外 文 文 献 翻 译
毕业论文
译文题目： The contribution of attachment security and social
support to depressive symptoms in patients with metastatic cancer
学生姓名： 专 业： 指导教师：
2011年10月 15日
咸宁学院本科毕业论文：
The contribution of attachment security and social support to depressive symptoms in patients with
metastatic cancer
Gary Rodin1,2,3*, Andrew Walsh1,2, Camilla Zimmermann1,3, Lucia Gagliese1,2,3,4, Jennifer Jones1,2,3,Frances A. Shepherd3,5, Malcolm Moore3,5, Michal Braun6, Allan Donner7 and Mario Mikulincer8
Abstract
The present study examines the association between disease-related factors, perceived social support, attachment security (i.e. attachment anxiety and avoidance), and the occurrence of depressive symptoms in a sample of patients with metastatic gastrointestinal or lung
cancer.Results from a sample of 326 cancer outpatients with advanced disease indicate that diseaserelated factors are significantly associated with the occurrence of depressive symptoms, and the latter are inversely related to the degree of attachment anxiety and
avoidance, and perceived social support.Attachment security (on the dimension of anxious attachment) significantlybuffered the effect of disease-related factors on depressive symptoms, and perceived social support mediated the relationship between attachment security and depressive symptoms. The buffering effect of attachment security on depressive symptoms and its partial mediation through social support suggest that the interaction of individual, social, and disease-related factors contribute to the emergence of depressive symptoms in patients with metastatic cancer.
Keywords: cancer; oncology; attachment security; social support; depression; burden of illness
Introduction
Depression in cancer and other medical conditions has been regarded as a final common pathway of distress, which may occur in response to the multiple stressors and adaptive
challenges associated with these illnesses [1,2]. The extent to which physical suffering leads to depression may depend not only upon disease-related variables, but also on individual factors and on the relational milieu. In particular, the expectation and perception of support from others may play an important role in determining to what extent depressive symptoms emerge in the context of progressive disease. The present study was designed to examine the
association between these relational factors and the occurrence of depressive symptoms in patients with metastatic gastrointestinal (GI) and lung cancer.There is now considerable research investigating the extent to which supportive relationships may protect from the
psychological distress associated with medical illness [3–5] and with other traumatic life events [6–8].
In particular, a buffering effect has been observed in which the perception of social support tends to diminish the tendency of stressful life events to produce emotional distress [9,10]
. This has been observed among patients with cancer[11,12] as well as with other medical conditions[13,14]. However, it has also been noted that the perception of social support is a complex phenomenon which may be determined not only by its availability and provision by significant others but also by temperamental and other individual factors which affect the elicitation and experience of support from human relationships [14–16]. The concept of ‘attachment style,’ which refers to internalized expectations and preferences regarding proximity to significant others and protection in times of need [17], is an example of an individual factor which affects the availability and experience of human support.
Attachment relationships are emotional bonds between individuals which provide a sense of felt security in response to threat [18] and which constitute a mutual system for the
regulation of emotions [19]. Those with secure attachment styles are postulated to have had
18
咸宁学院本科毕业论文：
more positive early caregiving experiences [20] and to have a greater capacity to ask for and to accept care and to express and acknowledge their emotions [21]. Such individuals may be more protected from psychological distress in the context of stress because they have a greater capacity to regulate affect states and to experience, anticipate and elicit support in relationships in times of threat [17, 22].
This protection, therefore, may derive not only from the experience of current support provided by significant others, but also from expectations and preferences regarding such support. Compatible with this view, a secure attachment style has been associated with less depression in community samples [23] and in patients with chronic pain[24] and migraine [25]. However, its role has not been previously examined in patients with cancer. Insecure attachment has been represented on dimensions of anxiety (i.e. the degree to which
individuals worry about being rejected, abandoned or unloved by significant others) and
avoidance (i.e. the degree to which individuals limit intimacy and interdependence with others) [26]
. Categories of ‘secure,’ ‘preoccupied,’ and ‘avoidant’ (either dismissing or fearful) attachment have been constructed to represent the clinical prototypes for these dimensions [27,28]
. Secure individuals are low in both anxiety and avoidance, preoccupied individuals are high in anxiety and low in avoidance, and avoidant individuals are high in avoidance and either high (fearful) or low (dismissing) in anxiety [26]. Although it is increasingly recognized that anxious and avoidant attachment styles are traits that exist on a continuum rather than discrete categories [29,30], a categorical approach may still have heuristic value in clinical settings [31]. Attachment style may be linked to the perception of social support in several different ways [26].More insecurely attached individuals, both anxious and avoidant, tend to appraise support as less available from relationship partners, to be less satisfied with the support they receive, and to be less likely to mobilize or cultivate mutually supportive relationships (e.g. [32–35]). Such individuals are also less likely to benefit from supportive interactions when coping with stress and distress and to express more doubts and distrust
about their relationship partner’s supportiveness (e.g.[26,36]). In that regard, Collins and Feeney [26]
found that anxious and avoidant participants who received standardized
ambiguous-support messages from their romantic partner perceived this partner as less
supportive in a subsequent interaction than did securely attached participants. These findings suggest that people with attachment insecurity are predisposed to negative appraisals of support from their partners, in accordance with their chronic negative internal ‘working
models,’ in which support from such others is not expected.Although some evidence suggests that symptom burden in cancer patients increases toward the end of life [37,38], there is
substantial physical distress in many cancer patients before that time [39]. In that regard, the SUPPORT study revealed that more than 25% of patients with metastatic colon cancer and advanced non-small cell lung cancer report severe uncontrolled pain 3–6 months before the end of life [40]. Other studies have shown that pain and other physical symptoms are often associated with the development and exacerbation of psychological distress [39,41], most
notably depression [42–45]. The present study was designed to evaluate the relationship between disease burden and depressive symptoms in individuals with advanced cancer, and to evaluate the relative contribution and protective role of attachment security and social support to
depressive symptoms in this population. We hypothesized that:(1) Attachment security (lower scores along attachment anxiety and avoidance dimensions) and social support would independently buffer(or moderate) the relationship between disease burden and depressive symptoms.(2) Perceived social support would mediate the relationship between attachment security and depressive symptoms.
Methods
Measures and data collected
Medical and demographic data abstracted from the medical record of each participant and
19
咸宁学院本科毕业论文：
from the recruitment interview included: age, gender, marital status, current living
arrangements, highest level of education, country of birth, primary language, socioeconomic status (family income based on postal code), cancer site, cancer stage,and date of diagnosis.Cognitive functioning was assessed using the Short
Orientation-Memory-Concentration Test (SOMC; [46]), a 6-item mental status examination validated for use in acute care settings. A cut-off score of 520 was used to screen for cognitive impairment at the time of recruitment. Psychological distress was assessed using the Beck Depression Inventory-II (BDI-II; [47]). The BDI-II is a 21-item self-report measure of
depressive symptoms that is consistent with the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR; [48]) criteria for major depressive disorder. Participants are asked to rate their feelings during the preceding two weeks with regard to each symptom domain by indicating their agreement with one of four related statements. Each domain is scored on a 4-point scale ranging from 0 to 3, with scores of 15 or greater on the BDI-II being clinically significant for a diagnosis of major depression [49]. This measure has been widely used in cancer populations [49–53]. Attachment security was rated using the Experiences in Close Relationships Inventory (ECR; [29]), a 36-item self-report measure tapping attachment anxiety and avoidance (18 items per dimension) in romantic and marital relationships. This measure was derived from a factor analysis of 482 items designed to assess 60 attachment-related constructs from existing self-report measures of adult romantic
attachment [29]. Participants are asked to indicate their agreement with each statement based on their feelings and experiences in romantic relationships on a 7-point Likert-type scale ranging from 1 (disagree strongly) to 7 (agree strongly). The ECR yields two subscales, attachment avoidance(the discomfort with closeness and dependence on others) and
attachment anxiety (the fear of rejection and abandonment). For each subscale, a total score was computed by averaging the relevant 18 items. Based on the ECR, attachment security is operationalized as lower scores on the attachmentanxiety and avoidance subscales. Social support was rated using the Medical Outcomes Study Social Support Survey
(MOSSSS;[54]), a 20-item self-report measure developed for use in chronically ill populations. This scale measures multiple domains of social support including: (i) emotional/informational support;(ii) tangible support; (iii) affectionate support; (iv) positive social interaction; and (v) global social support. Participants are asked to quantify the number of friends and relatives they feel close to and then to indicate the degree to which support from either would be
available as it relates to each statement included in the questionnaire. Response formats range from 1 (none of the time) to 5 (all of the time). For each subscale of the MOS-SSS, a transformed score was computed from the raw scores, as described by Sherbourne and Stewart[54]. Scores range from 0 to 100, with higher scores representing greater perceived social support.Disease burden was assessed using a shortened version of the Memorial
Symptom Assessment Scale (MSAS; [52]), a multidimensional self-report measure designed to assess the presence, frequency,severity, and distress associated with common cancer
symptoms (23 physical, 6 psychological).Symptoms are rated as present or absent; if present, symptom frequency is rated on a 4-point scale, ranging from 1 (rarely) to 4 (almost
constantly), symptom severity is rated on a 4-point scale, ranging from 1 (slight) to 4 (very severe), anddistress caused by the symptom is rated on a 5-point scale, ranging from 1 (not at all) to 5 (verymuch). Five separate indices were generated from this measure: (i) total number of physical symptoms;(ii) average frequency of endorsed symptoms;(iii) average severity of endorsed symptoms; (iv)average distress caused by the symptoms; and (v) a Global Distress Index (GDI), computed as the mean symptom frequency ratings of four items:feeling sad, worrying, feeling irritable and feeling nervous, and the mean symptom distress ratings of five items: lack of energy, pain, feeling drowsy,constipation, and dry mouth.
Participants and procedure
20
咸宁学院本科毕业论文：
The research protocol for this study was reviewed and approved by the Research Ethics Board of the University Health Network. Participants were recruited from medical and
radiation oncology outpatient clinics at Princess Margaret Hospital(PMH) (a member of the University Health Network),a comprehensive cancer center in Toronto, Canada. Patients were eligible for this study if they had a confirmed diagnosis of Stage IV GI or Stage IIIA, IIIB, or IV lung cancer, were 18 years of age or older, and were able to speak and read English sufficiently to provide informed consent and complete self-report questionnaires. Patients were excluded if they had a confirmed diagnosis of carcinoid or neuroendocrine carcinoma, or if significant cognitive impairment was documented in the medical chart, identified by their physician,or demonstrated by a failure to meet the predetermined cut-off on the SOMC [46] at the time of recruitment. Potential participants were presented with an introductory letter by a clinic volunteer or nurse asking permission to allow a research assistant to approach them regarding study participation.If the patient gave permission, the research assistant described the study and informed written consent was obtained. Participants who provided consent were then assessed using the SOMC. Those scoring 420 on the SOMC were then given a
Karnofsky Performance Status (KPS;[55]) rating and demographic information was recorded. The patient was given a questionnaire package, which was completed either in the waiting room or at home and returned by mail (postage and return envelope were provided) (KPS ratings and other data from this study have been reported elsewhere) [56]. A research team member assisted patients who requested or required assistance to complete the questionnaire package. Telephone reminder calls were made if the questionnaire package had not been returned within two weeks.Referral to the PMH Psychosocial Oncology Program was offered to patients who requested psychosocial assistance, or who reported significant emotional distress or suicidality.
Statistical analysis
Data were analyzed using the Statistical Package for the Social Sciences (SPSS), version 12.0 for Windows 2000 [57]. Descriptive statistics were calculated for the demographic and medical data,and for the attachment security, social support,disease burden, and psychological distress (i.e.depression) variables. The relationship between depressive symptoms and demographic and medical characteristics, attachment security, social support, and disease burden was evaluated using ordinary least-squares linear regression. Statistical tests were two tailed with alpha set at 0.05. Mean score imputation within cancer site (i.e. GI or lung) was used to estimate missing scale scores for psychological, psychosocial, and physical
measures.Mean scores were imputed only if 575% of the study assessment package had been completed. Mean estimation resulted in 3.4% imputed scores for the MSAS, 4.3% for the ECR, 5.2% for the BDI-II, and 7.7% for the MOS-SSS. Mean estimation resulted in little change to the standard deviations and similar results were obtained in comparative analyses using non-imputed data.Systematic response omissions were not detected.To test the first hypotheses that attachment security and social support buffer the association between disease burden (i.e. number of physical symptoms, mean symptom frequency, severity and distress, and GDI) and depressive symptoms,regression analysis was utilized according to Aiken and West [58]. Depression (i.e. BDI-II total score) was regressed onto disease burden indicators (tested separately), and attachment security indicators (i.e. anxious and avoidant attachment) or social support domains (i.e. emotional/informational,tangible, affectionate, positive social interaction, and global social support, tested separately)along with their cross-product
interactions (between each disease burden indicator and each attachment/support indicator) in one blocked regression step. To facilitate description of the interactions, regression predictors were centered to their means. Each individual interaction between the attachment
security/social support indicators and disease burden was evaluated separately to elucidate the individual pathways.To test the second hypotheses that social support would mediate the
21