Evaluation of the Relation between Deliberate Self-Harm Behavior and Childhood Trauma Experiences in Patients Admitted to a Secondary-Care Psychiatric Outpatient Clinic for Adolescents and Young Adults

Abstract

Introduction

The aim of this study was to determine the prevalence and the features of deliberate self-harm (DSH) behavior in patients admitted to the psychiatric outpatient clinic for adolescents and young adults and also to detect the association between the act of DSH and childhood traumas.

Method

In this study, we included all patients who were admitted to the secondary-care psychiatric outpatient clinic for adolescents and young adults in Bakırkoy Research and Training Hospital Psychiatry, Neurology and Neurosurgery throughout a month. A sociodemographic data form, the Childhood Trauma Questionnaire (CTQ-28), Childhood Abuse and Neglect Question List and the Deliberate Self-Harm and Intent Screening Form were applied to three hundred participants.

Result

The prevalence of DSH was 50.0% among the participants (56.8% for females and 28.8% for males). Childhood abuse was detected in 57.0% of all participants (60.4% of females and 46.6% of males). Among patients with act of DSH, the rate of childhood abuse was 71.3%, while it was 42.7% in the subjects without act of DSH.

Conclusion

DSH is a common behavior among adolescent psychiatric patients. It is more common in females than in males. The prevalence of experience of childhood abuse and neglect is remarkably high and is associated with self-harm behavior.

Introduction

Deliberate self harm (DSH) is defined as any behaviour which causes to self-injury independent of suicidal thought. DSH may occur at any age. However, it has been found with remarkably high rates in adolescents and young adults (1). In adolescents, DSH may reflect a transient period of trouble and not carry further risk or may be a significant indicator related with psychiatric problems in the future periods of life and increased risk of suicidal behavior in the future (2). This behaviour is thought to be a method for alleviating emotions which give pain and for decreasing stress or a non-adaptive coping mechanism or a method to arrange emotions by many investigators. It prevents treatment and inter-personal relations and may result in death even if it is unintentional (3).

Separation of the mother and father, intra-familial violence, physical and sexual abuse by the parents have been reported with a significantly higher rate in patients who display self-harm behaviour (4).

Abuse is deliberate malicious use of one side by another side in inter-personal relations such as that side is harmed. The concept of childhood abuse corresponds to emotional, physical and sexual abuse and injury. In addition, encouraging such a relation, allowing the relation and being inconsiderate or profiting from this is also an abuse (5).

There is no consensus on the prevalence of childhood abuse. Rates ranging from 6% to 62% for women and from 3% to 16% for men have been proposed related with the prevalence of sexual abuse in the population. The findings show that girls are exposed to sexual abuse with higher rates compared to boys and the exploiters are mostly family members. It is understood that boys are abused with a lower rate by a relative and with a higher rate by a stranger (6).

The relation between physical self-harm and suicidal behaviour with childhood traumas has been demonstrated clearly in many different studies. This relation in the childhood was recognized by Green for the first time (7). Afterwards, in the first controlled study, Green showed that 41% of the children and adolescents who were exposed to physical abuse displayed suicidal behaviour and deliberate self-harm behavior and this rate was significantly lower in the healthy children in the control group (8). A similar relation has been shown in preschool children, adolescents and adults in many studies (9,10). Some studies in which measurement tools assessing severity and frequency were used showed that there was a dose-response relation between trauma and self-harm behaviour (11). Physical self-harm behaviour occurring as a result of destructive psychological experiences is not observed only in humans, but also in more simple living beings. For example, rhesus monkeys who were isolated in infancy and prevented to be cared by their mothers were reported to display behaviour including beating themselves, hitting their heads on hard surfaces and shaking their heads. Thus, self-harm behavior is reported to be a primitive behavior pattern which occurs as a result of traumatic relation experienced with the caregiver during infancy and childhood rather than dynamic factors including conflict, guilt, superego pressure and self-agression and which can be observed in more simple living beings (12).

In this study, it was aimed to determine the prevalences of deliberate self-harm behaviour and childhood traumas in patients presenting to the adolescent and young adult psychiatry outpatient clinic and to determine if there was a relation between them.

Method

All patients who presented to the secondary-care psychiatric outpatient clinic for adolescents and young adults in Bakırkoy Ord. Prof. Dr. Mazhar Osman Mental Health Hospital consecutively for a period of one month and treated and followed up were included in this study. The secondary-care psychiatric outpatient clinic for adolescents worked as a unit where adolescents who were examined in the primary care psychiatric outpatient clinic for adolescents and in whom it was thought that psychotherapy would be beneficial in addition to pharmacotherapy after a certain period of follow-up, who had no psychotic disorder and who could comply with therapy were followed up and treated. The participants were informed that any personal information would be kept secret, they could withdraw from the study at any time and not participating in the study would not affect the treatment program in any way.

After being informed about the study 300 patients in the 14–20-year age group who gave consent for participation, who had no active psychotic symptoms and who had the mental capacity to understand and answer the questions in the questionnaire were included in the study group. During this process, one female patient was excluded from the study because she was illiterate and 13 female and 3 male patients were excluded because they did not accept to participate in the study.

The deliberate self-harm and intent screening form was applied to the patients who were being followed up in the secondary-care psychiatric outpatient clinic for adolescents. The subjects who had deliberate self-harm behaviour and who did not have deliberate self-harm behaviour were compared in terms of childhood traumas.

Tools

1. Sociodemographic Data Form

Gender, age, education level, working status, togetherness of the parents, number of siblings, number of people living at home, the person who brougth up the patients, education levels of the parents, time of psychiatric treatment and presence of familial history of psychiatric disease were questioned.

2. Deliberate Self-harm and Intent Screening Form

This form is an internationally accepted standardized anonymous questionnaire form prepared by the members of a multi-center study which investigated the self-harm behaviour in children and adolescents in Europe and was used to collect data in all centers where the study was conducted. Its Turkish translation was made by the investigators from the article of Morey et al. published in 2008 in Ireland which was one of the centers where the study was conducted and was used to screen deliberate self-harm and intent. The definition of deliberate self-harm was as follows: Action of an individual to realize one or more of the following without a fatal outcome.

Interventional behaviour with the aim of self-harm (cutting oneself, jumping from a high place)

Ingestion of prescription substances in excessive doses or generally in recognizeble theraputic doses.

Ingestion of illegal drugs for the aim of amusement or for the aim of self-harm

Ingestion of a substance or object which cannot be digested

The following questions were used to define deliberate self-harm: “Have you ever deliberately taken high dose of drug or have you ever attempted to harm yourself (for example, cutting yourself)” For the aswer the options “No/Yes, once./yes, more than once” were used. Intent of self-harm was investigated using the following questions: “Has there been an occasion in the last one month or year that you seriously thought of taking a high dose of drug or harming yourself (for example, cutting yourself) but did not realize it” (13).

4. Childhood Traumas Questionnaire (CTQ-28)

This questionnaire the original form of which was developed by David P. Bernstein (14) in 1995 was adapted to Turkish by Professor Vedat Şar in 1996 with permission of the author. It is a five-point likert type self-report scale. It includes questions which evaluate emotional, physical and sexual abuse and physical and emotional neglect in the childhood.

5. Childhood Abuse (CA) Question List

This form questions physical, sexual and emotional abuse, neglect and incest experiences in the childhood and focuses on the issue if there is abuse and incest behaviour before the age of 18 years. The original name of this semi-structured scale is Childhood Trauma Questionnaire and it was developed by Bernstein in 1997. It was adapted to Turkish by Yargıç et al. and its Turkish validity and reliability study was performed by Yargıç et al. (15). The definitions of Brown and Anderson were used for physical and sexual abuse and incest (16). The definition of Walker, Bonner and Kaufman was used for emotional abuse (17).

Statistical Evaluation

SPSS 13.0 program was used for all statistical assessments. Sociodemographic and diagnostic variables were expressed as numerical and percent values. Biostatistical evaluation of numerical and categorical variables was done with chi-square test based on frequency and percent ratios. Comparison of mean values was performed using independent samples t-test. A p value of <0.05 was considered significant. In addition, logistic regression analysis was performed to determine the relation between DSH behaviour and trauma types. The variables which were found to be insignificant in the logistic regression analysis were excluded and backward elimination model was used in the second stage.

Results

75.7% (n=227) of a total of 300 patients were female and 24.3% (n=73) were male. The mean age was 17.35±1.62 years. Most of the patients were students (71.3%, n=214) and the mean education time was 10.48±1.91 years.

The rate of deliberate self-harm was found to be 50.0% (n=150) in all patients included in the study. This rate was found to be 56.8% (n=129) in girls and 28.8% (n=21) in boys. The sociodemographic properties of the patients with self-harm behaviour are shown in (Table 1).

Table 1.

Sociodemographic properties in the subjects with DSH behaviour and childhood abuse (CA)

	DSH present (n=150, 50%)		CA present (n=171, 57%)
	n	%	n	%
Gender
Female	129	86.0	137	80.1
Male	21	14.0	34	19.9
Education level
Education is not continuing	49	57.0	65	75.6
Education is continuing	101	47.2	106	49.5
Employment
Student	101	67.3	106	62.0
Working regularly	16	10.7	22	12.9
Working irregularly	6	4.0	9	5.3
Not working	27	18.0	34	19.9
Status of the parents
They live together	118	78.7	133	77.8
Divorced/seperate	19	12.7	26	15.2
The mother is dead	2	1.3	3	1.8
The father is dead	11	7.3	8	4.7
Both the mother and the father are dead	118	78.7	1	0.6
Education level of the father
Illiterate	2	1.3	2	1.2
Primary scgool	73	48.7	91	53.2
Secondary school	29	19.3	24	14.0
High school	39	26.0	43	25.1
University	7	4.7	11	6.4
Education level of the mother
Illiterate	12	8.0	16	9.4
Primary school	83	55.3	93	54.4
Secondary school	25	16.7	29	17.0
High school	25	16.7	27	15.8
University	5	3.3	6	3.5
Number of siblings
No sibling	20	13,3	12	7,0
Siblings present	130	86.7	159	93.0
Psychiatric diagnosis
Depressive disorder	63	42.0	55	32.2
Anxiety disorder	39	26.0	60	35.1
Obssesive compulsive disorder	6	4.0	12	7.0
Conversion disorder	12	8.0	15	8.8
Somatoform disorder	3	2.0	3	1.8
Behaviour disorder	27	18.0	25	14.6
Impulse control disorder	0	0	1	0.6

Abuse in the childhood was found in 57.0% (n=171) of all the patients included in the study. It was found that childhood abuse was present in 60.4% of the girls (n=137) and in 46.6% of the boys (n=34). The distribution of the sociodemographic data by presence of childhood abuse is summarized in (Table 1).

The rate of childhood abuse was found to be 71.3%(n=107) in the group with DSH behavior and 42.7% (n=64) in the group without DSH behaviour (Table 3). The difference was found to be statistically significant between the groups (p<0.001, p=0.0001).

Table 3.

Logistic regression rates of the types of childhood trauma

	p	Odds ratio	95,0% CI (Odds ratio)
			Upper	Lower
Physical abuse	0.560	0.820	0.421	1.597
Emotional abuse	0.006	2.270	1.272	4.050
Physical ans emotional neglect	0.022	1.991	1.106	3.586
Sexual abuse	0.142	1.702	0.837	3.460
Intra-familial sexual abuse	0.002	4.379	1.694	11.320

When the relation between CA types and DSH was examined, physical abuse was found with a rate of 28.7% (n=43), sexual abuse was found with a rate of 22.9% (n=33), emotional abuse was found with a rate of 44.7% (n=67), physical and emotional neglect was found with a rate of 38.0% (n=57), intra-familial sexual abuse was found with a rate of 18.7% (n=28) in the group with DSH behaviour. These rates were higher compared to the group without DSH behaviour and the difference between the groups was found to be statistically significant (Table 2). When the relation between DSH and CA types was analysed with logistic regression analysis, a statistically significant relation was found between all trauma types and DSH and this significance was observed to be mostly related with intra-familial sexual abuse (p=0.02, odds ratio=4.379) and emotional neglect (p=0.06, odds ratio=2.270). P value was found to be 0,001 and R2 value was found to be 0,157 for logistic regression model. It was observed that the significance found in the logistic regression model was maintained when the variables with a low significance were excluded and backward elimination model was used (p=0.003, odds ratio=2.295 for emotional abuse, p=0.021, odds ratio=l.962 for physical and emotional neglect, p=0.002, odds ratio=4.468 for intra-familial sexual abuse.

Table 2.

Distribution of presence and types of CA by groups

	DSH absent DSH present (n=150) (n=150)
	n	%	n	%	p	χ2
CA	Yok	86	57.3	37	28.7	0.0001	25.146
	Var	64	42.7	107	71,3
Physical abuse	Yok	123	82.0	107	71.3	0.029	4.770
	Var	27	18.0	43	28.7
Emotionalabuse	Yok	118	78.7	83	55.3	0.0001	18.468
	Var	32	21.3	67	44.7
Physical and emotional neglect	Yok	123	82.0	93	62.0	0.0001	14.881
	Var	27	18.0	57	38.0
Sexual abuse	Yok	134	89.3	117	78.0	0.008	7.049
	Var	16	10.7	33	22.0
Intra-familial sexual abuse	Yok	144	96.0	122	81.3	0.0001	16.055
	Var	6	4.0	28	18.7

According to the Childhood Trauma Questionnaire findings the total score was found to be 9.14±3.18 in the group with DSH behaviour and 7.14±2.20 in the group without DSH behaviour (Table 4). The difference was found to be statistically significant to a great extent (p=0.0001). According to the Childhood Trauma Questionnaire findings the scores of emotional neglect, emotional abuse and sexual abuse were higher in the group with DSH behavior compared to the group without DSH behavior and the difference was found to be statistically significant to a great extent (p=0.0001). The scores of physical neglect and physical abuse were also higher in the group with DSH behavior compared to the group without DSH behavior and the difference was found to be statistically significant (p=0.01, p=0.001, respectively) (Table 4).

Table 4.

Distribution of the Childhood Trauma Questionnaire (CTQ-28) findings by groups

	DSH absent (n=150)		DSH present (n=150)
CTQ-28	Mean	SD	Mean	SD	p
Emotional neglect	2.06	0.95	2.60	1.10	0.0001
Physical neglect	1.23	0.40	1.35	0.45	0.01
Emotional abuse	1.51	0.62	2.15	0.93	0.0001
Physical abuse	1.16	0.40	1.38	0.63	0.001
Sexual abuse	1.18	0.54	1.65	1.02	0.0001
Weighted total score	7.14	2.20	9.14	3.18	0.0001

The rate of DHS was found to be 53.2% in individuals who were exposed to one type of trauma, 67.4% in individuals who were exposed to two types of trauma, 69% in individuals who were exposed to three types of trauma, 75% in individuals who were exposed to four types of trauma and 100% in individuals who were exposed to five types of trauma. It was observed that the rates increased in the group with DSH behavior compared to the group without DSH behaviour as the number of trauma types increased, but the difference was not found to be statistically significant (p=0.117).

The rate of presence of CA was found to be 73.2% in the groups with DSH intent and 43.2% in the group without DSH intent (Table 5). The difference of the rates of presence of CA between the groups was found to be statistically significant (p=0.0001).

Table 5.

Number of trauma types exposed and CA according to presence of DSH thought

Number of trauma types	DSH intent absent (n=150)			DSH intent present (n=150)
	n	%	n	%	p	χ2
1	39	49.4	40	50.6	0.270	5.270
2	16	37.2	27	62.8
3	10	34.5	19	65.5
4	4	25.0	12	75.0
5	1	25.0	3	75.0
CA present	92	56.8	37	26.8	0.0001	27.324
absent	70	43.2	101	73.2

It was observed that the rates increased in the group with DSH intent compared to the group without DSH intent as the number of trauma types increased, but the difference was not found to be statistically significant (p=0.270) (Table 5).

Discussion

The patients included in the study were divided into two groups as the ones with and without DSH behaviour using DSH and intent from. The rate of the patients with DSH behaviour was found to be 50%.

Fliege et al. reported that the rate of DSH history was 30% in a study they conducted with university students (18). In the study performed by Morey et al. in schools, the rate of lifelong DSH was found to be 12.2% in adolescents, 45.9% of the ones who harmed themselves performed this action more than once and the rate of girls was 3-fold higher than boys (13).

In our study, the rates of self-harm were found to be higher compared to the literature. This may be explained with the fact that we conducted our study with our patient population and used a wider definition for self-harm behaviour.

One of the primary weak points in the present literature related with DSH is the fact that DHS has no consistent and well accepted definition (19). One of the reasons of inconsistence is the fact that the terms deliberate self-harm, self-injury and self-mutilation are used in place of each other to express the same condition (20,21,22,23). Another significant problem and reason of inconsistency is the fact that the term DSH is used to define behaviours with different natures. For example, many investigators use the term of DSH to differentiate self-harm and suicide-related behaviour and to conceptualize that self-harm is the antithesis of suicidal attempts (24,25), while others have not been able to differentiate the aim of self-harm and the aim to die (26,27,28,29). Therefore, they include suicidal attempt into the concept of self-harm behavior.

The rate of DHS was found to be 56.8% in girls and 28.8% in boys. It was found that deliberate self-harm behaviour was observed with a higher rate in girls compared to boys. In many research articles, it has been reported that slef-harm behaviour is observed more frequently in women compared to men (2,28,30,31,32). This gender difference was porposed to be related with the fact that women behave less aggressively against other people (33,34). However, there are also studies which have reported that there is no gender difference in occurrence of self-injury behaviour (35,36).

When we examined the psychiatric diagnoses of the patients with DSH behaviour in our study, depressive disorder was in the first order (42%) and anxiety disorder was in the second order (26%). In some publications, it has been reported that the rates of self-harm behavior, suicidal thought and attempt are increased in adolescents with depressive disorder (37,38,39).

The majority of studies related with DSH behavior have investigated the traumatic events which created stress in the childhood. Emotional neglect, psychological or physical abuse and especially sexual abuse experienced in the childhood were related with self-harm behavior in adolescence and adulthood with a high rate (40). In a study performed by Aydın et al. in university students, the most commonly observed childhood trauma was found to be separation from the caregiver (46.1%) and this was followed by witnessing violence (33.1%) (41).

In our study, the rate of CA (childhood abuse) was found to be 71.3% in the group with DSH behavior and 42.7% in the group without DSH behavior. According to the Childhood Trauma Questionnaire findings the scores of emotional neglect, emotional abuse and sexual abuse were found to be significantly higher in the group with DSH behaviour compared to the group without DSH behaviour. The scores of physical neglect and physical abuse were also higher in the group with DSH behaviour compared to the group without DSH behaviour. These results were found to be compatible with the present publications. In recent studies, the functions and forms of DSH behaviour have been examined systematically. There are also single studies investigating the relation between history of CA and DSH behavior. For example, sexual abuse shows a strong relation with many types of personal harms including DSH (42,43). Physical abuse has also been found in some studies (43,44). According to the study performed by Evren et al. there was a relation between physical abuse and self-harm when the demographic properties, familial history and clinical variables were adjusted (45). In our study, a strong relation was found between DSH and intra-familial sexual abuse and emotional neglect.

In some studies in which measurement tools evaluating severity and frequency were used, a dose-response relation was shown between trauma and self-harm behavior (11). In a study performed by Zorog̃lu et al., the rate of self-harm was found to be 24.5% in individuals who were exposed to one type of trauma, 51.5% in individuals who were exposed to two types of trauma and 66.7% in individuals who were exposed to three or more types of trauma (46). In accordance with the present literature, the rates were observed to be increased in the group with DSH behavior compared to the group without DSH behavior as the number of types of trauma increased in our study.

The number of population screening studies related with the prevelance of childhood abuse and neglect experiences is limited (46). In this study, childhood abuse and neglect experiences were determined retrospectively based on self-reports of the patients in the way as they were described. Thus, it cannot be argued that the data obtained reflect the actual frequency of abuse and neglect experiences in this population and “false positive” reports may also be present with a certain rate. The fact that this study was not conducted with patients who presented to the outpatient clinic of psychiatry for the first time and included patients without psychotic properties in whom psychotherapy was thought to be beneficial renders the findings more reliable. However, our study group represented only a certain adolescent group, since it was consisted of patients who were being treated with a certain diagnosis in the outpatient clinic of psychiatry. Large-scale studies comparing healthy and patient populations will present healthier data in this area.

Conclusion

In this study, deliberate self-harm behavior was found in one of two adolescents who were being treated in the outpatient clinic of psychiatry and who had no psychotic disorder. A strong relation was found between deliberate delf-harm behavior and emotional neglect and sexual abuse in the childhood.

These results obtained show that adolescents who present to psychiatric outpatient clinics for adolescents and young adults should be addressed in terms of self-harm behavior and traumatic experiences in addition to their present psychiatric diseases, should be evaluated in a way to cover other psychiatric diseases which may occur as a result of psychological trauma and treatment approaches should be planned accordingly.