Is the prevalence of major depression increasing in the Canadian adolescent population? Assessing trends from 2000 to 2014

Highlights

• The prevalence of MDE in adolescents has not changed over the past 15 years.

• An increase in reported diagnosis of mood disorders was observed from 2003 to 2014.

• Current strategies have not been effective at reducing MDE prevalence.

• Need for mental health services may increase with the rise in reported diagnoses.

Abstract

Background

Major depressive disorder is a relatively common diagnosis with onset across the lifespan. There is a recent belief that major depressive episodes (MDE) are increasing in adolescence; however, it is not clear if this is truly an increase in prevalence or reflective of other causes such as change in diagnostic patterns. This study aimed to determine whether evidence supports an epidemic of MDE in Canadian adolescents.

Methods

Past year MDE prevalence estimates were derived from a series of nationally representative surveys. Random effects meta-regression and graphical analyses were used to evaluate trends. A post hoc analysis compared trends in MDE prevalence to trends in self-reported mood disorder diagnosis (made by a health professional). The sample was split into 9 birth cohorts to examine whether MDE prevalence increased in more recent cohorts.

Results

Prevalence of MDE did not significantly change between 2000 and 2014 (β=0.001; p=0.532), and there was no modification of trends by sex or age. However, prevalence of self-reported mood disorder diagnosis by a health professional increased from 2003 to 2014 (β=0.001; p=0.024). There was no indication that MDE prevalence differed by birth cohort.

Limitations

Limitations include reduced precision in subgroup analyses, lack of clinical judgement in the structured diagnostic interview, and inability to differentiate mild, moderate and severe episodes of depression.

Conclusion

These findings do not support an epidemic of MDE in adolescents, however as more individuals report diagnoses by a health professional, future policy may need to incorporate an increase in need of mental health services.

Introduction

Major depressive episodes (MDE) are a characteristic feature of major depressive disorder, believed to be the second leading cause of disability worldwide (Ferrari et al., 2013). Development of a MDE during adolescence is critical, as early onset of depression can predict adverse mental health outcomes in adulthood, such as MDE recurrence, anxiety, substance use, and suicidal behaviours (Fergusson et al., 2007). The vulnerability of adolescents to MDE may result from a combination of genetic predisposition, age related changes, and psychosocial risk factors (i.e. peer conflict, bullying) that often occur during this period of development (Maughan et al., 2013). Further, symptoms of MDE are strongly associated with substance abuse and academic problems, and are risk factors for suicidal ideation and behaviour, which contribute to both morbidity and premature mortality in adolescents (Deykin et al., 1987, Goldston et al., 2009, Mayes et al., 2015, Owens et al., 2012).

In order to properly allocate resources in Canada, the prevalence of MDE in the adolescent population must be better understood. Canadian literature on trends in depression is limited (McMartin et al., 2014), and international reports are inconclusive. Some reports from Iceland, Finland and the United Kingdom demonstrate prevalence to be increasing in both males and females (Sigfusdottir et al., 2008, Sweeting et al., 2009, Torikka et al., 2014, Wijlaars et al., 2012), while additional reports specifically from England and the Netherlands support a decrease in males (Collishaw et al., 2010, Tick et al., 2008), and reports from Canada and Norway exhibit no change at all (McMartin et al., 2014, von Soest and Wichstrom, 2014). While the evident inconsistency may result from cross national difference or methodological variation between studies, there were some common limitations that emerged. In many studies, minimal time points were included for comparison, which provided insufficient information to document a trend over time. It was also common for different sampling or measurement strategies to be used over time, making it difficult to compare information. Finally, non-specific measures of depression (i.e. composite measures, distress scales) were often used, which did not address standardized diagnostic criteria. In order to properly evaluate trends, data must be collected at multiple time points, using comparable methods at each time point, and all samples must be representative of the same population. Assessment of trends in adolescent depression would also benefit from the use of a validated measure that is specific to MDE.

Another way to assess change over time is to evaluate past year MDE prevalence over subsequent birth cohorts. A birth cohort effect can be conceptualized as variation in the risk of disorder according to the year an individual is born, often coinciding with a shift in population exposure to risk factors over time (Keyes et al., 2010). This approach was popularized in the 20th century (Kessler et al., 1994), and has changed substantially since then, incorporating various statistical techniques (Keyes et al., 2010). Rather than attempting to disentangle the effect of birth cohort from age and period effects, this can be explored visually to examine prevalence of MDE in different birth cohorts as they age.

The Canadian Community Health Survey (CCHS) is a series of nationally representative health questionnaires that collect data on the general health of the Canadian household population. The sampling frames and interview strategies remained largely consistent each cycle, which provides a valuable opportunity to compare data over time (Statistics Canada, 2015a). The aim of this study was to assess the existence of trends in past year prevalence of MDE using nationally representative samples of Canadian adolescents between 2000 and 2014.