Distribution and frequency of categories
In this study, Assert was implemented in a primary mental health service context that focuses on prevention, early intervention and counselling for less severe adolescent mental illnesses. Although we used an idiographic measure, there were similarities among the concerns presented by the adolescents.
Notably, adolescents most frequently reported concerns related to Self, followed by concerns about Life domains and, lastly, concerns about Relationships. Concerns relating to the sub-category Autonomy (under the main category Self) were clearly the most frequent. Within the main category Life domains, concerns about school recurred most often. Concerns about Improving relationships were the most frequent sub-category pertaining to Relationships. About one of four concerns relating to Self were sub-categorised as apparent mental health issues, equalling over one in ten of all concerns presented by the adolescents.
The high frequency of concerns within the sub-category Autonomy indicates adolescents’ need to be more in control of their lives and their desire to better regulate their emotions, assert themselves, trust in their capabilities and be more independent. As most standardised measures are designed to assess diagnostic criteria or symptoms of mental health impairment [27], however, it is reasonable to assume that these issues would not have been detected by other types of assessment that are commonly conducted in primary mental health care settings.
The relative low occurrence of concerns directly related to mental health provides evidence that the sample was sub-clinical, displaying mental health symptoms below the clinical cut-off, yet with a higher quality of life than outpatient populations [34]. However, the dimensions under the mental health sub-category show that certain internalising symptoms were prevalent, which is in compliance with other studies of similar samples (e.g. 48). Symptoms of internalising disorders often have their onset in adolescence, and early detection is a key to prevention. Even if their symptom levels did not warrant a diagnosis, all adolescents in this sample had a subjective need for help and chose to contact a counsellor. The adolescents were using their own words to describe issues impacting their wellbeing, and these words are not necessarily equivalent to the concepts professionals use to describe mental illness. Thus, independent of adolescents’ objective symptoms of mental illness, it would still be meaningful to measure the alleviation of subjective concerns that the adolescents decide are important for their wellbeing.
Most of the concerns around Relationships were about improving existing relationships; these concerns were similarly distributed between friends and family. It has been well documented that both family and peer relationships impact adolescent health and wellbeing (e.g. 49, 50). Concerns about Relationships may reflect that, during adolescence, individuals may begin to acknowledge that the quality of relationships is also their responsibility. For example, spending more time without their parents and changing schools pushes adolescents to take responsibility for maintaining—and sometimes redefining—old relationships, while seeking out and forming new ones.
Another noteworthy sub-category within the Relationship category was the need to feel safe around people the adolescents are close to. The majority of concerns in this area related to family relationships. Discovering potential violence or abuse at home is crucial to the security and wellbeing of the adolescent. Using idiographic measures to supplement therapeutic interviewing, in the detection and follow-up of these problems, could be beneficial. However, in the further assessment and handling of situations of this nature, one should not rely on idiographic measures alone, but should also implement appropriate efforts immediately.
Within the category Life Domains, concerns regarding School were the most frequent. Twenty-five of 40 school related concerns were about improving performance. Increased pressure to perform in school can be stressful, and stress and demands are related to mental and physical health complaints [21]. In addition, improving performance in the workplace—or in leisure activities like sports or keeping fit—was a frequently reported concern. The remaining concerns around Life domains were related to motivation and wellbeing, both of which are central to performance [51,52,53] and should therefore be seen in conjunction.
The results show that boys more frequently reported concerns about the Life domain than did girls. As this category includes concerns about School, it is possible that boys tend to have more difficulties with academic performance than girls [54, 55]. Admittedly, the sample includes a minority of males, so no firm conclusions should be drawn from these results. Moreover, age differences were found in a number of concerns related to Self and Relationships. Younger adolescents more often reported no concerns about Self, than did older adolescents. Older adolescents have a greater capacity for introspection and are more able to imagine their future identity and life choices—which could make them more concerned about who they are and who they will become [56, 57]. The opposite was seen for age differences in number of concerns around Relationships; older adolescents more often reported no concerns about Relationships than younger adolescents. Conflicts in the parent–child relationship may temporarily increase in early adolescence [58]. In addition, there are two school transitions for Norwegian adolescents in early adolescence. These often involve establishing new friendships and adapting to new social contexts. In contrast, older adolescents may have achieved more mature relationships with their parents and may have friendships that are more established. Still, since the analyses involved dividing a small sample into two groups, putting too much emphasis on these results could be misleading.
Some adolescent concerns were presumably not reported. For example, no adolescent reported concerns about social media, even though social media is widely important to many adolescents and has an influence on adolescents’ mental health (e.g. 59). One hypothesis could be that social media platforms in themselves do not feel very important to the individual adolescent—but rather the consequences of using them, such as a negative self-image [60, 61] or relationship concerns, which were reported by several adolescents in this study.
The relatively low level of concerns around Relationships, in this sample, is also noteworthy. Forming new relationships, both romantic and platonic—as well as renegotiating the relationship with one’s parents—is commonly assumed to be a large part of adolescence. In this sample, however, the adolescents reported two concerns about Self for every concern about Relationships. This could possibly be explained by the primary mental health care setting, which invites more concerns related to oneself and one’s wellbeing. Also, issues of self and identity are common among this age group [14]. The service setting could also explain why none of the adolescents reported larger issues, like worries about the environment or broader social or political concerns. It could be that the adolescents assumed that what mattered most to them had to be personal issues.
In summary, one could argue that a general tendency across all concerns is the wish for personal improvement, to feel better about oneself, to improve relationships and increase one’s school performance. This constant striving for improvement could potentially lead to feelings of insufficiency—that one is never good enough. Adolescence is a formative step in the human development, which is marked by leaps in social, cognitive and emotional capabilities, making it challenging to navigate through this phase. Despite increases in the adult population’s happiness and advances in health care services, the mental health of adolescents in the Western world has deteriorated over the last decades [16, 20]. Including youths' mental health concerns, as voiced by themselves, in low-threshold primary mental health care settings may contribute to limiting that deterioration.
Alleviation of concerns during counselling
While Assert is designed to systematically capture what is important to adolescents at each session and allow for changes, the utility of Assert may be diminished if adolescents change all of their concerns each session. However, changes in concerns were only found in a minority of the participants. When the adolescents define what matters to them in their own words, they are presumably closer to the core of the problem than if a counsellor defined their concerns. This was also something the counsellors experienced while using Assert [39]. Measures used to assess and track progress in counselling need to be dynamic and develop along with the adolescent. After using Assert, adolescents experienced that their mental health was less dependent on external factors [34]. Defining their concerns in each session may have increased their sense of control over what was happening in counselling.
The multilevel models provide preliminary support that Assert could be sensitive to changes in self-defined concerns over the course of counselling. The score on each concern on Assert showed a significant alleviation of the burden or distress throughout counselling. This was true for all three main categories, as well as for five out of nine sub-categories. Concerns about Self were significantly more relieved during counselling than concerns about Relationships. The counselling was most often individual, and relational concerns may be more challenging to address in this service setting. Alleviating concerns about Self may be easier to navigate without involving other parties. Yet the scores on the adolescents’ concerns about their Relationships and Life domains also showed a significant alleviation from start to end of counselling. The adolescents may have managed to change their perceptions regarding their role in relationships or their efforts to improve—in school, for example. However, the absence of a comparison group in this study, means that no firm conclusions about the sensitivity of Assert can be made from this study.
The number of times that Assert was used had a significant effect on the alleviation of the adolescents’ self-defined concerns. This was particularly true when Assert was used in seven sessions; in these cases, the alleviation was greater than when it was used in either fewer or more sessions. These results may suggest that the adolescents were not in need of long-term treatment. Still, as less alleviation was achieved when Assert was used fewer than seven times, it also indicates that they did need a moderate amount of counselling.
The adolescents in this study all contacted a counsellor due to a subjective need for help. When they defined the concerns that mattered to them, and when these were addressed consistently over time, they felt less bothered by their concerns within relatively few counselling sessions. However, the variability was large, especially in the higher part of the range, possibly affecting the results.
Limitations
The sample is eighty percent female. This is a common challenge in mental health research, as males are less prone to seek help for mental issues [62, 63]. Furthermore, the sample size is small, which also could limit generalisations about the adolescents’ concerns. Even though the number of participants was limited, the number of concerns that were used for the categorisation was fairly high. However, statistical analyses of smaller sub-categories and dimensions were challenging.
An obvious limitation was the lack of information regarding the counselling the adolescents received, as this could have shed light on differences in the alleviation of adolescent concerns according to the type of counselling received. Due to the high levels of variability between and within services—along with the study's naturalistic setting—it would have been too time-consuming and costly to collect the appropriate data to accommodate and control for the type of counselling, within the limits of this study.
Finally, it is possible that the alleviation of concerns could be due to regression towards the mean. Unfortunately, it is not possible to control for this with the available data, due to the design of this study. Thus, it cannot be ruled out that regression towards the mean accounted for at least some of the alleviation measured with Assert. Further, the aim of this study was not to examine the effect of Assert on alleviation of self-defined concerns, rather to examine if there was evidence that Assert was sensitive to changes in concerns over the course of counselling.