Psycho-Oncology Psycho-Oncology 23: 1014–1020 (2014) Published online 3 April 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pon.3523
Social support and health-related quality of life in women with breast cancer: a longitudinal study Janni Leung1*, Nancy A. Pachana2 and Deirdre McLaughlin1 1 2
The University of Queensland, School of Population Health, Brisbane, Queensland, Australia The University of Queensland, School of Psychology, Brisbane, Queensland, Australia
*Correspondence to: School of Abstract Population Health, The University Objectives: A breast cancer diagnosis is a distressing event that impacts on physical and psychological of Queensland, Herston Rd, functioning. This study examined the longitudinal relationships among a diagnosis of breast cancer, Herston, Brisbane, Queensland social support, and health-related quality of life (HRQOL). 4006, Australia. E-mail: j. Methods: Participants were 412 women from the 1946–1951 birth cohort of the Australian [email protected] Longitudinal Study on Women’s Health who self-reported a new diagnosis of breast cancer between 1998 and 2007. The three surveys of longitudinal data analyzed included data 3 years before diagnosis, at diagnosis (baseline), and 3 years after diagnosis (follow-up). Social support was measured using the 19-item Medical Outcomes Study Social Support Survey; HRQOL was measured using the Medical Outcomes Study 36-item Short-Form Health Survey. Results: Compared with pre-diagnosis HRQOL, women newly diagnosed with breast cancer reported significantly poorer HRQOL in subscales related to pain, physical functioning, and health and vitality. At 3-year follow-up, HRQOL had improved in most domains to levels consistent with pre-diagnosis. Levels of social support remained stable across time. The structural equation model showed that social support was positively predictive of better physical and mental HRQOL at 3-year follow-up. Conclusions: Longitudinal analyses indicate that social support appears to be an important predictor of HRQOL in women diagnosed with breast cancer. In particular, positive emotional and informational support that may normally be provided by a partner is important in maintaining HRQOL. Identification of those lacking social support, especially patients without partners, will enable them Received: 26 July 2013 to be guided to appropriate support networks and programs. Revised: 25 February 2014 Copyright © 2014 John Wiley & Sons, Ltd. Accepted: 26 February 2014
Introduction Breast cancer is the second most prevalent cancer among women and is one of the leading causes of death in women worldwide [1]. Despite this, advances in screening and diagnostic procedures mean that most breast cancers can be diagnosed at an early stage, leading to effective treatments and improved survival rates [2]. Nonetheless, a diagnosis of breast cancer is a distressing event that affects physical and psychological functioning and impacts on lifestyle and relationships with family and friends [3–5]. Breast cancer treatments, including systemic adjuvant therapy, lead to long-term worsened quality of life in breast cancer survivors [5]. Furthermore, treatment-related changes to women’s physical appearance may affect self-image, and other effects of treatment, including fatigue, nausea, and pain, may decrease opportunities for social interactions and limit the ability of women with breast cancer to continue their normal lifestyle and social activities [6]. This in turn curtails their opportunities to maintain supportive social networks and to receive appropriate emotional and instrumental support. Copyright © 2014 John Wiley & Sons, Ltd.
Social support has been identified as being protective for health and in particular for reducing cancer-related distress [7–9]. Studies have found that social support is associated with better adjustment to disease and better quality of life; however, the subjective appropriateness of the support offered is important [10,11]. Not all social relationships are sources of support; although women possess extensive and robust social networks, these can come at a price (such as maintaining the networks and managing expectations of providing support to others) [10,12]. The stress-buffering model of social support focuses on the perceived availability, quality, and appropriateness of support [13]. In particular, this model suggests that when confronted with a situation of high stress, the availability of appropriate levels of subjective support is more important than the number of contacts in the social network. Perceived social support in health-related contexts has been categorized as (a) emotional/informational support, (b) tangible support, (c) affectionate support, and (d) positive social interaction [14]. The provision of satisfactory emotional and informational support has been associated with better health and lower symptom intensity in women
Social support, quality of life, and breast cancer
with breast cancer [15]. Both tangible support, in the form of a psycho-educational program, and affectionate support, from family and friends, have been associated with better well-being and improved health-related quality of life (HRQOL) among breast cancer patients [16–18]. Although previous studies have reported that positive social support can enhance adjustment to breast cancer and improve HRQOL [18–20], these studies are largely cross-sectional. Little evidence has been presented that examines either the longitudinal impact of social support on HRQOL in women with breast cancer or changes over time in social support and HRQOL throughout the course of the disease. A small number of previous studies that have reported on patterns of social support associated with breast cancer have found that levels of social support fluctuated over time [21,22]. However, these studies did not have information on pre-diagnosis levels of social support, and analyses were based on short follow-up times (up to 6 months). Given the increasing incidence and higher survival rate from breast cancer in recent years, a focus on factors associated with HRQOL, especially from a longitudinal perspective, can provide valuable insight to inform the development of strategies to support women during this distressing experience. The aim of this study is to examine the impact of a breast cancer diagnosis on social support and quality of life over time. Additionally, this paper will examine the role of social support as a predictor of better HRQOL in women newly diagnosed with breast cancer. Consistent with previously reported cross-sectional research, it is expected that women with high levels of social support will report higher HRQOL scores.
Method Participants The Australian Longitudinal Study on Women’s Health (ALSWH) is a prospective study of the health and wellbeing of three cohorts of women who were aged 18–23 years (1973–1978 birth cohort), 45–50 years (1946–1951 birth cohort), and 70–75 years (1921–1926 birth cohort) when recruited in 1996. The study sample was selected randomly from the Medicare Australia database (which covers all citizens and permanent residents of Australia, including refugees and immigrants), with intentional oversampling of women living in rural and remote areas in order to achieve an adequate sample size in this population. The project uses mailed questionnaires to collect self-report data on health and related variables every 3 years. Details of the ALSWH recruitment and study design have been fully described elsewhere [23], and details of the study can be found at www. alswh.org.au. Participation is voluntary, and the ALSWH is approved by the University of Newcastle and University of Queensland Human Research ethics committees. Copyright © 2014 John Wiley & Sons, Ltd.
1015
Data for these analyses are drawn from surveys 1 to 6 (1996 to 2010) of the 1946–1951 birth cohort. Because of uncertainties about the accuracy of the Medicare Australia database, response rates to the initial invitation cannot be exactly specified. However, it has been estimated that 53–56% of women in this birth cohort responded with 14,072 women responding to survey 1 [24]. Of the total eligible women at each survey, responses to subsequent surveys were 12,338 (90.7%) at survey 2 (1998), 11,226 (84.3%) at survey 3 (2001), 10,905 (84.0%) at survey 4 (2004), 10,638 (83.8%) at survey 5 (2007), and 10,011 (83.3%) at survey 6 (2010). Participants were women who responded ‘yes’ to the following question in survey 2 (1998), survey 4 (2004), or survey 5 (2007): ‘In the past three years, have you been diagnosed or treated for breast cancer’. A baseline breast cancer diagnosis was a positive response at any of the three surveys. This self-report item has been validated against Cancer Registry data with 93% sensitivity and 98% specificity [25]. Pre-diagnosis data were drawn from the survey immediately preceding the baseline (breast cancer diagnosis) survey, and follow-up data were drawn from the survey immediately after the baseline survey. A total of 519 women reported a new diagnosis of breast cancer between 1998 and 2007: 82 in 1998, 199 in 2004, and 238 in 2007. There were 433 women (83%) who completed the follow-up survey, in which 21 (5%) had missing data. The final sample included 412 women newly diagnosed with breast cancer who had valid data. Missing data were excluded pair-wise for statistical analyses.
Measures Social support was measured by the 19-item Medical Outcomes Study Social Support Survey [14], with a stem question that asks, ‘How often is each of the following kinds of support available to you if you need it?’ Each item was answered on a five-point scale, ranging from none of the time to all of the time. Scores were from 1 to 5 with higher scores indicating greater levels of social support. The original four functional support subscales included (1) emotional/informational support (eight items), (2) tangible support (four items), (3) affectionate support (three items), and (4) positive social interaction (three items), and an additional item [14]. Emotional support is the expression of positive affect, empathic understanding, and the encouragement of expressions of feelings (e.g., share your most private worries and fears with); informational support is the offering of advice, information, guidance, or feedback that can provide a solution to a problem (e.g., turn to for suggestions about how to deal with a personal problem); positive social interaction is companionship or the availability of other persons to share leisure and recreational activities (e.g., have a good time with); affectionate support is the expression of love and affection Psycho-Oncology 23: 1014–1020 (2014) DOI: 10.1002/pon
J. Leung et al.
1016
(e.g., love and make you feel wanted); and tangible support is the offering of material aid or behavioral assistance (e.g., take you to the doctor if you need it). Psychometric testing and scale validation in the current sample was previously conducted [26]. Factor analysis and parallel analysis in the current sample revealed three factors (Cronbach’s alphas 0.90–0.96), supporting the combination of the affectionate support and positive social interaction subscales, and the inclusion of the additional item in the emotional/informational support subscale [26]. Consequently, in our sample, the three subscales of social support that can be measured separately are mean emotional/informational support (EMI), mean affectionate support/positive social interaction (AFF), and mean tangible support (TAN). Health-related quality of life was measured using the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) [27]. The SF-36 assesses general physical and psychological health and well-being over the preceding 4 weeks. The survey covers eight dimensions of physical and mental health: physical functioning, role limitations due to physical or emotional problems, bodily pain, general health perceptions, vitality, social functioning, and emotional well-being or mental health. The responses to items within each dimension are summed and linearly transformed to produce scores ranging from 0 to 100 with higher scores indicating better health. Sociodemographic characteristics examined included marital status (single/divorced/widowed and married/ cohabiting), education (no formal, high school, and trade/ certificate/degree), country of birth (Australia, other English-speaking countries, and non-English-speaking countries), and area of residence (major cities, inner regional, outer regional, remote, and very remote).
Statistical analysis Statistical analyses were conducted using SPSS 19 (IBM Corp, Armonk, NY, USA). The women’s social support was compared over time, from the survey before diagnosis of breast cancer (pre-diagnosis) to the survey at which breast cancer was diagnosed (baseline) and at 3 years follow-up. Repeated-measures analyses of variance (ANOVAs) were conducted to compare levels of social support over time. Within-subject contrasts were conducted to compare pre-diagnosis social support with baseline social support and baseline social support with social support reported at follow-up. Similarly, repeated-measures ANOVAs were conducted to compare HRQOL over time. One-way between-subject ANOVAs were conducted to compare social support and HRQOL by sociodemographics (marital status, education, country of birth, and area of residence) at baseline. The bivariate associations between baseline social support and follow-up HRQOL were estimated using Pearson’s correlation. Partial correlation was used to estimate the association between social support Copyright © 2014 John Wiley & Sons, Ltd.
and HRQOL adjusting for sociodemographic characteristics found to be significantly associated with both social support and HRQOL. Structural equation modeling in Amos 17 was used to model the effects of social support on both physical and mental HRQOL outcome in the same model. Social support and HRQOL variables were standardized. Paths were entered between latent variables from baseline social support to physical and mental HRQOL. The observations on bodily pain, general health, physical functioning, and role physical were loaded onto the latent variable physical HRQOL. The observations on mental health, role emotional, social functioning, and vitality were loaded onto the latent variable mental HRQOL. The three social support subscales, EMI, AFF, and TAN, were loaded onto the latent variable social support. Covariance paths were entered between the three social support subscales. Sociodemographic characteristics were controlled for by entering the observed characteristics with paths to latent social support, physical, and mental HRQOL. A good model fit is indicated by a root mean square error of approximation ≤0.05, and normed-fit index and comparative fit index >0.90.
Results Social support and quality of life over time Women newly diagnosed with breast cancer had moderately high levels of EMI, AFF, and TAN social support overall, indicated by a mean of 3.97–4.16 out of a fivepoint scale at baseline (Table 1). Repeated-measures ANOVA results showed that there was no change in social support pre-diagnosis, at baseline, and after breast cancer diagnosis. Compared with pre-diagnosis HRQOL, women newly diagnosed with breast cancer had significantly worse quality of life on the subscales of bodily pain, general health, physical functioning, role physical, and vitality. However, compared with when they were newly diagnosed, at follow-up, the women had increased scores in some domains of HRQOL. Specifically, significant improvements were observed in their general health, role physical, role emotional, social functioning, and vitality (Table 1).
Social support by sociodemographic characteristics Analysis of variance results by sociodemographic characteristics showed that at baseline, women who were married/cohabiting reported higher EMI, AFF, and TAN support scores and had better HRQOL scores (Table 2). Therefore, marital status was included as a control variable in the analysis between social support and HRQOL. There were no significant differences in social support scores by education, country of birth, and area of residence. Psycho-Oncology 23: 1014–1020 (2014) DOI: 10.1002/pon
Social support, quality of life, and breast cancer
1017
Table 1. Social support and health-related quality of life at pre-diagnosis, at baseline diagnosis, and at 3-year follow-up after breast cancer diagnosis Before diagnosis M
[95% CI]
Social support (N = 124) EMI 3.98 [3.80, 4.16] AFF 4.20 [4.03, 4.37] TAN 4.01 [3.80, 4.22] Health-related quality of life (N = 345) Bodily pain 68.08 [65.40, 70.76] General health 69.74 [67.42, 72.06] Physical functioning 80.84 [78.70, 82.98] Role physical 73.90 [69.97, 77.83] Mental health 73.42 [71.46, 75.37] Role emotional 77.38 [73.49, 81.28] Social functioning 79.45 [76.81, 82.08] Vitality 57.89 [55.56, 60.22]
Baseline (1)
M 3.97 4.16 4.10
* *** ** **
*
65.12 63.70 77.80 65.18 72.60 73.80 77.04 55.86
Follow-up M
[95% CI]
[3.79, 4.16] [3.98, 4.34] [3.90, 4.29] [62.38, [61.21, [75.49, [60.61, [70.49, [69.63, [74.15, [53.30,
4.00 4.16 4.01
67.85] 66.18] 80.11] 69.74] 74.70] 77.96] 79.93] 58.41]
67.24 67.22 78.48 72.73 74.05 79.99 82.14 58.55
[95% CI]
Repeated-measures ANOVA
[3.82, 4.18] [3.99, 4.32] [3.81, 4.22] [64.61, [64.70, [76.15, [68.61, [72.07, [76.20, [79.60, [56.14,
69.88] 69.74] 80.82] 76.84] 76.03] 83.78] 84.68] 60.96]
F
Partial eta squared
0.06 0.30 0.66
Social support and health-related quality of life in women with breast cancer: a longitudinal study Janni Leung1*, Nancy A. Pachana2 and Deirdre McLaughlin1 1 2
The University of Queensland, School of Population Health, Brisbane, Queensland, Australia The University of Queensland, School of Psychology, Brisbane, Queensland, Australia
*Correspondence to: School of Abstract Population Health, The University Objectives: A breast cancer diagnosis is a distressing event that impacts on physical and psychological of Queensland, Herston Rd, functioning. This study examined the longitudinal relationships among a diagnosis of breast cancer, Herston, Brisbane, Queensland social support, and health-related quality of life (HRQOL). 4006, Australia. E-mail: j. Methods: Participants were 412 women from the 1946–1951 birth cohort of the Australian [email protected] Longitudinal Study on Women’s Health who self-reported a new diagnosis of breast cancer between 1998 and 2007. The three surveys of longitudinal data analyzed included data 3 years before diagnosis, at diagnosis (baseline), and 3 years after diagnosis (follow-up). Social support was measured using the 19-item Medical Outcomes Study Social Support Survey; HRQOL was measured using the Medical Outcomes Study 36-item Short-Form Health Survey. Results: Compared with pre-diagnosis HRQOL, women newly diagnosed with breast cancer reported significantly poorer HRQOL in subscales related to pain, physical functioning, and health and vitality. At 3-year follow-up, HRQOL had improved in most domains to levels consistent with pre-diagnosis. Levels of social support remained stable across time. The structural equation model showed that social support was positively predictive of better physical and mental HRQOL at 3-year follow-up. Conclusions: Longitudinal analyses indicate that social support appears to be an important predictor of HRQOL in women diagnosed with breast cancer. In particular, positive emotional and informational support that may normally be provided by a partner is important in maintaining HRQOL. Identification of those lacking social support, especially patients without partners, will enable them Received: 26 July 2013 to be guided to appropriate support networks and programs. Revised: 25 February 2014 Copyright © 2014 John Wiley & Sons, Ltd. Accepted: 26 February 2014
Introduction Breast cancer is the second most prevalent cancer among women and is one of the leading causes of death in women worldwide [1]. Despite this, advances in screening and diagnostic procedures mean that most breast cancers can be diagnosed at an early stage, leading to effective treatments and improved survival rates [2]. Nonetheless, a diagnosis of breast cancer is a distressing event that affects physical and psychological functioning and impacts on lifestyle and relationships with family and friends [3–5]. Breast cancer treatments, including systemic adjuvant therapy, lead to long-term worsened quality of life in breast cancer survivors [5]. Furthermore, treatment-related changes to women’s physical appearance may affect self-image, and other effects of treatment, including fatigue, nausea, and pain, may decrease opportunities for social interactions and limit the ability of women with breast cancer to continue their normal lifestyle and social activities [6]. This in turn curtails their opportunities to maintain supportive social networks and to receive appropriate emotional and instrumental support. Copyright © 2014 John Wiley & Sons, Ltd.
Social support has been identified as being protective for health and in particular for reducing cancer-related distress [7–9]. Studies have found that social support is associated with better adjustment to disease and better quality of life; however, the subjective appropriateness of the support offered is important [10,11]. Not all social relationships are sources of support; although women possess extensive and robust social networks, these can come at a price (such as maintaining the networks and managing expectations of providing support to others) [10,12]. The stress-buffering model of social support focuses on the perceived availability, quality, and appropriateness of support [13]. In particular, this model suggests that when confronted with a situation of high stress, the availability of appropriate levels of subjective support is more important than the number of contacts in the social network. Perceived social support in health-related contexts has been categorized as (a) emotional/informational support, (b) tangible support, (c) affectionate support, and (d) positive social interaction [14]. The provision of satisfactory emotional and informational support has been associated with better health and lower symptom intensity in women
Social support, quality of life, and breast cancer
with breast cancer [15]. Both tangible support, in the form of a psycho-educational program, and affectionate support, from family and friends, have been associated with better well-being and improved health-related quality of life (HRQOL) among breast cancer patients [16–18]. Although previous studies have reported that positive social support can enhance adjustment to breast cancer and improve HRQOL [18–20], these studies are largely cross-sectional. Little evidence has been presented that examines either the longitudinal impact of social support on HRQOL in women with breast cancer or changes over time in social support and HRQOL throughout the course of the disease. A small number of previous studies that have reported on patterns of social support associated with breast cancer have found that levels of social support fluctuated over time [21,22]. However, these studies did not have information on pre-diagnosis levels of social support, and analyses were based on short follow-up times (up to 6 months). Given the increasing incidence and higher survival rate from breast cancer in recent years, a focus on factors associated with HRQOL, especially from a longitudinal perspective, can provide valuable insight to inform the development of strategies to support women during this distressing experience. The aim of this study is to examine the impact of a breast cancer diagnosis on social support and quality of life over time. Additionally, this paper will examine the role of social support as a predictor of better HRQOL in women newly diagnosed with breast cancer. Consistent with previously reported cross-sectional research, it is expected that women with high levels of social support will report higher HRQOL scores.
Method Participants The Australian Longitudinal Study on Women’s Health (ALSWH) is a prospective study of the health and wellbeing of three cohorts of women who were aged 18–23 years (1973–1978 birth cohort), 45–50 years (1946–1951 birth cohort), and 70–75 years (1921–1926 birth cohort) when recruited in 1996. The study sample was selected randomly from the Medicare Australia database (which covers all citizens and permanent residents of Australia, including refugees and immigrants), with intentional oversampling of women living in rural and remote areas in order to achieve an adequate sample size in this population. The project uses mailed questionnaires to collect self-report data on health and related variables every 3 years. Details of the ALSWH recruitment and study design have been fully described elsewhere [23], and details of the study can be found at www. alswh.org.au. Participation is voluntary, and the ALSWH is approved by the University of Newcastle and University of Queensland Human Research ethics committees. Copyright © 2014 John Wiley & Sons, Ltd.
1015
Data for these analyses are drawn from surveys 1 to 6 (1996 to 2010) of the 1946–1951 birth cohort. Because of uncertainties about the accuracy of the Medicare Australia database, response rates to the initial invitation cannot be exactly specified. However, it has been estimated that 53–56% of women in this birth cohort responded with 14,072 women responding to survey 1 [24]. Of the total eligible women at each survey, responses to subsequent surveys were 12,338 (90.7%) at survey 2 (1998), 11,226 (84.3%) at survey 3 (2001), 10,905 (84.0%) at survey 4 (2004), 10,638 (83.8%) at survey 5 (2007), and 10,011 (83.3%) at survey 6 (2010). Participants were women who responded ‘yes’ to the following question in survey 2 (1998), survey 4 (2004), or survey 5 (2007): ‘In the past three years, have you been diagnosed or treated for breast cancer’. A baseline breast cancer diagnosis was a positive response at any of the three surveys. This self-report item has been validated against Cancer Registry data with 93% sensitivity and 98% specificity [25]. Pre-diagnosis data were drawn from the survey immediately preceding the baseline (breast cancer diagnosis) survey, and follow-up data were drawn from the survey immediately after the baseline survey. A total of 519 women reported a new diagnosis of breast cancer between 1998 and 2007: 82 in 1998, 199 in 2004, and 238 in 2007. There were 433 women (83%) who completed the follow-up survey, in which 21 (5%) had missing data. The final sample included 412 women newly diagnosed with breast cancer who had valid data. Missing data were excluded pair-wise for statistical analyses.
Measures Social support was measured by the 19-item Medical Outcomes Study Social Support Survey [14], with a stem question that asks, ‘How often is each of the following kinds of support available to you if you need it?’ Each item was answered on a five-point scale, ranging from none of the time to all of the time. Scores were from 1 to 5 with higher scores indicating greater levels of social support. The original four functional support subscales included (1) emotional/informational support (eight items), (2) tangible support (four items), (3) affectionate support (three items), and (4) positive social interaction (three items), and an additional item [14]. Emotional support is the expression of positive affect, empathic understanding, and the encouragement of expressions of feelings (e.g., share your most private worries and fears with); informational support is the offering of advice, information, guidance, or feedback that can provide a solution to a problem (e.g., turn to for suggestions about how to deal with a personal problem); positive social interaction is companionship or the availability of other persons to share leisure and recreational activities (e.g., have a good time with); affectionate support is the expression of love and affection Psycho-Oncology 23: 1014–1020 (2014) DOI: 10.1002/pon
J. Leung et al.
1016
(e.g., love and make you feel wanted); and tangible support is the offering of material aid or behavioral assistance (e.g., take you to the doctor if you need it). Psychometric testing and scale validation in the current sample was previously conducted [26]. Factor analysis and parallel analysis in the current sample revealed three factors (Cronbach’s alphas 0.90–0.96), supporting the combination of the affectionate support and positive social interaction subscales, and the inclusion of the additional item in the emotional/informational support subscale [26]. Consequently, in our sample, the three subscales of social support that can be measured separately are mean emotional/informational support (EMI), mean affectionate support/positive social interaction (AFF), and mean tangible support (TAN). Health-related quality of life was measured using the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) [27]. The SF-36 assesses general physical and psychological health and well-being over the preceding 4 weeks. The survey covers eight dimensions of physical and mental health: physical functioning, role limitations due to physical or emotional problems, bodily pain, general health perceptions, vitality, social functioning, and emotional well-being or mental health. The responses to items within each dimension are summed and linearly transformed to produce scores ranging from 0 to 100 with higher scores indicating better health. Sociodemographic characteristics examined included marital status (single/divorced/widowed and married/ cohabiting), education (no formal, high school, and trade/ certificate/degree), country of birth (Australia, other English-speaking countries, and non-English-speaking countries), and area of residence (major cities, inner regional, outer regional, remote, and very remote).
Statistical analysis Statistical analyses were conducted using SPSS 19 (IBM Corp, Armonk, NY, USA). The women’s social support was compared over time, from the survey before diagnosis of breast cancer (pre-diagnosis) to the survey at which breast cancer was diagnosed (baseline) and at 3 years follow-up. Repeated-measures analyses of variance (ANOVAs) were conducted to compare levels of social support over time. Within-subject contrasts were conducted to compare pre-diagnosis social support with baseline social support and baseline social support with social support reported at follow-up. Similarly, repeated-measures ANOVAs were conducted to compare HRQOL over time. One-way between-subject ANOVAs were conducted to compare social support and HRQOL by sociodemographics (marital status, education, country of birth, and area of residence) at baseline. The bivariate associations between baseline social support and follow-up HRQOL were estimated using Pearson’s correlation. Partial correlation was used to estimate the association between social support Copyright © 2014 John Wiley & Sons, Ltd.
and HRQOL adjusting for sociodemographic characteristics found to be significantly associated with both social support and HRQOL. Structural equation modeling in Amos 17 was used to model the effects of social support on both physical and mental HRQOL outcome in the same model. Social support and HRQOL variables were standardized. Paths were entered between latent variables from baseline social support to physical and mental HRQOL. The observations on bodily pain, general health, physical functioning, and role physical were loaded onto the latent variable physical HRQOL. The observations on mental health, role emotional, social functioning, and vitality were loaded onto the latent variable mental HRQOL. The three social support subscales, EMI, AFF, and TAN, were loaded onto the latent variable social support. Covariance paths were entered between the three social support subscales. Sociodemographic characteristics were controlled for by entering the observed characteristics with paths to latent social support, physical, and mental HRQOL. A good model fit is indicated by a root mean square error of approximation ≤0.05, and normed-fit index and comparative fit index >0.90.
Results Social support and quality of life over time Women newly diagnosed with breast cancer had moderately high levels of EMI, AFF, and TAN social support overall, indicated by a mean of 3.97–4.16 out of a fivepoint scale at baseline (Table 1). Repeated-measures ANOVA results showed that there was no change in social support pre-diagnosis, at baseline, and after breast cancer diagnosis. Compared with pre-diagnosis HRQOL, women newly diagnosed with breast cancer had significantly worse quality of life on the subscales of bodily pain, general health, physical functioning, role physical, and vitality. However, compared with when they were newly diagnosed, at follow-up, the women had increased scores in some domains of HRQOL. Specifically, significant improvements were observed in their general health, role physical, role emotional, social functioning, and vitality (Table 1).
Social support by sociodemographic characteristics Analysis of variance results by sociodemographic characteristics showed that at baseline, women who were married/cohabiting reported higher EMI, AFF, and TAN support scores and had better HRQOL scores (Table 2). Therefore, marital status was included as a control variable in the analysis between social support and HRQOL. There were no significant differences in social support scores by education, country of birth, and area of residence. Psycho-Oncology 23: 1014–1020 (2014) DOI: 10.1002/pon
Social support, quality of life, and breast cancer
1017
Table 1. Social support and health-related quality of life at pre-diagnosis, at baseline diagnosis, and at 3-year follow-up after breast cancer diagnosis Before diagnosis M
[95% CI]
Social support (N = 124) EMI 3.98 [3.80, 4.16] AFF 4.20 [4.03, 4.37] TAN 4.01 [3.80, 4.22] Health-related quality of life (N = 345) Bodily pain 68.08 [65.40, 70.76] General health 69.74 [67.42, 72.06] Physical functioning 80.84 [78.70, 82.98] Role physical 73.90 [69.97, 77.83] Mental health 73.42 [71.46, 75.37] Role emotional 77.38 [73.49, 81.28] Social functioning 79.45 [76.81, 82.08] Vitality 57.89 [55.56, 60.22]
Baseline (1)
M 3.97 4.16 4.10
* *** ** **
*
65.12 63.70 77.80 65.18 72.60 73.80 77.04 55.86
Follow-up M
[95% CI]
[3.79, 4.16] [3.98, 4.34] [3.90, 4.29] [62.38, [61.21, [75.49, [60.61, [70.49, [69.63, [74.15, [53.30,
4.00 4.16 4.01
67.85] 66.18] 80.11] 69.74] 74.70] 77.96] 79.93] 58.41]
67.24 67.22 78.48 72.73 74.05 79.99 82.14 58.55
[95% CI]
Repeated-measures ANOVA
[3.82, 4.18] [3.99, 4.32] [3.81, 4.22] [64.61, [64.70, [76.15, [68.61, [72.07, [76.20, [79.60, [56.14,
69.88] 69.74] 80.82] 76.84] 76.03] 83.78] 84.68] 60.96]
F
Partial eta squared
0.06 0.30 0.66
