REFLECTIONS ON THE EXPERIENCES AND NEEDS OF ADOLESCENTS WHO HAVE ATTEMPTED SUICIDE: A QUALITATIVE STUDY

Suicide and attempted suicide as a response to crisis are growing alarmingly among the young. Thomerson (2002:30) states that “Youth suicide is a growing epidemic... It is currently the third leading cause of death of 15to 24-year-olds, and the fourth leading cause of death among 10to 14-year-olds. Nearly 4600 kids killed themselves in the United States, and approximately 46 000 others tried”. Schlebusch (1995:3) has confirmed the high incidence of attempted suicides by postulating that for every completed suicide there are at least eight to ten attempted suicides. In a study by Terblanche, Bates, Bokwana, Brits, Corder, Cuendet, Davis, Donson, Mbengu, Ngesi, Olivier, Roger, Stokwe, Totana and Viljoen (1999:148) on risk behaviours of high school learners in Port Elizabeth, South Africa, it was found that 25% of a sample of 2198 learners that had responded to the specific item in the questionnaire on having thoughts the past year, had such thoughts in the year preceding the study. From the total 25% who had thoughts about suicide, 16% had told someone about such thoughts. 14% of the sample of 2201 who responded to the specific item in the questionnaire on attempting suicide in the year of the study, in fact did attempted suicide, and 42% (or 122 respondents) of the learners who attempted suicide needed to seek professional treatment. Terblanche et al. (1999:81) concluded that suicide ideation was higher amongst female respondents than among their male counterparts. However, more male respondents actually attempted suicide which necessitated professional treatment following such incidents.


INTRODUCTION AND PROBLEM FORMULATION
Suicide and attempted suicide as a response to crisis are growing alarmingly among the young. Thomerson (2002:30) states that "Youth suicide is a growing epidemic… It is currently the third leading cause of death of 15-to 24-year-olds, and the fourth leading cause of death among 10-to 14-year-olds. Nearly 4600 kids killed themselves in the United States, and approximately 46 000 others tried". Schlebusch (1995:3) has confirmed the high incidence of attempted suicides by postulating that for every completed suicide there are at least eight to ten attempted suicides. In a study by Terblanche, Bates, Bokwana, Brits, Corder, Cuendet, Davis, Donson, Mbengu, Ngesi, Olivier, Roger, Stokwe, Totana and Viljoen (1999:148) on risk behaviours of high school learners in Port Elizabeth, South Africa, it was found that 25% of a sample of 2198 learners that had responded to the specific item in the questionnaire on having thoughts the past year, had such thoughts in the year preceding the study. From the total 25% who had thoughts about suicide, 16% had told someone about such thoughts. 14% of the sample of 2201 who responded to the specific item in the questionnaire on attempting suicide in the year of the study, in fact did attempted suicide, and 42% (or 122 respondents) of the learners who attempted suicide needed to seek professional treatment. Terblanche et al. (1999:81) concluded that suicide ideation was higher amongst female respondents than among their male counterparts. However, more male respondents actually attempted suicide which necessitated professional treatment following such incidents.
On the basis of their research findings Terblanche et al. (1999:167) concluded that the unprecedented rise in suicide and attempted suicide, especially amongst females and African culture groups, necessitated closer scrutiny in follow-up research studies, particularly through a qualitative lens.
The researcher supports this recommendation for further qualitative research on adolescent suicide and attempted suicide. This is in view of the fact that empirically based literature on adolescents' own accounts of their attempted suicides, with specific reference to their experiences and needs in this regard, seems to be lacking. Literature on the epidemiology and etiology of adolescent suicide seems to be more available (Ackerman, 1998;Gutierrez, 1999;Rabe, 1993;Thomerson, 2002). A major limitation in these studies is the fact that the suicideprone adolescent and the circumstances surrounding his or her suicide have been the focus of study. The voice of the particular adolescent, as self-knowing subject speaking authoritatively about his or her thoughts, feelings, behaviour related to the attempted suicide, and the specific needs resulting from this experience, has been largely silent. Although the efforts of researchers up to this point have informed our understanding of parasuicide, the participants' perspective, or voice, is increasingly being recognised as essential for a comprehensive or holistic understanding of the topic. This is even more apparent if intervention and prevention strategies which are youth-informed and youth-driven are to be planned for addressing the needs of this particular client group (Kienhorst, De Wilde, Diekstra & Wolters, 1995:623;Paulson & Worth, 2002:86).
From these introductory remarks a research problem needs to be derived. The research problem becomes the foundation for the entire research process. It refers to the general or substantive area of focus for the research and includes situations characterised by doubt and ignorance (Grinnell & Williams, 1990:62;Rothery in Grinnell, 1993:17;Strauss & Corbin, 1998:35;Yegidis & Weinbach, 1996:42, 49).
The problem statement for this research was postulated as follows: suicide and attempted suicide are becoming a growing epidemic among youths (Fritz, 2001;Thomerson, 2002). The researcher therefore embarked upon this research project in the light of the research recommendation by Terblanche et al. (1999) and in view of the paucity of literature articulating the experiences and needs of adolescents who had attempted suicide, as selfknowing subjects speaking authoritatively about their experiences.
The objective of the research project was to explore and describe qualitatively adolescents' experiences and needs as related to their attempted suicide. Creswell (1994:70) states that, whereas quantitative research is based on hypotheses, qualitative research stems from research questions. According to Strauss and Corbin (1998:35), the research question refers to the specific query to be addressed by this research and sets the parameters of the project, and suggests the methods to be used for data-gathering and analysis. The research question that delineated the focus of this study was: what are the experiences and needs of adolescents who have attempted suicide?

RESEARCH QUESTION
The qualitative paradigm appeared to be the appropriate choice because of the apparent dearth of basic information and research, particularly within the ambit of social work pertaining to this subject area (that is, the experiences and needs of adolescents who had attempted suicide). This paucity is especially apparent in South Africa and in terms of information from the perspective of self-knowing adolescents who had attempted suicide. The researcher found that qualitative research aided him in the quest to obtain subjective, personal data pertaining to the research problem. The research employed an inductive, contextual, interpretive and idiographic stance in its interpretation of data. Within this holistic and discovery-orientated approach (from the "what?" perspective) it was primarily geared towards a hermeneutic stance (that is, an interest in meaning). This research was concerned with how people make sense of their lived experiences, which provides a postmodern or social constructionist perspective to this study (De Vos in De Vos et al., 1998:242).
Phase three: Selecting the qualitative research design Rubin and Babbie (1993:1) state that "... a research design is viewed as a blueprint or formula to be selected at a certain stage within the broad research process where the research question ... and the corresponding theoretical paradigm or base selected for the study are linked to strategies upon which the study can be executed, and not as the whole research process". Denzin and Lincoln (1994:202-203) identify seven strategies of inquiry, or tools that can be used to design qualitative research. The researcher chose phenomenology as the particular qualitative strategy of inquiry. According to Crabtree and Miller (1999:28), "… phenomenology seeks to understand the lived experiences of individuals and their intentions within their 'life-world'. It is the search for essence. It answers the questions, 'What is it like to have a certain experience?' 'What is the essence of this particular experience?". In order to accomplish this, researchers must "bracket" their own preconceptions, enter the subject's lifeworld; place themselves in the "shoes" of the subject by means of naturalistic methods of study, and use the self as an experiencing interpreter.
The rationale for the use of this design was to investigate, as thoroughly as possible, the phenomena of the experiences and needs of adolescents who had attempted suicide. Coupled with the phenomenological strategy of inquiry were explorative and descriptive modes of investigation. The exploration of the experiences and the needs of adolescents who had attempted suicide enabled the researcher to provide a description from the perspective of adolescents who had first-hand experience.

Phase four: Preparing for data collection
This phase embraced several decisions that were taken before the implementation of data collection (Phase five). They included the following: Boundaries or parameters for data collection were identified, as well as the setting. The geographical boundary was delineated as the Southern Cape and included the towns of George, Oudsthoorn, Mossel Bay, Riversdale and Ladysmith. The interviews took place at the homes of the research participants, or of the research assistants, or any other appropriate location of the participants' choice. The actors were adolescents between the ages of 13 and 18 years who had attempted suicide. The events (what the actors would be interviewed about) and the process entailed investigating the evolving nature of events undertaken by the actors within the setting. The two latter parameters are discussed in detail in Phase five.
The population or universum was defined to include all adolescents who had attempted suicide in the Southern Cape, particularly in the towns of George, Oudsthoorn, Mossel Bay, Riversdale and Ladysmith. (The rationale for inclusion of only these towns in the Southern Cape was that the research assistants were rendering social work services to client systems in these towns.) A sample was framed and developed to reflect the population or the total set. Because of the stigma attached to suicide and attempted suicide, adolescents who make up this particular group are an almost hidden or invisible population. Therefore it becomes difficult, if not impossible, to arrive at a conclusion as to what number would constitute a representative sample of such adolescents. For this reason, together with the fact that this study is qualitative in nature, findings cannot be generalised to a broader population.
Therefore a purposive, non-probability sampling technique was used (Rubin & Babbie, 1993:367). The research assistants made use of informants to bring them into contact with potential participants. Fourteen social workers, a teacher and a psychologist assisted in this regard.
The roles which the research assistants would assume were delineated and planned. The roles of the research assistants in this study can be described as "observer-as-participants" or as "non-concealed, minimal participants" who identified themselves as researchers and interacted casually with the informants during the interviewing process (Gold in Rubin & Babbie, 1997:379;Yegidis & Weinbach, 1996:151-152).
Informants involved were contacted and the objectives and the procedures of the research project were explained to them. The informants then had the task of identifying potential participants, making contact with them telephonically and informing them of the research project, as well as inquiring about their willingness to participate. If willingness was indicated, the identifying particulars and a contact telephone number were forwarded to the particular research assistant. The research assistants then had the task of making contact with the potential research participants and obtaining their co-operation. In this case the contact process began with a phone call and was followed up with a letter requesting an interview. In this sample letter the researcher and research assistants explained the following points: the purpose of the study; under whose auspices the study would be conducted; why the adolescents in question were requested to participate; what would be asked of them; where the interviews would take place; the duration of interviews (for example, the length and number of interviews); the use and purpose of any mechanical recording devices; what else might be expected of them (such as, reviewing transcripts or notes, discussing the interpretation); ethical issues of confidentiality management of the information (that is, where tapes, notes and/or transcripts would be stored and what would happen to them when the study was concluded); and what would become of the findings. Handling volatile or illegal information and informing research participants of any personal risks connected to their participation was also discussed. A voluntary consent form was also issued to each adolescent to be completed before the interviews took place and parental consent was also secured before the interviews commenced.
The researcher, in collaboration with the research assistants, designed a tentative protocol for noting observations in the field, namely a single page with a dividing line down the centre, separating descriptive and reflective notes which would serve as a guideline for the organisation of raw data obtained during interviews. The research assistants also made use of a dictaphone to record informants' verbal and non-verbal comments. For ethical reasons, this took with the agreement with the adolescents being interviewed.

Phase five: Data collection and data analysis
The in-depth semi-structured (or focused) interview was selected as the method of data collection. Interviews were transcribed verbatim immediately after they were completed (De Vos in De Vos et al., 1998:48). Interviewees were given an opportunity to "warm up" by being asked some general questions that gave them some practice in talking to the research assistants and also allowed them to organise their thoughts. The interviews were scheduled for a maximum of 1 hour per interview, but were terminated when information was perceived as becoming redundant, or when the interviewees began to repeat themselves (Lincoln and Guba, 1985:271).
Data collection and data analysis took place simultaneously. According to Yin in Krueger (1994:140), data analysis consists of "…examining, categorizing, tabulating or otherwise recombining the evidence, to address the initial propositions of a study". This is a creative, eclectic process and no set recipes exist (Marshall & Rossman, 1989:23;Schoeman & Botha, 1991:56;Tesch in Creswell, 1994:153). Creswell (1994:153) contends that the researcher is engaged in several activities during qualitative data analysis. These include collecting the data; sorting the data into categories; formatting the data into a coherent story or picture, and writing the qualitative text. Rose (1982:118) states that the process of moving from data to conceptualisation and theorisation is the most distinguishable aspect of qualitative research. This study employed the eight steps provided by Tesch (in Creswell, 1994:155) to analyse the data systematically, by segmenting it into words or categories that subsequently formed the basis of the emerging story or picture.

Phase six: Data verification
The process of data verification followed Guba's model of trustworthiness (in Krefting, 1991:215-222), which identifies four criteria and strategies for ensuring and establishing trustworthiness and which are therefore used to assess the qualitative research process undertaken.
The first criterion addressed in establishing trustworthiness is that of truth value, that is determining to what extent the findings are a true reflection of the "life-worlds" of the participants, as experienced and described by them. The strategy for establishing truth value is credibility. The particular actions taken to achieve credibility include triangulation (that is, triangulation of data sources and triangulation of investigators), peer examination, interviewing techniques and research assistants' authority.
The second criterion is applicability, or the degree to which findings can be applied to other contexts or settings and groups (that is, generalisability). Transferability was the strategy employed to attain applicability. Two perspectives of applicability for qualitative research were delineated (Krefting, 1991:216). The first perspective holds that applicability is not seen as being relevant to qualitative research, as it proposes to describe experiences or phenomena which cannot be generalised to other experiences or phenomena (Krefting, 1991:216). The second perspective proposed by Guba in Krefting (1991:216), however, claims that "fittingness" is the criterion against which the applicability of qualitative research can be assessed. As was the case in the study in question, findings fit into contexts outside the study situation that can be determined by transferability was achieved by documenting dense descriptions of the research methodology and by working contextually so that procedures could be duplicated accurately.
The third criterion of Guba's model (in Krefting, 1991:215-222) is termed consistency, which is concerned with the extent to which the replication of the study in a similar context or with similar informants will produce the same results. Dependability was the strategy used to ensure consistency. The actions that were taken to ensure dependability in this study included: peer examination of the research methodology and implementation, triangulation, a dense description of the research methodology, and finally the implementation of a code-recode procedure (Krefting, 1991:216-217).
Neutrality is the fourth and final criterion, and refers to the extent to which the study's findings are free from bias. Lincoln and Guba in Krefting (1991:217) propose that neutrality in qualitative research should consider the neutrality of the data rather than that of the researcher, and thus suggested confirmability as the strategy to achieve neutrality. In this study, triangulation was employed to achieve confirmability (Krefting, 1991:221-222).

DISCUSSION OF FINDINGS
Data were collected through semi-structured interviews with a total of 17 adolescents who participated in this study. The grand tour question that delineated the focus of this study was as follows: what are the experiences and needs of adolescents who have attempted suicide?
The following open-ended statements and questions were asked at each interview: Tell me (the story of) how you tried to commit suicide.
How did you feel before the attempt… during the attempt… and after the suicide attempt?
Tell me how and by whom you were supported after the suicide attempt, and if you would have liked be supported differently.
The answers to these questions were transcribed verbatim and analysed according to the eight steps for analysing qualitative generated data provided by Tesch in Creswell (1994:153).
The ensuing discussion on the research findings will be divided and presented in two sections, namely: a biographical profile of the participants; and a discussion of the themes, sub-themes, categories and sub-categories that emerged from the process of data analysis and the consensus discussion between the researcher and the research assistants. Table 1 below gives a biographical profile of the participants who participated in this research study.

Themes, sub-themes, categories and sub-categories: A thematic discussion
The following themes emerged from the data collected: Reasons for the suicide attempts; Feelings associated with the suicide attempts; Sources and nature of support received after the suicide attempts; Recommendations on alternative and/or complementary support desired after the suicide attempt.
The first theme, reasons for the suicide attempts, emerged spontaneously from the request to the participants to tell the stories of their attempts at suicide and did not form part of the objective of this research project.
The theme of the feelings associated with the suicide attempts derived from the question, "How did you feel before the attempt… during the attempt… and after the suicide attempt?" The themes on the nature and sources of support offered to the participants after the suicide attempts, and the recommendations from them about alternative and/or complementary support they would have wished for, emerged from the answers to the request: "Tell me how and by whom you were supported after the suicide attempt, and if you would have liked be supported differently".
Each of these themes with their accompanying sub-themes, categories and sub-categories will be discussed in the remainder of this presentation and be subjected to a literature control.

Reasons for the suicide attempts
According to Ackerman (1998:19) and Joan (1986:38), there is no one reason for attempted suicide, but rather a combination of complex and contributing factors, although one incident can serve as a trigger for a suicide attempt. Table 2 below indicates the reasons cited by the participants for attempting suicide and provides excerpts from the interviews to support the various sub-themes.
In his study on problem drinking in the family and youth suicide, Fernquist (2000:551) refers to Husain and Vandiver, who note that family violence may increase youth suicidality. He further notes that the data revealed a link between youth suicide rates and problem drinking in the home (Hurry, 2000). Roberts (in Bezuidenhout, 1998:76) is of the opinion that in families where parents are not living together (are separated or divorced) the chance of adolescent suicide increases. Pfeffer (in Stoelb and Chiriboga, 1998:361) verifies this trend that adolescents who have been exposed to sexual abuse and molestation are prone to suicide and attempted suicide. A study on parasuicide confirms that when an adolescent is in a relationship which his or her parents disapprove of, he or she might commit or attempt suicide, especially if he or she is pressurised by the parents to end this relationship (Schlebusch, 1995). De Jager (1998:260) found in her research project, which focused on adolescents' involvement in the occult, that Satanism could be a contributing factor to committing or attempting suicide. Adolescents who participated in the study and were involved in the occult were found to be preoccupied with death in general, as well as their own death, and some of them had already attempted suicide on several occasions.

Feelings associated with the suicide attempts
This sub-theme on the feelings associated with the suicide attempts will be divided into the following three categories: Feelings experienced before the suicide attempt; Feelings experienced during the suicide attempt; Feelings experienced after the suicide attempt. Feelings experienced before the suicide attempt Table 3 below depicts the feelings the participants experienced prior to the suicide attempt. Berman and Jobes (1992:107, 109, 141) mention that real interpersonal isolation and estrangement, or the experience of such feelings, could stimulate suicidal behaviour. These authors also refer to the fact that feelings of rejection, worthlessness, irritation and aggression experienced by an adolescent might be seen as warning signs or be indicative of possible suicidal behaviour. Russell andJoyner (2001:1278) found that adolescents who attempt suicide experience feelings of hopelessness and depression (Paulson & Worth, 2002:88.) These feelings were verified in studies undertaken by Bancroft as early as 1976, as reported by Kienhorst et al. (1995:623). Table 4 below depicts the feelings the participants experienced during the suicide attempt.  The feelings of numbness are a common experience amongst people who are suicide prone. Cotton and Range (in Westefield, Range, Rogers, Maples, Bromley & Alcorn, 2000:445) confirm this: "Suicidal individuals are less attracted to life and more repulsed by it, and at the same time they are less repulsed by death and more attracted to it". The literature also confirms these as feelings experienced by adolescents planning a suicide attempt (Posener et al. in Berman & Jobes, 1992:107). Table 5 below depicts the feelings the participants experienced after the suicide attempt. In surveying the existing literature to verify the feelings described by participants after the suicide attempt, the researcher found a paucity of research on the feelings adolescents experience after an attempted suicide. The feelings of relief and remorse reported by participants in this study are, however, congruent with the thoughts and feelings reported by Kienhorst et al. (1995:627) as occurring during the attempt. Valach, Michel, Dey and Young (2002:1, 4) confirm that feelings of shame often accompany this phase of the suicide attempt.

The sources and nature of support offered to adolescents after the suicide attempts
The participants reported having received help from a variety of people (sources) after the suicide attempt, namely their mothers, a brother in-law, a cousin, neighbours, bystanders, teachers and a caregiver, as well as the hospital staff, a social worker in the community and a minister of religion. The nature of the support received can be divided into two categories, namely: Support that was experienced as positive after the suicide attempts; and Support that was experienced as negative after the suicide attempt.
These two categories are discussed below.

Support that was experienced as positive after the suicide attempt
One participant experienced the support and the attention received from the hospital staff as positive. She expressed this as follows: "They have given me the best care in the hospital. I got everything, everything… they have given me everything I was looking for. You know that attention-seeking story, they have given me attention, every day, every time, and food and a safe place".
Other participants also experienced the help offered by the general practitioner and the nursing staff as positive and echoed the latter sentiment. One participant said: "…the nurses came to me to talk with me". A second participant said: "She [the nurse] spoke comforting words to me, she told me that I have my whole future ahead of me".
The help offered by the hospital social worker and the psychologist was also experienced as positive. The following actions by the social worker in the hospital setting made a positive impression on the participants: The fact that she referred one participant for psychological counselling; The counselling she provided on the pros and cons of taking an overdose of medication; Counselling around the suicide attempt with specific reference to the reason for the suicide attempt and the thought processes of the person contemplating suicide; Involving the participants in group activities.
These positive experiences of the participants regarding the quality of support they received from the staff in a health care setting is in sharp contrast with the findings of Nirui and Chenoweth (1999:362), who state that: "…research reveals that people at risk of suicide perceive the range of health care services to be unhelpful. Hospital services are often regarded unfavorably because of poor staff attitudes to suicide attempts".
Outside the hospital environment the participants reported having experienced as positive the help and support received from friends, neighbours, a minister of religion, a social worker in the community, extended family, teachers, a caregiver and their mothers. One participant described how a teacher assisted her: "She asked me why I did it, why I want to kill myself. She said things to me, positive things that encouraged me. My class teacher really supported me. Even up to today she asks me how I am doing, and everything".
The following two quotes from the transcripts testify to the support the participants received from their mothers, which they experienced as positive.
"My mother also tried to commit suicide, and she talked with me about the experience and it helped".
"My mother prayed for me".

Support that was experienced as negative after the suicide attempt
The nature of support received by the participants after the suicide attempt was mainly positive, but there were references in the transcripts referring to support that was experienced as negative.
Some of the participants experienced the support form the hospital staff (that is, the general practitioner and the nurses) as negative. The following excerpts from the interviews demonstrate this: [The General Practitioner said] "…can't we give him a 'death' injection, so that he can die, because he keeps on attempting suicide?" "He [the General Practitioner] does not have time for a person. Everything he does, he does so quickly".
This experience is congruent with the view expressed by a participant in the study by Nirui and Chenoweth (1999:366), who expressed her views on the quality of service provision by a GP for suicidal patients as follows: "He always gave short consultations… in and out in a short time".
The following statements from the research participants testify to their negative reaction to support from the nursing staff: "They were angry and they don't treat you well. They don't treat you with care".
The treatment was not good. When I asked them to bring me something so that I can rinse my mouth, they said why don't you go to the bathroom and fetch it yourself'" Apart from the support of the hospital staff that was experienced as negative, some of the participants also experienced as negative the support from parents, a teacher and a social worker.
The following excerpts provide evidence of the negative experiences some participants encountered from their parents after the suicide: "She did not even know that I was in hospital, she [the mother] was too drunk. The morning when I got home, she hit me in the face".
One participant experienced the following reactions of a teacher as negative: "She came to talk with me and asked me why I did it [attempt suicide]. Then she started to swear at me, and then she said to me that I sleep around with all the men".
In the next section of this paper the last theme, namely the recommendations suggested by adolescents on alternative and/or complementary support desired after the suicide attempt, will be discussed.

Recommendations on alternative and/or complementary support desired after the suicide attempt
The participants made recommendations about alternative and/or complementary support they would have wished after the suicide attempts, with specific reference to the following systems: Alternative support from parents; Alternative support from the hospital staff; Alternative support from the social worker.
With regard to the alternative support from parents, 4 of the 17 participants said that they would have liked it if their mothers were more involved in their lives and available to talk to. Other participants longed for a greater involvement from their father's side, while some had the need for both parents to spend time with them. Two quotes to support this are as follows: "…I want us to be together more, to talk about things. When I tell them that I want to talk about this and that, they say they don't have time, they have to go here and there".
"…my father should have supported me! I would have felt satisfied if he also came to help me…and said to me, you must not have done it… then I would know that he loves me".
The need of adolescents for greater involvement from parents in their lives is supported by Borowsky, Ireland and Resnick (2001:487) in their research on adolescent suicide attempts focusing on risk and protective factors. It emerged that parent-family connectedness is a protective factor against attempting suicide which cuts across the gender and racial/ethnic lines of the group of adolescents studied, since family relationships were found to be one of the causal factors for attempted suicide (Kienhorst et al., 1995:628).
The alternative support wanted from the hospital staff focused mainly on medical staff-patient relationships. They would have liked more communication from the medical staff; more time and attention from them, and more caring and supportive treatment. This is summed up by the following comment from a participant: "…they need to talk nicely to you. The staff must be friendlier with you. They must take time to come and sit with you…" This comment about support points to the fact that services offered by hospital staff to suicidal patients are unsatisfactorily. Nirui and Chenoweth (1999:362) support this view: "Hospital services are often regarded as unfavorable because of poor staff attitudes to suicide attempts. General practitioners' lack of interest and/or experience in dealing with people who present with suicide ideation and depression and their over-prescription of medication for suicidal symptoms are also areas of concern". With regard to the alternative support they desired from the social worker, the participants mentioned that they would prefer more time alone with the social worker to talk about their suicide attempt. They also expressed the need for a support group where they could talk to other adolescents who had had the experience of attempting suicide.

CONCLUSIONS AND RECOMMENDATIONS
The conclusions and recommendations that can be made, based on the research findings highlighted above, are as follows: The research found that although only 17 participants were interviewed, a total of 25 suicides had been attempted in the sample. The literature confirms that if a person has attempted suicide once, he or she is likely to repeat it (Bernardo, 1996:3;Levinsohn et al., 1994:302, 303).
From the data collected from the participants by means of semi-structured interviews, the following themes emerged: -Reasons for the suicide attempt; -Feelings associated with the suicide attempt (before, during and after); -The sources and the nature of support received after the suicide attempt; -Recommendations on alternative and/or complementary support desired after the suicide attempt.
The participants cited the following as the causes of their suicide attempts: stressful family circumstances with specific reference to family disorganisation, abuse of alcohol by parents, absence of parental role models, abuse by parents and family violence; sexual abuse; relationships of children with the opposite sex not approved by parents, and Satanism. These reasons for attempted suicide given by the sample group are in accordance with the reasons for attempted suicide cited in the literature (De Jager, 1998:20;Hurry, 2000;Husain & Vandiver in Fernquist, 2000:551;Pfeffer in Stoelb & Chiriboga, 1998:361;Roberts in Bezuidenhout, 1998:76;Schlebush, 1995;).
Before the suicide attempts a variety of negative feelings were experienced, namely feelings of loneliness, rejection, desperation/hopelessness, inner anger, irritation, meaninglessness and inferiority. These feelings experienced by the participant before the suicide attempts are characteristic of the feelings linked with suicidal behaviour that are described in the literature (Berman & Jobes, 1992:107,109,141;Paulson & Worth, 2002:88;Russell & Joyner, 2001:1278. The participants could describe only two feelings experienced during the suicide attempt, namely feelings of numbness and uncertainty. These feelings were confirmed in the literature as being present during the suicide attempt, but also prior to the attempt (Cotton & Range in Westefield et al., 2000:445;Kienhorst et al., 1995:623;Posener et al. in Berman & Jobes, 1992:107).
The feelings experienced by the participants after the suicide attempt included regret, disappointment, guilt, sadness, fearfulness, shame, relief and a renewed zest for life, disappointment because the suicide attempt had failed, and discontent because the suicide attempt was interrupted. The researcher has noted the apparent paucity of information in the literature on the feelings experienced by adolescents after an attempted suicide. Regret and remorse, however, are reported in a study by Kienhorst et al. (1995:627). Feelings of shame are reported by Valach et al. (2002:1, 4) as being typically associated with this phase of the suicide attempt.
The participants experienced as positive some of the support received from hospital staff (that is, GPs, nursing staff, hospital social worker and psychologist) and some as negative. The help they received from friends, neighbours, a minister of religion, a social worker in the community, extended family, teachers, a caregiver and their mothers was also viewed as positive by some of the participants. Although the majority of the research participants experienced the support offered by these people as positive, others experienced the help and reactions of the hospital staff, parents, teachers and a social worker as negative. This negative experience is also reported in the research study conducted by Nirui and Chenoweth (1999:362) that aimed to explore the kind of experiences that suicidees had when seeking support from health care services in the period leading up to their death, as perceived by close family and friends. In that study it was found that those who had committed suicide had not felt comfortable talking about their emotional problems, because of a perceived lack of interest on the part of the general medical practitioner in helping them to deal with the suicidal thoughts and behaviour.
Based on the above findings, the researcher recommends that further qualitative research be undertaken amongst adolescents who have attempted suicide specifically to explore (i) their perceptions and experiences regarding the quality of health care services and (ii) the medical staff-patient relationships of people at risk of committing suicide.
The recommendations from the participants with regard to the alternative and/or complementary support desired after the suicide attempt were directed towards the parents, hospital staff and social workers. The researcher endorses the participants' call for greater parental involvement in the lives of their children. This is relevant not only as a reactive measure in the aftermath of a suicide attempt, but also as a preventative and protective measure (Borowsky et al., 2001:487).
The researcher endorses the call by the participants for quality care and empathic treatment of suicide-prone individuals.
In conducting the literature review the researcher became acutely aware of the fact that much research has been conducted in the areas of the epidemiology and etiology of adolescent suicide (Ackerman, 1998;Gutierrez, 1999;Rabe, 1993;Thomerson, 2002). A major limitation of such studies, which focus predominantly on the epidemiology and etiology surrounding adolescent suicide, is the fact that the suicide-prone adolescent and the circumstances surrounding his or her suicide have been the object of the study. The voice of this particular group of adolescents, as the self-knowing subject speaking authoritatively about thoughts, feelings, and behaviour surrounding suicide attempts, and specific needs emanating from the attempted suicide experience, has been silenced. Typically, it has been the researchers' points of view that have informed our understanding of the phenomena of suicide and attempted suicide. However, it is increasingly being recognised that what is essential for a comprehensive or holistic understanding of the topic is the voice and perspective of the participants. This is even more relevant if intervention and prevention strategies are to be planned for addressing the needs of this particular client group (Paulson & Worth, 2002:86). Therefore, it is recommended that more research projects, such as the one under discussion, be undertaken, not only to broaden our understanding of the needs and experiences of the adolescent who has attempted suicide, but also to include these experiences and expressed needs of these self-knowing subjects in policies and programmes that can truly claim to be youth-informed, youth-sensitive and youth-centred.
It is further recommended that the findings from these studies be communicated to the different stakeholders in the health, welfare and educational sectors through information sessions, talks and seminars, in order to sensitise service providers to the experiences and needs of adolescents who have attempted suicide.

FINAL REMARK
Although in this study no comments were made by the research participants about the prevention of this growing epidemic of suicide and attempted suicide among the youth, the researcher is of the opinion that the health, welfare, educational, religious and parental systems have a pivotal role to play in the prevention of suicide and attempted suicide. These multiple systems share the responsibility of promoting protective factors in the lives of young people. Health care and welfare professionals have the responsibility to inquire about emotional health, family interactions, school achievements and connectedness. Clinicians and teachers should educate parents early on about the importance of nurturing children and promoting parenting skills that will improve the life and interpersonal skills of children. Clinicians can play an important role in identifying patients at risk of suicide and in treating or referring them for help in good time (Borowsky et al., 2001:487). These preventative measures are necessary, because suicidal behaviour remains a multi-factorial phenomenon that demands multilevel approaches (Nirui & Chenoweth, 1999:370-371).