A Population-Based Study of the Quality of Life of Cancer Survivors and Their Family Caregivers

Suzanne Mellon, PhD, RN; Laurel L. Northouse, PhD, RN, FAAN; Linda K. Weiss, PhD

Disclosures

Cancer Nurs. 2006;29(2):120-131. 

In This Article

Results

The demographic characteristics of the sample are listed in Table 1 . In general, a wide range of educational levels and income levels were represented in this population-based sample. The average age of the cancer survivor was 65 and the average age of the family caregiver was 55. Most of the family members were either spouses (53%) or adult children (29%). Approximately half of the family dyads were white and half were African American.

In regards to the medical characteristics, the stratified sampling yielded approximately equal numbers of survivors in the 4 cancer sites, with 31 breast, 31 colon, 31 uterine, and 30 prostate patients. The types of treatment that survivors had undergone included surgery (88%), radiation (39%), chemotherapy (23%), and hormone therapy (3%). Nearly half of the prostate survivors had only radiation therapy. Most of the survivors (56%) had only 1 type of treatment (surgery or radiation), whereas one third of the patients had a combination of 2 treatments (surgery/radiation, surgery/hormone, surgery/chemotherapy, radiation/hormone). Eleven survivors had 3 types of treatment, and 4 survivors had a total of 4 types. The average time since diagnosis was 3.39 years, and that since completion of treatment was 3.01 years. Other health problems, such as heart disease, arthritis, or diabetes were reported by approximately 73% of survivors and 50% of family caregivers.

A Repeated Measure Multivariate General Linear Model with survivor-family caregiver dyad treated as a within-subject independent variable was conducted to assess whether or not survivors and their family caregivers differed on the major study variables. Multivariate results showed that survivors and family caregivers significantly differed on the set of variables, F6,117 = 6.68, P <.001. Follow-up univariate analyses were conducted to assess which variables differed (see Table 2 ). Univariate results indicated significant differences on quality of life, fear of recurrence, family hardiness, and social support.

In this population-based sample, cancer survivors reported significantly higher quality of life than their family caregivers (see Table 2 ). Cancer survivors also reported having significantly less fear of recurrence than did their family caregivers (see Table 2 ). To get a better understanding of why there were differences in the amount of fear reported by survivors and their family caregivers, we examined individual items on the Fear of Recurrence Scale and found that family caregivers reported significantly more fear on items pertaining to worrying and uncertainty about the survivor's future health status.

Cancer survivors also reported higher levels of family hardiness and higher levels of social support than did their family caregivers. No differences, however, were found in the amount of family stressors or the meaning of illness reported by survivors and family caregivers in this study. There was considerable variation in the number of family stressors reported, with some survivors and family caregivers reporting many family stressors, whereas others reported very few.

Several demographic variables were related to the cancer survivor's quality of life. Survivors who had older caregivers (r = 0.19, P <.05), had a higher family income (r = 0.25, P<.05), were married (r = -0.38, P <.01), or were male (r = -0.22, P <.05) reported higher quality of life. Survivors who had spouses as the primary caregiver reported higher quality of life (F3, 119 = 4.87, P <.001) than those survivors who identified adult children. Male survivors also reported significantly higher quality of life (M = 143.88) than female survivors (M = 134.01), t = 2.49, P <.01. Survivors who were retired (F2, 120 = 6.18, P <.01) or who had family members who were retired (F2, 120 = 3.60, P <.05) reported higher quality of life than those who were employed or homemakers. Because the sample was stratified by race, we also examined if quality of life differed by race. No significant difference in quality of life was found between white and African American cancer survivors.

Analysis of medical characteristics and quality of life revealed only 2 significant differences. Survivors who did not have other health problems, such as heart disease or arthritis, reported higher quality of life (r = 0.19, P <.05) than survivors with other health problems. The type of treatment that survivors received initially also made a significant difference on survivors' long-term quality of life. Survivors who received only radiation reported the highest quality of life, whereas those survivors who received a combination of treatments reported the lowest quality of life, F2, 120 = 4.37, P< .05. The type of cancer (breast, colon, prostate, uterine), time since diagnosis, or time since completion of treatment were not significantly related to quality of life. In addition, no interaction effect was found between the disease site and the time since treatment.

Analyses were conducted to determine the interrelationships between the major study variables (see Table 3 ). Several variables showed moderate correlations without evidence of collinearity. All major study variables were significantly related to quality of life. To ascertain the possibility of item content overlap, relationships between major study variables and quality of life subscales were further examined and showed low to moderate correlations across several subscales. Cancer survivors who had less somatic concerns, less family stressors, and less fear of recurrence had higher quality of life. Concomitantly, those survivors with higher family hardiness and social support and a more positive meaning of the cancer illness also had higher quality of life.

Multiple regression was used to determine the amount of variance that the major study variables contributed to the cancer survivors' quality of life (see Table 4 ). Only those demographic and medical variables significantly related to quality of life were used in this analysis, and they were entered first to control their influence on quality of life. The major study variables were entered next as a block. Because meaning of illness was considered a potential mediator in the model, it was entered last in the equation. The multiple regression equation accounted for 54% of the variance in survivors' quality of life, with family stressors, social support, meaning of the illness, and survivor employment status each making unique contributions (see Table 4 ).

Demographic and medical characteristics also were analyzed to determine their relationships with family caregivers' quality of life. Similar to the cancer survivor, family caregivers who were older (r = 0.50, P <.01), had higher family income (r =0.31, P <.01), were married (r = -0.35, P <.01), and were male (r = -0.19, P <.05) reported higher quality of life. About the type of relationship to the survivor, those family caregivers who were spouses reported higher quality of life (F3, 119 = 9.69, P <.001) than those family caregivers who were adult children to the survivor. Male family caregivers reported significantly higher quality of life (M = 139.23) than female family caregivers (M = 129.93), t = 2.12, P = .03. In addition, family caregivers who were retired had higher quality of life (F2, 122 = 9.42, P <.001) than family caregivers who were employed or homemakers. No significant relationship was found between family caregivers who had other health problems and quality of life, and no differences were found in quality of life of family caregivers according to race (white and African American).

Correlation analyses were conducted to examine the interrelationships between the major study variables and the family caregiver's quality of life (see Table 5 ). With the exception of the survivor's somatic concerns, all other variables showed significant relationships with quality of life. To explore possible item content overlap between major study variables and subscales of quality of life, additional analysis showed low to moderate relationships of the quality of life subscales with major study variables and no evidence of collinearity. Family caregivers who reported fewer family stressors and less fear of recurrence had higher quality of life. Conversely, those family caregivers who reported higher family hardiness, more social support, and a more positive meaning of the illness also reported higher quality of life.

Multiple regression was used to determine the amount of variance that the major study variables contributed to family caregivers' quality of life (see Table 4 ). Only those demographic variables that were significantly related to quality of life were entered, followed by the major study variables, which were entered as a block, with the last entry being the potential mediator, meaning of the illness. These sets of variables accounted for 56% of the variance in family caregivers' quality of life scores. Fear of recurrence and social support each made a significant contribution to the family caregivers' quality of life.

The third purpose of the study was to examine the interrelationships between cancer survivors' and family caregivers' quality of life. Using a Pearson r correlation coefficient, a significant relationship was found between survivor and family caregiver's quality of life (r = 0.55, P <.001). When cancer survivors reported higher quality of life, family caregivers also reported higher quality of life.

Multiple regression was used to determine if any additional variance in the cancer survivors' quality of life could be accounted for by entering the family caregivers' quality of life score into the regression analysis reported in Table 4 . Results indicated that the quality of life of the family caregiver made a significant, independent contribution to the quality of life reported by the cancer survivor (t = 3.71, P =.001), raising the amount of variance accounted for to 61%, an increase of 7%. In the final step of this regression analysis, social support (t = 4.01, P =.001) and survivor employment status (t = 2.76, P =.007) continue to make unique contributions to survivors' quality of life. However, family stressors and meaning of the illness no longer make a significant independent contribution when the caregivers' quality of life score is added to the regression equation.

Similar regression analyses were conducted with family caregiver data to determine if any additional variance in caregivers' quality of life could be accounted for by adding the cancer survivors' quality of life scores to caregivers' regression equation. Results indicated that the cancer survivors' quality of life did make a significant, unique contribution to family caregivers' quality of life (t = 2.98, P =.004), increasing the overall variance accounted for by 4% to 60%. The final regression equation indicated that fear of recurrence (t =-2.20, P = .03), social support (t = 2.16, P =.03), family caregiver age (t = 2.27, P =.03), and cancer survivor's quality of life (t = 2.98, P =.004) each made a unique contribution to the overall variance in family caregiver's quality of life.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....