The re-development of the guidelines was conducted in three stages: literature search, questionnaire development and Delphi consensus survey rounds.
Literature search
A systematic literature search was conducted to find statements about how someone can help a person who is engaging in deliberate NSSI, including how to determine if someone is deliberately injuring themselves, how to offer short-term assistance to the person, and how to help them seek appropriate professional support. The literature searched included online materials, research publications, and self-help books.
Websites and online materials were searched using the Google search engines of English-speaking countries (Google.com, Google.com.au, Google.co.uk, Google.nz, Google.ca). The search terms 'self injury', 'self harm', `cutting’, as well as help (truncated to include terms such as `helping’ and `helped’) and `friend’ or `family’. The websites returned in the top 50 results from each search were reviewed. Overall, 146 unique websites were reviewed for potential first aid helping actions, with statements found on 57 of these sites.
The research literature was searched through PsycInfo and PubMed, with the terms `self harm’, `self injury’ or `NSSI’ searched for in the title and abstract, and `help’, `prevent’, `assist’, `support’ or `care’, as well as `friend’ or `family’ searched for throughout. Exclusion terms, comprising `cell suicide’, `assisted suicide’, and `suicide attack’ were also entered to improve relevance of results. Results were also limited to articles published after 2004, as the searches aimed to find new articles that not been included in the literature search in the earlier literature search for the original version of the guidelines. Search results returned 834 unique articles, with 22 considered relevant for review, and eligible statements found in 2 of these.
To locate relevant books, a search of Amazon.com was also conducted using the search terms `self harm’, `cutting’, `self- injury’, `help’ and `friend’ or `family’. Nine books were returned, with 5 of these considered relevant. These 5 books were purchased and read, with eligible statements found in all but one.
Questionnaire development
Relevant helping statements that were found in the literature search, as well as the statements included in the previous Delphi questionnaires [
25] formed the content of the first questionnaire. The statements included in the questionnaire were agreed upon by all three authors as being actionable by the first aider, as relevant to the role of a first aider, as well as being clear and non-ambiguous in its meaning. Examples of the types of statements included in the questionnaire include `The first aider should discuss their concerns with the person in a private place’ and `The first aider should let the person know the ways in which they are willing to help the person’. These statements were grouped into categories based on common thematic content. Statements were edited so that those with similar content were combined in order to reduce repetition throughout the questionnaire. Statements were also edited to improve clarity by systematic re-wording or elaboration through examples. This editing occurred in meetings of a working group, which were held to edit and develop a draft of the questionnaire, including its categories and structure of statements. The working group comprised the authors of this paper who are all researchers with previous experience in conducting research using the Delphi methodology and on MHFA training programmes.
The questionnaire was completed online through an online survey website, SurveyMonkey. Participants were given a two to three week time period to finish the questionnaire for each of the three rounds of the Delphi survey process. The questionnaires were able to be completed at times that were convenient to participants, and in multiple sessions if desired.
Delphi consensus survey rounds
The consensus survey was conducted using the Delphi method [
32]. The Delphi method involved identifying and recruiting panels of experts in the field of NSSI to rate the importance of helping statements. Statements that achieved substantial consensus regarding their importance for inclusion in the guidelines were considered as the recommended actions to take to help someone who is self-injuring.
Participants were recruited from developed English-speaking countries (Australia, Canada, Ireland, New Zealand United Kingdom, and the United States) to join one of two expert panels representing two areas of expertise: professionals or consumers. To be considered as having expertise in NSSI, panellists were required to have past personal experience in self-injuring, or professional experience working in the field of NSSI prevention and intervention (i.e. as a researcher, clinician, mental health nurse, social worker). Potential professional panellists were identified as experts through their involvement with NSSI research, prevention and intervention organisations, while potential consumer panellists were identified through their advocacy roles in NSSI prevention.
The profession panel was recruited through editorial boards of relevant academic journals and suicide prevention organisations. The heads of these boards and organisations were emailed an invitation to participate and a copy of the project's plain language statement, asking that these be forwarded on to the relevant members. The academic journal editorial boards contacted included `Crisis’, `Suicide and Life-Threatening Behavior’ and the `Journal of Clinical Psychology’ special NSSI edition (November 2007, Volume 63, Issue 11). Professional panellists were also recruited through NSSI and suicide prevention organisations, such as the International Association of Suicide Prevention, Suicide Prevention Australia, the Australian Suicide Prevention Advisory Council, the American Foundation for Suicide Prevention, the American Association of Suicidology, the Canadian Association for Suicide Prevention, the Suicide Prevention Resource Center, the University of Oxford Centre for Suicide Research and Suicide Prevention Information New Zealand. Members of editorial boards and prevention organisations who were interested in participating were asked to give an expression of interest by contacting the authors and to provide an outline of their experience working with NSSI populations. Contact details for the authors were provided to potential participants in the Plain Language Statement sent to the editorial boards and prevention organisations. Interested persons who proved to have direct professional experience working in the field of NSSI prevention and intervention (i.e. as a researcher, clinician, mental health nurse, social worker) were added to the expert panel. Professionals were also asked to nominate any colleagues who they felt would also be appropriate panel members. The professional panel comprised 28 panellists, some of whom had multiple roles, including 7 professors and associate professors in psychiatry or psychology, 7 psychologists, 6 researchers, 6 social workers, 3 counsellors, 2 psychiatrists, 2 mental health service directors, and 3 who worked in other mental health support roles. This panel represented global professional opinions in NSSI prevention, demonstrated through their demographics presented in Table
1.
Table 1
Participant characteristics (data collected in Round 1)
Mental health professionals | 28 | 28-69 | 40 | 75 | 13 | 10 | 3 | 2 |
Consumers | 33 | 18-71 | 32 | 91 | 13 | 16 | 2 | 2 |
The consumer panel was recruited through depression and mental disorder advocacy organisations, including
beyondblue: the national depression initiative (Australia), Depression and Bipolar Support Association (United States), National Alliance of Mental Illness (NAMI) (United States), Depression Alliance (United Kingdom), and Depression Support Network (New Zealand). Email invitations and plain language statements were emailed to the advocacy group coordinators for the information to be forwarded to the group members. Consumers who had written websites that offered support and information to other consumers, as well as promoted recovery from NSSI, were also identified as potential panellists and were invited to participate through email invitation. Consumers who were interested in participating were asked to give an expression of interest by contacting the authors and providing an outline of their first-hand experiences of NSSI. Contact details for the authors were provided to potential participants in the Plain Language Statement sent in the email invitations. Interested persons who claimed to have past personal experience in self-injuring and were comfortable reflecting on these experiences were added to the expert panel. Consumers were also asked to nominate anyone they knew who they felt would also be appropriate panel members. Thirty-three NSSI consumer advocates were recruited to this panel, with demographic characteristics also included in Table
1.
The outcome for each item was determined using the following criteria:
1.
Statements that were rated as essential or important by 80% or more of the members in both panels were endorsed as helping actions to be included into the guidelines.
2.
Statements were re-rated in a subsequent round of the questionnaire if:
a.
Statements were rated as essential or important by 70-79.9% of the panel members
b.
Statements were rated as essential or important by 80% of more of one panel, but less than 80% by the other panel
3.
Statements that were rated as essential or important by less than 70% of both panel members were excluded.
In Round 1, panel members were also asked to provide feedback through a textbox at the end of each section of the questionnaire. This feedback textbox was intended for use by panellists to suggest helping actions that were not covered in the questionnaire, but generally panellists used the textboxes to provide rationales for their ratings. The comments made were reviewed by the working group. Suggestions that contained novel ideas were used to create new helping statements to be included in the subsequent Round 2 questionnaire. Also, statements that received feedback suggesting ambiguity in the interpretation of its meaning were re-phrased to make them clearer and included in Round 2. Statements from Round 1 that met the criteria to be re-rated were also included in the Round 2 questionnaire.
The third and final questionnaire was comprised of new statements that were developed from Round 1 feedback and presented for the first time in Round 2, but required re-rating in a further round. Items that still did not achieve consensus after being re-rated were rejected from inclusion in the guidelines.
Following each of the three rounds, each panellist was sent a report containing a summary of the results from the previous round. The report included a list of the statements that had been endorsed for inclusion in the guidelines, as well as a list of the statements that had been rejected from inclusion. The statements to be re-rated in the subsequent round were also included, with the report personalised to include the individual panellist's rating for each statement, as well as a table summary of each panel's ratings for the statement.
The statements that were endorsed across the three survey rounds were compiled. These statements were then used to form the guidelines, with working group meetings held to finalise structure and wording. The final draft copy of the guidelines was then disseminated to panellists for their final comment on the document.