Garnefski, N & Diekstra, RR.
Department of Clinical and Health Psychology,
University of Leiden,
Abstract Introduction Method Results Conclusions References
Keywords: Adolescents, comorbidity, co-occurrence, suicidal-behaviour
Accepted 28 March 1995
The co-occurrence of addiction-risk behaviour, aggressive/criminal behaviour, emotional problems and cognitive problems was investigated in a population-based sample of 477 secondary school students with a history of suicidal behaviour and a similar matched sample of 477 students without such a history. More non-attempters than attempters had problems in a `single' problem category, whereas more attempters than non-attempters had problems in `multiple' categories. The number or accumulation of emotional, behavioural and/or cognitive problems seems of more importance in relation to a history of suicidal behaviour than the specific type of problem. Within the subgroup of suicide attempters, the co-occurrence of symptoms of different disorders was rather the rule than the exception, which raises doubts upon the validity of the prevailing diagnostic and classification systems, such as DSM or ICD.
Suicidal behaviour in youngsters of secondary school age in the form of one or more non-fatal suicide attempts has received a great deal of attention recently. Estimates of the life-time prevalence vary from 2% to 11% in this population (Andrews Lewinsohn, 1990; Diekstra et al., 1991; Garnefski et al., 1992; Harkavy-Friedman et al., 1987; Kienhorst et al., 1989; Pronovost et al., 1990; Smith et al., 1986), while estimates of annual prevalence range from 7% to as high as 20% (CDC, 1991; Dubow et al., 1989; Nagy et al., 1990; Pronovost et al., 1990; Rubenstein et al., 1989). Reasons for the variation in estimates between the studies include differences in definitions, operationalizations, research methods and sample characteristics.
Parasuicide is a potentially lethal health event and a risk for completed suicide in the future. Therefore, even the lowest percentages reported in the literature are alarming enough to warrant obtaining in-depth information about behavioural, emotional and cognitive risk factors.
The majority of studies on suicidal behaviour in secondary school students have focused on the identification of characteristics that differentiate between suicidal youngsters and non-suicidal youngsters, who are usually defined as those with a history of suicidal behaviour or ideation and those without. Associations between suicidal behaviour and cognitive problems and/or specific types of emotional problem, such as depressed mood and low self-esteem in secondary school students have been clearly established in studies which investigated these fields (Clark, 1993; Lewinsohn et al., 1993; Smith, 1990; Spirito et al., 1989; Pronovost et al., 1990; Smith, 1990; Smith Crawford, 1986; Garrison, 1989; Rubenstein et al., 1989; Spirito et al., 1989; Kienhorst et al., 1990). It has also been found that a variety of behavioural problems, such as substance abuse, aggressive or criminal behaviour, are related to suicidal behaviour (Andrews Lewinsohn, 1992; Dubow et al., 1989; Garnefski et al., 1992; Kienhorst et al., 1990; Lewinsohn et al., 1993; Smith, 1990; Spirito et al., 1989). Although there is agreement about the types of problem that differ\-entiate between youngsters who demonstrate suicidal behaviour and those who do not in most studies, in general the percentages of explained variance are low. One explanation for this might be the fact that for many years, studies on the relationship between emotional and/or behavioural problems and suicidal behaviour have largely been based on the so-called single problem approach, i.e., they have focused on only one specific problem at a time. Fairly\linebreak recently, studies on emotional and behavioural problems have presented findings which indicate that making a distinction between single problems should be considered artificial, both conceptually and statistically, because the co-occurrence or interrelatedness of such problems is more often the rule than the exception (Capaldi, 1991; Choquet Menke, 1987; Donovan Jessor, 1985; Donovan, Jessor Costa, 1988; Elliott, Huizinga Menard, 1989; Garnefski Diekstra, submitted; Grube Morgan, 1990; Ingersoll Orr, 1989; Irwin Vaughan, 1988; Jessor Jessor, 1977; Jessor, Donovan Costa, 1991; Magnusson Bergman, 1990; McGee Newcomb, 1992; Osgood, Johnston, O'Malley Bachman, 1988; Pritchard, Diamond, Fielding, Cox Choudry, 1987; Reinherz, Frost Pakiz, 1991; Verhulst, Achenbach, Althaus Akkerhuis, 1988; Vingilis Adlaf, 1990; White, 1992). Puig-Antich et al. (1989) in a study of prepubertal children with major depressive disorder concluded that the search for pure, i.e. noncomorbid forms of early affective disorders may be a futile undertaking since comorbidity possibly is an intrinsic characteristic of children with major affective illness. It could be argued that such a statement also applies to other psychiatric diagnosis in children and youngsters. Clearly, the issue of comorbidity represents a central theoretical concern for defining the phenomenology, i.e. symptomatology of mental disorders in youth.
Studies on differences between suicide attempters and non-attempters which have focused on comorbidity or the co-occurrence of problems, however, are noticeably absent in the literature. The aim of this study was therefore to compare the presence of sets of interrelated problems, i.e. the comorbidity or co-occurrence of emotional, behavioural and/or cognitive problems, in a population-based sample of secondary school students with and without a history of suicidal behaviour. Subsequently, the co-occurrence of problems was studied in more detail in the group of suicide attempters.
The most suitable method for investigating differences between attempters and non-attempters is the matched control design. Owing to the large size of the original population used in this study, we were able to match suicide attempters and non-attempters on relevant social and educational criteria (sex, age and academic level) and obtain two highly comparable samples.
The sample comprised 954 12 to 19-year-old (mean age 15) secondary school students: 477 with and 477 without a history of suicidal behaviour, defined as one or more previous self-reported suicide attempts.
These youngsters form a subset of an original sample which comprised 15,245 adolescents from 212 randomly selected secondary schools in the Netherlands, of whom 49.7% were boys and 50.1% were girls, varying in age from 11 to 23 years (M = 15 years and 5 months; SD = 1 year and 8 months). This sample appeared to be representative of the entire population of students receiving lower general secondary education, higher general secondary educa\-tion, pre-university education, lower vocational education and intermediate vocational education in the Netherlands. To obtain the most homogeneous sample possible within the framework of the present study, only the 12 to 19 year-old boys and girls of Dutch nationality who were not receiving intermediate vocational education, and who had a limited number of missing values on the relevant variables, were considered ( N = 9,491).
A total number of 514 (5.4%) youngsters in this selected subsample had reported at least one previous suicide attempt: 211 boys (4.6%) and 303 girls (6.1%). These percentages are clearly in agreement with the range of life-time prevalence rates reported in the literature (Diekstra et al., 1994). To compare suicide attempters and non-attempters, each student with a history of suicidal behaviour was matched with a student without such a history, using school and grade, type of education, age and sex as matching criteria. We succeeded in matching 200 male suicide attempters with 200 male non-attempters and 277 female suicide attempters with 277 female non-attempters.
All the subjects filled out an extensive self-report questionnaire in the period between October 1992 and January 1993 within the framework of a long-term project which involves three bodies: the National Institute for Budget Information. The Hague, Leiden University and the University of Rotterdam. The central aim of this project, called `Monitoring the Future: Behaviour and Health of Secondary School Students', is to assemble information on trends in lifestyles, attitudes, health status, emotional and behavioural characteristics on a regular, i.e. bi-annual basis, from the secondary school student population in the country. The questionnaire was filled out during normal school hours under the supervision of a teacher.
The questionnaire was constructed on the basis of the Monitoring-the-Future Questionnaire (Bachman et al., 1986). It covers a number of areas such as: physical and mental health, life-style, behavioural patterns, risk behaviours, attitudes towards various social and political topics, income, consumption pattern and leisure activities. This paper deals only with the data on emotional, behavioural and cognitive problems. Previous suicide attempts were assessed by the question: Have you ever made a serious attempt to end your life? with possible answer categories `never', `once' and `more than once'. The other variables which were used to operationalize the problems, are described in more detail below. First the name of the variable is given, followed by a short description of the operationalization and, in brackets, the literal text of the question.
Addiction risk behaviour
Frequent smoking : an average of more than 5 cigarettes a day during the past month (yes/no) (`How many cigarettes did you smoke during the past month on average a day?')
Frequent alcohol use : more than 20 glasses in the past month (yes/no) (`How many glasses of alcohol, such as beer, wine, spirits, did you drink during the past month?')
Frequent drunkenness : 2 or more incidents of drunkenness during the past month (yes/no) (`How many times did you get drunk because of the use of alcohol in the past month?')
Marijuana use : at least once during the past year (yes/no) (`Did you use soft drugs such as hash and marijuana at any time during the past year?')
Felony assault : one or more incidents during the past 12 months (yes/no) (`How many times in the past 12 months did you 1) participate in a serious fight, or 2) hit/hurt somebody in such a way that he or she had to be bandaged or had to see a doctor?')
Vandalism : one or more incidents during the past 12 months (yes/no) (`How many times in the past 12 months did you deliberately vandalize 1) school or 2) public property?')
Theft : one or more incidents during the past 12 months (yes/no) (`How many times in the past 12 months did you 1) go shoplifting or 2) steal a bike or 3) steal something else with a value of more than 50 guilders or 4) with a value of less than 50 guilders?')
Involvement with police : one or more incidents during the past 12 months (yes/no) (`How many times in the past 12 months were you involved with the police because you `had done something?')
Loneliness : agreement or disagreement with the statement `I often feel lonely'
Low self-esteem : agreement or disagreement with the statement `Generally speaking, I have a positive image of myself'
Anxiety : agreement or disagreement with the statement `I have a tendency to feel anxious when the things don't go as I had expected they would'
Depressed mood : agreement or disagreement with the statement `I am often in a depressed mood'
Concentration problems : twice or more during the past month (yes/no) (`Could you indicate how many times during the past 30 days you were bothered by memory or concentration problems?')
Learning difficulties : twice or more during the past month (yes/no) (`Could you indicate how many times during the past 30 days you had difficulty in mastering new material?')
Although most of the data were available in a continuous form, the items were used in a dichotomized form for three reasons. First, this enabled us to focus on the actual percentages of adolescents who had co-occurring behavioural and emotional problems, which was one of the aims of this study. Second, by dichotomizing uniformity of all the data was attained (some of the questions were originally posed in a dichotomized form (agree/disagree), while others were not). And third, the problem of statistical unmanage\-ability due to the strong skewness of many of the continuous variables was avoided.
Furthermore we chose to analyse single-item answers to be able to present rather straightforward patterns of co-occurrence. As a consequence a direct conclusive test of validity is hardly feasible. However it has been proven by many studies that simple direct questions tend to show high construct validity (for example, Andrews et al., 1976).
The classification of the items into the four separate categories is based on the results of Principal Component Analyses, which are described elsewhere (Garnefski Diekstra, submitted). Therefore, although the category `cognitive problems' could be both considered as a separate entity as well as a subcategory of emotional problems (e.g., concentration problems could both be seen as a symptom of depressed mood, but also as a cognitive problem on its own), it was treated as a separate category here.
Differences between specific problems reported by suicide-attempters and non-attempters All of the specific behavioural, emotional and cognitive problems were reported by more of the suicide attempters than non-attempters. A Bonferroni correction was applied because of the large number of bivariate tests conducted, but all the differences remained significant. In such a large sample, the value of evaluating significance levels alone is dubious, so we also calculated the percentage of variance shared by two variables by dividing the chi-square value by the sample size. The percentages of explained variance were low to moderate, ranging from 0.01 to 0.09. The table shows that although without exception the attempters had higher scores than the non-attempters, the non-attempters also reported emotional, behavioural and cognitive problems. The percentage of suicide attempters who reported specific problems ranged from 21.7% for frequent alcohol use to 43.7% for low self-esteem, whereas in the non-attempters the percentages ranged from 10% for marijuana use to 25.1% for low self-esteem.
Differences between specific problem categories in which one or more problems were reported by suicide-attempters and non-attempters Significantly higher percentages of suicide attempters reported one or more specific problems within the four categories than non-attempters. The percentages for suicide attempters ranged from 56.0% for addiction risk behaviour to 71.5% for emotional problems. Also here the percentages of variance explained are low to moderate, and the percentage of non-attempters reporting one or more problems within these categories rather high (ranging from 30.4 (addiction-risk) to 46.1 (emotional problems).
Differences between single and multiple problem categories in which one or more problems were reported by suicide-attempters and non-attempters
Irrespective of the content of a particular problem category in which one or more problems were reported, a) more non-attempters than attempters had only one problem in one of the categories or no problems at all, and b) more attempters than non-attempters reported problems in two or more problem categories.
The percentage of suicide attempters who reported problems in all four categories, which indicates serious emotional, behavioural and cognitive disturbances, was five times higher than that in the non-attempters. In contrast, the percentage of attempters who did not report any problems was five times lower than that in the non-attempters. The fact that almost 60% of the suicide attempters reported problems in three or four problem categories shows that they were very heterogeneous regarding the types of problem they reported, but fairly homogeneous regarding their reports of multiple problems.
Mann-Whitney-U tests revealed that the suicide attempters reported significantly more simultaneous problems in the problem categories than the non-attempters. The percentage of explained variance was 18%, which is fairly high.
More non-attempters than attempters reported a total of 0, 1 or 2 specific problems, while more attempters than non-attempters reported 3 to 10 specific problems.
One-way analysis of variance (ANOVA) showed that the suicide attempters reported significantly more specific problems than the non-attempters and explained a considerable amount of the variance.
It can be concluded that the difference between suicide attempters and non-attempters lies more in the number of problems present than in the type of problem or problem category.
Relationship with sex and age, single and multiple attempts The stability of the results reported above was tested across sex and age. Also the results of youngsters reporting one single suicide attempt and those reporting more than one suicide attempt were compared. Hiloglinear analyses showed that the direction and significance of the relationship and interactions between suicide attempters and the presence of single or multiple problems did not vary across sex and age groups (younger than 16 and 16 years and older) Also no significant difference was found between the frequency of problems reported by the students who reported one single suicide attempt ( N = 354) and those who reported multiple attempts ( N = 123).
However, since it might still be that within the suicide-attempters subsample sex differences do occur with regard to specific patterns of co-occurrence, in the next section separate analyses were carried out for male and female suicide attempters. The percentage of male and female suicide attempters with single and multiple problems As a group, the suicide attempters were characterized by reporting more than one problem in more than one problem category. It was more the rule than the exception that adolescent suicide attempters reported the co-occurrence of problems in several categories. Generally speaking, in about 80 to 99% of the attempters, reporting problems in one problem category also implicated reporting problems in one or more of the other categories. This proves that there was a large degree of co-occurrence on an individual level. Between the sexes, there were considerable differences in the types of co-occurring problem in the problem categories.
In the group of female suicide attempters , the category emotional problems was reported most frequently, not only as a single problem category (19%), but also as a co-occurring problem category (74 to 87%). Of the girls who reported emotional problems, 47% reported addiction risk behaviour, 49% reported aggressive/criminal behaviour and 55% reported cognitive problems as co-occurring problem categories.
Only a very small percentage of the female suicide attempters reported aggressive/criminal behaviour as a single problem category without the co-occurrence of other problem categories (1.4%). Of the girls who reported \linebreak aggressive/criminal behaviour, 87% reported emotional problems, 70% addiction risk behaviour and almost 60% cognitive problems as co-occurring problem categories.
Addiction-risk problems was reported as a single problem category by 1 in 14 (7%) of the female suicide attempters. Overlap percentages ranged from 65% for cognitive problems to 74% for emotional problems.
Only a very small percentage reported cognitive problems as a single problem category (2.1%). There was co-occurrence with addiction risk in 55%, aggressive-criminal behaviour in 57%, and emotional problems in 84%.
In the group of male suicide attempters , the category aggressive-criminal behaviour was reported most frequently, not only as a `single' problem category (8%), but also as a co-occurring problem category (75 to 87%). Of the boys reporting aggressive-criminal problems, 71% reported addiction risk behaviour, 59% reported emotional problems and 47% reported cognitive problems as co-occurring problem category.
Addiction risk behaviour was reported as a single problem category by a very low percentage of male suicide attempters (1.6%). There was overlap with other problem categories in 98%. Of the boys who reported addiction risk behaviour, 87% reported aggressive/criminal behaviour, 59.8% emotional and 48% cognitive problems as an additional problem category.
Emotional problems were reported as a single problem category by 7% of the male suicide attempters. Overlap percentages with other problem categories ranged from 51% for cognitive problems to 63% for addiction risk behaviour.
Only 3% of the male suicide attempters reported cognitive problems as a single problem category. There was co-occurrence with addiction risk behaviour in 66%, aggressive/criminal behaviour in 78% and emotional problems in 67%.
Conclusions and discussion
Although many studies are available on the type of problems that distinguish adolescents with a history of suicidal behaviour from those without and there is a great deal of agreement between them, very little is known about whether such problems belong to a single problem category or whether there is comorbidity with other problems. This study was performed to obtain data on single and multiple emotional, behavioural and cognitive problems in a large sample of secondary school students aged 12 to 19 years.
When all the specific emotional, behavioural and cognitive problems were analysed separately, they were observed significantly more often in the suicide attempters than in the non-attempters, but this did not mean that problems were reported infrequently by the non-attempters. In fact, analyses on all the possible combinations of the four problem categories showed that more non-attempters than attempters reported problems in one single category, irrespective of its content. However, when we analysed the number of students who reported problems in two or more categories, the attempters outnumbered the non-attempters.
The differences between the number of categories in which problems were reported and between the total number of specific problems reported by attempters and non-attempters, appeared to be highly significant. No differences were found between male and female reporting patterns or age groups, or between the students who had made only one attempt and those who had made multiple attempts. In conclusion, on a group level, more attempters than non-attempters reported the co-occurrence of emotional, behavioural and/or cognitive problems.
On an individual level, we found that in 80 to 99% of the suicidal males and females, reporting problems in one category implicated reporting problems in one or more other categories. The problem category reported most often by the male suicide attempters was aggressive/criminal behaviour (77%) and there was co-occurrence with other behavioural, emotional and/or cognitive problems in 92%. The problem category reported most often by the female \linebreak suicide attempters was emotional problems (79%) and there was co-occurrence with other emotional, behavioural and/or cognitive problems in 81%. In conclusion, on an individual level there was a large degree of comorbidity in suicidal adolescents, although the types of problem which co-occurred differed between males and females.
Several limitations can be mentioned in association with this study. Our population comprised a sample of white secondary school students, so it is difficult to estimate the extent to which the findings can be generalized to other samples. For example, adolescents who have dropped-out of school or have left school were missed by our sampling procedures. Another limitation was that the data were obtained using a self-report questionnaire. There were no independent ratings via relevant others or via more objective methods, either for measuring suicide attempts or for measuring the prevalence of emotional, behavioural and cognitive problems. In a retrospective design, nothing can be concluded about the causal direction of the relationships. In this study, emotional, behavioural and cognitive problems were assessed as point prevalences in clearly defined periods in the questionnaire, i.e., in the past month or past year, whereas the question of previous suicide attempts was assessed as life-time prevalence. Therefore, it is possible that in chronological order, the reported suicide attempt(s) may have preceded the reported problem(s). Hypothetically, suicide attempts can just as easily be the cause of the piling-up of emotional, behavioural and/or cognitive problems as the consequence.
To test the validity of the conclusions drawn from the findings in this study and to determine the extent to which the accumulation of problems plays a predictive or even causal role in suicide attempts, prospective studies should be performed on large samples using several different methods of data collection.
This study also had several strong points. For example, using a large and representative community sample increased the generalizability of the results. The large sample size meant that it contained a substantial number of adolescents who reported one or more suicide attempts, which increased the statistical power and enabled us to study differences between the sexes and age groups. The validity of the study findings was enhanced by using a matched control design to investigate differences between adolescent suicide attempters and non-attempters. We also used emotional, behavioural and cognitive variables which enabled us to examine their comorbidity. In our opinion, the percentual approach applied in this study provided valuable insight into `true' relationships and deserves further attention and application in future studies. Using this strategy, we were able to demonstrate differences between suicide attempters and non-attempters which are not usually revealed by `traditional' correlational methods, such as discriminant or regression analyses.
We found a clear-cut association between a history of suicide attempts and a multi-problem situation. Our results indicate that suicidal behaviour is related to a complex intermingling of emotional, behavioural and cognitive problems indicative of general psychological dysfunctioning.
Our conclusions are roughly consistent with the results published in a recent study by Lewinsohn et al. (1993), although we used a different method. They found that the more difficulties an adolescent has (besides depression), the more likely it is that he or she will have engaged in suicidal behaviour. Within their sample they demonstrated that the probability of past suicide attempts could be seen as a function of the number of risk factors. These results are also consistent with those of Jacobs (1971), who compared adolescent suicide attempters to a matched control group. He concluded that adolescent suicide attempters were generally facing many more problems in their lives than non-suicidal adolescents.
It is fairly usual for adolescents without a history of suicide attempts to report single emotional, behavioural or cognitive problems. In contrast, adolescents with a history of suicide attempts are to a large extent characterized by the co-occurrence of emotional, behavioural and/or cognitive problems. It can therefore be argued that the number or accumulation of emotional, behavioural and cognitive problems is of more importance in relation to a history of suicidal behaviour than the specific type of problem. This finding has important implications for the prediction and prevention of suicide attempts among adolescents. Our results suggest that screening procedures to detect adolescents at risk for suicidal behaviour should attribute more significance to patterns of multiple emotional, behavioural and cognitive problems than to single problems, such as addictive behaviour or emotional problems.
The fact that a large degree of co-occurrence of symptoms from distinct disorders was found, raises doubts upon the validity of the prevailing diagnostic and classification systems such as DSM or ICD. On base of our data, we are inclined to forward the hypothesis that the associations of attempted suicide (usually regarded as symptomatic for the existence of a clinical or subclinical affective disorder) with other affective symptoms are not necessarily stronger than with symptoms of other diagnostic entities such as conduct or anxiety disorders. If this hypothesis proves to be rightly, a phenomenological or phenotypical `reshuffling' of disorder entities would be called for.
Andrews, F. M., Crandall, R. (1976). The validity of measures of self-reported well-being. Social Indicators Research, 1 , 1--26.
Andrews, J. A., Lewinsohn, P. M. The prevalence, lethality and intent of suicide attempts among adolescents . Paper presented at the 98th Annual Convention of the American Psychological Association, Boston, August 10--14, 1990.
Andrews, J. A., Lewinsohn, P. M. (1992). Suicidal attempts among older adolescents: Prevalence and co-occurrence with psychiatric disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 31 , 655--662.
Andrews, F. M., Whitey, S. B. (1976). Social indicators of well-being . New York, NY: Plenum Press.
Bachman J. G., Johnson, L. D., O'Malley, P. M. (1987). Monitoring the future: Questionnaire responses from the nation's high school seniors 1986 . The University of Michigan, Ann Arbor, Michigan: Institute for Social Research.
Capaldi, D. M. (1992). Co-occurrence of conduct problems and depressive symptoms in early adolescent boys: II. A 2-year follow-up at Grade 8. Development and Psychopathology, 4 , 125--144.
Centers of Disease Control (1991). Attempted suicide among high school students -- United States 1990, leads from the Morbidity and Mortality Weekly Report. JAMA, 266 , 14, 1911.
Choquet, M., Menke, H. (1987). Development of self-perceived risk behaviour and psychosomatic symptoms in adolescents: A longitudinal approach. Journal of Adolescence, 10 , 291--308.
Clark, D. C. (1993). Suicidal behaviour in childhood and adolescence: Recent studies and clinical implications. Psychiatric Annals, 23 , 271--283.
Diekstra, R. F. W., Garnefski, N., Heus, P., de Zwart, R., Praag, B. M. S. van Warnaar, M. (1991). Scholierenonderzoek 1990: Gedrag en gezondheid van scholieren uit het voortgezet onderwijs. ( Behaviour and health of high school students ). State University of Leiden: Department of Clinical and Health Psychology.
Diekstra, R. F. W., Kienhorst, C. W. M., Wilde, E. J. de (in press). Suicide and suicidal behaviour among adolescents. In M. Rutter D. Smith (Eds.), Psychosocial disorder in young people. Time trends and their causes (pp. x--n). London: John Wiley.
Donovan, J. E., Jessor, R. (1985). Structure of problem behavior in adolescence and young adulthood. Journal of Consulting and Clinical Psychology, 53 , 890--904.
Donovan, J. E., Jessor, R., Costa, F. M. (1988). Syndrome of problem behavior in adolescence: A replication. Journal of Consulting and Clinical Psychology, 56 , 762--765.
Dubow, E. F., Kausch, D. F., Blum, M. C., Reed, J. Bush, E. (1989). Correlates of suicidal ideation and attempts in a community sample of junior high and high school students. Journal of clinical child psychology, 18 , 158--166.
Elliott, D. S., Huizinga, D., Menard, S. (1989). Multiple problem youth: Delinquency, substance use, and mental health . New York: Springer-Verlag.
Garnefski, N., Diekstra, R. F. W., de Heus, P. (1992). A population-based survey of the characteristics of high school students with and without a history of suicidal behaviour. Acta Psychiatrica Scandinavica, 86 , 189--196.
Garrison, C. Z. (1989). The study of suicidal behavior in the schools. Suicide and life threatening behaviour, 19 , 120--130.
Grube, J. W., Morgan, M. (1990). The structure of problem behaviours among Irish adolescents. British Journal of Addiction, 85 , 667--675.
Harkavy-Friedman, J. M., Asnis, G. M., Boeck, M., DiFiore, J. (1987). Prevalence of specific suicidal behaviors in a high school sample. American Journal of Psychiatry, 144 , 1203--1206.
Ingersoll, G. M., Orr, D. P. (1989). Behavioral and emotional risk in early adolescents. Journal of Early Adolescence, 9 , 396--408.
Irwin, C. E., Vaughan, E. (1988). Psychosocial context of adolescent development. Journal of Adolescent Health Care, 9 , 11s--19s.
Jacobs, J. (1971). Adolescent suicide . New York: Wiley-Interscience.
Jessor, R., Donovan, J. E., Costa, F. M. (1991). Beyond adolescence: Problem behavior and young adult development . Cambridge: Cambridge University Press.
Jessor, R., Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth . New York: Academic Press.
Kienhorst, C. W. M., De Wilde, E. J., Van den Bout, J., Diekstra, R. F. W., Wolters, W. H. G. (1990). Self-reported suicidal behavior in Dutch secondary education students. Suicide and life threatening behavior, 20 , 101--112.
Kienhorst, C. W. M., De Wilde, E. J., Van den Bout, J., Diekstra, R. F. W., Wolters, W. H. G. (1990). Characteristics of suicide attempters in a population-based sample of Dutch adolescents. British Journal of Psychiatry, 156 , 243--248.
Lewinsohn, P. M., Rohde, P., Seeley, J. R. (1993). Psychosocial characteristics of adolescents with a history of suicide attempt. Journal of the American Academy of Child and Adolescent Psychiatry, 32 , 60--68.
Magnusson, D., Bergman, L. R. (1990). A pattern approach to the study of pathways from childhood to adulthood. In L. Robins M. Rutter (Eds.), Straight and devious pathways from childhood to adulthood (pp. 101--115). Cambridge: Cambridge University Press.
McGee, L., Newcomb, M. D. (1992). General deviance syndrome: Expanded hierarchical evaluations at four ages from early adolescence to adulthood. Journal of Consulting and Clinical Psychology, 60 , 766--776.
Nagy, S., Adcock, A. (1990). The Alabama Adolescent Health Survey: Health knowledge and behaviors. Summary Report II . Alabama: The University of Alabama and Troy State University.
Osgood, D. W., Johnston, L. D., O'Malley, P. M., Bachman, J. G. (1988). The generality of deviance in late adolescence and early adulthood. American Sociological Review, 53 , 81--93.
Pritchard, C., Diamond, I., Fielding, M., Cox, M., Choudry, N. (1987). Drug misuse and truancy in normal 4th and 5th year comprehensive schoolchildren. Maladjustment and Therapeutic Education, 5 , 2--17.
Pronovost, J., Cote, L., Ross, C. (1990). Epidemiological study of suicidal behaviour among secondary-school students. Canada's Mental Health, 38 , 9--14.
Puig-Antich, J., Goetz, D., Davis, M., Kaplan, T., Davies, S., Ostrow, L., Asnis, L., Twomey, J., Lyengar, S. Ryan, N. D. (1989). A controlled family history study of prepubertal major depressive disorder. Archives of General Psychiatry, 46 , 406--418.
Reinherz, H. Z., Frost, A. K., Pakiz, B. (1991). Changing faces: Correlates of depressive symptoms in late adolescence. Family and Community Health, 14 , 52--63.
Rubenstein, J. L., Heeren, T., Housman, D., Rubin, C., Stechler, G. (1989). Suicidal behavior in `normal' adolescents: Risk and protective factors. American Journal of Orthopsychiatry, 59 , 59--71.
Smith, K. Crawford, S. (1986). Suicidal behavior among `normal' high school students. Suicide and Life threatening behavior, 16 , 313--325.
Smith, K. (1990). Suicidal behaviour. School Psychology Review, 19 , 186--195.
Spirito, A., Brown, L., Overholser, J., Fritz, G. (1989). Attempted suicide in adolescence: A review and critique of the literature. Clinical Psychology Review, 9 , 335--363.
Verhulst, F. C., Achenbach, T. M., Althaus, M., Akkerhuis, G. W. (1988). A comparison of syndromes derived from the Child Behavior Checklist for American and Dutch girls aged 6--11 and 12--16. Journal of Child Psychology and Psychiatry and allied Disciplines, 29 , 879--895.
Vingilis, E., Adlaf, E. (1990). The structure of problem behaviour among Ontario high school students: A confirmatory-factor analysis. Special Issue: Traffic injury prevention. Health Education Research, 5 , 151--160.
White, H. R. (1992). Early problem behavior and later drug problems. Journal of Research in Crime and Delinquency, 29 , 412--429.
All pages copyright ©Priory Lodge Education Ltd 1994-2000.