Youths’ attitudes toward open discussion of suicide,preferred contexts,and the impact of Internet use: An exploratory sequential mixed-methods study in Hong Kong

Abstract

Background:

Although the necessity and benefits of having the open and public discussion about suicide have been recognized, youths’ opinions regarding such discussion remain unknown.

Aims:

To explore youths’ attitudes toward open suicide discussion, particularly concerning their preference of discussion contexts and impacts of Internet use.

Method:

This exploratory sequential mixed-methods study targeted Hong Kong youths aged 15 to 19. A total of six focus groups and 12 individual interviews were held (N = 40). Topics included adolescents’ views of open suicide discussion, concerns, perceived benefits, and preferred contexts. The questionnaire survey (N = 1,676) was conducted subsequently to investigate the prevalence of youths’ perspectives on public discussion of suicide and relationships with discussion contexts, use of social media platforms, and motivations of online expression.

Results:

Qualitative findings revealed three types of attitudes toward open suicide discussion: reluctance, support, and indifference. Major barriers included cultural norms, topic sensitivity, privacy concerns, contagion effect, fear of embarrassment, and unpleasant experiences in school programs. Results of quantitative analyses showed that reluctance was the dominant attitude among adolescents, and taboo was the top concern. Variations in youths’ attitudes were related to gender, school academic banding, and suicide-related experiences. Notably, adolescents who had been exposed to suicide messages in contexts of peer networks and online platforms were more likely to endorse open suicide discussion. In addition, an increased likelihood of engaging in public suicide discussion was associated with the use of Instagram, WhatsApp, and Snapchat, and the motive of ‘expressing emotions and opinions’ online.

Conclusions:

Our results indicated a prevalent rejection among adolescents toward open suicide discussion, suggesting the issue of stigma, the need for tailored programs, the value of appropriate contexts, and the impact of Internet use. These findings may facilitate the development of school-based suicide prevention initiatives and the efficacy of online services for suicide-related communication.

Keywords

Open suicide discussion preferred contexts Internet use youths’ attitudes Hong Kong Chinese

Introduction

Suicide has become a serious public health concern in recent decades. According to the World Health Organization, suicide is the fourth leading cause of death among adolescents aged 15 to 19-year-old (World Health Organization, 2021). However, despite being well recognized, suicide was rarely addressed openly (Poijula et al., 2001). The lack of public education and open discussion has been identified as the principal trigger of the proliferation of misconceptions about suicide and mental health (Shahtahmasebi, 2014), which may have contributed to the increasing suicide rates and prompted ‘more of the same’ interventions (Pridmore et al., 2016; Shahtahmasebi & Cassidy, 2014).

Per the Social Norms Theory, social norms are defined by people’s perceptions of ‘normal behavior’ (Dempsey et al., 2018). Promotion of public discussion about suicide may reconstruct the descriptive and prescriptive norms associated with emotional coping, leading to the establishment of ‘new standards’ in society. The policy of ‘keeping suicide discussions out of the public domain’ has been criticized for impeding the replacement of outdated social norms (Shahtahmasebi, 2016), and ‘the need to initiate a dialog about suicide’ has emerged as a major theme in north American articles (Carmichael & Whitley, 2019). Open discussion of suicide has also been involved as a crucial element in suicide prevention programs for young people, whilst it remains unclear whether the idea resonates with audiences (Ftanou et al., 2017). Insofar, little is known about youths’ attitudes toward open suicide discussion.

The call for a more open discussion of suicide has been denoted by journalists, mental health professionals, and social organizations, considering its benefits, including breaking the silence surrounding suicide, improving help-seeking (Wiggins et al., 2016), and redressing longstanding cultural norms and taboos (Fitzpatrick, 2020). On the other hand, concerns and challenges to open suicide talk were also underlined, such as eliciting overwhelming emotions, lack of support and appropriate guidance, and privacy issues (Gibson et al., 2019; Ranahan & Keefe, 2022). Variations in peoples’ attitudes were shown to be related to their gender (Murali & Smitha, 2022), academic achievement (Calver & Michael-Fox, 2021), and the previous history of suicide (Eskin et al., 2016; Vossos, 2017), whereas evidence regarding the relationships is mixed.

It is anticipated that the degree of transparency and comprehension during communication would vary upon the targets with whom adolescents talk about suicide, hence determining their willingness to participate (Huisman & van Bergen, 2019). However, much uncertainty exists about the contexts that would facilitate adolescents’ engagement in open suicide discussion. The preference for a certain context often builds on the prior portrayals of suicide information from the sources. For instance, some adolescents believe that adults are incapable of understanding youths’ struggles, and their responses to youth suicide are usually flawed and even stigmatizing (Gibson et al., 2019). Discussing suicide with families, professionals, or other known adults was thus deemed risky, unwanted, and unhelpful (Frey & Cerel, 2015; Han et al., 2018). On the other hand, exposure to suicide messages via media and the Internet may encourage youths to join open suicide discussions. Adolescents would proactively initiate dialog about suicide following media reporting of celebrity suicidal deaths (Niederkrotenthaler et al., 2020) and the launch of the prevailing Netflix series ‘13 Reasons Why’ (Carmichael & Whitley, 2018).

With the rapid development of the Internet, access to online platforms enables young people to discuss sensitive issues without adult surveillance or face-to-face encounters (Mishara & Kerkhof, 2013). Previous study has revealed that youths preferred to talk about suicide within their peer networks, particularly through digital communication (Gibson et al., 2019). Online conversation about suicide has been valued for its openness and broader opportunities for connection and recognition (Niezen, 2013). These findings demonstrated the benefits of the online environment for youths to discuss suicide, whereas research on the impact of Internet use on youths’ attitudes toward open suicide discussion is still in relative infancy. Most prior studies only evaluated the effect of online forums (Harris et al., 2009; Paulus & Varga, 2015; Varga & Paulus, 2014), failing to address the increasing prevalence of social media (SM) platforms and potential impacts of adolescents’ motivations of online expression.

Aims and objectives

Although most researchers show support for public discussion of suicide, there is a dearth of information on people’s attitudes toward open suicide discussion, particularly from youths’ perspectives. This study is part of a larger project adopting an exploratory sequential mixed-methods design to examine youth suicide in Hong Kong, with specific aims to (1) explore adolescents’ attitudes toward open discussion of suicide, (2) determine youths’ concerns, perceived benefits, and preferred settings of open suicide discussion, and (3) identify factors contributing to differences in youths’ attitudes toward open suicide discussion. Besides, we are particularly interested in examining: (4) the association between the contexts of discussion and youths’ attitudes, assessed by their previous exposure to suicide messages, and (5) the impact of Internet use, including SM platforms and motivations, on youths’ attitudes toward open discussion of suicide.

Methods

Ethical approval was attained from the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (UW 18-338).

Qualitative phase

Participant recruitment

A total of 40 participants were recruited (mean age = 16.3 years, range 15–18), including 31 focus group participants (12 males, 19 females) and 12 individual interviewees (four males, eight females). Three female participants attended both individual interviews and focus group discussions. Of the 40 participants, 35 participants had lived experience of suicide, including the incidence of suicidal ideations (SIs; 20 participants having SI only), self-harm, suicide attempts (SAs; eight participants having SA), and providing support in a crisis. Purposive sampling was adopted to ensure the diversity in participants’ backgrounds. We sought referrals from those who served or had close contact with at-risk youths to facilitate participants’ recruitment, such as teachers, general practitioners, psychologists, psychiatrists, and school social workers. Written informed consents were obtained from all the participants or, if they were under the age of 18, from their parents or guardians.

Procedure

The qualitative data were collected by semi-structured interviews, where participants discussed their attitudes, perceived concerns and benefits, and preferred settings for open discussion of suicide. In total, six focus group interviews were conducted (3–7 participants in each session lasting for 1–1.5 hours). Members of each group were acquainted and of the same gender. In addition, 12 individual interviews (each lasting for 1 hour) were then held with participants who were unsuitable to join the focus groups, or whose views were poorly articulated in the group setting owing to confidentiality concerns and time limits. A pilot focus group has been carried out to test the sensitivity of topics for vulnerable participants.

Analysis

All interviews were conducted in Chinese, audio-recorded, and transcribed verbatim. Two experienced qualitative researchers were responsible for data collection and analysis. Reflexive thematic analysis was performed in NVivo (Version 12) using the inductive approach. We followed the six-step framework: familiarization, generating codes, constructing themes, revising themes, naming and defining themes, and reporting findings (Braun & Clarke, 2019). Disagreements on codes and themes were addressed by team discussion until consensus was attained. Subthemes and themes, upon which the survey questionnaire was constructed, were categorized and reviewed until both researchers approved.

Quantitative phase

Sample

In the quantitative phase, a cross-sectional survey was conducted with students aged 15 to 19 in Grades 10 to 12. Invitations were sent to all Hong Kong secondary schools and nine of them (covering all academic bandings: Band 1, 2, and 3) responded positively. Students from Band 1 schools, on average, score highest in the university admission examination. A sample size of at least 1,153 responses was required, considering the total registered number of eligible students in 2019 (N = 135,669) (School Education Statistics Section, 2020) and the reported prevalence of suicidality among Hong Kong adolescents (estimated to be as high as 25%) (Siu, 2019), to achieve a maximum estimation error (absolute precision) of d = 0.025 from the true prevalence rate with a 95% confidence.

Questionnaire

The questionnaire was self-administered and anonymous which took about 15 to 20 minutes to complete. School teachers assisted in the distribution of questionnaires to students at participating schools. The items were generated and amended according to qualitative findings, previous literature, and feedback from the research team. The conditional branching was used on items to assess respondents’ attitudes toward open suicide discussion. The parent question was mandatory and multiple-choice, asking respondents ‘Are you willing to discuss suicide openly’ (dichotomous item coded as Yes = 1 or No = 0). The subsequent child question was optional and multiple-answer, investigating perceived benefits (following Yes responses; three dichotomous items) or concerns (following No responses; five dichotomous items) for such discussion.

Other sections included contexts of previous exposure to suicide messages (five dichotomous items), use of SM platforms (eight dichotomous items), motivations (nine dichotomous items) of online expression, SIs and SAs in the past 12-month, and sociodemographic information. Respondents’ suicide risk was examined by questions on prior suicidal behaviors (scored as Yes = 1 or No = 0). We asked respondents ‘Have you considered killing yourself in the past 12 months?’ and ‘Have you attempted to kill yourself during the past 12 months?’. Non-affirmative responses to both questions were categorized as ‘low risk’, affirmative responses to only the first question as ‘medium risk’, and affirmative responses to both questions as ‘high risk’. The reliability and validity of the questionnaire were shown as satisfactory (Cronbach’s α = .74) through the pilot test (N = 30). Further details of the questionnaire design have been described in a previous publication (Chen et al., 2022).

Analysis

Statistical analyses were performed in SPSS (Version 27.0). Descriptive statistics are demonstrated for the response distribution of items. The Pearson Chi-squared test was conducted to identify differences in attitudes toward open suicide discussion by respondents’ background characteristics. Multivariable logistic regression analyses were performed to examine the associations between attitudes toward open discussion of suicide and contexts of previous exposure to suicide messages as well as the Internet use of platforms and motivations. Effects of gender, school academic banding, and suicidal status were adjusted in regression analyses, and the estimated adjusted odds ratios (AORs), as well as the corresponding 95% confidence intervals (CIs), were reported. Statistical significance was denoted by p-value < .05.

Results

Qualitative findings

Participants showed three major types of attitudes toward open discussion of suicide: reluctance, support, and indifference. Their concerns, perceived benefits, and preferred contexts of such discussion were also revealed. Details of themes and subthemes are elaborated with sample quotes in Table 1.

Table 1.

Illustrative quotes for each theme and subtheme (N = 40).

Themes/subthemes	Example quotes
Theme 1: Reluctance – barriers and concerns
(a) Taboo on death	Quote 1. Hong Kong is a Chinese society after all, and it is still a taboo to talk about death in the local population. It’s normal that people are afraid of unknown things. (Individual-9, F, 18, secondary school student, with SI, attempter) Quote 2. School programs rarely focused on ‘death’ since most life education aimed to teach us how to find purposes of life and tended to be positive. Probably because of the taboo, death was usually not discussed. (Individual-10, F, 17, secondary school student, with SI, non-attempter)
(b) Sensitivity of the topic	Quote 3. I would be nervous, especially when I face a large group of people of my age, and we need to sit in a circle to discuss this topic. As [suicide] is a sensitive topic, it seems that I have to share my emotional feelings, which would make me even more nervous. (Individual-5, F, 18, university freshman, with SI, non-attempter)
(c) Privacy issue	Quote 4. In fact, one’s problems are often very personal and private, hence most of them are discussed and resolved in a small group. It is rare to see a person raise such an [private] issue [of suicide] and share it in public. It will increase their pressure due to the broad range of audience. (Individual-11, M, 15, secondary school student, with SI, non-attempter)
(d) Contagion effect	Quote 5. Maybe at first, only one person was sharing some unpleasant experiences, yet eventually [the negative emotions spread among] the whole group of people who were listening. They would definitely provide the distressed person with some comforts, but they might also get dispirited [after engaging in the discussion]. (Group 3 – Participant-3, F, 18, university freshman, no SI, non-attempter)
(e) Embarrassment and feeling overwhelmed	Quote 6. In real life, when I discuss [suicide] issues with others, one concern is that people will make comments, but I may not know how to respond, another is the embarrassment to have someone talking to me face-to-face. Therefore, I tend to discuss it online given the extra time for thinking and reaction. I have time to consider what I should say and how to avoid hurting others’ feelings or making them happier. (Individual-8, F, 17, secondary school student, no SI, non-attempter)
(f) Negative experiences in previous school programs	‘Oversimplification of coping with self-distress’ Quote 7. I think the school talks [on suicide prevention] are not helpful and sometimes make me feel uncomfortable. They invited a professional to be the speaker who constantly simplified the issue and described emotional control as a minor issue. But we, as students, don’t believe it is such an easy task. (Individual-4, F, 16, secondary school student, with SI, attempter) ‘Repetition of general knowledge’ Quote 8. The speakers just told us that, ‘Be happy and don’t get sidetracked. If you have any problem, go and find the social worker or your teacher’. They were repeating something that we already knew, so we all fell asleep. (Individual-6, M, 16, secondary school student, no SI, non-attempter) ‘Nonspecific and unindividualized approaches’ Quote 9. Suggestions [given by professionals who were invited to deliver] life education targeted a group of people rather than a particular person’s problem. This approach was too general and too formal without considering individual characteristics and the focus of suicidality development. (Group 5 – Participant-2, M, 16, secondary school student, with SI, non-attempter)
Theme 2: Support – perceived benefits
(a) Motivating self-reflection	Quote 10. I think [the invited speaker’s] words were very inspirational. He used a distinct perspective to explore the suicide issue that I had never thought of. He sometimes gave you the responses that would make you feel very surprised. (Group 5 – Participant-3, M, 16, secondary school student, no SI, non-attempter)
(b) Skills and efficacy development	Quote 11. Some school talks were useful. For instance, people who had undergone a suicidal episode were invited to share their experiences. At least we can learn from a successful example. (Individual-4, F, 16, secondary school student, with SI, attempter)
Theme 3: Indifference
‘None of my business’	Quote 12. Suicide is not such a solemn matter. It’s none of my business. I don’t want to discuss it, maybe because I haven’t experienced it myself, and I don’t have any friends who have ever tried to attempt suicide. (Group 1 – Participant-1, M, 16, secondary school student, with SI, non-attempter)

Note. Participants are identified by the interview type and number (‘Group’ and ‘Individual’ – types of interviews), gender (‘M’ and ‘F’ refer to males and females respectively), age, educational status, and suicide-related experiences (‘with SI’/ ‘no SI’ indicates whether they had any suicidal ideation; ‘attempter’/‘non-attempter’ indicates SA experiences).

Theme I: Reluctance – barriers and concerns

Reluctance was the predominant kind of attitude, that most participants were hesitant or against talking about suicide in public. They identified a range of barriers and concerns, including:

i. ‘Taboo on death’ (Quote 1–2). Several participants emphasized the cultural impact in Chinese society, where the taboo still exists regarding public discussion on death and suicide.

ii. ‘Sensitivity of the topic’ (Quote 3). Some participants worried that suicide was a sensitive topic for youths at risk. Engaging them in an open discussion about suicide might cause overwhelming distress and emotional disturbances.

iii. ‘Privacy issue’ (Quote 4). Certain participants expressed concerns for the security of their privacy and a preference for discussing suicide in a smaller group with known individuals. They argued that the diverse backgrounds of the audience and a lack of mutual trust might diminish their confidence in the confidentiality of the information disclosed during public conversation, therefore inhibiting the sharing of personal experiences and challenges.

iv. ‘Contagion effect’ (Quote 5). Some participants were concerned that openly discussing suicide might encourage copycat behaviors and induce contagion effects among adolescents.

v. ‘Embarrassment and feeling overwhelmed’ (Quote 6). Several participants felt embarrassed when discussing suicide-related topics face-to-face. They were uncertain about the appropriateness of their responses or emotional reactions to those who indicated distress or suicidality during the discussion. Online platforms, which sustain greater anonymity and provide more buffers for communication, were favored among participants for open suicide discussion.

vi. ‘Negative experiences in previous school programs’. Most participants who had attended ‘life education’ or equivalent programs in schools were unwilling to attend another one. They reported various issues based on prior experiences, such as ‘oversimplification of coping with self-distress’ (Quote 7), ‘repetition of general knowledge’ (Quote 8), and ‘nonspecific and unindividualized approaches’ (Quote 9).

Theme II: Support – perceived benefits

Despite the resistance by the majority, some participants supported open discussion of suicide and considered it beneficial for promoting the mental health of adolescents. Perceived benefits of public suicide discussions consisted of:

i. ‘Motivating self-reflection’ (Quote 10). Certain participants denoted that public discussion of suicide could be inspirational when novel angles were suggested and might motivate them to reflect on mental health issues.

ii. ‘Skills and efficacy development’ (Quote 11). Some participants deemed the open discussion as a learning opportunity. They were enabled to exchange ideas with people who had effective coping. It might enhance their skills of handling emotions and their confidence in interfering with self- and peer-suicide.

Theme III: Indifference – ‘None of my business’

Aside from explicit opposition or support for public suicide discussion, a few participants were apathetic (Quote 12). They felt that it had nothing to do with them since they were not directly affected or lacked relevant experiences.

Quantitative findings

In total, we collected 1,676 valid responses. Table 2 displays respondents’ sociodemographic characteristics and suicide-related experiences. Respondents’ suicidality was classified into three groups based on their suicidal status: ‘low risk’ (no SI), ‘medium risk’ (with SI only), and ‘high risk’ (having SA). The top sources for respondents to obtain suicide-related messages were self-disclosure from suicidal peers (32.7%) and online ventilation posts (27.6%).

Table 2.

Descriptive statistics for respondents’ backgrounds (N = 1,676).

Variables	%
Sociodemographic characteristics
Gender
Male	49.6
Female	50.4
School bandings
Band 1	26.4
Band 2	34.5
Band 3	39.1
Suicide-related experiences
Suicidal status
Low risk: no SI	74.7
Medium risk: with SI only	21.5
High risk: with SA	3.7
Contexts of exposure to suicide messages
Self-disclosure of suicidal peers	32.7
Online ventilation posts	27.6
Information from the third party	11.4
Media report	2.4
School announcement	0.8

Note. SA = suicide attempt; SI = suicidal ideation.

Table 3 demonstrates responses to items on attitudes toward open discussion of suicide, as well as the variances in respondents’ background factors. Over half of respondents opposed open suicide discussion (53.6%). Taboo was the top concern that suicide might be too heavy of a topic to be discussed publicly (55.8% among opposition), while emotional support (69.5% among supporters) was perceived as the primary benefit. In general, respondents from Band 3 schools were more likely to oppose open suicide discussion (χ² = 8.5, p = .01). Of those against, respondents from Band 3 schools were less likely to worry about the ‘burden of the suicide topic’ (χ² = 11.1, p = .004) yet more likely to show concern on the ‘contagion effect’ (χ² = 7.0, p = .03). Among supporters, female respondents were more likely to recognize the benefit of ‘emotional support’ (χ² = 10.1, p < .001) of open suicide discussion, whereas male respondents were more likely to endorse the ‘skill and knowledge development’ (χ² = 9.7, p = .002). Respondents at low suicide risk were more likely to be concerned about the ‘panic effect’ (χ² = 14.1, p < .001) and acknowledge the ‘enhancement of skills and knowledge’ (χ² = 10.4, p = .006), while those who identified the benefit of ‘raising awareness’ were more likely to have SA experiences (χ² = 8.0, p = .02).

Table 3.

Response frequency of items on attitudes toward open discussion of suicide and differences by sociodemographic characteristics and suicidality (N = 1,676).

Items ^a	Overall (%)	Gender (%)			School banding (%)				Suicidality (%)^b
Items ^a	Overall (%)	Male	Female	χ ²	Band 1	Band 2	Band 3	χ ²	Low risk	Medium risk	High risk	χ ²
Attitudes
Objection	53.6	55.5	51.5	2.6	54.4	48.9	57.1	8.5^*	53.7	53.8	47.5	0.9
Concerns (among opposition)
Topic too heavy	55.8	53.8	58.3	1.8	61.1	60.0	49.2	11.1^**	56.5	51.9	60.7	1.6
Never reach a conclusion	33.3	33.9	32.6	0.2	30.1	35.4	33.9	1.7	32.3	36.2	39.3	1.4
Cause panic	25.0	27.5	22.6	2.8	24.8	23.3	26.4	0.9	27.8	14.4	21.4	14.1^***
Safer to talk within one’s own circle	21.8	19.2	24.5	3.5	24.4	21.1	20.7	1.3	20.2	25.1	32.1	3.9
Contagion effect	15.3	14.3	16.1	0.6	13.8	11.6	19.0	7.0^*	15.1	12.8	25.0	3.0
Perceived benefits (among supporter)
Emotional management	69.5	64.1	74.7	10.1^***	72.0	74.7	62.2	11.3^**	69.6	69.9	65.6	0.3
Skill and knowledge learning	48.4	54.4	43.2	9.7^**	49.0	47.1	49.3	0.3	52.1	38.0	43.8	10.4^**
Raising awareness	32.2	30.9	33.0	0.4	36.5	33.4	27.7	4.5	30.2	35.0	53.1	8.0^*

Note. Descriptive statistics and Pearson Chi-squared test were used to analyze the data.

The proportion of ‘yes’ responses to the items are reported. Respondents were able to choose only one option of ‘Attitudes’ (objection/support) and were able to choose one or more options of given items of concerns (only presented to those who reported ‘objection’) and perceived benefits (only presented to those who reported ‘support’).

Three groups for suicidality: Low risk = those with no suicidal ideation; Medium risk = those who had suicidal ideation only; High risk = those who had attempted suicide.

p < .05, **p < .01, ***p < .001.

Multivariable logistic regression analyses were conducted to examine the association between attitudes toward open discussion of suicide and contexts of prior exposure to suicide messages (Table 4). Adjusted for gender, school banding, and suicidal status, respondents who were exposed to suicide messages through suicidal peers’ self-disclosure (AOR = 1.30, 95% CI 1.05–1.61) and online ventilation posts (AOR = 1.61, 95% CI 1.29–2.01) were at greater odds of favoring open discussion of suicide.

Table 4.

Association of attitudes toward open discussion of suicide with contexts of previous exposure to suicide messages (N = 1,676).

Attitudes toward open discussion^a	Model 1				Model 2^b
Attitudes toward open discussion^a	Parameter estimates	SE	p	Crude OR(95% CI)	Parameter estimates	SE	p	Adjusted OR (95% CI)
Contexts of previous exposure to suicide messages
Self-disclosure of suicidal peers (ref: No)	0.26	0.11	.01	1.30 (1.06–1.60)	0.26	0.11	.02	1.30 (1.05–1.61)
Online ventilation posts (ref: No)	0.48	0.11	<.001	1.62 (1.30–2.01)	0.48	0.11	<.001	1.61 (1.29–2.01)
Information from the third party (ref: No)	0.20	0.16	.19	1.22 (0.90–1.66)	0.19	0.16	.22	1.21 (0.89–1.65)
Media report (ref: No)	0.38	0.33	.25	1.46 (0.77–2.79)	0.54	0.35	.13	1.71 (0.85–3.43)
School announcement (ref: No)	−0.18	0.59	.76	0.84 (0.26–2.64)	−0.19	0.65	.77	0.82 (0.23–2.95)

Note. Multivariable logistic regression analysis was used to analyze the data. SE = standard error; OR = odds ratio; CI = confidence interval.

Attitudes toward open discussion of suicide has two outcome levels: ‘objection’ and ‘support’. The response of ‘objection’ is chosen as the reference category.

Model 2 adjusted for gender, school bandings, and suicidal status.

Table 5 presents the results of multivariable logistic regression analyses of the association between attitudes toward open suicide discussion and Internet use concerning platforms and motivations (Table 5). Adjusted for gender, school banding, and suicidal status, using Instagram (AOR = 1.50, 95% CI 1.21–1.85), WhatsApp (AOR = 1.25, 95% CI 1.02–1.53), and Snapchat (AOR = 1.35, 95% CI 1.07–1.71) were associated with increased likelihood of embracing open discussion of suicide. With regard to use motivations, respondents who expressed online for emotional expression (AOR = 1.28, 95% CI 1.04–1.56), life documentation (AOR = 1.42, 95% CI 1.16–1.75), and expression of opinions (AOR = 1.40, 95% CI 1.11–1.76) were more likely to accept open suicide discussion.

Table 5.

Associations of attitudes toward open suicide discussion with use of social media platforms and motivations for online expression (N = 1,676).

Attitudes toward open discussion^a	Model 1				Model 2^b
Attitudes toward open discussion^a	Parameter estimates	SE	p	Crude OR(95% CI)	Parameter estimates	SE	p	Adjusted OR (95% CI)
Social Media Platforms
Instagram (ref: No)	0.43	0.10	<.001	1.54 (1.26–1.89)	0.40	0.11	<.001	1.50 (1.21–1.85)
WhatsApp (ref: No)	0.27	0.10	.006	1.31 (1.08–1.60)	0.22	0.10	.03	1.25 (1.02–1.53)
Snapchat (ref: No)	0.36	0.12	.002	1.44 (1.14–1.81)	0.30	0.12	.01	1.35 (1.07–1.71)
WeChat (ref: No)	−0.27	0.15	.07	0.76 (0.57–1.02)	−0.26	0.16	.10	0.77 (0.57–1.05)
Facebook (ref: No)	0.11	0.16	.46	1.12 (0.83–1.52)	0.11	0.16	.50	1.12 (0.81–1.53)
TikTok (ref: No)	−0.02	0.26	.93	0.98 (0.59–1.63)	0.16	0.28	.57	1.17 (0.68–2.01)
Weibo (ref: No)	−0.41	0.28	.14	0.66 (0.38–1.15)	−0.41	0.29	.15	0.67 (0.38–1.16)
Twitter (ref: No)	0.11	0.27	.69	1.12 (0.65–1.91)	0.04	0.28	.88	1.04 (0.60–1.80)
Motivations for online expression
Express emotions (ref: No)	0.27	0.10	.007	1.31 (1.08–1.59)	0.24	0.10	.02	1.28 (1.04–1.56)
Share lives (ref: No)	0.14	0.10	.17	1.15 (0.94–1.39)	0.14	0.10	.19	1.15 (0.94–1.40)
Record lives (ref: No)	0.38	0.10	<.001	1.46 (1.20–1.78)	0.35	0.10	<.001	1.42 (1.16–1.75)
Express opinions (ref: No)	0.34	0.12	.003	1.41 (1.12–1.76)	0.34	0.12	.004	1.40 (1.11–1.76)
Check-in (ref: No)	0.26	0.13	.05	1.30 (1.00–1.69)	0.26	0.14	.06	1.29 (0.99–1.69)
Build images (ref: No)	0.07	0.18	.69	1.07 (0.76–1.52)	0.08	0.18	.68	1.08 (0.76–1.54)
Attract ‘likes’ (ref: No)	0.06	0.19	.77	1.06 (0.73–1.54)	0.11	0.20	.60	1.11 (0.76–1.63)
Make friends (ref: No)	−0.06	0.21	.79	0.95 (0.62–1.43)	−0.08	0.22	.71	0.92 (0.60–1.42)
Seek help (ref: No)	0.48	0.24	.04	1.61 (1.01–2.57)	0.44	0.24	.07	1.55 (0.97–2.47)

Note. Multivariable logistic regression analysis was used to analyze the data. SE = standard error; OR = odds ratio; CI = confidence interval.

Attitudes toward open discussion of suicide has two outcome levels: ‘objection’ and ‘support’. The response of ‘objection’ is chosen as the reference category.

Model 2 adjusted for gender, school bandings, and suicidal status.

Discussion

To our knowledge, this is the first study that has explored youths’ attitudes toward open suicide discussion, with further examination of associated correlates, favored contexts, and impacts of the Internet. With the mixed-methods approach, three types of attitudes toward public suicide discussion were revealed and reluctance was the dominant attitude among adolescents. Taboo was the major concern, while other barriers identified were consistent with prior research (Fitzpatrick & Kerridge, 2013; Ranahan & Keefe, 2022), including privacy, contagion, embarrassment, and negative experiences. The greater likelihood of supporting open suicide discussion was associated with prior exposure to suicide messages disclosed in peer networks and online postings. Users of Instagram, WhatsApp, and Snapchat, and those who expressed online for ‘self-expression’ were more likely to engage in open suicide discussion.

Our results demonstrated a predominant reluctance among youths to discuss suicide in public, which was contrary to recent research on the general population (Murali & Smitha, 2022), whilst consistent with the study that revealed adolescents’ rejection to have open discussion with suicidal individuals (Schlichthorst et al., 2018). These findings manifest an urgent need to adjust the stigmatized perception of suicide and the call for tailored programs for adolescents. Our results on perceived barriers and benefits corroborated this conclusion. Taboo was recognized as the primary concern owing to the sensitivity of the suicide topic, suggesting the enduring impact of cultural norms in Chinese society (Chan & Cheung, 2022). However, an earlier study showed willingness of discussing the topic of death in most Chinese youths (Kok et al., 2011), and the discrepancy in views might be attributed to the common belief in suicide myths (Pridmore et al., 2016). Therefore, further education is crucial to equip adolescents with appropriate knowledge of the suicide phenomena and prevention. In addition, in line with prior research (Wiggins et al., 2016), emotional support was perceived as the most essential benefit of open suicide discussion. It corresponded to the negative experiences reported in qualitative findings due to the ‘oversimplified and excessively general’ approach, thus emphasizing the significance of offering timely emotional support in parallel with knowledge education in school-based suicide prevention initiatives.

In accordance with previous research, gender differences in the perception of principal facilitators have been identified. Females are more likely to participate in suicide-related communication to achieve external support (Balt et al., 2021), whereas males tend to embrace self-reliance and seek skill development for emotional management (Calver & Michael-Fox, 2021). It suggested the inhibited help-seeking for emotional distress among male adolescents, which might be influenced by masculinity norms and should be addressed in future campaigns. In addition, our findings contribute to the understanding that students from schools with lower academic attainment level showed more opposition to public suicide discussion with a stronger fear of the ‘contagion effect’. It suggested the greater stigma among these students, which might be explained by their overall weak socioeconomic backgrounds, resulting in fewer opportunities to acquire suicide-related knowledge (Park et al., 2015). Hence, additional resources should be directed to lower Band schools for mental health education.

Research on attitudes toward open discussion of suicide and suicidal risks remains scarce. Our results demonstrated that low-risk adolescents were more concerned about the ‘panic effect’. It suggested a poor public tolerance for the discourse on mental health issues and latent discrimination toward suicidal individuals (Fitzpatrick & Kerridge, 2013; Pridmore et al., 2018). This finding aligned with the emphasis on the benefit of ‘raising public awareness’ among youths at high risk of suicide. Previous research found that most at-risk people were more receptive to the open discussion of suicide compared to those who had never attempted suicide (Eskin et al., 2016; Vossos, 2017), indicating a strong need for public attention, empathy, and understanding as well as the importance of relevant public education.

In addition, our findings on the relationship between a supportive attitude and previous exposure to suicide messages in peer networks and online postings highlighted that adolescents preferred specific contexts for open suicide discussion. In conformity with previous research (Morris, 2015), adolescents were more willing to share emotional and suicide issues in a smaller social group, particularly with peers. Trained peer specialists have been found to realize the significance of their role in breaking the taboo around suicide and making it a ‘normal’ topic to discuss (Huisman & van Bergen, 2019). In lieu of professionals, school-based programs should thus consider involving peers with gatekeeper training to be moderators in open suicide discussion.

Besides, we noted an emerging trend of digital communication for suicide discussion. The positive feedback and reassuring messages from the online community might be the reasons for youths’ preference that may motivate their help-seeking for emotional distress (McDermott & Roen, 2016). Prior research has also shown the increasing prevalence of online suicide communication (Gibson et al., 2019). It substitutes conventional (e.g. newspaper) and popular (e.g. television and movie) media in the dissemination of suicide messages, which has been criticized for having detrimental effects on suicide contagion among vulnerable groups (Ortiz & Khin Khin, 2018). Suicide issues and mental health concerns also become more prevalent among children and young adolescents due to increasing online activities (Coyne et al., 2021). Electronic game use, in particular, is associated with higher risks of SAs, psychological distress, and problematic behaviors in those aged 11 to 17 (Rikkers et al., 2016). Hence, suicide prevention studies should involve the younger population, and targeted strategies should be employed. Further research is needed to facilitate a better understanding of their needs for mental health support and to investigate the differences in the younger age groups with regards to their characteristics of Internet use and online communication.

Although the popularity of online suicide communication is well documented, the platform use and motives of young people in relation to public discourse about suicide remain unknown. The present study revealed a correlation between the favorable attitude and the usage of Instagram, WhatsApp, and Snapchat. All three platforms were characterized by their higher level of privacy protection (Waddell, 2016), which ensured anonymity during online suicide discussion. Regarding Instagram and Snapchat, recordability affordance feature offers gratification of retaining privacy while the modality affordance feature allows richer expressive formats than text-based technologies (Alhabash & Ma, 2017). Most prior research adopted traditional SM platforms, such as Facebook (Gomes de Andrade et al., 2018; Prescott et al., 2020), for suicide prevention and intervention initiatives, whereas our findings suggested the suitability of including these youth-oriented fast-evolving SM platforms for future studies on suicide communication considering their high prevalence among adolescents (Plaisime et al., 2020). In addition, our results indicated that the motivation for ‘self-expression’ online was associated with a greater acceptance to open suicide discussion. It demonstrated that most youths engaging in such discussion took it seriously but not for attentional or social purposes. Previous research underscored adolescents’ uncertainty over the use of appropriate language for suicide discussion, which might restrict online expressions (Ftanou et al., 2018) and lead to negative trolling (Niezen, 2013). Thus, instructions for generating destigmatizing messages without normalizing or sensationalizing suicide should be offered and embedded in school education on proper Internet use.

Strengths and limitations

The key strength of the study was its mixed-methods design that enabled us to gain a comprehensive and integrated understanding of this novel topic. However, since the cross-sectional design was used in the quantitative phase, establishing causalities was hardly possible. Future research should evaluate the potential outcomes of public suicide discussion and incorporate population-level statistics for national or cross-cultural analysis. Reliability of the self-reported data was another concern due to the recall biases. Prospective study design should be employed in future investigations to monitor the behavioral patterns concerning participation in suicide talk among youths with various kinds of views. Besides, the findings of this study might not be applicable to adolescents of non-Chinese background or rural regions. In future research, it would be informative to examine and compare the opinions of youths from diverse cultural and socioeconomic backgrounds on the topic of open suicide discussion.

Conclusions

Open discussion of suicide has been encouraged worldwide, yet little is known about the opinions of youths toward this topic. Our findings shed light on the prevalent reluctance toward open suicide discussion, preferred settings including peer networks and online discussion, and facilitative effects of youth-oriented Internet platforms and online ‘self-expression’ motive among Hong Kong Chinese adolescents. These findings would provide insight for the design of school programs, public education, and online interventions for a more robust and honest discussion on suicide among adolescents. Future research should focus on reducing the stigma of public suicide discussion and developing effective training to build youths’ confidence and comfort while discussing suicide in public.

Footnotes

Declaration of Conflicting Interest

The authors declare that there is no conflict of interest.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Public Policy Research Funding Scheme, Policy Innovation and Co-ordination Office, HKSAR, PRC (grant number 2017.A8.075).

ORCID iD

Tai Pong Lam

References

Alhabash

(2017). A tale of four platforms: Motivations and uses of Facebook, Twitter, Instagram, and Snapchat among college students? Social Media + Society, 3(1), 1–13.

Balt

Mérelle

van Bergen

Gilissen

van der Post

Looijmans

Creemers

Rasing

Mulder

van Domburgh

Popma

(2021). Gender differences in suicide-related communication of young suicide victims. PLoS One, 16(5), e0252028.

Braun

Clarke

(2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport Exercise and Health, 11(4), 589–597.

Calver

Michael-Fox

(2021). Under pressure: Representations of student suicide in British documentary television. Mortality, 26(4), 376–393.

Carmichael

Whitley

(2018). Suicide portrayal in the Canadian media: Examining newspaper coverage of the popular Netflix series ‘13 Reasons Why’. BMC Public Health, 18(1), 1086.

Carmichael

Whitley

(2019). Media coverage of Robin Williams’ suicide in the United States: A contributor to contagion? PLoS One, 14(5), e0216543.

Chan

T. M. S.

Cheung

(2022). The “men in grief” phenomenon among suicide bereaved Chinese men in Hong Kong. Death Studies, 46(8), 1845–1852.

Chen

S. S.

Lam

T. P.

Lam

K. F.

T. L.

Chao

D. V. K.

Mak

K. Y.

Lam

E. W. W.

Tang

W. S.

Chan

H. Y.

Yip

P. S. F.

(2022). Motivations for online expression, willingness of online help-seeking, and the risk of suicide among Hong Kong youths: A mixed-methods study. Cyberpsychology Behavior and Social Networking, 25(6), 384–391.

Coyne

S. M.

Hurst

J. L.

Dyer

W. J.

Hunt

Schvanaveldt

Brown

Jones

(2021). Suicide risk in emerging adulthood: Associations with screen time over 10 years. Journal of Youth and Adolescence, 50(12), 2324–2338.

10.

Dempsey

R. C.

McAlaney

Bewick

B. M.

(2018). A critical appraisal of the social norms approach as an interventional strategy for health-related behavior and attitude change. Frontiers in Psychology, 9, 2180.

11.

Eskin

Kujan

Voracek

Shaheen

Carta

M. G.

Sun

J. M.

Flood

Poyrazli

Janghorbani

Yoshimasu

Mechri

Khader

Aidoudi

Bakhshi

Harlak

Ahmead

Moro

M. F.

Nawafleh

Phillips

, . . . Tsuno

(2016). Cross-national comparisons of attitudes towards suicide and suicidal persons in university students from 12 countries. Scandinavian Journal of Psychology, 57(6), 554–563.

12.

Fitzpatrick

S. J.

(2020). Epistemic justice and the struggle for critical suicide literacy. Social Epistemology, 34(6), 555–565.

13.

Fitzpatrick

S. J.

Kerridge

I. H.

(2013). Challenges to a more open discussion of suicide. The Medical Journal of Australia, 198(9), 470–471.

14.

Frey

L. M.

Cerel

(2015). Risk for suicide and the role of family. Journal of Family Issues, 36(6), 716–736.

15.

Ftanou

Cox

Nicholas

Spittal

M. J.

Machlin

Robinson

Pirkis

(2017). Suicide prevention public service announcements (PSAs): Examples from around the world. Health Communication, 32(4), 493–501.

16.

Ftanou

Skehan

Krysinska

Bryant

Spittal

M. J.

Pirkis

(2018). Crafting safe and effective suicide prevention media messages: Outcomes from a workshop in Australia. International Journal of Mental Health Systems, 12, 23.

17.

Gibson

Wilson

Grice

J. L.

Seymour

(2019). Resisting the silence: The impact of digital communication on young people’s talk about suicide. Youth & Society, 51(8), 1011–1030.

18.

Gomes de Andrade

N. N.

Pawson

Muriello

Donahue

Guadagno

(2018). Ethics and artificial intelligence: Suicide prevention on Facebook. Philosophy & Technology, 31(4), 669–684.

19.

Han

Batterham

P. J.

Calear

A. L.

Randall

(2018). Factors influencing professional help-seeking for suicidality. Crisis, 39(3), 175–196.

20.

Harris

K. M.

McLean

J. P.

Sheffield

(2009). Examining suicide-risk individuals who go online for suicide-related purposes. Archives of Suicide Research, 13(3), 264–276.

21.

Huisman

van Bergen

D. D.

(2019). Peer specialists in suicide prevention: Possibilities and pitfalls. Psychological Services, 16(3), 372–380.

22.

Kok

J. K.

Gan

C. C.

Goh

L. Y.

(2011). Youths talk about death: Causes for suicide. International Journal of Social Science and Humanity, 1(2), 145–149.

23.

McDermott

Roen

(2016). Queer youth, suicide and self-harm. Palgrave Macmillan.

24.

Mishara

B. L.

Kerkhof

A. J. F. M

. (2013). Suicide prevention and new technologies: Evidence based practice. Palgrave.

25.

Morris

(2015). Risky bodies: Making suicide knowable among youth. In White

Marsh

Kral

M. J.

Morris

(Eds.), Critical suicidology: Transforming suicide research and prevention for the 21st century (pp. 62–78). University of British Columbia Press.

26.

Murali

T. N. N.

Smitha

(2022). Attitude towards suicide among junior residents- A cross-sectional study. Kerala Journal of Psychiatry, 35(1), 1–8.

27.

Niederkrotenthaler

Braun

Pirkis

Till

Stack

Sinyor

Tran

U. S.

Voracek

Cheng

Arendt

Scherr

Yip

P. S. F.

Spittal

M. J.

(2020). Association between suicide reporting in the media and suicide: systematic review and meta-analysis. BMJ, 368, m575.

28.

Niezen

(2013). Internet suicide: Communities of affirmation and the lethality of communication. Transcultural Psychiatry, 50(2), 303–322.

29.

Ortiz

Khin Khin

(2018). Traditional and new media’s influence on suicidal behavior and contagion. Behavioral Sciences & the Law, 36(2), 245–256.

30.

Park

Kim

M. J.

Cho

M. J.

Lee

J. Y.

(2015). Factors affecting stigma toward suicide and depression: A Korean nationwide study. International Journal of Social Psychiatry, 61(8), 811–817.

31.

Paulus

T. M.

Varga

M. A.

(2015). “Please know that you are not alone with your pain”: Responses to newcomer posts in an online grief support forum. Death Studies, 39(10), 633–640.

32.

Plaisime

Robertson-James

Mejia

Núñez

Wolf

Reels

(2020). Social media and teens: A needs assessment exploring the potential role of social media in promoting health. Social Media + Society, 6(1), 1–11.

33.

Poijula

Wahlberg

K. E.

Dyregrov

(2001). Adolescent suicide and suicide contagion in three secondary schools. Int J Emerg Ment Health, 3(3), 163–168.

34.

Prescott

Rathbone

A. L.

Brown

(2020). Online peer to peer support: Qualitative analysis of UK and US open mental health Facebook groups. Digital Health, 6, 2055207620979209.

35.

Pridmore

Money

T. T.

Pridmore

(2018). Suicide: What the general public and the individual should know. Malaysian Journal of Medical Sciences, 25(2), 15–19.

36.

Pridmore

Varbanov

Aleksandrov

Shahtahmasebi

(2016). Social attitudes to suicide and suicide rates. Open Journal of Social Sciences, 4(10), 39–58.

37.

Ranahan

Keefe

(2022). The bounds of suicide talk: Implications for qualitative suicide research. Health: An Interdisciplinary Journal for the Social Study of Health Illness and Medicine, 26(1), 81–99.

38.

Rikkers

Lawrence

Hafekost

Zubrick

S. R.

(2016). Internet use and electronic gaming by children and adolescents with emotional and behavioural problems in Australia - results from the second child and adolescent survey of mental health and wellbeing. BMC Public Health, 16, 399.

39.

Schlichthorst

King

Turnure

Sukunesan

Phelps

Pirkis

(2018). Influencing the conversation about masculinity and suicide: Evaluation of the man up multimedia campaign using Twitter data. JMIR Mental Health, 5(1), e14.

40.

School Education Statistics Section. (2020). Student Enrolment Statistics, 2019/20 (Kindergartens, Primary and Secondary Schools). https://www.edb.gov.hk/attachment/en/about-edb/publications-stat/figures/Enrol_2019.pdf

41.

Shahtahmasebi

(2014). Suicide survivors. Dynamics of Human Health, 1(4), 596.

42.

Shahtahmasebi

(2016). Do current suicide prevention policies neglect suicide survivors? Journal of Socialomics, 5(4), 1–4.

43.

Shahtahmasebi

Cassidy

(2014). Suicide, mental illness and Maori people. Open Journal of Social Sciences, 2(8), 87–94.

44.

Siu

A. M. H.

(2019). Self-harm and suicide among children and adolescents in Hong Kong: A review of prevalence, risk factors, and prevention strategies. Journal of Adolescent Health, 64(6), S59–S64.

45.

Varga

M. A.

Paulus

T. M.

(2014). Grieving online: Newcomers’ constructions of grief in an online support group. Death Studies, 38(7), 443–449.

46.

Vossos

(2017). Collaborative interprofessional practice to prevent college student suicide. Journal of Interprofessional Education & Practice, 8, 42–46.

47.

Waddell

T. F.

(2016). The Allure of privacy or the desire for self-expression? Identifying users’ gratifications for ephemeral, photograph-based communication. Cyberpsychology Behavior and Social Networking, 19(7), 441–445.

48.

Wiggins

McQuade

Rasmussen

(2016). Stepping back from crisis points: The Provision and acknowledgment of support in an online suicide discussion forum. Qualitative Health Research, 26(9), 1240–1251.

49.

World Health Organization. (2021). Suicide: fact sheets. Retrieved May 24, 2022, from https://www.who.int/news-room/fact-sheets/detail/suicide