Behavior problems in children with mild intellectual disabilities:An initial step towards prevention

Petri J.C.M. Embregts a,b,c,*, Marleen Grimbel du Bois d, Nathalie Graef d

a Tilburg University, Clinical Psychology, Tilburg, The Netherlandsb HAN University of Applied Sciences, Faculty of Health and Social Studies, Nijmegen, The Netherlandsc SZ Dichterbij, Oostrum, The Netherlandsd OPSY, GGz Eindhoven, Eindhoven, The Netherlands

Children with mild intellectual disabilities (MID) have a greater risk for developing psychopathology and behaviorproblems compared to children without MID (Dekker & Koot, 2003). According to this Dutch study of Dekker and Koot,executed by 474 children, 38.6% of the participants had a DSM-IV diagnose, compared to 21.5% of the normal population.Beside the higher levels of psychopathology, behavior problems are also more prevalent in children with MID. A study ofEinfeld and Tonge (1996) showed that well over 40% of the children with MID have serious emotional and behavioralproblems. Moreover, the majority of children and adolescents with MID referred for mental health care, suffer fromdisruptive behavior problems and/or aggressiveness, oppositionality, defiance and conduct disorders (Wallander, Dekker &Koot, 2003). The prognosis for these children and adolescents is unfavorable: behavior problems co-occur with internalizingand social problems, minimize opportunities in society and predict a host of unfavorable adult outcomes. The burden of the

Research in Developmental Disabilities 31 (2010) 1398–1403

A R T I C L E I N F O

Article history:

Received 19 June 2010

Accepted 22 June 2010

Keywords:

Prevention

Mild intellectual disability

Behavior problems

Parental stress

A B S T R A C T

To develop prevention activities, an analysis is conducted of child and parent

characteristics that occur significantly more often among children with a mild intellectual

disability and behavior problems than among children with a mild intellectual disability

and no behavior problems and their families. The sample consisted of 45 children attenting

schools for special education. Data were collected from the children, their parents, and

their teachers. The instruments used are the Dutch version of the Parenting Stress Index,

the Nijmegen Child-Rearing Situation Questionnaire and the Strenghts and Difficulties

Questionnaire for parents, teachers and children. On the basis of the results of parents on

the Strenghts and Difficulties Questionnaire, the research sample was divided into one

group of children with behavior problems and one group without behavior problems.

Parents of the children with behavior problems were found to feel less competent, more

socially isolated, less satisfied about their relationship with their partner, and indicate

more negative life occurrences than the parents of the children without behavior

problems. Characteristics in the area of adaptability, mood, distractibility/hyperactivity,

demandingness, reinforcement of parents, and acceptability were found to contribute to

the total stress in the child–parent relationship for those children with behavior problems

and their parents. On the basis of these results prevention activities will be developed and

tested on their effectiveness.

� 2010 Elsevier Ltd. All rights reserved.

* Corresponding author at: Department of Clinical Psychology, Tilburg University, P.O. Box 90153, 5000 LE Tilburg, The Netherlands. Tel.: +31 134664126;

fax: +31 134662370.

E-mail address: p.j.c.m.embregts@uvt.nl (Petri J.C.M. Embregts).

Contents lists available at ScienceDirect

Research in Developmental Disabilities

0891-4222/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.ridd.2010.06.020

behavior problems on families and friends is such that they have been reported to consider these problems the mostimportant issue for treatment.

When children with MID and/or their parents need help, the Dutch government strives to see that support, guidance andtreatment is provided within the immediate environment and family system of the child whenever is possible. One of thefirst steps in this process is offering prevention programs for the children with ID and/or their parents. Nevertheless, withinthe care for people with ID prevention programs are still in its infancy, as well as incorporating family processes in theseprograms, although several studies highlight the impact of behavior problems on families (e.g., Femmie, Bakermans-Kranenburg & IJzendoorn, 2005). Meta-analyses within mental health care conclusively show that prevention programsresulted in a reduction of 25% in symptoms of depression, fear, stress and behavioral problems. This effect is comparable tothe decrease achieved by psychological, educational and behavioral treatment (Jane-Llopis, 2002; Jane-Llopis, Hosman,Jenkins & Anderson, 2003). In addition, it appeared that effects were stable until at least 1-year follow-up and thatinterventions were twice as effective if performed by specialized professionals compared to exclusively offered by non-professionals (Jane-Llopis, 2002).

Prevention programs can be used autonomously or be a part of a extensive outpatient or residential treatment. Becauseoutpatient treatment is increasingly undertaken to prevent the need for residential treatment, Embregts (2009) exploreddifferences in child and family characteristics of children with a MID who were placed in residential treatment followingoutpatient treatment and children with a MID who only followed the outpatient treatment. The results showed that there weresignificantly more children placed in residential treatment having educationally incapable parents, parents with alcohol/drugproblems and/or psychiatric problems than in the group of children without residential treatment following outpatienttreatment. There is also a strong evidence base for positive associations between the frequency and/or severity of behaviorproblems in children with ID and parental psychological difficulties (e.g., Beck, Hastings, Daley, & Stevenson, 2004; Hastings,2003). Parents of children with ID report more parenting stress and mental health problems such as depression than parents ofchildren without disabilities (e.g., Emerson, 2003). High levels of stress influence whether families remain in treatment and, forthose who do remain, the extent to which children improve and maintain treatment gains over time (e.g., Kazdin, 1995).

The aim of the present study therefore was to identify variables (i.e., child characteristics, parent characteristics, andsituations that are directly related to the role of being a parent) related to children with MID and behavior problems. TheDutch version of the Parenting Stress Index and the Nijmegen Child-Rearing Situation Questionnaire were administered tothe parents of 45 children with MID for this purpose. The results were judged to be of use for the design of an effectivepreventive program aimed at the occurrence of behavior problems of children with MID.

1. Method

1.1. Participants

A total of 45 children with MID met the following criteria: (1) MID or borderline functioning; (2) age between 12 and 16years and (3) visiting a school for special education for at least three months. They participated in this study based oninformed consent: their parents were willing to fulfill the questionnaires. The average age of the children was 13.5 years(range 12–16); 27 of the children were males and 18 were females. The intellectual disabilities were described as borderlinein 2 cases, mild in 42 cases and unknown in 1 case. In 71.1% (n = 32) of the cases the family situation consisted of 2 parents. Atotal of 19 children (42.2%) had additional psychiatric disorders according to the SDQ fulfilled by their parents, in 40.0%(n = 18) of the cases according to the teachers version, and only in 31.1% (n = 14) according to the children themselves.

1.2. Instruments

1.2.1. The Strengths and Difficulties Questionnaire (SDQ)

The SDQ (Goodman, 1997) is a 25-item questionnaire for assessing the psychosocial adjustment of children andadolescents with three response categories (not true, somewhat true, certainly true), and 5 scales: (1) emotional symptoms,(2) conduct problems, (3) hyperactivity, (4) peer problems, and (5) prosocial behavior. A high score on the scale prosocialbehavior is a reflection of strengths, while high scores on the other scales show difficulties. A total problem score is obtainedby adding up scales 1, 2, 3 and 4. The sum of these scores may vary from 0 to 40. A score equal to or higher than 17 indicatespsychopathology; a score between 14 and 16 implies borderline.

The SDQ can be filled in by parents, teachers and children themselves. Several studies (including the Dutch version) haveevaluated the internal consistency of the SDQ total difficulties scales and subscales in community populations (Goodman,2001; Goodman & Scott, 1999; Muris, Meesters, Van den Berg, 2003; Widenfelt, Goedhart, Treffers, & Goodmann, 2003). Forthe total difficulties scale, all studies reported Cronbach a� .70 for all types of participating informants. For the subscalesemotional symptoms, conduct problems, hyperactivity-inattention, peer problems and prosocial behavior, the internalconsistencies ranged from a is .63 to .78, .45 to .77, .66 to .89, .39 to .74 and .57 to .84, respectively.

1.2.2. Parenting Stress Index (PSI)

The parent–child relationship was assessed using a Dutch version (Nijmegen Parental Stress Index; de Brock, Vermulst,Gerris & Abidin, 1992) of the Parenting Stress Index (Abidin, 1983). The PSI was designed to assess the degree of stress relatedto parenting. The Dutch translation of the PSI consists of 123 items divided in two domains, the parent and child domain

P.J.C.M. Embregts et al. / Research in Developmental Disabilities 31 (2010) 1398–1403 1399

(58 and 65 items respectively), and a scale measuring life occurrences (40 items). The parent domain is composed out ofseven dimensions, including depression, attachment, restriction of role, sense of competence, social isolation, relationshipwith spouse, and parent health. The child domain consists of six dimensions: adaptability, acceptability, demandingness,mood, distractibility/hyperactivity, and reinforces parent. The items are scored on a 6-point likert scale ranging from ‘agreecompletely’ to ‘disagree completely’. The scale measuring life occurrences indicates dramatic and stressful circumstancesthat have taken place in the family during the last twelve months, such as increasing number of conflicts with partner orfinancial problems. The PSI yields individual scores for each child and parent dimension and overall scores for the child andparent domains that are obtained by calculating the means of the scores on the respective dimensions. The higher the score,the more stress reported. Both the original PSI and its Dutch version demonstrate a good content validity, and show sufficientfactorial, concurrent, discriminant and construct validity and internal reliability, Cronbach’s alpha for the PSI dimensionsranged between a is .75 and .91 (Abidin, 1983; de Brock et al., 1992).

1.2.3. The Nijmegen Child-Rearing Situation Questionnaire (NCSQ)

The NCSQ (Wels & Robbroeckx, 1996) was developed with the aim of measuring parental cognitive appraisals of the child-rearing situation that are believed to reflect different aspects of stress. The NCSQ considers measures of appraisals on child-rearing in four different sections: (1) subjective parenting stress (46 items), (2) global appraisal of the child-rearing situation(8 items), (3) parental attribution of child-rearing outcomes (34 items) and (4) expectations for help (36 items). In this studyonly section four ‘expectations for help’ was used, because of the unique information not included in the PSI dimensions. Thissection contains items defining different levels of parental need for expectations about help, including the following scales:(1) degree of satisfaction; (2) wish for change; (3) wish for help; (4) internal help expectancy; (5) external help expectancy.The items were scored on a 5-point likert scale ranging from ‘agree completely’ to ‘disagree completely’. Results from theDutch psychometric testing of the NCSQ indicated good psychometric properties (Wels & Robbroeckx, 1996), internalconsistency for the test ranged from a is .68 to .92 and correlation coefficients for test retest ranged from r = .60 to .94.Internal consistency for the section expectations about help was between a is .70 and .86 and the test retest ranged from .64to .92 (Robbroeckx & Wels, 1988).

1.3. Statistical analysis

First the internal consistency of all three scales was measured. Mean scores on the SDQ were used to split op the data file,into children with and without behavior problems. T-tests were executed to assess whether or not their were differencesbetween the two groups, on the PSI and NCSQ.

2. Results

2.1. Parenting Stress Index (PSI)

Good internal consistency was found for the PSI (a = .84), and for the scales ranging from a is .75 to .91.

2.2. Nijmegen Child-Rearing Situation Questionnaire (NCSQ)

Good internal consistencies were found for the scale ‘expectation of help’; Cronbach’s alfa ranges from a is .82 to .96.

2.3. The Strengths and Difficulties Questionnaire (SDQ)

Cronbach’s alfa coefficients for the SDQ scales were computed for all three informants. Good internal consistencies werefound for the teacher SDQ scales, ranging from a is .72 to .85. For the parent SDQ scales, the internal consistencies were lowerthan the teacher scales, ranging from a is .64 to .83. The lowest internal consistencies were found for the self-report SDQscales (range .11–.58).

Mean problem score of the SDQ for children was 11.57 (range 2–25), 12.6 for the teachers (range 1–24) and 13.30 for theparents (range 0–28). As the internal consistencies found for the self-report SDQ scales were low, children were divided ingroups with and without behavior problems based on the SDQ parents’ version.

2.4. Differences between children with and without behavior problems on the PSI and NCSQ

Table 1 shows the scores for children with and children without behavior problems and the p- and t-values on the PSI andNCSQ.

2.5. PSI

Parents of children with behavior problems had a significantly higher score on the following dimensions: sense ofcompetence (t(28.21) =�2.68, p< .05), social isolation (t(22.07) =�2.97, p< .01) and relationship with spouse

P.J.C.M. Embregts et al. / Research in Developmental Disabilities 31 (2010) 1398–14031400

(t(26.60) =�2.89, p< .01) than parents of children without behavior problems. There were no significant differencesbetween parents of children with and without behavior problems with respect to the subscales restriction of role(t(29.16) =�2.01, p = .08), attachment (t(26.81) =�1.62, p = .12), depression (t(29.80) =�1.83, p = .08) and parent health(t(26.43) =�1.94, p = .06). Furthermore, parents of children with behavior problems had a significantly higher mean score onall scales of the child domain (i.e., adaptability (t(39) =�4.48, p< .001), acceptability (t(28.87) =�3.65, p = .001),demandingness (t(39) =�3.52, p = .001), mood (t(27.25) =�4.31, p< .001), distractibility/hyperactivity (t(39) =�4.23,p< .001), and reinforces parent (t(28.33) =�3.88, p< .001)) than parents of children without behavior problems. Parents ofchildren with behavior problems report more negative life experiences (t(28.21) =�2.27, p = .03) for their children thanparents of children without behavior problems; they report a comparable degree of positive life experiences (t(39) = 0.54,p = .59).

2.6. NCSQ

Parents of children with behavior problems had a significantly higher score on the scales internal help expectancy(t(39) =�3.05, p< .01) and external help expectancy (t(39) =�3.21, p< .01). On the scales degree of satisfaction (t(38) = 1.17,p = .25), wish for help (t(23.62) =�1.31, p = .20) and wish for change (t(39) =�1.45, p = .15), there were no significantdifferences between parents of children with and without behavior problems.

3. Discussion

The aim of the present study was to identify variables (i.e., child characteristics, parent characteristics, and situations thatare directly related to the role of being a parent) that are related to children with MID and behavior problems.

Results of the Dutch translation of the PSI and NCSQ show that parents of children with MID and behavior problems feelless competent, are more socially isolated and more dissatisfied about the relation with their partner than parents of childrenwithout behavior problems. Also, qualities and characteristics of children with behavior problems in the areas of adaptation,mood, distractibility, exactingness, positive confirmation and acceptance, according to parents contribute to the overallstress level in the parent–child relationship. Parents of children with behavior problems indicate significantly more negativelife experiences than parents of children without behavior problems. They also are clearly in need of help, both aimed atthemselves (internal) and at others (external). These results show that the relationship, as found in literature, betweendeterminant factors and the risk of behavior problems is also confirmed in our study. Based on our findings, these resultstherefore offer starting points for the development of well-founded prevention interventions.

Table 1

Mean scores for children with and without behavior problems on the PSI and NCSQ (expectations for help), t and p values.

Questionnaire Domain Variables Behavior problems No behavior problems t p

Mean Mean

PSI

Parents domain

Sense of competence 2.28 1.57 �2.68 0.012*

Restriction of role 2.79 1.99 �2.01 0.077

Attachment 2.12 1.63 �1.62 0.117

Depression 2.32 1.73 �1.83 0.077

Parent health 2.41 1.67 �1.94 0.064

Social isolation 2.31 1.40 �2.97 0.007**

Relationship with spouse 2.69 1.62 �2.89 0.007**

Child domain

Adaptability 4.19 2.52 �4.48 0.000***

Mood 3.19 1.88 �4.31 0.000***

Distractability/hyperactivity 3.41 0.47 �4.23 0.000***

Demandingness 3.43 2.16 �3.52 0.001**

Reinforces parent 3.02 1.86 �3.88 0.000***

Acceptability 3.36 2.28 �3.65 0.001**

Life occurences

Negative life experiences 2.26 0.91 �2.27 0.031*

Positive life experiences 0.21 0.32 0.54 0.594

NCSQ

Expectations for help

Degree of satisfaction 3.94 4.14 1.17 0.250

Wish for help 2.19 1.86 �1.31 0.203

Wish for change 2.42 2.10 �1.45 0.154

Internal help expectancy 3.22 2.43 �3.05 0.004**

External help expectancy 2.94 2.24 �3.21 0.003**

* p< .05.** p< .01.*** p< .001.

P.J.C.M. Embregts et al. / Research in Developmental Disabilities 31 (2010) 1398–1403 1401

A first and certainly not unimportant indicator in the context of prevention is the fact that our results point to a relationbetween parental- and family factors, behavior problems and treatment. The same relation also comes up in a study byEmbregts (2009). She explored characteristics of 62 children with MID and their parents following outpatient and/orresidential treatment. Results showed that parental problems and educational inability can lead to a disturbed balancebetween the support needs and means of a family, and such an imbalance is often the reason for residential treatment. Morespecifically parents of children with behavior problems in our study experience problems in the handling and acceptance ofthe child’s limitations and these parents feel less competent in the raising and education of their child. A multisystemictreatment model in which multiple therapies are undertaken simultaneously, has been shown to be more effective thantreatment of child behavior problems alone (Curtis, Ronan & Borduin, 2004; Gorske, Srebalus & Walls, 2003). Multisystemictreatment is based upon the theory that behavior problems are caused and maintained by the convergence of risk factorswithin the individual, family, school, peer group, extended family and neighborhood.

Parents of children with MID and behavior problems in our study are in need of support. The clearly expected need forexternal help is greater among the parents of children with MID and behavior problems; this goes for the help focused onhim/herself as well as for help focused on others in his/her direct surroundings. In earlier research, it is shown that this needfor support increases as the children get older. Parents of older children, ages 19–21, felt less supported, more isolated andmore in need of expanded services than parents of younger ones (Suelzle & Keenan, 1981). This parental well-being cansubsequently influence the treatment outcomes and is thus an important aspect that needs to be considered whendeveloping an adequate prevention program.

In our study we screened behavior problems by using the Strengths and Difficulties Questionnaire (SDQ) (Goodman,1997). Although the SDQ seems to be a useful alternative for more extensive lists such as the Child Behavior Checklist (CBCL;Achenbach & Edelbrock, 1983), we have some critical comments. First of all, Arseneault, Kim-Cohen, Taylor, Caspi, andMoffitt (2005) found in a survey among 2232 normally gifted children that self-reports about behavior problems are validonly if valid and useful instruments are being used for the target group. The SDQ is initially meant for normally gifted youths,which may be a possible explanation for its low reliability of the pupil’s version in our study. Secondly, a poor concurrencebetween the parent and teacher versions was found. Concurrence between different informants proved also to be poor toaverage using other questionnaires to measure psychopathology under youngsters, such as the Child Behaviour Checklist(Embregts, 2000).

Given the fact that prevention programs, within the care for children with MID, are still in its infancy and that familyprocesses are not incorporated in these programs, this study can be used as a starting point for developing and implementingsuch programs. Because of the dependence of most individuals with MID and the vulnerability of their families as well, it isimportant to develop a program with a low threshold. Promoting competent parenthood, the relationship between the childand his family and the motivation of all participants, will lead to reduction of behavior problems in children of families withdifferent problems and backgrounds.

Acknowledgements

We would like to thank the children, parents and teachers for their participation. We thank Lianne van den Eijnden for herhelp with data collection and Kim van den Bogaard for her comments on the manuscript.

References

Abidin, R. R. (1983). Parenting Stress Index-Manual (PSI). Charlottesville, VI: Pediatric Psychology Press.Achenbach, T. M., & Edelbrock, C. S. (1983). Manual for the Child Behavior Checklist and revised Profile. Burlington, VT: University of Vermont Department of

Psychiatry.Arseneault, L., Kim-Cohen, J., Taylor, A., Caspi, A., & Moffitt, T. E. (2005). Psychometric evaluation of 5- and 7-year-old children’s self-reports of conduct problems.

Journal of Abnormal Child Psychology, 33(5), 537–550.Beck, A., Hastings, R. P., Daley, D., & Stevenson, J. (2004). Pro-social behaviour and behaviour problems independently predict maternal stress. Journal of Intellectual

and Developmental Disabilities, 29, 339–349.de Brock, A. J. L. L. , Vermulst, A. A., Gerris, J. R. M., & Abidin, R. R. (1992). Nijmeegse ouderlijke stress index. Handleiding experimentele versie. Lisse: Swets & Zeitlinger.Curtis, N. M., Ronan, K. R., & Borduin, C. M. (2004). Multisystematic treatment: A meta-analysis of outcome studies. Journal of Family Psychology, 18, 411–419.Dekker, M. C., & Koot, H. M. (2003). Borderline to moderate intellectual disability. II: Child and family predictors. Journal of the American Academy of Child &

Adolescent Psychiatry, 42, 923–931.Einfeld, S. L., & Tonge, B. J. (1996). Population prevalence of psychopathology in children and adolescents with intellectual disability: II epidemiological findings.

Journal of Intellectual Disability Research, 40, 99–109.Embregts, P.J.C.M.. (2000). Reliability of the Child behavior Checklist for the assessment of behavioral problems of children and youth with mild mental

retardation. Research in Developmental Disabilities, 21, 31–41.Embregts, P.J.C.M.. (2009). Residential treatment following outpatient treatment for children with mild to borderline intellectual disabilities: A study of child and

family characteristics. Research in Developmental Disabilities, 30, 1062–1067.Emerson, E. (2003). Prevalence of psychiatric disorders. Journal of Intellectual Disability Research, 47, 51–58.Femmie, J., Bakermans-Kranenburg, M. J., & IJzendoorn, M. (2005). The importance of parenting in the development of disorganized attachment: Evidence from a

preventive intervention study in adoptive families. Journal of Child Psychology and Psychiatry, 46, 263–274.Goodman, R. (1997). Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581–586.Goodman, R. (2001). Psychometric properties of the Strength and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40,

1337–1345.Goodman, R., & Scott, S. (1999). Comparing the Strengths and Difficulties Questionnaire and the Child Behavior Checklist: Is small beautiful? Journal of Abnormal

Child Psychology, 1, 17–24.

P.J.C.M. Embregts et al. / Research in Developmental Disabilities 31 (2010) 1398–14031402

Gorske, T. T., Srebalus, D. J., & Walls, R. T. (2003). Adolescents in residential centers: Characteristics and treatment outcome. Children and Youth Services Review, 25,317–326.

Hastings, R. P. (2003). Child behavior problems and partner menta health as correlates of stress in mothers and fathers of children with autism. Journal ofIntellectual Disability Research., 47, 231–237.

Jane-Llopis, E. (2002). What makes the ounce of prevention effective? A meta-analysis of mental health promotion and mental disorder prevention programmes.Proefschrift Nijmegen: Katholieke Universiteit Nijmegen.

Jane-Llopis, E., Hosman, C., Jenkins, R., & Anderson, P. (2003). Predictors of efficacy in depression prevention programmes. Meta-analysis. British Journal ofPsychiatry, 183, 384–397.

Kazdin, A. E. (1995). Child, parent, and family dysfunction as predictors of outcome in cognitive behavioral treatment of antisocial children. Behaviour Research andTherapy, 33, 271–281.

Muris, P., Meesters, C., Berg, F., & van den, (2003). The strengths and difficulties questionnaire (SDQ). Further evidence its reliability and validity in a communitysample of Dutch children and adolescents. European Child & Adolescent Psychiatry, 12, 1–8.

Robbroeckx, L. M. H., & Wels, P. M. A. (1988). De Nijmeegse vragenlijst voor de opvoedingssituatie (NVOS). Derde versie. Nijmegen: Katholieke Universiteit, instituutvoor de orthopedagogiek.

Suelzle, M., & Keenan, V. (1981). Changes in family support networks over the life cycle of mentally retarded persons. American Journal of Mental Deficiency, 86,267–274.

Wallander, J. L., Dekker, M. C. M., & Koot, H. M. (2003). Psychopathology in children and adolescents with intellectual disability: Measurement, prevalence, courseand risk. Issues in studying. International Review of Research in Mental retardation, 26, 93–134.

Wels, P. M. A., & Robbroeckx, L. M. H. (1996). NVOS: Nijmeegse Vragenlijst voor de Opvoedingssituatie. Handleiding. Lisse: Swets & Zeitlinger.Widenfelt, B. M., Goedhart, A. W., Treffers, P. D. A., & Goodman, R. (2003). Dutch version of the Strengths and difficulties Questionnaire (SDQ). European Child and

Adolescent Psychiatry, 12, 281–289.

P.J.C.M. Embregts et al. / Research in Developmental Disabilities 31 (2010) 1398–1403 1403