Suicidal behaviors in adolescents
http://www.100md.com
《美国医学杂志》
Department of Psychiatry, G. B. Pant Hospital, New Delhi, India
Abstract
Objective . Suicidal ideation and suicide attempt are generally considered as non-fatal suicidal behaviors (NFSB) by most researchers. Across different cultures, the prevalence of NFSB has been found to be alarmingly high among adolescents. As there is no published study estimating prevalence rates in India, we conducted a study on adolescents in Delhi to find prevalence of NFSB and other related behaviors and to identify risk factors for NFSB. Methods : We collected data from 1205 adolescents in the age group from 12 to 19 years from 2 schools through semi-structured questionnaire on demographic variables, NFSB, death wish, deliberate self harm (DSH), Adjustment Inventory for School Students (AISS) by Sinha et al. & BDI. Results : Prevalence of suicidal ideation (lifetime), suicidal ideation (last year), suicide attempt (lifetime), suicide attempt (last year) were 21.7%, 11.7%, 8% and 3.5%, respectively. All the significant variables were entered into a logistic regression analysis model, and the adjusted odds ratios, with 95% confidence intervals, were obtained for them. Hindu religion, female sex, older adolescent, physical abuse by parents, feeling neglected by parents, history of running away from school, history of suicide by a friend, death wish and DSH were found to be significant risk factors for NFSB. Conclusions : There is high prevalence of suicidal ideation, suicide attempt, death wish and deliberate self- harm in adolescent population of two schools in Delhi. Clinical implications of these findings are discussed.
Keywords: Suicidal ideation; Suicide attempt; Adolescents
Suicidal behaviors represent a spectrum, ranging from suicidal ideation, to suicidal plan, to suicide attempt, to completed suicide. Non-fatal suicidal behaviors (NFSB) include all these behaviors except the completed suicide. Attempted suicide, as conceptualized currently, is a potentially self injurious action with a non-fatal outcome for which there is evidence, either explicit or implicit, that the individual intended to kill himself or herself.[1] The most frequently endorsed motives for self-harm, reported by attempters, were to die, to escape, and to obtain relief.[2] As prospective studies have emphasized the high subsequent suicide rates in clinically presenting suicide attempters[3], the evidence, linking NFSB to later completed suicide, has brought a particular focus on the study of NFSB. Across different cultures, the prevalence of NFSB has been found to be alarmingly high among adolescents. Lifetime suicidal ideation rates have ranged from 20 to 54%, and lifetime suicide attempt rates mostly lie between 7 and 10%, with some studies showing higher rates of up to 15%.[4] One-year prevalence rates for suicide attempts have ranged from 1.5 to 3%. Some studies have shown higher rates of up to 9%.[4] Suicidal ideation rates for one-year have shown a wide range from 11 to 62%. The estimation of prevalence of NFSB in adolescents in India has not been reported so far. In India, several risk factors for NFSB in adolescents have been reported which have not been reported in Western studies. Studies reported failure in examination, anticipated punishment, social conflict, neuro-psychiatric condition, physical illness, family dysfunction, and impending loss of love object as risk factors.[5],[6],[7],[8]
Materials and methods
We approached several schools in the central Delhi where our institution is located. Only two schools- one private (school 1) and another government-aided (school 2) gave permission for the study. So, no random sampling for the schools was conducted. Students studying in classes 7th to 12th and aged 12 to 19 years of age formed the universe of the study. Distribution of our sample is depicted in
Tools
Semistrutured questionnaire: Demographic data and clinical variables related to NFSB were listed based on clinical experience of the consultant psychiatrist (SJ) providing adolescent mental health service at the institute. Then a semistructured questionnaire, with 2 sections, was prepared. This is available with the corresponding author (SJ) on request.
Section-1: It has 58 questions. It covered most of the demographic variables.
Section-2: It has 27 questions on NFSB and other related behaviors.
Adjustment Inventory for School Students (AISS) by Sinha et al.[9] : It has 60 questions, which require a "yes" or "no" response from the students. The inventory measures the adjustment of the students in three areas- emotional, social and educational. Coefficient of reliability, as determined by the test re-test method, has been reported as 0.93 by the authors.
Beck Depression Inventory (BDI): This is a well-established self-assessment scale for depression.[9]a Written permission was obtained from the school authorities to conduct the study prior to the data collection. A 90-minute slot (2 periods) was used to do the data collection. All students, 12 - 19 years of age of either sex, present on the day of the data collection and willing to participate, were included. The study was cross-sectional in design and anonymous in nature. All the sections from classes 8th to 12th were taken by turn to conduct the study. Prior to the distribution of the Pro forma the students were informed about the purpose of the study, and the anonymous nature of the study was explained and emphasized. The students were given the office address of the investigator (TS) and encouraged to contact him, if after filling up the Pro forma they felt that they needed professional help. The necessary instructions required to fill up the Pro forma were given and then they were distributed to all students. The same procedure was repeated in each section.
Statistical analysis
One thousand three hundred and eighty-four students formed the population of the study and 1205 completed the Pro forma. Statistical Package for Social Sciences (SPSS) Version 9.0 was used to perform the statistical analysis. The lifetime and one year prevalence rates for NFSB were calculated using items in Section-2 of the Performa. Those having a lifetime history of suicidal ideation (n =261) with or without history of suicide attempts (n = 97) were grouped together to form one group. This was done because all attempters were also ideators. This group was called the NFSB group. The group comprised of a total of 261 students. This group was compared to the rest of the students who did not report NFSB on all variables. NFSB was taken as the dependent variable and all other variables were taken as independent variables. Chi-square values, 2-sided significance and Odds ratios (OR) with 95% confidence intervals were calculated to estimate the differences in the two groups for each of the variables. The variables for which a significant difference (two sided significance less than .05 and confidence intervals for odds ratio not including the value 1) was found between the two groups were entered into a Logistic Regression Analysis model. This method allowed us to assess the associations of these variables with NFSB after adjusting for other variables found significant on univariate analysis. The variables, which continued to show an independent significant association with the presence of NFSB after controlling in the rest of the variables, were identified and presented separately.
Results
Students were uniformly distributed in all the classes except for class 7, which contributed only 3.8 % of the total sample. This was because most of the students in 7th class were under 12 years of age and overall strength of students in this class was lower than other classes in both the schools. The mean age of the sample was 14.73 (S.D.-1.44). Younger adolescents (12-14 years) were 545 (45.2%), while older adolescents (15-18 years) were 660 (54.8%). 82.9 % of the sample fell between the age groups of 13 to 16 years. 84.6% of the sample were Hindus, 8.0% were Muslims, 0.7% were Christians, 5.6% were Jains and 1.1% were Sikhs. Males formed approximately 60% of the sample, whereas females formed 40% of the sample. Majority (68.2%) of students were living in smaller (5 members or less) size families. The parents of most students had had high levels of education. Nine hundred forty-eight (78.6%) of fathers and 797 (66.1%) mothers were educated up to graduation or above. Regardless of these high levels of literacy, 75% of mothers were housewives. Approximately 5% of the adolescents reported that their parents were separated or divorced while 4.1% of them reported that either or both of their parents were not alive. Only 2% of the students reported that they were not living with their parents.
As all life time suicide attempters were also life time suicidal ideators, NFSB was considered if a student was an ideator with or without being an attempter. Total number of students with NFSB was 261. The lifetime prevalence of suicidal ideation was higher in females (25.4% vs 19.1%) and lifetime prevalence of suicide attempt was also higher in females (11% vs 6.1%), while the prevalence of deliberate self-harm (DSH) was found to be equal in both sexes. Lifetime prevalence of suicidal ideation was higher (24.2% vs 17.8%) among the older adolescents and so also the lifetime prevalence of suicide attempt (8.6% vs 7.3%) among the older adolescents, while lifetime prevalence of DSH was higher in the younger adolescents (19.1% vs 17.1%). Of the 97 respondents who reported having made a suicide attempt, 25.8% of them had done so by consuming poisons, 26.8% had done so by taking an overdose of medicines. 17.5% said that they had attempted suicide by cutting themselves, 7.2% and 2.1% had attempted suicide by hanging and jumping from a height, respectively. 50.6% had attempted suicide after thinking about the act for a long time and 42.3% did so without giving it much thought. Seven suicide attempters did not answer this question. 20.6% of the suicide attempters said that they were intoxicated at the time of the suicide attempt. A substantial number (75.3%) of attempters did not seek medical care after the attempt. 33% of the suicide attempters reported that they had done something to prevent rescue by others.
School adjustment was assessed using AISS. table4 shows types of school adjustment and their relationship with NFSB.
"Excellent/Good" & "Average" categories were merged in to "Good adjustment" and "Unsatisfactory" & "Very unsatisfactory" categories were grouped in to "Poor adjustment" for further analysis.
Depression was assessed by BDI. Students with BDI score of 16 or more were labeled as depressed. They were 508 (42.2%) in number. As depression is a very powerful risk factor for NFSB, inclusion of it could have eliminated a number of relatively less powerful risk factors. So, depression, as a variable, was not included in the univariate analysis and the logistic regression analysis. Data were collected on forty variables, and after the univariate analysis, 23 variables were found to be statistically significant. All the 23 significant variables were entered into a logistic regression analysis model, and the adjusted odds ratios (OR), with 95% confidence intervals (CI), were obtained for them. The adjusted odds ratios were the independent OR for each variable, controlling for all other variables entered into the logistic regression analysis. The variables, which continued to have OR greater than 1, with 95% CI not including the value 1, were taken to be the variables which showed the strongest and the most direct association with NFSB. There were 9 such variables table5.
Discussion
Study design: Our study was cross-sectional in design, and information was collected retrospectively using a self-report anonymous Pro forma. The reliability of such anonymous self-report method of data collection about high-risk behavior in the youth in the US: Youth Risk Behavior Survey (YRBS) has been tested and established.[10] Two schools, a government aided and another private, were chosen to get a fair representation of adolescents from poor, low middle class, high middle class and rich sections of the society. 90.74% (1205) of the total students (1328) of the two schools were included in the study, as mentioned in table1. This rate is better than most of the published studies on similar type of populations. [4],[11]
Prevalence of NFSB and other related behaviors: The lifetime prevalence of suicidal ideation (SI) was 21.7% and one-year prevalence of SI was 11.7%. Choquet et al[12] reported the lifetime prevalence of SI in a community sample of adolescents in France to be 14% for boys and 23% for girls. The one-year prevalence of SI in YRBS-2003 was 16.9%.[13] The lifetime prevalence of suicide attempt (SA) was 8% and one-year prevalence of SA was 3.5%. The lifetime prevalence rates for SA have been found to be between 2.2% and 8.8%. [11],[14] Our prevalence of 8% falls within this range. The one-year prevalence of SA in the YRBS study in 2003 was 8.5%.[13] Borowsky et al[13] reported one-year prevalence of SA as 5.1% for girls and 2% for boys. High rates of NFSB in the US (YRBS study) could reflect the breaking of families resulting in frequent change in parental figure and great number of single parent families. The lifetime prevalence of death wish was 47.2%. Paykel et al[16] had found a prevalence rate of death wish as 8.2%. Most of the researchers working on suicidal behavior have not studied death wish probably because it is considered to be a symptom of depression. For this reason we have not included death wish as a form of NFSB. We found the lifetime and one-year prevalence of DSH were 18% and 6.1%, respectively. Patton et al[17] reported one-year prevalence of DSH as 5.1% in 15-16 year old school students. It is known that SA and DSH are two overlapping behaviors. In our study, predominant method employed for DSH was self-injury (52%), where as the predominant methods employed by suicide attempters were self-poisoning and overdose of medicines (52.65%). Most researchers understand DSH as suicidal behavior with or without intent to end life, while other researchers identify DSH as behavior different from SA. In our study, we have not included DSH as a type of NFSB. 50.6% of the suicide attempters contemplated suicide for several hours to more than a day before attempting it, while 42.3% of the attempters reported that they only thought of attempting it for a few moments to few minutes. These two groups of attempters may indicate two different groups- one group could be depressed and other group could be impulsive. 30% of attempters informed that they had done something to prevent rescue by others, indicating high intent of suicide in them. In our study, we found that 20% of the attempters were intoxicated at the time of the attempt. Wichstrom found 25% were intoxicated at the time of the attempt. Alcohol and drug use may increase likelihood of impulsive or violent act; intoxication may also be instrumental in gaining courage or to avoid pain. Also, intoxication may increase the occurrence of precipitating events, such as break up in relationship or quarrel with a friend or parents.
Risk factors: Many studies on risk factors for NFSB used univariate analysis, which provides spurious association between them. A multivariate analysis like, logistic regression analysis, done in our study, eliminates false associations as one risk factor is tested controlling the effects of other risk factors.
Demographic risk factors: In the present study, older age group (15-18 years), female sex and Hindu religion were found to be significant risk factors for NFSB. Guyer[18] observed that older adolescents were at greater risk for NFSB than younger ones in a univariate analysis, but did not reach significant level in multivariate analysis. Kienhorst et al[11] and Grossman et al[4] found female sex to be significant for NFSB, even after controlling for other demographic and psychosocial variables. One study in India by Kumar et al[19] found males outnumbered females as far as attempted suicide was concerned. Singh et al[20] found Sikh community to be over-represented in completed suicides in general population. Gupta et al[21], in a study of suicide attempters reaching the hospital, had found that 75% of the attempters belonged to the Hindu community.
Family environment: We found physical abuse by parents and feeling neglected by parents to be significant risk factors for NFSB. Dubow et al[22] found marital discord, use of physical discipline by parents to be significant associates of NFSB. Kienhorst et al[11], in their cross-sectional survey of 9393 Dutch adolescents, had found poor perceived relationship with the parents to be a significant risk factor.
School environment: We found history of running away from the school and history of suicide by a friend to be important risk factors. Dubow et al (1989) had found poor school grades to be significantly associated with NFSB. Grossman et al (1991) found that a suicide attempt or suicide by a friend was significant risk factor. Choquet et al. (1989) found that the presence of conduct problems and delinquent behavior to be significantly related to suicidal ideation.
Death wish and DSH: We found death wish and DSH to be significant risk factors for suicidal ideation and suicide attempt. Death wish has been considered as a symptom of depression, so it has not been studied as a risk factor for NFSB. As we have not assessed depression, we cannot say whether death wish was part of depression or not. DSH, as different from suicide attempt, has been a controversial issue for most suicide researchers, as many believe that later is a sub-group of the former. We separated both the groups on the basis of intent to end life, and found DSH to be a significant risk factor for NFSB.
Limitations of the study
We did not conduct random sampling to choose the two schools. There might have been unknown bias while choosing them.
Few students might have given wrong information about certain past events due to problem in recall.
Our study being cross-sectional in nature, a temporal relationship between certain risk factors and NFSB may not have been possible in certain cases.
The results and conclusion of our study cannot be generalized to whole adolescent population in Delhi or India.
Conclusion
There is high prevalence of suicidal ideation, suicide attempt, death wish and deliberate self- harm in adolescent population in two schools in Delhi. Hindu religion, female sex, older adolescent, physical abuse by parents, feeling neglected by parents, history of running away from school, history of suicide by a friend, death wish and deliberate self harm were found to be significant risk factors for NFSB.
Clinical implications
While dealing with adolescents, a health professional should look into risk factors for NFSB, like physical abuse or neglect by parents, conduct problems, death wish and history of DSH in them. Adolescents harboring death wish and DSH should be assessed thoroughly by mental health professionals and managed for underlying mental disorder, like depression. Appropriate steps should be taken to prevent further abuse or neglect of adolescents by their parents. Conduct problems need to be treated by mental health professionals. Teachers, school health professionals, parents, pediatricians need to be sensitized about how to identify NFSB and underlying risk factors in adolescents, and to take appropriate steps to deal with them. By doing so, we can lower the rate of adolescent suicide in our society.
However, a long-term prospective study, using larger, representative sample, is required to find the true prevalence of NFSB in adolescents and their risk factors, in India.[23]
References
1. Moscicki EK. Identification of suicide risk factors using epidemiological studies. Psychiatr Clin North Am 1997; 20: 499-517.
2. Boergers J, Spirito A, Dolandson D. Reasons for suicide attempts: associations with psychological functioning. J Am Acad Child Adolesc Psychiatry 1988; 37(12) : 1287-1293.
3. Hawton K, Fagg J, Platt S. Factors associated with suicide after parasuicide in young people. BMJ 1993; 306: 1641-1644.
4. Grossman DC, Milligan C, Deyo RA. Risk factors for suicide attempts among Navajo adolescents. Am J Public Health 1991; 81: 870-874.
5. Kar NM, Choudhury P. Characteristics of suicidal behaviour by adolescents. Paper presented at the 49th Annual National Conference of the Indian Psychiatric Society 1997.
6. Jayaramiah C, Gunde R, Reddy V. Attempted suicide among young. Paper presented at the 49th Annual National Conference of the Indian Psychiatric Society 1999.
7. Kar NM, Swain AK, Naik, M.S, Dash PS. Suicide and attempted suicide in children and adolescents - some observations. Paper presented at the 52nd Annual National Conference of the Indian Psychiatric Society 2000.
8. Kar NM, Pursty GS, Mohapatra BN, Swain AK, Pattnaik P. Suicide attempts in adolescents- affective and cognitive predictors of precipitation. Paper presented at the 54th Annual National Conference of the Indian Psychiatric Society 20002.
9. Sinha AKP, Singh RP. The Adjustment Inventory for School Students (AISS). Agra. National Psychological Corporation 1993.
10. 9a. Whisman MA, Perez JE, Ramel W. Factor structure of the Beck Depression Inventory-Second Edition (BDI-II) in a student sample. J Clin Psychol 2000; 56 (4): 545-551.
11. Brenner ND, Collins J, Kann L. Reliability of the YRBS. Am J Epidemiol 1995; 141: 575-580.
12. Kienhorst CWM, De Wilde EJ, Van Den Bout J. Characteristics of suicide attempters in a population based sample of Dutch adolescents. Br J Psychiatry 1990; 156: 243-248.
13. Choquet M, Menke H. Suicidal thoughts during early adolescence: prevalence associated troubles and help seeking behaviour. Acta Psychiatr Scand 1989; 81 : 170-177.
14. Centre for Disease Control and Prevention. YRBSS: Youth Online: comprehensive results. [online] 2003. Available: http://apps.nccd.cdc.gov/yrbss/index.asp.
15. Wichstrom L. Predictors of adolescent suicide attempts: a nationally representative longitudinal study of Norwegian adolescents. J Am Acad Child Adolesc Psychiatry 2000; 39(5): 603-610.
16. Borowsky IW, Ireland M, Resnick MD. Adolescent suicide attempts: risks and protectors. Pediatrics 2001; 107(3): 486-493.
17. Paykel ES, Myers JK, Lindenthal JJ. Suicidal feelings in the general population: a prevalence study. Br J Psychiatry 1974; 124: 460-469.
18. Patton GC, Harris R, Carlin JB. Adolescent suicidal behaviors: a population based study of risk. Psychol Med 1997; 27: 715-724.
19. Guyer B. Intentional injuries among children and adolescents in Masachusetts. N Engl J Med 1989; 321: 1584.
20. Kumar CTS, Chandrasekaran R. A study of psychosocial and clinical factors associated with adolescent suicide attempts. Indian J Psychiatry 2000; 42(3): 237-242.
21. Singh K, Jain NR, Khullar BMR. A study of suicide in Delhi state. Indian J Psychiatry 1971; 57(11): 412-419.
22. Guptal SC, Singh H. Psychiatric illness in suicide attempters. Indian J Psychiatry 1981; 23 (1): 69-74.
23. Dubow EF, Kausch DF, Blum MC, Reed J. Correlates of suicidal ideation and attempts in a community sample of junior high school students. J Consult Clin Psychol 1989; 18: 158-166.(Sidhartha Tanuj, Jena Shi)
Abstract
Objective . Suicidal ideation and suicide attempt are generally considered as non-fatal suicidal behaviors (NFSB) by most researchers. Across different cultures, the prevalence of NFSB has been found to be alarmingly high among adolescents. As there is no published study estimating prevalence rates in India, we conducted a study on adolescents in Delhi to find prevalence of NFSB and other related behaviors and to identify risk factors for NFSB. Methods : We collected data from 1205 adolescents in the age group from 12 to 19 years from 2 schools through semi-structured questionnaire on demographic variables, NFSB, death wish, deliberate self harm (DSH), Adjustment Inventory for School Students (AISS) by Sinha et al. & BDI. Results : Prevalence of suicidal ideation (lifetime), suicidal ideation (last year), suicide attempt (lifetime), suicide attempt (last year) were 21.7%, 11.7%, 8% and 3.5%, respectively. All the significant variables were entered into a logistic regression analysis model, and the adjusted odds ratios, with 95% confidence intervals, were obtained for them. Hindu religion, female sex, older adolescent, physical abuse by parents, feeling neglected by parents, history of running away from school, history of suicide by a friend, death wish and DSH were found to be significant risk factors for NFSB. Conclusions : There is high prevalence of suicidal ideation, suicide attempt, death wish and deliberate self- harm in adolescent population of two schools in Delhi. Clinical implications of these findings are discussed.
Keywords: Suicidal ideation; Suicide attempt; Adolescents
Suicidal behaviors represent a spectrum, ranging from suicidal ideation, to suicidal plan, to suicide attempt, to completed suicide. Non-fatal suicidal behaviors (NFSB) include all these behaviors except the completed suicide. Attempted suicide, as conceptualized currently, is a potentially self injurious action with a non-fatal outcome for which there is evidence, either explicit or implicit, that the individual intended to kill himself or herself.[1] The most frequently endorsed motives for self-harm, reported by attempters, were to die, to escape, and to obtain relief.[2] As prospective studies have emphasized the high subsequent suicide rates in clinically presenting suicide attempters[3], the evidence, linking NFSB to later completed suicide, has brought a particular focus on the study of NFSB. Across different cultures, the prevalence of NFSB has been found to be alarmingly high among adolescents. Lifetime suicidal ideation rates have ranged from 20 to 54%, and lifetime suicide attempt rates mostly lie between 7 and 10%, with some studies showing higher rates of up to 15%.[4] One-year prevalence rates for suicide attempts have ranged from 1.5 to 3%. Some studies have shown higher rates of up to 9%.[4] Suicidal ideation rates for one-year have shown a wide range from 11 to 62%. The estimation of prevalence of NFSB in adolescents in India has not been reported so far. In India, several risk factors for NFSB in adolescents have been reported which have not been reported in Western studies. Studies reported failure in examination, anticipated punishment, social conflict, neuro-psychiatric condition, physical illness, family dysfunction, and impending loss of love object as risk factors.[5],[6],[7],[8]
Materials and methods
We approached several schools in the central Delhi where our institution is located. Only two schools- one private (school 1) and another government-aided (school 2) gave permission for the study. So, no random sampling for the schools was conducted. Students studying in classes 7th to 12th and aged 12 to 19 years of age formed the universe of the study. Distribution of our sample is depicted in
Tools
Semistrutured questionnaire: Demographic data and clinical variables related to NFSB were listed based on clinical experience of the consultant psychiatrist (SJ) providing adolescent mental health service at the institute. Then a semistructured questionnaire, with 2 sections, was prepared. This is available with the corresponding author (SJ) on request.
Section-1: It has 58 questions. It covered most of the demographic variables.
Section-2: It has 27 questions on NFSB and other related behaviors.
Adjustment Inventory for School Students (AISS) by Sinha et al.[9] : It has 60 questions, which require a "yes" or "no" response from the students. The inventory measures the adjustment of the students in three areas- emotional, social and educational. Coefficient of reliability, as determined by the test re-test method, has been reported as 0.93 by the authors.
Beck Depression Inventory (BDI): This is a well-established self-assessment scale for depression.[9]a Written permission was obtained from the school authorities to conduct the study prior to the data collection. A 90-minute slot (2 periods) was used to do the data collection. All students, 12 - 19 years of age of either sex, present on the day of the data collection and willing to participate, were included. The study was cross-sectional in design and anonymous in nature. All the sections from classes 8th to 12th were taken by turn to conduct the study. Prior to the distribution of the Pro forma the students were informed about the purpose of the study, and the anonymous nature of the study was explained and emphasized. The students were given the office address of the investigator (TS) and encouraged to contact him, if after filling up the Pro forma they felt that they needed professional help. The necessary instructions required to fill up the Pro forma were given and then they were distributed to all students. The same procedure was repeated in each section.
Statistical analysis
One thousand three hundred and eighty-four students formed the population of the study and 1205 completed the Pro forma. Statistical Package for Social Sciences (SPSS) Version 9.0 was used to perform the statistical analysis. The lifetime and one year prevalence rates for NFSB were calculated using items in Section-2 of the Performa. Those having a lifetime history of suicidal ideation (n =261) with or without history of suicide attempts (n = 97) were grouped together to form one group. This was done because all attempters were also ideators. This group was called the NFSB group. The group comprised of a total of 261 students. This group was compared to the rest of the students who did not report NFSB on all variables. NFSB was taken as the dependent variable and all other variables were taken as independent variables. Chi-square values, 2-sided significance and Odds ratios (OR) with 95% confidence intervals were calculated to estimate the differences in the two groups for each of the variables. The variables for which a significant difference (two sided significance less than .05 and confidence intervals for odds ratio not including the value 1) was found between the two groups were entered into a Logistic Regression Analysis model. This method allowed us to assess the associations of these variables with NFSB after adjusting for other variables found significant on univariate analysis. The variables, which continued to show an independent significant association with the presence of NFSB after controlling in the rest of the variables, were identified and presented separately.
Results
Students were uniformly distributed in all the classes except for class 7, which contributed only 3.8 % of the total sample. This was because most of the students in 7th class were under 12 years of age and overall strength of students in this class was lower than other classes in both the schools. The mean age of the sample was 14.73 (S.D.-1.44). Younger adolescents (12-14 years) were 545 (45.2%), while older adolescents (15-18 years) were 660 (54.8%). 82.9 % of the sample fell between the age groups of 13 to 16 years. 84.6% of the sample were Hindus, 8.0% were Muslims, 0.7% were Christians, 5.6% were Jains and 1.1% were Sikhs. Males formed approximately 60% of the sample, whereas females formed 40% of the sample. Majority (68.2%) of students were living in smaller (5 members or less) size families. The parents of most students had had high levels of education. Nine hundred forty-eight (78.6%) of fathers and 797 (66.1%) mothers were educated up to graduation or above. Regardless of these high levels of literacy, 75% of mothers were housewives. Approximately 5% of the adolescents reported that their parents were separated or divorced while 4.1% of them reported that either or both of their parents were not alive. Only 2% of the students reported that they were not living with their parents.
As all life time suicide attempters were also life time suicidal ideators, NFSB was considered if a student was an ideator with or without being an attempter. Total number of students with NFSB was 261. The lifetime prevalence of suicidal ideation was higher in females (25.4% vs 19.1%) and lifetime prevalence of suicide attempt was also higher in females (11% vs 6.1%), while the prevalence of deliberate self-harm (DSH) was found to be equal in both sexes. Lifetime prevalence of suicidal ideation was higher (24.2% vs 17.8%) among the older adolescents and so also the lifetime prevalence of suicide attempt (8.6% vs 7.3%) among the older adolescents, while lifetime prevalence of DSH was higher in the younger adolescents (19.1% vs 17.1%). Of the 97 respondents who reported having made a suicide attempt, 25.8% of them had done so by consuming poisons, 26.8% had done so by taking an overdose of medicines. 17.5% said that they had attempted suicide by cutting themselves, 7.2% and 2.1% had attempted suicide by hanging and jumping from a height, respectively. 50.6% had attempted suicide after thinking about the act for a long time and 42.3% did so without giving it much thought. Seven suicide attempters did not answer this question. 20.6% of the suicide attempters said that they were intoxicated at the time of the suicide attempt. A substantial number (75.3%) of attempters did not seek medical care after the attempt. 33% of the suicide attempters reported that they had done something to prevent rescue by others.
School adjustment was assessed using AISS. table4 shows types of school adjustment and their relationship with NFSB.
"Excellent/Good" & "Average" categories were merged in to "Good adjustment" and "Unsatisfactory" & "Very unsatisfactory" categories were grouped in to "Poor adjustment" for further analysis.
Depression was assessed by BDI. Students with BDI score of 16 or more were labeled as depressed. They were 508 (42.2%) in number. As depression is a very powerful risk factor for NFSB, inclusion of it could have eliminated a number of relatively less powerful risk factors. So, depression, as a variable, was not included in the univariate analysis and the logistic regression analysis. Data were collected on forty variables, and after the univariate analysis, 23 variables were found to be statistically significant. All the 23 significant variables were entered into a logistic regression analysis model, and the adjusted odds ratios (OR), with 95% confidence intervals (CI), were obtained for them. The adjusted odds ratios were the independent OR for each variable, controlling for all other variables entered into the logistic regression analysis. The variables, which continued to have OR greater than 1, with 95% CI not including the value 1, were taken to be the variables which showed the strongest and the most direct association with NFSB. There were 9 such variables table5.
Discussion
Study design: Our study was cross-sectional in design, and information was collected retrospectively using a self-report anonymous Pro forma. The reliability of such anonymous self-report method of data collection about high-risk behavior in the youth in the US: Youth Risk Behavior Survey (YRBS) has been tested and established.[10] Two schools, a government aided and another private, were chosen to get a fair representation of adolescents from poor, low middle class, high middle class and rich sections of the society. 90.74% (1205) of the total students (1328) of the two schools were included in the study, as mentioned in table1. This rate is better than most of the published studies on similar type of populations. [4],[11]
Prevalence of NFSB and other related behaviors: The lifetime prevalence of suicidal ideation (SI) was 21.7% and one-year prevalence of SI was 11.7%. Choquet et al[12] reported the lifetime prevalence of SI in a community sample of adolescents in France to be 14% for boys and 23% for girls. The one-year prevalence of SI in YRBS-2003 was 16.9%.[13] The lifetime prevalence of suicide attempt (SA) was 8% and one-year prevalence of SA was 3.5%. The lifetime prevalence rates for SA have been found to be between 2.2% and 8.8%. [11],[14] Our prevalence of 8% falls within this range. The one-year prevalence of SA in the YRBS study in 2003 was 8.5%.[13] Borowsky et al[13] reported one-year prevalence of SA as 5.1% for girls and 2% for boys. High rates of NFSB in the US (YRBS study) could reflect the breaking of families resulting in frequent change in parental figure and great number of single parent families. The lifetime prevalence of death wish was 47.2%. Paykel et al[16] had found a prevalence rate of death wish as 8.2%. Most of the researchers working on suicidal behavior have not studied death wish probably because it is considered to be a symptom of depression. For this reason we have not included death wish as a form of NFSB. We found the lifetime and one-year prevalence of DSH were 18% and 6.1%, respectively. Patton et al[17] reported one-year prevalence of DSH as 5.1% in 15-16 year old school students. It is known that SA and DSH are two overlapping behaviors. In our study, predominant method employed for DSH was self-injury (52%), where as the predominant methods employed by suicide attempters were self-poisoning and overdose of medicines (52.65%). Most researchers understand DSH as suicidal behavior with or without intent to end life, while other researchers identify DSH as behavior different from SA. In our study, we have not included DSH as a type of NFSB. 50.6% of the suicide attempters contemplated suicide for several hours to more than a day before attempting it, while 42.3% of the attempters reported that they only thought of attempting it for a few moments to few minutes. These two groups of attempters may indicate two different groups- one group could be depressed and other group could be impulsive. 30% of attempters informed that they had done something to prevent rescue by others, indicating high intent of suicide in them. In our study, we found that 20% of the attempters were intoxicated at the time of the attempt. Wichstrom found 25% were intoxicated at the time of the attempt. Alcohol and drug use may increase likelihood of impulsive or violent act; intoxication may also be instrumental in gaining courage or to avoid pain. Also, intoxication may increase the occurrence of precipitating events, such as break up in relationship or quarrel with a friend or parents.
Risk factors: Many studies on risk factors for NFSB used univariate analysis, which provides spurious association between them. A multivariate analysis like, logistic regression analysis, done in our study, eliminates false associations as one risk factor is tested controlling the effects of other risk factors.
Demographic risk factors: In the present study, older age group (15-18 years), female sex and Hindu religion were found to be significant risk factors for NFSB. Guyer[18] observed that older adolescents were at greater risk for NFSB than younger ones in a univariate analysis, but did not reach significant level in multivariate analysis. Kienhorst et al[11] and Grossman et al[4] found female sex to be significant for NFSB, even after controlling for other demographic and psychosocial variables. One study in India by Kumar et al[19] found males outnumbered females as far as attempted suicide was concerned. Singh et al[20] found Sikh community to be over-represented in completed suicides in general population. Gupta et al[21], in a study of suicide attempters reaching the hospital, had found that 75% of the attempters belonged to the Hindu community.
Family environment: We found physical abuse by parents and feeling neglected by parents to be significant risk factors for NFSB. Dubow et al[22] found marital discord, use of physical discipline by parents to be significant associates of NFSB. Kienhorst et al[11], in their cross-sectional survey of 9393 Dutch adolescents, had found poor perceived relationship with the parents to be a significant risk factor.
School environment: We found history of running away from the school and history of suicide by a friend to be important risk factors. Dubow et al (1989) had found poor school grades to be significantly associated with NFSB. Grossman et al (1991) found that a suicide attempt or suicide by a friend was significant risk factor. Choquet et al. (1989) found that the presence of conduct problems and delinquent behavior to be significantly related to suicidal ideation.
Death wish and DSH: We found death wish and DSH to be significant risk factors for suicidal ideation and suicide attempt. Death wish has been considered as a symptom of depression, so it has not been studied as a risk factor for NFSB. As we have not assessed depression, we cannot say whether death wish was part of depression or not. DSH, as different from suicide attempt, has been a controversial issue for most suicide researchers, as many believe that later is a sub-group of the former. We separated both the groups on the basis of intent to end life, and found DSH to be a significant risk factor for NFSB.
Limitations of the study
We did not conduct random sampling to choose the two schools. There might have been unknown bias while choosing them.
Few students might have given wrong information about certain past events due to problem in recall.
Our study being cross-sectional in nature, a temporal relationship between certain risk factors and NFSB may not have been possible in certain cases.
The results and conclusion of our study cannot be generalized to whole adolescent population in Delhi or India.
Conclusion
There is high prevalence of suicidal ideation, suicide attempt, death wish and deliberate self- harm in adolescent population in two schools in Delhi. Hindu religion, female sex, older adolescent, physical abuse by parents, feeling neglected by parents, history of running away from school, history of suicide by a friend, death wish and deliberate self harm were found to be significant risk factors for NFSB.
Clinical implications
While dealing with adolescents, a health professional should look into risk factors for NFSB, like physical abuse or neglect by parents, conduct problems, death wish and history of DSH in them. Adolescents harboring death wish and DSH should be assessed thoroughly by mental health professionals and managed for underlying mental disorder, like depression. Appropriate steps should be taken to prevent further abuse or neglect of adolescents by their parents. Conduct problems need to be treated by mental health professionals. Teachers, school health professionals, parents, pediatricians need to be sensitized about how to identify NFSB and underlying risk factors in adolescents, and to take appropriate steps to deal with them. By doing so, we can lower the rate of adolescent suicide in our society.
However, a long-term prospective study, using larger, representative sample, is required to find the true prevalence of NFSB in adolescents and their risk factors, in India.[23]
References
1. Moscicki EK. Identification of suicide risk factors using epidemiological studies. Psychiatr Clin North Am 1997; 20: 499-517.
2. Boergers J, Spirito A, Dolandson D. Reasons for suicide attempts: associations with psychological functioning. J Am Acad Child Adolesc Psychiatry 1988; 37(12) : 1287-1293.
3. Hawton K, Fagg J, Platt S. Factors associated with suicide after parasuicide in young people. BMJ 1993; 306: 1641-1644.
4. Grossman DC, Milligan C, Deyo RA. Risk factors for suicide attempts among Navajo adolescents. Am J Public Health 1991; 81: 870-874.
5. Kar NM, Choudhury P. Characteristics of suicidal behaviour by adolescents. Paper presented at the 49th Annual National Conference of the Indian Psychiatric Society 1997.
6. Jayaramiah C, Gunde R, Reddy V. Attempted suicide among young. Paper presented at the 49th Annual National Conference of the Indian Psychiatric Society 1999.
7. Kar NM, Swain AK, Naik, M.S, Dash PS. Suicide and attempted suicide in children and adolescents - some observations. Paper presented at the 52nd Annual National Conference of the Indian Psychiatric Society 2000.
8. Kar NM, Pursty GS, Mohapatra BN, Swain AK, Pattnaik P. Suicide attempts in adolescents- affective and cognitive predictors of precipitation. Paper presented at the 54th Annual National Conference of the Indian Psychiatric Society 20002.
9. Sinha AKP, Singh RP. The Adjustment Inventory for School Students (AISS). Agra. National Psychological Corporation 1993.
10. 9a. Whisman MA, Perez JE, Ramel W. Factor structure of the Beck Depression Inventory-Second Edition (BDI-II) in a student sample. J Clin Psychol 2000; 56 (4): 545-551.
11. Brenner ND, Collins J, Kann L. Reliability of the YRBS. Am J Epidemiol 1995; 141: 575-580.
12. Kienhorst CWM, De Wilde EJ, Van Den Bout J. Characteristics of suicide attempters in a population based sample of Dutch adolescents. Br J Psychiatry 1990; 156: 243-248.
13. Choquet M, Menke H. Suicidal thoughts during early adolescence: prevalence associated troubles and help seeking behaviour. Acta Psychiatr Scand 1989; 81 : 170-177.
14. Centre for Disease Control and Prevention. YRBSS: Youth Online: comprehensive results. [online] 2003. Available: http://apps.nccd.cdc.gov/yrbss/index.asp.
15. Wichstrom L. Predictors of adolescent suicide attempts: a nationally representative longitudinal study of Norwegian adolescents. J Am Acad Child Adolesc Psychiatry 2000; 39(5): 603-610.
16. Borowsky IW, Ireland M, Resnick MD. Adolescent suicide attempts: risks and protectors. Pediatrics 2001; 107(3): 486-493.
17. Paykel ES, Myers JK, Lindenthal JJ. Suicidal feelings in the general population: a prevalence study. Br J Psychiatry 1974; 124: 460-469.
18. Patton GC, Harris R, Carlin JB. Adolescent suicidal behaviors: a population based study of risk. Psychol Med 1997; 27: 715-724.
19. Guyer B. Intentional injuries among children and adolescents in Masachusetts. N Engl J Med 1989; 321: 1584.
20. Kumar CTS, Chandrasekaran R. A study of psychosocial and clinical factors associated with adolescent suicide attempts. Indian J Psychiatry 2000; 42(3): 237-242.
21. Singh K, Jain NR, Khullar BMR. A study of suicide in Delhi state. Indian J Psychiatry 1971; 57(11): 412-419.
22. Guptal SC, Singh H. Psychiatric illness in suicide attempters. Indian J Psychiatry 1981; 23 (1): 69-74.
23. Dubow EF, Kausch DF, Blum MC, Reed J. Correlates of suicidal ideation and attempts in a community sample of junior high school students. J Consult Clin Psychol 1989; 18: 158-166.(Sidhartha Tanuj, Jena Shi)