A systematic review of the prevalence rates of chronic health conditions in populations of children with intellectual disability was provided. We identified 2,994 relevant studies by searching Medline, Cinahl, and PsycINFO databases from 1996 to 2008. We included the 31 studies that had sufficient methodological quality. The 6 most prevalent chronic health conditions in children with intellectual disability were epilepsy (22.0/100), cerebral palsy (19.8/100), any anxiety disorder (17.1/100), oppositional defiant disorder (12.4/100), Down syndrome (11.0/100), and autistic disorder (10.1/100). The reported prevalence rates of chronic health conditions in this population was much higher than in the general population. However, both the number of studies that were included and the number of chronic health conditions they reported about were limited. There is an urgent need for better evidence on the prevalence of chronic health conditions among children with intellectual disability.
In the past three decades, there has been an increase in knowledge about chronic health conditions in children with intellectual disabilities. In several studies researchers found that many of these children have with a range of chronic health conditions (Emerson & Hatton, 2008; Goddard, Davidson, Daly, & Mackey, 2008; Sturmey, Lindsay, & Didden, 2007; Van Schrojenstein Lantman-de Valk & Walsh, 2008; World Health Organization, 2000).
A major difficulty in studies on this subject is the wide variability of prevalence rates that are reported for specific chronic health conditions in children with intellectual disability. This variability is a result of heterogeneity of several factors, such as size of the sample, definition of the study population, recruitment of the study population, response rate, method of diagnosis, and the use of different classification frameworks for diagnosing certain disorders in children with intellectual disability. As a result, the validity of prevalence rates has been disputed (Borthwick-Duffy, 1994; Bradley, Summers, Wood, & Bryson, 2004; Cooper, Smiley, Morrison, Williamson, & Allen, 2007; Davis & Brosco, 2007).
Both policymakers and healthcare professionals value and need valid prevalence rates. Policymakers require these data for the planning and financing of adequate care arrangements (e.g., health, education, work) in order to enhance the well-being and societal participation of children with intellectual disability and their families. Professionals need these data for the early detection and adequate treatment of chronic health conditions and prevention of the burden of these conditions for children with intellectual disability and their family (Goddard et al., 2008; Hou, Wang, & Chuang, 1998; Luckasson et al., 2002; McDermott, Durkin, Schupf, & Stein, 2007; Newacheck, Rising, & Kim, 2006; Strømme & Hagberg, 2000; van der Lee, Mokkink, Grootenhuis, Heymans, & Offringa, 2007; Yeargin-Allsopp, Boyle, Van Naarden Braun, & Trevathan, 1997).
Our aim in this study is to provide an overview of the prevalence rates of chronic health conditions in populations of children with intellectual disability by presenting a systematic literature review. This review is part of a research project on chronic health conditions in a population of children with intellectual disability in the Netherlands.
Data Sources and Identification of Studies
We searched the following electronic databases for relevant studies from 1996 to 2008: Medline, Cinahl, and PsycINFO. Our search terms were mental retardation, intellectual disability or disabilities, mentally disabled persons, mental development, intellectual development, developmental disability or disabilities, learning disability or disabilities, or learning development in Major Medical Subject Headings Descriptor (MJME), and learning disorders in all subheadings. We used the International Classification of Diseases (ICD-9) classification chapter entitled 1 to 16 (World Health Organization, 1977), the ICD-10 classification chapter entitled I to XVIII (World Health Organization, 1992), the Diagnostic and Statistical Manual of Mental Disorders (DSM III and III-R) classification chapter entitled “Disorders Usually First Evident in Infancy, Childhood, Adolescence” (American Psychiatric Association, 1980, 1987), and the DSM IV and IV-TR classification chapter entitled “Disorders Usually First Diagnosed in Infancy, Childhood, or Adolescence” (American Psychiatric Association, 1994, 2000) for the selection of chronic health conditions in children. The diagnostic titles were translated in Medical Subject Headings with complications, diagnosis, or epidemiology in all subheadings. Additional studies were obtained from the reference lists of the included studies.
In order to be considered for inclusion, researchers had to report mainly about children (<19 years) diagnosed with borderline, mild, moderate, severe, or profound intellectual disability and with a chronic health condition. Intellectual disability was ideally established by a validated intelligence and/or adaptive behavior test and ideally classified according to the DSM classifications (American Psychiatric Association, 1980, 1987, 1994, 2000) or the ICD (World Health Organization, 1977, 1992). Although borderline intellectual disability has not been an accepted level for quite a long time, studies in which researchers reported on children with borderline intellectual disability were included because this group faces substantially elevated cognitive and morbidity risks as well as problems in adaptive behavior (Ferrari, 2009; Masi, Marcheschi, & Pfanner, 1998).
A chronic health condition was defined as a chronic physical, developmental, behavioral, or emotional condition in children ages 0 to 18 years. The condition had to be present for longer than 3 months, would probably last longer than 3 months, or had occurred three times or more during the past year and would probably recur. The chronic health condition had to be diagnosed according to professional standards or extracted from medical case files or registers and ideally classified according to the DSM or ICD (Mokkink, van der Lee, Grootenhuis, Offringa, & Heymans, 2008; van der Lee et al., 2007).
Eligible studies had to be in English; have a cohort, patient-control, or cross-sectional design; and had to be focused on a population of children with intellectual disability. We excluded studies that were focused exclusively on subpopulations of children with a specific biomedical cause of intellectual disability, such as Down syndrome, because there are reviews available about the prevalence of chronic health conditions related to specific conditions (Merrick, Kandel, & Vardi, 2004; Prasher, 1999; Roizen & Patterson, 2003). In addition, studies in which participants were all adults, those with other designs (e.g., case studies), studies that were not original research (e.g., reviews or comments), and studies in which the diagnoses were based on screening instruments were also excluded.
Three of the authors independently screened the titles and abstracts of the relevant studies from 1996 to 2008 for eligibility. Each reviewer screened two thirds of the titles and abstracts; therefore, each title and abstract was screened by two reviewers. Any discrepancies between reviewers were resolved by discussion; if necessary, by obtaining the full text article; and by consultation of a fourth author.
The interrater reliability statistics were very good (Cohen, 1960). Cohen's weighted kappas were 0.81 (second and last author), 0.82 (first and last author) and 0.82 (first and second author) for the screening of titles and abstracts (outcomes: inclusion, exclusion, or doubt). There was disagreement among the reviewers in 8 to 10% of the 2,994 studies. Most disagreements were resolved by discussion between the reviewers. In 12 unsure cases, the full text of the article was obtained because eligibility could not be determined from the title and abstract alone. In 2 cases of remaining doubt, the fourth author was consulted.
We obtained and analyzed full texts of potentially eligible studies. Teams of two reviewers independently assessed all studies identified for full-text analysis; the first author analyzed all the studies, whereas the second and last author analyzed about half of these studies. The reviewers used a structured data form based on Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist to extract the key data for the assessment of the methodological quality of the included studies (von Elm et al., 2007).
For 570 studies eligible for methodological quality assessment (values: low, weak to medium, good and high), reviewers disagreed on 16% to 17% of the studies. However, Cohen's (1960) weighted kappas were very good .85 (first and last authors) and .89 (first and second authors) because the interrater disagreements were smaller compared to the interrater disagreements when screening titles and abstracts. All disagreements were resolved by discussion between the reviewers.
Assessment of Methodological Quality
There is no gold standard for assessing the quality of observational studies (Mallen, Peat, & Croft, 2006); therefore, we developed a method based on the domains that are of importance for the internal and external validity of the studies. We assessed the methodological quality of the eligible studies by examining the following characteristics: contextual information, population (bias), selection (bias), outcome measures reliability, results, and confounders. We assigned weights to each characteristic, taking into account the importance of each characteristic for internal and external validity (Mallen et al., 2006; Nguyen, Bezemer, Habets, Prahl-Andersen, 1999; Sanderson, Tatt, & Higgines, 2007). The appraisal of the overall methodological quality of each study was based on a total score on the different indicators of these characteristics (see Table 1).
We calculated the total quality scores for each study by summing up the score on each indicator (range from 0 to 11 points) and classified the quality scale scores into the following ordinal categories: low quality (score < 6), weak to medium quality (score 6 and 7), good quality (score 8 and 9), and high quality (score 10 and 11) (El Baz et al., 2007; Verhagen et al., 2000). We excluded studies with a low methodological quality from further analyses.
Analysis and Synthesis
We first assessed interrater reliability by calculating Cohen's weighted kappas with MEDCALC v188.8.131.52. for the screening of titles and abstracts and for the methodological quality assessment (Cohen, 1960). Subsequently, we grouped the prevalence rates of the included studies by somatic chronic health conditions (epilepsy, cerebral palsy, visual problems, hearing problems, and miscellaneous somatic chronic health conditions); congenital malformations (genetic–chromosomal/sex-linked malformations and other chromosomal, endocrine/metabolic diseases malformations); pervasive developmental disorder (autistic disorder, pervasive developmental disorder other than autistic disorder); attention-deficit/hyperactivity disorder /hyperkinetic disorder; and miscellaneous psychiatric disorders. The weighted mean prevalence rate and the 95% confidence intervals (CIs) were calculated if more than two studies reported about a disease. The weighted mean was used to aggregate the prevalence rates of the different chronic health conditions that we found in each study to a single resultant score. Studies with large populations contribute more to the weighted mean than do smaller ones. The following equation was the formula weighted mean:
Search for Trials
Figure 1 shows the study selection process. The database search resulted in 3,216 potentially relevant studies: 2,478 Medline, 510 Cinahl, and 228 PsycINFO; 241 of these were duplicates. The screening of the reference lists of the included studies added another 19 studies, resulting in a total number of 2,994 studies. After screening the titles, we excluded 2,424 studies because they did not meet the inclusion criteria. After full-text analysis, we eliminated another 539 studies because (a) study exclusively about subpopulations of children with a specific biomedical cause of intellectual disability (n = 170); selective population or low response rate (n = 107); did not report about chronic health condition or intellectual disability (n = 92); review, case report, or comment/letter (n = 82); age-range too broad (n = 65); diagnoses made based on screening instruments (n = 23). Finally, 31 studies were included (Table 2).
Description of the Studies
In Table 2, the characteristics of the 31 included studies are presented. Sixteen studies were classified as high quality methodological studies; 13, as good quality; and 2, as weak to medium quality. The studies were predominantly conducted in Europe (n = 17). Six studies were from North America; 2, South Africa; 2, Australia, 2, Asia; and 2, Israel. Most of the studies (n = 22) were focused on more than one chronic health condition: epilepsy (n = 14); cerebral palsy (n = 11); visual problems (n = 4); hearing problems (n = 5); miscellaneous somatic chronic health conditions (n = 7); congenital malformations genetic–chromosomal/sex-linked (n = 10); congenital malformations, other chromosomal, endocrine/metabolic diseases (n = 9); autistic disorder (n = 11); pervasive developmental disorder other than autistic disorder (n = 7); attention-deficit/hyperactivity disorder/hyperkinetic disorder (n = 8); and miscellaneous psychiatric disorders (n = 7). All researchers reported about children (< 19 years). However, 7 studies also included young adults with a maximum age of 22 years. There was a huge variation in the size of the population with intellectual disability surveyed (range: 64 to 11,892). Response rates were generally high (67% to 100%); only 6 studies had a response rate below 80%.
In most studies the researchers recruited the children from special schools and certain health or social services: special schools and health services (n = 8), special schools, health and social services (n = 7), special schools (n = 3), private households (n = 3), health and social services (n = 2), health services (n = 2), special schools and social services (n = 1), special schools day-care and residential centers (n = 1), special schools and training center (n = 1), residential centers (n = 1), community registers (n = 1), and department of developmental services (n = 1).
Methods of Diagnosis and the Classification Framework
The various methods used in the assessment of the diagnoses of intellectual disability and of chronic health conditions are presented in Table 3 in addition to their classification frameworks.
In 20 studies intellectual disability was assessed with intelligence tests; in 14 studies, in combination with developmental tests. Six studies included children and adolescents who were eligible for health, education, and social services for people with intellectual disability. In 2 studies the diagnosis of intellectual disability was based on primary caregiver and/or teacher reports. One study included children and adolescents eligible for social services for people with intellectual disability or diagnosed with intellectual disability by school achievement and/psychological tests, researchers in one study used developmental tests for establishing intellectual disability, and in one study the intellectual disability diagnosis was extracted from the family history and medical records.
Chronic health conditions
In 17 studies the diagnoses of chronic health conditions was made by clinical examination or extracted from registers, medical files, or from blood samples. In 14 studies investigators used multiple methods of diagnosis, in different combinations, for establishing the diagnosis. In most studies that were focused on somatic chronic health conditions or congenital malformations, the researchers did not use a classification system for the diagnoses they reported. The International League Against Epilepsy criteria were used in 3 of the 14 studies about epilepsy. The International Association for Prevention of Blindness criteria were used in 1 of the 4 studies in which researchers reported about visual problems. Only 3 studies used the ICD 9 or 10 as a classification system for different somatic chronic health conditions or congenital malformations. The DSM or the ICD was used in most studies about pervasive developmental disorder or psychiatric disorders. In 4 studies, the investigators did not use a classification system for the pervasive developmental disorder or psychiatric disorders diagnoses they reported.
Chronic Health Conditions
Table 3 presents the prevalence rates of chronic health conditions in the included studies and the methods of diagnosis and classification framework that were used by the researchers. The weighted mean prevalence rate and the 95% CI of the chronic health conditions that were reported by investigators in more than 2 studies are presented in Table 4.
Prevalence rates of epilepsy were reported in 14 studies and ranged from 5.5% to 35.0%. The weighted mean prevalence rate was 22.0/100 (CI 20.8–23.2).
Prevalence rates of cerebral palsy were reported in 11 studies and ranged from 8.4% to 33.8%. The weighted mean prevalence rate was 19.8/100 (CI 18.6–21.1).
In 4 studies researchers reported about visual problems such as refractive errors, strabismus, visual acuity, visual field, or visual impairment in general. Prevalence rates ranged from 2.2% to 26.8%. We were not able to calculate a weighted mean prevalence rate for the different visual problems because of the variability between the studies regarding outcomes.
Five studies were about hearing impairment or disability in general. Prevalence rates ranged from 0.0% to 7.1%. The weighted mean prevalence rate was 4.5/100 (CI 3.4–5.7).
Miscellaneous somatic chronic health conditions
Researchers in 7 studies reported prevalence rates of miscellaneous somatic chronic health conditions, such as chronic obstructive pulmonary disease (8.9%), gastric and esophageal diseases (6.9%), back and neck disorders (6.9%), osteoarthropathia (2.5%), cerebrovascular accident (range: 0.3% to 2.5%), Reye syndrome (0.3%), human immunodeficiency virus (0.0%), and other chronic health conditions (4%). The weighted mean prevalence rate for cerebrovascular accident was 2.0/100 (CI 1.8–2.3). We were not able to calculate a weighted mean prevalence rate for the other chronic health conditions.
Congenital malformations genetic–chromosomal and sex-linked disorders
Ten studies were focused on congenital chromosomal malformation. Prevalence rates of Down syndrome, fragile X, cri-du-chat, and disorders other than Down syndrome chromosomal malformation and prevalence ranged from 0.1% to 20.3% for the different disorders. The weighted mean prevalence rate for Down syndrome was 11.0/100 (CI = 10.5–11.4); for fragile X, 1.9/100 (CI = 1.6–2.1); for cri-du-chat, 0.2/100 (CI = 0.2–0.3); and for malformations other than Down syndrome, 3.1/100 (CI = 2.9– 3.3).
Congenital malformations other chromosomal, endocrine/metabolic diseases
Prevalence rates of congenital malformations other chromosomal, endocrine/metabolic diseases were reported in 9 studies. Researchers reported prevalence rates of malformations of the nervous system (e.g., spina bifida, hydrocephaly), other malformations (e.g., Prader-Willi, tuberous sclerosis, malformations of the musculoskeletal, genital/urinary, digestive, or circulatory system) and metabolic, endocrine/thyroid gland disorders (e.g., phenylketonuria, hypothyroidis) (range = 0.8% to 13.1% for the different disorders). The weighted mean prevalence rate for malformations of the nervous system was 5.1/100 (CI = 4.6–5.6); for other malformations, 6.1/100 (CI = 5.7–6.4); and for metabolic, endocrine/ thyroid gland disorders, 0.5/100 (CI = 0.4–0.7).
Prevalence rates of autistic disorder were reported in 11 studies and ranged from 4.5% to 25.1%. The weighted mean prevalence rate was 10.1/100 (CI = 8.8–11.6).
Pervasive developmental disorders other than autistic disorder
Prevalence rates of this disorder, including Rett syndrome, Asperger syndrome, atypical autism, pervasive developmental disorder–not otherwise specified (NOS), autistic symptoms, autistic-like disorder and autistic disorder-NOS, were reported in 7 studies and ranged from 0.6% to 7.9% for the different disorders. Prevalence rates of the pervasive developmental disorder categories that are NOS, such as pervasive developmental disorder-NOS, atypical autism, and autistic disorder–NOS, were reported in 4 studies and ranged from 2.5% to 7.9%. The weighted mean prevalence rate for pervasive developmental disorder categories–NOS was 7.1/100 (CI = 5.8–8.6). We were not able to calculate a weighted mean prevalence rate for the other diseases.
Attention-deficit hyperactivity disorder/Hyperkinetic disorder
Prevalence rates were reported in 8 studies (range = 5.9% to 30.0%. The weighted mean prevalence rate was 9.5/100 (CI = 8.2–10.9).
Miscellaneous psychiatric disorders
Researchers in 7 studies reported prevalence rates of miscellaneous psychiatric disorders. Most frequently mentioned were conduct disorder, with prevalence rates ranging from 0.6% to 8.4%; oppositional defiant disorder, with prevalence rates ranging from 11.1% to 13.9%; any anxiety disorder, with prevalence rates ranging from 11.4% to 39.0%; and tic disorder, with prevalence rates ranging from 0.8% to 4.6%. The weighted mean prevalence rate for conduct disorder was 5.1/100 (CI = 4.1–6.4); for oppositional defiant disorder, 12.4/100 (CI = 10.7–14.4); for any anxiety disorder, 17.1/100 (CI = 15.1–19.4); and for tic disorder, 1.1/100 (CI = 0.6–2.0). Prevalence rates of enuresis/encopresis, affective, behavioral/emotional, anxiety/phobic/obsessive-compulsive, depressive, emotional, mood, psychotic, eating, impulse control, and somatoform disorders were also reported, ranging from 0.6% to 17.2% for the different disorders. We were not able to calculate a weighted mean prevalence rate for these disorders.
This systematic literature review shows high prevalence rates of a wide range of chronic health conditions in children with intellectual disability. We included 31 studies that were focused on a limited number of chronic health conditions that had a sufficient methodological quality. In general, the quality of these studies was good to high. However, the prevalence rates for most chronic health conditions varied among the included studies. The characteristics of the sample, recruitment of the study population, method of diagnosis, classification framework used, and factors not examined and appraised in this study, such as the quality of registers, diagnostic overshadowing, or the accessibility of healthcare for children with intellectual disability in different countries, can contribute to this variation (Borthwick-Duffy, 1994; Bradley et al., 2004; Cooper et al., 2007; Fletcher & Fletcher, 2005; Jopp & Keys, 2001). Nevertheless, the prevalence rates of chronic health conditions in children with intellectual disability are higher than the prevalence rates reported in studies of children without intellectual disability (Bowley & Kerr, 2000; Dykens, 2000; Fombonne, 2003, 2005; Froehlich et al., 2007; Hastings, Beck, Daley, & Hill, 2005; Jansen, Krol, Groothoff, & Post, 2004; Sillanpaa, 1999; Simonoff, Pickles, Wood, Gringras, & Chadwick, 2007; Volkmar, Lord, Bailey, Schultz, & Klin, 2004; Yeargin-Allsopp, Boyle, Van Naarden Braun, & Trevathan, 2008). We note that in contrast to the reviews of Fombonne (2003, 2005, 2009), we found higher prevalence rates of autistic disorder compared to pervasive developmental disorders other than autistic disorder. An explanation could be the inclusion of studies in Fombonne's reviews in which the researchers reported findings about children who do not have intellectual disability. Children with autistic disorder are more likely to have intellectual disability compared with children who have pervasive developmental disorder other than autistic disorder (Yeargin-Allsopp et al., 2003). However, Bertrand et al. (2001) and Williams, Thomas, Sidebotham, and Emond (2008) also reported higher prevalence rates of autistic disorder compared to pervasive developmental disorder-NOS in population studies with children with and without intellectual disability.
Fit With Previous Studies
To our knowledge this is the first systematic review on chronic health conditions in children with intellectual disability. Comparison of our results with other reviews on chronic health conditions in children with intellectual disability (Dykens, 2000; Kerker, Owens, Zigler, & Horwitz, 2004; Lhatoo & Sander, 2001; Masi, 1998; Owens, Kerker, Zigler, & Horwitz, 2006; Roizen & Patterson, 2003; Van Schrojenstein Lantman-de Valk & Walsh, 2008; Whitaker & Read, 2006) is difficult. In contrast to traditional narrative reviews, we appraised the studies on methodological characteristics reflecting the internal and external validity.
Strengths and Limitations
A major strength of our study is the very thorough search strategy that covered all relevant literature databases and included a check of the references of the papers that were found. Despite this, we may still have missed some publications (e.g., studies that are not indexed well in the databases). Another strength is that we used a consensus on chronic health conditions and health conditions in childhood (0–18 years of age) to define chronic health condition (Mokkink et al., 2008), which is operationalized in a list of chronic health conditions in the ICD (World Health Organization, 1977, 1992) and the DSM (American Psychiatric Association, 1980; American Psychiatric Association, 1987, 1994, 2000).
We limit the generalization of the results to the chronic health conditions reported in the studies included in our systematic review and to the particular European “industrialized countries.” We did not find many studies from the less industrialized countries.
Another limitation is our focus on studies in which researchers reported about chronic health conditions in children with intellectual disability in general. By excluding studies that were focused exclusively on subpopulations of children with a specific biomedical cause of intellectual disability, such as Down syndrome, we are not addressing the prevalence of syndrome-related health problems that are associated with these conditions. Although diseases such as congenital heart disorders, obesity, celiac disease, and thyroid disorders are very prevalent in children with Down syndrome, these authors did not sufficiently address them. Inclusion of studies on children with Down syndrome in the present study would increase the prevalence of these conditions for the population with intellectual disability. However, there are already reviews available about the prevalence of chronic health conditions in Down syndrome (Merrick et al., 2004; Prasher, 1999; Roizen & Patterson, 2003).
Implications for Clinicians
The high prevalence rates of chronic health conditions in children with intellectual disability should alert clinicians, who are crucial in the identification and registration of chronic health conditions. Identification begins with a complete history, including a pre-, peri-. and postnatal medical history, physical examination, and psychological and social evaluation (Hou et al., 1998; Luckasson et al., 2002; Ru, 2008; van Schrojenstein Lantman-de Valk et al., 2008). Early detection and adequate treatment of chronic health conditions is important because these conditions have a significant negative impact on the well-being and social participation of children with intellectual disability and their families (Goddard et al., 2008; Hou et al., 1998; Luckasson et al., 2002; McDermott et al., 2007; Newacheck et al., 2006; Strømme & Hagberg, 2000; van der Lee et al., 2007; Yeargin-Allsopp et al., 1997). Moreover, a good diagnostic work-up according to professional standards accompanied with a good registration system are useful tools to delineate (probable) causes (Strømme & Diseth, 2000). Evidence on the association between intellectual disability and chronic health conditions may identify etiologic clues that are necessary for the early identification of these conditions and the development and implementation of effective programs to increase opportunities for children with intellectual disability (Cans et al., 1999; Kirby, 2002; Luckasson et al., 2002; Petterson, Bourke, Leonard, Jacoby, & Bower, 2007).
Implications for Research
To validate evidence on the prevalence of chronic health conditions in children with intellectual disability, researchers should first reach consensus about the classification framework that should be used in prevalence studies for psychiatric disorders in children with intellectual disability. A promising development is the future alignment of the DSM-V with the ICD-11, which will present an opportunity to unify and strengthen knowledge of global mental health. However, the debate about the validity of both classification systems as clinical and research tools has not yet been resolved (Banzato, 2004; First & Westen, 2007; Kupfer, Regier, & Kuhl, 2008; Moller, 2008). Other developments in this area that could be used are the Diagnostic Criteria for Psychiatric Disorders for Use With Adults With Learning Disabilities/Mental Retardation, developed by the Royal College of Psychiatrists (Cooper, Melville, & Einfeld, 2003; Royal College of Psychiatrists, 2001) or the Developmental Psychiatric Assessment (Dosen, 2005a, 2005b). Second, the use of multiple data sources, in contrast to the one data source strategy used in most studies, will improve the validity of the prevalence rates (Yeargin-Allsopp et al., 1997). Third, transnational comparative population studies are needed that include less industrialized countries. Fourth, researchers should improve the way they report their studies to enable a proper assessment of internal and external validity. This may be accomplished by using the recently developed STROBE guidelines (von Elm et al., 2007).
Our review shows that caring for children with intellectual disability could be improved by the availability of better evidence on the occurrence of chronic health conditions.
Indicates study included in our review.
Editor-in-Charge: Steven J. Taylor
Barth Oeseburg, PhD (e-mail: firstname.lastname@example.org), Researcher; Geke J. Dijkstra, PhD, Researcher; Johan W. Groothoff, PhD, Professor; Sijmen A. Reijneveld, PhD, Professor; Daniëlle E. M. C. Jansen, PhD, Researcher, Department of Health Sciences and Graduate School for Health Research (SHARE), PO Box 30.001, University of Groningen, Groningen 9700 RB, The Netherlands. The first author is also affiliated with the Wenckebach Institute and the last author, with the Department of Sociology, University of Groningen.