Background: Research has established that the media can influence suicidal behaviour [5]. This may occur through processes of contagion, imitation and transmission of methods. The Internet is a relatively new medium that is transforming the environment for the communication of information about suicide. As well as providing an outlet for traditional sources of media such as news reports, which are known to influence suicidal behaviour, it also poses additional and unique challenges. It allows instantaneous, repeated and global sharing of a broad array of information, including detailed 'how to' information about suicide methods. It also enables creation and interactive exchange of user-generated content through social media platforms, such as chat rooms, blogging, video sites and social networking sites [6]. Information may derive from individuals as well as institutions, meaning that new and anonymous voices can contribute, uncensored to dialogue about suicide [1], and create pro-suicide 'extreme communities' [7] which may encourage suicide as a response to difficulties. Internet use may therefore increase access to suicide [8] both physically, though transmission of methods, and cognitively by endorsing this as a culturally available response to distress. The Internet and social media are fundamental to the way that many people share information, opinions and ideas. There are over two billion Internet users worldwide and use is most prevalent amongst young adults (aged 16-24years) [1].

Concern about the potential for the Internet to cause suicide is at an all time high with mounting cases of internet-related suicide reported in the popular and academic press (e.g. [9] [10] [11]). These provide examples of individuals using suicide methods discovered online, accessing drugs for overdose from unregulated online pharmacies, and being encouraged to attempt suicide through participation in chat room discussions. Social networking has been implicated in recent suicide clusters such as Bridgend. While such examples demonstrate that the Internet can contribute to suicidal behaviour in a range of ways, much of this literature in this field is impressionistic. There is little systematic empirical evidence. Studies of Internet content reinforce the notion that detailed information about suicide methods can be accessed easily [12] [13], however, these studies cannot inform about actual user behaviour in accessing these sites or how such information is interpreted. Prevalence estimates of suicide-related internet use are currently lacking. Existing studies of users tend to be based on self-selecting samples responding to online adverts and questionnaires [2] [3]. While furthering concerns about possible harmful effects of the Internet, these studies also draw attention to the possibility that for some individuals, the Internet provides a constructive and supportive environment for dealing with suicidal feelings. This raises the prospect of exploiting the Internet for prevention purposes but to achieve this requires understanding of what users seek online and what material will engage them. Detailed empirical study examining how, when and for what purpose the internet is used by individuals with suicidal feelings and how this shapes suicidal behaviour is now required to inform attempts to both restrict harmful use and to capitalise on the Internet's potential to offer population level suicide prevention.

Aims: To provide detailed empirical evidence about the use of the Internet for suicide-related purposes by young adults and how this influences suicidal feelings and behaviour.

Design: This project will involve using data from the section on deliberate self-harm in the forthcoming ALSPAC 20 years questionnaire, including questions about suicide-related Internet use formulated by the applicants (LB and DG). The data will be used to conduct quantitative analysis and derive a purposive sample for further qualitative data collection. This project forms part of a larger application, which will also explore Internet amongst other groups.

1. Quantitative analysis

This is a descriptive, exploratory study and so there are not specifc hypotheses. The main objectives will be to:

1) provide prevalence estimates of suicide-related internet use amongst young adults, including those who self-harm and/ or experience suicidal thoughts

2) to investigate socio-demographic and clinical characteristics associated with use.

Exposure variables: deliberate self-harm, suicidal thoughts

The outcome variable will be whether or not the Internet has been used for suicide-related purposes.

Data analysis: Descriptive statistics, bivariate analysis and logistic regression techniques will be used to:

* Estimate the prevalence of suicide-related Internet use by young adults

* Describe the nature of Internet use including: frequency; material accessed; use of search engines; engagement with chat room and other discussion forums; use of support/ prevention sites.

* Investigate the association between Internet use and socio-demographic characteristics; self-harm (including frequency, method and intent); suicidal thoughts; and formal help-seeking behaviour.

After considering total prevalence and patterns of Internet use amongst all responders, analysis will focus on 3 sub-groups, individuals reporting (past year): 1) self-harm; 2) self-harm with suicidal intent; 3) suicidal thoughts. Results will be presented as percentages or odds ratios with 95% confidence intervals and p-values. Data analysis will be performed using STATA (v12).

2. Qualitative interviews

The main objectives will be to:

1) derive in-depth qualitative accounts of Internet use exploring: how, why and when the Internet is accessed; the material that is viewed and how this is interpreted; how use changes over time; and how use shapes suicidal behaviour.

2) investigate how supportive/ preventative web material is evaluated by Internet users who experience suicidal thoughts/ behaviour.

Sampling: Respondents who report recent (past year) self-harm/ suicidal thoughts and that they have used the Internet for suicide-related purposes in the 20 years questionnaire will form a sampling frame for in-depth qualitative interviews. We will seek to use maximum variation sampling to obtain a diverse sample in terms of gender, nature of self-reported self-harm and suicidal thoughts (e.g. frequency, method used, suicidal intent), and nature of Internet use. In particular, we will seek to include participants who report visiting support and prevention sites as well as those visiting pro-suicide sites. The aim will be to conduct around 25 interviews though exact sample size will be determined according to the number of participants required to achieve a consistent and refined understanding of the main themes across a diverse group. Based on our previous experience of qualitative research in this area, we anticipate that some respondents will yield extremely rich and complex data as multiple episodes of suicidal behaviour and Internet use are likely to be presented within single cases. Two of our previous studies with mentally distressed young adults and near-fatal suicide attempters have been based on sample sizes of 23 [14] and 22 [15] participants.

Data collection and analysis: ALSPAC assistance will be required to obtain a sample only. All further data collection will be undertaken independently by the research team headed by Biddle et al. Interviews will be in-depth and semi-structured. They will be conducted within the interpretive tradition[16] and in keeping with an ethnographic grounded theory approach [17]. Detailed narrative accounts of Internet use in relation to suicidal feelings and behaviour will be sought, focusing on:

* How (including search strategies), why and when (for example, along the suicidal pathway) the Internet is accessed;

* how Internet use evolves over time;

* the material that is viewed (including types of site and participation in chat rooms) and the meanings/ evaluations attached to this;

* the impact of use on suicidal feelings and behaviour;

* the relative import of the Internet alongside other sources of information in shaping suicidal behaviour

Examples will be gathered of Internet use that has encouraged or facilitated suicidal behaviour and of that which has prevented this and facilitated coping or recovery. Computer access will be available at the interviews and where appropriate, participants will be asked to take the interviewer through sites accessed, identifying material that they viewed and their perceptions of this. Participants will also be shown selected prevention/ support sites and asked for their evaluations of these.

A topic guide will be used to ensure consistency and that the research aims are fully addressed, but minimal prompting will be used. Participants will be encouraged to raise issues they consider of importance, including any relevant additional areas not covered by the guide. In this way, data collection will be grounded in the viewpoints and experiences of interviewees without imposing meanings. Interviews will be conducted in small batches with analysis performed simultaneously. This will allow for an iterative process in which emerging questions and ideas will be incorporated into the topic guide and explored with subsequent respondents until a refined level of understanding is reached. Interviews will be conducted at the respondent's home or the University according to their preference.

Full ethical approval will be sought before undertaking the study and a protocol will be developed for use should a situation arise where a respondent became distressed or disclosed information indicating possible risk. This will be based upon protocols developed by the team for use in similar studies. The study team has clinical expertise (JP) and extensive experience of conducting research with vulnerable populations. These include studies exploring non-help-seeking in young adults with mental distress [14] and factors influencing choice of suicide method amongst individuals who have survived a near-fatal attempt [15]. In the latter study we assessed respondent well-being before and after the qualitative interview and found that most respondents found participation cathartic and reported an improvement in mood [18].

Interviews will be audio-recorded with participants' consent and transcribed in full. First, transcripts will be coded for emerging themes and analysis will proceed according to the method of constant comparison, data relating to each code being retrieved and compared within and across individuals. Analytical grids will be used to map out relationships between codes. Second longitudinal case studies will be prepared to explore how Internet use may evolve and impact upon suicidal behaviour over time. Case studies of negative and positive Internet use will also be prepared and contrasted to identify their key features.

References

[1] Durkee T, Hadlaczky G, Westerlund M CV. Internet pathways in suicidality: a review of the evidence. International Journal of Environmental Research and Public Health 2011. 8:3938-3952.

[2] Harris K, McLean J, Sheffield J. Examining Suicide-Risk Individuals who go online for

suicide related purposes. Archives of Suicide Research. 2009. 13:3, 264-276.

[3] Eichenberg C. Internet message boards for suicidal people: a typology of users. Cyberpsychology and Behaviour. 2008. 11:1, 107-113

[4] Hawton K, Rodham K, Evans E, Weatherall R. Deliberate self-harm in adolescents: self-report survey in schools in England. BMJ 2002. 325: 1207-11

[5] Hawton K, Williams K. Influences of the media on suicide. BMJ 2002. 325: 1374-5

[6] Luxton D, June J, Fairall J. Social media and suicide: a public health perspective. American Journal of Public Health 2012. 102:195-199.

[7] Bell V. Online information, extreme communities and internet therapy: Is the interne good for our mental health? Journal of Mental Health 2007.16:445-457.

[8] Florentine J, Crane C. Suicide prevention by limiting access to methods: A review of theory and practice. Social Science and Medicine 2010. 70:1626-1632

[9] Prior T. Suicide methods from the Internet. American Journal of Psychiatry 2004. 161(8): 1500-1501

[10] Alao A, Soderberg M, Pohl E, Alao A. Cybersuicide: review of the role of the Internet on suicide. 2006. Cyberpsychology and Behavior. 9(4):489-493

[11] Becker K, Mayer M, Nagenborg M, El-Faddagh M, Schmidt M. Parasuicide online: can suicide websites trigger suicidal behaviour in predisposed adolescents? Nordic Journal of Psychiatry 2004. 58:111-114.

[12] Biddle L, Donovan J, Hawton K, Kapur N, Gunnell D. Suicide and the Internet. BMJ. 2008. 336(7648): 800-802

[13] Recupero R, Harmss E, Nobel JM. Googling suicide: surfing for suicide information on the Internet. Journal of Clinical Psychiatry. 2008. 69(6): 878-888

[14] Biddle, LA, Donovan, J, Sharp, D & Gunnell, D. 'Explaining non-help-seeking amongst young adults with mental distress: a dynamic interpretive model of illness behaviour', Sociology of Health and Illness. 2007. 29 (7): 983-1002.

[15] Biddle, L, Donovan, JL, Owen-Smith A, Potokar, JP, Longston D, Hawton K, Kapur N & Gunnell, DJ. 'A qualitative study of factors influencing the decision to use hanging as a method of suicide', British Journal of Psychiatry. 2010. 197 (4): 320-325.

[16] Schwandt T. Three epistemological stances for qualitative inquiry: interpretivism, hermeneutics, and social constructionism. In: Handbook of qualitative research (ed N Denzin, Y Lincoln) Sage, 2000.

[17] Glaser B, Strauss A. The discovery of grounded theory: strategies for qualitative research. New York: Aldine De Gruyter, 1967.

[18] Biddle L, Cooper J, Owen-Smith A, Klineberg E, Bennewith O, Hawton K, Kapur N, Donovan J, Gunnell D. Qualitative interviewing with vulnerable populations: individuals' experiences of participating in suicide and self-harm based research. Submitted to Journal of Affective Disorders.