Self-harm attempters' perception of community services and its implication on service provision.

OBJECTIVE: This study aimed at exploring the attempters' perception of community social services included any barriers to seeking help and services.
METHOD: The participants were patients with self-harming behavior aged 15 years or above. A set of guiding questions were designed to explore the general barriers and accessibility to community social services. A voice recording was made, which was later converted into a text transcript and then preceded for content analysis with co-occurrence and similarity matrix interpretation. Two males and nine females with a history of self-harm aged between 24 and 58 years were recruited for the interviews. RESULT: The participants had diverse experiences and backgrounds, and attitudes toward community social services. However, there was a shared perception of the need to enhance community social services. There were four main themes and 12 sub themes identified. The main theme included the service availability, service accessibility, affordability and acceptability. For details, participants were unaware of the available types of care/social services in the community, and were unaware about the nearby social services. They also suggested extending service hours and focused services should be offered to help people with different backgrounds and needs. Actually, those with experience of service utilization had both positive and negative perspectives and they gave suggestions for service delivery, mainly extending service hours and offering focused services such as for gambling control and financial planning. In view of interaction with service providers, counseling skills and trust were highly appreciated by the participants.
CONCLUSION: The results identified common circumstances of falling into financial hardship (gambling) and social fragmentation (divorce, poor family relationships, and poor marital relationships), which also suggested to enhance services on center location, service arrangement, and skill of caregivers.

Introduction

The results identified independency between the geographical accessibility of service and the rate of self-harm. Further exploration was carried out to explore the individual's perception of community services, which was expected to have an influence on service usage and rate of self-harm. Previous studies identified an association between psychiatric disease and self-harm attempt [[1], [2], [3]]. It was also found that most self-harm attempters had multiple psychosocial problems, which facilitated subsequent repeated attempts [4,5]. In addition, the assessment of self-harm attempters' psychosocial needs was associated with the outcome and recognized as an important part of self-harm management [[5], [6], [7]]. Addressing self-harm attempters' psychosocial needs was closely related to the reduction of subsequent repetition [8]. On the other hand, self-harm attempters reflected the varied hospital experiences [9]. To date, the service users' experience and perception of community service have not been fully explored. This aspect has been relatively neglected in comparison to the many epidemiological studies on self-harm. Though the assessment of psychosocial needs has been addressed, patients' perception and experience of using community services needs to be explored, which would help to improve resources and focus of interventions and to reduce self-harm repetition.

This study explored attempters’ perception on existing social and health services across the study area. The data collection was done by structured interview in the emergency department. A set of interview question were developed with reference to the National Collaborating Center for Mental Health and Royal college of Psychiatrists. The focus was on the availability and accessibility of related services, and included any barrier to seeking help and services. In addition, the perception on the quality of service was explored from the perspective of the process of care, information delivery, privacy protection, staff attitude, and appropriateness of professionals.

Methods

Participants

The target participants were patients with self-harming behavior aged 18 years or above, who was staying in the emergency medial ward. The self-harming behavior was defined as intended behavior that resulted to the physical injury or type of self-poisoning. Inclusion criteria included not suffering from severe physical or life-threatening illness, conscious and emotionally stable, and Chinese speaking. Children under the age of 18 might not be suitable because it may be difficult to get consent from parents. In addition, children or patients with mental retardation were not included because they might not be able to answer the questions and express their feelings clearly. The study was approved by the institutional research ethics committee.

Interview process

We first informed patients about the purpose of the study and details of the interview such as duration of the interview and their right to withdraw, etc. After allowing them to consider participating in the study, written informed consent was obtained. If not, we asked them their reasons for refusal. If the reasons were related to misunderstanding something about the interview, then we would further explain the details to them and try to get their consent. In cases of refusal, the reasons for refusal were recorded for later analysis.

The interview was conducted in a quiet environment so that the patients and the interviewers could hear the questions and answers clearly. Apart from a quiet environment, visual disturbances were reduced to avoid distraction. The A&E department was not a suitable place for interview as it is always noisy and many patients and medical personnel are walking around. A meeting room was arranged for the interview to avoid both auditory and visual disturbances, and to provide a comfortable environment for the interviewer and patients.

A suitable semi-structured interview was used as it provided some flexibility and allowed the interviewer to further explain any question and to ask follow-up questions. First, warm-up questions were asked at the start of the interview, such as asking about the patient's background and letting them introduce themselves. Second, the interviewer asked the prepared questions in an organized way. Open-ended questions were offered in this situation as the patients could express their opinion more freely. At the same time, the interviewer recorded the answer for further analysis. In addition to verbal communication, the interviewer also paid attention to the patients' non-verbal cues such as facial expressions, intonation, and eye contact as they might be related to their underlying psychological states. After asking questions, the interviewer asked the patients if they had any queries, which they tried to answer. Finally, the interviewer politely thanked the patients for their participation.

The following guiding questions explored the local community factors to determine the general barriers and accessibility to social and health service access and utilization. The participant was encouraged to provide suggestions on the service arrangements that could improve the accessibility of social and health services.(1) Welcome and introductions(3 min):Settling in and getting comfortable; Introduction–brief overview of purpose, format of interview, assurance of confidentiality. (2) Background questions (5 min): When did you first seek help for self-harm and whom did you contact? (Please describe this first contact.) What means of transport were used (e.g., taxi, bus, ambulance, etc.)? (3) Access and barrier to services (15 min): Were you involved in thinking through what care/services you might need? Did you have enough information about your nearby health and social service facilities? What helped or did not help you gain access to services? In what ways has your family/friend support worker helped you with your health care? What are some other things he/she could do to help? Is there anything you would like to say, either positive or negative, about the availability of service in your residential area? Did you feel that you were offered the appropriate social and health service within your residential area? Any other suggestion about the access of health and social service within your residential area?(4) General Satisfaction(5 min): Are you happy with the availability of social and health service you receive in the community? What sorts of things/information are helpful/useful to you? What do you like about visiting the emergency department? What would you like to see happen that would be different for you so that you could access more or better services?(5) Wrap – up/Close(2 min): Other comments/questions.

As the target patients have self-harming behaviors, their psychological states might not be stable, even though they may seem to be calm. Hence, there may be a worry that some questions may elicit emotional distress or, in serious cases, sudden self-harming behaviors during the interview. Therefore, all tools/items were removed from the interview room to prevent any harm such as glass and sharp instruments. In addition, some self-harming patients may have a background of mental illness such as depression and anxiety disorder, so we needed to pay more attention to them to avoid worsening their situation.

Data analysis

The voice recording made during the interview was transcribed as text. Thematic analysis was done by the QDA Miner 4.0, and categories for each item and descriptions were coded. Repeated procedures were done on the recruited cases to identify sub-themes and main themes. Similarity index and co-occurrence matrix were explored for associations between major categories and minor categories. After the transcript data was analyzed for minor and major themes, a review was conducted to ensure the categorization of data was appropriate. In addition, similar data were merged into sub-categories. The analyses were then summarized into main themes and sub-themes.

Results

Sample characteristics

Two males and nine females with a history of self-harm aged between 24 and 58 years were recruited for the interviews (Table 1). Same as in previous studies [10], the number of intent to suicide was low. There was only one patient who had the intent to end his life. All the attempters described an upset family life (e.g., divorce, poor marital relationship, separation). Only two of the 11 participants were employed and ten had only primary education. Five of them had a psychiatric history (depression, adjustment disorder, impulsive/hot tempered).

Table 1

Demorgraphics of participants in the semi-structured interview.

Case Code	Gender	Age	Occupation	History	Social Factors	Family Factors	Place of Self-harm	Event
1	F	24	Customer services	Nil	Poor social skill	Conflictual Parents Relationship	Home	Intend to jump
2	F	39	Housewife	Nil	Pathological gambler	Nil	Home	Intend to jump
3	M	45	Chief	Nil	Hx of using Ketamine/Cocaine	Poor relationship with 2nd marriage wife	Home	Drug overdose
4	F	58	Clerk	Depression	Nil	Poor relationship with busband	Home	Drug overdose
5	F	50	Housewife	Impulsive/hot tempered	Handling of a son with Asperger	Nil	Home	Holding chopper with suicida idea
6	F	48	Government Servant	Adjustment disorder	After a workplace violence and recent workload increased	Nil	Home	Drug overdose
7	M	57	Ship worker	Nil	Pathological gambler	Devoiced already due to gambling	Home	Burning Charcoal
8	F	32	Library assistant	Adjustment disorder	Nil	Nil	Refer from clinic	Expressed suicidal idea
9	F	54	Nil	Substance abuse	Quarrel with boy friend/cheated by boy friend	Nil	Home	Drug overdose
10	F	53	Housewife	Nil	Nil	Poor relationship with ex husband	Home	Expressed suicidal idea
11	F	32	Housewife	Adjustment disorder	Financial Tightness	Poor relationship with husband wo is a beavy gamber	Home	Expressed suicidal idea

Personal circumstance: co-occurrence and similarity

The interview provided a picture of the patient's circumstances and how this contributed to self-harm. The interview transcript was prepared and imported to QDA Miner 4.0 for coding and content analysis, which included co-occurrence analysis and similarity matrix analysis (Table 2).

Table 2

Distribution of codes (personal circumstance) being used throughout the content analysis.

Main themes	Sub themes	Number of time this code has been used	Percentage of coding associated with this code(%)	Number of cases in which this code appears	Percentage of cases containing this code(%)
Social Factors	Gambling	14	5.90	3	27.30
	Affected by suicidal behavior from others	1	0.40	1	9.10
	Already with social workers follow up	16	6.70	8	72.70
	Support from friends	8	3.40	6	54.50
Family Factors	Devoiced	2	0.80	2	18.20
	Poor marriage relationship	18	7.60	4	36.40
	Poor family relationship	13	5.50	8	72.70
	Poor family economic features	8	3.40	3	27.30
Individual factors and experiences	Psycho-emotional problems	6	2.50	3	27.30
	Stress management	9	3.80	5	45.50

From the content analysis, ten sub-themes were identified as “already with social worker follow-up” “divorced” “gambling” “poor family economic features” “poor family relationship” “poor marriage relationship” “Psycho-emotional problems” “stress management” “suicidal behavior from others” and “having support from friends”. The majority of the attempters had problems of poor family relationship (n = 8,72.7%), already had social worker follow-up (n = 8, 72.7%), and were supported by friends (n = 6, 54.5%).

Further categorizing and merging of the sub-themes was done and three main themes were identified: social factors (4 sub-themes), family factors (4 sub-themes), and individual factors and experience (2 sub-themes) (Table 2).

From the co-occurrence matrix, the top four highest co-occurrences were found between poor family relationships and already with social worker follow-up (7 episodes); support from friends and already with social worker follow-up (5 episodes); stress management and poor family relationship (5 episodes); and stress management and support from friends (5 episodes).On the other hand, no co-occurrences were found between psycho-emotional problems and three sub-themes (divorced, gambling, and poor marital relationship); suicidal behavior from others and six sub-themes (already with social worker follow-up, divorced, gambling, poor family relationship, poor marital relationship, and stress management); and support from friends and suicidal behavior from others.

The similarity matrix by Jaccard's coefficient (occurrence) was calculated from a four-fold table as a/(a+b + c), ‘a’ represented cases where both items occur, and ‘b’ and ‘c’ represented cases where one item was found but not the other. From the transcript, the high similarity index was identified between poor family economic features and divorced (0.667); poor family relationship and already with social worker follow-up (0.778); poor marital relationship with already with social worker follow-up (0.5) or with poor family relationship (0.5); stress management and poor family relationship (0.625); and support from friends and stress management (0.833) (Table 3).

Table 3

Similarity matrix by Jaccard's coefficient of sub-themes in personal circumstance.

Sub themes	Already with social workers follow up	Devoiced	Gambling	Poor family economic features	Poor family relationship	Poor marriage relationship	Psycho-emotional problems	Stress management issue	Suicidal behavior from others	Having support from friends
Already with social workers follow up	1.00
Devoiced	0.25	1.00
Gambling	0.22	0.25	1.00
Poor family economic features	0.22	0.67	0.20	1.00
Poor family relationship	0.78	0.11	0.10	0.10	1.00
Poor marriage relationship	0.50	0.20	0.17	0.17	0.50	1.00
Psycho-emotional problems	0.10	0.00	0.00	0.20	0.22	0.00	1.00
Stress management	0.44	0.17	0.14	0.14	0.63	0.29	0.33	1.00
Suicidal behavior from others	0.00	0.00	0.00	0.33	0.00	0.00	0.33	0.00	1.00
Having support from friends	0.56	0.14	0.13	0.13	0.75	0.25	0.29	0.83	0.00	1.00

Through the average-linkage hierarchical clustering approach, clusters were constructed from the similarity matrix. The clusters linkage was identified by a dendrogram, sub-themes that tended to co-occur were combined at the early stage and those with higher independency were combined at the end of the procedures. The most dependent sub-themes were found between stress management and support from friends (0.833); already with social worker follow-up and poor family relationship (0.778); and divorced and poor family economic features (0.667).

Perception on accessibility: co-occurrence and similarity

From the content analysis (Table 4), 12 sub-themes were identified as “unawareness on the type of care/social service” “unawareness on available nearby care/social service” “positive feedback on the nearby services” “negative feedback on the nearby services” “suggestion on the offer from nearby services” “location” “attend private practice” “staff skills, including trust” “expectation” “stigma” “unawareness of service need” and “happy with the available service in the community”. The top three highest number of occurrences were found in “suggestion on the offer from nearby services” (31 episodes), “negative feedback on the nearby services” (27 episodes) and “positive feedback on the nearby services” (21 episodes).

Table 4

Distribution of codes(perception of community social service) being used throughout the content analysis.

Main themes	Sub themes	Number of time this code has been used	Percentage of coding associated with this code(%)	Number of cases in which this code appears	Percentage of cases containing this code(%)
Service availability	Unawareness on the type of care/social service	14	5.90	6	54.50
	Unawareness on available nearby care/social service	17	7.10	6	54.50
	Positive feedback on the nearby services	21	8.80	7	63.60
	Negative feedback on the nearby services	27	11.30	8	72.70
	Suggestion on the offer from nearby services	31	13.00	8	72.70
Service accessibility	Location	6	2.50	3	27.30
Affordability	Attend private practice	3	1.30	3	27.30
Acceptability	Staff skills, including trust	5	2.10	5	45.50
	Expectations	9	3.80	5	45.50
	Stigma	1	0.40	1	9.10
	Unawareness of service need	5	2.10	3	27.30
	Happy with the available service in the community	4	1.70	3	27.30

Further categorization of the themes was made to identify four main themes on the perspective of accessibility, which included service availability (5 sub-themes), service accessibility (1 sub-theme), affordability (1 sub-theme), and acceptability (5 sub-themes). The majority of the codes were assigned to service availability (110 episodes). There were three patients attending private psychiatric service under the main theme of “affordability”.

From the co-occurrence matrix, the top three highest co-occurrences were found between the negative feedback on the nearby services and positive feedback on the nearby services (6 episodes); suggestion on the offer from nearby services and negative feedback on the nearby services (7 episodes); and unawareness on the type of care/social service and suggestion on the offer from nearby services (5 episodes). On the other hand, there were no co-occurrences between location and attended private practice; stigma and three sub-themes (attend private practice, location and staff skills including trust); unawareness of service need and attended private practice; unawareness on available nearby care/social service and stigma; and unawareness on the type of care/social service and two sub-themes (stigma and unawareness of service need).

From the similarity matrix by Jaccard's coefficient (occurrence) (Table 5), a high similarity index was identified between positive feedback on the nearby services and negative feedback on the nearby services (0.667); suggestion on the offer from nearby services and negative feedback on the nearby services (0.778); and unawareness on the type of care/social service and suggestion on the offer from nearby services (0.556).

Table 5

Similarity matrix by Jaccard's coefficient of sub-themes in the perception of community social services.

Sub themes	Attend private practice	Expectations	Happy with available service in the community	Location issue	Negative feedbackon the nearby services	Positive feedbackon the nearby services	Staff skills, including trust	Stigma	Suggestion on the service provided by nearby services	Unawareness of service need	Unawarenesson available nearby care/social service	Unawarenesson the type of care/social service
Attend private practice	1.00
Expectations	0.33	1.00
Happy with available service in the community	0.20	0.33	1.00
Location issue	0.00	0.14	0.20	1.00
Negative feedback on the nearby services	0.38	0.44	0.22	0.22	1.00
Positive feedback on the nearby services	0.25	0.50	0.25	0.11	0.67	1.00
Staff skills, including trust	0.33	0.25	0.33	0.60	0.44	0.33	1.00
Stigma	0.00	0.20	0.33	0.00	0.13	0.14	0.00	1.00
Suggestion on the service provided by nearby services	0.38	0.44	0.22	0.22	0.78	0.50	0.44	0.13	1.00
Unawareness of service need	0.00	0.14	0.20	0.20	0.38	0.25	0.14	0.33	0.22	1.00
Unawareness on available nearby care/social service	0.29	0.10	0.29	0.29	0.40	0.30	0.57	0.00	0.40	0.29	1.00
Unawareness on the type of care/social service	0.29	0.22	0.13	0.29	0.40	0.30	0.38	0.00	0.56	0.00	0.33	1.00

Through the average-linkage hierarchical clustering approach, clusters were constructed from the similarity matrix.The cluster linkage was illustrated by dendrogram, sub-themes that tended to co-occur were combined at the early stage and those with higher independency were combined at the end of the procedures. The most dependent sub-themes were found between negative feedback on the nearby service and suggestion on the offer from nearby services (0.778); and location and staff skill, including trust (0.6).

Perception on service coverage: four sub-themes

Service availability

In view of the awareness of the types of available care and social services, six participants indicated that they were not aware of the types of available services in the community, especially services helping with debt problems.Contact by telephone was seen to be an acceptable means of seeking help. On the other hand, the perception of service availability was seen to be insufficient and types of services were limited.

Six participants were unaware of the available nearby care/social services within their residential area. However, they suggested changes to the service hours and type of services being offered. For instance, they preferred the integration of social services with medical/nursing to help their situation, and proposed extended service hours to fit around their work.

“ I do not think the social worker could help me with my debt problem. They would have no solution on it. I think phone contact would be enough for me.”(Case 1)

“I did not want to seek help from others. I always found people to help me in the past, but I do not want to seek help now. As I am an adult, I should handle my problems by myself.” (Case 2)

“I know that there is a social service center nearby, but I did not want to explore which services could help me. I think telephone contact would be enough.” (Case 6)

“I do not have the channel for getting the information. I could not see any promotional activities.“(Case 8)

“I have never thought about the social services that could help me.“(Case 9)

“I do not have information / idea about the available social services. I think that social workers would not help in my situation. I would like to seek help from psychiatric doctor more.” (Case 10)

Five participants recounted positive experiences from the utilization of community care/social services. Regular meetings with designated nurses or social workers were much appreciated by the participants. Phone contact with the case nurse or social worker was helpful in the case of frustration.

“The opening hours are enough for me. I could meet the social worker before the closing time. The availability of social worker would be fine in each service center.“(Case 5)

“I could meet my case social worker once a month, they could solve the problems before some serious incident happened. I feel positive after talking with the community health nurse.“(Case 6)

“I have phone contacts, so it is not difficult to contact the social worker .Social worker helped me to apply for divorce, he helps me a lot. I thank him for his help to apply for resource support e.g., milk powder and security assistance.“(Case 11)

“I could ask the social worker to help when encountering problem, he could give me advice. If a social worker follows my situation, it would be better for supporting my life.“(Case 8)

“I have called the social worker and said that I have bought several packs of charcoal. They called the police to save me. I went to meet the social worker, she gave me some noodles and canned food to eat, she also helped me to apply for urgent funds in Kwai Hing Government Building. I think the hotline could help me to improve my emotions, but it could not solve my problem of debt. The social worker referred me to a financial company and helped me to analyze my situation (bankruptcy).“(Case 7)

In contrast, they also had negative experiences of using community services. The accounts from eight participants included long waiting times before meeting the social worker, and that their attitudes should be more empathetic. In addition, more information about the social services should be provided.

“I think that the waiting time should be shorter. I needed to wait for 3 months to meet them. I think the social worker should be empathetic.“(Case 2)

“I think the social worker could not help me anymore.I have been followed up by a social worker, he said the attitude of my wife could not be handled as she always nagging and arguing with me.“(Case 3)

“I think that the information about the social service is not enough.If I could seek help from social worker earlier, the incidence would not have happened yesterday.“(Case 4)

“I had two meetings with the social worker before. They are too busy that I don't think they can help me.My son does not want to go there, ‘very old fashioned style’“.(Case 5)

“I tried to see the social worker in West Kowloon. As I am a civil servant, she gave the feeling to me that I wanted to seek benefits, so I don't like the social worker.I have negative feeling about their attitude.They have stopped following me after 1 year.“(Case 6)

“It is difficult to talk to someone or find activities of the nearby community center.” (Case 8)

“I do not have confidence in Hong Kong social workers, I prefer to seek medical help instead. I think the social worker could only help with finance/resource issues.They help people from the mainland more than local people.I have negative feelings about their attitude.“(Caser 10)

“I think that social workers could not help me to solve my problem. I have been followed by the social worker for 6 months. The social worker is busy, I just see him for a short time each visit.” (Caser 11)

The eight participants provided suggestions to improve the nearby services, which include more nursing services, extend service hours to meet working schedules, and more focus on patients with special needs.

“I have confidence in nurses, more available nursing services are needed in social service centers.It would be good to have someone follow my situation after discharge from hospital. Also, I prefer to seek help from psychiatric doctor instead.“(Case 10)

“Early appointment would be good for me. Opening hours between 9 – 6 would be good for me, I could go there after work.If they have some ways to help people to relax, I think that there will not be so many people committing suicide.Many people do not know they have a social Center in Kwai Chung.“(Case 4)

“ I think that health and social services could be combined at community centers, it can be a one-stop service. The center could arrange some target activities for groups of people.I think that social worker could develop more types of services.” (Case 2)

“I think regular follow-up with arranged activities would be good for the youth.The social worker should spend more time to take care of us.If there were specialized centers to handle problem children, it would be very helpful for those parents.“(Case 5)

“If a nearby community center is available, we can go there to find some people to talk to or meet some new friends. If I feel unwell or sad, I will go there to find someone to talk to. I think the center can organize some activities as well.Those activities could help me to relax and calm down my mind. If medical service available in center was good, the social worker could refer you to a doctor if necessary.“(Case 8)

“The opening hours may not fit my working time.” (Case 3)

“Phone contact is enough for me.I think they should have more centers, but not necessary for extended service hours for all centers.“(Case 6)

Service accessibility

Two participants suggested enhancing the accessibility of existing community social services center in Kwai Chung. The existing center was too small or they were not aware there was a community center located in Kwai Chung, where the rate of self-harm is high.

“The center in Kwai Chung is too small, it is far from my home.“(Case 2)

“I was not aware there was a community center located in Kwai Chung.” (Case 1)

Affordability

In view of affordability, all participants reported no financial implications in accessing the nearby services as the transport fees were cheap enough for them. Three participants reported that they had been followed up by private psychiatric doctors

“I attended a private psychiatric doctor.” (Case 10)

“I have seen the doctor, and we are friend. I usually go to find her.” (Case 4)

“I have seen the government doctor and private doctor.” (Case 6)

Acceptability

Five participants appreciated the staff counseling skills and a rapport was built between the social workers and participants. In case of frustration, they were helped to be relaxed after talking with social workers.

“I think that she is listening to me and gives me advice. For example, I believe her and talk to her about something that happened before or something that makes me feel nervous.” (Case 6)

“They would help me to relax and talk to me.“(Case 1)

“He knows it. We have gone to family court already and we have made an application.” (Case 3)

“She has met my husband and talked to him.We talked in the Kwai Fong Garden to solve the problem together.” (Case 4)

“He will give me suggestions, we are just like friends.“(Case 2)

Participants also expected to have a hotline that could give them advice in case of emergency. Post-hospital discharge follow-up was also welcomed by them.

“I have called the hotline to help me give up gambling. I hope to contact the social worker as soon as possible.” (Case 7)

“If someone could follow me after hospital discharge.“(Case 10)

“I want her to talk with my son and give him advice.Contact special organization to handle hyperactivity disorder in children.” (Case 5)

“Change some services to target people.” (Case 2)

“I hope that the social worker could help me after hospital discharge.“(Case 7)

However, participants were concerned about being stigmatized when using the community psychiatric service.

“ I worry that my son would be stigmatized by people.“(Case 5)

In case of emergency, participants did not consider going to social service center, instead they sought medical help at hospital.

“I do not go to the center in an emergency.” (Case 3)

“If I have any problem, I will call and go to psychiatric outpatient clinic.” (Case 11)

“May be my son and I do not get along well.I seldom go to the social service center.“(Case 5)

Discussion

The participants had diverse experiences and backgrounds, and attitudes toward community social services. However, there was a shared perception of the need to enhance community social services. The preferences complied with the existing literature. Among the 11 participants, only two were male, which reflected the difficulty in seeking male self-harm victims for participation in this interview. This may be due to a cultural or social norm issues in the Chinese. Psychiatric illness was found to be one of the major factors of self-harm, which include depression, adjustment disorder, and substance abuse. In view of social circumstances, pathological gambling, poor social relationships, drug abuse, psychological trauma, and financial hardship were the main contributing factors for self-harm. Four participants reported poor relationships with their family including poor marital relationships. In view of their background, the three main themes reported by participants were social factors, family factors, and individual factors/experience. From the interview transcript, social circumstances of gambling, affected by suicidal behavior of others, and support from friends were identified. In view of family support, patients reported experiences of poor marital relationships, poor family relationships, poor family economic features, or were divorced. This concurred with previous literature on social fragmentation and financial hardship. On the other hand, psycho-emotional problems and poor stress management were associated with self-harm. Although some participants were followed up by social workers, participants with poor family relationship were linked to self-harm. From the similarity matrix, most participants with poor family relationships were being followed up by social workers; those with poor stress management were being supported by friends; those with poor family economic features were most likely divorced; and those with psycho-emotional problems were most affected by the suicidal behavior from others.

The patient's perception on accessibility could be classified into availability, accessibility, affordability, and acceptability. In view of service availability, participants were unaware of the available types of care/social services in the community, and were unaware about the nearby social services. Actually, those with experience of service utilization had both positive and negative perspectives and they gave suggestions for service delivery, mainly extending service hours and offering focused services such as for gambling control and financial planning.

Because of the small urban setting, the location of the community social service was not recognized as a problem. In addition, some participants also attended private psychiatric services, so financial costs might not be a factor. In view of interaction with service providers, counseling skills and trust were highly appreciated by the participants. They also suggested extending service hours and focused services should be offered to help people with different backgrounds and needs.

Limitations and future studies

This study consisted of a semi-structured interview. Although interviews can cover a wide range of information, it is better that the interviewer obtain answers to some specific questions. However, using a semi-structured interview might lead to a certain level of bias. Bias might occur when the interview was restricted by sets of guiding questions. Although some important points may still have been missed with structured questions, the interviewer was allowed to elaborate or add new questions to allow the participant to give more information. Indeed, it was suggested that the interviewer should first ask the interviewee to describe the overall picture, followed by the semi-structured interview with added supplementary questions. This was to ensure that the patient provided the information necessary and at the same time allowed the patient to divulge what on his/her mind. Yet, the suggested procedure may take too much time, hence become less effective. On the other hand, the main method of obtaining important themes in this research was by assigning codes to phrases and words with similar meaning. Though this could help turn the interview transcript into some quantifiable data, there might be errors within this method, leading to misinterpretation and subjective bias. This problem might occur when relying on the occurrence of certain key words, co-occurrence, and similarity analysis.

In Hong Kong, the small urban areas allow short traveling time with convenient available transport. However, the limited utilization rose from culture and belief especially in attending community psychiatric services, which might include fear of being stigmatized in the community. This highlights the need for public promulgation and education on the availability of community social services, which includes the scope and types of services available. In addition, people should be educated to self-initiate seeking help from services when necessary.

In agreement with the previous findings, pathological gambling, financial hardship, and stress management need to be addressed. Community programs on prevention and intervention would be expected to help to relieve these social problems to mitigate potential self-harm. In particular, financial hardship could be tackled through organizations and advice centers to help people to manage the consequences of their financial burden. The findings recommend extending service hours to fit around the normal working hours, which would provide more opportunities for people to seek help. Shortening waiting times would definitively help as earlier intervention would prevent repetition. Since participants appreciated counseling and trust from some social workers, ensuring they are skilled in handling self-harm and can establish a rapport with clients would be highly recommended.

Conclusions

The studies highlighted the attempters’ perception on existing community services in relation to their personal and social circumstances. The findings highlighted common circumstances of financial hardship (gambling) and social fragmentation (divorced, poor family relationships, and poor marital relationships), and common recommendations included enhanced service types and extended service hours. There were some limitations in the study such as the small number of samples and potential subjective bias in the content analysis. Further large-scale studies are suggested in other districts of Hong Kong, which would help to identify any differences and provide suggestion for self-harm prevention.