Social Psychiatry and Psychiatric Epidemiology

, Volume 42, Issue 12, pp 945–952

Adult and peer involvement in help-seeking for depression in adolescent population

A two-year follow-up in Finland

Authors

    • Tampere School of Public HealthUniversity of Tampere
  • Mauri Marttunen
    • Dept. of PsychiatryKuopio University Hospital and University of Kuopio
    • Dept. of Mental Health and Alcohol ResearchNational Public Health Institute
    • Dept. of Adolescent Psychiatry, Hospital for Children and AdolescentsHelsinki University Hospital
  • Mirjami Pelkonen
    • Dept. of Mental Health and Alcohol ResearchNational Public Health Institute
    • Dept. of Adolescent Psychiatry, Hospital for Children and AdolescentsHelsinki University Hospital
  • Bettina von der Pahlen
    • Dept. of Mental Health and Alcohol ResearchNational Public Health Institute
    • Åbo Akademi University
  • Riittakerttu Kaltiala-Heino
    • Tampere School of Public HealthUniversity of Tampere
    • Psychiatric Treatment and Research Unit for Adolescent Intensive Care (EVA)Tampere University Hospital
ORIGINAL PAPER

DOI: 10.1007/s00127-007-0254-4

Cite this article as:
Fröjd, S., Marttunen, M., Pelkonen, M. et al. Soc Psychiat Epidemiol (2007) 42: 945. doi:10.1007/s00127-007-0254-4

Abstract

Help-seeking among adolescents and adults is often perceived as a process involving only the health care system and the individual suffering from psychiatric symptoms. The present study attempted to create a model of help-seeking also including members of adolescents social networks. A prospective follow-up design included a school-based survey on all ninth grade students (mean age 15.5) during the academic year 2002–2003 in two Finnish towns (N = 3,278; response rate 94%); and a follow-up conducted two years later (N = 2,080; response rate 63%). The respondents were assessed for current depression and help-seeking for depression and other mental health problems each time. In the follow-up they were asked if they felt that they need help for depression and if their mother, father, sibling, peers, boy- or girl-friend or teacher had been worried about changes in their mood or behaviour. The associations of perceived need for help, help-seeking behaviour and concerns of different people, with depression at baseline were studied. Stepwise logistic regression models were computed to find the best predictors for help-seeking for depression. One third of the adolescents meeting R-BDI-13 criteria for depression at baseline still perceived a need for help for depression two years later but only a fifth of them had sought professional help. Depression at baseline was significantly associated with concerns about changes in mental health or behaviour among parents and significant others; and the concerns of mother, peers and teacher were among the best predictors of recent help-seeking for depression. To ensure adequate intervention for depressed adolescents without a social network capable of prompting referral, routine screening for depression should be applied in primary health care services and specialized services for adolescents.

Key words

adolescent health services/utilizationfamilymental health services/utilizationhealth services needs and demand

Introduction

With approximately 10% of adolescents currently suffering from moderate to severe depression and up to 25% of adolescents experiencing a depressive episode during adolescence [1, 2], depression can be considered a major public health issue among adolescents. Research on adults has clearly suggested that the origins of many psychiatric disorders are in childhood or adolescence [3, 4]. Treatment of childhood disorders may thus be preventive of adult mental disorders. Adolescence is a developmental stage when the individual prepares for adulthood. Depression may severely disrupt this development. Despite current impairment, potentially deleterious effect on future career and social roles, and relatively high recurrence rates associated with adolescent depression, it is a commonly under-treated condition [1, 5, 6, 7].

Several aspects for child, adolescent, and young adult help-seeking behaviour have been addressed in the literature. Some are characteristics of the individual or the illness; others are related to the health care system. Type and severity of the disorder and impairment have been suggested to be significantly associated with adolescent help-seeking [8, 9, 10, 11]. Since less than half of adolescents with major depression seek treatment by the age of 18 [1, 12, 13], factors other than the presence of a disorder, or impairment [14] can be hypothesised to be significantly associated with help-seeking.

In voluntary treatment settings a depressed adolescent or his/her parent must perceive a need for intervention in order to contact mental health services. Young people actually seeking or receiving help are a minority of those perceiving a need for help [11, 15]. Thus, perceived need is an essential but not sufficient predictor of help-seeking.

Several types of perceived barriers to care have been identified in the literature. Among adolescents these include economic disadvantage, concerns about confidentiality and stigma, perceiving professionals as being inaccessible, willingness to solve the problem unaided and unfavourable beliefs about mental illness and treatment [16, 17, 18]. In Finland, the health care system is publicly funded. Primary mental health services for children and adolescents are available free of charge at schools and other educational institutions and also community services for adults are free of charge. Also the costs of specialist level services are covered by the municipalities except for a nominal entrance fee for some services. The public is well aware of the funding of health and social services. Thus, financial barriers to access of services are unlikely. No association between socioeconomic status and mental health service use was found in a Finnish population study on children and adolescents [9]. Other barriers to Finnish adolescent mental health services have not been addressed in the literature.

Among children, recognition of mental health problems by parents and/or a teacher is suggested to be of importance [19, 20, 21, 22, 23]. During adolescence the problem recognition of the adolescent him/herself may be hypothesized to become more important predictor of help-seeking, although parental perception of mental health problems has been suggested to be more strongly associated with mental health service use than the perception of adolescents as old as 16 [9].

Many studies on help-seeking behaviour of children and adolescents have used cross-sectional data, been conducted with children or preadolescents or have covered a wide age range from childhood to adolescence. Help-seeking patterns, however, are suggested to change during adolescence [24, 25]. Informal help-seeking may shift with age from parents to peers [26]. The proportion of perceived social support received from parents and peers may also change in the course of adolescence [27]. An interesting question is whether these changes will also be reflected in the roles of parents and peers in help-seeking for depression. In order to study the significance of different actors in the adolescent’s social network in recognising problems in mental health and initiating help-seeking for depression, we conducted a follow-up study on Finnish adolescents.

The aim of the study was to answer the following questions:
  1. 1.

    What is the proportion of depressed middle adolescents perceiving a need for professional help?

     
  2. 2.

    What is the proportion of depressed adolescents reporting having sought professional help for their ailment?

     
  3. 3.

    Are the concerns about changes in adolescent mood or behaviour among parents or other people close to the adolescents associated with adolescent depression?

     
  4. 4.

    Are the concerns about changes in adolescent mood or behaviour among parents or other people close to the adolescent associated with formal help-seeking for depression?

     

Methods

Procedures

The present study is a part of an ongoing prospective follow-up study entitled “Adolescent Mental Health Cohort Study”. Ninth grade students (aged 15–16 years) of all the Finnish-speaking secondary schools in two Finnish cities, Tampere (200,000 inhabitants) and Vantaa (180,000 inhabitants) filled in a person-identifiable questionnaire during a school lesson supervised by a teacher. Students absent from school received the questionnaires by post. The final T1 sample consisted of 1,609 girls and 1,669 boys whose mean age was 15.5 (SD 0.39). The response rate was 94%. The procedures of baseline (T1) data collection are described in detail elsewhere [28].

The subjects responding to the first survey were reached for a two-year follow-up (T2) through their current educational institutes, by post and finally through the Internet. A total of 2,082 responses were received. Two subjects responded both at the educational institute and on the Internet. For these two, the answers given through the internet were excluded from the data. Ten responses were excluded as they were judged by the researchers to have been completed facetiously. The final T2 sample represents 63% of the adolescents having responded to the first survey, and consisted of 2,070 respondents, whose mean age was 17.6 (SD 0.41). Of the responses in the final data, 54% were given at survey occasions organized by educational institutes, 44% by post and 2% via the Internet.

Measures

Depression

A Finnish modification [29] of the 13-item Beck Depression Inventory [30, 31] was used to assess depression in both surveys (T1 and T2). The Finnish version of the BDI-13 (R-BDI-13) has shown good psychometric properties in a large population sample of adolescents [32]. The 13 items indicate feelings, cognitions, and physical symptoms related to depression and were scored 0–3 (“3” indicating the highest severity). The scores of all 13 items were computed into a sum-score (range 0–39 points) in which a score of 8 or more points indicates moderate to severe depression [30]. This cutoff point was used to define depression in the present study.

Perceived need for help

At T2 the adolescents were asked if they felt that they needed help for (a) depression (yes/no), (b) for another mental health problem (yes/no). These items have been used extensively in school surveys of Finnish adolescents. Both items of perceived need for help were used for the analyses, since the adolescents feeling a need for help may not have known whether they were suffering from depression or from another mental health problem.

Others concerned about changes in the adolescent’s mood or behaviour

The adolescents were asked if people close to them had been concerned about changes in their mood or behaviour (separately: mother, father, sibling, peers, boy/girl-friend, teacher; yes/no). The respondents were instructed to complete every item even if they did not think that there had been any reason for concern.

Help-seeking for depression and other mental health problems

At T2 the adolescents were asked if they had during the past two years sought professional help (“for example from the school health services, primary health care, family counselling services, or adolescent psychiatric services”) for (a) depression (yes/no), (b) another mental health problem (yes, no). Help-seeking for other mental health problems was controlled for in the final model, since adolescents may not have recognised the nature of their mental health problems, and also because help-seeking for other mental health problems may enhance help-seeking for depression (comorbidity, good experience of help, propensity to seek help).

Family structure

At T2 the adolescents were asked about their home family structure. Responses were coded into three categories: “both parents”, “single or stepparent family”, and “apart from parents”. This final category included living alone, with a spouse or friend, or a legal guardian.

Socioeconomic variables

At T1, the educational levels of the parents were elicited with two analogous items: “What is the highest educational level obtained by your father (mother)”. The response alternatives were “Comprehensive school only”, “Comprehensive school with vocational education”, “Secondary school with/without vocational education”, and “Academic degree”. The alternative “comprehensive school only” was selected to indicate low educational level both in mothers and fathers.

Statistical analyses

Frequencies of depression at T1 and T2, perceived need for help, help-seeking behaviour, others concerned about changes in the adolescent’s mood or behaviour, family structure, socioeconomic variables, and sex are given. Crosstabulations with chi square statistics were computed in order to investigate associations of perceived need for help, previous and recent help-seeking behaviour, and concerns of others about changes in the adolescent’s mood or behaviour, with depression at T1. The percentage of adolescents seeking help for depression when perceiving a need for help for depression was also computed.

Variables assessing depression, perceived need for help, help-seeking for other mental health problems, concerns of others about changes in the adolescent’s mood or behaviour, family structure, socioeconomic variables, sex and age were first entered into a logistic regression equation one at a time to calculate the unadjusted risk (OR with 95% CI) for help-seeking for depression associated with them. Next, a stepwise logistic regression model was computed testing the significance of perceiving the need for help when sociodemographic background and seeking help for other mental health problems were controlled for (model 1). Finally, to test the significance of people close to the adolescent in help-seeking for depression, variables indicating concerns of other people were added into the stepwise logistic regression model (model 2). Forward LR statistics was applied to deal with multicollinearity. The final model was also tested with enter and backward LR procedures. Nagelkerke R Square statistics was used to ascertain the percentage of probability of help-seeking for depression at T2 explained by the stepwise models.

Attrition

Respondents differed significantly (P < 0.005) from non-respondents. According to the T1 survey, non-responders at T2 survey were more commonly males (63% vs. 44%), came less often from families with both parents (65% vs. 72%), and more often from families with low parental educational levels (low educational level of the father: 22% vs. 16%; of the mother: 18% vs. 13%). A larger proportion of the drop-outs were depressed (12% vs. 9%).

Results

Characteristics of the sample

At both T1 and T2, 9% of adolescent participants met R-BDI-13 criteria for depression. At T2, perceiving a need for help for depression was more common than perceiving a need for help for other mental health problems. The percentage of adolescents seeking help for depression and also for other mental health problems was larger at T2 than at T1. Of the people close to the adolescents, the one most commonly worried about changes in the mood or behaviour of the adolescent was the mother, followed by peers. (Table 1).
Table 1

Frequencies (n,%) of sex, family structure, depression, perceived need for help, help-seeking, concerns of others about the adolescent mood or behaviour, and proportion of missing data among 17–18-year-old adolescents responding to a follow-up survey on mental health

Characteristics of the sample

n

%

% missing

Depression

  At T1

188

9

0.5

  At T2

177

9

0.5

Perceived need for help (T2)

  Thinks that needs help for depression

166

8

0.8

  Thinks that needs help for other mental health problem

100

5

1.4

Previous help-seeking behaviour (T1)

  Sought professional help for depression

101

5

1.5

  Sought professional help for other mental health problem

82

4

2.0

Recent help-seeking behaviour (T2)

  Sought professional help for depression

140

7

0.7

  Sought professional help for other mental health problem

169

8

1.4

Others concerned about changes in mood or behaviour (T2)

  Mother

485

24

3.4

  Father

231

12

4.2

  Sibling

160

8

5.0

  Peers

314

16

4.3

  Boy/girl-friend

261

13

6.1

  Teacher

120

6

4.8

Family structure (T2)

  

1.0

  Both parents

1279

62

 

  Single or stepparent family

628

31

 

  Apart from parents

142

7

 

Socioeconomic variables (T1)

  Low educational level of the father

312

16

8.5

  Low educational level of the mother

252

13

7.5

Sex

  

0.0

  Male

903

44

 

  Female

1167

56

 

Associations of perceived need for help, help-seeking behaviour, and concerns of others with depression at T1

Of the adolescents meeting R-BDI-13 criteria for depression at T1, 41% met the criteria two years later. Approximately one third of the adolescents meeting R-BDI-13 criteria for depression at T1 still perceived a need for help for depression at T2 but only a fifth had sought professional help for it in the past two years. Twenty per cent of the adolescents depressed at T1 had sought help for depression some time in the past. The parents and other people close to the adolescent were more commonly concerned about changes in the mood or behaviour of adolescents meeting R-BDI-13 criteria for depression at T1 than of adolescents not meeting the criteria. (Table 2).
Table 2

Percentages of perceived need for help, help-seeking behaviour, and concerns of others in 17–18-year-old adolescents with and without depression at T1

 

Depressed at T1

Yes

No

P

Previous help-seeking behaviour (T1)

  Sought professional help for depression

20

3

<0.001

  Sought professional help for other mental health problem

12

3

<0.001

Depressed at T2

41

5

<0.001

Perceived need for help (T2)

  For depression

28

6

<0.001

  For other mental health problem

12

4

<0.001

Recent help-seeking behaviour (T2)

  Sought professional help for depression

23

5

<0.001

  Sought professional help for other mental health problem

9

3

<0.001

Others concerned about changes in mood or behaviour (T2)

  Mother

41

22

<0.001

  Father

22

11

<0.001

  Sibling

18

7

<0.001

  Peers

30

14

<0.001

  Boy/girl-friend

30

12

<0.001

  Teacher

15

5

<0.001

Predictors of recent help-seeking for depression

Of the adolescents perceiving a need for help for depression, 39% had sought help for it in the past two years. When alone in logistic regression depression, perceived need of help, help-seeking for other mental health problems, concern of the people close to the adolescent, living apart from the parents, sex and age were associated with help-seeking for depression, while parental educational levels did not. Living with both parents was associated with not seeking help for depression. Perceiving a need for help for depression (T2) and having recently sought professional help for some other mental health problem were associated with an approximately 15-fold probability of seeking help for depression. (Table 3, unadjusted) In model 1 depression at T1, perceived need for help for depression (T2), recent help-seeking for other mental health problems, living apart from parents and female sex were associated with help-seeking for depression. This model explained 33% of the probability of recent help-seeking for depression. (Table 3, model 1).
Table 3

Help seeking for depression among Finnish adolescents (ORs with 95% CI)

 

Unadjusted

Model 1

Model 2

OR

95% CI

OR

95% CI

OR

95% CI

Depression

  At T1

5.4

3.6–8.1

2.7

1.6–4.7

2.2

1.2–3.9

Perceived need for help (T2)

  For depression

14.9

10.1–22.0

8.7

5.2–14.3

6.0

3.5–10.3

  For other mental health problem

7.3

4.5–11.7

ns

 

ns

 

Recent help-seeking for other mental health problems (T2)

  Sought professional help for other mental health problem

14.9

9.2–23.9

19.0

10.1–35.8

9.8

4.8–20.1

Others concerned about changes in mood or behaviour (T2)

  Mother

7.1

4.9–10.3

not in the model

2.1

1.3–3.5

  Father

6.3

4.3–9.2

not in the model

ns

 

  Sibling

6.6

4.4–9.9

not in the model

ns

 

  Peers

6.9

4.8–9.8

not in the model

2.2

1.3–3.7

  Boy/girl-friend

5.1

3.5–7.5

not in the model

ns

 

  Teacher

10.4

6.8–16.0

not in the model

3.5

1.8–6.5

Family structure (T2)

  Both parents

1.00

 

1.00

 

1.00

 

  Single or stepparent family

1.5

1.0–2.3

1.3

0.8–2.2

ns

 

  Apart from parents

4.9

3.0–7.9

2.3

1.2–4.4

ns

 

Socioeconomic variables (T1)

  Low educational level of the father

1.5

0.9–2.3

ns

 

ns

 

  Low educational level of the mother

0.7

0.4–1.2

ns

 

ns

 

Sex

  Female

3.6

2.3–5.6

2.4

1.4–4.1

2.3

1.3–4.0

Age (continuous)

1.7

1.2–2.5

 

ns

ns

 

Unadjusted probability of seeking help for depression during the past two years associated with depression at the age of 15–16 years and currently perceived need for help, help-seeking for other mental health problems, concerns of people close to the adolescent, demographic variables and socioeconomic background. Model 1: Best predictors of help-seeking in a stepwise logistic regression model including depression, perceived need for help, help-seeking for other mental health problems, demographic variables and socioeconomic background. Model 2: Best predictors of help-seeking for depression in a stepwise logistic regression model including also concerns of people close to the adolescent (model 2)

Adding concerns of the people close to the adolescents resulted in family structure loosing significance. Concerns of the mother, peers and the teacher were significantly associated with help-seeking for depression. Model 2 explained 38% of the probability to seek help for depression. (Table 3, model 2).

Discussion

Of the adolescents depressed at T1, 41% were still or again depressed two years later. At T2, only 23% of the adolescents meeting R-BDI-13 criteria of depression at baseline reported having sought professional help for depression during the past two years. The proportion is similar to that for help-seeking adolescents having an episode of DSM-major depression, in population samples of middle to late adolescents in Finland [33, 34] and elsewhere [24, 35].

Of the adolescents meeting R-BDI-13 criteria for depression, one third reported a perceived need for help for it two years later. The finding that a minority (39%) of the adolescents perceiving a need for help for depression had actually sought help in the past two years was in accordance with earlier reports on community adolescents [36] and adults [37]. Perceiving a need for help was, nevertheless, one of the strongest variables associated with help-seeking in the present study.

Meeting R-BDI-13 criteria for depression at baseline was significantly associated with reporting that parents and significant others (siblings, peers, boy/girl-friends, teacher) had been concerned about changes in the adolescent’s mood or behaviour at follow-up. This may indicate that people close to the adolescent had recognised the adolescent’s current depression or residuals of a previous episode. The percentage of mothers concerned about the changes in the adolescent’s mood or behaviour in the present study is consistent with earlier findings among younger adolescents: more than 60% of American parents (mainly biological mothers) of depressed 9–13-year-olds reported that their child “has a problem” or “needs help” [20]. The effect of recognising the problem or being worried about a depressed peer, on help-seeking initiation has, to the best of our knowledge, not been previously addressed in the literature. Adolescents have, however, been reported to be able to recognise well the most obvious symptoms of depression (suicidality, sense of worthlessness) and to indicate concern for a depressed peer presented in a vignette [38].

The concerns of parents and significant others were also associated with seeking professional help for depression in the present study. Even when depression, perceived need for help, and sociodemographic variables were controlled for, the concerns of mother, peers and teacher were associated with greater probability of recent help-seeking for depression. As discussed earlier, parental and teacher recognition of mental health problems have previously been found to be associated with service use [22, 39]. The present study attempted to create an even more comprehensive model of help-seeking by simultaneously studying the significance of perceived need and concerns of parents and significant others, including the peers and intimate partners of the adolescents. The explanatory power of the statistical model for help-seeking was enhanced with adding the concerns of people close to the adolescent, which suggests that the social network may play an important role in the help-seeking process.

In univariate analysis depression at baseline was associated with more than a five-fold increase in probability of seeking professional help. Adding perceived need for help and controlling for sociodemographic background in a stepwise logistic regression model decreased the odds ratio considerably, and adding concerns of people close to the adolescent to the model decreased the odds ratio further. Adolescent help-seeking was more significantly associated with perceived need for help than with symptoms of depression. Concerns of the people close to the adolescent were found to be equally significant to the symptoms of depression. The help-seeking process seems, thus, to include negotiations about the meaning of the symptoms experienced. These negotiations may concern not only the adolescent but also his/her social network.

Help-seeking for other mental health problems considerably increased the probability of help-seeking for depression. This finding suggests that young people already in contact with mental health services have an advantage in receiving treatment for relapses or incidents of new mental health problems. It may also be an indication of the significance of comorbidity in enhancing help-seeking reported also in previous studies of adolescents [33].

Of the adolescents with no depression at T1, 5% reported help-seeking for depression during the past two years. This indicates little amount of new incidences, also reported in other longitudinal studies on adolescents [2], or help-seeking for mild depression or other problems related to adolescent development.

According to the data gathered at T1, the drop-outs fared worse than the responders. The results should, thus, be interpreted carefully in populations at high risk of mental health problems. Both logistic regression methods (backward and forward LR) tested yielded similar models, which is a strong indication of the models robustness.

The Finnish version of R-BDI-13 is a tool validated for adolescent populations. It cannot be used to produce a DSM diagnosis of major depression but is probably robust enough to be used for identifying the subjects most probably in need of an intervention. Self-reported internalising and externalising problems in adolescence have been reported to be associated with DSM diagnoses in follow-up [40]. Items assessing “perceived need for help”, and “help-seeking for depression” have previously been used in national surveys on adolescents. The concerns about changes in an adolescent’s mood or behaviour among parents and significant others were elicited with items tailored specifically for this survey. With adolescents as the sole informants we cannot determine whether the parents and significant others actually were concerned or not. Minor concerns may not have been noticed by the adolescent. Finally, our results may have been affected by the stigma associated with mental health services, resulting in a reluctance to report such contacts. On the other hand, the independent variables such as the perceived need for help for mental health problems and concerns of others would also probably have been similarly affected. Thus, the hypothesised non-response would not have affected the association between the explanatory variables and the outcome.

Conclusion

Access to services is important but not sufficient in ensuring adequate intervention for adolescents with emotional problems. Depression may affect the decision-making capabilities of adolescents otherwise mature and independent. Self recrimination associated with depression may also lead the adolescent to consider the behavioural changes as shortcomings in personal characteristics, rather than as signs of problems in mental health. Parents, teachers and other adults should thus not leave young people alone to cope with mental health problems. The mental health literacy of both adolescents and adults must be enhanced to achieve better awareness of the effectiveness of treatment, symptoms deserving psychiatric evaluation and the fact that adolescent depression is not necessarily a transient phenomenon. To guarantee adequate intervention for depressed adolescents without a social network capable of prompting referral, routine screening for depression should be applied in primary health care services and specialised services for adolescents.

Copyright information

© Springer-Verlag 2007