bs_bs_banner
Journal of Intellectual Disability Research
doi: 10.1111/jir.12127
1
Risk factors associated with the mental health of fathers of children with an intellectual disability in Australia R. Giallo,1,2,3 M. Seymour,1 J. Matthews,1 S. Gavidia-Payne,2 A. Hudson1 & C. Cameron1 1 Parenting Research Centre, East Melbourne, Vic., Australia 2 RMIT University, Bundoora, Vic., Australia 3 Murdoch Children’s Research Institute, Royal Children’s Hospital, Parkville, Vic., Australia.
Abstract Background Raising a child with a disability places considerable demands and stress on parents, which can contribute to mental health difficulties. Research has primarily focused on mothers’ mental health, and our understanding of the effects on fathers remains limited. The factors that place fathers at increased risk of mental health difficulties are also poorly understood. This study aimed to redress these gaps by reporting on the mental health of a large sample of fathers of children with an intellectual disability (ID) (aged 3–15 years), comparing this to published Australian norms and mothers of children with ID. The second aim of the study was to explore risk factors associated with fathers’ mental health. Method The data for this study come from 315 Australian fathers of children (aged 3–15 years) with ID, who participated in the large-scale evaluation of the Signposts for building better behaviour programme. Fathers completed a range of self-report questionnaires at baseline including the Depression Anxiety Stress Scale (DASS).
Correspondence: Ms Monique Seymour, Parenting Research Centre, 5/232 Victoria Parade, East Melbourne, Vic. 3002, Australia (e-mail: [email protected]).
Results Fathers in the present sample reported significantly more symptoms of depression and stress than the Australian normative data, with approximately 6–8% reporting symptoms in the severe to extremely severe range. The strongest predictors of fathers’ mental health difficulties were children’s behaviour problems, daily stress arising from fathers’ own needs and children’s care needs, and low parenting satisfaction. Socio-economic factors did not predict mental health difficulties. Conclusion This study is among one of the first to report the mental health of fathers of children with a disability in Australia. Findings highlight that some fathers of children with ID are at heightened risk of experiencing mental health difficulties, underscoring the importance of the provision of information and interventions to promote their mental health. Keywords children with disabilities, father mental health, risk and protective factors, socio-ecological
Parents raising children with disabilities experience considerable caregiving demands and stressful life circumstances that can contribute to mental health difficulties. Research to date however, has primarily focused on mothers (Resch et al. 2012; Seymour et al. 2013b) and our understanding of the extent to
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 2 R. Giallo et al. • Mental health of fathers
which fathers experience mental health difficulties such as depression, stress and anxiety, remains limited. Similarly, studies seeking to understand the factors associated with fathers’ mental health difficulties are scarce (Olsson & Hwang 2001; Hastings & Brown 2002; Hastings 2003; Herring et al. 2006; Gerstein et al. 2009). This might be in part due to the focus on mothers as they are often the primary caregivers of children with disabilities (Hastings 2003; Olsson & Hwang 2006), and the difficulties associated with recruiting fathers in research (Costigan & Cox 2001; Sherr et al. 2006). We had an opportunity to redress these gaps in our knowledge by drawing upon data from a large sample of over 300 Australian fathers of children with intellectual disabilities (ID; aged 3–15 years) who participated in a large-scale evaluation of an intervention to help families manage their children’s behavioural difficulties.
The mental health of fathers of children with disabilities Studies reporting on the mental health difficulties of fathers of children with disabilities are scarce. Of those available, estimates range between 12% and 18% for depressive symptoms (Olsson & Hwang 2001; Hastings & Brown 2002) and 25% for symptoms of anxiety (Olsson & Hwang 2001). For example, it was reported that 18% of 167 fathers of children with ID (aged 1–16 years) in Sweden had clinically significant depressive symptoms on the Beck Depression Inventory (Olsson & Hwang 2001), and 12% of 56 fathers of young children with developmental disabilities (aged 2–6 years) had clinically significant depressive symptoms on the Center for Epidemiologic Studies-Depression Scale (Bristol et al. 1988). In another study of 20 fathers of primary school aged children with an autism spectrum disorder (ASD) in the UK, 25% and 15% of fathers had scores in the at risk or clinical ranges for anxiety and depression, respectively, on the Hospital Anxiety and Depression Scale (Hastings & Brown 2002). Despite these findings, other studies have reported no differences in depressive symptoms between fathers of children with disabilities (i.e. ASD and Down syndrome) and without disabilities (Fisman et al. 1989; Dumas et al. 1991).
The variation in estimates and outcomes reported across studies is likely due to methodological differences associated with the measures used to assess mental health (i.e. Beck Depression Inventory, Hospital Anxiety and Depression Scale), the nature of the samples (i.e. parents of children with ASD, ID or Down syndrome; variable age ranges), and whether fathers’ mental health has been compared with normative data or a comparison sample of fathers with typically developing children. Furthermore, the existing literature is limited on several counts. First, fathers are typically embedded within studies primarily about mothers (e.g. Resch et al. 2012; Feldman et al. 2007). While this is often due to difficulties in recruiting fathers, these studies often lack sufficient numbers to make meaningful comparisons to mothers or to fathers of typically developing children, limiting generalisability of findings. Second, there is an almost exclusive focus on depressive symptoms despite research with fathers of typically developing children indicating that they tend to underreport the extent to which they experience distress when asked specifically about depression compared with other symptoms such as stress or fatigue (Gay et al. 2004; Giallo et al. 2012a; Seymour et al. 2013a). Focusing on a broader range of distress symptoms may provide a more accurate assessment of the extent to which fathers experience mental health difficulties. Finally, another limitation pertains to the lack of Australian-based studies. While there are social, economic and political similarities between Australia and other developed countries (i.e. the UK, Canada, the USA, Sweden), previous studies are based on small, convenient and relatively homogenous samples. These findings may not be generalisable to fathers of children with disabilities in Australia. To date, one large population-based study of 4983 Australian families reported no difference in the rates of psychological distress as assessed by the Kessler-6 among fathers of children (aged 5 years) at risk of disability (n = 297) and fathers of children at no risk (Emerson & Llewellyn 2008). It is important to note that these findings were based on children with a probable diagnosis of disability or chronic health condition based upon a range of physical, behavioural and cognitive functioning outcomes. It is likely that some children
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 3 R. Giallo et al. • Mental health of fathers
were misclassified, and the sample likely reflects a very diverse and heterogeneous group of children and their families. We had an opportunity to redress these limitations using data from over 300 fathers of children with ID (aged 3–15 years) who participated in a large-scale evaluation of an intervention to help parents manage their children’s behavioural difficulties. The Depression Anxiety Stress Scale (DASS; Lovibond & Lovibond 1995), a well-established measure with clinical cuts offs and Australian-based normative data, was used to assess mothers’ and fathers’ depressive, anxiety and stress symptoms. Therefore, the first aim of the study was to assess the extent to which fathers reported depressive, anxiety and stress symptoms in the established clinically elevated ranges of the DASS, and how this compared with Australian normative data (Crawford et al. 2011).
Comparisons between mothers’ and fathers’ mental health Although the prevalence of mental health problems among fathers of children with disabilities remains unclear, studies consistently report that fathers tend to experience fewer mental health difficulties than mothers (Olsson & Hwang 2001; Mugno et al. 2007; Emerson & Llewellyn 2008; Olsson et al. 2008). For example, Gray & Holden (1992) found that Australian mothers of children with an ASD reported significantly higher levels of depression and anxiety than fathers. Similarly, in Vesson’s (1999) Northern European study of 151 mothers and 57 fathers of children with disabilities, mothers reported feeling more worried, sad, tired, helpless, depressed and nervous than fathers. These findings suggest that fathers may adjust better to, or be less affected by, disability in the family than mothers. It has been proposed that mothers are at greater risk of mental health difficulties because they take on more domestic and caregiving responsibilities than fathers (Olsson & Hwang 2006). Fathers, on the other hand, are more likely to be employed, and may be buffered by the benefits of participation in paid employment (Olsson & Hwang 2006). Although this is yet to be substantiated by research, it does suggest that mothers’ and fathers’ mental
health may be influenced by different factors. Some evidence for this comes from Olsson & Hwang (2008). They found that negative perceptions of the impact of disability on the family, and low satisfaction in participation in work, leisure and social activities predicted depressive symptoms for mothers; while the only factor predicting fathers’ depressive symptoms was their self-rated quality of health. It is evident that the context and experience of caregiving, parenting, family roles and responsibilities may differ considerably for mothers and fathers. It stands to reason that there may be a different set of risk factors for mothers’ and fathers’ mental health difficulties, yet there is little research in this area.
Factors associated with fathers’ mental health Studies identifying the risk factors associated with fathers’ mental health are scarce, and those available are typically embedded in research about mothers, examine a narrow range of influences, and fail to account for the impact of socio-economic disadvantage often experienced by families of children with disabilities (Emerson et al. 2006; Emerson & Llewellyn 2008). A review of the literature within the context of these key limitations will now be discussed. Compared with mothers of children with disabilities (Hastings & Brown 2002; Emerson et al. 2006; Feldman et al. 2007; Gerstein et al. 2009; Giallo et al. 2011; Seymour et al. 2013b), few studies have identified risk factors specifically for fathers’ mental health. Of the research available, child behaviour difficulties and partner mental health have received the most attention and produced mixed findings. For example, some studies have reported associations between child behaviour problems and fathers’ overall health (Herring et al. 2006) and perceptions of parenting stress (Dumas et al. 1991; Hastings 2003), but not anxiety and depressive symptoms. Several studies have found no association between mothers’ mental health and fathers’ stress (Herring et al. 2006), depressive or anxiety symptoms (Hastings & Brown 2002; Hastings 2003), while others have. For instance, Olsson & Hwang (2001)
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 4 R. Giallo et al. • Mental health of fathers
found a small to moderate correlation between the depressive symptoms of 216 mothers and fathers of children (aged 7–10 years) with ASD and/or ID. Similarly, a longitudinal study found that the course of fathers’ parenting stress when their children with an ID were 36 to 60 months was predicted by mother’s mental health (Gerstein et al. 2009). The daily demands and challenges arising from parenting and parent-child interactions that may have an impact on fathers’ mental health have also received some limited attention. For example, Gavidia-Payne et al. (1997) found that daily stress associated with children’s behaviour, communication and caregiving needs, as well as parents’ own needs (i.e. having time for themselves or others in the family) were associated with increased anxiety, stress and depressive symptoms among 96 mothers and 12 fathers of children (aged 4–19 years) with disabilities. Parenting stress and child behaviour difficulties have the potential to undermine parental self-efficacy (PSE), which refers to parents’ sense of competence in their parenting role (Coleman & Karraker 1997). Hastings & Brown (2002) found a complex relationship between children’s behaviour problems, PSE and mental health for 26 mothers and 20 fathers of preschool aged children with ASD. They found that fathers of children with behaviour difficulties were more anxious when they had a lowered sense of PSE. This is consistent with research with fathers of typically developing children in the postnatal period, whereby low PSE has been associated with increased psychological distress (Giallo et al. 2013). While relationships between child behaviour difficulties, parenting stress, PSE and fathers’ mental health are likely to be complex and bi-directional, this small body of research highlights that they are also key factors associated with fathers’ mental health that warrant further investigation in larger samples of fathers. Finally, a key limitation of previous research is the failure to consider the influence of socioeconomic disadvantage on fathers’ mental health. This is important because it is well established that families of children with disabilities are more likely than other families to experience significant economic hardships (Emerson 2003; Emerson et al. 2006; Emerson & Llewellyn 2008). Such differences in economic and social resources have been shown to partially account for differences between families
of children with and without disabilities on a range of outcomes (Emerson & Llewellyn 2008; Emerson et al. 2010). For example, limited participation in paid employment (Olsson & Hwang 2006), poverty and socio-economic disadvantage (Emerson et al. 2006; Emerson & Llewellyn 2008) have been associated with mental health difficulties among mothers. These factors have received little attention with fathers.
Aims of the study In summary, studies exploring the extent to which fathers of children with disabilities experience mental health difficulties are scarce, and the factors that heighten their risk remains poorly understood. The current study sought to redress these significant gaps in our knowledge by reporting on the mental health of a large sample of fathers of children with ID (aged 3–15 years) who participated in an evaluation of an Australian intervention to help families manage their children’s behavioural difficulties. The first aim of the study was to report on the extent to which fathers reported mental health difficulties, and compare this to published Australian norms (Crawford et al. 2011) and mothers who also participated in the intervention study. It was hypothesised that fathers would report fewer symptoms of depression, anxiety and stress than mothers, but significantly more symptoms than adults in the general population. The second aim of the study was to explore a broader range of risk factors associated with fathers’ mental health than previously available. Socioecological models of health (Brofenbrenner 1979; McLaren & Hawe 2005; World Health Organisation 2013) provide useful frameworks to advance this research. These theories focus on identifying factors at different levels of an individual’s immediate and broader social environment. Examples of risk and protective factors at the level of the individual can include age, personality, values, beliefs and coping skills. At the interpersonal level, quality of relationships, social support and stress in the family environment are key factors, while socio-economic influences, employment and availability of services are potential risk factors in the broader social environment. This model has been widely used in many
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 5 R. Giallo et al. • Mental health of fathers
fields (e.g. health behaviours and promotion, violence prevention, weight management and nutrition). It has also been used as a guiding framework to identify the factors associated with men’s health (e.g. prostate examination; Woods et al. 2006), as well as postnatal mental health difficulties among Australian fathers of typically developing children (Giallo et al. 2012b, 2013; Seymour et al. 2013a). Drawing upon this model, we took advantage of the available baseline data from the intervention study to explore risk factors for fathers’ mental health difficulties at the individual level (i.e. parenting efficacy and satisfaction) and interpersonal level (i.e. child behaviour difficulties, parenting stress, partner mental health, social support). Given previous research indicating that socio-economic disadvantage experienced by families of children with disabilities partially accounts for mothers’ mental health difficulties (Emerson & Llewellyn 2008; Emerson et al. 2010), first we sought to examine the unique contribution of indicators of socio-economic disadvantage (i.e. socio-economic factors such as employment, education, neighbourhood disadvantage) to fathers’ mental health. Then we examined the contribution of the individual and interpersonal level factors over and above socio-economic disadvantage factors. We hypothesised that socioeconomic disadvantage would be associated with fathers’ mental health difficulties, and that individual and interpersonal level factors would predict fathers’ mental health difficulties over and above the contribution of the socio-economic factors.
Method Study design The data for this study forms part of a large-scale evaluation of the Signposts for building better behaviour programme for parents of children (aged 3–15 years) with ID (Hudson et al. 2003, 2008). The aim of this programme was to teach parents skills to manage any difficult behaviours of their child before those behaviours escalate to a level that requires intensive intervention by specialists. Both mothers and fathers from all families were invited to participate in the programme. Details of the study design and sampling information are detailed elsewhere (Hudson et al. 2003, 2008). Ethics approval for the
evaluation study was granted by the Department of Human Services Ethics Committee, Victoria, Australia. Dissemination of the programme was jointly funded by the Australian Commonwealth Government’s former Department of Families, Community Services and Indigenous Affairs (now known as the Department of Social Services) and the Victorian State Government’s Department of Human Services. The current sample consisted of fathers for who self-report data on the variables of interest at pre-intervention was available.
Measures Demographic and family background information Information about the parent’s age, gender, language spoken, level of education and employment status were collected. Additionally, information pertaining to the age and gender of the child with an ID, specific disability diagnosis, severity of disability and educational placement was obtained. Families’ socio-economic status was assessed using a measure of neighbourhood disadvantage, the Index of Relative Socio-economic Disadvantage from the Australian Bureau of Statistics, Socio-economic Indexes for Areas (SEIFA; Trewin 2003). This index is derived from 2001 population census data pertaining to income, educational attainment and employment. The SEIFA is constructed so that the mean score is 1000, with scores below 1000 indicative of being less advantaged (Australian Bureau of Statistics 2006). The Depression, Anxiety and Stress Scale (DASS; Lovibond & Lovibond 1995) The DASS was developed and normed in Australia, and assesses the negative emotional states of depression, anxiety and stress over the past week. Both mothers and fathers participating in the Signposts evaluation were asked to complete this measure. There are 42 items that are rated on a four-point scale ranging from 0 = ‘Did not apply to me at all’ to 3 = ‘Applied to me very much, or most of the time’. Although the DASS is not a diagnostic tool for mood disorders, it does provide clinical cut-offs to indicate whether a respondent is experiencing significant levels of distress. Internal consistency for the current sample of both mothers and fathers was Cronbach’s α = 0.97.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 6 R. Giallo et al. • Mental health of fathers
The Developmental Behaviour Checklist (DBC; Einfeld & Tonge 2002) The DBC is a 96-item scale designed to assess difficult behaviour of children with disabilities, and was also developed in Australia. Items are rated on a three-point scale, where 0 = ‘not true as far as you know’ to 2 = ‘very true or often true’. It consists of five sub-scales: Disruptive, Self-Absorbed, Communication Disturbance, Anxiety and Social Relationships. It also gives a Total Problems score, which was used in this study. Although normative data or clinical cut-offs have not been established for this measure, data from a large sample of 454 Australia children with mild to profound levels of ID have been published (Einfeld & Tonge 1996a,b). The mean scores on the Total Problems scales for children with mild, moderate, severe and profound ID were reported to be 43.9, 42.5, 43.1 and 26.0 respectively. Internal consistency for the current sample of fathers was Cronbach’s α = 0.95. Parenting Hassles Scale (PHS; Gavidia-Payne et al. 1997) The PHS is a 50-item scale specifically designed to assess potential stressors experienced by families of children with disabilities on a daily basis, as distinct from stress symptoms measured by the DASS. It consists of 12 sub-scales, however, given that the Signposts programme focused on assisting parents to manage their children’s behaviour, only the Child Behaviour/Needs, Parent Needs/Characteristics, and Education and Child Development sub-scales were included in the evaluation study. Each item is also rated on a five-point scale ranging from 1 = ‘no hassle’ to 5 = ‘major hassle’, with higher scores indicating a greater impact of hassles. The internal consistency for the current sample of fathers was Cronbach’s α = 0.95. Parenting Sense of Competence Scale (PSOC; Johnston & Mash 1989) The PSOC is a 16-item scale used to measure parents’ views of their competence as parents on two scales: the Satisfaction sub-scale, measuring the extent to which parents are satisfied with their parenting role; and the Efficacy sub-scale, measuring the extent to which parents feel they are managing
the role of being a parent. The items were rated on a six-point scale, ranging from 1 = ‘strongly agree’ to 6 = ‘strongly disagree’. Satisfaction and Efficacy items are added together to obtain a total score (range 16–96), where higher scores indicate greater perceived competence. The internal consistency for the current sample of fathers was Cronbach’s α = 0.70. Sources of Support Sources of Support was designed specifically for this study to assess a comprehensive range of sources of support for parents of children with disabilities. It consists of 20-items asking whether parents use these sources of support and how helpful the support received is. Example sources of support include from parents, friends, co-workers, parent support group, early intervention teacher, school or childcare educator. Items are rated on a six-point scale ranging from 0 = ‘not available’ to 5 = ‘extremely helpful’. The internal consistency for the current sample of fathers was Cronbach’s α = 0.75.
Results Sample characteristics Of the 1551 families who participated in the Signposts evaluation, 1323 were mothers and 315 were fathers. Demographic information for fathers is provided in Table 1. The majority of fathers were English speaking, employed and had an educational attainment of high school or above. Of the fathers participating in this study, 252 completed the Signposts programme with their partner. There were significantly more unemployed fathers and fathers from non-English-speaking backgrounds who participated in the programme by themselves compared with those who participated with their partner (P < 0.05). Demographic information for the children with a disability is presented in Table 2. The majority of the children were male, had a diagnosis of ASD and were attending a specialist or early intervention programme. The mean score for the DBC Total Difficulties scale was 49.48 (25.06), which is consistent with previous studies reporting mean scores for
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 7 R. Giallo et al. • Mental health of fathers
Table 2 Child with disability characteristics (N = 315)
Table 1 Father characteristics (N = 315)
Variable Language spoken English only Employment status Employed Not in paid employment Not reported Highest level of education completed Some high school Completed high school TAFE, Trade Certificate, Diploma Tertiary (degree, post-graduate degree) Not reported SEIFA – Index of Relative Socio-economic Disadvantage (M, SD)
n (%)
286 (90.8%) 261 (82.9%) 44 (14.0%) 10 (3.2%) 97 (30.8%) 26 (8.3%) 39 (12.4%) 58 (18.4%) 95 (30.1%) 1009.68 (58.70)
M, mean; TAFE, technical and further education; SD, standard deviation; SEIFA, Socio-Economic Indexes for Areas.
Australian children with mild to profound ID ranging from 26.0 to 43.9 (Einfeld & Tonge 1996a,b).
The extent to which fathers’ report depressive, anxiety and stress symptoms Descriptive statistics for the DASS sub-scales are presented in Table 3, and Table 4 presents the percentage of fathers in the Normal, Mild, Moderate, Severe and Extremely Severe ranges on each of the DASS sub-scales. While the majority of fathers scored in the Normal range on all sub-scales, a small proportion reported symptoms of depression (7.9%), anxiety (6.0%) and stress (7.9%) in the Severe to Extremely Severe ranges. Comparisons between fathers’ and mothers’ distress symptoms were made using paired samples t-tests for a subset of the sample in which both father and mother data from the same family were available (n = 220). Mothers reported significantly higher depressive, anxiety and stress symptoms than fathers (see Table 5), and this was associated with moderate to large effect sizes for Cohen’s d; where 0.2, 0.5 and 0.8 are small, medium and large effect sizes respectively. Finally, one-sample t-tests were used to compare fathers’ DASS sub-scale scores with published Australian norms (Crawford et al. 2011). The normative
Variable
n (%)
Age of child with disability (years) (M, SD) Gender of child with disability Male Female Not reported Intellectual disability plus other condition Autism spectrum disorder Vision impairment Hearing impairment Down syndrome Cerebral palsy Epilepsy or other seizure disorder Psychiatric disorder Attention deficit and hyperactivity disorder Acquired brain injury Health problems (i.e. asthma, food allergies) Severity of intellectual disability – formally assessed Mild Moderate Severe Profound Not reported How well child communicates with speech Very well Well Reasonably well Poorly Very poorly Not reported How well child understands speech Very well Well Reasonably well Poorly Very poorly Not reported Educational placement Specialist and early intervention programme Regular preschool/childcare Special school Special development school Home tutoring Regular school with integration aide Regular school without integration aide Other (i.e. speech therapy, family day care)
7.83 (5.52) 220 (69.8%) 89 (28.3%) 6 (1.9%) 141 (44.8%) 19 (6.0%) 17 (5.4%) 15 (4.8%) 8 (2.5%) 26 (8.3%) 6 (1.9%) 46 (14.6%) 10 (3.2%) 24 (7.6%)
54 (17.1%) 23 (7.3%) 7 (2.2%) 1 (0.3%) 230 (73.1%) 46 (14.6%) 66 (21.0%) 87 (27.6%) 46 (14.6%) 31 (9.8%) 39 (12.7%) 51 (16.2%) 81 (25.7%) 100 (31.8%) 20 (6.3%) 8 (2.5%) 55 (17.4%) 114 (36.2%) 97 (30.8%) 64 (20.3%) 44 (14.0%) 5 (1.6%) 41 (13.0%) 29 (9.2%) 24 (7.6%)
M, mean; SD, standard deviation.
sample was comprised of 497 men and women from the general adult population, with distributions by age and gender consistent with Australian census data. Men with an education level of less than Year
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 8 R. Giallo et al. • Mental health of fathers
Table 3 Descriptive statistics for the current sample and normative data on the Depression Anxiety Stress Scale (DASS)
Current sample (n = 315)
Depression Anxiety Stress
Normative data (n = 497)
M
SD
M
SD
t
Cohen’s d (95% CI)
6.95 3.49 11.00
7.89 5.24 8.24
5.02 3.36 8.10
7.54 5.07 8.40
4.35* 0.42 6.24*
0.25 (0.11, 0.39) 0.03 (−0.12, 0.17) 0.35 (0.21, 0.49)
* P < 0.001. CI, confidence interval; M, mean; SD, standard deviation.
Table 4 Percentage of fathers in the normal and clinical ranges on the Depression Anxiety Stress Scale (DASS) (N = 315)
Range
Depression n (%)
Anxiety n (%)
Stress n (%)
Normal Mild Moderate Severe Extremely severe
228 (72.4) 33 (10.5) 32 (10.2) 14 (4.4) 8 (2.5)
261 (82.9) 22 (7.0) 13 (4.1) 11 (3.5) 8 (2.5)
210 (66.9) 39 (12.4) 41 (13.0) 11 (3.5) 14 (4.4)
12 were slightly under-represented. Given that no gender differences were found between men and women, it was recommended by the authors that combined normative data for men and women is used. Fathers of children with disabilities in the present sample reported significantly higher depression and stress than the normative sample, and this was associated with small to moderate effect sizes for Cohen’s d; where 0.2, 0.5 and 0.8 are small, medium and large effect sizes respectively.
Factors associated with fathers’ mental health difficulties First, bivariate correlation analyses were conducted to examine the relationships between the symptoms of depression, stress and anxiety, and the predictor variables (see Table 6). Next, a series of hierarchical regression analyses were conducted separately for fathers’ depressive, anxiety and stress symptoms. In the first block, indicators of socio-economic status
were entered. This included fathers’ employment status, educational attainment, neighbourhood disadvantage and whether they were from a nonEnglish-speaking background. In the second block, the individual and interpersonal factors (excluding mothers’ mental health difficulties) were entered to assess their contribution over and above socioeconomic factors. Finally, given that 95 cases did not have mental health data for mothers, the regression analyses were conducted with a small subset of cases for which mothers’ mental health data was available. Depressive symptoms At the first step of the hierarchical regression analysis, the socio-economic variables were entered, and these did not significantly predict fathers’ depressive symptoms (R2 = 0.02, Adj R2 = 0.01, F4,310 = 1.39, P = 0.23). In the second step, the individual and interpersonal variables were entered, resulting in a significant change in R2 (ΔR2 = 0.32, F7,303 = 21.21, P < 0.001), with the model now significantly accounting for 32% of the variance in fathers’ depressive symptoms (R2 = 0.34, Adj R2 = 0.32, F11,303 = 14.24, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were child behaviour difficulties, parenting stress arising from the child’s behaviours and needs, as well as the fathers’ own needs, and low parenting satisfaction. The pattern of results was similar when mothers’ mental health symptoms were entered into the model (R2 = 0.34, Adj R2 = 0.30, F14,219 = 7.56, P < 0.001); however, mothers’ depressive, anxiety
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 9 R. Giallo et al. • Mental health of fathers
Fathers
Depression Anxiety Stress
Mothers
M
SD
M
SD
t
Cohen’s d (95% CI)
6.52 3.19 10.84
7.85 5.16 8.20
9.70 5.64 15.40
9.82 7.60 9.61
−4.34* −4.38* −6.27*
−0.36 (−0.54, −0.19) −0.39 (−0.56, −0.22) −0.52 (−0.69, −0.34)
Table 5 Descriptive statistics for fathers and mothers in the current sample on the Depression Anxiety Stress Scale (DASS) (N = 220)
* P < 0.001. CI, confidence interval; M, mean; SD, standard deviation.
and stress symptoms did not provide a significant amount of unique predictive variability in the model. Anxiety symptoms At the first step of the regression analysis with the socio-economic variables entered, the model did not significantly predict fathers’ anxiety symptoms (R2 = 0.03, Adj R2 = 0.02, F4,314 = 2.21, P = 0.067. In the second step, the individual and interpersonal variables were entered, resulting in a significant change in R2 (ΔR2 = 0.22, F7,303 = 12.62, P < 0.001), with the model now significantly accounting for 22% of the variance in fathers’ anxiety symptoms (R2 = 0.25, Adj R2 = 0.22, F11,303 = 9.05, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were child behaviour difficulties, parenting stress arising from the child’s behaviours and fathers’ own needs, and low parenting satisfaction. The model including mothers’ mental health symptoms remained significant, R2 = 0.25, Adj R2 = 0.20, F14,219 = 4.80, P < 0.001, although parenting satisfaction and parenting stress related to their child’s needs were no longer significant predictors. Mothers’ depressive symptoms provided a significant amount of unique variability in this model (β = 0.25, t = 2.50, P = 0.013). Stress symptoms At the first step of the analyses with the socioeconomic variables entered, the model did not significantly predict fathers’ stress symptoms (R2 = 0.01, Adj R2 = 0.01, F4,314 = 0.74, P = 0.563. In the next step with the individual and interpersonal variables entered, there was a significant
change in R2 (ΔR2 = 0.38, F7,303 = 26.38, P < 0.001), with the model now significantly accounting for 36% of the variance in fathers’ anxiety symptoms (R2 = 0.39, Adj R2 = 0.36, F11,303 = 17.21, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were fathers’ employment status, child behaviour difficulties, parenting stress arising from the child’s behaviours and needs as well as their own needs, and low parenting satisfaction. The model also accounting for mothers’ mental health symptoms remained significant, R2 = 0.41, Adj R2 = 0.37, F14,219 = 10.06, P < 0.001; however, mothers’ depressive, anxiety and stress symptoms did not provide a significant amount of unique predictive variability in the model.
Discussion This study is one of the first to examine the mental health of fathers of children with disabilities in Australia. While the majority of fathers reported good mental health within the normal range on the DASS sub-scales, the proportion of fathers who reported severe to extremely severe levels of depression, anxiety and stress was 6–8%. Overall, fathers in the present study reported significantly higher levels of depression and stress symptoms compared with normative data for the adult Australian population. These findings are generally consistent with estimates for fathers of children with disabilities reported in previous studies in other countries (Olsson & Hwang 2001; Hastings & Brown 2002), highlighting that a small proportion of fathers are at heightened risk of mental health difficulties. The
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Depression Anxiety Stress Child age Child gender† Education‡ Employment§ Neighbourhood disadvantage¶ Non-English speaking Difficult behaviour (DBC) Child behaviour/needs (PHS) Parental needs or characteristics (PHS) Education & child development (PHS) Parental self-efficacy Parenting satisfaction Social support Mothers’ depressive symptoms Mothers’ anxiety symptoms Mothers’ stress symptoms Range Mean Standard deviation –
3
0.23***
−0.18** −0.05 −0.42*** −0.32*** −0.09 −0.05 0.26*** 0.26*** 0.24*** 0.20** 0.23** 0.18** 0–41 0–29 6.95 3.49 7.89 5.24
0.32***
– −0.05 −0.09 −0.03 −0.01 0.06 0.14
Journal of Intellectual Disability Research
doi: 10.1111/jir.12127
1
Risk factors associated with the mental health of fathers of children with an intellectual disability in Australia R. Giallo,1,2,3 M. Seymour,1 J. Matthews,1 S. Gavidia-Payne,2 A. Hudson1 & C. Cameron1 1 Parenting Research Centre, East Melbourne, Vic., Australia 2 RMIT University, Bundoora, Vic., Australia 3 Murdoch Children’s Research Institute, Royal Children’s Hospital, Parkville, Vic., Australia.
Abstract Background Raising a child with a disability places considerable demands and stress on parents, which can contribute to mental health difficulties. Research has primarily focused on mothers’ mental health, and our understanding of the effects on fathers remains limited. The factors that place fathers at increased risk of mental health difficulties are also poorly understood. This study aimed to redress these gaps by reporting on the mental health of a large sample of fathers of children with an intellectual disability (ID) (aged 3–15 years), comparing this to published Australian norms and mothers of children with ID. The second aim of the study was to explore risk factors associated with fathers’ mental health. Method The data for this study come from 315 Australian fathers of children (aged 3–15 years) with ID, who participated in the large-scale evaluation of the Signposts for building better behaviour programme. Fathers completed a range of self-report questionnaires at baseline including the Depression Anxiety Stress Scale (DASS).
Correspondence: Ms Monique Seymour, Parenting Research Centre, 5/232 Victoria Parade, East Melbourne, Vic. 3002, Australia (e-mail: [email protected]).
Results Fathers in the present sample reported significantly more symptoms of depression and stress than the Australian normative data, with approximately 6–8% reporting symptoms in the severe to extremely severe range. The strongest predictors of fathers’ mental health difficulties were children’s behaviour problems, daily stress arising from fathers’ own needs and children’s care needs, and low parenting satisfaction. Socio-economic factors did not predict mental health difficulties. Conclusion This study is among one of the first to report the mental health of fathers of children with a disability in Australia. Findings highlight that some fathers of children with ID are at heightened risk of experiencing mental health difficulties, underscoring the importance of the provision of information and interventions to promote their mental health. Keywords children with disabilities, father mental health, risk and protective factors, socio-ecological
Parents raising children with disabilities experience considerable caregiving demands and stressful life circumstances that can contribute to mental health difficulties. Research to date however, has primarily focused on mothers (Resch et al. 2012; Seymour et al. 2013b) and our understanding of the extent to
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 2 R. Giallo et al. • Mental health of fathers
which fathers experience mental health difficulties such as depression, stress and anxiety, remains limited. Similarly, studies seeking to understand the factors associated with fathers’ mental health difficulties are scarce (Olsson & Hwang 2001; Hastings & Brown 2002; Hastings 2003; Herring et al. 2006; Gerstein et al. 2009). This might be in part due to the focus on mothers as they are often the primary caregivers of children with disabilities (Hastings 2003; Olsson & Hwang 2006), and the difficulties associated with recruiting fathers in research (Costigan & Cox 2001; Sherr et al. 2006). We had an opportunity to redress these gaps in our knowledge by drawing upon data from a large sample of over 300 Australian fathers of children with intellectual disabilities (ID; aged 3–15 years) who participated in a large-scale evaluation of an intervention to help families manage their children’s behavioural difficulties.
The mental health of fathers of children with disabilities Studies reporting on the mental health difficulties of fathers of children with disabilities are scarce. Of those available, estimates range between 12% and 18% for depressive symptoms (Olsson & Hwang 2001; Hastings & Brown 2002) and 25% for symptoms of anxiety (Olsson & Hwang 2001). For example, it was reported that 18% of 167 fathers of children with ID (aged 1–16 years) in Sweden had clinically significant depressive symptoms on the Beck Depression Inventory (Olsson & Hwang 2001), and 12% of 56 fathers of young children with developmental disabilities (aged 2–6 years) had clinically significant depressive symptoms on the Center for Epidemiologic Studies-Depression Scale (Bristol et al. 1988). In another study of 20 fathers of primary school aged children with an autism spectrum disorder (ASD) in the UK, 25% and 15% of fathers had scores in the at risk or clinical ranges for anxiety and depression, respectively, on the Hospital Anxiety and Depression Scale (Hastings & Brown 2002). Despite these findings, other studies have reported no differences in depressive symptoms between fathers of children with disabilities (i.e. ASD and Down syndrome) and without disabilities (Fisman et al. 1989; Dumas et al. 1991).
The variation in estimates and outcomes reported across studies is likely due to methodological differences associated with the measures used to assess mental health (i.e. Beck Depression Inventory, Hospital Anxiety and Depression Scale), the nature of the samples (i.e. parents of children with ASD, ID or Down syndrome; variable age ranges), and whether fathers’ mental health has been compared with normative data or a comparison sample of fathers with typically developing children. Furthermore, the existing literature is limited on several counts. First, fathers are typically embedded within studies primarily about mothers (e.g. Resch et al. 2012; Feldman et al. 2007). While this is often due to difficulties in recruiting fathers, these studies often lack sufficient numbers to make meaningful comparisons to mothers or to fathers of typically developing children, limiting generalisability of findings. Second, there is an almost exclusive focus on depressive symptoms despite research with fathers of typically developing children indicating that they tend to underreport the extent to which they experience distress when asked specifically about depression compared with other symptoms such as stress or fatigue (Gay et al. 2004; Giallo et al. 2012a; Seymour et al. 2013a). Focusing on a broader range of distress symptoms may provide a more accurate assessment of the extent to which fathers experience mental health difficulties. Finally, another limitation pertains to the lack of Australian-based studies. While there are social, economic and political similarities between Australia and other developed countries (i.e. the UK, Canada, the USA, Sweden), previous studies are based on small, convenient and relatively homogenous samples. These findings may not be generalisable to fathers of children with disabilities in Australia. To date, one large population-based study of 4983 Australian families reported no difference in the rates of psychological distress as assessed by the Kessler-6 among fathers of children (aged 5 years) at risk of disability (n = 297) and fathers of children at no risk (Emerson & Llewellyn 2008). It is important to note that these findings were based on children with a probable diagnosis of disability or chronic health condition based upon a range of physical, behavioural and cognitive functioning outcomes. It is likely that some children
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 3 R. Giallo et al. • Mental health of fathers
were misclassified, and the sample likely reflects a very diverse and heterogeneous group of children and their families. We had an opportunity to redress these limitations using data from over 300 fathers of children with ID (aged 3–15 years) who participated in a large-scale evaluation of an intervention to help parents manage their children’s behavioural difficulties. The Depression Anxiety Stress Scale (DASS; Lovibond & Lovibond 1995), a well-established measure with clinical cuts offs and Australian-based normative data, was used to assess mothers’ and fathers’ depressive, anxiety and stress symptoms. Therefore, the first aim of the study was to assess the extent to which fathers reported depressive, anxiety and stress symptoms in the established clinically elevated ranges of the DASS, and how this compared with Australian normative data (Crawford et al. 2011).
Comparisons between mothers’ and fathers’ mental health Although the prevalence of mental health problems among fathers of children with disabilities remains unclear, studies consistently report that fathers tend to experience fewer mental health difficulties than mothers (Olsson & Hwang 2001; Mugno et al. 2007; Emerson & Llewellyn 2008; Olsson et al. 2008). For example, Gray & Holden (1992) found that Australian mothers of children with an ASD reported significantly higher levels of depression and anxiety than fathers. Similarly, in Vesson’s (1999) Northern European study of 151 mothers and 57 fathers of children with disabilities, mothers reported feeling more worried, sad, tired, helpless, depressed and nervous than fathers. These findings suggest that fathers may adjust better to, or be less affected by, disability in the family than mothers. It has been proposed that mothers are at greater risk of mental health difficulties because they take on more domestic and caregiving responsibilities than fathers (Olsson & Hwang 2006). Fathers, on the other hand, are more likely to be employed, and may be buffered by the benefits of participation in paid employment (Olsson & Hwang 2006). Although this is yet to be substantiated by research, it does suggest that mothers’ and fathers’ mental
health may be influenced by different factors. Some evidence for this comes from Olsson & Hwang (2008). They found that negative perceptions of the impact of disability on the family, and low satisfaction in participation in work, leisure and social activities predicted depressive symptoms for mothers; while the only factor predicting fathers’ depressive symptoms was their self-rated quality of health. It is evident that the context and experience of caregiving, parenting, family roles and responsibilities may differ considerably for mothers and fathers. It stands to reason that there may be a different set of risk factors for mothers’ and fathers’ mental health difficulties, yet there is little research in this area.
Factors associated with fathers’ mental health Studies identifying the risk factors associated with fathers’ mental health are scarce, and those available are typically embedded in research about mothers, examine a narrow range of influences, and fail to account for the impact of socio-economic disadvantage often experienced by families of children with disabilities (Emerson et al. 2006; Emerson & Llewellyn 2008). A review of the literature within the context of these key limitations will now be discussed. Compared with mothers of children with disabilities (Hastings & Brown 2002; Emerson et al. 2006; Feldman et al. 2007; Gerstein et al. 2009; Giallo et al. 2011; Seymour et al. 2013b), few studies have identified risk factors specifically for fathers’ mental health. Of the research available, child behaviour difficulties and partner mental health have received the most attention and produced mixed findings. For example, some studies have reported associations between child behaviour problems and fathers’ overall health (Herring et al. 2006) and perceptions of parenting stress (Dumas et al. 1991; Hastings 2003), but not anxiety and depressive symptoms. Several studies have found no association between mothers’ mental health and fathers’ stress (Herring et al. 2006), depressive or anxiety symptoms (Hastings & Brown 2002; Hastings 2003), while others have. For instance, Olsson & Hwang (2001)
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 4 R. Giallo et al. • Mental health of fathers
found a small to moderate correlation between the depressive symptoms of 216 mothers and fathers of children (aged 7–10 years) with ASD and/or ID. Similarly, a longitudinal study found that the course of fathers’ parenting stress when their children with an ID were 36 to 60 months was predicted by mother’s mental health (Gerstein et al. 2009). The daily demands and challenges arising from parenting and parent-child interactions that may have an impact on fathers’ mental health have also received some limited attention. For example, Gavidia-Payne et al. (1997) found that daily stress associated with children’s behaviour, communication and caregiving needs, as well as parents’ own needs (i.e. having time for themselves or others in the family) were associated with increased anxiety, stress and depressive symptoms among 96 mothers and 12 fathers of children (aged 4–19 years) with disabilities. Parenting stress and child behaviour difficulties have the potential to undermine parental self-efficacy (PSE), which refers to parents’ sense of competence in their parenting role (Coleman & Karraker 1997). Hastings & Brown (2002) found a complex relationship between children’s behaviour problems, PSE and mental health for 26 mothers and 20 fathers of preschool aged children with ASD. They found that fathers of children with behaviour difficulties were more anxious when they had a lowered sense of PSE. This is consistent with research with fathers of typically developing children in the postnatal period, whereby low PSE has been associated with increased psychological distress (Giallo et al. 2013). While relationships between child behaviour difficulties, parenting stress, PSE and fathers’ mental health are likely to be complex and bi-directional, this small body of research highlights that they are also key factors associated with fathers’ mental health that warrant further investigation in larger samples of fathers. Finally, a key limitation of previous research is the failure to consider the influence of socioeconomic disadvantage on fathers’ mental health. This is important because it is well established that families of children with disabilities are more likely than other families to experience significant economic hardships (Emerson 2003; Emerson et al. 2006; Emerson & Llewellyn 2008). Such differences in economic and social resources have been shown to partially account for differences between families
of children with and without disabilities on a range of outcomes (Emerson & Llewellyn 2008; Emerson et al. 2010). For example, limited participation in paid employment (Olsson & Hwang 2006), poverty and socio-economic disadvantage (Emerson et al. 2006; Emerson & Llewellyn 2008) have been associated with mental health difficulties among mothers. These factors have received little attention with fathers.
Aims of the study In summary, studies exploring the extent to which fathers of children with disabilities experience mental health difficulties are scarce, and the factors that heighten their risk remains poorly understood. The current study sought to redress these significant gaps in our knowledge by reporting on the mental health of a large sample of fathers of children with ID (aged 3–15 years) who participated in an evaluation of an Australian intervention to help families manage their children’s behavioural difficulties. The first aim of the study was to report on the extent to which fathers reported mental health difficulties, and compare this to published Australian norms (Crawford et al. 2011) and mothers who also participated in the intervention study. It was hypothesised that fathers would report fewer symptoms of depression, anxiety and stress than mothers, but significantly more symptoms than adults in the general population. The second aim of the study was to explore a broader range of risk factors associated with fathers’ mental health than previously available. Socioecological models of health (Brofenbrenner 1979; McLaren & Hawe 2005; World Health Organisation 2013) provide useful frameworks to advance this research. These theories focus on identifying factors at different levels of an individual’s immediate and broader social environment. Examples of risk and protective factors at the level of the individual can include age, personality, values, beliefs and coping skills. At the interpersonal level, quality of relationships, social support and stress in the family environment are key factors, while socio-economic influences, employment and availability of services are potential risk factors in the broader social environment. This model has been widely used in many
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 5 R. Giallo et al. • Mental health of fathers
fields (e.g. health behaviours and promotion, violence prevention, weight management and nutrition). It has also been used as a guiding framework to identify the factors associated with men’s health (e.g. prostate examination; Woods et al. 2006), as well as postnatal mental health difficulties among Australian fathers of typically developing children (Giallo et al. 2012b, 2013; Seymour et al. 2013a). Drawing upon this model, we took advantage of the available baseline data from the intervention study to explore risk factors for fathers’ mental health difficulties at the individual level (i.e. parenting efficacy and satisfaction) and interpersonal level (i.e. child behaviour difficulties, parenting stress, partner mental health, social support). Given previous research indicating that socio-economic disadvantage experienced by families of children with disabilities partially accounts for mothers’ mental health difficulties (Emerson & Llewellyn 2008; Emerson et al. 2010), first we sought to examine the unique contribution of indicators of socio-economic disadvantage (i.e. socio-economic factors such as employment, education, neighbourhood disadvantage) to fathers’ mental health. Then we examined the contribution of the individual and interpersonal level factors over and above socio-economic disadvantage factors. We hypothesised that socioeconomic disadvantage would be associated with fathers’ mental health difficulties, and that individual and interpersonal level factors would predict fathers’ mental health difficulties over and above the contribution of the socio-economic factors.
Method Study design The data for this study forms part of a large-scale evaluation of the Signposts for building better behaviour programme for parents of children (aged 3–15 years) with ID (Hudson et al. 2003, 2008). The aim of this programme was to teach parents skills to manage any difficult behaviours of their child before those behaviours escalate to a level that requires intensive intervention by specialists. Both mothers and fathers from all families were invited to participate in the programme. Details of the study design and sampling information are detailed elsewhere (Hudson et al. 2003, 2008). Ethics approval for the
evaluation study was granted by the Department of Human Services Ethics Committee, Victoria, Australia. Dissemination of the programme was jointly funded by the Australian Commonwealth Government’s former Department of Families, Community Services and Indigenous Affairs (now known as the Department of Social Services) and the Victorian State Government’s Department of Human Services. The current sample consisted of fathers for who self-report data on the variables of interest at pre-intervention was available.
Measures Demographic and family background information Information about the parent’s age, gender, language spoken, level of education and employment status were collected. Additionally, information pertaining to the age and gender of the child with an ID, specific disability diagnosis, severity of disability and educational placement was obtained. Families’ socio-economic status was assessed using a measure of neighbourhood disadvantage, the Index of Relative Socio-economic Disadvantage from the Australian Bureau of Statistics, Socio-economic Indexes for Areas (SEIFA; Trewin 2003). This index is derived from 2001 population census data pertaining to income, educational attainment and employment. The SEIFA is constructed so that the mean score is 1000, with scores below 1000 indicative of being less advantaged (Australian Bureau of Statistics 2006). The Depression, Anxiety and Stress Scale (DASS; Lovibond & Lovibond 1995) The DASS was developed and normed in Australia, and assesses the negative emotional states of depression, anxiety and stress over the past week. Both mothers and fathers participating in the Signposts evaluation were asked to complete this measure. There are 42 items that are rated on a four-point scale ranging from 0 = ‘Did not apply to me at all’ to 3 = ‘Applied to me very much, or most of the time’. Although the DASS is not a diagnostic tool for mood disorders, it does provide clinical cut-offs to indicate whether a respondent is experiencing significant levels of distress. Internal consistency for the current sample of both mothers and fathers was Cronbach’s α = 0.97.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 6 R. Giallo et al. • Mental health of fathers
The Developmental Behaviour Checklist (DBC; Einfeld & Tonge 2002) The DBC is a 96-item scale designed to assess difficult behaviour of children with disabilities, and was also developed in Australia. Items are rated on a three-point scale, where 0 = ‘not true as far as you know’ to 2 = ‘very true or often true’. It consists of five sub-scales: Disruptive, Self-Absorbed, Communication Disturbance, Anxiety and Social Relationships. It also gives a Total Problems score, which was used in this study. Although normative data or clinical cut-offs have not been established for this measure, data from a large sample of 454 Australia children with mild to profound levels of ID have been published (Einfeld & Tonge 1996a,b). The mean scores on the Total Problems scales for children with mild, moderate, severe and profound ID were reported to be 43.9, 42.5, 43.1 and 26.0 respectively. Internal consistency for the current sample of fathers was Cronbach’s α = 0.95. Parenting Hassles Scale (PHS; Gavidia-Payne et al. 1997) The PHS is a 50-item scale specifically designed to assess potential stressors experienced by families of children with disabilities on a daily basis, as distinct from stress symptoms measured by the DASS. It consists of 12 sub-scales, however, given that the Signposts programme focused on assisting parents to manage their children’s behaviour, only the Child Behaviour/Needs, Parent Needs/Characteristics, and Education and Child Development sub-scales were included in the evaluation study. Each item is also rated on a five-point scale ranging from 1 = ‘no hassle’ to 5 = ‘major hassle’, with higher scores indicating a greater impact of hassles. The internal consistency for the current sample of fathers was Cronbach’s α = 0.95. Parenting Sense of Competence Scale (PSOC; Johnston & Mash 1989) The PSOC is a 16-item scale used to measure parents’ views of their competence as parents on two scales: the Satisfaction sub-scale, measuring the extent to which parents are satisfied with their parenting role; and the Efficacy sub-scale, measuring the extent to which parents feel they are managing
the role of being a parent. The items were rated on a six-point scale, ranging from 1 = ‘strongly agree’ to 6 = ‘strongly disagree’. Satisfaction and Efficacy items are added together to obtain a total score (range 16–96), where higher scores indicate greater perceived competence. The internal consistency for the current sample of fathers was Cronbach’s α = 0.70. Sources of Support Sources of Support was designed specifically for this study to assess a comprehensive range of sources of support for parents of children with disabilities. It consists of 20-items asking whether parents use these sources of support and how helpful the support received is. Example sources of support include from parents, friends, co-workers, parent support group, early intervention teacher, school or childcare educator. Items are rated on a six-point scale ranging from 0 = ‘not available’ to 5 = ‘extremely helpful’. The internal consistency for the current sample of fathers was Cronbach’s α = 0.75.
Results Sample characteristics Of the 1551 families who participated in the Signposts evaluation, 1323 were mothers and 315 were fathers. Demographic information for fathers is provided in Table 1. The majority of fathers were English speaking, employed and had an educational attainment of high school or above. Of the fathers participating in this study, 252 completed the Signposts programme with their partner. There were significantly more unemployed fathers and fathers from non-English-speaking backgrounds who participated in the programme by themselves compared with those who participated with their partner (P < 0.05). Demographic information for the children with a disability is presented in Table 2. The majority of the children were male, had a diagnosis of ASD and were attending a specialist or early intervention programme. The mean score for the DBC Total Difficulties scale was 49.48 (25.06), which is consistent with previous studies reporting mean scores for
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 7 R. Giallo et al. • Mental health of fathers
Table 2 Child with disability characteristics (N = 315)
Table 1 Father characteristics (N = 315)
Variable Language spoken English only Employment status Employed Not in paid employment Not reported Highest level of education completed Some high school Completed high school TAFE, Trade Certificate, Diploma Tertiary (degree, post-graduate degree) Not reported SEIFA – Index of Relative Socio-economic Disadvantage (M, SD)
n (%)
286 (90.8%) 261 (82.9%) 44 (14.0%) 10 (3.2%) 97 (30.8%) 26 (8.3%) 39 (12.4%) 58 (18.4%) 95 (30.1%) 1009.68 (58.70)
M, mean; TAFE, technical and further education; SD, standard deviation; SEIFA, Socio-Economic Indexes for Areas.
Australian children with mild to profound ID ranging from 26.0 to 43.9 (Einfeld & Tonge 1996a,b).
The extent to which fathers’ report depressive, anxiety and stress symptoms Descriptive statistics for the DASS sub-scales are presented in Table 3, and Table 4 presents the percentage of fathers in the Normal, Mild, Moderate, Severe and Extremely Severe ranges on each of the DASS sub-scales. While the majority of fathers scored in the Normal range on all sub-scales, a small proportion reported symptoms of depression (7.9%), anxiety (6.0%) and stress (7.9%) in the Severe to Extremely Severe ranges. Comparisons between fathers’ and mothers’ distress symptoms were made using paired samples t-tests for a subset of the sample in which both father and mother data from the same family were available (n = 220). Mothers reported significantly higher depressive, anxiety and stress symptoms than fathers (see Table 5), and this was associated with moderate to large effect sizes for Cohen’s d; where 0.2, 0.5 and 0.8 are small, medium and large effect sizes respectively. Finally, one-sample t-tests were used to compare fathers’ DASS sub-scale scores with published Australian norms (Crawford et al. 2011). The normative
Variable
n (%)
Age of child with disability (years) (M, SD) Gender of child with disability Male Female Not reported Intellectual disability plus other condition Autism spectrum disorder Vision impairment Hearing impairment Down syndrome Cerebral palsy Epilepsy or other seizure disorder Psychiatric disorder Attention deficit and hyperactivity disorder Acquired brain injury Health problems (i.e. asthma, food allergies) Severity of intellectual disability – formally assessed Mild Moderate Severe Profound Not reported How well child communicates with speech Very well Well Reasonably well Poorly Very poorly Not reported How well child understands speech Very well Well Reasonably well Poorly Very poorly Not reported Educational placement Specialist and early intervention programme Regular preschool/childcare Special school Special development school Home tutoring Regular school with integration aide Regular school without integration aide Other (i.e. speech therapy, family day care)
7.83 (5.52) 220 (69.8%) 89 (28.3%) 6 (1.9%) 141 (44.8%) 19 (6.0%) 17 (5.4%) 15 (4.8%) 8 (2.5%) 26 (8.3%) 6 (1.9%) 46 (14.6%) 10 (3.2%) 24 (7.6%)
54 (17.1%) 23 (7.3%) 7 (2.2%) 1 (0.3%) 230 (73.1%) 46 (14.6%) 66 (21.0%) 87 (27.6%) 46 (14.6%) 31 (9.8%) 39 (12.7%) 51 (16.2%) 81 (25.7%) 100 (31.8%) 20 (6.3%) 8 (2.5%) 55 (17.4%) 114 (36.2%) 97 (30.8%) 64 (20.3%) 44 (14.0%) 5 (1.6%) 41 (13.0%) 29 (9.2%) 24 (7.6%)
M, mean; SD, standard deviation.
sample was comprised of 497 men and women from the general adult population, with distributions by age and gender consistent with Australian census data. Men with an education level of less than Year
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 8 R. Giallo et al. • Mental health of fathers
Table 3 Descriptive statistics for the current sample and normative data on the Depression Anxiety Stress Scale (DASS)
Current sample (n = 315)
Depression Anxiety Stress
Normative data (n = 497)
M
SD
M
SD
t
Cohen’s d (95% CI)
6.95 3.49 11.00
7.89 5.24 8.24
5.02 3.36 8.10
7.54 5.07 8.40
4.35* 0.42 6.24*
0.25 (0.11, 0.39) 0.03 (−0.12, 0.17) 0.35 (0.21, 0.49)
* P < 0.001. CI, confidence interval; M, mean; SD, standard deviation.
Table 4 Percentage of fathers in the normal and clinical ranges on the Depression Anxiety Stress Scale (DASS) (N = 315)
Range
Depression n (%)
Anxiety n (%)
Stress n (%)
Normal Mild Moderate Severe Extremely severe
228 (72.4) 33 (10.5) 32 (10.2) 14 (4.4) 8 (2.5)
261 (82.9) 22 (7.0) 13 (4.1) 11 (3.5) 8 (2.5)
210 (66.9) 39 (12.4) 41 (13.0) 11 (3.5) 14 (4.4)
12 were slightly under-represented. Given that no gender differences were found between men and women, it was recommended by the authors that combined normative data for men and women is used. Fathers of children with disabilities in the present sample reported significantly higher depression and stress than the normative sample, and this was associated with small to moderate effect sizes for Cohen’s d; where 0.2, 0.5 and 0.8 are small, medium and large effect sizes respectively.
Factors associated with fathers’ mental health difficulties First, bivariate correlation analyses were conducted to examine the relationships between the symptoms of depression, stress and anxiety, and the predictor variables (see Table 6). Next, a series of hierarchical regression analyses were conducted separately for fathers’ depressive, anxiety and stress symptoms. In the first block, indicators of socio-economic status
were entered. This included fathers’ employment status, educational attainment, neighbourhood disadvantage and whether they were from a nonEnglish-speaking background. In the second block, the individual and interpersonal factors (excluding mothers’ mental health difficulties) were entered to assess their contribution over and above socioeconomic factors. Finally, given that 95 cases did not have mental health data for mothers, the regression analyses were conducted with a small subset of cases for which mothers’ mental health data was available. Depressive symptoms At the first step of the hierarchical regression analysis, the socio-economic variables were entered, and these did not significantly predict fathers’ depressive symptoms (R2 = 0.02, Adj R2 = 0.01, F4,310 = 1.39, P = 0.23). In the second step, the individual and interpersonal variables were entered, resulting in a significant change in R2 (ΔR2 = 0.32, F7,303 = 21.21, P < 0.001), with the model now significantly accounting for 32% of the variance in fathers’ depressive symptoms (R2 = 0.34, Adj R2 = 0.32, F11,303 = 14.24, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were child behaviour difficulties, parenting stress arising from the child’s behaviours and needs, as well as the fathers’ own needs, and low parenting satisfaction. The pattern of results was similar when mothers’ mental health symptoms were entered into the model (R2 = 0.34, Adj R2 = 0.30, F14,219 = 7.56, P < 0.001); however, mothers’ depressive, anxiety
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 9 R. Giallo et al. • Mental health of fathers
Fathers
Depression Anxiety Stress
Mothers
M
SD
M
SD
t
Cohen’s d (95% CI)
6.52 3.19 10.84
7.85 5.16 8.20
9.70 5.64 15.40
9.82 7.60 9.61
−4.34* −4.38* −6.27*
−0.36 (−0.54, −0.19) −0.39 (−0.56, −0.22) −0.52 (−0.69, −0.34)
Table 5 Descriptive statistics for fathers and mothers in the current sample on the Depression Anxiety Stress Scale (DASS) (N = 220)
* P < 0.001. CI, confidence interval; M, mean; SD, standard deviation.
and stress symptoms did not provide a significant amount of unique predictive variability in the model. Anxiety symptoms At the first step of the regression analysis with the socio-economic variables entered, the model did not significantly predict fathers’ anxiety symptoms (R2 = 0.03, Adj R2 = 0.02, F4,314 = 2.21, P = 0.067. In the second step, the individual and interpersonal variables were entered, resulting in a significant change in R2 (ΔR2 = 0.22, F7,303 = 12.62, P < 0.001), with the model now significantly accounting for 22% of the variance in fathers’ anxiety symptoms (R2 = 0.25, Adj R2 = 0.22, F11,303 = 9.05, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were child behaviour difficulties, parenting stress arising from the child’s behaviours and fathers’ own needs, and low parenting satisfaction. The model including mothers’ mental health symptoms remained significant, R2 = 0.25, Adj R2 = 0.20, F14,219 = 4.80, P < 0.001, although parenting satisfaction and parenting stress related to their child’s needs were no longer significant predictors. Mothers’ depressive symptoms provided a significant amount of unique variability in this model (β = 0.25, t = 2.50, P = 0.013). Stress symptoms At the first step of the analyses with the socioeconomic variables entered, the model did not significantly predict fathers’ stress symptoms (R2 = 0.01, Adj R2 = 0.01, F4,314 = 0.74, P = 0.563. In the next step with the individual and interpersonal variables entered, there was a significant
change in R2 (ΔR2 = 0.38, F7,303 = 26.38, P < 0.001), with the model now significantly accounting for 36% of the variance in fathers’ anxiety symptoms (R2 = 0.39, Adj R2 = 0.36, F11,303 = 17.21, P < 0.001). Table 7 shows that the predictor variables providing a significant amount of unique predictive variability in the model were fathers’ employment status, child behaviour difficulties, parenting stress arising from the child’s behaviours and needs as well as their own needs, and low parenting satisfaction. The model also accounting for mothers’ mental health symptoms remained significant, R2 = 0.41, Adj R2 = 0.37, F14,219 = 10.06, P < 0.001; however, mothers’ depressive, anxiety and stress symptoms did not provide a significant amount of unique predictive variability in the model.
Discussion This study is one of the first to examine the mental health of fathers of children with disabilities in Australia. While the majority of fathers reported good mental health within the normal range on the DASS sub-scales, the proportion of fathers who reported severe to extremely severe levels of depression, anxiety and stress was 6–8%. Overall, fathers in the present study reported significantly higher levels of depression and stress symptoms compared with normative data for the adult Australian population. These findings are generally consistent with estimates for fathers of children with disabilities reported in previous studies in other countries (Olsson & Hwang 2001; Hastings & Brown 2002), highlighting that a small proportion of fathers are at heightened risk of mental health difficulties. The
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Depression Anxiety Stress Child age Child gender† Education‡ Employment§ Neighbourhood disadvantage¶ Non-English speaking Difficult behaviour (DBC) Child behaviour/needs (PHS) Parental needs or characteristics (PHS) Education & child development (PHS) Parental self-efficacy Parenting satisfaction Social support Mothers’ depressive symptoms Mothers’ anxiety symptoms Mothers’ stress symptoms Range Mean Standard deviation –
3
0.23***
−0.18** −0.05 −0.42*** −0.32*** −0.09 −0.05 0.26*** 0.26*** 0.24*** 0.20** 0.23** 0.18** 0–41 0–29 6.95 3.49 7.89 5.24
0.32***
– −0.05 −0.09 −0.03 −0.01 0.06 0.14
