Risk factors for psychological stress in children of parents with a chronic medical condition
U. Ambachtsheer S1350293
Master Thesis Clinical Psychology Supervisor: Dr. D. Sieh
Clinical Psychology Universiteit Leiden 31-08-2017§
2 Abstract
Children of parents with a chronic medical condition are known to be at risk for psychological stress. We assessed possible risk factors and categorized them with factor analysis using data from a sample of 139 children aged 10–20 years and their corresponding 86 parents. Parents filled in questionnaires about their functioning (physical, mental, and social), and children filled in questionnaires about the consequences of the parental chronic medical condition on their daily life (daily hassles, limitation of social activity, social isolation, caregiving), and the amount of stress they experienced. Both parents and child rated the unpredictability of the illness. Results showed that girls experience more stress than boys. Additionally, a higher age was positively correlated with informal caregiving, daily hassles concerning the ill parent, daily hassles concerning personal life, social isolation, and child stress. Factors that determine stress levels were daily hassles, limitation of social activity, and social isolation. These factors also negatively correlated with parental mental and social functioning. Caregiving was not related to stress but a lower parental physical functioning was related to more caregiving.
3 Internationally, 10% of the children grow up with a parent who has a chronic medical condition (Worsham, Compas, & Ey, 1997, as cited in Sieh, Meijer, Oort, Visser-Meily, & Van der Leij, 2010). This amount is rising due to an increase in autoimmune system diseases among parents, an improvement in techniques to prolong life expectancy of parents, and parents giving birth at a later stage of life (Bach, 2008, as cited in Sieh, 2012; Lubkin & Larsen, 2006; Shifren & Kachorek, 2003, as cited in Sieh, 2012; Sieh, 2012). A chronic medical condition (CMC) is defined as “a syndrome involving one or more organ systems that impairs health and psychological functioning for at least 3 months”. The four most common types of CMC are cardiovascular diseases, cancer, lung diseases (chronic respiratory diseases), and diabetes (World Health Organization, 2016).
Children of parents with a chronic medical condition (target group) are at risk for elevated psychological stress due to their family situation being negatively altered by the illness of the parents (Huizinga et al., 2005; O'Connor, Dunn, Jenkins, Pickering, & Rasbash, 2001). An example of an adverse alteration is when the target group has less time to play with friends because they need to spend more time in taking care of their parents. Elevation of psychological stress has adverse health outcomes for both children and adolescents. Children with high psychological stress may have a hyperactive immune system in the future, leading to chronic inflammation which in turn results in poor functioning of the immune system and increased blood pressure (Miller et al., 2011, as cited in Farrell, Simpson, Carlson, Englund, & Sung, 2017). For adolescents, a high amount of psychological stress predicts future headaches (Waldie, 2001, as cited in Farrell et al., 2017), metabolic syndrome (Gustafsson et al., 2012, as cited in Farrell et al., 2017), obesity, and proinflammatory tendencies (Ehrlich, Miller, Rohleder, & Adam, 2016, as cited in Farrell et al., 2017). Farrell and colleagues (2017) also found that stress cumulates, as children who experienced stress in their childhood and adolescence had a higher chance of developing health problems than children who experienced stress in only childhood or adolescence.
The transactional Stress and Coping (TSC) Model of Hocking and Lochman (2005) is widely used to explain how the target group adjusts to its negatively altered situation caused by parental CMC. It gives a broad view on how different factors interrelate with the parental CMC; the present study will focus solely on factors that are suggested to be related to child stress. The TSC model suggests that parental CMC causes child stress in two ways. One way is that parental CMC causes stress by impacting family functioning. Another way parental illness causes stress is by increasing the amount of daily hassles, informal caregiving, and
4 caregiving responsibility. The first way suggests that parental CMC can differ in type and severity, which influences parental health: The physical, mental, and social well-being of the parent. Parental physical health is defined by how well the parent is physically functioning, such as being able to climb stairs or walk to the shopping mall. Parental mental health in this study is determined by parental depression, and parental helplessness. Parental social health is defined by how well a parent is functioning in the social domain of life such as spending time with friends or having a healthy, and stable relationship with a partner. Functioning poorly in these three areas may result in negatively altering family functioning (e.g. child alienation from parent, quality of marital relationship, and quality of child rearing). Negative change in child rearing is particularly suggested to cause child stress (Armistead et al., 1995; Sieh, Meijer, & Visser-Meily, 2010). Another factor highly related to parental physical health is illness unpredictability, which involves uncertainty about the course of the parental CMC and therefore, the future. Illness unpredictability also causes stress by influencing family functioning, as it interferes with making plans for the future, hindering the family in engaging in an appropriate coping style (Rolland, 1999). Illness unpredictability is also directly related to stress, as it causes anxiety and ambiguity. Moreover, unpredictable events during the illness such as a heart attack also cause a high amount of stress for children due to being fast, and intensive changes, which hinders in utilizing an efficient coping style (Sieh, Meijer, & Visser-Meily, 2010). In short, the TSC model suggests that parental health and illness unpredictability cause stress through family functioning, Figure 1 gives an overview of this mediation.
5 Figure 1. Suggested relationship between parental health, child stress, and family functioning. Thickened lines = included in the present study and discussed relations; faded lines = included but not (yet) discussed; dotted lines = not included in the present study. DIP = daily hassles concerning ill parent; DHP = dailly hassles concerning healthy parent; DPL = daily hassles concerning personal life.
The second way parental CMC causes stress is by increasing the amount of psychosocial stressors a child has to face. Psychosocial stressors are defined by the negative appraisal of social events such as getting low grades at school, losing a football match, or having a fight with a friend. The TSC model suggests two types of psychosocial stressors, namely daily hassles, and caregiving. Daily hassles are stresses and strains of daily life (Pedersen & Revenson, 2005) and the present study takes into account daily hassles concerning the ill parent, healthy parent, and personal life. Examples of daily hassles are seeing the ill or healthy parent being sad, or witnessing and achieving poor grades at school. The present study takes into account two types of caregiving namely informal caregiving, and caregiving responsibility. Informal caregiving consists of taking care of the parent (e.g. washing the parent, and feeding the parent), and doing household chores. Whereas informal caregiving is defined by the amount of care one has to provide, caregiving responsibility is characterized by a subjective feeling of being expected of taking care of the family, and providing emotional support to family members. Studies suggest that caregiving is stressful because it interferes with fulfilling needs for social contact, play, or performing at school (Perkins et al., 2013). Even though the TSC model only considers daily hassles, and
6 caregiving as stressors, other research suggests that the target group also faces an elevation in limitation of social activity, and social isolation (Sieh, Visser-Meily, & Meijer, 2012). Social isolation is characterized by feeling lonely, being unable to express worries to someone, and lacking meaningful bonds. People who are socially isolated fail to receive social support (Hawthorne, 2008). As social support acts as a buffer for stress and reduces stress responses, socially isolated people are at risk for stress (Hawthorne, 2008). Additionally, the loneliness hypothesis claims that people have an internally regulated need to belong to a group (Hawthorne, 2008). This need becomes even stronger during adolescence, where the adolescent brain makes changes to become more socially active and increase social competence (Crone, & Dahl, 2012). A plethora of research confirms the loneliness hypothesis and has successfully linked social isolation with depression, psychological stress, physical stress, and even morbidity (Cruces, Venero, Pereda-Perez, & De La Fuente, 2014; Grant, Hamer, & Steptoe, 2009; Hawthorne, 2008). Figure 2 shows the second way the negatively altered situation may cause child stress.
Figure 2. Suggested relation between child stress and psychosocial stressors. Thickened lines = included in the present study and discussed relations; faded lines = included but not (yet) discussed; dotted lines = not included in the present study. DIP = daily hassles concerning ill parent; DHP = daily hassels concerning healthy parent; DPL = daily hassles concerning personal life.
7 Finally, the TSC model suggests that child age, and gender are related to child stress. In regards to child age, it has been reported that the target group experiences more psychological stress than adults with chronically ill parents (Schrag, Mrely, Quinn, And Jahanshahi, 2004). However, a recent study by Sieh, Meijer, & Visser-Meily (2010) found no relation between age, and stress within the target group. While it is unclear whether child age is related to child stress, studies consistently confirm a relation between gender, and child stress. In general, girls are more sensitive to psychological stress than boys (Kreutzer, Gervasio, & Camplair, 1994; Linn, Allen, & Willer, 1994. As cited in Verhaeghe, Defloor, & Grypdonck, 2004; Sieh, Meijer, & Visser-Meily, 2010; Sieh, Dikkers, Visser-Meily, Meijer, 2012). The most prominent explanation is that girls might use a less effective coping style than boys which causes an increase in stress (Matud, 2004). Another explanation is that girls are more prone to experiencing stress from people around them due to girls being more emotionally involved in family networks, as well as social networks (Matud, 2004).
The TSC model highlights the many ways in which child stress and its potential risk factors (parental health, illness unpredictability, and psychosocial stressors) can be related to one another. However, one weakness of the TSC model is that it lacks specific pathways; as a result it is difficult to interpret which factors determine child stress. The present study aims to solve this problem by categorizing the potential risk factors of child stress through factor analysis and subsequently relating them to child stress through structural equation modeling. The present study hypothesizes that parental illness unpredictability, along with their physical, social, and mental health (depression, and helplessness), social health, and illness unpredictability will amalgamate, forming a factor named parental health stressors (H1), and that daily hassles (DIP, DHP, DPL), caregiving (informal caregiving, and caregiving responsibility), limitation of social activity, and social isolation, will amalgamate to form a factor named social stressors (H2). It is further hypothesized that parental health stressors, and social stressors are related to child stress (H3). The hypothesized model is shown in figure 3.
8 Figure 3
Hypothesized model of relating child stress with parental health stressors, and social stressors
Green colored = hypothesis 1; red colored = hypothesis 2; blue colored = hypothesis 3. DIP = daily hassles concerning ill parents; DHP = daily hassles concerning healthy parents; DPL = daily hassles concerning personal life.
9 Method
Sample and participant selection
An existing dataset by Sieh (2012) was used for the data analysis. The data set includes 161 Dutch children between the ages of 10 and 20 years old with a mean age of 15 where 52% of the participants are female, and each has at least one parent with a CMC. Spouses (with or without CMC) were also included in the analysis. The types of CMC's that the parents contracted included multiple sclerosis, rheumatoid arthritis, brain damage, neuromuscular disease, spinal cord injury, inflammatory bowel disease, Parkinson disease, and type 1 diabetes (Sieh, 2012). Exclusion criteria for the children included lack of fluency in the Dutch language, having a serious chronic illness, or cognitive impairments and behavioral problems prior to when the parent got a CMC.
Assessment and Measures
Parents. Parental physical, and parental social health were measured through the Medical Outcome Study Short-Form 36 (MOS-SF36) from Stewart (1988) as cited in Sieh (2012) by using the subscales physical functioning, and social functioning. The score for each subscale ranges from 0 to 100, with a higher score indicating a better functioning. There are no cut-off scores for the MOS-SF36. Physical functioning consists of 10 items (α = .92), which measure limitations in carrying out physical activities due to physical health problems. Score per item ranges from 0 to 100 with three answer possibilities: 0 (Yes, I am completely limited), 50 (Yes, I am partly limited) and 100 (No, I am not limited). The total score is calculated by averaging the score of the ten items. Social functioning consists of 2 items (α = .80) and measures how well the parent is social functioning (i.e. engaging in social activities and relationships). Both items have five answer categories. The first item is "How often has your physical health or emotional problems limited you in engaging in social activities (e.g. visiting friends or family) last week" and has five answer categories which are scored 0, 25, 50, 75 and 100, namely 0 (always), 25 (mostly), 50 (sometimes), 75 (rarely), and 100 (never). The second item "How much has your physical health or emotional problems restricted you in engaging in normal social intercourse with family, friends, neighbors or group activities last week?" is answered with 0 (very much), 25 (much), 50 (quite), 75 (somewhat), and 100 (not at all). The total score is calculated by averaging the score of the two items.
10 Parental mental health was assessed by measuring depression, and helplessness. Depression was measured by the Dutch version of the Beck Depressive Inventory (BDI), a 21 item (α = .85) measurement instrument that measures cognitive, behavioral and somatic signs of depression (Bouman, Luteijn, Albersnagel, & Ploeg, 1985). The BDI consists of questions about how the person felt about themselves in the past week. Every item can be scored from 0 to 3 and the total score of the BDI ranges from 0 to 63, where a higher score indicates more signs of depression. Cut-off scores are 0-9 (minimal depression), 10-18 (mild depression), 19-29 (moderate depression) and 30-63 (severe depression). Helplessness was measured by using the helplessness subscale of the Illness Cognition questionnaire (Bakker, 2005, as cited in Sieh, 2012; Evers, Kraaimaat, Lankhorst, Jacobs, & Bijlsma, 1998). Helplessness consists of 6 items (α = .81) with scores ranging from 6 to 24. A higher score indicates an increased experience of helplessness. Items are statements that can be answered 4-point scale stretching from 1 (not at all), 2 (a little), 3 (strongly), and 4 (entirely). Two example items of the helplessness subscale are "your illness causes you to feel useless", and "your illness controls your life".
Family. All family members in the household (mother and children) reported the illness unpredictability of the chronically ill parent through the Illness Unpredictability Questionnaire (Pakenham, Bursnall, Chiu, Cannon, & Okochi, 2006). The questionnaire contains 5 items on a 5-point scale stretching from 0 (strongly disagree) to 4 (strongly agree). Score ranges from 0 to 25 with a higher score indicating a higher illness unpredictability. In order to calculate the family score, the score of all respondents in the family were added up, and afterwards divided by the amount respondents who filled in the questionnaire. For example in a family consisting of a mother with two children, the score of the mother and children were summed up and then divided by three. It has been reported that the Cronbach's alpha was .89 in a family consisting of three family members (Sieh et al., 2012).
Children. Daily hassles were measured by using the subscales daily hassles concerning the ill parent (DIP), daily hassles concerning the healthy parent (DHP), and daily hassles concerning personal life (DPL), from the Dutch version of the Daily hassle Questionnaire (Dufour, Meijer, Van de Port, & Visser-Meily, 2006). The questionnaire contains different statements about daily hassles and asks how often the child has to deal with them with four answer possibilities: Never (0), minimum of one time per month (1), per week (2), or per day (3) DIP consists of 6 items (α = .61) with questions such as "do you sometimes see your ill parent being sad". DHP consists of 5 items (α = .71) with questions such as "do you
11 sometimes see your healthy parent being exhausted or overstressed". The final scale, DPL, consists of 9 tems (α = .80) and contains questions such as "do you find that your homework / school are negatively affected due to your situation at home" and " Do you find that you do not receive enough recognition from your ill parent in regards to the things you do?". Caregiving was measured through the caregiving responsibility subscale of the Young Caregiver of Parents Inventory by Pakenham and colleagues (2006), and the extent of informal care subscale of the Informal Care Questionnaire by Meijer, Van Oostveen, and Stams (2008). Caregiving responsibility consists of 8 items (α = .76) with a score ranging from 9 to 45. It measures the extent to which the child believes that the family relies on them in; housekeeping tasks, taking care of the ill parent, siblings, and providing emotional support. Caregiving responsibility contains questions such as “my parents rely on me to ensure things go well in our family” or “my parents rely on me regarding to emotional support”. The extent of informal care contains 16 items (α = .74) measuring how often the child has to provide housekeeping tasks or informal care. The score ranges between 0 to 64. It contains questions such as “how often do you need to do the laundry?” or “how often do you need to help your parent going to the toilet?”. The questions are rated on 4-point scale ranging from 0 (never) to 4 (every day).
Limitations of social activities and feelings of isolation were measured by using the subscales limitation of social activity, and feelings of isolation of the Young Caregiver of Parents Inventory (Pakenham et al. 2006). The items in the subscales are rated on a 5-point scale varying from 1 (completely false) to 5 (completely true). A higher score on a subscale indicates a higher amount of limitation of social activity or feelings of isolation. Limitation of social activity consists of 8 items (α = .73) with a score ranging from 8 to 40. Two examples of items are “Helping my parents prevent me from doing things” and “Because I need to help my parent(s) I sometimes feel too exhausted”. Feelings of isolation consists of 3 items (α = .76) with a score ranging from 3 to 15. An example of an item is “I would like to be able to talk about my feelings with someone”.
Child stress was measured by using the Dutch version of the Stress Questionnaire for children (SQ-C) from Hartong, Krol, Maaskant, Te Plate, and Schuszler (2003). The questionnaire consists of 17 items (α = .88) on a 4 point-scale from 1 (completely false) to 4 (completely true). The score ranges from 17 to 68, where a higher sum score is related to a higher stress level. Two examples of items are “I find that I think a lot about things” and “I often feel rushed”.
12 Procedure
Gathering of participants took place between 2008 and 2011. All families were recruited through health organizations, rehabilitation, community centers, hospitals, schools, public places and general health practitioners through advertisement with posters and brochures. Families could use e-mail, phone or post to contact the project manager. After participants were screened over the phone, they received an information package and both the parent and child received an informed consent form. Afterwards, research assistants visited the families to conduct the questionnaires. As a reward for filling in the questionnaires, the adolescents received a gift voucher, a cinema ticket, or a mobile phone cover. The ethical commission of the research institute of Child Development and Education of the University of Amsterdam approved the study in which Sieh (2012) gathered the dataset.
Data analysis
All analyses were computed by using IBM SPSS Statistics (Version 23; IBM Corp, 2015), except for SEM where we made use of Stata (Version 13; StataCorp, 2013). Means and standard demographic data of the participants (e.g. age, gender and education level) were analyzed by using descriptive statistics. A Pearson correlation was then used to check for associations between age, psychosocial stressors, and child stress. Gender differences in psychosocial stressors, and child stress were also analyzed by using an independent sample-t-test. Normally, when gender differences are present, or when age is related to variables, these effects are taken into account in the SEM analysis. However, adding age and gender to the SEM analysis was too complex at this stage of research and were therefore excluded from the analysis.
Before building the model, it was important to inspect the relations of the variables, which served as a guide for building the model. A Pearson correlation matrix was therefore designed in order to inspect the relationship between parental health, psychosocial stressors, and child stress, as significant correlations suggest possible paths or covariations between variables. Categorization of the psychosocial stressors and parental health were checked for by using exploratory factor analysis (EFA) with the principal component method (PCA), and varimax rotation. By using EFA, the components (variables) were related to one another and sorted into factors. Significant factors had an Eigenvalue (λ) greater than 1.0 (Kline, 2016).
13 Afterwards, SEM was used to construct a model based on the data provided by the Pearson correlation and EFA. The model was constructed in two steps. Firstly, a confirmary factor analysis was computed based on the results of the EFA. Factors with significant eigenvalues, and their corresponding components with the highest loadings formed the initial CFA model. Additionally, we checked for common-method bias by using Harman’s single factor test, and the common latent factor method (Podsakoff, Mackenzie, & Lee, 2003). Harman's single factor test suggested common-method bias but the common latent factor method showed that this effect was negligible, as the amount of common-method variance was only .05%. Afterwards, the CFA model was built and a good goodness of fit (GOF) was achieved by removing observed variables with weak relations with latent variables, and by following modification indices. GOF was checked for by using four different types of goodness-of-fit statistics, as Kline (2016) states that GOF is most accurately determined by using Model chi-square fit (χ2), Steiger-Lind root mean square error of approximation (RMSEA), Bentler comparative fit index (CFI), and standardized root mean square residual (SRMR). Good GOF is indicated by an insignificant χ2 (p = >.05), significant RMSEA (p = <.05), a CFI greater than .95, and SRMR smaller than .05. After having reached a good GOF, child stress was added to the CFA model. Paths from the latent variables to child stress were drawn based on the Pearson correlation matrix. Again, modification indices were followed to improve GOF. No variables were removed from the model at this final step.
14 Results
Demographic characteristics
The analysis consisted of 139 children (51.8% female). Their current mean age was 15.20 (SD = 2.36) with a minimum age of 10.72 and maximum age of 20.88. Roughly 43% of the children followed HAVO or a higher form of education (VWO, Higher education, or University). The remaining 57% either followed primary education, lower vocational education or intermediate vocational education.
The corresponding 87 parents (62.1% female) were also included in the analysis. Their current mean age was 47.38 (SD = 5.61) with a minimum age of 32.07 and maximum age of 63.22. In total, 58% of the parents followed high school or a higher education level (pre-university and higher vocational education or university). Parental CMC's included were MS (28.7%), brain damage due to stroke or accidents (23.0%), joint diseases such as reuma and Addison’s disease, (17.2%), myopathy (13.8%), rheumatism (11.5%), Parkinson’s disease (4.6%), and other illnesses (9.2%) including chronic sleeplessness, Crohn’s disease, Addison’s disease, chronic pain syndrom, Colitis Ulceroca, and hemiparesis.
Gender differences and child age.
An independent-samples t-test was computed to check gender differences in the psychosocial stressors, and child stress. There was only a significant difference in scores for child stress in boys (M = 34.54, SD = 7.32) and girls (M = 37.49, SD = 7.85); t (137) = -2.29, p = .02 Next a Pearson correlation was computed to check a possible relation between age, psychosocial stressors, and child stress. A higher age was positively corrrelated to informal care (r = .19, p = .03), DIP (r = .19, p = .03), DPL (r = .20, p = .02), social isolation (r = .35, p = <.01), and child stress (r = .23, p = < .01). In contrast, no significant correlation was found between illness unpredictability, and age. (r = .10, p = .25).
15 Correlations between parental health, illness unpredictability, psychosocial stressors, and child stress
The correlation matrix is shown in Figure 1. Parental physical health correlated weakly with parental helplessness, and social health, and did not correlate with illness unpredictability. Moderate to strong correlations were found between, parental depression, helplessness, social health, and illness unpredictability. Parental depression, and social health weakly correlated with child stress.
DIP, DHP, DPL, limitation of social activity, and social isolation, moderately to strongly correlated with each other. They also moderately to strongly correlated with child stress. Informal care, and caregiving responsibility moderately correlated with each other, but weakly to moderately correlated with the other psychosocial stressors. Informal care, and caregiving responsibility correlated weakly with child stress.
Computing correlations between parental health, and psychosocial stressors showed that parental physical health moderately correlated with informal care, and caregiving responsibility. Parental physical health did not correlate with the other psychosocial stressors, with an exception of a weak correlation with DIP. The other parental health factors (depression, helplessness, social health, and illness unpredictability) all weakly to moderately correlated with DIP, DHP, DPL, limitation of social activity, and social isolation.
16 Table 1
Intercorrelations among parental health, illness unpredictability, psychosocial stressors, and child stress.
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Parental physical health _
2. Parental depression -.16 _
3. Parental helplessness -.36** .65** _
4. Parental social health -.28** -.64** -.56** _
5. Illness unpredictability .04 .32** .27** -.21* _ 6. DIP -.21* .34** .33** -.28** .38** _ 7. DHP -.14 .26** .37** -.30** .08 .49** _ 8. DPL -.05 .28** .16 -.31** .16 .56** .50** _ 9. Informal caregiving -.41** .12 .17 -.22** .14 .49** .35** .36** _ 10. Caregiving responsibility -.26** .18* .27** -.22* .12 .37** .45** .29** .38** _ 11. Limitation of social activity -.07 .29** .23** -.19* .19* .46** .46** .65** .34** .38** _ 12. Social isolation -.04 .23** .15 -.20* .22** .47** .37** .68** .24** .26** .55** _ 13. Child stress -.01 .22* .14 -.25** .12 .45** .39** .73** .17* .18* .53** .69** _
Note. DIP = Daily hassles concerning ill parent; DHP = Daily hassles concerning healthy parent; DPL = Daily hassels concerning personal life. * p < .05, ** p < .01.
17 Exploratory Factor Analysis
The Kaiser-Meyer Olkin measure of sampling adequacy suggested that it was possible to compute a factor analysis (χ2 (66) = .80, p = <.01). Table 2 shows the results from the EFA Table 2
Varimax rotated component loadings of the variables
Component Social stressors Parental health stressors Caregiving
Parental physical health -.35 .36 .67
Parental depression .61 -.55 .34
Parental helplessness .60 -.62 .04
Parental social health -.59 .55 -.12
Illness unpredictability .39 -.15 .45 DIP .76 .16 -.00 DHP .69 .15 -.14 DPL .73 .43 .18 Informal care .57 .14 -.52 Caregiving responsibility .57 .09 -.38
Limitation of social activity .70 .38 .11
Social isolation .64 .43 .26
Eigenvalues 4.49 1.73 1.33
Percentage of total variance 37.36 14.42 11.05
Number of test measures 5 3 4
Note. Bolded values are components that are proposed to be suitable to form a factor. DIP = Daily hassles concerning ill parent, DHP = daily hassles concerning healthy parent, DHPL = daily hassles concerning personal life.
18 Structural equation modeling
A CFA based on the results of the EFA shown in Table 2 resulted into a poor GOF : χ2 = 140.40 (df =51, p = <.01), RMSEA = .112 (90% CI = [.09, .14]), CFI = .854, and SRMR = .083. After following modification indices, GOF still remained poor: χ2 = 96.23 (df =45, p = <.01), RMSEA = .090 (90% CI = [.07, .12]), CFI = .917, and SRMR = .063. Since the components illness unpredictability, and parental physical health had the lowest path coefficients to the factor caregiving (ϐ = .26 for illness unpredictability, and ϐ = -.46 for parental physical health), both were consecutively removed from the analysis to improve GOF. Afterwards, modification indices suggested to draw paths from the components DIP, DHP, DPL, and social isolation to the factor caregiving, which lead to a model with a good GOF. To prevent confusion, we decided to change the name from the latent variable
caregiving, to caregiving + social stressors. The CFA model is shown in Figure 4.
Next, child stress was added to the model and was related to social stressors, which caused GOF to decrease to: χ2 = 63.69 (df =37, p = <.01), RMSEA = .072 (90% CI = [.04, .1]), CFI = .960, and SRMR = .047. Modification indices suggested to draw a path from limitation of social activity to caregiving + social stressors. This resulted into the final model with a good GOF: χ2 = 48.25 (df =36, p = .08), RMSEA = .049 (90% CI = [.00, .83]), CFI = .982, and SRMR = .036. The final model is shown in Figure 5.
19 Figure 4.
CFA model of parental health stressors, caregiving + social stressors, and social stressors
Note. Standardized estimates. GOF: χ2 = 42.82 (df =27, p = <.05), RMSEA = .062 (90% CI = [.02, .1]), CFI = .973, and SRMR = .036. DIP = daily hassles concerning ill parent; DHP = Daily hassles concerning healthy parent; DPL = daily hassles concerning personal life. * p < .05, ** p < .01.
20 Figure 5.
Final model of parental health stressors, social stressors, caregiving + social stressors, and child stress
Note. Standardized estimates. GOF: χ2 = 48.25 (df =36, p = .08), RMSEA = .049 (90% CI = [.00, .83]), CFI = .982, and SRMR = .036. DIP = daily hassles concerning ill parent; DHP = Daily hassles concerning healthy parent; DPL = daily hassles concerning personal life. * p < .05, ** p < .01
21
discussion
The present study aimed to identify risk factors for stress in the target group. Potential risk factors (parental health, illness unpredictability, and psychosocial stressors) of child stress were categorized and their relationships with child stress were inspected. Results showed that daily hassles (DIP, DHP, and DPL), limitation of social activity, and social isolation loaded on the factor "social stressors", and were directly related to child stress. The relationship between daily hassles, and child stress can be explained through the concept of stress. Stressful events are characterized as uncontrollable, and unpredictable (Koolhaas et al., 2011). The measured daily hassles contained these characteristics. For example, the child has little control over seeing the ill parent in pain, or being sad. Another way daily hassles are related to child stress is that they cause worry (Thielsch, Andor, & Ehring, 2015). Worry is related to physical stress, as well as adverse mental, and somatic health (Verkuil, Brosschot, Borkovec, & Thayer, 2009). Furthermore, the relationship between limitation of social activity, social isolation, and child stress, is in line with the loneliness hypothesis which claims that every human being has an innate need for social contact (Hawthorne, 2008). The results are also in line with earlier studies where social isolation was shown to cause stress, as well as physical and psychological problems (Cruces, Venero, Pereda-Perez, & De La Fuente, 2014; Grant, Hamer, & Steptoe, 2009; Hawthorne, 2008).
Parental depression, helplessness, and social health loaded on the factor "parental health stressors" but was not directly related to child stress. Instead, more parental health stressors were related to more social stressors. This is in line with prior research where children from depressed parents experienced more negative life events (e.g. daily hassles) than children from healthy parents (Hirsch, Moos, & Reischl, 1985). One way to explain this relation in the target group, is that parental depression hinders parental functioning by increasing the parents their pain sensitivity, or causing additional psychosomatic complaints (Gerrits, Marwijk, Van Oppen, Van der Horst, Penninx, 2015; Li, 2015). Moreover, parents with depressive symptoms show more pain, and sadness than non-depressed parents (Gerrits, Marwijk, Van Oppen, Van der Horst, Penninx, 2015; Li, 2015). As a result, the target group may perceive their parents to be in peril, increasing DIP. The relationship between parental health stressors and DHP can be explained by the finding that depression is related to poorer marital functioning (Gotlib, & Whiffen, 1989). This in turn, leads to irritation and sadness in the healthy parent. With regards to DPL, and parental health stressors, it is reported that
22 parental depression causes internalizing problems in the target group (Sieh, Visser-Meily, & Meijer, 2012). These internalizing problems may cause difficulties in school functioning, and social functioning. School functioning is impaired because internalizing problems such as depression, and anxiety, cause concentration problems and truancy (Hunt, & Hopko, 2009; Richardson, Richards, & Barkham, 2008; Visser et al., 2004). With regards to social functioning, children with internalizing problems show more social withdrawal and worries (Antonucci, & Bayer, 2017; Strauss, Forehand, Smith, & Frame, 1986). The link between internalizing problems and social withdrawal may also explain why the target group faces limitation of social activity. Due to developing internalizing problems they may lack interest to engage in social activities, which also leads to social isolation. Parental depression also leads to an increase in social isolation as it leads to neglect of the child (Mustillo, Dorsey, Conover, & Burns, 2011). As a result, the parent no longer functions as a source of social support.
Contrary to expectations, parental physical health, illness unpredictability, and caregiving were not related to child stress. The absence of a relationship between parental physical health and child stress suggests that parental physical health does not always lead to deprivation of family functioning (Visser et al., 2004). Moreover, Armistead and colleagues (1995) also mentioned that a poor parental physical health does not necessarily hinder parents in carrying out important parental functions such as child rearing and providing emotional support. The absence of a relationship between caregiving and child stress, can be explained through the method caregiving was measured. The present study only measured the extent to which the target group provided informal caregiving, or experienced caregiving responsiblity. However, Ireland and Pakenham (2010) found that these are not the attributes from caregiving that cause stress. Instead, caregiving only causes stress if it the target group to behave more maturely, or when they are not offered a choice in providing caregiving. Additionally, Reinhard and Horwitz found that caregiving is considered stressful if the parent shows disruptive behavior related to the illness (e.g. feeling pain, and sadness).
The finding that caregiving, and all the components from social stressors load together on one factor (caregiving + social stressors) is difficult to explain. Earlier research already found that an increase in caregiving is paired with limitation of social activity (Ireland & Pakenham, 2010). However, this does not explain the finding that caregiving also loads together with daily hassles on a common factor. Another explanation is that caregiving, daily hassles, limitation of social activity, and social isolation, were all measured from the same
23 participant, which results into higher correlations between the variables. Yet, this explanation is not sufficient, and future should focus on the possible relation between daily hassles, and caregiving.
As the result of keeping the analysis simple, the current study has some major weaknesses. Many relevant variables were excluded from the model such as the child’s coping style, attachment style, and the parents’ rearing style. Based on the TCS model and Rolland’s family-systems model it is expected that parental health stressors are related to parents’ rearing style, as Rolland (1999) stated that parental depression interferes with rearing the child effectively. As a result, this may cause all kinds of problems such as negatively altering the child's coping style and influencing the child's attachment style. A second weakness is that the present study did not use multilevel modeling, even though the data from the present study consists of two levels; family (1), and child (2). As siblings share the same parents and environment, variables showed high covariance and the results showed larger effects (Kline, 2016). A final weakness of the present study is that gender and age were not taken into account in the SEM model. The results showed that girls experienced more stress than boys. Moreover, age was related to many of the variables included in the analysis used variables namely, informal caregiving, DIP, DPL, social isolation, and child stress. Even though the study has some weaknesses it also contains many strengths such as a relatively high sample size, and careful data acquisition. Moreover, the analysis took into account the possible effect of common-method variance.
Future studies should focus on researching how parental mental, and social health effects child rearing, and child attachment, as they are related to child adjustment, and child stress. Another important topic of research is which factors cause parental mental, and social health, to drop in parents. Examining risk factors of low parental mental, and social health, in parental CMC will aid health professionals and social workers in recognizing the risk factors, and developing appropriate interventions. Moreover, the present study and additional research shows that parental mental, and social health are partially related to psychosocial stressors, but do not fully explain the increase in psychosocial stressors. Therefore, it is important to investigate which other mechanisms cause an increase in social stressors. Finally, as age has shown to be related to many of the social stressors and child stress, it is important to research how parental CMC impacts different age groups. Perhaps age groups differ in vulnerability and reactivity to parental CMC.
24 The present study has shown that the target group is at risk for psychological stress through social stressors. Based on the results, social workers should primarily aim to protect or improve the social functioning of the target group. Social workers should also observe the amount of daily hassles the target group is facing, and find ways to reduce these daily hassles. Moreover, the present study shows the importance of professional mental health care for parents with a CMC. Ensuring the mental and social well-being of the parent will improve parents' their functioning, and reduce the social stressors in the target group. Health psychologists play an important role in this task, as they can teach parents, and the target group with coping with adverse effects of the CMC.
25 References
Antonucci, M., & Bayer, J. K. (2017). Children's Moods, Fears and Worries Questionnaire: Validity with young children at risk for internalizing problems. Infant and Child Development, 26. doi: 10.1002/icd.1966
Armistead, L., Klein, K., & Forehand, R. (1995). Parental physical illness and child functioning. Clinical Psychology Review, 15, 409–422.
doi: 10.1016/0272-7358(95)00023-I
Bouman, T. K., Luteijn F. A., & van der Ploeg, F. A. E. (1985). Enige ervaringen met de Beck Depression Inventory. Tijdschrijft voor Psychologie, 13, 13–24.
Cruces, J., Venero, C., Pereda-Perez, I., & De la Fuente, M. (2014). The effect of psychological stress and social isolation on neuroimmunoendocrine communication. Current Pharmaceutical Design, 20(29), 4608–4628.
Dufour, M. J., Meijer, A. M., Van de Port, I., & Visser-Meily, J. M. A. (2006). Daily hassles and stress in the lives of children with chronically ill parents. Netherlands Journal of Psychology, 61, 54–64
Farrell, A. K., Simpson, J. A., Carlson, E. A., Englund, M. M., & Sung, S. (2017). The impact of stress at different life stages on physical health and the buffering effects of maternal sensitivity. Health Psychology, 36, 35–44. doi: 10.1037/hea0000424 Gerrits, M. M. J. G., Van Marwijk, H. W. J., Van Oppen, P., Van der Horst, H., & Penninx,
B. W. J. H. (2015). Longitudinal association between pain, and depression and anxiety over four years. Journal of Psychosomatic Research, 78, 64–70.
doi: 10.1016/j.jpsychores.2014.10.011
Gotlib, I. H., & Whiffen, V. E. (1989). Depression and marital functioning – An examination of specificity and gender differences. Journal of Abnormal Psychology, 98, 23–30 doi: 10.1037/0021-843X.98.1.23
Grant, N., Hamer, M., & Steptoe, A. (2009). Social isolation and stress-related cardiovascular, lipid, and cortisol responses. Annals of Behavioral Medicine, 37, 29–37. doi: 10.1007/s12160-009-9081-z
26 Hartong, I., Krol, M., Maaskant, A., Plate, A., Schuszler, D., & Meijer, A. M. (2003). Psst . . .
Are you asleep? Study on the quality of sleep. Unpublished manuscript, University of Amsterdam.
Hawthorne, G. (2008). Perceived social isolation in a community sample: Its prevalence and correlates with aspects of peoples’ lives. Social Psychiatry and Psychiatric Epidemiology, 43, 140–150. doi: 10.1007/s00127-007-0279-8
Hirsch, B. J., Moos, R. H., & Reischl, T. M. (1985). Psychosocial adjustment of adolescent children of a depressed, arthritic, or normal parent. Journal of Abnormal Psychology, 94, 154–164. doi: 10.1037/0021-843X.94.2.154
Hocking, M. C., & Lochman, J. E. (2005). Applying the transactional stress and coping model to sickle cell disorder and insulin-dependent diabetes mellitus: Identifying psychosocial variables related to adjustment and intervention. Clinical Child and Family Psychology Review, 8, 221–246. doi: 10.1007/s10567-005-6667-2
Huizinga, G. A., Visser, A., Van der Graaf, W. T. A., Hoekstra, H. J., Klip, E. C., Pras, E., Hoekstra-Weebers, J. E. H. M. (2005). Stress response symptoms in child and young adult children of parents diagnosed with cancer. European Journal of Cancer, 41, 288–295. doi: 10.1016/j.ejca.2004.10.005
Hunt, M. K., & Hopko, D. R. (2009). Predicting high school truancy among students in the Appalachian South. The Journal of Primary Prevention, 30, 549–567. doi: 10.1007/s10935-009-0187-7
Ireland, M. J., & Pakenham, K. I. (2010). Youth adjustment to parental illness or disability: the role of illness characteristics, caregiving, and attachment. Health and Medicine, 15, 632–645. doi: 10.1080/13548506.2010.498891
Kline, R. B. (2016). Principles and practice of structural equation modeling (4thed.) New York: The Guildford Press.
Koolhaas, J. M., Bartolomucci, A., Buwalda, B., De Boer, S. F., Flugge, G., Korte, S. M., . . . Fuchs, E. (2011). Stress revisited: A critical evaluation of the stress concept. Neuroscience & Biobehavioral Reviews, 35, 1291–1301.
27 Li, J. (2015). Pain and depression comorbidity: A preclinical perspective. Behavioural Brain
Research, 276, 92–98. doi: 10.1016/j.bbr.2014.04.042
Matud, M. P. (2004). Gender differences in stress and coping styles. Personality and Individual Differences, 37, 1401–1415. doi: 10.1016/j.paid.2004.01.010
Meijer, A. M., Van Oostveen, S. J. E., & Stams, G. J. J. M. (2008) De relatie tussen mantelzorg, ziekte van de ouder en gedragsproblemen bij kinderen [Caring for an ill parent: The relationship between caring, parental disease, and the child's problem behavior]. Kind en Adolescent, 29, 208–220.
Mustillo, S. A., Dorsey, S., Conover, K., & Burns, B. J. (2011). Parental depression and child outcomes: The mediating effects of abuse and neglect. Journal of Marriage and Family, 73, 164–180. doi: 10.1111/j.1741-3737.2010.00796.x
O’Connor, T. G., Dunn, J., Jenkins, J. M., Pickering, K., & Rasbash, J. (2001). Family settings and children’s adjustment: Differential adjustment within and across families. The British Journal of Psychiatry, 179, 110–115. doi: 10.1192/bjp.179.2.110
Pakenham, K. I., Bursnall, S., Chiu, J., Cannon, T., & Okochi, M. (2006). The psychosocial impact of caregiving on young people who have a parent with an illness or disability: Comparisons between young caregivers and noncaregivers. Rehabilitation
Psychology, 51, 113-126. doi: 10.1037/0090-5550.51.2.113
Pedersen, S., & Revenson, T. A. (2005). Parental illness, family functioning, and adolescent well-being: A family ecology framework to guide research. Journal of Family Psychology, 19, 404–419. doi: 10.1037/0893-3200.19.3.404
Perkins, M., Howard, V. J., Wadley, V. G., Crowe, M., Safford, M., Haley, W. E., . . . Roth, D. L. (2013). Caregiving strain and all-cause mortality: Evidence from the REGARDS study. Psychological Sciences and Social Sciences, 68, 504–512.
doi: 10.1093/geronb/gbs084.
Podsakoff, P. M., MacKenzie, S. B., & Lee, J. Y. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879–903. doi: 10.1037/0021-9101.88.5.879
28 Reinhard, S. C., & Horwitz, A. V. (1995). Caregiver burden: Differentiating the content and
consequences of family caregiving. Journal of Marriage and Family, 57, 741–750. doi: 10.2307/35928
Richardson, R., Richards, D. A., & Barkham, M. (2008). Self-help books for people with depression: A scoping review. Journal of Mental Health, 17, 543–552.
doi: 10.1080/09638230802053334
Rolland, J. S. (1999). Parental illness and disability: A family systems framework. Journal of Family Therapy, 21, 242–266. doi: 10.1111/1467-6427.00118
StataCorp. (2013). Stata Statistical Software: Release 13 [Computer software]. Stata College Station, TX: StataCorp LP
Schrag, A., Morley, D., Quinn, N., & Jahanshahi, M. (2004). Impact of parkinson’s disease on patients' adolescent and adult children. Parkinsonism and Related Disorders, 10, 391–397. doi: 10.1016/j.parkreldis.2004.03.011
Sieh, D. S., Meijer, A. M., & Visser-Meily, J. M. A. (2010). Risk factors for stress in children after parental stroke. Rehabilitation Psychology, 55, 391–397.
Sieh, D. S., Meijer, A. M., Oort, F. J., Visser-Meily, J. M. A., & Van der Leij, D. A. V. (2010). Problem behavior in children with a chronically ill parent: A meta-analysis. Clinical Child and Family Psychology Review, 3, 384–397
Sieh, D. S. (2012). The impact of parents’ chronic medical condition on children. Unpublished doctoral dissertation. University of Amsterdam.
Sieh, D. S., Visser-Meily, J. M. A., & Meijer, A. M. (2012). Differential outcomes of adolescents with chronically ill and healthy parents. Journal of Child and Family Studies. doi: 10.1007/s10826-012-9570-8.
Strauss, C. C., Forehand, R., Smith, K., Frame, C. L. (1986). The association between social withdrawal and internalizing problems of children. Journal of Abnormal Child Psychlogy, 14, 525–535. doi: 10.1007/BF01260521
29 Thielsch, C., Andor, T., & Ehring, T. (2015). Do metacognitions and intolerance of uncertainty predict worry in everyday life? An ecological momentary assessment study. Behavior Therapy, 46, 532–543. doi: 10.1016/j.beth.2015.05.001
Verhaeghe, S., Defloor, T., Grypdonck, M. (2004). Stress and coping among families of patients with traumatic brain injury: A review of the literature. Journal of Clinical Nursing, 14, 1004–1012. doi: 10.1111/j.1365-2702.2005.01126.x
Verkuil, B., Brosschot, J. F., Borkovec, T. D., & Thayer, J. F. (2009). Acute autonomic effects of experimental worry and cognitive problem solving: Why worry about worry? International Journal of Clinical and Health Psychology, 9(3), 439–453. Visser, A., Huizinga, G. A., Van der Graaf, W. T. A., Hoekstra, H. J., & Hoekstra-Weebers,
J. E. H. M. (2004). The impact of parental cancer on children and the family: A review of the literature. Cancer Treatment Reviews, 30, 683–694.
doi: 10.1016/.j.ctrv.2004.06.001
World health organization (2016): Noncommunicable diseases country profiles 2014 Netherlands. Retrieved august 2, 2017, from
http://www.who.int/nmh/countries/nld_en.pdf?ua=1
Worsham, N. L., Compas, B. E., & Ey, S. (1997). Children's coping with parental illness. In S. A. Wolchik & I.N. Sandler (Eds.), Handbook of children's coping: Linking theory and intervention (pp. 195–213). NewYork: Plenum Press.
30 Acknowledgment
Foremost, we would like to show our gratitude to our supervisor, Dominik Sieh, for providing us with great guidance and feedback through his insight, wisdom, and expertise, which increased the level of the present study. We also thank James Sheerin, for providing excellent feedback, and tutoring in the correct use of grammar, spelling, and APA style, Lieke van Wensveen, for providing feedback on our research proposal, Fruzsina Szanyo, for providing feedback on the tables and figures, and Fatih Bogaards, for giving us feedback on the scientific poster, and guiding us with the technical aspects of Microsoft Office Word.
