Adolescence is associated with a marked increase in the onset of mental disorders (Solmi et al., 2022), of which depression is among the most prevalent globally (Lu, 2019; Wartberg et al., 2018). Research also indicates that there has been a notable increase in reported anxiety and depression among young people over the past decades, internationally (Bor et al., 2014; Collishaw & Sellers, 2020; Patalay & Gage, 2019) and in Ireland, where the current study is conducted (Dooley et al., 2019). Adolescent depression is linked to significant impairment, recurrence in adulthood, heightened risk of suicide, and increased all-cause mortality (Schubert et al., 2017; Shore et al., 2018). However, early identification and treatment may mitigate this impact (Davey & McGorry, 2019; Hetrick et al., 2008). Parents, given their key role within a young person’s microsystem (Bronfenbrenner, 1994), can play an important role in facilitating early intervention. Parents are well-placed to monitor risk, encourage alternative coping strategies, provide emotional support, and enable engagement with mental health services (Bouwman et al., 2018; Logan & King, 2001, 2002). They also frequently initiate treatment-seeking (Crouch et al., 2019; Hassett et al., 2018; Platell et al., 2023). Parents thus bear significant influence on a young person’s illness trajectory, and are an integral part of a young person’s wider system of care (Feinstein et al., 2009). Family members, especially parents, have also been identified by both young people, and those who work with young people, as a key target group for youth mental health and suicide prevention research (Sharma et al., 2021; Taylor et al., 2022). The importance of families and carers in supporting young people is also one of the key principles of the International Youth Mental Health Research Network (Mei et al., 2020).
Mental Health Literacy
For parents to identify symptoms and facilitate treatment-seeking for their adolescent children, they must possess the appropriate knowledge and skills – or “mental health literacy” (MHL). MHL was originally defined as the “knowledge and beliefs about mental disorders which aid their recognition, management or prevention… the ability to recognise specific disorders; knowing how to seek mental health information; knowledge of risk factors and causes; of self-treatments, and of professional help available; and attitudes that promote recognition and appropriate help-seeking” (Jorm et al., 1997, p. 182). More recent operationalisations of the concept have incorporated knowledge of mental health (as distinct to mental disorders) and help-seeking (Kutcher et al., 2016). Authors have also advocated for applying the concept of MHL to specific disorders, as this may be valuable for producing pragmatic, actionable targets for intervention (Johnco & Rapee, 2018; Singh et al., 2019). For example, assessing depression literacy (DL) may have more utility than an assessment of MHL more broadly conceived.
Yet, despite the evidently crucial role that parents play in supporting adolescents experiencing depression, and the clear importance of DL in underpinning parents’ responses to adolescents experiencing depression, little MHL or DL research has focused on parents of adolescents specifically. Indeed, two recent systematic reviews highlighted the limited evidence base on parental MHL, and reported that the existing research presents mixed findings on parents’ knowledge and understanding of mental health issues and help-seeking (Hurley et al., 2020; Johnson et al., 2023). Further investigation is therefore required to consolidate this evidence base. Moreover, the Hurley et al. (2020) systematic review, which focused primarily on parents of adolescents, reported that just three of these studies (Deitz et al., 2008; Hurley et al., 2018; Jeong et al., 2011) assessed parents’ knowledge of depression specifically (as opposed to mental illness more broadly). A more recent Chinese study, which involved the participation of parents of adolescents who had received a formal diagnosis of major depressive disorder, found that participants exhibited moderate levels of knowledge about adolescent depression (Gao et al., 2024). Given the fact that depression is one of the most prevalent mental health issues affecting adolescents (Lu, 2019; Wartberg et al., 2018), more research is required.
The present study therefore aims to address the gaps in the literature through investigating DL amongst parents of adolescents. By applying the concept of MHL to depression specifically, this study aims to increase understanding in a precise and practicable way. In line with calls for further research on the relationship between parents’ knowledge of mental health and their attitudes towards various help-seeking strategies (Hurley et al., 2020), the current study explores parents’ ability to identify depression, assessment of severity of symptoms, identification of disorder, recognition of symptoms, and assessment of prognosis (length of time until recovery), and beliefs about the helpfulness or harmfulness of different strategies. Furthermore, this study aims to not only describe parents’ DL but also to explore the associations between a range of key factors (specifically: gender, age, and empathy), and parents’ understanding of adolescent depression.
Gender and Parents’ Depression Literacy
Parents’ gender may influence their DL, as adult women have greater MHL than adult men in general (Angermeyer & Dietrich, 2006; Gibbons et al., 2015; Jorm et al., 2005; Swami et al., 2011), and also recognise mental health disorders in children more frequently than men (Pescosolido et al., 2008). Research specifically with adults who are parents has shown that fathers have poorer physical (Dey et al., 2015) and mental (Mendenhall & Frauenholtz, 2015) health literacy than mothers. In response to a vignette, female parents were more likely to indicate they would follow up a referral to a GP and specialised mental health services than male parents (Bignold & Anderson, 2023). In addition to the parent’s gender being associated with their DL, the extant literature suggests that the gender of their child may affect parents’ recognition of symptoms, as it may affect what kind of behaviours are seen as problematic or worrying. Parents have been found to respond to depressive symptoms differently whether these are expressed by boys or girls. For example, caregivers of boys with depression have been found more likely than caregivers of girls with depression to respond angrily to adolescent sadness (Shortt et al., 2016). Girls with depressive disorder have been shown to receive more punitive and minimising responses from fathers, in comparison to non-depressed girls (Shortt et al., 2016), and also report lower perceived paternal warmth and emotional availability than their non-depressed counterparts (Demidenko et al., 2014). Research has shown that mothers use emotion-related language, and discuss negative emotions and problems, more frequently with daughters than with sons (Saritas et al., 2013; Waller & Rose, 2010). However, adolescent boys are significantly less likely than girls to seek help for their mental health issues (Ando et al., 2018; Rice et al., 2018), meaning that, in order to engage with professional support, they may be more dependent upon parents’ early identification of distress and engagement with services. It thus appears that gender may be an important factor to examine when assessing parental DL, for several reasons. The current study aims to address this through analysing whether parents’ responses differed according to their gender, and/or their child’s gender.
Adolescents’ Age and Parents’ Depression Literacy
Research suggests that parents’ identification and understanding of their child’s symptoms of depression may also be impacted by their child’s age or developmental stage. Parents may be less likely to recognise certain kinds of behaviours as worrisome at different times in their child’s life. ‘Storm and stress’ during the pubertal phase of early adolescence is often perceived as normative, so certain symptoms may be considered by parents to be part of a phase (Bussing et al., 2003; Logan & King, 2001), or parents may have difficulty distinguishing what is normal from what is cause for concern (Boulter & Rickwood, 2014). A mother in Stapley et al.’s (2015, p. 623) study voiced such difficulties, stating ““… I don’t know if that’s a teenage thing or it’s a depression thing””. Indeed, a quarter of parents in the same study described “how they had wondered whether their child could simply be experiencing a more extreme version of the typical mood swings that adolescence brings, that their child could just be going through a phase that they would grow out of” (Stapley et al., 2015, p. 621). The prevalence of depressive disorders in adolescence increases with age (Bruckauf, 2017), such that the public may associate presentation of depressive symptoms with older rather than younger adolescents, or indeed believe that younger adolescents ‘don’t get depression’. This may be problematic as identification of early warning signs in young adolescents could facilitate an ‘upstream’ approach to intervention (Wyman, 2014), intervening before risk increases.
While rates of depression are higher among older adolescents, adolescence is a period of individuation, hallmarked by increasing independence from parents and family, and increasing dependence on friends (Bokhorst et al., 2010; Smetana, 2010). Parents of older adolescents have been found to be less supportive and more punitive towards their child’s emotional displays (Klimes‐Dougan et al., 2007). Parent-child communication involving older adolescents also tends to be lower in quality when compared with younger adolescents (Zhang et al., 2021), which may impact on parents’ ability to identify symptoms. Some research has also found a relationship between older age of the child and greater discrepancies between parental and child accounts of the child’s mental health (Braaten et al., 2001; Handwerk et al., 1999; Kiss et al., 2007), although others have not supported this trend (Baumgartner et al., 2020; Lauth et al., 2010; Logan & King, 2002). One study found that having older children was associated with parents’ mental health knowledge (Mendenhall & Frauenholtz, 2015), but mental health professionals report that they do not believe child age influences parents’ understanding (Frauenholtz et al., 2015). More research is therefore required to elucidate whether parents’ responses to the display of symptoms of depression are associated with their child’s age.
Empathy and Parents' Depression Literacy
Empathy is thought to incorporate both empathic emotions (e.g., sympathy, anger) or ‘state empathy’, as well as the cognitive adoption of another’s perspective, sometimes referred to as ‘trait empathy’ (Betancourt, 1990). Empathy has long been linked with mental health stigma – for example, lower ‘trait empathy’ has been reported to be associated with greater prejudice against people experiencing depression and ‘nondescript mental illness’ (Foster et al., 2018). In investigating the link between empathy and stigma (considered by some to be a component of MHL; Kutcher et al., 2016), attribution theory has highlighted perceived controllability of the symptoms as a key factor (e.g.,Siegel et al., 2012).
Parental empathy is considered a core facilitator of the parental drive to care for offspring (Mikulincer et al., 2001; Mikulincer et al., 2005; Stern et al., 2015). Neuroimaging research has revealed that the brain regions associated with empathy are dramatically reconfigured following pregnancy (Hoekzema et al., 2017). One study of mothers of adolescents demonstrated highly similar neural representations of their own distress and their child’s distress; in other words, mothers mentalise their child’s distress as if it were their own (Lee et al., 2017). With regards to the empathic emotions, anger has been reported qualitatively by parents of children who have experienced mental and physical illnesses (Bonner et al., 2006; Shpigner et al., 2013; Svensson et al., 2013), and parents of children experiencing eating disorders and post-traumatic stress disorders have also reported feeling sympathy for their child (Honey & Halse, 2005; Williamson et al., 2019). Perceived controllability was also assessed in one study of parent-adolescent dyads, which found that parents’ perceptions that their child is able to control their mental health problems was associated with lower self-stigma among adolescents (Moses, 2010).
However, the link between empathy and MHL more broadly has not yet been thoroughly investigated: to the best of the authors’ knowledge, to date just one study has been conducted, which demonstrated that adults with greater MHL had greater empathy (Furnham & Sjokvist, 2017). To the best of the author’s knowledge, no studies have examined the association between empathy and DL specifically. While adults’ empathy has been assessed in abundant studies (e.g., Ze et al., 2014), empathy research has less often specifically targeted parents. Moreover, the study of parental empathy has typically focused on parents of infants and young children (e.g., Barnett et al., 1980), and rarely with parents of adolescents. Thus, the relationship between parental empathy and DL requires further exploration, particularly among parents of adolescents.
Current Study
The present study builds on the current literature and addresses some of its existing gaps, by examining parents’ DL as it relates to their child. In addition, the study aims to investigate if parents’ DL varies according to their gender, their child’s gender, their child’s age, and their own empathy; to our knowledge, this study is the first to explore the link between empathy and DL in parents of adolescents. Also to the best of the authors’ knowledge, the study is also the first to examine DL amongst parents of adolescents in Ireland. In response to the presentation of a vignette stimuli describing an adolescent experiencing emotional and behavioural difficulties indicative of depression, the research questions were:
1.
Which symptoms do parents identify as indicative of the vignette character experiencing difficulties?
2.
How do parents label the cluster of depression symptoms displayed by the vignette character?
3.
How concerned are parents for the vignette character?
4.
How long do parents perceive it will take for the character to feel better?
5.
How controllable do parents believe the character’s symptoms are?
6.
How much anger and sympathy do parents feel towards the vignette character?
7.
How helpful or harmful do parents believe certain responses would be?
8.
Do parents’ responses relating to degree of concern, length of time until recovery, controllability of symptoms, anger, sympathy and the helpfulness/harmfulness of certain actions differ by parent gender, child gender, child age, or parent dispositional empathy?
Based on prior research, it is hypothesised that female parents, parents of female adolescents, parents of older adolescents, and parents with higher dispositional empathy will be more likely to label the symptoms ‘depression’, feel more concern for the vignette character, believe it will take longer for the character to feel better, perceive the characters’ symptoms to be less controllable, and feel less anger and feel more sympathy towards the character.
Method
Participants
Participants (n = 321) were adults who self-identified as primary caregivers of adolescents aged 12–18 years and resided in the Republic of Ireland. Participants ranged in age from 20 to 63 years (M = 46.2 years, SD = 6.2 years); 73 participants (22.7%) did not give their date of birth. Three-quarters (76%) were female caregivers of an adolescent child, with 24% stating that they were male caregivers. Specifically, 73.8% reported they were a mother of an adolescent, 15.9% reported they were a father, and 2.2% reported they had a different caregiving relationship (i.e., an older sister, foster or host mother); 8.1% did not specify their relationship. Most participants were born in Ireland (75.4%) and identified as White Irish (82.2%). Approximately half (45.8%) stated that their adolescent child identified as male, and 54.2% stated that their child identified as female. Participants’ children ranged in age from 12 to 18 years (M = 15.4 years, SD = 1.6). See Table 1 for a full overview of participants’ demographic characteristics.
Table 1
Descriptive data for participants’ age, gender, relationship to child, child’s gender, participants’ country of birth, ethnic background, and child’s primary female caregiver’s highest level of education completed to date
Variable | % | |
|---|---|---|
Age – M (SD) | 46.2 (6.2) | |
Gender | Male | 15.9 |
Female | 76.0 | |
Missing | 8.1 | |
Relationship to child | Mother | 73.8 |
Father | 15.9 | |
Other | 2.2 | |
Missing | 8.1 | |
Child’s gender | Male | 45.8 |
Female | 54.2 | |
Country of birth | Ireland | 75.4 |
United Kingdom (UK) | 8.4 | |
Europe (other than Ireland or UK) | 2.8 | |
Asia | 1.6 | |
North America | 0.6 | |
South America | 0.3 | |
Australia | 0.3 | |
Africa | 1.6 | |
Missing | 9.0 | |
Ethnic backgrounda | White Irish | 82.2 |
Irish Traveller | 0.3 | |
Any other White background | 5.6 | |
Black Irish | 0.3 | |
African | 0.6 | |
Any other Asian background | 1.6 | |
Other | 0.6 | |
Missing | 8.7 | |
Child’s primary female caregivers' highest level of education completed to date | Primary or less | 1.9 |
Junior Certificate | 10.9 | |
Leaving Certificate | 21.2 | |
Diploma/Certificate | 27.7 | |
Primary Degree | 15.6 | |
Postgraduate/Higher Degree | 14.0 | |
Missing | 8.7 | |
Procedure
Ethical approval for the study was sought and granted from the Trinity College Dublin School of Psychology Research Ethics Committee. Participants were recruited in two ways: through their child’s school, or online. Principals of 200 secondary schools across the Leinster region of Ireland were informed about the study by post. Fourteen schools agreed to facilitate the study. A total of 1,561 study packs were distributed across the 14 schools, which included an information sheet, consent form, and parent questionnaire. A total of 257 participants were recruited in this manner. The remaining 64 were recruited online via parent groups and snowball sampling through social media. Differences in demographic characteristics between school- and online-recruited participants were investigated to ensure that recruitment source was not a potential confounding variable. While online-recruited participants were significantly older than school-recruited participants and significantly more likely to identify as White Irish, this was not hypothesised to impact upon analyses.
School-recruited participants were provided with a gender-matched questionnaire by the school staff, while online-recruited participants were asked to indicate the gender of their child, which was used to apply ‘display logic’ for the version of the vignette with which they were presented. Participants were informed that participation was voluntary, and their responses were confidential. School-recruited participants provided written informed consent, and completed a pen-and-paper version of the questionnaire in their own time, at home. Completed questionnaires and signed consent forms were returned to school in a sealed envelope, and collected by researchers. Online-recruited participants indicated their informed consent via the online questionnaire, which they completed in their own time.
Measures
Vignette stimulus
The vignette stimulus can be found in Supplementary File 1. This vignette was adapted from vignettes originally developed and validated by O’Driscoll (2013), see O’Driscoll et al. (2012; 2015a; 2015b), and further validated by Silke (2016). The vignette describes a young person’s behaviour; in the text, six of the nine DSM-IV (American Psychiatric Association, 2000) criteria for depression are mentioned. The vignette character was gender-matched to the participant’s adolescent child (i.e., parents of male children read about a character named Michael, while those with female children read about Michelle). Participants were also informed that the vignette character was “the same age as your child”. It is important to note here that no diagnostic labels accompanied the vignette, therefore facilitating assessment of participants’ ability to spontaneously identify or label the symptoms depicted.
Mental health literacy
An adapted version of the Friend in Need questionnaire (Burns & Rapee, 2006) was selected for use in the current study, and encompasses assessment of severity of symptoms, identification of disorder, recognition of symptoms, and assessment of prognosis (length of time until recovery). This has been used in numerous studies of MHL (Byrne et al., 2015; Dolphin & Hennessy, 2014; Doyle et al., 2017; Leighton, 2010). Beliefs about the helpfulness or harmfulness of various potential responses were assessed by asking participants to mark whether they believe each of ten potential ‘treatments’ would be ‘helpful’, ‘harmful’ or would ‘make no difference’ to the vignette character. Example items rated by participants are ‘Listen to X’s problems in an understanding way’ and ‘Encourage X to become more physically active’. This method of assessment has been used in multiple studies (Cotton et al., 2006; Jorm & Wright, 2008; Jorm et al., 2007; Wright et al., 2005; Wright et al., 2007). These questions do not form a scale, and as such reliability is not reported here. For the same reason, participants’ responses were analysed at an item level.
Empathy
Dispositional or ‘trait’ empathy was measured using the 20-item Basic Empathy Scale (Jolliffe & Farrington, 2006), which has been validated in adult samples (Carre et al., 2013; Schwenck et al., 2016). Items are rated on a five-point scale from ‘Strongly disagree’ to ‘Strongly agree’. A sample item is “I get caught up in other people’s feelings easily”. Four items are worded negatively and are thus reverse-scored. Cronbach’s α was 0.709. Participants’ mean BES score was 3.60 (SD = 0.35; range 1–5). Female participants had higher mean BES scores (M = 3.65, SD = 0.33) than did male participants (M = 3.43, SD = 0.35), and this difference was statistically significant, M = 0.22, SE = 0.05, t (293) = 4.227, p < 0.001. In order to supplement the assessment of empathy, the anger, sympathy, and controllability subscales of the Indicator Questions for Perceived Controllability scale (Reisenzein, 1986) were used, which measures participants’ anger and sympathy towards the vignette character as well as perceptions of the controllability of the vignette characters’ symptoms. Each subscale comprises three items answered on a nine-point scale. Cronbach’s α was 0.564 for the controllability subscale, 0.787 for the anger subscale and 0.524 for the sympathy subscale.
Data analysis
Responses to open-ended questions (i.e., “In five words or less, what do you think is the matter with X?” and “Please quote all the words/phrases from X’s story that suggest to you that he/she might be experiencing difficulties”) were analysed using conventional content analysis, according to the guidelines set out by Hsieh and Shannon (2005). Quantitative data were analysed using IBM Statistical Package for Social Sciences (SPSS) version 27.0 (I.B.M. Corp., 2020). Binomial logistic regressions were performed to ascertain the effects of participant gender, child age, child gender, and empathy on categorical outcome variables (e.g., identification of depression: Y/N). Ordinal logistic regressions were used to predict ordinal (ranked) categories (e.g., level of concern measured categorically), given the same predictor variables. Multiple linear regressions were used for continuous outcomes (perceived controllability, anger, and sympathy). A critical significance level of p = 0.01 is reported throughout due to the large number of tests conducted and the subsequent likelihood of Type 1 error. Note that as online- and school-recruited parents differed in age, a Pearson correlation was run to determine the relationship between participants’ age and the key continuous predictor, total Basic Empathy Scale (BES) scores, which was not statistically significant (r = −0.02, n = 248, p = 0.749), suggesting that it was not necessary to control for parent age in the relevant analyses.
Results
Which ‘Symptoms’ do Participants Identify as Indicative of the Character Experiencing Difficulties?
Six of the nine DSM-IV (American Psychiatric Association, 2000) symptoms were mentioned in the vignette; participants were asked “Please quote all the words/phrases from X’s story that suggest to you that he/she might be experiencing difficulties” and responses were coded for the presence of words relating to these symptoms. Responses were coded for each symptom mentioned; the mean number of symptoms identified was 3.13 (SD = 1.51). The most frequently mentioned symptom was depressed mood, mentioned by 80.7% of participants, while 65.4% mentioned diminished interest, 60.4% mentioned insomnia, 40.8% mentioned diminished concentration, 34.9% mentioned fatigue, and 30.5% mentioned feelings of worthlessness.
Do Parents Label a Cluster of Symptoms as ‘Depression’?
Participants were asked to explain in five words or less what they “think is the matter” with the vignette character and these responses were coded using content analysis. If participants gave multiple explanations, responses were coded for each (e.g., the response “He is feeling depressed or being bullied” was coded for both “depression” and “bullying”). The percentage of participants who mentioned each explanation is presented in Table 2 below. The most frequently mentioned label was “depression”, mentioned by 66.7% of participants, followed by “bullying” (20.1%) and “anxiety” (12.5%).
Table 2
Percentage of participants who attributed the symptoms to each cause
Attributed explanation for symptoms | Mentioned by X% of participants |
|---|---|
Depression | 66.7 |
Bullying | 20.1 |
Anxiety | 12.5 |
Low self-esteem | 8.9 |
Loneliness/isolation | 5.8 |
Stress | 3.8 |
“Something wrong at home”/Family problems | 3.5 |
Hormones | 3.2 |
Issues with friends | 2.9 |
Lack/loss of confidence | 2.6 |
A binomial logistic regression was performed to ascertain the effects of participant gender, child age, child gender, and empathy on the likelihood that participants would label the symptoms as ‘depression’. The overall logistic regression model was not statistically significant at the level of p < 0.01, χ2(4) = 12.85, p = 0.012. Of the four predictor variables, only one was statistically significant: empathy (p = 0.003). With every unit increase in BES score, there was a 3.26 increase in the odds of labelling the symptoms as ‘depression’, see Table 3.
Table 3
Logistic regression predicting likelihood of labelling symptoms ‘depression’ based on gender, child’s age, child’s gender, and empathy
95% C.I. for odds ratio | ||||||||
|---|---|---|---|---|---|---|---|---|
B | SE | Wald | df | p | Odds ratio | Lower | Upper | |
Gendera | 0.104 | 0.334 | 0.097 | 1 | 0.755 | 1.110 | 0.576 | 2.137 |
Child age | 0.097 | 0.082 | 1.373 | 1 | 0.241 | 1.101 | 0.937 | 1.294 |
Child gendera | 0.004 | 0.267 | 0.000 | 1 | 0.987 | 1.004 | 0.595 | 1.695 |
Empathy | 1.180 | 0.398 | 8.808 | 1 | 0.003* | 3.255 | 1.493 | 7.096 |
Constant | −5.059 | 1.777 | 8.100 | 1 | 0.004 | 0.006 | ||
How Concerned are Parents for the Vignette Character?
The vast majority (92.2%) of participants believed the vignette character needed the help of another person, and two-thirds (64.7%) were “extremely worried” about the vignette character. An ordinal logistic regression was performed to ascertain the effects of participant gender, child gender, child age, and empathy on level of concern for the vignette character. The odds of participants with male children feeling a greater level of concern was similar to that of participants with female children (OR = 1.083 (95% CI, 0.650 to 1.802)), χ2(1) = 0.093, p = 0.760; the odds were also similar between male and female participants (OR = 1.474 (95% CI, 0.756 to 2.873)), χ2(1) = 1.299, p = 0.254. There was also no statistically significant increase in odds associated with the age of participants’ children (OR = 1.043 (95% CI, 0.890 to 1.221)), χ2(1) = 0.270, p = 0.603. However, an increase in empathy, as measured by the BES, was associated with a statistically significant increase in the odds of feeling a greater level of concern for the vignette character, with an odds ratio of 5.218 (95% CI, 2.387 to 11.408), Wald χ2(1) = 17.144, p < 0.001.
How Long do Participants Perceive it Will Take for the Character to Feel Better?
The majority (68.8%) of participants thought that it would take the vignette character “longer than a few months” to feel better. Again, an ordinal logistic regression was performed to examine the effects of participant gender, child gender, child age, and empathy on participants’ beliefs about how long it would take for the vignette character to feel better. However, the final model did not significantly predict the dependent variable over and above the intercept-only model, χ2(4) = 10.419, p = 0.034. There were no statistically significant differences in the odds of perceiving it would take longer for the character to feel better associated with participant gender, child gender, child age, or empathy.
How Controllable do Participants Perceive the Characters Symptoms to be?
Participants’ mean score on the perceived controllability subscale was 3.08 (SD = 1.35). A multiple regression was run to predict perceived controllability from participant gender, child gender, child age, and empathy. The multiple regression model did not predict perceived controllability, F(4, 289) = 1.5263, p > 0.16, adj. R2 = 0.007, see Table 4. While male parents (M = 3.40, SD = 1.28) and parents of male adolescents (M = 3.20, SD = 1.33) perceived that the vignette character had more control over their situation than did female parents (M = 2.98, SD = 1.36) and parents of female children (M = 2.97, SD = 1.36), these differences were not statistically significant.
Table 4
Multiple regression results for controllability
95% CI for B | ||||||||
|---|---|---|---|---|---|---|---|---|
Controllability | B | LL | UL | SE B | β | p | R2 | ΔR2 |
Constant | 4.403 | 2.247 | 6.559 | 1.095 | 0.021 | 0.007 | ||
Empathy | −0.089 | −0.562 | 0.385 | 0.241 | −0.022 | 0.712 | ||
Gendera | −0.338 | −0.765 | 0.089 | 0.217 | −0.093 | 0.120 | ||
Child gendera | −0.186 | −0.515 | 0.143 | 0.167 | −0.067 | 0.268 | ||
Child age | −0.043 | −0.145 | 0.060 | 0.052 | −0.050 | 0.415 | ||
How Much Anger did Participants Feel towards the Vignette Character?
Participants’ mean level of anger was 1.87 (SD = 1.26). A multiple regression model did not predict anger, F(4, 289) = 2.663, p = 0.03, adj. R2 = 0.022, see Table 5. Male parents (M = 2.13, SD = 0.20) and parents of male children (M = 1.93, SD = 1.39) responded with more anger than female parents (M = 1.77, SD = 0.07) and parents of female children (M = 1.81, SD = 1.14), but this difference was not statistically significant.
Table 5
Multiple regression results for anger
95% CI for B | ||||||||
|---|---|---|---|---|---|---|---|---|
Anger | B | LL | UL | SE B | β | p | R2 | ΔR2 |
Constant | 1.686 | −0.215 | 3.588 | 0.966 | 0.036 | 0.022 | ||
Empathy | −0.241 | −0.658 | 0.177 | 0.212 | −0.068 | 0.256 | ||
Gendera | −0.379 | −0.756 | −0.003 | 0.191 | −0.118 | 0.048 | ||
Child gendera | −0.176 | −0.467 | 0.114 | 0.147 | −0.072 | 0.233 | ||
Child age | 0.094 | 0.003 | 0.184 | 0.046 | 0.123 | 0.042 | ||
How Much Sympathy did Participants Feel towards the Vignette Character?
Participants’ mean level of sympathy was 8.14 (SD = 1.26). The multiple regression model statistically significantly predicted sympathy, F(4, 289) = 5.796, p < 0.001, adj. R2 = 0.061, see Table 6. Empathy and participant gender significantly contributed to the model, p < 0.01. Female parents responded to the vignette with significantly more sympathy (M = 8.25, SD = 0.06) than male parents (M = 7.56, SD = 1.23), t(319) = −3.29, p < 0.001.
Table 6
Multiple regression results for sympathy
95% CI for B | ||||||||
|---|---|---|---|---|---|---|---|---|
Sympathy | B | LL | UL | SE B | β | p | R2 | ΔR2 |
Constant | 6.202 | 4.595 | 7.809 | 0.816 | 0.074 | 0.061 | ||
Empathy | 0.507 | 0.154 | 0.860 | 0.179 | 0.166 | 0.005* | ||
Gendera | 0.464 | 0.146 | 0.782 | 0.162 | 0.167 | 0.004* | ||
Child gendera | 0.149 | −0.097 | 0.394 | 0.125 | 0.071 | 0.234 | ||
Child age | −0.023 | −0.099 | 0.054 | 0.039 | −0.035 | 0.559 | ||
How Helpful or Harmful did Parents Believe Certain Actions would be?
The percentage of participants who believed each action as ‘helpful’, ‘harmful’, or would ‘make no difference’ is depicted in Table 7 below. For most of the listed actions, parents’ beliefs tended to converge on whether the action would be helpful, harmful, or make no difference – for example, almost all (99.7%) of participants believed that “listening to X’s problems in an understanding way” would be helpful. However, certain actions showed more variation in beliefs: an equal proportion of parents believed that ‘Keeping X busy to keep his/her mind off his/her problems’ would be helpful (40.8%) as believed this would make no difference (40.5%).
Table 7
Proportion of participants that thought each action was ‘helpful’, ‘harmful’, or would ‘make no difference’
% of participants who selected each response | |||
|---|---|---|---|
Helpful | Harmful | Make no difference | |
Listen to X’s problems in an understanding way | 99.7 | 0.3 | 0.0 |
Talk to X firmly about getting his/her act together | 11.2 | 78.2 | 10.6 |
Suggest X seek professional help | 89.1 | 7.8 | 3.1 |
Make an appointment for X to see a GP | 87.9 | 6.5 | 5.6 |
Ask X whether he/she is feeling suicidal | 60.4 | 34.6 | 5.0 |
Suggest X has a few drinks to forget his/her troubles | 0.6 | 95.3 | 4.0 |
Rally friends to cheer X up | 54.8 | 15.9 | 29.3 |
Ignore X until he/she gets over it | 1.6 | 95.0 | 3.4 |
Keep X busy to keep his/her mind off his/her problems | 40.8 | 18.7 | 40.5 |
Encourage X to become more physically active | 89.1 | 0.6 | 10.3 |
Discussion
The literature suggests that parents’ responses to an adolescent’s expression of depressive symptoms may play a crucial role in determining the subsequent course of their illness (Herres & Kobak, 2015; Perry & Pescosolido, 2015; Perry et al., 2017). For example, parents may detect symptoms, identify the underlying cause (Lawrence & Ureda, 1990; Roberts et al., 2005), and play an important role in facilitating or encouraging adolescents’ access to professional mental health support (Frojd et al., 2007; Godoy et al., 2014). However, little is known about parents’ knowledge and understanding of adolescent depression, thus leaving a substantial gap in the literature.
This study aimed to address this gap by exploring depression literacy (DL) in a sample of parents of adolescents living in the Republic of Ireland through a vignette-based questionnaire. Participants read a vignette that was gender-matched to the participants’ adolescent child (i.e., read about Michelle or Michael), and participants were told that the vignette character was the same age as their child. Analyses explored whether participants’ labelling of the character’s symptoms as ‘depression’, level of concern for the character, beliefs about prognosis for recovery, perceived controllability, anger, and sympathy were associated with participants’ gender, the gender of their child (i.e., the gender of the vignette character), child’s age, and empathy. The study hypothesised that female parents, parents of female adolescents, parents of older adolescents, and parents with higher dispositional empathy would be more likely to label the symptoms ‘depression’, feel more concern for the vignette character, believe it will take longer for the character to feel better, perceive the characters’ symptoms to be less controllable, and feel less anger and feel more sympathy towards the character.
Identification of Depression
While the majority of parents in the current study (66.7%) labelled the cluster of ‘symptoms’ described in the vignette “depression”, this is a lower rate of identification of depression than reported in previous research with parents and other caregivers (Bonfield et al., 2009; Shah et al., 2004), although higher than a study of parents in the United States (Bonanno et al., 2021). Labelling of symptoms is often interpreted as indicative of MHL, and is a key target in many MHL programmes for parents of adolescents (Kusaka et al., 2022). As such, lower levels of identification among parents in the current sample may be cause for concern. However, it is important to note that while some studies have found that labelling symptoms in line with diagnostic criteria for mental disorders predicts less stigmatising attitudes (Yap et al., 2013), other research and theory suggests that labelling induces negative responses from others. For example, one study found that a diagnostic label increased the likelihood of parents responding in a stigmatised manner to a hypothetical child (Ohan et al., 2013). Often situated in a wider discussion of psychological essentialism, labelling is interpreted by some academics as a reification of Western, psychiatric (or more specifically, biomedical) understandings of human emotional experiences (Kokanovic et al., 2013; Kvaale et al., 2013; May, 2007; Spence, 2012). This was discussed in an editorial in the Journal of Child Psychology and Psychiatry, “How upset is too upset? How often is too often?” (Gregory, 2017, p. 649), and previously in the British Journal of General Practice, “Is diagnosing depression in young people just medicalising moodiness?” (Iliffe et al., 2009, p. 156). This has long been recognised as a core conundrum within mental health research (Romelli et al., 2015).
Other research has found, however, that labelling conveys little impact on responses or behaviour; for example, including a depression label when describing a vignette character (i.e., “it is important to note that X would receive a diagnosis of clinical depression from a psychiatrist or clinical psychologist”) was not associated with any significant differences in male adolescents’ emotional responses towards the character, nor their evaluations of the character’s mental health status (Dolphin & Hennessy, 2017). Research with adults has similarly found that labelling had ‘practically no effect’ on public attitudes towards people with major depression (Angermeyer & Matschinger, 2003), and individuals with depression and other mental health issues also qualitatively report mixed experiences associated with illness labels (see O’Connor et al., 2018 for one systematic review). Therefore, it is unclear whether interventions and psychoeducation strategies should continue to focus on improving parents’ ability to label symptoms, or whether other aspects of DL may be more important targets. Perhaps more importantly, the vast majority (92.2%) of parents in the current study believed that the vignette character needed help, which stands in stark contrast to just 27.3% of Australian parents in another study (Lawrence et al., 2015). On the whole, parents’ levels of DL in the current study appeared to be relatively good. Participants also reported low levels of anger and high levels of sympathy for the vignette character.
In contrast with our hypothesis and the results of Shortt et al. (2016), participants’ labelling of the symptoms as ‘depression’ did not differ according to the gender or age of the participants’ child. The current sample was comprised of parents whose children’s ages spanned a relatively short period, 12–18 years. It may be the case that greater differences would have been observed if the sample included parents of primary school aged children as well as parents of young adults. Nonetheless, this finding remains useful for developing education and intervention strategies, as it suggests that such strategies may be appropriately and effectively targeted uniformly to all parents of secondary-school aged children.
Empathy as a Key Factor Determining Parental DL
Indeed, the hypotheses that there would be gender or age differences in male and female participants’ DL was largely unsupported across variables; instead, empathy was the predictor variable associated with increased likelihood of identifying depression, greater concern for the vignette character, and greater sympathy. Given that other predictor variables in the current study were non-modifiable, this is a promising result, as research asserts that it is possible to increase empathy through the provision of training, and that receipt of empathy training is effective in increasing helping and prosocial behaviour among children, adolescents, and adults (Georgi et al., 2014; Gibbs et al., 1996; Greenberg et al., 2009; Jazaieri et al., 2012; Klimecki et al., 2013; Klimecki et al., 2014; Leiberg et al., 2011; Mascaro et al., 2013; Weng et al., 2013). Empathy is hypothesised to motivate parents to respond with support or care in response to a child’s emotions and is thus related to child outcomes (Borelli et al., 2021). To our knowledge, this study is the first to explore the link between empathy and DL in parents of adolescents, and as such these findings point to the need to further investigate the nuances of this association.
Beliefs about Asking about Suicide
A third (34.6%) of parents thought that asking the adolescent if they were feeling suicidal would be harmful. This is a slightly larger proportion than the results of a previous vignette-based study which used the same measurement tool, in which 23% of Australian parents believed this action would be harmful (Jorm et al., 2008). Beliefs regarding suicide risk assessment are important to highlight, as increased suicide-related knowledge has been found to exert a positive effect on intentions to provide adequate help to distressed individuals (Arendt et al., 2018). The literature is clear that suicide risk does not increase after assessment (Blades et al., 2018; Dazzi et al., 2014; de Beurs et al., 2016; DeCou & Schumann, 2018; Gould et al., 2005; Mathias et al., 2012; Peña & Caine, 2006; Ross et al., 2016). In fact, a review of the evidence concluded that asking about suicide may reduce, rather than increase, suicidal ideation (Dazzi et al., 2014), and suicide prevention experts (Kelly et al., 2008) and mental health professionals (Schurtz et al., 2010) endorse the helpfulness of this action. Despite this, concerns about the iatrogenic risk of asking about suicide are commonly expressed by laypeople (Yoshioka et al., 2014), and some parents report finding the topic of suicide difficult and uncomfortable (McKay et al., 2022). Parent suicide stigma has also been found to be significantly associated with more negative help-seeking attitudes and lower help-seeking intentions (Burke et al., 2023). While it is therefore encouraging that most participants in the current study appeared to recognise the helpfulness of suicide risk assessment, the fact that one-third thought this would be harmful indicates that this belief should be targeted in future interventions and suicide prevention messaging.
Strengths and Limitations
The findings of this study should be considered in light of several limitations. The majority of the sample were female; this is a common issue in research with parents (Costigan & Cox, 2001; Phares et al., 2005), but suggests that caution may be warranted when interpreting results regarding their generalisability to fathers and parents of other genders. In future, specific sampling strategies may be required to target the recruitment of fathers as participants. It is possible that this study may have been affected by self-selection bias, such that those parents with higher levels of MHL/DL were more likely to choose to participate. However, it must be acknowledged that this is true of all research on this topic that is ethically conducted (i.e., wherein participants voluntarily consent to participation), and therefore comparison of the current results to those obtained in other studies should not be unduly affected. Most participants were born in Ireland (75.4%) and identified as White Irish (82.2%). Further research is needed to explore DL amongst parents from diverse backgrounds.
This study asked participants to respond to a hypothetical vignette scenario; it is possible that participants may respond differently in real-world situations. However, it is important to note that this vignette has been previously validated in two studies, each of which included a review by a panel of clinical psychologists (O’Driscoll, 2013; Silke, 2016). Additionally, Rossetto et al. (2014, 2016) found that participants’ help-giving intentions towards a hypothetical person experiencing mental illness were associated with their real-life behaviours. Experimental research has also demonstrated that behavioural intentions in response to hypothetical vignettes tend to be under-, rather than over-reported (Eifler & Petzold, 2019). The literature additionally suggests that social desirability is reduced in vignette responses, in comparison to direct questioning (Hughes & Huby, 2002).
Research on MHL is hampered by a lack of a universal or consistent measure; this has been raised by a number of researchers (Bale et al., 2018; Hurley et al., 2020; O’Connor et al., 2014; Spiker & Hammer, 2019; Wei et al., 2016). The alpha level of the perceived controllability subscale in the current study was lower than desired, suggesting that its psychometric properties may be questioned. Furthermore, there are a number of additional variables that would have been interesting to explore in the current study. The source of participants’ knowledge of depression was not examined. Participants’ experience of help-seeking for their own mental health issues or their current mental health status may be useful to measure in future studies, as other research has found that higher levels of depressive symptoms among parents was associated with more negative responses to adolescent depression (Johnco & Rapee, 2018). Although participants’ perceptions of the controllability of symptoms was measured, the study did not measure participants’ beliefs about the causes of the vignette characters problems. These data were collected prior to the onset of the COVID-19 pandemic. Research suggests that rates of depression increased among adolescents throughout the pandemic (Octavius et al., 2020), and public health messaging for parents often featured information about looking after their child’s mental health (e.g., World Health Organization, 2020). Parents’ DL may therefore have changed over time; replication of this study in future could facilitate comparison.
Implications
This study highlights a number of potential future directions for research. The current study explored DL across a relatively homogenous group of participants (the majority being female and White Irish, with children’s ages ranging from 12 to 18). As such, further research is required to explore DL in a more diverse sample of parents, including parents of young people spanning a wider age range (e.g., parents of primary school aged children, and parents of young adults), and specifically targeting the participation of fathers and parents from diverse ethnic and cultural backgrounds. In order to gain a more comprehensive understanding of parents’ DL, it would also be useful to explore the source of parents’ knowledge of depression, as well as beliefs about contributory factors, appropriate responses and supports. Furthermore, research could explore how adolescents perceive parental responses to signs of depression, and whether these align with their own preferences or whether adolescents experience these responses as helpful.
The results of this study also have important implications for guiding these efforts to improve the support that parents may provide to young people experiencing symptoms of depression. In particular, the results highlight empathy which may be potentially modifiable, and therefore suitable to target through intervention, the robust evaluation of which should furthermore be a focus of future research. The translation of research evidence into interventions for parents regarding their role in supporting children with depression has been slow (Ong & Caron, 2008; Yap et al., 2015). This may be, in part, due to concerns about avoiding parent blame and criticism of parenting skills, which have been shown to decrease caregivers’ engagement (Brown, 2018; Iachini et al., 2015; Penzo & Harvey, 2008). Rather, it has been suggested that an effective approach to the development of parent training is one that views parents as “active agents of assistance and support for young people with mental illness… neither [focusing] on changing parents nor [implying] causal responsibility” (Honey et al., 2014, p. 201), but instead helping parents to find and use the most effective responses when supporting their child. This study indicates that targeting parents’ empathic responses to their child (e.g., engaging in perspective-taking) should also be emphasised.
As discussed earlier, research supports that asking about suicide does not increase risk (e.g., Blades et al., 2018), and this is addressed in public health messaging in Ireland (Health Service Executive, 2022). However, the results of the current study – specifically, that approximately one-third of parents thought that asking about suicide would be harmful – suggest that this messaging may not have effectively reached all parents. Suicide prevention gatekeeper training for parents is being evaluated in Australia (Calear et al., 2024; McKay et al., 2022) and this study suggests that similar education for parents may be beneficial in the Irish context. Clinicians and services working with parents may also wish to address these concerns.
Conclusion
To our knowledge, this study is the first to examine DL among parents of adolescents in Ireland, and overall found that parents expressed relatively good understanding of adolescent depression, and demonstrated their capacity for responding in a positive and supportive manner. This study significantly advances knowledge of parental understanding of adolescent depression in this context; by identifying the factors that are associated with parents’ DL, the study hopefully will enable the academic and clinical fields to more clearly comprehend and optimise parents as partners in supporting adolescents experiencing depression.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1007/s10826-024-02970-4.
Acknowledgements
S.J.B. was supported by a Government of Ireland postgraduate scholarship, awarded by the Irish Research Council, to complete this work. We gratefully acknowledge all participants, for taking the time to take part in this study.
Compliance with Ethical Standards
Conflict of Interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
