Self-control, self-efficacy, role overload, and stress responses among siblings of children with cancer.
This study focused on responses to sibling cancer and its aftermath, with particular scrutiny directed toward stress factors, duress responses, and coping resources. Adopting Lazarus and Folkman's (1984) theoretical definition of stress, we investigated role overload as these HSCC's stress factor, anxiety and physical symptoms as their duress responses, and self-control and self-efficacy as their coping resources.
According to Lazarus and Folkman (1984), a state of duress is a dynamic state of imbalance between oneself and one's surroundings, when the latter is perceived as placing too many demands on one's personal well-being. Childhood cancer poses a threat to the sick child and the entire family system, therefore evoking a stress response. A state of duress incorporates three processes: primary appraisals, wherein the person perceives the threat itself; secondary appraisals, wherein the person raises to consciousness a possible response to the threat; and coping, wherein the person rises out of the duress (Lazarus & Folkman, 1984). Coping involves cognitive and behavioral attempts to control, reduce, or tolerate the internal and external demands (Rosenbaum, 2000). Coping is affected by the characteristics of the pressure-causing event as well as by internal and environmental resources at the individual's disposal (Bush, 1998). Thus, coping plays two complementary functions: regulating emotions or distress (emotion-focused coping), and exercising control over the problem creating the distress (problem-focused coping) (Rosenbaum, 2000). These coping types encompass one's ability to control oneself (Lazarus & Folkman, 1984; Rosenbaum, 2000).
Role Overload: The Stress Factor
The need to cope with childhood cancer necessitates a redefinition of roles within the family (Koch, Harter, Jakob, & Siegrist, 1996). The HSCC may begin to fill more roles, such as helpers at home, as caregivers for younger siblings, as liaisons between the sick sibling and his or her school, and even as a supportive listening ear for parents in the case of adolescents (Gorman, 1998). Some roles are undertaken as a reaction formation for the resentment that HSCC may feel toward the situation or as a way of identifying with parents. Whether imposed or undertaken, these roles affect children's self-concept and self-assessment and may lead to the experience of role overload.
Role overload refers to an imbalance between the role demands placed on the individual and the resources at the person's disposal to meet those demands (French & Caplan, 1973). Overloaded individuals must do more than they can do in the time available to them or require knowledge or skills that lie beyond their capabilities. The literature recognizes several types of role overload. Quantitative overload (a large number of tasks in a given period of time) contrasts with qualitative overload (a demand to perform at a level exceeding the resources available to the individual). An objective load can be measured objectively (for example, in terms of time), whereas a subjective load is experienced, perceived, and reported by the individual (French & Caplan, 1973).
Novak and Guest (1989) suggested five indices for "burden" of role overload:
1. Time burden refers to the burden imposed by restrictions on the caregiver's time. The HSCC's need to fulfill household-related tasks and care for younger siblings decreases their time and availability for participating in social activities outside the home and for planning their personal daily routine.
2. Developmental burden refers to the sense of impaired development as compared with peers (for example, prevention of social activities because of the sibling's illness may impair the HSCC's social skills development).
3. Physical burden refers to feelings of tiredness and damage to physical health. HSCC have reported physical symptoms such as headaches, stomachaches, and sleeping difficulties.
4. Social burden refers to the caregiver's feelings toward the caregiving role. The HSCC may experience a conflict between their desire to resume their routine peer-group lifestyle on the one hand, and the role they undertake or are expected to undertake to compensate the sick sibling for his or her inability to participate in the peer group's social events and routine, on the other hand.
5. Emotional burden refers to the caregiver's negative feelings toward the person receiving care. Among HSCC, this is reflected in feelings of anger, jealousy, and aggression toward the sick sibling, the parents, or both (Dolgin & Phipps, 2000; Sourkes & Proulx, 2000).
Anxiety and Psychosomatic Symptoms: Duress Responses
HSCC experience a range of duress responses. In the emotional realm, they develop anxiety over the existential threat posed by the disease, their concern for the patient's fate, and the fear that they, too, will contract the disease (Sourkes & Proulx, 2000).This anxiety is apt to manifest itself in acting-out behaviors, daydreams, and physical symptoms such as headaches, stomachaches, and decreased appetite.
This study focused on two of the most common responses among children: anxiety and psychosomatic symptoms. In clinical interviews conducted with HSCC by Liat Hamama, a high range of anxiety and various physical maladies, such as headaches and stomachaches, were found (Hamama, Ronen, & Feigin, 2000).
Anxiety is a cloudy, unpleasant emotional state consisting of concerns, fright, distress, and restlessness. It is a response to physical or psychological danger or to the threat of such danger, and it comprises both state anxiety and trait anxiety (Spielberger, 1972). State anxiety refers to a temporary emotional state characterized by tension, worry, fear, disquiet, and restlessness, accompanied by physiological arousal related to the autonomic nervous system, perceived as an unpleasant experience. This anxiety may be triggered either by stimuli originating from the external world or by internal stimuli in one's emotional world, perceived or interpreted as dangerous or threatening. Trait anxiety refers to the person's predisposition to feel anxious and sensitive in situations perceived as dangerous or threatening and to respond with increased state anxiety. Previous studies showed that HSCC experienced a high level of state anxiety with regard to the patient's fate, social isolation, an undermining of their routine, and a threat to the wholeness of the family system (Murray, 2000).
Psychosomatic symptoms are normative responses to experiences that arouse stress or duress (Kazdin, 1988; Ronen, Rahav, & Rosenbaum, 2003). Physical manifestations such as stomachaches, headaches, and fatigue may indicate that an unusual problem exists in the child's environment (Margalit, 1983). Israeli children during the Gulf War or terrorist attacks demonstrated more psychosomatic symptoms than did children during more peaceful times (Rahav & Ronen, 1994; Ronen, Rahav, & Appel, 2003; Ronen, Rahav, & Rosenbaum, 2003; Rosenbaum & Ronen, 1990), underscoring children's tendency to express tension or stress through psychosomatic symptoms.
Self-Control and Self-Efficacy: Coping Resources
A stressful situation can either create a crisis and traumatic response or mobilize the individual for coping. In coping with a stressful situation, the individual sustains primary appraisal, secondary appraisal, and coping phases (Lazarus & Folkman, 1984). In the current study, we selected self-efficacy (SE) as a variable representing the secondary appraisal phase, when the individual becomes aware of possible responses to a threat and mentally explores different paths of action, coping strategies, and their probability of success. We selected self-control (SC) as a variable representing the coping phase, when the individual implements the planned response by thinking or doing something to cope with the stressful situation. Both variables relate to cognitive processes. Theoretically, SE can be distinguished from SC, but in reality, it is difficult to separate the two, and the perception emerged in the 1990s that the two concepts are interconnected. SE emphasizes expectations, whereas SC emphasizes skills.
Perceived SE is individuals' subjective assessment of their capacity to carry out an action necessary to achieve hoped-for results (Bandura, 1997). It relates to a person's belief about whether he or she can use that certain behavior to reach the desired outcome ("Can I do this?") and to a person's capabilities to organize and execute the courses of action required to produce given attainments (Bandura, 1997). This belief may differ from a person's expectations about the probability that a certain type of behavior will lead to the result ("What will happen if I do this?"). Children's SE develops through their interactions with their surroundings and as they learn about their own abilities in diverse areas of functioning (Bandura, 1997).
SE has been examined in a wide range of problem areas regarding anxiety, phobias, depression, and learning processes (Bandura, 1997; Beckham, Burker, Lytle, Feldman, & Costakis, 1997), as well as regarding chronic health conditions, arthritis, and cancer in adults (Beckham et al., 1997). The current study focuses on HSCC's specific SE related to the tasks evoked by the cancer of their siblings.
SC is a set of skills enabling a person to shape behavior without external coercion, out of his or her own free will, while replacing one type of behavior with another more desirable one (Rosenbaum, 2000). SC derives from the assumption that a person's behavior is goal-oriented and is always undergoing a process of change and development. SC is especially important when a person must learn new behavior patterns or make decisions, or when a previous behavior is no longer as effective as it was in the past (Rosenbaum, 2000). Self-controlled behavior fosters a feeling of power, confidence, comfort, and independence in terms of the ability to direct one's own life.
Studies have shown a correlation between SC and behavior problems in a broad range of problem areas such as cancer, anxieties, nightmares, medical procedures, test anxiety, depression, sleeping problems, enuresis, encopresis, and stuttering (Hamama et al., 2000: Ronen, Rahav, & Appel, 2003). Hamama et al. found that SC contributed to a decrease in HSCC's anxiety and loneliness.
We hypothesized that
* healthy siblings with higher role overload will also present a higher level of anxiety and a higher number of psychosomatic symptoms.
* healthy siblings with a higher level of SC skills and a higher level of illness-related SE will present a lower level of role overload.
* healthy siblings with a higher level of SC and a higher level of illness-related SE will present lower levels of role overload and duress responses (anxiety and psychosomatic symptoms).
* healthy siblings who report higher levels of SC and illness-related SE will present a lower correlation between role overload and duress responses (anxiety and psychosomatic symptoms) than will HSCC reporting lower levels of coping resources.
Participants comprised 100 Israeli Jewish HSCC, recruited through pediatric oncology departments at five hospitals in various parts of Israel during a period of 19 months. Criteria for participation were an ill brother or sister who was receiving active treatment for cancer (chemotherapy, radiotherapy, bone marrow transplant) during the recruitment period; the healthy sibling's age (8 to 19 years) and fluency in speaking and understanding Hebrew (thus excluding immigrant families and Israeli Arab families); and parental consent for participation.
During the recruitment period, we identified 245 families who met the criteria but excluded 47 families wherein the cancer had recurred. Of the remaining 198 families, 48 families were unwilling to expose the children to contents that the questionnaire could elicit, and 50 families lacked time or interest. Accessibility to the ill children's medical and social work files revealed no significant differences between participating families and nonparticipating ones, in terms of the type of disease, type of treatment, or duration of illness. Of the 100 ill children included in the study, 50 had leukemia, 11 had bone tumor, 15 had soft-tissue sarcoma, 12 had cancer of the cerebral nervous system, 11 had lymphoma, and 1 did not mention the type. Regarding duration of sickness, 37 children were ill less than six months, 49 were ill between six to 12 months, and 14 were ill more than 14 months.
The impact on study variables of the ill child's type of cancer, type of treatment, and duration of illness were examined. No significant differences emerged for type of cancer or treatment, corroborating previous research on HSCC (Hamama et al., 2000). The lack of significant findings for illness duration differed from Hamama et al.'s, where anxiety levels decreased with longer durations of illness. Analyses controlling for these three illness characteristics yielded findings similar to analyses that did not control for these variables; therefore, we related to all the sick children as one group.
The 100 participating HSCC (53 boys, 47 girls) comprised 34 children ages 8 to 10 years, 41 children ages 11 to 14, and 25 children ages 15 to 19 (M = 12.06, SD = 2.69). No significant difference emerged in the three age groups' gender distributions. Regarding rank in the family, 40 were oldest, 43 were middle children, and 17 were youngest. As for family size, nine children came from families of two children, 41 from families of three children, 23 from families of four children, and 27 from families of five children or more. Regarding parents' employment, 82 fathers and 49 mothers were employed, and 18 fathers and 51 mothers were unemployed. Twenty-four fathers had less than 12 years of education and 38 had 12 years or more.
Preliminary analyses of the sociodemographic variables characterizing the five groups of healthy siblings (from each hospital) revealed no significant differences; therefore, all the analyses addressed the HSCC as one group. Family characteristics (for example, family size, birth order, employment, and education) did not significantly affect the findings and, therefore, are not presented here.
Parents completed a scale on demographic data constructed for this study. It included information on the participant's gender and age, family size, the healthy sibling's place in the birth order, head of household's employment and education level, housing congestion (number of bedrooms), and details on the sibling's illness (diagnosis, duration, and hospitalizations during the questionnaire completion period).
HSCC completed scales on role overload, anxiety, psychosomatic symptoms, SC, and illness-related SE.
Role Overload. For this study, we adapted Zarit and Zarit's (1982) 29-item Burden Interview. We translated the scale into Hebrew using a back-and-forth translation procedure, with six expert judges. After examining inter-judge agreement, we selected 12 items that described the feelings of burden or overload that children with an ill sibling may experience (for example, "I am satisfied with the roles imposed on me," "I feel that my social life is being hurt by the roles I am filling at home"). Participants rated items on a five-point scale ranging from 1 = not at all to 5 = to a great extent; thus, higher mean scores indicated higher role overload. Internal consistency for the adapted version in this study was high ([alpha] = .82).
Anxiety. We used the Hebrew adaptation (Teichman & Melnick, 1979) of the state anxiety subscale of the State-Trait Anxiety for Children Scale developed by Spielberger, Edwards, Montouri, and Lushene (1970). The scale was found valid for children, and its internal consistency was high ([alpha] = .89). Ben-Raphael (1981) found internal consistency of [alpha] = .77. In its abridged version, the state anxiety subscale comprises 10 statements on the child's emotional state at a given moment. Participants rated items on a four-point Likert-type scale ranging from 1 = not at all to 4 = to a great extent; thus, higher mean scores indicated higher state anxiety. Cronbach's alpha for the abridged version was .82 in a previous study (Hamama et al., 2000) and .78 in the current study for children ages 8 to 19 years.
Psychosomatic Symptoms. To assess the incidence of both physical and emotional (psychosomatic) symptoms considered prevalent enough to warrant children's referral for therapy, we used the frequent symptoms scale developed by Lapouse and Monk (1958) and translated into Hebrew by Rahav and Ronen (1994). The scale constitutes a problem identification tool (Kazdin, 1988) and has been distributed to school-age children in previous studies (Rahav & Ronen, 1994). Participants rated whether each of the 18 items (for example, crying, headaches, enuresis, and stuttering) had occurred in the past month (Yes = 1 or No = 2). Scores ranged from 18 to 36 ([alpha] = .70), where higher scores indicated more physical symptoms in the past month. Prior means in studies of Israeli children were 21.75 and 24.75 (Rahav & Ronen, 1994; Ronen, Rahav, & Appel, 2003), respectively.
Self-Control. To examine children's ability to exert SC, we used the Children's Self-Control Scale, a 17-item self-report scale developed for children by Rosenbaum and Ronen (1990). This scale depicts everyday situations in the children's lives: school, games, and discipline at home and at school (for example, "When I get headaches at school, I find it very difficult to continue participating at school and to forget the pain"). Participants rated items on a six-point scale ranging from -3 = not very typical of me to +3 = very typical of me. A sampling of 2,000 children from the general Israeli population produced a Cronbach's alpha of .65. Scores were assigned to each of the six rankings on the scale, such that -3 indicated low SC and +3 indicated high SC, with the possible total score for the entire 17-item scale ranging from 1 to 6. The present sample's Cronbach alpha was .65. Despite the moderate alpha level, direct reporting by the child holds great importance in light of the limited validity attributed to parental or teacher reports on children's emotional states. Prior means in study of HSCC were 3.60 (Hamama et at., 2000).
Perceived SE Regarding the Sibling's Illness. On the basis of a specific and highly focused test of perceived SE suggested by Bandura (1997, pp. 42-44), this study included a single item tapping the healthy siblings' specific belief in their ability to cope with the tasks evoked by the abnormal situation they confronted in having an in sibling. As proposed by Bandura, we asked only one question: "To what extent do you feel you are managing to cope with your sibling's disease?" We asked participants to rate this item on a 10-point scale ranging from 1 = unable to cope to 10 = coping well.
The first author located the suitable families in the five sampled hospitals, obtained their agreement to participate, and collected sociodemographic data. After receiving formal agreement from families, she gathered data from the sick children's medical flies concerning diagnosis, treatment, and duration of illness. Then she met with most of the healthy siblings in their homes (n = 76) and attended the children as they completed the questionnaires. Questionnaires were completed in approximately 30 minutes. At least one parent was present at home when the child completed the questionnaire. The remaining 24 families asked that the questionnaire be mailed to them, and all of them returned their completed questionnaires by mail. These parents signed a statement attesting that the child had completed the questionnaire without help. Analysis showed no differences in demographic data between participants' completion during home visits or in writing; hence, analyses related to them as one group.
Descriptive Statistics for Study Variables
The study variables' means and standard deviations are presented in Table 1. As seen in the table, participants reported a moderate level of role overload--the stress factor--and of the two duress responses: anxiety and psychosomatic symptoms. The mean score of 1.79 for anxiety indicated, on average, feeling anxiety "sometimes," which was lower than reported in other studies (Hamama et al., 2000). The mean score of 25.30 for psychosomatic symptoms was higher than in previous studies (Rahav & Ronen, 1994). Note that the current participants reported duress responses during a time of stress, whereas in earlier research children were interviewed after stressful events had concluded (after the Gulf War, after a terrorist attack), so that earlier samples knew they were out of danger. Regarding the coping resources, the mean score of 4.0 for SC was slightly lower than data from normative same-age samples but within the norms (Rosenbaum & Ronen, 1990) and similar to those of children after a terrorist attack (Ronen, Rahav, & Appel, 2003).
Intercorrelations between Variables
Significant positive correlations emerged between role overload and each of the two duress responses: anxiety and psychosomatic symptoms (correlating moderately with anxiety and highly with psychosomatic symptoms). Significant Pearson correlations emerged between role overload and anxiety (r = .23, p < .05) and between role overload and psychosomatic symptoms (r = .26, p < .01), thus supporting the study's first hypothesis that healthy siblings who experience role overload will report a higher level of anxiety and a greater number of symptoms. Moreover, role overload revealed a high significant negative correlation with one of the two coping resources: Healthy siblings with a higher rate of SC skills experienced less role overload (r = -.21, p < .01). Anxiety correlated positively with psychosomatic symptoms (r = .49, p < .001) and negatively with SC (r = -.44, p < .001) and with SE (r = -.24, p < .05). Psychosomatic symptoms correlated negatively with SC (r = -.29, p < .01) and SE (r = -.41, p < .001). Inasmuch as the correlation with SE was no significant, this finding only partially supports this study's second hypothesis. Overall, the significant negative correlations suggest a link between stronger coping resources and milder duress responses.
Hierarchical Regression Analyses
We performed hierarchical regression analyses to examine the third hypothesis, that HSCC who report high levels of SC and SE will reveal a lower correlation between role overload and duress responses (anxiety and psychosomatic symptoms) than will HSCC who report low levels of coping resources. In the first hierarchical regression, the dependent variable was anxiety. In the first step, we entered the sociodemographic variables: healthy sibling's gender (male = 1, female = 2), healthy sibling's age, number of children in family, father's education, and housing congestion. In the second step, we entered personal resources: SC and SE. In the third step, we entered the stress factor (role overload), and in the fourth step, we entered the interactions between the study variables. Variables in steps 1 and 2 were entered using the simultaneous regression method, that is, variables were entered in the same step without relating to their significance level. Conversely, in step 3, the criterion for entry was the significance level.
This hierarchical regression for anxiety explained 42 percent of the variance. Two sociodemographic variables (step 1) contributed significantly (p < .05) to the variance in anxiety: gender (.19) and family size (.12). Girls reported more anxiety than did boys, and children from larger families reported stronger anxiety feelings. In step 2, SE contributed significantly (22 percent) to the variance: Greater illness-related SE predicted milder feelings of anxiety (.46, p < .001). In step 3, role overload added 10 percent to the explanation of the variance in anxiety, evidencing a significant positive contribution (.21,p < .001). Greater role overload predicted higher anxiety. In step 4, the SC x Age interaction added 4 percent to the explanation of variance. To examine this, we divided each participant group according to median age and calculated Pearson coefficients for each group. A significant negative correlation emerged for older participants (older than age 12), r = -.33, p < .01, but not for younger participants (younger than age 12), r = -.05, p < .75. Thus, among children ages 12 to 19, greater SC predicted lower anxiety levels. [R.sup.2] was .06 (ns) for step 1, and was .28 for step 2, .38 for step 3, and .42 for step 4, p < .001. [DELTA][R.sup.2] for the four steps was .06 (ns), .22 (p < .01), .10 (p < .05), and .04 (p < .05), respectively.
The second hierarchical regression with psychosomatic symptoms as the dependent variable explained 31 percent of the variance. Sociodemographic variables (step 1) showed no significant contribution. Both personal coping resources (step 2) contributed significantly (19 percent) to the variance: SC was -.35 (p < .01) and SE was -.19 (p < .05). Higher levels of SC and illness-related SE predicted fewer psychosomatic symptoms. Role overload (step 3) added 10 percent to the explained variance of psychosomatic symptoms: Higher role overload predicted more psychosomatic symptoms. No interactions (step 4) reached statistical significance (.18, p < .05). [R.sup.2] was .02 (ns) for step 1, .21 (p < .01) for step 2, and .31 (p < .005) for step 3. [DELTA][R.sup.2] was .02 (ns) for the first step, .19 (p < .01) for the second, and .10 (p < .05) for the third.
The results of these regression analyses did not confirm this study's third hypothesis that the correlation between the stress factor and the two duress responses would be affected by the two personal coping resources. Positive correlations emerged between the stress factor and the two duress responses (anxiety, psychosomatic symptoms) as well as between both personal resources (SC and SE) and the two duress responses.
The hierarchical regression analyses added information on the contribution made by the sociodemographic variables to variance in the duress response of anxiety. Regarding gender, girls reported more anxiety than did boys. Regarding age, in older children older than 12 years, more SC predicted lower anxiety. Finally, regarding family size, a larger number of children in the family predicted higher anxiety.
The study focused on the stress factors, duress responses, and personal coping resources of HSCC ages 8 to 19. Outcomes revealed that these siblings' stress (role overload) correlated significantly with their duress responses: An experience of greater role overload was linked with higher levels of state anxiety and more psychosomatic symptoms. Likewise, these siblings' stress factor correlated significantly with one of their personal resources: Greater SC was linked with lower role overload. Furthermore, personal coping resources correlated significantly with healthy siblings' duress responses: Greater SC and SE were linked with lower levels of anxiety and fewer psychosomatic symptoms. In addition, the hierarchical regression analyses showed that, among children older than age 12, greater SC was linked with milder anxiety. Although these data do not show the influence or direction of the correlations, we suggest that the ability to use personal resources of SC and SE likely enabled the children to cope better with anxiety and therefore to present fewer behavioral symptoms and a lower rate of role overload. However, before stating this as a final outcome, more studies are needed to support this conclusion.
Regression analyses highlighted the contribution of sociodemographic variables--gender and family size--to the explained variance in siblings' anxiety. The gender outcome resembles previous studies emphasizing that girls tend to report anxiety more than boys (Hamama et al., 2000; Sahler et al., 1994). The gender differences may be rooted in different sex roles and cultural expectations, allowing girls to express their emotions more than boys. As for family size, anxiety was higher in families with more children. This outcome contradicts previous studies, which reported that larger families cope better with stress responses following sickness (Horowitz & Kazak, 1990). One explanation might relate the higher anxiety level in larger families to the increase in role overload because such families face more tasks to accomplish, and parents expect more help from the children. Also, behavior therapists suggest that talking about anxiety (rather than about coping skills) may actually increase anxiety (Meichenbaum, 1985). Perhaps in larger families the children have more opportunity to talk with their siblings about their anxiety, because parents are often absent or preoccupied. Obviously, further research is needed to examine these speculations and to determine how family size affects anxiety.
To summarize the role overload variable, which relates to the familial aspect of childhood cancer stressors, greater overload was associated with more intense duress responses. The outcomes regarding the association between high role overload and high levels of anxiety and psychosomatic symptoms support previous findings that showed a link between a change in the healthy siblings' roles at home and a high level of symptoms (Sahler et al., 1994). Other studies reported that changes in the family system were associated with behavior problems like aggression, crying, or a drop in scholastic achievement, and with emotional responses such as anxiety, anger, and envy. These findings point to the importance of maintaining routine to help children cope (Madan-Swain, Sexson, Brown, & Ragab, 1993).
The link found between the role overload experienced by HSCC and their anxious and somatic responses coincides with Novak and Guest's (1989) conceptualizations of the myriad of burdens placed on caregivers in the temporal, developmental, social, and emotional domains. Yet, it is important to understand what the healthy siblings deem significant. Park and Folkman (1997) emphasized "situational meaning," which specifically assesses events in terms of their relevance and significance for one's life. Perhaps healthy siblings experience disturbances to their routine because of their role overload at home and their relationships with the family. When these changes are assessed as relevant and significant to their development and normal daily routine, increased anxiety and symptoms are reported. Future qualitative research can help to explore the specific issues these siblings find important.
The association between role overload and stress responses can be related to social-environmental variables. Social support is a main resource helping to reduce psychological and physiological symptoms (Bloom, 2000). Social support helps human beings perceive an event as less intimidating and assists in problem solving (Bush, 1998). Murray (2000) claimed that the ability to furnish the healthy siblings' needs and help them continue their social activities is necessary for their well-being. Role overload prevents HSCC from continuing their normal routine, decreases their normal social activity, and thus may impair the social support they could receive.
The emphasis on role overload is a new direction in studies of stress within the family. This new approach proposes that not only the situation itself (the illness), but also the concrete assignments with which siblings must cope increase their distress. This outcome's importance lies in pinpointing possible directions for guiding families and helping them to help their children. More studies are needed to determine the roles of social support and routine activities as possible mediators between stress and coping.
The present finding that HSCC who were characterized by a high level of SC reported less role overload may be explained using Meichenbaum's (1985) three-stage developmental model of SC: Self-observation, occurs when individuals learn to recognize external signs of their behavior. HSCC may note changes in their environment, such as parental absences, a reduction in social and family activities, familiarization with a new environment (the hospital), entrance of people from a wider social network into the home, changes in the sick sibling's appearance, and so forth. Internal self-observation occurs when individuals learn to recognize their thoughts and the interpretations they give those thoughts. HSCC are likely preoccupied with thoughts such as "What will happen to me?," "Who will take care of me and the house?," "How will my parents manage?," and "What will happen to my sick brother/sister?." The interpretations of these thoughts may manifest themselves in worry, anxiety, sadness, and uncertainty. Formation of new patterns occurs when, through internal dialogue, the individual terminates undesirable thought patterns and forms desirable competing thoughts that give rise to new thought patterns. Possibly, healthy siblings will introduce desirable thoughts such as "I will manage by myself," "My parents will keep taking care of me," "I can get help from others, too, if I want," or "My parents are strong enough to keep our family together." In other words, HSCC with a high level of SC use external and internal information to intensify their feelings of strength and control over the situation. Hence, healthy siblings' subjective perception of changes in their roles in the aftermath of their sibling's illness may likely be affected by their ability to activate SC skills, which give them hope and confidence in their ability to deal with the novel additional roles. Therefore, children characterized by a high SC level may experience a lower sense of role overload. This outcome has important implications for intervention. Developing intervention programs to increase SC may offer a good direction for reducing children's anxiety, role overload, and duress symptoms.
No significant correlation emerged between illness-related SE and role overload. Bandura (1997) suggested that a person's belief in his or her ability to use resources influences that person's ability to comply with various assignments. A possible explanation for the lack of correlation might be that these healthy participants, whose sibling with cancer had been undergoing active treatment for an average of 6.2 months (SD = 1.64), may have already passed the stage of assessing their ability to implement the behavior required to achieve desirable results (= SE). At this stage, where the family had already geared for life "in the shadow of cancer," the HSCC may have already reached a stage of action or behavior implementation (= SC skills), aiming to cope actively with the stressful situation. In other words, HSCC routinely cope with the illness's repercussions for both the family system and their own ability to function. Therefore, an assessment of the belief in their ability to cope may be irrelevant to their current situation, which requires behaviors intended to regulate the role overload and facilitate normal functioning at both the individual and family levels. Also, Stoneman and Brody (1993) suggested that some of the tasks that children fulfill are naturally elicited during interactions with siblings, whereas some are dictated by parents. Perhaps changes in roles and tasks in the family are not related to the children's SE belief concerning their ability to complete these assignments, and they just view such tasks as part of their expected roles.
Self-Control and Self-Efficacy
The current findings whereby greater SC and SE correlated with lower anxiety and psychosomatic symptoms among the HSCC supported Hamama et al.'s (2000) reported link between strong SC and mild anxiety among these children. Bandura (1997) asserted that a person's level of belief in his or her ability to use physical, intellectual, and emotional resources affects that person's motivation to cope with tasks, performance, and results. People who do not believe in their abilities do not prepare themselves to implement effective behavior, even if they are capable of doing so: therefore, greater SE correlates with more active coping efforts (Bandura, 1997).
According to Rosenbaum (2000), SC is activated when individuals cannot continue to function in the manner in which they are accustomed, either because of a disruption in the environment or because they were acting ineffectively. The acquisition of SC skills changes perceptions of the disruptive conditions, leading to a shift from helplessness to learned resourcefulness (Meichenbaum, 1985). Learned resourcefulness consists of beliefs as well as skills and behavior, which the individual develops during life to cope with disruptive situations and factors (Rosenbaum, 2000). Hence, possibly, the correlation between SC and the duress responses may result from learned resourcefulness. Greater learned resourcefulness may be linked with stronger feelings of control over the stressful situation, lessening the arousal of anxiety responses and psychosomatic symptoms. This implies that learned resourcefulness constitutes the stage of active coping with the stressful situation, preceded by cognitive--behavioral efforts involving SC and SE. Therefore, the expectation that children characterized by high levels of SC and SE would report fewer duress responses was confirmed.
In sum, when SC and SE were greater, anxiety responses and psychosomatic symptoms were milder. These findings may elucidate how SC and SE contribute to people's ability to cope with stressful situations in general, and with an illness in the family in particular. Children are continually exposed to complex physical and emotional situations. Their level of confidence in their ability to cope with these situations gives them hope and strength to cope with them. Children who feel more confident will implement more active coping patterns (McKernon et al., 2001).
Apart from the aforesaid, the hierarchical regression analyses found that among siblings older than age 12 with a high level of SC, feelings of anxiety were milder. This finding is supported in SC theory. SC increases as children get older and expand their field of social contact and behavioral repertoire. Differences in SC between younger and older children manifest in the amount of time a child is able to continue behaving in an SC-led fashion, and in the methods the child uses to activate SC processes (Ronen, Rahav, & Rosenbaum, 2003). Therefore, the older healthy siblings are, the longer they can succeed in making those cognitive and behavioral efforts that aid them in controlling the feelings aroused by the abnormal situation (illness).
The connection between SC and role overload can also be explained from the familial point of view. Shapiro (1990) suggested that children learn about sickness and how to relate to it from parents' behaviors and parents' responses. Thus, those families who took the time to share with their children, explain situations to them, and model effective coping behaviors may have also helped increase the children's SC and their experience of less role overload. Future research should carefully study this connection to learn what variables mediate this link--social support, social control, or others.
Limitations of the Study
This study has three main methodological deficiencies: its lack of a matched control group, its focus on self-report measurements, and the use of only one question to tap SE. First, it is difficult to design a control group when the study focuses on a specific group of children facing a concrete, acute stressful situation. To address this limitation, we used means from previous studies of normative samples as a frame of reference, but future research designs would do well to attempt to find a matched control group. Previous studies using control groups in comparison with HSCC have been inconsistent. Researchers such as Dolgin and Phipps (2000) and Saltier et al. (1994) reported higher distress levels among the siblings, but others (for example, Hamama et al., 2000) found differences between siblings and a control group regarding psychosomatic symptoms only and not in their anxiety or sense of loneliness. Labay (2002) reported differences between siblings and a control group regarding academic and social achievements but not in behavior problems.
Second, concerning the validity and reliability of self-assessments, especially for children, debates have abounded. Our focus on self-assessment coincides with that of researchers who contend that measurement of thoughts, intentions, and beliefs can only be based on self-reports, especially while relating to internal events such as SC or emotional components like anxiety or internalizing symptoms (Kazdin, 1988; Rosenbaum, 2000). Future studies could add parental or teacher reports as well as direct observation to highlight differences between self-reports and other measures.
Third, our measure of SE comprised one question. Although Bandura (1997, pp. 46-47) emphasized the importance of the direct question regarding one's efficacy belief, future research should try to extend this question.
Implications for Research and Application
Despite the methodological limitations, this study on families' responses during the most stressful events of their lives also offers some unique contributions and advantages. Different from many other studies, HSCC completed this multipart questionnaire not retrospectively or prospectively but during an ongoing active stressful situation. While meeting families who are currently coping with stress, one must limit the amount and kinds of questions and time demands. Although some data are lacking, there are advantages to studying families' responses in real time.
This study's main finding focuses on the importance of SC as a coping mechanism. Current speculation regarding SE and SC requires further empirical scrutiny in a longitudinal rather than a cross-sectional study, spanning the period from the sibling's diagnosis and following development of healthy siblings' coping over ensuing months. Likewise, intervention aiming to increase SE and SC skills should use pre--post design to evaluate the intervention's impact on coping throughout disease stages.
Note also the importance of family support for children and adolescents who experience a stressful situation. Children learn how to behave and relate to disease from their parents' behavior (Shapiro, 1990). Many researchers have highlighted the importance of family in coping with a chronic disease, as well as the support provided through the family to its members (Bloom, 2000; Sourkes & Proulx, 2000). We recommend that, in the future, researchers shed more light on the contribution of familial support for children and adolescents trying to cope with stressful situations.
This study suggests two main directions for intervention: (1) facilitating SC and SE skills as coping resources, and (2) fostering family strengths to cope with a stressful situation, as a supportive and protective factor. Therefore, intervention planning at the individual, family, or group level should aim to increase SC skills and reinforce expectations of success (SE). The literature on childhood cancer describes the effectiveness of group interventions targeting healthy siblings and their parents, which emphasize emotional sharing with people undergoing similar experiences as well as the learning of problem-solving skills (Barrera, Chung, Greenberg, & Fleming, 2002). Thus, these recommendations for intervention must be supported by future research on pediatric oncology in particular, and on chronic childhood diseases in general.
Original manuscript received February 16, 2005
Final revision received February 9, 2007
Accepted August 14, 2007
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman.
Barrera, M., Chung, J. Y., Greenberg, M., & Fleming, C. (2002). Preliminary investigation of group intervention for siblings of pediatric cancer patients. Children's Health Care, 31(2), 131-142.
Beckham, J. C., Burker, E. J., Lytle, B. L., Feldman, M. E., & Costakis, M. J. (1997). Self-efficacy and adjustment in cancer patients: A preliminary report. Behavioral Medicine, 23, 138-142.
Ben-Raphael, M. (1981). Anxiety reactions of the hospitalized child and his or her parents. Unpublished master's thesis, Faculty of Humanities, Tel Aviv University. (Hebrew)
Bloom, J. R. (2000). The role of family support in cancer control. In L. Baider, C. L. Cooper, & A. K. De-Nour (Eds.), Cancer and the family (pp. 55-67). Chichester, England: John Wiley & Sons.
Bush, N. J. (1998). Coping and adaptation. In R. M. Carroll-Johnson, L. M. Gorman, & N. J. Bush (Eds.), Psychosocial nursing care along the cancer continuum (pp. 35-52). Pittsburgh: Oncology Nursing Press.
Cohen, M. S. (1999). Families coping with childhood chronic illness: A research review. Families, System & Health, 17(2), 149-164.
Dolgin, M. J., & Phipps, S. (2000). Maternal problem-solving training in childhood cancer. In L. Baider, C. L. Cooper, & A. K. De-Nour (Eds.), Cancer and the family (pp. 289-303). Chichester, England: John Wiley & Sons.
French, J.R.P., & Caplan, R. D. (1973). Organizational stress and individual strain. In A. J. Marrow (Ed.), The failure of success (pp. 30-66). New York: Amacom.
Gorman, L. M. (1998). The psychosocial impact of cancer on the individual, family, and society. In R. M. Carroll-Johnson, L. M. Gorman, & N. J. Bush (Eds.), Psychosocial nursing care along the cancer continuum (pp. 3-25). Pittsburgh: Ontology Nursing Press.
Hamama, R., Ronen, T., & Feigin, R. (2000). Self-control, anxiety, and loneliness in siblings of children with cancer. Social Work i, Health Care, 31(1), 63-85.
Horowitz, W. A., & Kazak, A. E. (1990). Family adaptation to childhood cancer: Sibling and family system variables. Journal of Clinical Child Psychology, 19, 221-228.
Kazdin, A. E. (1988). Child psychotherapy: Developing and identifying effective treatment. New York: Pergamon Press.
Koch, U., Harter, M., Jakob, U., & Siegrist, B. (1996). Parental reactions to cancer in their children. In L. Baider, C. L. Cooper, & A. Kaplan De-Nour (Eds.), Cancer and the family (pp. 149-170). Chichester, England: John Wiley & Sons.
Labay, L. E. (2002). Empathy and psychological adjustment in siblings of children with cancer. Dissertation Abstract International. 62(7-B), 3381.
Lapouse, R., & Monk, M. A. (1958). An epidemiology study of behavior characteristic in children. American Journal of Public Health, 48, 1134-1144.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer.
Madan-Swain, A., Sexson, S. B., Brown, R.T., & Ragab, A. (1993). Family adaptation and coping among siblings of cancer patients, then: brothers and sisters, and nonclinical controls. American Journal of Family Therapy, 21, 60-70.
Margalit, M. (1983). Coherence and helplessness in the learning disability syndrome. Iyunim Bechinuch, 37 & 38, 153-158. (Hebrew)
McKernon, W. L., Holmbeck, G. N., Colder, C. R., Hommeyer, J. S., Shapera, W., & Westhoven, V. (2001). Longitudinal study of observed and perceived family influences on problem-focused coping behaviors of preadolescents with spina bifida. Journal of Pediatric Psychology, 26, 41-54.
Meichenbaum, D. H. (1985). Stress inoculation training. New York: Pergamon Press.
Murray, J.S. (2000). Attachment theory and adjustment difficulties in siblings of children with cancer. Issues in Mental Health Nursing,, 21, 149-169.
Novak, M., & Guest, C. (1989). Application of a multidimensional caregiver burden inventory. Journal of Gerontologist, 29, 798-803.
Park, C. L., & Folkman, S. (1997). Meaning in the context of stress and coping. Review of General Psychology, 1(2), 115-144.
Rahav, G., & Ronen, T. (1994). Children's perception of their behavior problems during the Gulf War. Anxiety, Stress & Coping, 7, 241-252.
Ronen, T., Rahav, G., & Appel, N. (2003). Adolescent stress response to a single acute stress and to continuous external stress: Terrorist attack. Journal of Loss and Trauma, 8, 261-282.
Ronen, T., Rahav, G., & Rosenbaum, M. (2003). Children's response to war situation as determined by variables of age and sex. Anxiety, Stress and Coping, 16, 59-69.
Rosenbaum, M. (2000). The self-regulation of experience: Openness and construction. In P. Dewe, A. M. Leiter, & T. Cox (Eds.), Coping and health and organizations (pp. 51-67). London: Taylor & Francis.
Rosenbaum, M., & Ronen, T. (1990 November, 10). The development of the Children's Self-Control Scale. Paper presented at the American Association of Behavior Therapy. New York.
Sahler, O.J.Z., Roghmann, K. J., Mulhern, R. K., Carpenter. P. J., Sargent, J. R., Copeland, D. R., Barbarin, O. A., Zeltzer, L. K., & Dolgin, M. J. (1994). Sibling adaptation to childhood cancer collaborative study: The association of sibling adaptation with maternal well-being, physical health, and resource use. Developmental and Behavioral Pediatrics, 18, 233-243.
Shapiro, J. (1990). Family reaction and coping strategies in response to the physically in handicapped child: A review. In M. Nagler (Ed.), Perspectives on disabilities (pp. 260-280). Oxford: Pergamon Press.
Sourkes, B. M., & Proulx, R. (2000). "My family and I are in this together": Children with cancer speak out. In L. Baider, C. L. Cooper, & A. K. De-Nour (Eds.), Cancer and the family (pp. 273-284). Chicester. England: John Wiley & Sons.
Spielberger, C. D. (1972). Anxiety: Current trends in theory and research. New York: Academic Press.
Spielberger, C. D.. Edwards, C. D., Montouri, J., & Lushene, R. (1970). The state trait anxiety inventory for children. Palo Alta, CA: Consulting Psychologists Press.
Stoneman, Z., & Brody, G. H. (1993). Sibling relations in the family concept. In Z. Stoneman, & P.W. Berman (Eds.), The effect of mental retardation disability and illness on sibling relationships. (pp. 3-30). New York: Paul H. Brookes.
Teichman, Y., & Melnick, H. (1979). SHACHMAT: State and trait anxiety assessment questionnaire. Tel Aviv, Israel: Department of Psychology, Tel Aviv University. (Hebrew)
Zarit, J. M., & Zarit, S. H. (1982, November). Measuring burden and support in families with Alzheimer's disease elders. Paper presented at the Meeting of the Gerontological Society of America, Boston.
Liat Hamama, PhD, is a lecturer, Tammie Ronen, PhD, is professor, and Giora Rahav, PhD, is professor, Bob Shapell School of Social Work, Tel Aviv University, Ramat Aviv, Tel Aviv, Israel 69978; e-mail: email@example.com.
Table 1: Study Variables' Means, Standard Deviations, and Ranges for Healthy Siblings of a Child with Cancer, and Prior Means for Normative Samples Current Study M from Prior Variable M SD Range Studies Role overload 2.03 66.00 1-5 -- State anxiety 1.79 56.00 1-4 3.59 (a), 2.60 (b) Psychosomatic symptoms 25.30 3.62 15-36 21.75 (c), 24.75 (b) Self-control 4.00 0.65 1-6 3.60 (a), 5.47 (d) Self-efficacy 7.90 2.17 1-10 -- Note: Dashes indicate that there are no data from earlier studies. (a) (Hamama, Ronen, & Feigin, 2000). (b) (Rouen, Rahav, & Appel, 2003). (c) (Rahav & Ronen, 1994). (d) (Rosenhaum & Ronen, 1990).
|Printer friendly Cite/link Email Feedback|
|Author:||Hamama, Liat; Ronen, Tammie; Rahav, Giora|
|Publication:||Health and Social Work|
|Date:||May 1, 2008|
|Previous Article:||Cultural competence in a group intervention designed for Latino patients living with HIV/AIDS.|
|Next Article:||Perceptions of biopsychosocial services needs among older adults with severe mental illness: met and unmet needs.|