|Home | About | Journals | Submit | Contact Us | Français|
Delivering mental health services to children and their families through schools has many potential advantages. However, little is known about the characteristics of children referred to such services. This study aimed to determine the pattern of mental health and learning difficulties of children referred to one school mental health service.
An identity stripped administrative database of all new referrals (n=353) to a school mental health program in southern Alberta between September 2006 and June 2009 was used. Teacher Strengths and Difficulties Questionnaire responses and questions about learning and other developmental problems were included.
Hyperactivity-inattention was the most prevalent mental health concern, and spelling was the most common learning concern. Higher rates of hyperactivity-inattention concerns and pro-social deficits were observed for boys and more emotional problems were observed for girls. Hyperactivity-inattention was higher at lower grades. Hyperactivity-inattention and conduct problems were often comorbid as were several learning problems.
Understanding the typical patterns of concerns among referrals to school mental health services may guide the prioritization of assessment and intervention approaches within these programs. Findings suggest assessments and interventions for ADHD and other disruptive behaviours should be prioritized, as well as the provision of cognitive and academic testing.
La prestation de services de santé mentale aux enfants et à leur famille dans un contexte scolaire présente de nombreux avantages. Toutefois, on dispose de peu d’informations sur les caractéristiques des enfants référés à un service scolaire de santé mentale. Cette étude présente le schéma mental et les difficultés d’apprentissage des enfants référés à un service scolaire de santé mentale.
Utilisation de la base de données (anonyme) de 353 sujets référés au programme de santé mentale d’une école du sud de l’Alberta, de septembre 2006 à juin 2009. Les réponses au Teacher Strengths and Difficulties Questionnaire et aux questions sur les problèmes d’apprentissage et de comportement ont été incluses.
L’hyperactivité/inattention était le principal problème de santé mentale, et l’orthographe la principale difficulté d’apprentissage. Les garçons présentaient davantage de symptômes d’hyperactivité/inattention et de difficultés de socialisation que les filles qui, elles, avaient davantage de problèmes émotionnels. L’hyperactivité/inattention était plus marquée dans les petites classes. Les comorbidités les plus fréquentes étaient les troubles de conduite et les troubles d’apprentissage divers.
Comprendre le type de difficultés rencontrées chez les enfants référés à des services scolaires de santé mentale aide à prioriser les techniques d’évaluation et d’intervention proposées. Il convient de prioriser l’évaluation et le traitement du TDAH et des autres troubles du comportement, de faire passer des tests cognitifs et de vérifier les connaissances scolaires des enfants.
School mental health clinics are an important approach to service delivery in child mental health. Expansion of school mental health services was flagged within the recent Canadian Senate report on mental health services in Canada (Kirby & Keon, 2006). School mental health clinics have been recognized as a potential solution to removing common barriers that impede children from receiving mental health services, such as financial concerns, transportation problems, poor accessibility, inflexible office hours, stigma, and long waitlists (American Academy of Pediatrics, 2004; Walrath et al., 2004; Weist, 2000). Additionally, school personnel may be positioned to (i) offer detailed information about a student’s behaviour and functioning across various circumstances and over a significant period of the day, (ii) aid in the identification of deviant behaviour due to familiarity with typical child behaviour and access to a large normative sample, and (iii) have ready contact with the student’s family (McLennan et al., 2008; Walrath et al., 2004). However, a lack of information as to what are typical concerns identified by teachers for students referred to school mental health services impairs attempts to optimally design these services.
Despite the growing body of literature on school mental health services, there are only a limited number of reports describing the mental health characteristics of referred students. One United States (US) study compared a sample of school mental health assessed and treated children and youth aged 5 to 18 years to a sample of children and youth receiving similar services in a community clinic (Armbruster & Lichtman, 1999). In the school sample, 43.5% were referred for externalizing problems versus 35.3% in the clinical sample. Internalizing problems were comparable for the two samples, with 27.9% of children in the school sample and 24.9% in the clinical sample. Children referred to the clinic were also significantly more likely than children referred to the school service to be referred for both externalizing and internalizing reasons, 34.3% versus 21.9% respectively. The authors also noted demographic differences of students in the school sample whereby 92.2% were of ethnic minority status compared to only 48.2% in the clinical sample. Moreover, a higher number of students in the clinical sample had private insurance, which may indicate higher socio-economic status.
A second US study also compared users of school mental health services with users of community mental health centers (Weist et al., 1999). Two samples of children and youth, aged 10 to 19, were compared on numerous measures including general background, violence exposure, life stress, family supportive behaviour, self concept, and behavioural and emotional problems. Differences between the two samples on these psychosocial measures were not found to be significant. One difference, however, was that children with internalizing problems in the school program were significantly less likely to have had past involvement with the mental health system when compared to the community sample. In terms of demographic differences, 66% of students were African American in the school sample which was significantly higher than the 28% in the community sample.
Another US study examined the characteristics of children identified by their school as having behavioural or emotional problems which resulted in their placement into a special education program (Kutash & Duchnowkski, 2004). The study took place across 10 different urban schools, ranging from elementary to high school. Results from the Child Behaviour Checklist found that 73% of the 158 children in this sample scored within the borderline or clinical range on the Total Problems scale. Of children in the clinical range, 63% had externalizing problems while 39% had internalizing problems.
A recent meta-analysis conducted by Reddy et al. (2009) examined the effectiveness of school based prevention and intervention programs for children and youth under 18 years of age with emotional disturbances. In the analysis, the authors briefly reported on the characteristics of the disorders of the children and youth, which was available for 14 of the 28 articles reviewed (552 subjects). A majority of the children in this analysis had disruptive behaviour problems (64%), while a much smaller percentage had internalizing problems (6.5%). Learning disabilities and adjustment disorders were flagged in only 3 to 4% of these children.
Whether findings from these US studies can be generalized to the Canadian context is unknown given differences in health service structures and funding. In addition, referral patterns may differ as a function of the nature of the school mental health service, as well as the age, gender, and other demographic variables of children in the sample. Furthermore, patterns found in non-school clinic samples may not generalize to school mental health services. In particular, school referrals may be more teacher than parent-driven, the latter case seen for more traditional clinical samples.
The objective of this study was to determine the patterns of mental health and learning difficulties reported by teachers for children referred to a school mental health program by using administrative data. Specific research questions included: (i) what is the frequency of reported mental health and learning problems (including comorbid patterns), and (ii) do the frequency patterns vary by grade and/or child gender? This information may inform efforts to improve school mental health initiatives.
Participants were referrals to the Community Outreach in Pediatrics/Psychiatry and Education (COPE program). COPE is a school mental health program serving all elementary schools in the Calgary and the Rockyview school districts (located around Calgary), Alberta, Canada. COPE is a partnership program between the schools and the health and social service sectors. Participating schools identify children with suspected emotional, behavioural and/or developmental problems and refer prioritized children to COPE. This process entails completing referral forms, as well as a screening meeting with COPE and school personnel. Following a screening meeting, most referred children are provided a physician-based assessment coordinated with the school and family. Additional details about the processes within this program are described elsewhere (McLennan et al., 2008). The COPE program was developed, in part, as an attempt to provide earlier access to developmental and mental health assessments for children and families who may not have ready access to timely care.
Data were extracted from a database containing all new referrals of children, ranging from kindergarten to grade 10, to COPE from September 2006 to June 2009. Complete data were available for mental health problems for 353 out of a possible 361 children in the database (97.8%). Complete data for learning concerns was available for 236 out of 353 children (66.9%). Learning problem data were only available for September 2006 to June 2008. There were no gender, grade, or Strengths and Difficulties Questionnaire -Total Difficulty Score differences between those with and without learning problem data.
The referral package from the teacher contains the extended version of the Strengths and Difficulties Questionnaire (SDQ) for each referred child. The SDQ is a commonly used screening measure for mental and social difficulties seen in children. The core 25 items, with response options of “not-true”, “somewhat true,” and “certainly true,” contain five subscales with five items each: (i) hyperactivity-inattention, (ii) conduct problems, (iii) emotional symptoms, (iv) peer problems, and (v) pro-social problems. The SDQ has satisfactory reliability suggested by acceptable internal consistency, inter-rater agreement and test-retest stability (Goodman, 2001). Cut points proposed by the instrument developers allow for score categorization into normal, borderline, and abnormal categories for all subscales. The abnormal cutpoints were developed to correspond to approximately the extreme 10% of community samples (Goodman, 1997, 2001). The extent to which these cutpoints predict psychiatric disorders have been reported elsewhere (Goodman, 2001).
Within the same referral form, teachers rate concerns about learning (math, printing/writing, reading, and spelling), speech and language, and fine and gross motor skills. A single item for each is included in the referral form using the same response options as the SDQ, i.e., “not true”, “somewhat true”, and “certainly true.” Additional information included on the teacher referral form included child gender and current grade. No other socio-demographic information was available in this administrative dataset.
Prevalence of problems was first examined by computing the percentage of students who fell into the abnormal range for these categories. For this analysis, only those meeting the cut point for “abnormal” were identified as possible “cases” of a mental health disorder. Similarly, only those rated as “certainly true” were scored as positive for a learning, speech, and fine or gross motor difficulty. Variation by gender and age were then determined using Pearson’s chi-square test. Finally, comorbidity patterns were studied by examining the co-occurrence of elevated scores on the SDQ subscales and the relationship between these and the learning difficulties.
The study was approved by the ethics board of the University of Calgary/Calgary Health Region and entailed the release of the administrative database containing the above elements without child, parent or teacher identifiers.
The majority of children in the sample were male (79.2%). Break down by grade was 27.6%, 34.8%, 25.6% and 12.0% for K-1, 2–3, 4–5, and 6+ respectively. Frequency distribution of problems by gender is presented in Table 1. Hyperactivity-inattention was the most frequently endorsed symptom cluster at the abnormal level for both genders though was significantly higher for boys. Deficits in pro-social behaviours were also more frequent in boys. In contrast, emotional problems were more frequently endorsed for girls. The distribution of the number of different mental health problems were similar for boys and girls with mean values of 2.8 (S.D. 1.3) and 2.6 (S.D. 1.3) respectively.
Spelling was the most common learning problem endorsed by teachers (Table 1). There were no significant differences in learning problems by gender. Mean number of learning concerns were 1.8 (S.D. 1.5) for both boys and girls. Fine motor concerns and speech and language concerns were reported in about one-third of the sample with no gender differences.
In examination of problems by grade, the only significant pattern was an overall trend for lower prevalence of hyperactivity-inattention with increasing grades (Table 2). There was an increase in the prevalence of emotional problems with higher grades though this was not statistically significant. No other patterns were apparent by grade grouping including total number of mental health or learning problems.
Co-occurring hyperactivity-inattention and conduct disorder symptoms were the most common mental comorbidities (Table 3). The next largest overlaps were hyperactivity-inattention and almost all the learning problems, as well as with fine motor concerns, each occurring in over 25% of children. There were no significant gender difference patterns.
Hyperactivity-inattention symptoms were the most common problems endorsed for children referred to this school mental health service, with hyperactivity-inattention and conduct problem clusters the most common combination. This is consistent with other school studies identifying a predominance of referrals for externalizing students (Armbruster & Litchman, 1999; Kutach & Duchnowski, 2004; Reddy et al., 2009). The gender analysis finding higher prevalence of some disruptive behaviour in the boys and more emotional symptom clusters in girls is also consistent with previous findings (Ford et. al, 2003). However, overall there were few gender differences. Concerns of learning problems were common and frequently comorbid with the mental health problems.
Prevalence of hyperactivity-inattention in the overall sample peaked in grades 2 to 3 and subsequently declined with grade. This finding is similar to findings in the Great Smokey Mountain study, a community-based epidemiological study, wherein some childhood disorders, such as attention-deficit/hyperactivity disorder (ADHD), seem to decline around age 12 (Costello, et al., 2003). The British Child and Adolescent Mental Health Survey found ADHD hyperactive type tending to peak at age 11 to 12, and declining thereafter (Ford et al., 2003). However, caution is needed in contrasting our referred clinical sample and these representative community samples.
High levels of conduct disorder within those children identified as having hyperactivity concerns is similar to past research (Barkley, 2006, Biederman et al., 2002). However, gender differences, in particular the tendency for boys with hyperactivity concerns to be significantly more likely to have a comorbid externalizing problem compared to girls, was not replicated in this study (Barkley, 2006; Biederman et al., 2002).
It is important to consider limitations of this study. First, these results were based on only one mental health service system in one district and it is unknown whether they can generalize to other jurisdictions. However, this service covers all elementary schools across a large urban district and a peri-urban/rural region and includes all school types, i.e., public, Catholic, private. Second, the study is based on a screening instrument, not a diagnostic instrument. Though the screening instrument has strong psychometric properties, a more detailed assessment would be required to generate specific diagnostic patterns. Similarly specific cognitive and academic testing would be required to determine the actual presence of learning disorders versus teacher concerns.
The particularly high rates of hyperactive-inattention symptom clusters may partly be an awareness by the referring agents that the service has physician-based assessments and hence the potential for review of the role of medication. The high rate of conduct symptoms and combination of hyperactive-inattention and conduct symptoms is likely driven by the obvious impact these difficulties have in the classroom and school setting and hence serve a very immediate press to seek help. These same factors may explain the relatively low rate of identified emotional symptoms in the referred group. Children with prominent emotional symptoms are potentially missed or are prioritized lower given less direct impact in the school setting and/or they are referred to other services that may not be predominately physician-based (e.g., community mental health services, psychotherapists).
The high rates of concerns of learning problems in the hyperactive-inattentive group may be linked to a common referral question within this service as to whether a child’s poor academic achievement is a function of ADHD or a learning disability. Given the lag in children receiving timely cognitive and academic testing in the schools to more definitely identify learning disorders, the latter concern has often not been examined prior to the physician-based assessment. The lack of cognitive and academic testing resources in COPE impairs the ability to adequately address this type of questions at the time of the physician-based assessment.
Given the high prevalence of hyperactive-inattention concerns, this program and others like it should be well-prepared to provide systematic assessments for ADHD and other disruptive behaviours and facilitate access to evidence-based services for these difficulties. ADHD may be a particularly good challenge to prioritize within school-mental health partnerships for several reasons including teachers’ excellent vantage point for helping to identify ADHD symptoms, as well as the potential for feedback on treatment response monitoring, the evidence for improvement through school-based behavioural modification (Pelham & Fabiano, 2008), and the level of comorbid learning concerns that require a strong school-based response.
Currently the program uses parent and teacher MTA-SNAP-IVs as a second stage screening for possible ADHD cases to increase the identification of possible ADHD patterns (Swanson et al., 2001). Efforts to increase access to evidence-based treatment include linkage to a medication assessment service modeled on the Texas Children’s Medication Algorithm Project (Pliszka et al., 2000a, 2000b, 2006). A pilot project is also underway, in collaboration with schools, to increase the use of the Daily Behavioural Report Card (Center for Children and Families, 2009). Approaches to systematically addressing other problems are less developed.
This study offers a partial description of the prevalence and patterns of mental health and learning problems of a sample of children referred to a school mental health service. This knowledge may inform the prioritization for service refinement for this particular program and others that share a similar referral pattern. In addition to the need for replication and expansion of this type of inquiry, there are other priority research areas. In particular there is a need to determine the pattern of school-based child mental health services in Canada and the extent to which evidence-based practices are employed within such services and to what effect.
The second author received financial support through personnel awards from the Alberta Heritage Foundation for Medical Research and the Canadian Institutes of Health Research. The authors have no financial relationships to disclose.