QUALITY OF LIFE EVALUATIONS OF CAREGIVERS OF OVARIAN CANCER PATIENTS DURING CHEMOTHERAPY TREATMENT

Transcription

1 QUALITY OF LIFE EVALUATIONS OF CAREGIVERS OF OVARIAN CANCER PATIENTS DURING CHEMOTHERAPY TREATMENT T. Le, MD,I A. Leis, PhD,2 P. Pahwa, PhD,2 K.Wright, PhD,2 K.AIi, MD,2 B. Reeder, MD,2 M. Kinderchuk, LPN,l K. Ward, LPNl I Department of Obstetrics and Gynaecology, Division of Gynaecologic Oncology, University of Ottawa, Ottawa ON 2Department of Community Health and Epidemiology, University of Saskatchewan, Saskatoon SK 3Department of Oncologic Nursing, Saskatoon Cancer Centre, Saskatoon SK Abstract Objectives: (I) To examine the quality of life (QOL) of caregivers of women undergoing chemotherapy for advanced ovarian cancers, and (2) to correlate the QOL measures of caregivers to those of the women undergoing chemotherapy. Methods: Over a 9-month period, all women undergoing chemotherapy for ovarian cancer at the Saskatoon Cancer Centre, and their caregivers, were offered participation. Two well-validated instruments were used to measure the "quality of life" concept. Women with ovarian cancer completed the Functional Assessment of Cancer Therapy - Ovarian (FACT-O) questionnaire in the clinic prior to each course of chemotherapy. Each caregiver completed a Caregiver Quality of Life Index - Cancer (CQOL-q questionnaire at home at the start and conclusion of each chemotherapy regimen (cycle I and after the last chemotherapy treatment). The demographics of the caregivers were described. A paired t test was used to detect changes to caregivers' QOL scores before and after chemotherapy treatment. Correlation analysis was carried out to examine the relationship between the caregivers' total QOL scores and the various subscale and total scores of the FACT-O questionnaires completed by the women with ovarian cancer. Multivariate regression models were constructed to examine the relative importance of each of the QOL domain measures of the woman with cancer in predicting the effect on her caregiver's QOL. Results: Thirty different patient-caregiver pairs participated in the study, providing 50 separate assessments since not all pairs had completed the post-chemotherapy assessments. There was improvement (P <.05) in the caregiver's QOL scores at the conclusion of the chemotherapy treatment compared to the baseline assessments. The improvement was unrelated to the performance status or response to chemotherapy of the woman undergoing treatment. There was also a correlation (P <.05) between an increase in a caregiver's distress and worsening scores in the "emotional;' "functional;' and "concerns" QOL domain assessments of the woman undergoing treatment. Stepwise regression analysis showed the "concerns" KeyWords Ovarian neoplasms, caregivers, quality of life Competing interests: None declared. Received on August 7, 2003 Revised and accepted on November 23, 2003 score, measuring specific ovarian cancer-related symptoms, to be the only predictor of a caregiver's distress (P <.05). Conclusions: Standard chemotherapy for ovarian cancer does not worsen a caregiver's QOL. There is a direct relationship between the QOL of women with cancer and that of their caregivers. Future research is required to identify how best to integrate the results of QOL assessments in cancer treatment protocols and to examine the long-term effects of ovarian cancer and its treatment on both caregivers and the women for whom they care. Resume Objectifs : (I) Etudier la qualite de vie (QDV) des fournisseurs de soins dont les patientes subissent une chimiotherapie pour Ie traitement d'un cancer de I'ovaire avance; (2) etablir une correlation entre les mesures de QDV des fournisseurs de soins et celles des patientes subissant une chimiotherapie. Methodes : Au cours d'une peri ode de 9 mois, au Saskatoon Cancer Centre, toutes les patientes subissant une chimiotherapie pour Ie traitement d'un cancer de I'ovaire, ainsi que leurs fournisseurs de soins, se sont vu offrir I'occasion de participer a I'etude. Deux instruments bien valides ont ete utilises pour mesurer Ie concept de «qualite de vie». Les femmes atteintes d'un cancer de I'ovaire ont rempli Ie questionnaire Funaional Assessment of Cancer Therapy - Ovarian (FACT-O) a la clinique avant chaque seance de chimiotherapie. Chaque fournisseur de soins a, quant a lui, rempli Ie questionnaire Caregiver Quality of Life Index - Cancer (CQOL-q a la maison, et ce, au debut et a la fin de chaque traitement de chimiotherapie (cycle I et apres la derniere seance de chimiotherapie). Les caracteristiques demographiques des fournisseurs de soins ont ete dec rites. Un test T jumele a ete utilise pour deceler les modifications subies par les scores de QDV des fournisseurs de soins avant et apres la chimiotherapie. Une analyse de correlation a ete menee, en vue d'etudier la relation entre les scores totaux de QDV des fournisseurs de soins et les divers scores de sous-categorie et totaux obtenus a I'aide des questionnaires FACT-O remplis par les femmes atteintes d'un cancer de I'ovaire. Des modeles de regression multivaries ont ete construits afin d'etudier I'importance relative de chacune des mesures de domaine de QDV (obtenues chez la patiente atteinte d'un cancer) en matiere de prediction de leurs effets sur la QDV du fournisseur de soins de la patiente en question. Resultats : Trente paires «patiente-fournisseur de soins» differentes ont participe a I'etude, ce qui a permis I'obtention de 50 evaluations distinctes (puisque ce ne sont pas toutes les loge 2004

2 paires qui ont rempli les questionnaires post-chimiotherapies). Une amelioration (P < 0,05) a ete constatee en ce qui a trait aux scores de QDV du fournisseur de soins a la fin du traitement de chimiotherapie, par comparaison avec les evaluations de depart. Cette amelioration n'etait pas liee au rendement ni a la reaction de la patiente a la chimiotherapie. Une correlation (P < 0,05) a egalement ete constatee entre la hausse de la detresse du fournisseur de soins et la deterioration des scores de la patiente dans les domaines de QDV suivants : emotional, functional et concerns. Une analyse de regression par degres a indique que Ie score concerns, lequel mesure les symptomes particulierement lies au cancer de I'ovaire, etait Ie seul predicteur de la detresse d'un fournisseur de soins (P < 0,05). Conclusions : La chimiotherapie standard pour Ie traitement du cancer de I'ovaire n'entraine pas une deterioration de la QDV des fournisseurs de soins. II existe une relation directe entre la QDV des patientes atteintes d'un cancer et celie de leurs fournisseurs de soins. D'autres recherches s'averent necessaires pour identifier la meilleure fa~on d'integrer les resultats des evaluations de la QDV aux protocoles de traitement du cancer, ainsi que pour etudier les effets a long terme du cancer de I'ovaire et de son traitement sur les fournisseurs de soins et leurs patientes. INTRODUCTION J Obstet Gynaecol Can 2004;26(7): Epithelial ovarian cancer is the second most common gynaecologic cancer in North America and the fifth most common cause of cancer death in women. I,2 Women with ovarian cancer are living longer with better treatment. 3 The burden that the diagnosis and treatment of cancer places on close family members has been recognized since the early 1980s. A diagnosis of advanced cancer invariably causes significant disturbances to the lives of both the patients diagnosed and their family caregivers.4-9 Caring for an individual with cancer is a stressful experience that can have major emotional and physical impacts on caregivers. The impact of the cancer and its treatments on caregivers has been poorly studied, even though each patient's overall sense of well-being is believed to be largely affected by her psychosocial support base. As most aspects of general cancer care are being shifted towards outpatient management, there is an increasing reliance on family, community, and social services resources to care for the cancer patient. These responsibilities commonly include emotional support, helping with activities of daily living, administration of medications, provision of a special nutrition plan, and assistance with other physical aspects of care. I 0 Caregivers often are asked to provide care even when their physical, emotional, social, and financial status are not adequate. Attention to the well-being of caregivers, and provision of adequate direct supportive services for them, must become an essential component of comprehensive cancer care. The goals of this study were to assess the quality of life (QOL) of caregivers of women undergoing chemotherapy for advanced ovarian cancer, and to correlate QOL assessments of the women with cancer to that of their caregivers. METHODS This study was carried out by the Gynecologic Oncology service based at the Saskatoon Cancer Centre and the Royal University Hospital, Saskatoon, Canada. Over a 9-month period, a prospective cohort of women diagnosed with advanced or recurrent ovarian cancers requiring chemotherapy was recruited to participate in this study, together with their caregivers. Each woman was asked to identify 1 person to be her "immediate caregiver," from whom she would expect to derive the most support during her chemotherapy. Well-validated questionnaires were used to measure Quality of Life of the caregivers ll and patients. 12 QOL assessments of the caregivers were performed using the Caregivers Quality of Life - Cancer (CQO L-C) questionnaire, II at the beginning (cycle 1) and the conclusion of each chemotherapy regimen (usually cycle 3 if disease progression was encountered or cycle 6 if response to treatment was observed). The CQOL-C questionnaire consisted 005 items to be assessed for QOL using a 5-point Likert scale ranging from 0 (not at all) to 4 (very much). The questions were designed to measure the emotional, functional, and social impact of disease and its treatment on the caregivers. A global score was obtained by summing all 35 items after appropriate recoding. The Functional Assessment of Cancer Therapy - Ovarian (FACT-O) questionnaire l2 was used to assess prior to each chemotherapy cycle the QOL of women undergoing treatment. Paired assessments of each woman and her caregiver were made at 2 points during each chemotherapy regimen: once at cycle 1 and once more at the time of the last chemotherapy cycle (usually cycle 3 or 6 at the time of tumour response assessments). The FACT-O questionnaire is a standardized self-administered questionnaire measuring all aspects of QOL as related to women with ovarian cancers. The questionnaire is relatively short and can be easily completed without assistance in 5 minutes by a person with a sixth grade reading level. Through a series of7 to 12 questions, assessments were made of several domains: physical (such as lack of energy, nausea, and pain), social (such as support from family and friends, family acceptance of illness, and closeness to partner), emotional (such as sadness, nervousness, and worry that condition will worsen), functional (such as ability to work and enjoy life, sleeping pattern, and acceptance of illness), and specific concerns about ovarian cancer (such as nausea, ascites, and loss of femininity and fertility). All questions used a Likert-type scale ranging from 0 (not at all) to 4 (very much). For each chemotherapy regimen, the QOL variables of interest were the total FACT-O score, obtained by summing all the QOL domains' subscales of physical well-being, sociavf.unilywell-being,

3 emotional well-being, functional well-being, and "concerns" as related to ovarian cancer, as well as each specific QOL domain score after appropriate recoding. The standard chemotherapy protocol at the Saskatoon Cancer Centre during the study time period was platinum-based for initial treatment and for those women with a progressionfree interval of more than 6 months. In the event of disease progression on the platinum-based regimen, liposomal doxorubicin, topotecan, and cispiatinumnp16 combinations were used as salvage chemotherapy as appropriate, pending drug availability and convenience of administration. Descriptive statistics were used to summarize caregivers' demographics including age, gender, occupation, self-rated health, and estimated time spent daily with the patient. Performance starus of the woman undergoing treatment and her response to the current chemotherapy regimen were also recorded. An overall score from the CQOL-C questionnaire was computed after appropriate recoding, with a higher score reflecting a poorer QOL. The difference in caregivers' QOL scores at the beginning and conclusion of each chemotherapy session was computed when data was available. A paired t test was used to detect significant changes to the caregivers' QOL scores during the course of chemotherapy treatment. With respect to chemotherapy response, the direction of the change in the caregiver's QOL score was examined only in the 13 patient-caregiver pairs with available assessments during both pre- and post-chemotherapy visits, using the Fisher exact test. Correlation analysis was carried out to examine the relationship between all available caregivers' overall QOL scores and the various subscale and total scores of the FACT-O questionnaires. A multivariate regression model was constructed to examine the relative importance of each of the QOL domain measures of the woman undergoing treatment in relation to her caregiver's QOL. In the proposed model, the caregiver's overall QOL score was the dependent variable, and the corresponding patient-related QOL domain-specific scores (physical, social, emotional, functional, and concerns) on the FACT-O questionnaire were the predictor variables. Stepwise regression analysis was used to identify the most parsimonious model. All statistical significance was set at P <.10 in consideration of the exploratory nature of the project. RESULTS There were 30 different patient-caregiver pairs providing 50 separate assessments, due to post-chemotherapy data not being available for all pairs. Data for pre- and post-chemotherapy caregivers' assessments were available for 13 patient-caregiver pairs, due to the premature termination of this study due to logistic reasons. Seventy percent of the women were undergoing treatment after their primary diagnosis or first recurrence of cancer, and 30% had been treated with 2 or more chemotherapy regimens in the past. Seventy-three percent of the women had either stage III or N disease. All had a performance status of between 0 and 1. The mean ± SD caregiver age was ± years (range, years). Of the 30 caregivers assessed, 28 were men, with half of them reponed to be currently working full-time. Twentyone of these caregivers were spouses of the women in therapy. Of the 2 caregivers who were women, 1 was a first-degree relative (daughter) and the other was a close friend of the family. Twenty-nine of the caregivers rated their own health as being good to fair. Thiny percent reponed spending between 6 and 12 hours each day with the woman under their care, and another 57% reported spending more than 12 hours caring for the woman undergoing treatment. When comparing the caregiver total CQOL-C score at the conclusion of chemotherapy to the baseline score assessments for the 13 patient-caregiver pairs for whom pre- and postchemotherapy data were available, a decrease in total caregiver CQOL-C score at the end of the chemotherapy regimen suggested an improvement in the QOL of 10 pairs. The other 3 pairs experienced an increase in total scores at the conclusion of chemotherapy. There was no significant relationship between how the woman responded to chemotherapy, change in performance status, and other patient- and disease-related factors, with respect to the change in the caregivers' scores. Of the 10 patient-caregiver pairs with improvement in the caregivers' QOL score, 3 of the women undergoing treatment had a complete response to chemotherapy, 3 had a partial response to chemotherapy, 3 had stable disease, and 1 had progressive disease during treatment. Of the 3 patient-caregiver pairs with deterioration in QOL of the caregivers, all treated women had partial response to chemotherapy with no change in their performance status. There was no association between the tumour response status (complete or partial vs stable or progression) and change in score (increase vs decrease) (P =.497). A significant improvement in the caregiver QOL scores during the course of chemotherapy treatment was demonstrated by the mean ± SD caregiver score post-chemotherapy (mean of24.23 ± 10.55) being lower than the mean ± SD score at the baseline of29.08 ± (P =.007). This might reflect the improved coping abilities of caregivers. Whether this change is clinically significant will require further in-depth studies. To study the effect of perceived QOL of treated women on their caregivers, an initial bivariate correlation analysis was performed on all 50 observations. As expected, the total FACT-O score was significantly and positively associated with the caregivers' global QOL score, with a Pearson product moment correlation coefficient of (P=.015). Furthermore, specific FACT-O domain scores of emotional (Pearson r = 0.383; P =.008), functional (Pearson r = 0.332; P =.023), and concerns (Pearson r = 0.457; P =.001) were all significantly positively associated with the caregivers' global QOL score. Regression modelling using data on all 50 assessments showed JULY 2004

4 a number of patient-related QOL domain scores positively associated with the caregivers' QOL: physical (P=.082), emotional (P =.064), and concerns (P =.025). Using a stepwise elimination approach, however, only the "concern" domain score remained as the most significant predictor of the caregivers' QOL score (P=.001). Further, the regression model with "concern" score as the sole predictor explained only 21 % of the variance in the prediction of the CQOL-C total score. DISCUSSION In the current study, 93% of caregivers were the husbands of the women undergoing treatment, and 97% of caregivers reported to be in a fair to good health status. These percentages likely reflect how women with cancer decide who should be their "immediate caregiver." Most caregivers (77%) experienced improvements in their QOL at the conclusion of the chemotherapy regimen compared to the base line. Lundh reviewed the sources of satisfaction and coping strategies used among Swedish caregivers, and, consistent with our findings, found that despite the very demanding nature of the caregiving process, the majority of caregivers experienced some feelings of reward. The sources of satisfaction that developed during the caregiving process were numerous and diverse. 13 Newly developed coping mechanisms to help caregivers with the stress of caring for persons undergoing cancer treatment could account for this observation. Caregivers often employed a complex range of coping behaviours, in response to the stress encountered. 14 The feelings of satisfaction may also result ftom an indirect positive effect of chemotherapy on the physical, emotional, and functional well-being of the women undergoing treatment. There was a significant direct correlation between the QOL scores of the caregiver and the woman undergoing treatment. The domains of "emotional," "functional," and "concern" were identified to have significant positive correlations with distress in the caregiver. This finding suggests that emotional distress together with an increase in cancer-related symptoms will increase the dependency of the woman on her caregiver, causing significant increase in stress and worsened QOL for the caregiver. Other studies have also documented similar findings Northhouse et al. 17 reported that self-efficacy (ability to care for oneself), social support, and family hardiness had a positive effect on caregivers' QOL, whereas patients' symptom distress, concerns, hopelessness, and negative appraisal of the illness or the care given had detrimental effects on caregivers of women with recurrent breast cancers.17 In an effort to improve the well-being of caregivers, more attention should be directed to improve the related QOL domains of the women undergoing treatment. Increased attention to the provision of emotional and social support for the caregivers, in the form of support services that can provide help with the physical aspects of caregiving tasks, would also help to lessen the social and emotional impact of caring. The caregiver's own emotional responses to the patient's diagnosis and prognosis should be carefully studied and incorporated into the care plan. The need for accurate and understandable information on the disease and treatment, as well as instruction in technical and care skills at home, can lessen the caregiver's anxiety prior to the patient being discharged from the hospital. Research has suggested that interventions designed to teach caregivers to become proficient in the physical and psychological aspects of patient care will benefit both patients and caregivers. 18 The health-care system should regard the caregiver as a partrler in the patient-care plan by providing specific instructions and guidance to caregivers as they assume this role. 19 In the final model to predict caregiver distress, only the "concern" domain score achieved significant predictor status with a positive regression parameter estimate. This information could be used to predict the stress that the caregiver is under when the woman with cancer is assessed in the clinic, even if the caregiver is not present. It is not uncommon for the focus to be only on the woman experiencing symptoms due to advanced cancer. Supportive services should be extended to caregivers and the other family members. There were certain limitations with the current study. In the comparison of caregivers' QOL before and after patients finished their chemotherapy, data were available for only 13 of the 30 patient-caregiver pairs. Selection bias could account for the score difference being significant. A comparison between the patient-caregiver pairs with and without pre- and postchemotherapy data, however, did not show any significant differences in demographic characteristics. Because the amount of variance in the caregiver QOL score explained by the "concern" FACT-O domain score was only 21 %, there must have been many other factors that were not studied in this project that could have influenced the caregivers' well-being, such as their age, social support, religious belief, socioeconomic status, and personality, as well as unique coping styles. Due to the premature termination of this study and the small sample size, these factors could not have all been incorporated into the design and analysis. These factors should be explored further in future qualitative and quantitative studies. As timely, active, and aggressive symptom control interventions can greatly benefit the QOL of the woman and her family, it is appropriate to include palliative care and social work service early in the care of women with ovarian cancer. It is all too common that the caregiver's distress is not addressed and goes unnoticed by the health-care team. 20 Caregivers might not have as much contact with the team as the women being treated, and since most of the caregivers are male, they might not readily disclose their concerns and suffering to health-care professionals. Attempts should be made to incorporate caregivers and family members into the treatment planning process with goals to address all their major concerns. JULY 2004

5 A last potential weakness of the study is the instruments that were used to measure QOL. The component of spirituality in both the women and their caregivers was not captured in assessing the overall QOL and may have played an important role. Furthermore, the caregivers' questionnaire was not broken down into different domains, leading to possible loss of sensitivity in the detection of changes in these specific areas. Further qualitative and quantitative research should identify important areas in the assessment of caregivers' QOL that would allow for further refinements of these instruments. CONCLUSIONS For women with ovarian cancer, chemotherapy treatment does not appear to worsen their caregivers' QOL. There was a direct association of the QOL scores of the women with ovarian cancer and their caregivers. Only the combination of qualitative and quantitative research can be expected to provide a full assessment of the psychological impacts of treatment and disease on patients and caregivers. As it is expected that further prolongation of survival will be obtained in women with ovarian cancer, further study of the long-term effects of cancer and its treatment on patients and their caregivers is essential. REFERENCES I. Yancik R. Ovarian cancer: age contrasts in incidence. histology. disease stage at diagnosis. and mortality. Cancer 1993;71 (Suppl 2): American Cancer SOCiety. Cancer facts and figures Atlanta (GA): American Cancer Society; Barnholtz-Sloan JS, Schwartz AG. Qureshi F,Jacques S, Malone J, Munkarah AR. Ovarian cancer: changes in patterns at diagnosis and relative survival over the last three decades. Am J Obstet Gynecol 2003; 189: I Sabo D. Men, death, and anxiety: critical feminist interpretations of adjustment to mastectomy. In: Clark C, Fritz J, Rieder P, editors. Clinical sociological perspectives on illness and loss. Philadelphia (PA): Charles Press; p Lewis FM. Strengthening family support. Cancer 1990;65: Nijboer C,Triemstra M,Tempelaar R, Sanderman R, van den Bos GA. Determinants of caregiving experiences and mental health of partners of cancer patients. Cancer 1999;86(4): Weitzner MA, McMillan SC,Jacobsen PB. Family caregiver quality of life: differences between curative and palliative cancer treatment settings. J Pain Symptom Manage 1999; 17(6): Smeenk FW, de-witte LP, van-haastregt JC, Schipper RM. Transmural care of terminal cancer patients: effects on the quality of life of direct caregivers. Nurs Res 1998;47(3): McMillan Sc. Quality of life of primary caregivers of hospice patients with cancer. Cancer Pract 1996;4(4): Laizner AM,Yost LMS, Barg FK, McCorkle R. Needs of family caregivers of persons with cancer: a review. Semin Oncol Nurs 1993;9: II. Weitner MA,Jacobson PB.Wagner H Jr.The Caregiver Quality of Life Index - Cancer (CQOL-C) scale: development and validation of an instrument to measure quality of life of the family caregiver of patients with cancer. Qual Life Res 1999;8: Basen-Engquist F, Bodurka-Bevers D, Fitzgerald MA. Reliability and validity of the functional assessment of cancer therapy: ovarian. J Clin Oncol 2001 ;19(6): Lundh U. Family carers 2: sources of satisfaction among Swedish carers. Br J Nurs 1999;8(10): Lundh U. Family carers 3: coping strategies among family carers in Sweden. Br J Nurs I 999;8( I 0): Oberst MT,Thomas SE, Gass KA,Ward SE. Caregiving demands and appraisal of stress among family caregivers. Cancer Nurs 1989; 12: Carey PJ, Oberst MT, McCubbin MA, Hughes SH. Appraisal and caregiving burden in family members caring for patients receiving chemotherapy.oncol Nurs Forum 1991; 18: Northhouse LL. Mood D. Kershaw T, Schafenacker A, Mellon S, Walker J, et al. Quality of life of women with recurrent breast cancer and their family members.j Clin OncoI2002;20(19): McCorkle R, Pasacreta JY. Enhancing caregiver outcomes in palliative care. Cancer Control 200 I ;8( I ): Given BA, Given CWO Kozachik S. Family support in advanced cancer. CA Cancer J Clin 200 I ;51 : Soothill K, Morris SM, Harman JC.lnformal carers of cancer patients: what are their unmet psychosocial needs. Health Soc Care Community 200 I ;9(6): JULY 2004