Adolescent Alcohol Use, Suicidal Ideation, and Suicide Attempts

Article Outline

• Abstract
• Methods
  • Participants
  • Measures
  • Missing values
  • Data analysis
• Results
  • Overview
• Discussion
• Conclusion
• References
• Copyright

Abstract 

Purpose

To examine the association between self-reported alcohol use and suicide attempts among adolescents who did and did not report suicidal ideation during the past year.

Methods

Screening data from 31,953 students attending schools in the United States that implemented the Signs of Suicide (SOS) program in 2001–2002 were used in this analysis. Two types of alcohol use were investigated: heavy episodic drinking, and drinking while down. Self-reported suicide attempts were regressed on suicidal ideation and both measures of alcohol use, controlling for participants’ levels of depressive symptoms, and demographic characteristics.

Results

Logistic regression analyses indicated that both drinking while down and heavy episodic drinking were significantly associated with self-reported suicide attempts. Analyses examining the conditional association of alcohol use and suicidal ideation with self-reported suicide attempts revealed that drinking while down was associated with significantly greater risk of suicide attempt among those not reporting suicidal ideation in the past year. Heavy episodic drinking was associated with increased risk of suicide attempt equally among those who did and did not report suicidal ideation.

Conclusions

This study identified the use of alcohol while sad or depressed as a marker for suicidal behavior in adolescents who did not report ideating prior to an attempt, and hence, may not be detected by current strategies for assessing suicide risk. Findings from this study should provide further impetus for alcohol screening among clinicians beyond that motivated by concerns about alcohol and substance use.

Keywords: Suicide, Suicide attempts, Alcohol, Adolescent, Ideation, Sad, Blue, Down, Binge, Heavy Episodic drinking, Drinking

Suicide among adolescents and young adults is a pressing public health problem. Suicide is the third leading cause of death for young persons aged 10 to 24 years in the United States [1]. Approximately one-third of adolescents report having experienced suicidal ideation at some time in their life, and suicide attempts are made by about 1 in 10 [2]. Recent data collected by the Centers for Disease Control and Prevention have revealed a dramatic spike in suicide rates among older adolescents in 2004 following years of declines [3].

One important characteristic of attempted and completed suicide is the extent to which the act was planned. Unplanned, or impulsive, acts of suicide involve “little preparation or premeditation” [4]. Although estimates of impulsive suicides differ greatly, likely because of varying definitions of impulsivity [5], they constitute a substantial proportion of suicides. Several studies provide prevalence estimates exceeding 50% among both adults [5], [6], and adolescents [7]. In a study of 100 patients who made a severe suicide attempt, 84% reported no specific plan, and 69% reported no specific plan and had only fleeting thoughts of suicide or no suicidal ideation at all prior to their attempt [8]. Other studies identified one-quarter [5] and two-fifths [6] of hospitalized self-injured patients who attempted suicide with less than 5 minutes of premeditation. Recent research has found a trend of increasing prevalence of suicide attempts among adolescents in conjunction with decreasing prevalence of suicidal ideation and planning [9], suggesting that the prevalence of attempts in the absence of ideation may be increasing.

Planned and unplanned suicides and suicide attempts differ in important ways. Planned attempts have generally been associated with higher levels of depression, hopelessness, lethality, and better follow-through on treatment following the attempt [4]. Unplanned attempts are more prevalent in early as opposed to late adolescence [10] and are more common among males and those higher in aggressiveness [5]. In addition, impulsive suicidal behavior among adolescents may be more likely to occur following stressful life events [4], [7]. Because they may be less visibly depressed, those at risk for unplanned acts are less easily identifiable as at risk prior to their attempts, potentially thwarting prevention efforts focused on identifying individuals exhibiting the signs and symptoms of suicidal behavior [11].

One key differentiating factor in planned and unplanned suicide attempts may be alcohol use. Alcohol use is strongly associated with suicide among adolescents [10], and adolescent alcohol abuse has been blamed for the increase in suicides among young persons from 1956 to 1994 [12]. From a theoretical perspective, alcohol intoxication may play a particularly important role in unplanned suicides because of: (a) increased disinhibition and impulsivity, (b) increased aggression and negative affectivity, and (c) increased cognitive constriction (“alcohol myopia”) that limits the production of alternative coping strategies [13], [14]. For youths higher in aggression and impulsivity, such as those with attention deficit hyperactivity disorder and other disruptive behavior disorders, research suggests that alcohol may directly or indirectly increase the risk of suicide [15], [16]. Although provocative, much of this evidence is inferential, with little empirical research investigating the association between drinking and unplanned suicidal behavior.

The current study investigated the association between alcohol use and impulsive suicide attempts. Based on literature reviewed above, we expected that alcohol use would play a greater role in impulsive compared to nonimpulsive attempts after taking into account youths’ depressive symptoms and demographic characteristics. Two measures of problematic alcohol use were included in our analyses: self-reported heavy episodic or “binge” drinking, and drinking while feeling down. Binge drinking has been linked to suicide attempts in young people [17] and adolescents who drink to cope with negative emotional states are more likely to drink heavily and to be problem drinkers [18].

Back to Article Outline

Methods 

Participants 

Data for this study were collected from students (n = 33,889) participating in the Signs of Suicide (SOS) program during the 2001–2002 school year. SOS is a school-based prevention program developed by Screening for Mental Health, Inc., a nonprofit organization in Wellesley, Massachusetts. SOS incorporates two prominent suicide prevention strategies into a single program, combining a curriculum that aims to raise awareness of suicide and its related issues with a brief screening for depression and other risk factors associated with suicidal behavior [11]. Data for the present study consisted of screening forms from students attending 225 of the 594 schools in the United States implementing the program during the 2001–2002 school year. Table 1 presents demographic characteristics of this sample relative to national data [19], [20], [21].

Table 1. Demographic profile and prevalence of risk behaviors of the sample (N = 31,953)

	Sample distribution	National distribution
Race/ethnicity %a
White (non-Hispanic)	71.1	69.3
Black (non-Hispanic)	11.9	14.9
Hispanic	10.0	14.9
Asian	2.0	4.3
Indian	1.3	0.2
Multiracial	3.7	3.6
Other	0.0	6.7
	100%	100%
Gender %a
Male	48.3	50.9
Female	51.7	49.1
Age %a
13 and under	2.8	1.6
14	21.5	18.2
15	30.4	23.8
16	22.0	24.3
17	15.8	22.3
18 and over	7.6	9.9
	100%	100%
Graded
7	1.3	—
8	2.4	—
9	40.8	29.5
10	25.6	26.0
11	15.9	23.4
12	14.0	21.1
Mean depressive symptoms (SD)	9.0 (5.2)	—
Heavy episodic drinking, % past year b	29.2	—
Heavy episodic drinking, % past 30 daysb	—	28.3
Drinking while down, % past year	12.2	—
Suicide attempts, % past year c	4.9	8.5
Suicidal ideation, % past year c	16.6	16.9

Note: National distributions for race, gender, and age are from the population of U.S. students in grades 9–12. Sources:

Because this study involved a secondary analysis of anonymous data collected as part of the SOS program, it was declared exempt from human subjects’ approval by the University of Connecticut Health Center's institutional review board.

Measures 

All data for this study were obtained from the SOS program's student screening form. This form was completed anonymously by students during class time, scored privately by students themselves, and returned to the teacher or school counselor administering the program. The screening form included the Columbia Depression Scale (CDS) [22], a brief screening scale derived from the Diagnostic Interview Schedule for Children IV [23]. The dependent variable in our analyses, self-reported suicide attempt, was assessed in response to the following question on the CDS: “Have you tried to kill yourself in the last year? (yes or no).”

Independent variables included (a) heavy episodic drinking, (b) drinking while down, (c) depressive symptoms, and (d) suicidal ideation. Depressive symptoms were measured using the 19 CDS items assessing common symptoms such as sadness, anhedonia, and irritability. Because of the focus of our analysis, items including thoughts about suicide or suicidal behavior were not included in the measure of depressive symptoms. Responses to all items reflected the presence or absence of symptoms over the past year; the final scale consisted of the sum of the number of “yes” responses. The correlation between the full 22-item CDS and the truncated 19-item CDS (omitting the suicide questions) approached unity (Pearson r = .992). Suicidal ideation was assessed in response to the following CDS question: “Has there been a time (in the past year) when you thought seriously about killing yourself? (yes or no).” The SOS screening form included additional questions to assess alcohol use. Heavy episodic drinking (HED) was assessed with the following question: “In the past year, has there been a time when you had 5 or more alcoholic drinks in a row? (By “drinks” we mean any kind of beer, wine, or liquor? (yes or no),” and drinking while feeling sad or depressed was assessed with the following item: “In the past year, have you used alcohol because you were feeling down? (yes or no).” Items similar to this have been included in scales used to assess drinking motives related to coping with emotional distress or escaping negative emotions [18]. Finally, demographic variables that have been linked to variations in baseline suicide rates were included as predictors. These included students’ age, race/ethnicity, and gender. Respondents were asked to check all ethnic/racial categories that applied to them; respondents who chose more than one category were designated as multiracial.

Missing values 

For the measure of depressive symptoms, a mean score was calculated for participants with valid responses on at least 16 of the 19 CDS items. Participants with fewer than 16 valid responses were classified as missing on the scale. Participants with missing values on any of the single-item measures used in the analyses were excluded (i.e., case-wise deletion) from all analyses presented in the Results section below. After exclusions for missing data, the effective sample size was 31,953. To assess the robustness of results using case-wise deletion, two types of missing data assignment were performed. First, missing cases were assigned the overall sample mean (or mode for categorical variables) for each missing variable. Second, means (or modes) for each variable were calculated separately by cluster (e.g., school) and missing cases were then assigned the cluster-specific mean or mode for that variable.

Data analysis 

To account for the clustered sampling design in which students were nested within schools, we used SUDAAN 9.0.1 software [24]. SUDAAN was developed to address the complicated variance estimation required in the analysis of data obtained using complex sampling designs, including cluster-correlated data. The probability of attempting suicide within the past year was estimated in logistic regressions that included as predictors depressive symptoms, suicidal ideation, drinking while down, HED, and demographic controls. Demographic controls included gender, race/ethnicity (black, Hispanic, Asian, Indian, Multiracial vs. white), and age. Age ranged from 12 to 19; although youth under 12 and over 19 were included in the original sample, we excluded these from the analysis in order to limit the age range in this developmentally dynamic period of life. Two product terms representing the interactions of suicidal ideation with HED and drinking while down were included in subsequent models to test for differences in the association between alcohol use and self-reported suicide attempts among youth with and without suicidal ideation.

Back to Article Outline

Results 

Overview 

Simple chi-square tests revealed that drinking while down and HED were both significantly associated with suicide attempts. Among nonideating attempters, drinking while down was associated more strongly with suicide attempts than HED. This relationship was replicated in a multivariate context. Both heavy episodic drinking and drinking while down were significantly associated with increased risk for self-reported suicide attempts, controlling for demographics, depressive symptoms, and suicidal ideation. In addition, drinking alcohol while down was associated with a larger increase in the risk of suicide attempts among those who did not report suicidal ideation compared to those who did, but the risk of suicide for those who reported HED did not differ between ideators and nonideators.

The demographics of the sample and the prevalence of suicidal behaviors are presented in Table 1. Almost 5% of the sample reported attempting suicide and 16.6% reported thoughts about suicide in the past year. The 1-year prevalence rates of heavy episodic drinking and drinking while down were 29.2% and 12.2%, respectively. The overlap between suicide attempts and drinking behaviors among the adolescents in the sample are presented in Table 2. Both drinking while down and heavy episodic drinking were significantly associated with self-reported suicide attempts. Almost 18% of students who drank while down reported a suicide attempt in the past year, compared to only 3.1% of those who did not report drinking while down. The association between heavy episodic drinking and suicide attempts was statistically significant but of lesser magnitude: 8.8% of students who reported heavy episodic drinking reported a suicide attempt in the past year, compared to 3.3% of students who did not report heavy episodic drinking. Finally, suicidal ideation was strongly associated with self-reported suicide attempts: 27.3% of respondents who reported thinking about suicide (ideators), compared to 0.4% of those who did not (nonideators), reported a suicide attempt.

Table 2. Numbers and column percents for suicide attempts within ideation and reported alcohol use

To examine these associations in a multivariate context, we performed a logistic regression analysis in which self-reported suicide attempts were regressed on suicidal ideation and the two measures of alcohol use, along with controls for participants’ levels of depressive symptoms and demographic characteristics. Results from the main effects model are presented in Table 3 (Model A). Not surprisingly, reporting suicidal ideation in the past year dramatically increased the risk of an attempt (odds ratio [OR] = e^3.64 = 38.2, confidence interval [CI] = 29.6, 48.3). Results also indicate that both heavy episodic drinking and drinking while down were significantly associated with increased risk for self-reported suicide attempts, controlling for depressive symptoms and suicidal ideation. Drinking alcohol while down in particular conveyed more than a 75% increase in the risk of self-reported attempts in adolescents.

Table 3. Logistic regression models predicting suicide attempts

Note: HED = heavy episodic drinking; CI = confidence interval.

∗p < .05, n = 31,953.

To test whether the association of alcohol use and suicide attempts differed among those who had and had not reported suicidal ideation, the interactions of suicidal ideation with HED and drinking while down were added separately to the equation presented in Model A to yield Models B and C in Table 3. The coefficient for the interaction of suicidal ideation and drinking alcohol while down was significant and negative (B = −.59, SE = .21), indicating that drinking alcohol while down was associated with a higher risk of suicide attempts among those who did not report suicidal ideation. Odds ratios calculated from the coefficients in this model indicated that one or more episodes of drinking while down in the past year was associated with a threefold increase in risk of self-reported suicide attempts among those not reporting suicidal ideation (OR = 3.02, CI = 2.03, 4.49), but was associated with a 68% increase among those who did report ideation (OR = 1.68, CI = 1.46, 1.93). In contrast, the interaction of HED and suicidal ideation presented in Model C failed to achieve statistical significance at the .05 level (B = −.40, SE = .21), indicating that the association of heavy episodic drinking and suicide attempts did not differ between those who did and did not report suicidal ideation.

Additional models were estimated to examine whether the association of alcohol use and suicidal ideation on self-reported suicide attempts differed among demographic groups defined by gender, age, and race/ethnicity. None of the three-way interaction terms estimated in these equations achieved statistical significance at the .05 level. In addition, to assess the impact of casewise deletion of missing data, models were reestimated with the imputation strategies described above. Using these alternative approaches to the treatment of missing data, coefficients were virtually identical to those presented in Table 3, thus substantiating the robustness of our results.

Back to Article Outline

Discussion 

Previous research indicates that impulsive acts of self-harm, those acts that are not accompanied by a specific plan or even thoughts of killing oneself, constitute an important subtype of suicidal behavior among adolescents that comprise a substantial proportion of medically serious attempts [5], [6]. Youths who do not exhibit ideation prior to their attempt present a difficult challenge to prevention and early detection, because the absence of suicidal thoughts may signify the absence of depression or less severe depressive symptomatology that may evade detection through commonly used screening instruments, which typically include questions about suicidal thoughts and previous suicidal behavior (e.g., the Brief Screen for Adolescent Depression [25]). Our study has identified an important marker for impulsive suicidal behavior among nonideating adolescents: drinking alcohol while down conveyed a threefold increase in the risk of self-reported suicide attempts among youths not reporting suicidal ideation, thus providing an alternative avenue for identification and early intervention. This finding is consistent with previous research that has identified mechanisms through which alcohol may promote increased impulsivity and aggressiveness, leading to intentional self-harm [13]. In addition, it illustrates the potential danger of alcohol use as self-medication for depression [18] and reinforces recent research documenting a link between self-medication with alcohol and suicidal behavior [26].

Although interview-based studies provide extensive evidence that impulsive acts of suicide among nonideators constitute a real phenomenon, it is possible that retrospective self-report studies such as ours may be plagued by intentional or unintentional misreporting. In a recent study using data from the National Longitudinal Study of Adolescent Health, there was evidence of both inaccurate responders, whose inaccuracies were most likely the result of carelessness or confusion (2.4% of respondents), and “jokester” responders, whose inaccuracies were most likely the result of “intentional mischief” (about .6% of respondents) [27]. These results suggest that some of the suicide attempters in our study who did not report ideation, and thus did not report the typical progression of ideation, planning, and attempt [28] may have responded to these questions erroneously.

Alternatively, it is possible that some suicide attempters may be erroneously reporting the presence of prior suicidal ideation. Because autobiographical memory is constructive, and not reproductive, it is prone to errors and illusions [29]. People often make decisions about the occurrence of an autobiographical event based on its plausibility, given their general knowledge about their own life history and about the way they understand the memory process itself, and the world in general, to work [30], [31]. Mazzoni and Kirsch [30] differentiate between autobiographical memory—which involves actual memory of an event—and autobiographical belief—which involves a judgment that an event occurred based on sources of information other than specific recollections. In their model synthesizing recent research on memory, persons may judge that an autobiographical event occurred without having a specific memory of it, if the event was plausible based on preexisting or new knowledge. In addition, studies have found that thoughts are more likely to be falsely remembered than events [32]. Thus, it is plausible, perhaps even likely, that some respondents who knew that they had attempted suicide wrongly assumed that this was not an impulsive, unpremeditated act. Clearly, future research should seek to clarify the prevalence of this type of suicidal behavior in community samples and to determine the extent to which misreporting, measurement error, and misremembering impacts results obtained in self-report surveys assessing this phenomenon.

Findings from this study must be considered in the context of other limitations as well. This sample, although large and diverse, was not a systematic probability sample, and only 38% of schools implementing the SOS program returned screening forms for analysis. However, the race and sex distribution of the sample (Table 1) was very close to the U.S. high school population distribution reported in the 2000 census. Moreover, prevalence estimates of suicidal behaviors in this sample were very similar to those observed in previous population-based studies investigating adolescent suicide, suggesting that these data are broadly representative. For example, our prevalence of lifetime and past year suicide attempts (10.5% and 4.9%, respectively) were very similar to those calculated in a recent meta-analysis (9.7% and 6.4%, respectively) [2]. Second, this study was cross-sectional, and causal processes linking alcohol to suicidal ideation and behavior must be inferred and cannot be tested in the data. Third, although the measure of heavy episodic drinking used in this study has been widely employed in epidemiologic research (e.g., CDC's Youth Risk Behavior Survey; Monitoring the Future), it has a higher threshold for heavy episodic drinking among women than is contained in the National Institute of Alcohol Abuse and Alcoholism's low risk guidelines (i.e., four or more drinks on a single occasion) [33]. Use of a lower threshold to measure heavy drinking among girls in this study might have increased the sensitivity of analyses to the association between HED, suicidal ideation, and suicide attempts. Fourth, other types of substance use are more common among alcoholics who attempt suicide [34], and thus some of the effect we have attributed to alcohol could actually be the result of other substance use. Because we did not have information on substance use other than alcohol, we were unable to control for it. This limitation may be somewhat less important regarding impulsive suicide attempts, however, because previous research has found that use of more than one substance does not increase the risk of an impulsive attempt above the use of one substance, a finding consistent with disinhibition as a mechanism [35].

Back to Article Outline

Conclusion 

This study has identified the use of alcohol while sad or depressed as a marker for suicidal behavior in adolescents who may not engage in planning or ideating prior to an attempt, and hence, may not be detected by current strategies for assessing suicide risk. Although problematic use of alcohol among adolescents is readily detectable using current screening approaches (e.g., AUDIT), routine screening for adolescent alcohol use by pediatricians and family practitioners is not universally practiced [36] despite the recommendations of the American Academy of Pediatrics, the American Medical Association, and the Society for Adolescent Medicine [37], [38], [39]. Findings from this study should provide further impetus for alcohol and drug screening among pediatricians and family practitioners beyond that motivated by concerns about alcohol and substance use.

Back to Article Outline

References 

1. Centers for Disease Control and Prevention . Youth suicide and attempted suicide. MMWR Morb Mortal Wkly Rep. 2004;53:471–498
   • View In Article
2. Evans E, Hawton K, Rodham K, Deeks J. The prevalence of suicidal phenomena in adolescents: a systematic review of population-based studies. Suicide Life Threat Behav. 2005;35:239–250
   • View In Article
   • MEDLINE
   • CrossRef
3. Gibbons RD, Brown CH, Hur K, et al. Early evidence on the effects of regulators’ suicidality warnings on SSRI Prescriptions and suicide in children and adolescents. Am J Psychiatry. 2007;164:1356–1363
   • View In Article
   • CrossRef
4. Conner KR. A call for research on planned vs. unplanned suicidal behavior. Suicide Life Threat Behav. 2004;34:89–98
   • View In Article
   • MEDLINE
   • CrossRef
5. Simon TR, Swann AC, Powell KE, et al. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32:49–59
   • View In Article
   • MEDLINE
   • CrossRef
6. Williams CL, Davidson JA, Montgomery I. Impulsive suicidal behavior. J Clin Psychol. 1980;36:90–94
   • View In Article
   • MEDLINE
   • CrossRef
7. Shaffer D, Pfeffer CR. Practice parameter for the assessment and treatment of children and adolescents with suicidal behavior. J Am Acad Child Adolesc Psychiatry. 2001;40:24S–51S
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (340 KB)
   • CrossRef
8. Hall RCW, Platt DE, Hall RCW. Suicide risk assessment: a review of risk factors for suicide in 100 patients who made severe suicide attempts: evaluation of suicide risk in a time of managed care. Psychosom J Consult Liaison Psychiatry. 1999;40:18–27
   • View In Article
9. Brener ND, Krug EG, Simon TR. Trends in suicide ideation and suicidal behavior among high school students in the United States, 1991–1997. Suicide Life Threat Behav. 2000;30:304–311
   • View In Article
   • MEDLINE
10. Brent DA, Baugher M, Bridge J, et al. Age- and sex-related risk factors for adolescent suicide. J Am Acad Child Adolesc Psychiatry. 1999;38:1497–1505
    • View In Article
    • Abstract
    • Full-Text PDF (821 KB)
    • CrossRef
11. Aseltine RH, DeMartino R. An outcome evaluation of the SOS suicide prevention program. Am J Public Health. 2004;94:446–451
    • View In Article
    • MEDLINE
    • CrossRef
12. Sher L, Zalsman G. Alcohol and adolescent suicide. Int J Adolesc Med Health. 2005;17:197–203
    • View In Article
    • MEDLINE
    • CrossRef
13. Sher L. Alcohol consumption and suicide. QJM. 2006;99:57–61
    • View In Article
    • MEDLINE
    • CrossRef
14. Hufford MR. Alcohol and suicidal behavior. Clin Psychol Rev. 2001;21:797–811
    • View In Article
    • MEDLINE
    • CrossRef
15. James A, Lai FH, Dahl C. Attention deficit hyperactivity disorder and suicide: a review of possible associations. Acta Psychiatr Scand. 2004;110:408–415
    • View In Article
    • MEDLINE
    • CrossRef
16. Kelly TM, Cornelius JR, Clark DB. Psychiatric disorders and attempted suicide among adolescents with substance use disorders. Drug Alcohol Depend. 2004;73:87–97
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (118 KB)
    • CrossRef
17. Powell KE, Kresnow M-J, Mercy JA, et al. Alcohol consumption and nearly lethal suicide attempts. Suicide Life Threat Behav. 2001;32:30–41
    • View In Article
    • MEDLINE
    • CrossRef
18. Kuntsche E, Knibbe R, Gmel G, Engels R. Why do young people drink? A review of drinking motives. Clin Psychol Rev. 2005;25:841–861
    • View In Article
    • MEDLINE
    • CrossRef
19. Martinez GM, Curry AE. School Enrollment—Social and Economic Characteristics of Students (Update). Washington, DC: U.S. Bureau of the Census; 1999;
    • View In Article
20. U.S. Census Bureau . Enrollment status of the population 3 years old and over, by age, sex, race, Hispanic origin, nativity and selected educational characteristics. Washington, DC: U.S. Census Bureau, Population Division, Education & Social Stratification Branch, 2002; October 2002;
    • View In Article
21. Centers for Disease Control and Prevention. National Youth Risk Behavior Survey: 1991–2005: Trends in the Prevalence of Suicide Ideation and Attempts. Available at http://www.cdc.gov/HealthyYouth/yrbs/pdf/trends/2005_YRBS_Suicide_Attempts.pdf. Accessed February 11, 2008.
    • View In Article
22. Shaffer D, Garland A, Vieland V, Underwood M. The impact of curriculum-based suicide prevention programs for teenagers. J Am Acad Child Adolesc Psychiatry. 1991;30:588–596
    • View In Article
    • Abstract
    • Full-Text PDF (8964 KB)
    • CrossRef
23. Shaffer D, Fisher P, Lucas CP, et al. NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): description, differences from previous versions, and reliability of some common diagnoses. J Am Acad Child Adolesc Psychiatry. 2000;39:28–38
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1097 KB)
    • CrossRef
24. Research Triangle Institute. SUDAAN 9.0.1. Research Triangle Park, NC, 2004.
    • View In Article
25. Lucas CP, Zhang H, Fisher PW, et al. The DISC Predictive Scales (DPS): efficiently screening for diagnoses. J Am Acad Child Adolesc Psychiatry. 2001;40:443–449
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (117 KB)
    • CrossRef
26. Bolton J, Cox B, Clara I, Sareen J. Use of alcohol and drugs to self-medicate anxiety disorders in a nationally representative sample. J Nerv Ment Dis. 2006;194:818–825
    • View In Article
27. Fan X, Miller BC, Park K-E, et al. An exploratory study about inaccuracy and invalidity in adolescent self-report surveys. Field Methods. 2006;18:223–244
    • View In Article
28. Perez VW. The relationship between seriously considering, planning, and attempting suicide in the Youth Risk Behavior Survey. Suicide Life Threat Behav. 2005;35:35–49
    • View In Article
    • MEDLINE
    • CrossRef
29. Schacter DL, Rose AD. The cognitive neuroscience of constructive memory: remembering the past and imagining the future. Philos Trans R Soc Lond B Biol Sci. 2007;362:773–786
    • View In Article
    • MEDLINE
    • CrossRef
30. Mazzoni G, Kirsch I. Autobiographical memories and beliefs: a preliminary metacognitive model. In:  Perfect TJ,  Schwartz BL editor. Applied Metacognition. Cambridge: Cambridge University Press; 2002;p. 121–145
    • View In Article
31. Mazzoni G, Vannucci M. Hindsight bias, the misinformation effect, and false autobiographical memories. Soc Cognit. 2007;25:203–220
    • View In Article
32. Conway MA, Collins AF, Gathercole SE, Anderson SJ. Recollections of true and false autobiographical memories. J Exp Psychol Gen. 1996;125:69–95
    • View In Article
    • CrossRef
33. National Institute on Alcohol Abuse and Alcoholism (NIAAA) . Helping Patients Who Drink Too Much: A Clinician's Guide. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism; 2005;
    • View In Article
34. Hesselbrock M, Hesselbrock V, Syzmanski K, Weidenman M. Suicide attempts and alcoholism. J Stud Alcohol. 1988;49:436–442
    • View In Article
    • MEDLINE
35. Borges G, Walters EE, Kessler RC. Associations of substance use, abuse, and dependence with subsequent suicidal behavior. Am J Epidemiol. 2000;151:781–789
    • View In Article
    • MEDLINE
36. Millstein SG, Marcell AV. Screening and counseling for adolescent alcohol use among primary care physicians in the United States. Pediatrics. 2003;111:114–122
    • View In Article
    • CrossRef
37. American Medical Association. H-30.942 Screening and Brief Interventions for Alcohol Problems. Available at http://www.ama-assn.org/apps/pf_new/pf_online?f_n=resultLink&doc=policyfiles/HnE/H-30.942.HTM&s_t=alcohol&catg=AMA/HnE&catg=AMA/BnGnC&catg=AMA/DIR&&nth=1&&st_p=0&nth=4&. Accessed February 4, 2008.
    • View In Article
38. Rosen DS, Elster A, Hedberg V, Paperny D. Clinical preventive services for adolescents: position paper of the Society for Adolescent Medicine. J Adolesc Health. 1997;21:203–214
    • View In Article
    • Full-Text PDF (1177 KB)
    • CrossRef
39. American Academy of Pediatrics . Alcohol use and abuse: a pediatric concern. Pediatrics. 2001;108:185–189
    • View In Article
    • CrossRef
40. National Center for Education Statistics. Common Core of Data (CCD), “State Nonfiscal Survey of Public Elementary/Secondary Education.” Available at http://nces.ed.gov/ccd/bat/output.asp. Accessed July 8, 2008.
    • View In Article