Pathways to Care and Help-Seeking Experience Prior to Self-Harm

A Qualitative Study in Taiwan

Abstract

Background: 

Help-seeking behavior may protect people from harming themselves. However, few studies have discussed how and why people access lay or professional help prior to self-harm.

Purpose: 

This study explored the subjective experiences of individuals with deliberate self-harm in terms of help-seeking behavior and medical care pathways.

Methods: 

Researchers performed qualitative in-depth interviews and content analysis and used a sampling grid to recruit participants. Twenty emergency attendees between the ages of 18–55 years were interviewed on their help-seeking pathways and experiences using a standardized topic guide.

Results: 

Participants identified friends, family members, healthcare personnel, and their own initiative as the primary medical care pathways. Analysis showed help-seeking experiences significantly related to the physician–patient relationship, social support, and treatment adherence. These factors were also identified as prominent themes related to medical help-seeking behavior.

Conclusions: 

Supportive attitudes and continuous care from formal and informal sources of help may facilitate help-seeking behavior, whereas negative influences from close friends or relatives may trigger a self-harm episode. Medication stockpiling and the negative aspects of close relationships should be addressed and minimized to raise the effectiveness of self-harm or suicide prevention efforts.

Introduction

Deliberate self-harm (DSH) is an important predictor of suicide (^{Cooper et al., 2005}) and has been suggested to occupy different positions in a process potentially leading to completed suicide (^{Wasserman, 2001}). Empirical studies have shown characteristics of individuals with DSH to differ from those who die of suicide in terms of gender, psychiatric diagnosis, self-harm method, suicide intent, and prior health service use (^{De Leo et al., 2002; Harriss & Hawton, 2005; Kapur & Gask, 2009}). However, although it is a significant predictor of DSH (^{Kaslow et al., 2002}) and may potentially be modified by social support or social networks (^{Wu, Stewart, Huang, Prince, & Liu, 2011}), little is known about the help-seeking behavior of DSH individuals. Understanding medical and informal help-seeking behavior from the subjective perspective of DSH individuals is essential to our understanding and engagement of this population (^{Lakeman & Fitzgerald, 2008; Wu et al., 2011}) and our effectively preventing self-harm and death.

Around 30%–60% DSH adults were reported as having visited psychiatric or nonpsychiatric services around the time of index DSH acts implemented in Europe (^{Houston, Haw, Townsend, & Hawton, 2003; Suominen, Isometsa, Ostamo, & Lőnnqvist, 2002}). In Taiwan, a cross-sectional self-harm registration study indicated that 43.2% of DSH individuals had contact with psychiatric services within a 1.5-year period (^{Chiu & Lee, 2006}). Few studies have addressed the motivations of and pathways used by people to seek professional help, which is a key area of evidence for pathways to care. Seeking help for mental distress prior to DSH is clearly not confined to contact with health services, and the role of friends and family as sources of help and advice is important. Even fewer studies have investigated the topic of help offered by informal sources to DSH individuals in East Asia. Understanding factors related to general help-seeking behavior, particularly informal help sources, has been suggested as a key requirement in the process of engaging individuals with psychiatric morbidity or suicidal thoughts (^{Biddle, Gunnell, Sharp, & Donovan, 2004}). Therefore, exploring informal help-seeking experiences may provide further understanding of how to access such people and consolidate their help-seeking behavior.

Taiwan reported a high suicide rate of 16.8 per 100,000 in 2010 (^{Department of Health, Executive Yuan, Taiwan, ROC, 2011}). Although the use of charcoal burning as a self-harm method is a significant cause of death (^{Kuo et al., 2008}), it is a method of suicide not likely to be strongly associated with an extensive history of mental disorders (^{Chen, Lee, Chang, & Liao, 2009}). The literature has also highlighted the possible exacerbating role of negative media reporting on suicides in the rapid increase of fatal and nonfatal DSH in Taiwan (^{Cheng et al., 2007}). In considering opportunities for seeking help, an important context is provided by Taiwan’s healthcare system, which is affordable and widely accessible under a national health insurance program that covers 99.48% of the nation’s population (^{Bureau of National Health Insurance, Department of Health, Executive Yuan, Taiwan, ROC, 2011}). A second important feature is the minimal ‘gatekeeping’ role played by primary care. In other words, all health services can be approached directly without the need for interservice referral. In the pathway to care, family members and people in social networks play a role in help-seeking decision-making. Help-seeking behaviors, the healthcare system, and suicide characteristics in Taiwan all differ significantly from Western societies. Therefore, identifying protective factors helpful to the prevention of DSH and suicide is a critical and outstanding issue for ethnically Chinese populations.

This study examined the lived experience of DSH individuals in the Taiwan context of direct public access to healthcare. Research objectives were to explore care pathways and identify factors significantly associated with decisions to seek help.

Aim

The aim of this qualitative study was to explore how and why people with DSH accessed formal or informal help sources and related factors.

Methods

Design

This study adopted a qualitative methodology employed within the theoretical framework of Kleinman’s healthcare system model and an interpretive framework of sociology that emphasizes subjective experience and personal meaning (^{Kleinman, 1980; Pescosolido & Boyer, 1999}). Using qualitative interviews was recommended for exploring the health-seeking experiences of DSH individuals (^{Sinclair & Green, 2005}) and pathways to care (^{Morgan, Mallett, Hutchinson, & Leff, 2004}). Researchers conducted in-depth interviews within 1 week of emergency admission to explore healthcare-seeking and informal help-seeking phenomena.

Participants

This study was conducted between October 2005 and December 2006. Participants were recruited from among emergency room attendees who had harmed themselves and already participated in a previous quantitative survey (^{Wu et al., 2011}). Taiwan’s universal National Health Insurance program has provided access to affordable medical services and healthcare to the public since 1995. Enrolled individuals enjoy open access to all healthcare facilities and specialties at any time with no prior referral necessary and with unlimited access. DSH individuals may thus access any medical service level or specialty.

DSH has been defined as deliberate self-poisoning or self-injury with a nonfatal outcome regardless of the intent to die (^{Hawton et al., 2003}). All patients with overt or possible DSH acts included in this study were referred to the associated hospital’s suicide prevention center where they were screened for DSH status using a structured suicide risk assessment. Information provided by spouses or other close relatives was also considered. A case conference was held every week to confirm DSH caseness and discuss clinical treatments conducted at the suicide prevention center. The structured instrument Pierce Suicide Intent Scale measured suicide intent. The instrument assigns one of three intention-to-die scores, which relate to the index DSH episode. The instrument has previously indicated good predictive validity (^{Pierce, 1981}).

People referred to the suicide prevention center who were over 18 years old and in stable physical and psychological conditions were invited to participate in this study. To maximize sample representativeness, researchers applied no other inclusion or exclusion criteria with respect to mental disorder diagnoses. A sampling grid was designed to recruit participants, which distinguished participants by age groups (18–29, 30–44, and ≥45 years), gender, and suicide intent score (i.e., low, medium, and high; Table 1). Variables were chosen after reviewing relevant literature articles and in light of the authors’ clinical experience. Attempts were made to recruit equivalent numbers of participants in each cell of the grid to obtain a heterogeneous population.

TABLE 1: Sampling Grid and Actual Numbers Recruited

The interviewer served as a suicide case manager during the study period. The interviewer’s case manager role did not affect the participation of study subjects or the quality of interviews as this study adopted standardized procedures and a topic guide for data collection.

Data Collection

All interviews were conducted within 1 month of the index self-harm episode in a quiet and segregated place. In-depth interviews were carried out under the guidance of a topic guide using the three main topics developed during the pilot study. These topics were arranged in a sequence designed to facilitate responses. The authors further adopted a framework proposed by ^{Kleinman (1980)} as the strategy to maximize responses for particular help sources such as medical, lay/popular, and/or folk. ‘Lay/popular’ refers to help provided by oneself, family, or friends; ‘folk’ refers to help from traditional or community healers, and ‘professional’ refers to help sought from accredited medical staff and ancillary professions. Experiences and pathways of help-seeking behavior were probed as completely as possible to explore participant attitudes and thoughts in relation to the following: (a) events or stressors leading up to the DSH index episode, (b) suicide ideation and intention underlying the DSH index episode, and (c) access pathway followed to obtain formal or informal help and the motivation(s) for seeking/not seeking help (data not included in this study). If the participant denied seeking or could not recall seeking help, the interviewer would then enquire specifically about particular sources of help (medical, lay/popular, folk, or others) to clarify responses further.

Ethical Considerations

This study received research ethics approval from two hospital ethics committees in Taiwan and the United Kingdom, respectively. Full instructions for the study were provided prior to asking each subject to sign informed consent. Participation was wholly voluntary, and adequate time was given to make the decision whether or not to do so. Participants could withdraw at any time of the study without reason and without affecting their treatment rights.

Data Analysis

All interviews were recorded on a digital recorder. Recordings were transcribed, translated, and back-translated by a Taiwanese bilingual specialist. Two external auditors then reviewed the English transcripts to clarify translation and data analysis. All English transcripts were downloaded in Atlas.ti 5.0 for data integration and analysis.

Authors used content analysis to analyze interview transcripts and field notes. Initially, each interview was coded systematically to identify preliminary topics. Specific words and phrases related to help-seeking experiences or reflecting participants’ perceptions of seeking support were coded. For example, the description, ‘those friends helped me find the value of living’ was coded as ‘informal help-seeking from friends’ and ‘function of friends’. These codes were then compared, integrated, and refined to develop the main themes of this study. During formal data collection procedure, participants were encouraged to respond to questions based on the three topic guides. Further interview contents were compared with and incorporated into previous findings to further explore and retrieve help-seeking perceptions. This method was utilized until thematic saturation and agreement were reached between authors. The two authors performed coding and analysis for each interview transcript independently, with final coding themes and quotations then sent to an external auditor for content validation. This multiple coding method is a strong check on qualitative research rigor (^{Mays & Pope, 1995}).

Study Rigor

A local research team leader and British supervisor closely supervised study procedures. Credibility of study findings was established through triangulation with wider quantitative study results within which this study was nested, hospital computerized records (limited to those who sought medical services in the study hospital), rich description of the findings, weekly peer discussion for emerging concepts, and independent review by an external auditor.

Results

Twenty participants who had all contacted medical services at the study hospital agreed to take part in this study. The response rate was 74.1% (20 of 27). Men and people aged over 45 years old were difficult to recruit, resulting in nonequivalent number(s) in each cell in the sampling grid (Table 1). Reasons for nonparticipation included poor physical condition (n = 1), refusal to provide necessary information (n = 4), and unfinished interviews (n = 2). Mean participant age was 33.5 years, and the sample was a reasonably heterogeneous group with respect to gender, DSH method, suicide intent, past DSH history, help-seeking experience, and previous self-harm history (Table 2).

Data were fitted into the two main topics of pathways to care and health service experience. Pathways to care were illustrated in Figure 1 and described how people with DSH accessed medical care. Themes derived from the topic of health service experience include positive and negative views toward the physician–patient relationship, nonadherence to medical contact, and social support.

TABLE 2: Participant Profiles

Figure 1: Pathways to care among people with self-harm (N = 20). Arrows indicate different pathways to medical care, with line thickness correlating to utilization frequency. Three participants of the study did not seek any help before harming themselves and were excluded here.

Pathways to Care

Sixteen of the twenty participants had contacted psychiatrists or physicians within 1 year before their interviews. Four had never visited psychiatrists or physicians, including one who had sought for friends’ opinions and three who had never thought about seeking any help prior to self-harm. Notably, informal sources of help, including friends, colleague, or family, were the first line of pathway contact for the majority (Table 2). Friends or family served, at least for some cases, as a mediator in initial medical contact or a reminder to return to the medical system. Although this may mostly happen among those with enduring mental disorders (n = 5) or serious functional impairment (n = 6), they appeared to be potentially important pathways for facilitating psychiatric evaluations:

At first I had a hard time sleeping. Then a friend of mine who used to work at the hospital suggested I see a doctor, so I took his advice. (CS1, a 35-year-old man, DSH method: repeated charcoal burning)

I phoned my daughter and told her all my conditions [raising voice]. Then she said, ‘I didn’t realize it was so serious.’ She then brought me back to the hospital and visited my psychiatrist. (CS4, a 55-year-old woman, DSH method: drug overdose)

Self-referral was another important care access pathway. Thoughts of suicide, an unbearable state of mind, and mental health problems resulted in medical help-seeking pathway, respectively, for the three of the self-referral participants. External sources of information on seeking medical advice (e.g., media reports, family experiences of medical visits, and reputation of certain health institutes) and self-awareness led some to access health services on their own:

After I had done things to calm myself down, I still couldn’t bear the thought of killing myself. Because it wasn’t a great feeling when you were down. Plus sometimes I felt dizzy…so I figured why not see a psychiatrist. (CS2, a 32-year-old woman, DSH method: KCL injection and drug overdose)

I was wondering whether I was experiencing depression. I felt difficulty getting enough sleep, no confidence, things not going well for me. I kept fixating on some unresolved issues from childhood. I got the information from television and from my mom who suffered from depression when I was in jail. I thought why not go to the hospital that my mom used to visit and have a check. (CS16, a 43-year-old man, DSH method: burning charcoal)

To summarize, pathways to medical care identified in this study were through friends or family, healthcare personnel, and the person himself or herself (Figure 1). It was found that informal sources of help (mainly friends and family) represented the major pathway (70.6%) and healthcare personnel such as physicians and nurses represented a minor pathway. Among the motivations for help-seeking behavior, most participants disclosed they sought help because of mood disturbances (n = 8); other reasons included sleep problems (n = 4), somatic complaints such as dizziness and headache (n = 2), and perceived pain (n = 2; Table 2).

Help-Seeking Experience

The authors analyzed medical help-seeking experience to derive three themes. The physician–patient relationship stood out as a prominent theme with both positive and negative views and attitudes toward its effect revealed. Adherence was found to be a serious concern for people with DSH, particularly for those who harmed themselves via medication overdose. Social support was another main theme facilitating help-seeking behavior, particularly for those who sought help from informal sources, mainly friends and family.

Positive and negative views toward the physician–patient relationship

A confidential relationship built on continuing care and active inquiry by health professionals was found to enhance the likelihood of patients sharing or disclosing their inner thoughts. The reported positive therapeutic dynamic appeared to create an open climate for communication. Participants revealed they were more likely to talk about their problems to mental health professionals who were seen as trustworthy and/or caring and who were seen as professionals who would maintain confidentiality:

I trust my therapist that she won’t tell another soul…I talked about my mood swings and anything unfortunate. I felt that at least she would give me suggestions, and I could get things off my chest. (CS18, a 50-year-old woman, DSH method: charcoal burning)

I only told my current psychiatrist that I am suicidal…I disclosed my overdose behavior to him because he asked all about my situation, and he was always concerned how I was going to solve my problems. (CS6, a 33-year-old man, DSH method: drug overdose)

However, some participants reported less satisfactory experiences with or attitudes toward their contact with medical professionals. Several participants expressed that limitations of medical service dissuaded them from visiting doctors. They perceived doctors as not able to help them in certain respects, for example, in the extent to which staff could help solve participant problems and provide practical assistance for their situations. Participants also discussed how past unsatisfactory medical advice resulted in negative attitudes:

I think it would be difficult to get help from medical services…. Doctors only help us to survive, but they don’t help us do everything…. Solving my financial problems is even more important, but extremely hard. (CS9, a 41-year-old woman, DSH method: charcoal burning and drug overdose)

My psychiatrist prescribed drugs and encouraged me not to worry so much, and said that I had to figure out things for myself. He also wanted me to think how my family was supposed to live without me. I thought that his suggestions were useless, and I didn’t listen to him. (CS16, a 43-year-old man, DSH method: charcoal burning)

Seeking medical help is really not much help for suicidal people. You go to doctors when you are sick and want to cure yourself. But having suicidal ideas is more like everything is finished and the only thing you want is to die. (CS17, a 35-year-old man, DSH method: charcoal burning)

Nonadherence to medical contact

Nonadherence was a theme derived from participants’ descriptions that included the issues of irregular medication consumption, stockpiling medication, and doctor-shopping. Getting medication from doctors was a commonly reported reason for medical contact. However, some participants reported not fully adhering to prescribed dosages or frequencies:

Sometimes I took medications; sometimes I did not. When I was in bad mood, I took them all, just like this time. This has been going on since senior high. (CS8, a 21-year-old woman, DSH method: drug overdose)

I didn’t quite follow the prescribed dosage. My doctor prescribed me one tablet a day, but I took three almost every day. (CS13, a 46-year-old woman, DSH method: drug overdose)

Some participants perceived medical contact as simply an opportunity to obtain or renew medications. Unfortunately, some visited different doctors to stockpile medicines for further DSH. The availability of drugs for overdose in those with suicidal ideas is a serious concern. It may also be a source of DSH in combination with other lethal acts such as charcoal burning, as was the case for three participants in this study:

(Through medical visits) I thought I could get medication, but I only took the hypnotics and stacked up the rest of drugs. People said that if I took all those drugs with alcohol, my heart would stop. I wanted to try. So I collected everything I had left over. (CS18, a 50-year-old woman, DSH method: drug overdose)

I have been taking medications for a long time, gastric medicines and stuff. I always had the idea that if I took all the medications or if I didn’t follow doctor’s orders, my body would gradually rot away and I would finally be dead. That’s what I have been doing. (CS8, a 21-year-old woman, DSH method: drug overdose)

Some participants disclosed that they had visited different doctors because they felt dissatisfied with their psychiatrist(s), which led them to find others. The expressed reasons for multiple visits to different doctors included stockpiling medications, finding better treatment, and personal preference:

At first my doctor gave me light medication because I told him not to prescribe too heavily…I still felt uneasy, sad and too shy, kept drinking and started losing faith in him. After a while I changed to another psychiatrist who prescribed heavier medication…. (CS16, a 43-year-old man, DSH method: charcoal burning)

I would go to one of my psychiatrists regularly, be a good patient and have my prescription refilled, without telling him my relationship problems and suicidal ideas. I went to another to disclose my situation…. (CS6, a 33-year-old man, DSH method: drug overdose)

Taken as a whole, the above messages identified nonadherence to medication instructions or a single therapist as related to participant mood disturbance, irrational thinking (e.g., purposefully overdosing on drugs to destroy the body) and/or suicidal ideation. Irregular medication consumption or stockpiling behavior derived from thoughts of dying may develop into acting on such over the long term. Such was exacerbated by participants’ inability to relieve uneasiness and their dissatisfaction with medical consultation results. Identifying and exploring reasons for doctor-shopping revealed an unmet need for treatment among DSH individuals.

Social support

A recurrent theme derived from data addressed the role of informal help sources as derived from participant social networks. Responses indicated this type of help was the initial pathway to care used by 12 of the 20 participants. This pathway primarily involved contacts with friends or family (Table 2) who served as major and immediate sources of social and emotional support. Several participants shared that they felt being listened to and accompanied by friends as support, and it was a factor that reined in their self-harm desires:

At least I didn’t feel as painful after chatting, and the ‘I must hurt myself’ kind of thoughts disappeared temporarily. Actually, what they’d say was the same thing others suggested…. However, I appreciated their care for me. Those friends helped me find the values of being alive…. Only two of them know. They accompanied me when I was in trouble. I felt less alone when I was with them, which I perceived to be an important factor that stopped me from hurting myself several times. (CS5, a 30-year-old woman, DSH method: drug overdose)

Whenever I had suicidal thoughts, I used to look for my closest friends. I told them I had this painful feeling and talked about my suicidal thoughts. The thoughts wouldn’t just fade away through talking, but I just needed someone to listen to me at that moment… I don’t think it’s likely for me to call a stranger to talk about myself…I perceive the company of a close friend as rather important…I would like to have some companionship first, and then maybe go to the psychiatrists if I have to. (CS17, a 35-year-old man, DSH method: burning charcoal)

Many participants identified friends as confidants and risk detectors through their offering of nonjudgmental listening. They described their perspectives on close social relationships as follows:

WU: Why did you seek their [the case’s friends] advice?

CS: For one, I am close to them. They wouldn’t label me as problematic, and I figured they’d give me objective opinions instead of judging me. (CS7, a 23-year-old man, graduate student, DSH method: drug overdose)

They are my best friends. Those who listened to my warnings carefully would come over to accompany me…. (CS10, a 28-year-old man, self-employment, DSH method: attempt to jump from the second floor)

They are more aware of my emotional change, so I told them my feelings. Judging from my attitude and behavior, they asked me what had happened. And I would answer their questions because they asked. (CS11, a 26-year-old man, salesperson, DSH method: burning charcoal)

However, people who were close to the participants, although they were seen as having the potential to offer care and companionship, were also described as having the potential to trigger DSH. This has been called as the ‘negative aspects of support’ in previous studies (^{Stansfeld & Marmot, 1992}). For example, the following participant described ways in which friends were seen to have reacted negatively to their confiding attempts and how social support had broken down:

I was in bed all the time because of depression, and she [the case’s friend] knew what had happened to me…but she didn’t react much. After bickering several times, she stopped talking to me. She had said to me angrily that I should go kill myself. She once choked me for about four seconds. At the time, I felt nothing mattered anymore. (CS2, a 32-year-old woman, DSH method: KCL injection and drug overdose)

Friends’ responses and attitudes, particularly from those with whom participants shared intimate relations, appeared to be influential and was a potential factor in the pathway between intense extreme feelings and DSH. These contrast with the reports of support and care described earlier. So-called ‘close’ friends thus were not necessarily a protective factor. Some friends were able to offer help just in time, but others appeared to trigger anger and impulsivity, and the acts sometimes appeared to have purposes other than merely to end life.

In summary, the three main themes in the study were interrelated in the pathways to care and help-seeking experiences of people with DSH. It was social support that helped them disclose to friends and/or doctors to begin with, but further seeking for help depended on the quality and quantity of support. Negative physician–patient relationships may lead to seeking help from friends and vice versa.

Discussion

Social Support and Help-Seeking Behavior

A trustworthy and supportive physician–patient relationship as well as social support from friends and family provided a climate for DSH individuals to confide their problems and seek help. Informal resources such as friends or family and nurses may engage individuals to seek medical services and thus suggest opportunities for early intervention. Factors found to influence the likelihood of contacting friends, family, or physicians (i.e., trust, listening, active probing, and a confiding relationship) were consistent with previous studies (^{Biddle, Sharp, Gunnel, & Donovan, 2006; Chew-Graham, Bashir, Chantler, Burman, & Batsleer, 2002}). Although this study did not identify a causal relationship between social support and repeated DSH, our study supported previous findings that supportive relationships are crucial elements in informal networking (^{King et al., 2006}). Friends, family, and professional support that increase level of perceived support have also been reported to reduce risk of suicide (^{Cheng & Chan, 2007; Grohol, 2008; McLaren & Challis, 2009}) and are also likely to reduce risk of repeated DSH (^{Henggeler, Schoenwald, Rowland, & Cunningham, 2002}). An implication is that nurses should increase their awareness and maintain an active and supportive networking for those at risk of suicide or self-harm.

Medical Help-Seeking Behavior and Its Positive/Negative Consequences

The physician–patient relationship may either interfere with or augment help-seeking behavior. Participants described needing positive therapy, dynamic and active care, and in-depth questioning as well as trust from their physician. The trust and care element echoes previous findings and emphasizes the potential of doctors’ attitudes in modifying the medical help-seeking behavior of people with mental distress (^{Biddle et al., 2006}). It is therefore equally critical that physicians pay attention to the potential negative consequences of medical services, including non-adherence to medication and further DSH actions using prescriptions obtained from multiple doctors. Previous research has indicated the positive relationship between a collaborative therapeutic relationship and successful management of suicidal ideation (^{Ilgen et al., 2009}). Such highlights the need to create a supportive relationship in the medical environment. Reduced adherence has also been associated with increased suicide risk (^{Pompili et al., 2009}), but there have been fewer empirical studies on adherence for suicidal behavior. Nurses can serve as gatekeepers to detect medication hoarding among DSH individuals to reduce overdose risks and refer those at risk to professional services. Apart from the gatekeepers program (^{Knox, Litts, Talcott, Feig, & Caine, 2003}), interventions may best target those at particularly high risk for DSH (e.g., people who visit medical services repeatedly or do not comply with interventions). Moreover, future research should consider prospective observational or cohort studies and qualitative methodology to explore factors that contribute to nonadherence and the extent to which these can be minimized.

Strengths and Limitations

This study focused on a group of predominantly younger adults that were heterogeneous in terms of suicide intent and DSH method. We used a sampling grid when recruiting study subjects to enhance credibility and ensure relatively equal representation for a number of subject characteristics. Employing qualitative methodology within a challenging sample in terms of recruitment enabled the authors to deeply explore the help-seeking perceptions and experiences of people experiencing mental distress. The qualitative method has been shown to be appropriate for accessing and eliciting personal experiences (^{Morgan et al., 2004; Sinclair & Green, 2005}). Study results were data driven under the rubric of grounded theory (^{Glaser & Strauss, 1967}). Moreover, internal/external auditing and close supervision by the research team further added to study rigor (^{Whitley & Crawford, 2005}).

There were several limitations to this study. First, results generalizability was restricted by definition to those who contacted hospital services after a DSH episode. Also, as this was a clinical sample from an urban emergency department in Northern Taiwan, caution is advised regarding generalizing results to individuals in other Taiwan areas or to DSH individuals who do not contact health services. Second, help-seeking, as defined in this study, was a failed behavior, as all participants had carried out a DSH act. However, it is difficult to imagine a feasible alternative for investigating this issue. Third, results cannot be assumed valid for generalization to other nations or cultures. The study was carried out within an East Asian, specifically an urban Taiwan, context. Differences in health service provision as well as cultural and societal issues may lead to specific help-seeking patterns. This was a key reason why we assessed the issue in this population and accepted that results may differ from those in other cultures. A further consideration in this study is the timing of interviews. Interviews were all conducted within a month after a DSH episode. On one hand, this should help improve the likelihood of accurate event and emotion recall proximal to the index act. However, participants with more extreme views on help-seeking behavior may not reflect their real thoughts or situations, and answers may be affected by short-term turbulent emotions after the DSH event. Therefore, interpretations of more extreme responses should be made with caution. Finally, the focus for the interviews and analysis was on help-seeking behavior in terms of how and why such occurred. Other important issues such as culture and suicide intent were only considered in the analysis to the extent that they appeared to influence help-seeking behavior.

Conclusions

Within a Taiwan context, findings emphasize the importance of the therapeutic relationship in facilitating disclosure of distress, particularly in primary care and general medical services. The need to further enhance the public’s gatekeeping role to encourage access to specialist sources of help is also evident because a sizeable proportion of people planning DSH may communicate such only to informal sources. This is particularly so in the presence of strong social support and may be similarly so in the presence of negative attitudes toward health services as a source of help. Nurses can play a part in informal networks and assume the role of gatekeepers to help people engage psychiatric services or other continuous care services such as the nursing case management.

Acknowledgment

The authors thank the Ministry of Education of Taiwan Government for its financial support in drafting this manuscript.