We need better data to understand student mental health now

Half of young people enter higher education. The progress of our society depends on the educational attainment and employment outcomes of these individuals. Governments are looking towards these university students to form part of the highly-skilled and healthy workforce to meet future technological and industrial needs. In recent years, general concern for university students’ mental health and wellbeing has risen substantially, which is reflected in universities investing more on their mental health and wellbeing support provision. However, it is unclear whether:

1. This general concern is substantiated for university students versus non-students.
2. University students’ life outcomes (education and employment) are affected by their mental health and wellbeing.
3. The investment on mental health and wellbeing interventions by universities will improve future mental health and wellbeing outcomes.

There is a real paucity of good data to guide strategic decision-making. For example, while the Adult Psychiatric Morbidity Survey indicated that in England, the proportion of 16 to 24-year-olds experiencing a common mental disorder rose from 15 to 19 per cent between 1993 and 2014, representing an increase of over 25 per cent, we do not know whether the increase in prevalence is comparable between young people in and out of higher education (McManus et al., 2016). We do know that the number of students declaring a mental disorder has increased five-fold over a 10-year period (Thorley, 2017), however, this increase has occurred alongside a massive campaign to encourage students to disclose mental health difficulties, thus we do not know the extent to which this reflects improvements in disclosure. Further, there is a lack of understanding about how mental health and wellbeing affects life outcomes of young people, particularly when coupled with their social characteristics. We know that poor mental health of young people is associated with poorer life outcomes, such as a significantly poorer academic outcome at university (Eisenberg et al., 2009), poorer mental health later in life (Essau et al., 2014), and poorer employment outcomes (Calderwood & Henderson, 2018; Clark et al., 2017). We also know that the social characteristics that represent the inequalities in society (Codiroli Mcmaster & Cook, 2018), such as gender, ethnic background and socioeconomic status (SES), affect a young person’s educational and employment outcomes (Anders, 2012; Archer et al., 2012; Hosein, 2018; Schoon, 2014), even when those young people have the same experience of attending university (HEFCE, 2015).

We also know that individuals from a lower SES, as well as those from a Black, Asian and minority ethnic (BAME) background, have reported lower wellbeing and are more likely to suffer from a mental disorder (Cosco et al., 2016; McLaughlin et al., 2012; Stevenson & Rao, 2014), and that women are also more likely to declare mental health problems (Thorley, 2017). The category of discipline that university students belong to can also affect their mental health and wellbeing. For example, in the US, a report commissioned by the University of California, Berkeley (Graduate Assembly, 2014), showed that over half of their postgraduate students in the Arts and Humanities, Biological Sciences, Physical Sciences and Engineering rated themselves as depressed in comparison to only a third of Business students. What we need is more evidence about how the interaction of these social characteristics (i.e. gender, SES, ethnicity) and discipline, together with their mental health and wellbeing, affects young people’s life outcomes.

Finally, universities are investing in a range of interventions and support mechanisms for their students to improve their mental health and wellbeing. However, more evidence is needed about whether these interventions will actually work for university students. We know that the interventions that universities are implementing are based on mental health and wellbeing evidence from the general population (Clark et al., 2018; Dolan et al., 2017). For example, activities to allow students to feel higher levels of belonging have better support structures during students’ transitions into higher education and engage them in wellbeing activities, such as sports (see for example, What Works Wellbeing, 2019). However, many of these interventions are at their early stages. It is uncertain whether improving students’ belonging, creating more supportive structures, or engaging them in wellbeing activities, will enhance students’ future mental health and wellbeing outcomes. Therefore, we need more evidence to understand how the intended outcomes from these activities can affect students’ mental health and wellbeing outcomes when taking into account their social characteristics, their discipline, and their university environment.

New methods: we need to use an intersectional approach to analysis

In researching the life outcomes of young people, researchers have generally made use of quantitative analyses of panel cohort studies and cross-sectional surveys. However, most of these quantitative studies on the life outcomes of young people have investigated a single identity or social category perspective (such as gender or ethnicity or SES), which means they fail to investigate how the combination of young people’s multiple identities can impact on their life outcomes. In other words, these studies have lacked an intersectional approach (Crenshaw, 1991), which until recently was primarily an approach utilised in qualitative studies. In an intersectional or ‘inter-categorical’ approach, young people’s life outcomes are treated as being dependent on their combination of identities (Codiroli Mcmaster & Cook, 2018). For example, Strand (2014) found in his intersectional study of school age children that lower SES children from both White British and Black Caribbean families made substantially lower educational progress than middle and upper SES White British children, but similar progress to Black Caribbean children with high SES. Therefore, without an intersectional approach, there may be a tendency to ascribe all outcomes to one social identity such as SES or ethnicity, rather than capturing the nuances that arise from holding several social identities (Codiroli Mcmaster & Cook, 2018).