Journal ofPediatric Psychology, Vol. 17, No. 6, 1992, pp. 705-724
Functioning Among Mothers and Fathers of Children with Juvenile Rheumatic Disease: A Longitudinal Study1
Department of Veterans Affairs and Stanford University Medical Centers Received November 1, 1991; accepted May 12, 1992
Examined the adaptation of mothers and fathers of children with juvenile rheumatic disease on two occasions, 1 year apart, using 159 married couples at Time 1, and 111 of these couples at Time 2. A stress and coping model was tested in which parental functioning is determined by ongoing life stressors (patient and spouse dysfunction), family resources, and parents' illness-related coping. Mothers reported more depression than fathers did. However, poorer concurrent functioning among both mothers and fathers was explained partly by patients having more functional disability, pain, and psychosocial problems. In addition, spouse's dysfunction and the parent's use of avoidance coping were related to poorer parental adaptation, both concurrently and 1 year later. The implications of the findings for developing stress and coping models of parental adaptation to having a chronically ill child, and for intervention strategies with parents, patients, and families, are discussed. KEY WORDS: parental functioning; juvenile rheumatic disease; chronically ill children; spouse functioning; coping.
•This research was supported by Multipurpose Arthritis Center Grant AM20610 from the National Institute of Health and Department of Veterans Affairs Medical and Health Services Research and Development Services research funds. We thank John J. Miller for his help in recruiting the patients and his overall support for the project. We also thank Martha Baumgartner for her help in preparing this paper. 2 A1I correspondence should be sent to Christine Timko, Veterans Administration Medical Center (152), 3801 Miranda Avenue, Palo Alto, California 94304. 705 0I46-8693/92/I20O-O705SO6.5O/0 © 1992 Plenum Publishing Corporation
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Christine Timko,2 Katherine W. Stovel, and Rudolf H. Moos
706
Timko, Stovel, and Moos
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Chronic childhood disease or disability has ramifications for all members of the patient's family (Kazak, 1989) and increases the likelihood of psychosocial impairment among family members (Breslau, 1983). Most studies of the effects of chronic childhood illness on parental functioning have focused on the mothers of patients; very little is known about fathers' adjustment to caring for a chronically ill child. This study examined the adaptation of mothers and fathers of children with juvenile rheumatic disease on two occasions, 1 year apart. Juvenile rheumatic disease (JRD) refers to a group of conditions that are believed to be caused by abnormal immunological mechanisms that affect the connective tissues, particularly those, of the joints. Chronic inflammation and swelling of joints cause pain and stiffness and may cause permanent damage. Symptoms vary from day to day and from morning to afternoon. Although the causes of JRD are unknown, and its course is difficult to predict, treatment generally includes anti-inflammatory medication and balanced programs of rest and exercise (Miller, 1983). One purpose of the present study was to compare the mothers and fathers of the JRD patients on specific domains of functioning. A second purpose was to focus on the variation in the adaptation of mothers of chronically ill children (Kovacs et al., 1990; Wallander, Pitt, & Mellins, 1990). In light of the variability in outcome, it is important to identify factors that are related to adjustment, among both mothers and fathers of children with chronic disease. We attempted to identify concurrent and predictive factors associated with parents' functioning. Further, we examined the extent to which personal and family resources contributed to parents' adjustment when the chronic stressors of poor patient and spouse functioning were already considered.
MOTHERS VERSUS FATHERS Mothers of chronically ill children appear to have more difficulty than fathers do, suggesting that mothers are at higher risk of poor adaptation and have more need of formal support services. Specifically, mothers of chronically ill or developmentally disabled children score lower than fathers on sociability, are more depressed, and are at higher risk for stress-related psychopathology; mothers also perceive the child's illness as creating more problems for the family (Bristol, Gallagher, & Schopler, 1988; Kazak & Marvin, 1984; Tavormina, Boll, Dunn, Luscomb, & Taylor, 1981). These results may occur because mothers carry the brunt of the burden, both functionally and psychologically, for caring for chronically ill children (Bristol et al., 1988; Johnson, 1988). We compared mothers and fathers of children with JRD on four domains of functioning. The domains included depressed mood, social activities, and two aspects of functioning that were directly related to having a child with JRD: the
Functioning Among Parents of Children with JRD
707
personal strain parents experience due to the demands of the illness, and parents' sense of mastery over the stressors associated with having a chronically ill child. Thus the functioning criteria included both symptom-oriented indices (depressed mood and personal strain) and positive aspects of functioning (social activity and mastery).
A STRESS AND COPING MODEL OF PARENTAL ADAPTATION
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Several conceptual models have been proposed to account for the differential adaptation of chronically ill and handicapped children and their parents (Bristol, 1987; Kovacs et al., 1990; Vami & Wallander, 1988; Wallander & Varni, 1992). In these models, risk factors for poor adaptation include demographic characteristics, severity of the child's disease or disability, and the child's psychosocial status; resistance factors include patterns of coping and family environment. Among the resistance factors, coping is generally considered a more proximal influence on adjustment than family environment and other social-ecological influences. We drew on this previous work and on stress and coping theory (Lazarus & Folkman, 1984) to develop the conceptual model of parents' adaptation to having a child with JRD that guided this study. According to stress and coping theory, the adaptational consequences of an ongoing life stressor, such as having a chronically ill child, are influenced by characteristics of this ongoing stressor, other family-related stressors, stressor-related coping responses, and family resources. Our model is shown in Figure 1. The figure shows the hypothesized determinants of mothers' functioning; a parallel model applies to explain fathers' functioning. Solid arrows depict hypotheses that are tested here regarding the determinants of parents' functioning. Broken arrows depict our hypotheses regarding associations among the sets of determinants; however, these potential associations were not examined. In the model, mothers' functioning is determined in part by demographic factors. Mothers' functioning is also determined by the characteristics of the chronic stressor (i.e., the patient's disease-related and psychosocial functioning) as well as by the spouse's dysfunction. Further, the mother's functioning is determined by how she copes with the child having JRD. Finally, family resources are hypothesized to influence mothers' adjustment. Although not shown in the figure, relationships between parent functioning and its determinants may be bidirectional; in particular, parental dysfunction can influence the patient's psychosocial functioning (Timko, Stovel, Moos, & Miller, 1992b). Demographic Factors. The study examined the relationships between parents' functioning and the patient's age and gender, the parent's own age and education, and the family's annual income and size. We expected that parents'
Timko, Stovel, and Moos
708
graphic Faidors
y
Related Family Stressors Father's Dysfunction 1
Focal Stressor Characteristics Patient1 3 Disease Faidors iI \r Patient's Psychosocial Dysfunotion
Mother's Functiorling
t Mother's IllnessRelated Coping
/ /
1
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Family Resources
Note: (a) A conceptually comparable model applies to fathers. (b) The hypotheses tested in the present study are depicted by the solid arrows.
Fig. 1. Stress and coping model of the determinants of functioning among mothers of children with chronic disease.
functioning might be aided by having more personal and social resources in the form of a longer life history from which to draw, higher educational and income levels, and only one child to care for. Regarding family size, we found that JRD patients who were "only children" were at less risk for negative psychosocial outcomes than patients who had siblings (Timko, Stovel, Moos, & Miller, 1992a), possibly because only children tend to have more positive relationships with their parents (Falbo & Polit, 1986). Duration of Disease, Disease-Related Functioning, and Psychosocial Adaptation. We examined the associations between parents' functioning and indicators of the severity of the stressor of having a child with JRD: the duration of the patients' JRD, and the patients' disease-related outcomes and psychosocial adaptation. Specifically, the disease-related outcomes of functional disability and pain were assessed, because these are the central dimensions of patients' experience of rheumatic diseases such as arthritis (Brown et al., 1984; Fries, Spitz, & Young, 1982). We expected that a less severe stressor, indicated by better disease-related and psychosocial outcomes, would be associated with better parent functioning. In addition, JRD (and other chronic childhood diseases) may become less disruptive as time since diagnosis lengthens, because the patient and family adapt over time to managing the disorder (Kovacs et al., 1990; Timko et al., 1992a). Spouse Dysfunction. The study investigated relationships between each par-
Functioning Among Parents of Children with JRD
709
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
ent's functioning and his or her spouse's functioning in the same domains. Because the spouse's dysfunction is an additional ongoing stressor, we expected poor adaptation by one's marital partner to be related to poorer adaptation by oneself. Research on parents of chronically ill children has shown that spousal support facilitates mothers' and fathers' adjustment to this situation (Bristol et al., 1988). Illness-Related Coping. We also studied the associations between parental functioning and methods of coping. Specifically, approach and avoidance methods of coping (Billings & Moos, 1981) were assessed with respect to the stressors associated with having a child with JRD. In the context of the present study, approach coping refers to cognitive attempts to manage one's appraisal of the stressfulness of one's child having JRD, as well as behavioral attempts to deal directly with the child's JRD and its effects. Avoidance coping consists of attempts to avoid confronting the child's having JRD or to indirectly reduce emotional tension (e.g., by eating or smoking). Generally, approach coping should be associated with good adjustment to having a child with a medical disorder, because it facilitates the management of emotional distress, seeking out and considering alternative strategies for accommodating to and handling problems, and receiving positive feedback from effective coping. Reliance on avoidance coping should be linked with poor adjustment, because the child's health problems cannot be confronted directly and so necessary accommodations are not made. Studies of medical and psychiatric patients have found approach coping related to good adjustment, and avoidance coping associated with dysfunction (Keyes, Bisno, Richardson, & Marston, 1987; Moos, Finney, & Cronkite, 1990; Namir, Wolcott, Fawzy, & Alumbaugh, 1987; Roberts et al., 1987). Family Resources. We examined the relationships between family resources and parents' adaptation, reasoning that the availability of more family resources to support parents emotionally in the face of an ongoing stressor should reduce the risk of parental dysfunction. More family resources of cohesion and expressiveness, as well as reduced conflict among family members, are associated with better adjustment and treatment outcomes among chronically ill children (Anderson, Miller, Auslander, & Santiago, 1981; Hauser, Jacobson, Wertlieb, Brink, & Wentworth, 1985; Hauser et al., 1990). In summary, the purposes of this study were threefold. The first purpose was to compare the functioning of mothers and fathers of children with JRD. The second aim was to examine concurrent relationships between parental adaptation and its potential determinants, including demographic factors; the focal stressor, as indexed by the patient's functioning; other stressors, as reflected in spouse dysfunction; illness-related coping; and family resources. Specifically, we determined the extent to which the resistance factors of coping responses and family resources explained parental adaptation above and beyond the stressors associated with patient and spouse dysfunction. The third purpose of the study was to
710
Timko, Stovel, and Moos
similarly examine the relationships of the initial stressors and resources to parental functioning assessed at a 1-year follow-up.
METHODS Sample
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
A sample of 204 families of children with diagnosed rheumatic disease was recruited from the Pediatric Rheumatology Clinic at Children's Hospital at Stanford. All patients entered in the study (71 boys and 133 girls; mean age = 9.4 years) were diagnosed by a pediatric rheumatologist and met American Rheumatology Association (ARA) diagnostic criteria (Brewer et al., 1977) that include both physical examination and laboratory findings. Most of the patients (71%) had a form of juvenile arthritis (i.e., systemic juvenile, polyarticular, or pauciarticular arthritis); the rest (29%) were diagnosed with other forms of rheumatic disease, such as ankylosing spondylitis, systemic lupus erythematosus, and dermatomyositis. Of 224 consecutive families in which the patient met the diagnostic criteria, 204 (91%) agreed to participate. Project staff met at the clinic with each participating family to explain the study and obtain informed consent. Questionnaires were distributed to families who signed informed consent forms, which were returned by mail. Due to time-limited funding, we attempted to follow only the first 165 of the 204 families 1 year later. At Time 2, questionnaires were distributed during a clinic visit, if the timing of the visit coincided with the 1-year follow-up; otherwise, to save participants an unnecessary trip to the clinic, the family was contacted by telephone, and questionnaires were distributed by mail. At Time 1, 197 mothers (99% of 199 mothers who were living with patients) and 164 fathers (95% of 173 fathers who were living with patients) provided data; at Time 2, 141 mothers (99% of 143 mothers who were located and still living with patients) and 115 fathers (95% of 121 fathers who were located and still living with patients) returned questionnaires. All analyses reported in this paper are based on members of married couples in which both partners returned data: 159 mothers and fathers at Time 1, and 111 mothers and fathers at Time 2. To ensure that the parents who participated at Time 2 were representative of the initial sample, we conducted / tests on the initial functioning indices between mothers who provided data at both Time 1 and Time 2 and mothers who provided only Time 1 data; corresponding analyses were conducted for fathers. No differences were found between either set of parents. T tests were also conducted between mothers of patients with juvenile arthritis and mothers of patients with other rheumatic diseases on the Time 1 and
Functioning Among Parents of Children with JRD
711
Time 2 psychosocial outcomes; the same t tests were conducted for fathers. At Time 2, fathers of patients who had juvenile arthritis engaged in fewer social activities (A/s = 3.7 vs. 5.1) f(109) = 2.0; p < .05; otherwise, there were no significant differences between groups. On the basis of these and previous results finding minimal differences between juvenile arthritis patients and patients with other rheumatic diseases, and their families (Timko et al., 1992a, 1992b), parents of all patients were analyzed together. Measures Parent Functioning Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
The outcomes examined in this study reflected the parent's assessment of his or her own functioning in four areas. Two of the outcome measures were taken from the Health and Daily Living Form (HDL; Moos, Cronkite, & Finney, 1990), which mothers and fathers completed at both Time 1 and Time 2. The HDL indices have good validity and reliability, and have been used with over 1,500 families and in more than 100 studies (Moos, Cronkite, & Finney, 1990). The two HDL outcome measures were Depressed Mood (Cronbach's alpha = .93 at Time 1 and at Time 2) is the sum of 18 items measured on a 5-point scale ranging from never (0) to often (4) with reference to the past month; the items tap the presence and severity of symptoms involved in obtaining an RDC (see Spitzer, Endicott, & Robins, 1978) diagnosis of major or minor depression (e.g., feeling sad or blue, poor appetite or weight loss, sleeping problems). Social Activities (alpha = .79 at Time 1, .77 at Time 2) is the number of activities (out of 12) that the parent noted he or she had done or attended with friends in the past month (e.g., athletic events, card games, parties). The two additional outcome measures were taken from the Family Effects of Illness scale (Stein & Reissman, 1980), which was also completed by both mothers and fathers at Time 1 and Time 2: Personal Strain (alpha = .93 at Time 1, .85 at Time 2) is the sum of seven items answered on a 4-point scale ranging from strongly disagree (0) to strongly agree (3) that assess the parent's strain resulting from demands of the child's illness (e.g., fatigue is a problem for me because of my child's illness; I worry about what will happen to my child in the future when he/she grows up or when I'm not around). Mastery (alpha = .82 at Time 1, .69 at Time 2) is the sum of five items answered on the same 4-point scale that assess the parent's view of his or her own and the family's ability to handle the stressors associated with the child's illness (e.g., learning to manage my child's illness has made me feel better about myself; because of what we have shared we are a closer family).
712
Timko, Stovel, and Moos
Determinants of Parent Functioning
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Demographic Factors. Demographic factors were assessed on each parent's HDL at Time 1. They included whether or not the patient was an only child; the family's total annual income; the parent's education and age; and the patient's age and gender. Patient Dysfunction. The patient's disease Duration was measured at Time 1 by the pediatric rheumatologist as number of years since disease onset. The patient's disease-related dysfunction was assessed at Time 1 on the Child Health Assessment Questionnaire (Child HAQ; Billings, Moos, Miller, & Gottlieb, 1987). The Child HAQ is adapted from the Adult HAQ (Brown et al., 1984; Fries et al., 1982), which assesses arthritis disability and pain among adults. Patients who were old enough to self-report at Time 1 (i.e., were at least 10 years old) completed the child version of the Child HAQ. If the patient completed a Child HAQ, his or her Functional Disability and Pain scores were used; otherwise, the mother's scores on the parent version of the Child HAQ were used. The Functional Disability index is based on items assessing difficulty in the past week, rated on a 4-point scale ranging from no difficulty (0) to unable to do (3), in eight categories (e.g., dressing and grooming, eating, walking). The scoring procedure is similar to that for the Adult HAQ (for details, see Fries et al., 1982), and results in a continuous scale from 0 to 3, with 3 representing the most severe disability. The Pain scale is a visual analog scale that measures pain due to rheumatic disease during the past week. Responses are indexed on a scale ranging from no pain (0) to pain as bad as it could be (3). Mothers and fathers completed an eight-item index on the HDL at Time 1 to describe the patient's psychosocial dysfunction. The index assesses the number of Psychosocial Problems (alpha = .73) during the past year (e.g., mental or emotional problems; problems getting along with other children). Spouse Dysfunction. Spouse dysfunction at Time 1 consisted of the spouse's responses to the same four indices used to assess parent functioning, that is, Depressed Mood, Social Activities, Personal Strain, and Mastery. Illness-Related Coping. Mothers' and fathers' methods of coping at Time 1 were measured on the HDL. Parents were asked to indicate what they did in connection with their child having rheumatic disease. Items measured the frequency of use on a 4-point scale of 32 coping responses. Approach Coping (alpha = .88) consists of items assessing active cognitive (e.g., prepared for the worst, tried to see the positive side of the situation) and active behavioral (e.g., tried to find out more about the situation, made a plan of action and followed it) coping. Avoidance Coping (alpha = .66) consists of items such as: refused to believe that it happened; and, tried to reduce tension by drinking more. Family Resources. The Family Resources index at Time 1 was assessed by both mothers and fathers (alpha = .69). It was a sum of three subscales: Cohesion (the degree of commitment, help, and support family members provide for
Functioning Among Parents of Children with JRD
713
one another); Expressiveness (the extent to which family members are encouraged to act openly and express their feelings directly) from the Family Environment Scale (Moos & Moos, 1986); and Family Arguments (reverse scored) from the HDL. The Cohesion and Expressiveness subscales are each the sum of nine true-false items. Family Arguments is the count of up to 14 items (e.g., relatives, helping with household tasks, money) that the parent marked as often causing disagreements in the family.
RESULTS Comparison of Mothers and Fathers
Concurrent Analyses of Parent Functioning First, we conducted multiple regression analyses in which the demographic factors were used to predict each aspect of functioning assessed at Time 1, for Table I. / Tests Comparing Mothers' and Fathers' Functioning" Mothers Functioning indices Time 1* Depressed mood Personal strain Social activities Mastery Time 2C Depressed mood Personal strain Social activities Mastery
Fathers
M
SD
M
SD
20.0 7.4 4.3 10.5
11.6 4.0 2.6 2.1
14.8 6.6 4.1 10.0
10.9 3.7 2.6 2.1
4.1&/" 1.84 0.70 1.86
19.5 6.1 4.4 10.3
12.2 4.0 2.3 2.5
14.8 5.8 3.9 9.6
10.6 4.1 2.7 3.0
3.04' 0.48 1.38 2.JC
"When the Bonferroni method is applied to consider that four comparisons were made at Time 1 and Time 2, alpha = .012. b n = 159 for each parent group. c n = 111 for each parent group. a p < .05. 'p < .01. fp < .001.
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Table I reports the results of / tests that compared mothers with fathers on functioning at Time 1 and Time 2. At Time 1 and Time 2, mothers experienced more depressed mood than fathers did. In addition, at Time 2, mothers experienced more mastery. No differences were found between mothers and fathers on personal strain or on social activities.
714
Timko, Stovel, and Moos
mothers and fathers separately. The patient's age and gender, the parent's age and education, and the family's annual income and size all failed to show significant associations with comparable aspects of both mothers' and fathers' functioning. Three scattered associations were significant. For mothers, having an older child (b = .27) and more education (b — .20) were related to more social activity; for fathers, having more than one child was related to more mastery (b = . 18, ps < .05). On the basis of these results, demographic factors were dropped from subsequent analyses. Mothers
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Table II presents the results of hierarchical multiple regression analyses using Time 1 determinants to predict each aspect of mothers' functioning at Time 1. In these regressions, indices of patient dysfunction were entered on Step 1, spouse dysfunction was entered on Step 2, mothers' illness-related coping responses were entered on Step 3, and the family resources index was entered on Step 4. When zero-order correlations were computed among mothers' Time 1 dependent variables, a significant positive association was found between depressed mood and personal strain (r = .41, p < .001); therefore, for each step in the regression analyses predicting depression and personal strain, the R2 change was considered significant only at p < .025. The results for patient dysfunction showed that patients' psychosocial problems were associated with mothers reporting more depression and personal strain and less mastery; patients' functional disability was also associated with mothers' personal strain. When patients had JRD for a longer duration and experienced less pain, mothers engaged in more social activities. The father's functioning was significantly positively related to each corresponding aspect of mother's functioning. Regarding mothers' illness-related coping, more use of approach coping was related to more social activity, while more reliance on avoidance coping was related to more depression and personal strain. When patient and spouse functioning factors were controlled, mothers' illness-related coping explained significant additional variance in depression, personal strain, and social activity. Finally, mothers in families with more resources experienced less depression and strain and more mastery. Taken together, patient, spouse, coping, and family factors explained 38% of the variance in depression, 50% in personal strain, 20% in social activity, and 29% in mastery. Fathers Table II presents parallel hierarchical multiple regression analyses for the fathers. As for mothers, intercorrelations among fathers' Time 1 functioning
"p < .001.
"p < .05. *p < .025. c p < .01.
Overall R Overall R2
Patients' dysfunction Duration Functional disability Pain Psychosocial problems Mothers' dysfunction Corresponding functioning index Fathers' coping Approach coping Avoidance coping Family resources
Overall R Overall R2
Patients' dysfunction Duration Functional disability Pain Psychosocial problems Fathers' dysfunction Corresponding functioning index Mothers' coping Approach coping Avoidance coping Family resources
-.18* .66" .44
.07 .47"
.29"
.17° .16 -.03 .14
.38
-.22' .62"
-.03 .30"
.28"
,01 -.03 .45"
.00
b
b
R2 change
.io
Functioning Among Mothers and Fathers of Children with Juvenile Rheumatic Disease: A Longitudinal Study1
Department of Veterans Affairs and Stanford University Medical Centers Received November 1, 1991; accepted May 12, 1992
Examined the adaptation of mothers and fathers of children with juvenile rheumatic disease on two occasions, 1 year apart, using 159 married couples at Time 1, and 111 of these couples at Time 2. A stress and coping model was tested in which parental functioning is determined by ongoing life stressors (patient and spouse dysfunction), family resources, and parents' illness-related coping. Mothers reported more depression than fathers did. However, poorer concurrent functioning among both mothers and fathers was explained partly by patients having more functional disability, pain, and psychosocial problems. In addition, spouse's dysfunction and the parent's use of avoidance coping were related to poorer parental adaptation, both concurrently and 1 year later. The implications of the findings for developing stress and coping models of parental adaptation to having a chronically ill child, and for intervention strategies with parents, patients, and families, are discussed. KEY WORDS: parental functioning; juvenile rheumatic disease; chronically ill children; spouse functioning; coping.
•This research was supported by Multipurpose Arthritis Center Grant AM20610 from the National Institute of Health and Department of Veterans Affairs Medical and Health Services Research and Development Services research funds. We thank John J. Miller for his help in recruiting the patients and his overall support for the project. We also thank Martha Baumgartner for her help in preparing this paper. 2 A1I correspondence should be sent to Christine Timko, Veterans Administration Medical Center (152), 3801 Miranda Avenue, Palo Alto, California 94304. 705 0I46-8693/92/I20O-O705SO6.5O/0 © 1992 Plenum Publishing Corporation
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Christine Timko,2 Katherine W. Stovel, and Rudolf H. Moos
706
Timko, Stovel, and Moos
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Chronic childhood disease or disability has ramifications for all members of the patient's family (Kazak, 1989) and increases the likelihood of psychosocial impairment among family members (Breslau, 1983). Most studies of the effects of chronic childhood illness on parental functioning have focused on the mothers of patients; very little is known about fathers' adjustment to caring for a chronically ill child. This study examined the adaptation of mothers and fathers of children with juvenile rheumatic disease on two occasions, 1 year apart. Juvenile rheumatic disease (JRD) refers to a group of conditions that are believed to be caused by abnormal immunological mechanisms that affect the connective tissues, particularly those, of the joints. Chronic inflammation and swelling of joints cause pain and stiffness and may cause permanent damage. Symptoms vary from day to day and from morning to afternoon. Although the causes of JRD are unknown, and its course is difficult to predict, treatment generally includes anti-inflammatory medication and balanced programs of rest and exercise (Miller, 1983). One purpose of the present study was to compare the mothers and fathers of the JRD patients on specific domains of functioning. A second purpose was to focus on the variation in the adaptation of mothers of chronically ill children (Kovacs et al., 1990; Wallander, Pitt, & Mellins, 1990). In light of the variability in outcome, it is important to identify factors that are related to adjustment, among both mothers and fathers of children with chronic disease. We attempted to identify concurrent and predictive factors associated with parents' functioning. Further, we examined the extent to which personal and family resources contributed to parents' adjustment when the chronic stressors of poor patient and spouse functioning were already considered.
MOTHERS VERSUS FATHERS Mothers of chronically ill children appear to have more difficulty than fathers do, suggesting that mothers are at higher risk of poor adaptation and have more need of formal support services. Specifically, mothers of chronically ill or developmentally disabled children score lower than fathers on sociability, are more depressed, and are at higher risk for stress-related psychopathology; mothers also perceive the child's illness as creating more problems for the family (Bristol, Gallagher, & Schopler, 1988; Kazak & Marvin, 1984; Tavormina, Boll, Dunn, Luscomb, & Taylor, 1981). These results may occur because mothers carry the brunt of the burden, both functionally and psychologically, for caring for chronically ill children (Bristol et al., 1988; Johnson, 1988). We compared mothers and fathers of children with JRD on four domains of functioning. The domains included depressed mood, social activities, and two aspects of functioning that were directly related to having a child with JRD: the
Functioning Among Parents of Children with JRD
707
personal strain parents experience due to the demands of the illness, and parents' sense of mastery over the stressors associated with having a chronically ill child. Thus the functioning criteria included both symptom-oriented indices (depressed mood and personal strain) and positive aspects of functioning (social activity and mastery).
A STRESS AND COPING MODEL OF PARENTAL ADAPTATION
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Several conceptual models have been proposed to account for the differential adaptation of chronically ill and handicapped children and their parents (Bristol, 1987; Kovacs et al., 1990; Vami & Wallander, 1988; Wallander & Varni, 1992). In these models, risk factors for poor adaptation include demographic characteristics, severity of the child's disease or disability, and the child's psychosocial status; resistance factors include patterns of coping and family environment. Among the resistance factors, coping is generally considered a more proximal influence on adjustment than family environment and other social-ecological influences. We drew on this previous work and on stress and coping theory (Lazarus & Folkman, 1984) to develop the conceptual model of parents' adaptation to having a child with JRD that guided this study. According to stress and coping theory, the adaptational consequences of an ongoing life stressor, such as having a chronically ill child, are influenced by characteristics of this ongoing stressor, other family-related stressors, stressor-related coping responses, and family resources. Our model is shown in Figure 1. The figure shows the hypothesized determinants of mothers' functioning; a parallel model applies to explain fathers' functioning. Solid arrows depict hypotheses that are tested here regarding the determinants of parents' functioning. Broken arrows depict our hypotheses regarding associations among the sets of determinants; however, these potential associations were not examined. In the model, mothers' functioning is determined in part by demographic factors. Mothers' functioning is also determined by the characteristics of the chronic stressor (i.e., the patient's disease-related and psychosocial functioning) as well as by the spouse's dysfunction. Further, the mother's functioning is determined by how she copes with the child having JRD. Finally, family resources are hypothesized to influence mothers' adjustment. Although not shown in the figure, relationships between parent functioning and its determinants may be bidirectional; in particular, parental dysfunction can influence the patient's psychosocial functioning (Timko, Stovel, Moos, & Miller, 1992b). Demographic Factors. The study examined the relationships between parents' functioning and the patient's age and gender, the parent's own age and education, and the family's annual income and size. We expected that parents'
Timko, Stovel, and Moos
708
graphic Faidors
y
Related Family Stressors Father's Dysfunction 1
Focal Stressor Characteristics Patient1 3 Disease Faidors iI \r Patient's Psychosocial Dysfunotion
Mother's Functiorling
t Mother's IllnessRelated Coping
/ /
1
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Family Resources
Note: (a) A conceptually comparable model applies to fathers. (b) The hypotheses tested in the present study are depicted by the solid arrows.
Fig. 1. Stress and coping model of the determinants of functioning among mothers of children with chronic disease.
functioning might be aided by having more personal and social resources in the form of a longer life history from which to draw, higher educational and income levels, and only one child to care for. Regarding family size, we found that JRD patients who were "only children" were at less risk for negative psychosocial outcomes than patients who had siblings (Timko, Stovel, Moos, & Miller, 1992a), possibly because only children tend to have more positive relationships with their parents (Falbo & Polit, 1986). Duration of Disease, Disease-Related Functioning, and Psychosocial Adaptation. We examined the associations between parents' functioning and indicators of the severity of the stressor of having a child with JRD: the duration of the patients' JRD, and the patients' disease-related outcomes and psychosocial adaptation. Specifically, the disease-related outcomes of functional disability and pain were assessed, because these are the central dimensions of patients' experience of rheumatic diseases such as arthritis (Brown et al., 1984; Fries, Spitz, & Young, 1982). We expected that a less severe stressor, indicated by better disease-related and psychosocial outcomes, would be associated with better parent functioning. In addition, JRD (and other chronic childhood diseases) may become less disruptive as time since diagnosis lengthens, because the patient and family adapt over time to managing the disorder (Kovacs et al., 1990; Timko et al., 1992a). Spouse Dysfunction. The study investigated relationships between each par-
Functioning Among Parents of Children with JRD
709
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
ent's functioning and his or her spouse's functioning in the same domains. Because the spouse's dysfunction is an additional ongoing stressor, we expected poor adaptation by one's marital partner to be related to poorer adaptation by oneself. Research on parents of chronically ill children has shown that spousal support facilitates mothers' and fathers' adjustment to this situation (Bristol et al., 1988). Illness-Related Coping. We also studied the associations between parental functioning and methods of coping. Specifically, approach and avoidance methods of coping (Billings & Moos, 1981) were assessed with respect to the stressors associated with having a child with JRD. In the context of the present study, approach coping refers to cognitive attempts to manage one's appraisal of the stressfulness of one's child having JRD, as well as behavioral attempts to deal directly with the child's JRD and its effects. Avoidance coping consists of attempts to avoid confronting the child's having JRD or to indirectly reduce emotional tension (e.g., by eating or smoking). Generally, approach coping should be associated with good adjustment to having a child with a medical disorder, because it facilitates the management of emotional distress, seeking out and considering alternative strategies for accommodating to and handling problems, and receiving positive feedback from effective coping. Reliance on avoidance coping should be linked with poor adjustment, because the child's health problems cannot be confronted directly and so necessary accommodations are not made. Studies of medical and psychiatric patients have found approach coping related to good adjustment, and avoidance coping associated with dysfunction (Keyes, Bisno, Richardson, & Marston, 1987; Moos, Finney, & Cronkite, 1990; Namir, Wolcott, Fawzy, & Alumbaugh, 1987; Roberts et al., 1987). Family Resources. We examined the relationships between family resources and parents' adaptation, reasoning that the availability of more family resources to support parents emotionally in the face of an ongoing stressor should reduce the risk of parental dysfunction. More family resources of cohesion and expressiveness, as well as reduced conflict among family members, are associated with better adjustment and treatment outcomes among chronically ill children (Anderson, Miller, Auslander, & Santiago, 1981; Hauser, Jacobson, Wertlieb, Brink, & Wentworth, 1985; Hauser et al., 1990). In summary, the purposes of this study were threefold. The first purpose was to compare the functioning of mothers and fathers of children with JRD. The second aim was to examine concurrent relationships between parental adaptation and its potential determinants, including demographic factors; the focal stressor, as indexed by the patient's functioning; other stressors, as reflected in spouse dysfunction; illness-related coping; and family resources. Specifically, we determined the extent to which the resistance factors of coping responses and family resources explained parental adaptation above and beyond the stressors associated with patient and spouse dysfunction. The third purpose of the study was to
710
Timko, Stovel, and Moos
similarly examine the relationships of the initial stressors and resources to parental functioning assessed at a 1-year follow-up.
METHODS Sample
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
A sample of 204 families of children with diagnosed rheumatic disease was recruited from the Pediatric Rheumatology Clinic at Children's Hospital at Stanford. All patients entered in the study (71 boys and 133 girls; mean age = 9.4 years) were diagnosed by a pediatric rheumatologist and met American Rheumatology Association (ARA) diagnostic criteria (Brewer et al., 1977) that include both physical examination and laboratory findings. Most of the patients (71%) had a form of juvenile arthritis (i.e., systemic juvenile, polyarticular, or pauciarticular arthritis); the rest (29%) were diagnosed with other forms of rheumatic disease, such as ankylosing spondylitis, systemic lupus erythematosus, and dermatomyositis. Of 224 consecutive families in which the patient met the diagnostic criteria, 204 (91%) agreed to participate. Project staff met at the clinic with each participating family to explain the study and obtain informed consent. Questionnaires were distributed to families who signed informed consent forms, which were returned by mail. Due to time-limited funding, we attempted to follow only the first 165 of the 204 families 1 year later. At Time 2, questionnaires were distributed during a clinic visit, if the timing of the visit coincided with the 1-year follow-up; otherwise, to save participants an unnecessary trip to the clinic, the family was contacted by telephone, and questionnaires were distributed by mail. At Time 1, 197 mothers (99% of 199 mothers who were living with patients) and 164 fathers (95% of 173 fathers who were living with patients) provided data; at Time 2, 141 mothers (99% of 143 mothers who were located and still living with patients) and 115 fathers (95% of 121 fathers who were located and still living with patients) returned questionnaires. All analyses reported in this paper are based on members of married couples in which both partners returned data: 159 mothers and fathers at Time 1, and 111 mothers and fathers at Time 2. To ensure that the parents who participated at Time 2 were representative of the initial sample, we conducted / tests on the initial functioning indices between mothers who provided data at both Time 1 and Time 2 and mothers who provided only Time 1 data; corresponding analyses were conducted for fathers. No differences were found between either set of parents. T tests were also conducted between mothers of patients with juvenile arthritis and mothers of patients with other rheumatic diseases on the Time 1 and
Functioning Among Parents of Children with JRD
711
Time 2 psychosocial outcomes; the same t tests were conducted for fathers. At Time 2, fathers of patients who had juvenile arthritis engaged in fewer social activities (A/s = 3.7 vs. 5.1) f(109) = 2.0; p < .05; otherwise, there were no significant differences between groups. On the basis of these and previous results finding minimal differences between juvenile arthritis patients and patients with other rheumatic diseases, and their families (Timko et al., 1992a, 1992b), parents of all patients were analyzed together. Measures Parent Functioning Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
The outcomes examined in this study reflected the parent's assessment of his or her own functioning in four areas. Two of the outcome measures were taken from the Health and Daily Living Form (HDL; Moos, Cronkite, & Finney, 1990), which mothers and fathers completed at both Time 1 and Time 2. The HDL indices have good validity and reliability, and have been used with over 1,500 families and in more than 100 studies (Moos, Cronkite, & Finney, 1990). The two HDL outcome measures were Depressed Mood (Cronbach's alpha = .93 at Time 1 and at Time 2) is the sum of 18 items measured on a 5-point scale ranging from never (0) to often (4) with reference to the past month; the items tap the presence and severity of symptoms involved in obtaining an RDC (see Spitzer, Endicott, & Robins, 1978) diagnosis of major or minor depression (e.g., feeling sad or blue, poor appetite or weight loss, sleeping problems). Social Activities (alpha = .79 at Time 1, .77 at Time 2) is the number of activities (out of 12) that the parent noted he or she had done or attended with friends in the past month (e.g., athletic events, card games, parties). The two additional outcome measures were taken from the Family Effects of Illness scale (Stein & Reissman, 1980), which was also completed by both mothers and fathers at Time 1 and Time 2: Personal Strain (alpha = .93 at Time 1, .85 at Time 2) is the sum of seven items answered on a 4-point scale ranging from strongly disagree (0) to strongly agree (3) that assess the parent's strain resulting from demands of the child's illness (e.g., fatigue is a problem for me because of my child's illness; I worry about what will happen to my child in the future when he/she grows up or when I'm not around). Mastery (alpha = .82 at Time 1, .69 at Time 2) is the sum of five items answered on the same 4-point scale that assess the parent's view of his or her own and the family's ability to handle the stressors associated with the child's illness (e.g., learning to manage my child's illness has made me feel better about myself; because of what we have shared we are a closer family).
712
Timko, Stovel, and Moos
Determinants of Parent Functioning
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Demographic Factors. Demographic factors were assessed on each parent's HDL at Time 1. They included whether or not the patient was an only child; the family's total annual income; the parent's education and age; and the patient's age and gender. Patient Dysfunction. The patient's disease Duration was measured at Time 1 by the pediatric rheumatologist as number of years since disease onset. The patient's disease-related dysfunction was assessed at Time 1 on the Child Health Assessment Questionnaire (Child HAQ; Billings, Moos, Miller, & Gottlieb, 1987). The Child HAQ is adapted from the Adult HAQ (Brown et al., 1984; Fries et al., 1982), which assesses arthritis disability and pain among adults. Patients who were old enough to self-report at Time 1 (i.e., were at least 10 years old) completed the child version of the Child HAQ. If the patient completed a Child HAQ, his or her Functional Disability and Pain scores were used; otherwise, the mother's scores on the parent version of the Child HAQ were used. The Functional Disability index is based on items assessing difficulty in the past week, rated on a 4-point scale ranging from no difficulty (0) to unable to do (3), in eight categories (e.g., dressing and grooming, eating, walking). The scoring procedure is similar to that for the Adult HAQ (for details, see Fries et al., 1982), and results in a continuous scale from 0 to 3, with 3 representing the most severe disability. The Pain scale is a visual analog scale that measures pain due to rheumatic disease during the past week. Responses are indexed on a scale ranging from no pain (0) to pain as bad as it could be (3). Mothers and fathers completed an eight-item index on the HDL at Time 1 to describe the patient's psychosocial dysfunction. The index assesses the number of Psychosocial Problems (alpha = .73) during the past year (e.g., mental or emotional problems; problems getting along with other children). Spouse Dysfunction. Spouse dysfunction at Time 1 consisted of the spouse's responses to the same four indices used to assess parent functioning, that is, Depressed Mood, Social Activities, Personal Strain, and Mastery. Illness-Related Coping. Mothers' and fathers' methods of coping at Time 1 were measured on the HDL. Parents were asked to indicate what they did in connection with their child having rheumatic disease. Items measured the frequency of use on a 4-point scale of 32 coping responses. Approach Coping (alpha = .88) consists of items assessing active cognitive (e.g., prepared for the worst, tried to see the positive side of the situation) and active behavioral (e.g., tried to find out more about the situation, made a plan of action and followed it) coping. Avoidance Coping (alpha = .66) consists of items such as: refused to believe that it happened; and, tried to reduce tension by drinking more. Family Resources. The Family Resources index at Time 1 was assessed by both mothers and fathers (alpha = .69). It was a sum of three subscales: Cohesion (the degree of commitment, help, and support family members provide for
Functioning Among Parents of Children with JRD
713
one another); Expressiveness (the extent to which family members are encouraged to act openly and express their feelings directly) from the Family Environment Scale (Moos & Moos, 1986); and Family Arguments (reverse scored) from the HDL. The Cohesion and Expressiveness subscales are each the sum of nine true-false items. Family Arguments is the count of up to 14 items (e.g., relatives, helping with household tasks, money) that the parent marked as often causing disagreements in the family.
RESULTS Comparison of Mothers and Fathers
Concurrent Analyses of Parent Functioning First, we conducted multiple regression analyses in which the demographic factors were used to predict each aspect of functioning assessed at Time 1, for Table I. / Tests Comparing Mothers' and Fathers' Functioning" Mothers Functioning indices Time 1* Depressed mood Personal strain Social activities Mastery Time 2C Depressed mood Personal strain Social activities Mastery
Fathers
M
SD
M
SD
20.0 7.4 4.3 10.5
11.6 4.0 2.6 2.1
14.8 6.6 4.1 10.0
10.9 3.7 2.6 2.1
4.1&/" 1.84 0.70 1.86
19.5 6.1 4.4 10.3
12.2 4.0 2.3 2.5
14.8 5.8 3.9 9.6
10.6 4.1 2.7 3.0
3.04' 0.48 1.38 2.JC
"When the Bonferroni method is applied to consider that four comparisons were made at Time 1 and Time 2, alpha = .012. b n = 159 for each parent group. c n = 111 for each parent group. a p < .05. 'p < .01. fp < .001.
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Table I reports the results of / tests that compared mothers with fathers on functioning at Time 1 and Time 2. At Time 1 and Time 2, mothers experienced more depressed mood than fathers did. In addition, at Time 2, mothers experienced more mastery. No differences were found between mothers and fathers on personal strain or on social activities.
714
Timko, Stovel, and Moos
mothers and fathers separately. The patient's age and gender, the parent's age and education, and the family's annual income and size all failed to show significant associations with comparable aspects of both mothers' and fathers' functioning. Three scattered associations were significant. For mothers, having an older child (b = .27) and more education (b — .20) were related to more social activity; for fathers, having more than one child was related to more mastery (b = . 18, ps < .05). On the basis of these results, demographic factors were dropped from subsequent analyses. Mothers
Downloaded from http://jpepsy.oxfordjournals.org/ at University of Victoria on May 25, 2014
Table II presents the results of hierarchical multiple regression analyses using Time 1 determinants to predict each aspect of mothers' functioning at Time 1. In these regressions, indices of patient dysfunction were entered on Step 1, spouse dysfunction was entered on Step 2, mothers' illness-related coping responses were entered on Step 3, and the family resources index was entered on Step 4. When zero-order correlations were computed among mothers' Time 1 dependent variables, a significant positive association was found between depressed mood and personal strain (r = .41, p < .001); therefore, for each step in the regression analyses predicting depression and personal strain, the R2 change was considered significant only at p < .025. The results for patient dysfunction showed that patients' psychosocial problems were associated with mothers reporting more depression and personal strain and less mastery; patients' functional disability was also associated with mothers' personal strain. When patients had JRD for a longer duration and experienced less pain, mothers engaged in more social activities. The father's functioning was significantly positively related to each corresponding aspect of mother's functioning. Regarding mothers' illness-related coping, more use of approach coping was related to more social activity, while more reliance on avoidance coping was related to more depression and personal strain. When patient and spouse functioning factors were controlled, mothers' illness-related coping explained significant additional variance in depression, personal strain, and social activity. Finally, mothers in families with more resources experienced less depression and strain and more mastery. Taken together, patient, spouse, coping, and family factors explained 38% of the variance in depression, 50% in personal strain, 20% in social activity, and 29% in mastery. Fathers Table II presents parallel hierarchical multiple regression analyses for the fathers. As for mothers, intercorrelations among fathers' Time 1 functioning
"p < .001.
"p < .05. *p < .025. c p < .01.
Overall R Overall R2
Patients' dysfunction Duration Functional disability Pain Psychosocial problems Mothers' dysfunction Corresponding functioning index Fathers' coping Approach coping Avoidance coping Family resources
Overall R Overall R2
Patients' dysfunction Duration Functional disability Pain Psychosocial problems Fathers' dysfunction Corresponding functioning index Mothers' coping Approach coping Avoidance coping Family resources
-.18* .66" .44
.07 .47"
.29"
.17° .16 -.03 .14
.38
-.22' .62"
-.03 .30"
.28"
,01 -.03 .45"
.00
b
b
R2 change
.io
