The 2014 Ontario Child Health Study

This issue of the Canadian Journal of Psychiatry features the first wave of results from the 2014 Ontario Child Health Study (OCHS). The OCHS is a large-scale epidemiologic survey that included 6537 families and 10,802 children aged 4 to 17 years. It represents an important advance in the current state of knowledge about child and youth mental health in Ontario, and it is very likely that many of the results can be generalized to other Canadian settings. The survey provides an opportunity for comparisons with an earlier iteration of the OCHS, conducted in 1983. To allow comparisons, the OCHS 2014 incorporated some of the same measures used in 1983. More importantly, however, the 2014 OCHS incorporates state-of-the-art research methods and, in terms of the sophistication of its methods, is one of the most advanced health surveys ever conducted in Canada. The series of articles presented in this issue does not stop at reporting high-quality population estimates. Rather, these results are extended further to look at social determinants (including area-based measures), health economics, and policy implications.

One of the advanced methodological features of the OCHS is an integration of data sources—for example, linkage of the survey data to area-level census information as well as data from government and health administrative data sources. The survey incorporates a school survey and collects data from families as well as the children themselves. The project was led by researchers at the Offord Centre for Child Studies at McMaster University but included contributions from leading academics at multiple institutions across the country, and from Statistics Canada, which implemented the protocol.

The sampling plan, based on the Canada Child Tax Benefit File, included deliberate oversampling of important subgroups as a means of better understanding these groups. The response rate, at 50.8%, is low by historical standards but reflects the current realities of conducting population surveys in Canada. Adjustments of sampling weights for nonresponse help to ensure that the risk of selection bias is greatly attenuated.

Another modern feature of the OCHS is the openness with which the data are managed. The data set is publicly available through Canada’s system of Regional Data Centres. Due to the complexity of the project, it will be a challenging data set for investigators to analyze. For example, replicate bootstrap weights must be used to ensure valid population estimates for some parameters (such as prevalence) but cannot be used with the multilevel modeling approaches that motivated aspects of the design of the study and represent one of its key opportunities. Specialized software may provide solutions, and the authors recommend careful consideration of these issues by researchers considering projects that may use the data.

The series of articles presented in this issue include an overview of methodology. ¹ Another article reports 6-month period prevalence estimates, ² whereas another focusses on changes that have occurred since 1983. ³ A series of additional manuscripts look at social determinants of mental health, through the lens of multilevel modeling, ⁴ the match-up of needs and expenditures within the health system, ⁵ and the issue of suicidality in this population. ⁶ In 2 final articles in this issue, implications of the OCHS for policy are explored. ^7,8

Each of these articles contains interesting, and sometimes unexpected, findings. They are too numerous to be listed here. Prevalence, for example, is found to depend on whether diagnoses are based on youth versus parent report, either or both. ² Schools were found to be the most common setting for mental health–related contacts with various professionals. Evidence of poor access to treatment for mood and anxiety disorders, especially in girls, is also presented. ² Immigrant children are another group that appears to encounter barriers in accessing health services.

Despite concerns about an “epidemic” of mental ill health among children and youth, another key finding is that the overall prevalence of mental disorders did not change between 1983 and 2014. ³ However, differences were observed in specific subgroups, driven by specific diagnoses, such as increased hyperactivity in 4- to 11-year-old boys and emotional disorders within the 12- to 16-year-old age group. These group-specific increases are consistent with the lack of an overall increase due to decreases in other categories, such as conduct disorder among 12- to 16-year-old boys.

The multilevel modeling results are complex due to statistical interactions in the data. ⁴ One of the interactions is intuitively easy to understand: in neighborhoods in which there is exposure to high levels of antisocial behaviour, household poverty has a larger impact on symptoms of externalizing disorders than in neighborhoods with lower levels of exposure. The other is more difficult to interpret: impoverished households seem to have lower externalizing symptoms when they are situated in high-poverty neighborhoods. These results emphasize the extent to which issues of person-context fit are important determinants of mental health. Comparing expenditures on child and youth mental health to identified imbalances should provide policy makers with just the kind of information they need to ensure better matching of expenditures and needs.⁵ A article on suicidality identifies determinants both of suicidal ideation and attempts—critical information for prevention. An important finding for clinical practice as well as policy is an important role for nonsuicidal self-injury as a risk factor for suicide attempts. ⁶

The paragraphs above merely scratch the surface of the wealth of policy-relevant data collected through the OCHS. In the spirit of integrated knowledge translation, a detailed discussion of policy implications of the OCHS findings is provided as a component of this issue. ^7,8 The OCHS is both an impressive example of state-of-the-art population research and a model for the kind of research capable of making a difference in the mental health of Canadian children and youth.