found that some variables had a direct path, others an indirect path, and still others both direct and indirect paths. The process (or path) by which variables affect quality of life is important because it provides an understanding about the complexity of quality of life and also gives some indication about where to target interventions. Patients’ symptom distress, for example, had a direct effect on patients’ quality of life and an indirect effect on family members’ quality of life (through appraisal of caregiving). Similarly, family members’ symptom distress had a direct effect on their own quality of life and an indirect effect on patients’ quality of life (through hopelessness). These findings are noteworthy because they suggest that symptom distress, in both patients and family caregivers, has a significant effect on each person’s own quality of life, and an indirect effect on their partners’ quality of life. Hence, medical and nursing interventions need to be targeted to reduce the symptom distress in both patients and family caregivers. Strategies to assist patients are fairly commonplace. Strategies to assist family caregivers may include determining whether caregivers have health problems of their own that may be interfering with their caregiving role, and encouraging them, if necessary, to seek assistance for these problems from their primary care provider. In the model testing, some variables had a positive effect on quality of life and others had a negative effect. For example, self-efficacy had a positive effect on both patients’ and family members’ quality of life through multiple routes. Patients and family members with more self-efficacy reported less hopelessness, less negative appraisal of the illness or caregiving, and better quality of life. Conversely, patients and family members who reported more concurrent concerns, more symptom distress, and more negative appraisals of the illness or caregiving reported poorer quality of life. From an intervention standpoint, these findings suggest that health professionals need to develop strategies to promote patient and caregiver self-efficacy. They also need to find ways to assist patients and caregivers to reduce competing demands and manage symptom distress, which in turn may lessen negative appraisals. Strategies may include cognitive-behavioral interventions, such as contracting and goal-setting activities to practice certain self-care or caregiving behaviors.31,76,77 Strategies also may include psychoeducational and skill-building activities that help patients and caregivers gain stress-management skills, symptom-management skills, or problem-solving skills so that they are better prepared to deal with the demands associated with recurrent breast cancer.78-80 Caregivers in particular may benefit from intervention programs because they are automatically assumed to be capable of providing complex care with little preparation and support. Furthermore, in current practice, there is little assessment, if any, of the extent to which caregivers’ own health problems or their accumulating caregiving burdens may interfere with their caregiving abilities. Although we had anticipated that the quality of life of the patient would have a significant effect on the family member, and vice versa, on the basis of our previous research,8,24 that was not observed in these data. The correlations among the indicators of patients’ and family members’ quality of life were small in magnitude and were generally in areas related to their social and emotional well-being. The low correlations between patients’ and family members’ physical well-being indicators can be attributed to the fact that the physical health of recurrent breast cancer patients was much worse than the physical health of their family members or the physical health of the newly diagnosed women in our earlier studies. However, the data were not void of evidence of reciprocal effects. Such effects were observed in patients’ and family members’ levels of symptom distress. These effects may have emerged in this study because we assessed not only the patients’ symptom distress but also the caregivers’ symptom distress (the latter is a variable that has seldom been assessed in prior dyadic studies). Further research is needed to explore the possible changes in the reciprocal relationships of patients and caregivers over the cancer trajectory. There were a few limitations to this study that should be taken into consideration. First, many analyses were conducted as a part of this study, and even though we based most of our findings on a more stringent alpha (P = .001), it is possible that some findings may have occurred by chance alone, and warrant further replication. Second, even though we discussed the effects of one factor on another, which is the common language of structural equation modeling, we cannot assume causality per se from this initial study based on concurrent measures obtained at one point in time. Similarly, we cannot identify at this time the possibility of reverse causality. Our interpretation of the findings is guided, at least in part, by our theoretical framework. However, evidence of causality will be more adequately supported if we find similar results when we are able to analyze time-ordered data from our longitudinal clinical trial. Third, there may have been some item overlap among the many predictor variables and outcome variables, which would be expected to some degree because of the constructs being assessed. However, the amount of variance accounted for by the correlations between individual predictors and the quality-of-life outcome variables (Tables 4 and 5) was only moderate, even under these conditions of concurrent measurement. Fourth, although the model we tested had an adequate fit, it is possible that there are other theoretical models that may yield equal or better fit and could be tested in future studies. Despite these limitations, the findings of this study indicate that recurrent breast cancer has a serious detrimental effect on the physical, functional, and emotional well-being of women with the disease and on the mental health of their family members who are caring for them. The stress-appraisal model accounted for a large amount of variance in both the patients’ and the family members’ quality of life. In addition, a number of personal, social, illness, and appraisal factors were identified that directly or indirectly affected the women’s and their caregivers’ quality of life. Programs of care need to be offered to both women with recurrent disease and their family caregivers to assist each person in improving their quality of life as they move through their cancer trajectory.
Supported by grant no. PBR-102 from the American Cancer Society (to L.L.N., principal investigator). We thank Marilyn Oberst, PhD, for her guidance in the development of this study.
1. Weisman AD, Worden JW: The emotional impact of recurrent cancer. J Psychosoc Oncol 3: 5-16, 1985
2. Silberfarb PM, Maurer LH, Crouthamel CS: Psychosocial aspects of neoplastic disease: I. Functional status of breast cancer patients during different treatment regimens. Am J Psychiatry 137: 450-455, 1980 3. McEvoy MD, McCorkle R: Quality of life issues in patients with disseminated breast cancer. Cancer 66: 1416-1421, 1990[CrossRef][Medline] 4. Northouse LL, Laten D, Reddy P: Adjustment of women and their husbands to recurrent breast cancer. Res Nurs Health 18: 515-524, 1995[Medline] 5. Given B, Given CW: Patient and family caregiver reaction to new and recurrent breast cancer. J Am Med Womens Assoc 47:201-206, 212, 1992 6. Manne SL, Pape SJ, Taylor KL, et al: Spouse support, coping, and mood among individuals with cancer. Ann Behav Med 21: 111-121, 1999[Medline] 7. Northouse LL: Social support in patients’ and husbands’ adjustment to breast cancer. Nurs Res 37: 91-95, 1988[Medline] 8. Northouse LL, Mood D, Templin T, et al: Couples’ patterns of adjustment to colon cancer. Soc Sci Med 50: 271-284, 2000[CrossRef][Medline] 9. Greenlee RT, Murray T, Bolden S, et al: Cancer statistics, 2000. CA Cancer J Clin 50: 7-33, 2000[Abstract] 10. Insa A, Lluch A, Prosper F, et al: Prognostic factors predicting survival from first recurrence in patients with metastatic breast cancer: Analysis of 439 patients. Breast Cancer Res Treat 56: 67-78, 1999[CrossRef][Medline] 11. Schwaibold F, Fowble BL, Solin LJ, et al: The results of radiation therapy for isolated local regional recurrence after mastectomy. Int J Radiat Oncol Biol Phys 21: 299-310, 1991[Medline] 12. Halverson KJ, Perez CA, Kuske RR, et al: Isolated local-regional recurrence of breast cancer following mastectomy: Radiotherapeutic management. Int J Radiat Oncol Biol Phys 19: 851-858, 1990[Medline] 13. Halverson KJ, Perez CA, Kuske RR, et al: Survival following locoregional recurrence of breast cancer: Univariate and multivariate analysis. Int J Radiat Oncol Biol Phys 23: 285-291, 1992[Medline] 14. Frost MH, Suman VJ, Rummans TA, et al: Physical, psychological and social well-being of women with breast cancer: The influence of disease phase. Psychooncology 9: 221-231, 2000[CrossRef][Medline] 15. Cella DF, Mahon SM, Donovan MI: Cancer recurrence as a traumatic event. Behav Med 16: 15-22, 1990[Medline] 16. Mahon SM, Casperson DS: Psychosocial concerns associated with recurrent cancer. Cancer Pract 3: 372-380, 1995[Medline] 17. Northouse LL, Dorris G, Charron-Moore C: Factors affecting couples’ adjustment to recurrent breast cancer. Soc Sci Med 41: 69-76, 1995[CrossRef][Medline] 18. Halliburton P, Larson PJ, Dibble S, et al: The recurrence experience: Family concerns during cancer chemotherapy. J Clin Nurs 1: 275-281, 1992[CrossRef] 19. Lewis FM, Deal LW: Balancing our lives: A study of the married couple’s experience with breast cancer recurrence. Oncol Nurs Forum 22: 943-953, 1995[Medline] 20. Lazarus RS: Emotion and Adaptation. New York NY, Oxford University Press, 1991 21. Lazarus RS: Evolution of a model of stress, coping, and discrete emotions, in Rice VH (ed): Handbook of Stress, Coping, and Health. Thousand Oaks CA, Sage, 2000, pp 195-222 22. Freeman A, Simon KM, Beuktler LE, et al: Comprehensive handbook of cognitive therapy, in Lazarus RS (ed): Constructs of the Mind in Mental Health and Psychotherapy. New York NY, Plenum, 1989, pp 99-121 23. Lazarus RS, Folkman S: Stress, Appraisal, and Coping. New York NY, Springer, 1984 24. Northouse LL, Templin T, Mood D: Couples’ adjustment to breast disease during the first year following diagnosis. J Behav Med 24: 115-136, 2001[CrossRef][Medline] 25. Mor V, Allen S, Malin M: The psychosocial impact of cancer on older versus younger patients and their families. Cancer 74: 2118-2127, 1994[CrossRef][Medline] 26. Northouse LL, Swain MA: Adjustment of patients and husbands to the initial impact of breast cancer. Nurs Res 36: 221-225, 1987[Medline] 27. Penman DT, Bloom JR, Fotopoulos S, et al: The impact of mastectomy on self-concept and social function: A combined cross-sectional and longitudinal study with comparison groups. Women Health 11: 101-130, 1986[CrossRef][Medline] 28. Vinokur AD, Threatt BA, Vinokur-Kaplan D, et al: The process of recovery from breast cancer for younger and older patients: Changes during the first year. Cancer 65: 1242-1254, 1990[CrossRef][Medline] 29. Northouse LL, Jeffs M, Cracchiolo-Caraway A, et al: Emotional distress reported by women and husbands prior to a breast biopsy. Nurs Res 44: 196-201, 1995[Medline] 30. Maunsell E, Brisson J, Deschenes L: Psychological distress after initial treatment of breast cancer: Assessment of potential risk factors. Cancer 70: 120-125, 1992[CrossRef][Medline] 31. Lev E, Paul D, Owen SV: Age, self-efficacy, and change in patients’ adjustment to cancer. Cancer Pract 7: 170-176, 1999[CrossRef][Medline] 32. Ell K, Nishimoto R, Mantell J, et al: Longitudinal analysis of psychological adaptation among family members of patients with cancer. J Psychosom Res 32: 429-438, 1988[CrossRef][Medline] 33. Ell K, Mantell JE, Hamovitch MB, et al: Social support, sense of control, and coping among patients with breast, lung or colorectal cancer. J Psychosoc Oncol 7: 63-89, 1989 34. Carey P, Oberst M, McCubbin M, Hughes S: Appraisal and caregiving burden in family members caring for patients receiving chemotherapy. Oncol Nurs Forum 18: 1341-1348, 1991[Medline] 35. McCubbin HI, Thompson AI, McCubbin MA: FHI: Family Hardiness Index, in Family Assessment: Resiliency, Coping and Adaptation—Inventories for Research and Practice. Madison WI, University of Wisconsin, 1996 36. Bloom JR: Social support, accommodation to stress and adjustment to breast cancer. Soc Sci Med 16: 1329-1338, 1982[CrossRef][Medline] 37. Pistrang N, Barker C: The partner relationship in psychological response to breast cancer. Soc Sci Med 40: 789-797, 1995[CrossRef][Medline] 38. Spiegel D, Bloom JR, Kraemer HC, et al: Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet 2: 888-891, 1989[CrossRef][Medline] 39. Koopman C, Hermanson K, Diamond S, et al: Social support, life stress, pain and emotional adjustment to advanced breast cancer. Psychooncology 7: 101-111, 1998[CrossRef][Medline] 40. Ferrell BR, Grant M, Funk B, et al: Quality of life in breast cancer: Part I. Physical and social well-being. Cancer Nurs 20: 398-408, 1997[Medline] 41. Ganz PA, Coscarelli A, Fred C, et al: Breast cancer survivors: Psychosocial concerns and quality of life. Breast Cancer Res Treat 38: 183-199, 1996[CrossRef][Medline] 42. Ganz PA, Rowland JH, Desmond K, et al: Life after breast cancer: Understanding women’s health-related quality of life and sexual functioning. J Clin Oncol 16: 501-514, 1998[Abstract] 43. Bull AA, Meyerowitz BE, Hart S, et al: Quality of life in women with recurrent breast cancer. Breast Cancer Res Treat 54: 47-57, 1999[CrossRef][Medline] 44. Given BA, Given CW, Helms E, et al: Determinants of family care giver reaction: New and recurrent cancer. Cancer Pract 5: 17-24, 1997[Medline] 45. Northouse LL, Caffey M, Deichelbohrer L, et al: The quality of life of African American women with breast cancer. Res Nurs Health 22: 449-460, 1999[Medline] 46. Compas BE, Worsham NL, Ey S, et al: When mom or dad has cancer: II. Coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol 15: 167-175, 1996[CrossRef][Medline] 47. Giese-Davis J, Hermanson K, Koopman C, et al: Quality of couples’ relationship and adjustment to metastatic breast cancer. J Fam Psychol 14: 251-266, 2000[CrossRef][Medline] 48. Northouse LL, Mood D, Oberst M: Breast Cancer Recurrence: Promoting Patient and Family Quality of Life. Atlanta GA, American Cancer Society, 1996 -1999
49. Brady MJ, Cella DF, Mo F, et al: Reliability and validity of the Functional Assessment of Cancer Therapy-Breast quality-of-life instrument. J Clin Oncol 15: 974-986, 1997 50. Ganz PA, Lee JJ, Sim M-S, et al: Exploring the influence of multiple variables on the relationship of age to quality of life in women with breast cancer. J Clin Epidemiol 45: 473-485, 1992[CrossRef][Medline] 51. Mood D, Bickes J: Strategies to enhance self-care in radiation therapy. Oncol Nurs Forum 16: 143, 1989 (suppl, abstr) 52. Worden JW: Psychosocial screening of cancer patients. J Psychosoc Oncol 1: 1-10, 1983[Medline] 53. Weisman AD, Worden JW: The existential plight in cancer: Significance of the first 100 days. Int J Psychiatry Med 7: 1-15, 1976[Medline] 54. Lewis FM: Family Home Visitation Study Final Report. Bethesda MD, National Cancer Institute, National Institutes of Health, 1996 55. McCubbin M, McCubbin H, Thompson A: Family Hardiness Index, in McCubbin H, Thompson A (eds): Family Assessment Inventories for Research and Practice. Madison WI, University of Wisconsin, 1987, pp 125-130 56. Brandt PA, Weinert C: The PRQ: A social support measure. Nurs Res 30: 277-280, 1981[Medline] 57. Weinert C: A social support measure: PRQ85. Nurs Res 36: 273-277, 1987[Medline] 58. Mood D, Northouse LL, Oberst M: Psychosocial Interventions in Radiation Therapy (grant no R01CA59013). Bethesda MD, National Cancer Institute, 1995 59. Munkres A, Oberst MT, Hughes SH: Appraisal of illness, symptom distress, self-care burden, and mood states in patients receiving chemotherapy for initial and recurrent cancer. Oncol Nurs Forum 19: 1201-1209, 1992[Medline] 60. Oberst MT: Appraisal of Illness Scale: Manual for Use. Detroit MI, Wayne State University, 1991 61. Oberst MT: Appraisal of Caregiving Scale: Manual for Use. Detroit MI, Wayne State University, 1991 62. Mishel M, Epstein D: Uncertainty in Illness Scales: Manual. Tucson AZ, University of Arizona, 1990 63. Beck AT, Weissman A, Lester D, et al: The measurement of pessimism: The hopelessness scale. J Consult Clin Psychol 42: 861-865, 1974[CrossRef][Medline]
64. Cella DF, Tulsky DS, Gray G, et al: The Functional Assessment of Cancer Therapy scale: Development and validation of the general measure. J Clin Oncol 11: 570-579, 1993 65. McHorney CA, Ware JE Jr, Raczek AE: The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 31: 247-263, 1993[Medline] 66. Anderson J, Gerbing D: Structural equation modeling in practice: A review and recommended two-step approach. Psychol Bull 103: 411-423, 1988[CrossRef] 67. Arbuckle JL: AMOS for Window: Analysis of Moment Structures (version 4.0). Chicago IL, Smallwaters, 1999 68. Jaccard J, Wan CK: Lisrel Approaches to Interaction Effects in Multiple Regression. Newbury Park CA, Sage, 1996 69. Byrne BM: A Primer of LISREL: Basic Applications and Programming for Confirmatory Factor Analytic Models. New York NY, Springer-Verlag, 1989 70. Loehlin JC: Latent Variable Models: An Introduction to Factor, Path, and Structural Analysis ( ed 3 ). Hilldale NJ, Lawrence Erlbaum, 1998 71. Hu L, Bentler PM: Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equation Model 6: 1-55, 1999 72. Browne MW, Cudek R: Alternative ways of assessing model fit, in Bollen K, Long JS (eds): Testing Structural Equation Models. Newbury Park CA, Sage, 1993, pp 136-162 73. Ware J, Kosinski M, Heller SD: SF-36 Physical and Mental Summary Scales: A User’s Manual. Boston MA, New England Medical Center, 1994 74. Roberts RL, Bengtson VL: Assessing familial and non-familial sources of parent-child attitude resemblance over two measurement occasions, in Horn JL, Collins L (eds): Best Methods for the Analysis of Change. Washington DC, American Psychological Association, 1991 75. Northouse LL, Templin T, Mood D, et al: Couples’ adjustment to breast cancer and benign breast disease: A longitudinal analysis. Psychooncology 7: 37-48, 1998[CrossRef][Medline] 76. Mood D: Head and neck cancer, in Varricchio C (ed): A Cancer Source Book for Nursing ( ed 7 ). Atlanta GA, American Cancer Society, 1997 77. Mood DW: Contingency contracting: An intervention strategy for moving patients from knowing to doing. Proceedings of the 21st Annual Conference of the Midwest Nursing Research Society, Minneapolis, MN, April 1997
78. Barg FK, Pasacreta JV, Nuamah I, et al: A description of a psychoeducational intervention for family caregivers of cancer patients. J Fam Nurs 4: 394-413, 1998 79. Braden C, Mishel M, Longman AJ: Self-help intervention project. Cancer Pract 6: 87-98, 1998[CrossRef][Medline] 80. Lewis FM: Behavioral research to enhance adjustment and quality of life among adults with cancer. Prev Med 26: S19-S29, 1997[CrossRef][Medline] Submitted February 13, 2001; accepted June 17, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
0.90 representing adequate fit.
