- Email: [email protected]
Posttraumatic Stress Disorder following childhood abuse increases the severity of suicide attempts
Journal of Affective Disorders 170 (2015) 7–14
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Posttraumatic Stress Disorder following childhood abuse increases the severity of suicide attempts Jorge Lopez-Castroman a,b,c,n, Isabelle Jaussent a, Severine Beziat a, Sebastien Guillaume a,c,d, Enrique Baca-Garcia b,e, Emilie Olié a,c,d, Philippe Courtet a,c,d a
Inserm U1061, Ho^pital La Colombiere, Montpellier, France IIS-Fundacion Jimenez Diaz, Department of Psychiatry, CIBERSAM, Madrid, Spain c Department of Emergency Psychiatry, CHRU Montpellier, France d University of Montpellier 1, Montpellier, France e Department of Psychiatry at the New York State Psychiatric Institute and Columbia University, NY, USA b
art ic l e i nf o
a b s t r a c t
Article history: Received 28 February 2014 Received in revised form 1 July 2014 Accepted 7 August 2014 Available online 15 August 2014
Objective: Posttraumatic Stress Disorder (PTSD) and childhood abuse are both consistently associated with a higher risk for suicide attempts. We hypothesize that among patients reporting childhood abuse, PTSD diagnoses are correlated with an increased severity of suicidal features. Method: We investigated 726 adult patients who had attempted suicide. These participants were assessed on lifetime clinical diagnoses and childhood abuse. The association of PTSD and childhood abuse dimensions with age at first suicide attempt, number of suicide attempts, violent attempts, serious attempts and suicide intent was studied. An adjusted multinomial logistic regression was performed to ascertain if childhood abuse and PTSD increased the severity of the suicidal behavior when combined. Results: Several types of childhood abuse (emotional, physical and sexual abuse) when combined with a lifetime diagnosis of PTSD showed an increased risk for more suicide attempts, serious attempts, and a higher level of suicidal intent compared with the absence of any or both risk factors. Conclusion: The combination of PTSD and childhood abuse should be investigated in clinical settings due to an augmented risk for more severe suicidal behavior. & 2014 Elsevier B.V. All rights reserved.
keywords: Posttraumatic Stress Disorders Suicidal behavior Childhood abuse
1. Introduction The development of Posttraumatic Stress Disorder (PTSD) depends on the type and severity of trauma exposure, as well as on individual vulnerability (Davidson et al., 2004). In Europe, the lifetime prevalence of PTSD has been estimated at 1.9% (Alonso et al., 2004), although rates may largely vary between countries (Hauffa et al., 2011). When present, PTSD has been repeatedly associated with a high suicidal risk (Krysinska and Lester, 2010). In fact, PTSD diagnoses, but not trauma exposure, were recently associated with suicide attempts (SA) in a community sample of urban American young adults (Wilcox et al., 2009). This association between PTSD and subsequent SAs appears to be independent of other mental disorders (Wilcox et al., 2009; Krysinska and Lester, 2010). Moreover, among mental disorders, PTSD has been estimated to be the fourth-ranked contributor to SAs with a 6.3% population-attributable fraction (Bolton and Robinson, 2010). This
n Corresponding author at: Inserm U1061, Hôpital La Colombiere, Pavillon 42, 39 Av Charles Flauhault, BP 34493, 34093 Montpellier, France. Tel. þ33 499614560. E-mail address: [email protected] (J. Lopez-Castroman).
http://dx.doi.org/10.1016/j.jad.2014.08.010 0165-0327/& 2014 Elsevier B.V. All rights reserved.
figure provides an approximation of the incidence of SAs that is directly linked to PTSD diagnoses. Childhood abuse is a prevalent public health problem that causes significant physical, psychological and societal consequences (Lu et al., 2008) and is also strongly associated with later SAs (Lopez-Castroman et al., 2012). In high-income countries, physical abuse may affect 4–16% of children, while 15–30% of girls and 5–15% of boys may suffer sexual abuse (Gilbert et al., 2009). Moreover, childhood abuse is a powerful predictor of future PTSD, in children and adolescents (Kearney et al., 2009) or adults (Ozer et al., 2003; Cloitre et al., 2010), and later-life revictimization, particularly among sexually-abused women (Coid et al., 2001). Several studies have investigated the link between childhood abuse and PTSD. The adjusted odds ratio for a lifetime diagnosis of PTSD following childhood maltreatment has been estimated to be 4.86 (Scott et al., 2010). Among individuals with PTSD, childhood abuse is associated with more comorbid mood disorders and more severe symptomatology (Teicher and Samson, 2013). Besides, a diagnosis of PTSD is more likely if the abused subject experiences another traumatic event as an adult (Xie et al., 2009). Although the mechanisms that link PTSD and childhood abuse with SAs are yet unknown, they may share some pathophysiological pathways.
8
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Autonomic and hypothalamic–pituitary–adrenal axis responses are altered in both conditions (Davidson et al., 2004; Carpenter et al., 2007) and may lead subjects experiencing childhood abuse and PTSD towards emotion dysregulation and interpersonal difficulties (Cloitre et al., 2010). Prior vulnerability traits, such as impulsive aggression, may be associated with the development of PTSD after a traumatic experience (Oquendo et al., 2005), and similar traits have been associated with childhood abuse and its transmission within families (Lopez-Castroman et al., 2012). Impulsivity might actually facilitate SAs through habituation to pain and fear, which impulsive individuals apparently experience more often (Bender et al., 2011). Therefore, an increased risk of suicide among abused subjects with subsequent PTSD can be expected. In this study we aim to investigate if the combination of childhood abuse and PTSD increases the severity of suicidal behaviors. We assessed a large sample of suicide attempters to ascertain a history of childhood abuse and a lifetime diagnosis of PTSD. We then examined the subtypes of childhood abuse, (i.e., emotional neglect, physical neglect, emotional abuse, physical abuse, sexual abuse) as well as the effect of PTSD as independent risk factors for more severe suicidal behaviors using five different indexes. We hypothesized that the occurrence of PTSD after childhood abuse would be associated with more severe suicidal behaviors during the lifetime of the participants, especially when reporting physical or sexual abuse.
2. Materials and methods 2.1. Participants Study participants were identified from a cohort of suicide attempters (n ¼1941), consecutively hospitalized and survivors of a current SA in a specialized unit of the Montpellier University Hospital. SAs were defined as self-injury behaviors with a nonzero level of suicidal intent (Silverman et al., 2007). Patients were aged 18 years old and over, French speaking, and had all four biological grandparents originating from Western European countries (for genetic purposes). All participants completed and returned a consent form. Overall, 726 patients had completed a diagnostic evaluation (including PTSD and major depression) and a questionnaire on childhood trauma. Excluded subjects presented a lower educational level and were less often diagnosed with nonPTSD anxiety disorders (po 0.05 for all comparisons), but did not statistically differ on the other socio-demographic and clinical variables. Trained psychiatrists or psychologists interviewed all patients. The local research ethics committee approved this study (CPP Sud Mediterranée IV, CHU Montpellier, France). 2.2. Assessment Patients were evaluated after remission of a potential mood episode (i.e. a current Hamilton Depression Rating Scale score below 15) (Hamilton, 1960). Either the French version of the Diagnostic Interview for Genetics Studies (DIGS) (Preisig et al., 1999) or the Mini International Neuropsychiatric Interview (MINI) (Sheehan et al., 1998) was used to obtain Axis I DSM-IV diagnoses. Lifetime diagnoses were determined using a best-estimate procedure: the psychiatrist in charge of the patient's care assigned the diagnosis based on MINI or DIGS interview, medical records and, when available, information from relatives (Kosten and Rounsaville, 1992). Lifetime PTSD diagnoses were assigned following MINI criteria for current or past PTSD. The assessment of the history of childhood trauma was performed using the short version of the
Childhood Trauma Questionnaire (CTQ) (Bernstein and Fink, 1998). It is a 28-item self-report questionnaire that investigates retrospectively five dimensions of child maltreatment: emotional abuse, emotional neglect, physical abuse, physical neglect and sexual abuse. Cut-off scores have been set for each type of trauma at four levels of maltreatment: None, Low, Moderate and Severe. The different cut-offs have been shown to have good specificity and sensitivity (Bernstein and Fink, 1998). Childhood trauma was considered only with moderate or severe scores. The suicide assessment procedure was based on the Columbia Suicide History Form (CSHF, Mann et al., 1999) and the Section O of the DIGS. The procedure is a semi-structured interview with validated questionnaires to collect information about sociodemographic features and characteristics of the suicide attempts. These questionnaires elicit in-depth information about lifetime suicide attempts, including suicidal methods and triggers, with questions such as “How many times have you tried to kill yourself?” (DIGS) or “When was the first time he/she ever made an attempt?” (CSHF). A suicide attempt was defined as violent, according to Asberg et al.'s (1976) criteria, when the method of suicide attempt was hanging, use of firearms, jumping from heights, several deep cuts, car crash, burning, gas poisoning, drowning, electrocution, or jumping under a train. Those suicide attempts that required intensive care interventions were considered serious. Age at first attempt was defined as the age at which the patient first made a SA. Age at first attempt was assessed by the interviewer and then blindly rated by an independent psychiatrist according to medical case notes and interviews. Cut-off for early age at first SA was set according to a mathematical modelization made in a previous study. In that study, the clinical picture and the history of childhood abuse differed between subjects younger or older than 26 years of age (Hamilton, 1960; Slama et al., 2009). The number of SAs for the analyses was also categorized using 1–2 and 42 as cut-off (Preisig et al., 1999; Lopez-Castroman et al., 2011). We further characterized suicidal behavior using the French version of the Suicide Intent Scale (SIS) (Beck et al., 1979; Sheehan et al., 1998), a 15-item semi-structured rating scale yielding a global score that indicates the severity of the suicidal intent. For each subject, only the highest SIS score of all previous SA was analyzed. Two subscales of the SIS were studied: expected lethality and planning (Brezo et al., 2008). The cut-off for the analyses of SIS scores was established using the higher tertile (SIS score r19 or 4 19). 2.3. Statistical analysis Associations between suicidal indexes (age at first SA, number of SA, SIS score, serious suicide attempt and violent suicide attempt) and subject characteristics, PTSD diagnosis and CTQ dimensions were quantified with odds ratios (OR) and their 95% Confidence Intervals (CI). Sociodemographic, and clinical variables associated with outcome variables (at p o0.05) were included in logistic regression models to estimate adjusted ORs for PTSD diagnosis and CTQ dimensions. When appropriate (i.e., when PTSD diagnosis or CTQ dimensions were significantly associated with the suicidal indexes), the interaction terms were tested using Wald χ2 tests given by the logistic regression model. If the interaction was not significant, we studied the additive effects of PTSD diagnosis and CTQ dimensions dividing the sample into 4 groups for each outcome variable: 1) subjects who reported both the childhood trauma and PTSD; 2) subjects who reported the childhood trauma but were not diagnosed with PTSD; 3) PTSD subjects not reporting childhood trauma; and 4) subjects who denied both risk factors. Significance level was set at p o0.05. Given the exploratory nature of our study, multiple test adjustments were not made (Rothman, 1990; Savitz
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
and Olshan, 1998; Bender and Lange, 2001). Analyses were performed using SAS statistical software (version 9.2; SAS Inc, Cary, North Carolina).
3. Results 3.1. Sample description Median age of the sample was 39.6 years (range: 18.0–83.4). Most of the sample was composed of women (n ¼540, 74.4%) with intermediate or high educational level (n ¼560, 79.5%). Less than one-third of the sample made more than two suicide attempts (28.5%) and these were rarely violent (11.0%) or severe (14.0%). Approximately half of the sample (46.5%) attempted suicide for the first time by the age of 26. Family history of suicidal behavior was common (38.8%). PTSD was diagnosed in 14.2% of the sample (n ¼103). The most frequent lifetime diagnoses were anxiety disorders (78.3%) and major depression (70.1%), followed by bipolar disorder (26.9%). Finally, a large majority of patients reported at least one type of childhood abuse according to the CTQ (68.9%, n¼ 500). One in every three subjects reported 3 or more different types of abuse (30.3%, n ¼ 220). The most frequent type of childhood abuse was emotional neglect (45.3%, n ¼329) followed by emotional abuse (42.0%, n¼ 305). Physical abuse was reported by 168 subjects (23.2%), sexual abuse by 212 subjects (29.2%) and physical neglect by 203 subjects (28.0%). Among subjects with lifetime PTSD, 88.3% (n ¼91) reported at least one type of childhood abuse and 58.2% reported three or more types of abuse (n¼ 60). The repetition of SAs (42) was associated with young age at the first SA ( o26 years; chi2 ¼40.3, df ¼1, p o0.0001), high suicidal intent (SIS score 419; chi2 ¼7.3, df ¼1, p ¼0.007), serious SAs (chi2 ¼44.8; df ¼1; po 0.0001) and violent SAs (chi2 ¼5.0; df ¼ 1; p¼ 0.025). A high suicidal intent was also associated with a history of serious SAs (chi2 ¼23.7; df ¼1; p o0.0001) or violent SAs (chi2 ¼3.7; df ¼1; p ¼0.053). No other significant associations were found between the suicidal indexes. In order to facilitate their presentation, henceforward all results will be adjusted by variables significantly associated with at least one outcome variable (age at first SA, number of SA, SIS score, serious SA, violent SA): sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder. Detailed results appear in Table 1. 3.2. Age at first SA Sexual abuse was associated with a young age at first SA (o26 years). Subjects with PTSD or emotional abuse were also younger at the time of their first SA, but these findings were borderline significant after adjustment. No other CTQ dimension was significantly associated with age at first SA and we found no interactions between CTQ dimensions and PTSD regarding age at first SA (data not shown). Sexual abuse without lifetime PTSD was associated with a younger age at first SA (OR¼1.67; 95% CI¼ 1.11–2.49). Similar, but non-significant, adjusted OR for early onset of SA were obtained in patients with PTSD but no sexual abuse (OR¼2.02; 95% CI¼0.99–4.12) and patients with a combination of both risk factors (OR¼1.70; 95% CI¼ 0.94–3.06) (Table 2). 3.3. Number of SAs PTSD and all types of childhood abuse, with the exception of physical neglect, were associated with a greater number of SAs (42 SAs). As CTQ dimensions and PTSD did not interact with respect to the number of SAs (data not shown), we studied the additive effects of their combination. Compared with subjects reporting no PTSD or
9
Table 1 Description of the sample (n¼ 726). Variables
n
%
Sex (female) Age median (range) Educational level (secondary or higher) Lifetime smoking Suicide attempt characteristics Violent suicide attempt Serious suicide attempt Number of suicide attempts median (range) Number of suicide attempts 42 SIS total median (range) SIS total 419 Age at first suicide attempt median (range) Age at first suicide attempt r 26 Family history of suicidal behavior Family history of suicide Lifetime diagnoses Posttraumatic Stress Disorder (PTSD) Major depression Bipolar disorder Anxiety disorders Alcohol use disorders Substance use disorders Cranial trauma Eating disorder Childhood Trauma Questionnaire CTQ - emotional neglect CTQ - physical neglect CTQ - emotional abuse CTQ - physical abuse CTQ - sexual abuse No abuse 1 abuse subtype 2 abuse subtype 3 abuse subtype 4 abuse subtype 5 abuse subtype
540 39.6 560 490
74.4 (18.0–83.4) 79.5 68.9
80 101 2 201 16 195 28 340 270 130
11.0 14.0 (1–21) 28.5 (0–30) 27.5 (7–74) 46.5 38.8 19.0
103 509 194 509 186 117 91 126
14.2 70.1 26.9 78.3 25.7 16.1 12.8 17.5
329 203 305 168 212 226 161 119 104 74 42
45.3 28.0 42.0 23.1 29.2 31.1 22.2 16.4 14.3 10.2 5.8
emotional neglect, patients with PTSD only (OR¼ 2.39; 95% CI¼1.17–4.86), emotional neglect only (OR¼1.72; 95% CI¼1.18– 2.52) or a combination of both (OR¼ 2.09; 95% CI¼1.13–3.87) were more likely to make a greater number of SAs in the adjusted analyses. Regarding emotional abuse, increased odds of making more SAs were found among subjects with emotional abuse but no PTSD (OR¼ 1.57; 95% CI¼ 1.07–2.30) or with a combination of both (OR¼2.56; 95% CI¼1.44–4.56). Subjects with PTSD and physical abuse combined were also more likely to make a greater number of SAs (OR¼3.43; 95% CI¼1.73–6.79) than those without risk factors, but not those with either physical abuse or PTSD alone. Finally, sexual abuse alone (OR¼1.96; 95% CI¼ 1.29–2.97) or in combination with lifetime PTSD (OR¼2.22; 95% CI¼1.24–3.96) was also associated with a greater number of SAs compared to the absence of these risk factors (Table 3, Fig. 1).
3.4. Suicide Intent Scale High Suicide Intent Scale scores were associated with lifetime PTSD diagnosis and with emotional, physical and sexual abuse. PTSD and these CTQ dimensions did not interact with regards to the suicidal intent (data not shown). After adjustment, PTSD diagnosis alone or emotional, physical or sexual abuses alone were not significantly associated with an increased suicidal intent compared with the absence of both risk factors. However, when coexisting with a PTSD diagnosis, all three types of abuse were significantly associated with increased suicidal intent: emotional (OR¼ 3.74; 95% CI¼2.08–6.72), physical (OR¼3.70; 95% CI¼1.85–7.38) and sexual (OR¼3.40; 95% CI¼1.89–6.13) (Table 4, Fig. 2).
10
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Table 2 Age at first suicide attempt compared by PTSD diagnosis and dimensions of the CTQ. Age at first attempt
426 N¼ 386
Variables
n
%
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
41 173 106 144 85 90
10.62 44.82 27.46 37.31 22.02 23.32
62 155 96 160
226 16 119 25 280 16 65 25
a
r 26 N ¼ 336 OR [95% CI]a
p-valuea
18.45 46.13 28.57 47.62
1.58 1.02 1.07 1.33
0.05 0.90
82 121
24.40 36.01
1.17 [0.80;1.69] 1.57 [1.10;2.24]
58.55 4.15 30.83 6.48
157 19 117 43
46.73 5.65 34.82 12.80
1 1.52 [0.72;3.19] 1.28 [0.90;1.81] 1.89 [1.05;3.40]
0.12
72.54 4.15 16.84 6.48
190 25 84 37
56.55 7.44 25.00 11.01
1 2.02 [0.99;4.12] 1.67 [1.11;2.49] 1.70 [0.94;3.06]
0.02
[0.99;2.51] [0.74;1.40] [0.75; 1.53] [0.96;1.84]
0.72 0.08 0.42 0.01
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Table 3 Number of suicide attempts compared by PTSD diagnosis and dimensions of the CTQ. Number of suicide attempts
SA¼ 1–2 N ¼ 515
Variables
n
%
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional neglect No–No PTSD–no emotional neglect No PTSD–emotional neglect Yes–Yes Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
59 215 144 190 104 121
11.46 41.75 27.96 36.89 20.19 23.50
43 113 57 112 64 90
275 25 181 34
53.40 4.85 35.15 6.60
300 25 156 34
a
SA42 N ¼205
OR [95% CI]a
p-valuea
20.98 55.12 27.80 54.63 31.22 43.90
1.70 1.60 0.87 1.70 1.68 1.90
0.03 0.008 0.49 0.003 0.009 0.0006
75 17 87 26
36.59 8.29 42.44 12.68
1 2.39 [1.17;4.86] 1.72 [1.18;2.52] 2.09 [1.13;3.87]
0.005
58.25 4.85 30.29 6.60
84 9 78 34
40.98 4.39 38.05 16.59
1 1.32 [0.58;3.01] 1.57 [1.07;2.30] 2.56 [1.44;4.56]
0.007
371 40 85 19
72.04 7.77 16.50 3.69
122 19 40 24
59.51 9.27 19.51 11.71
1 1.10 [0.58;2.06] 1.30 [0.83;2.05] 3.43 [1.73;6.79]
0.005
369 25 87 34
71.65 4.85 16.89 6.60
100 15 62 28
48.78 7.32 30.24 13.66
1 1.92 [0.93;3.99] 1.96 [1.29;2.97] 2.22 [1.24;3.96]
0.002
[1.07;2.72] [1.13;2.26] [0.59;1.29] [1.20;2.42] [1.14;2.47] [1.31;2.73]
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 1. Association of number of suicide attempts with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for greater number of suicide attempts ( 4 2) with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EN¼ emotional neglect (dark square); EA¼ emotional abuse (white square); PA ¼ physical abuse (triangle); SA¼sexual abuse (diamond).
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
11
Table 4 Suicide Intent Scale score compared by PTSD diagnosis and dimensions of the CTQ. Suicide Intent Scale score
r19 N ¼ 513
Variables
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
60 222 135 202 107 137
11.70 43.27 26.32 39.38 20.86 26.71
40 94 63 94 54 67
286 25 167 35
55.75 4.87 32.55 6.82
366 40 87 20 347 29 106 31
a
419 N ¼195 OR [95% CI]a
p-valuea
20.51 48.21 32.31 48.21 27.69 34.36
2.27 1.25 1.45 1.49 1.51 1.47
[1.42;3.64] [0.89;1.77] [1.00;2.11] [1.05;2.12] [1.03;2.24] [1.01;2.15]
0.0006 0.20 0.05 0.02 0.04 0.05
93 8 62 32
47.69 4.10 31.79 16.41
1 1.00 [0.43;2.33] 1.15 [0.78;1.70] 3.74 [2.08;6.72]
0.0002
71.35 7.80 16.96 3.90
121 20 34 20
62.05 10.26 17.44 10.26
1 1.75 [0.96;3.19] 1.22 [0.77;1.92] 3.70 [1.85;7.38]
0.002
67.64 5.65 20.66 6.04
117 11 38 29
60.00 5.64 19.49 14.87
1 1.22 [0.58;2.59] 1.04 [0.66;1.63] 3.40 [1.89;6.13]
0.0006
n
%
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 2. Association of high suicide intent with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for higher suicidal intent ( 419) with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EA ¼ emotional abuse (white square); PA ¼physical abuse (triangle); SA¼ sexual abuse (diamond).
3.5. Violent suicide attempt
4. Discussion
Violent suicide attempters were not associated with lifetime PTSD diagnosis or any CTQ dimension.
PTSD may be caused by childhood abuse or by other traumatic experiences, and both are powerful risk factors for suicidal behaviors. In this study we have tried to understand the relationship between these two exposures among suicide attempters by investigating the putative effect of their combination in a number of suicidal indexes. Some dimensions of childhood abuse were particularly associated with severity features of the suicide attempts when coexisting with a lifetime PTSD. Emotional neglect, either alone or in combination with lifetime PTSD, was associated with more SAs when compared with subjects without these conditions. However, no other significant association was found between emotional neglect or physical neglect and PTSD with regard to the severity of suicidal behavior. On the other hand, subjects with emotional abuse and PTSD presented higher odds of making serious attempts
3.6. Serious suicide attempt After adjustment, serious suicide attempts were still associated with lifetime PTSD and physical abuse but significant associations with sexual abuse and emotional abuse were lost. PTSD and these CTQ dimensions did not interact with regards to severe suicide attempts (data not shown). Subjects reporting PTSD and emotional abuse (OR ¼2.66; 95% CI ¼ 1.33–5.32), physical abuse (OR ¼4.14; 95% CI ¼ 1.99–8.61) or sexual abuse (OR ¼2.79; 95% CI ¼ 1.44–5.41) were more likely to make serious suicide attempts than subjects presenting only one or none of these risk factors (Table 5, Fig. 3).
12
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Table 5 Serious suicide attempters compared with non-serious attempters by PTSD diagnosis and dimensions of the CTQ. Serious suicide attempts
No N¼ 619
Variables
n
%
n
%
PTSD lifetime Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD– no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
79 279 170 249 133 171
12.76 45.07 27.46 40.23 21.49 27.63
23 48 32 53 35 40
340 30 200 49
54.93 4.85 32.31 7.92
435 51 105 28 413 35 127 44
a
Yes N¼ 101 OR [95% CI]a
p-valuea
22.77 47.52 31.68 52.48 34.65 39.60
1.86 0.98 1.24 1.50 2.08 1.55
0.03 0.92 0.38 0.07 0.002 0.06
43 5 35 18
42.57 4.95 34.65 17.82
1 1.27 [0.46;3.50] 1.30 [0.79;2.15] 2.66 [1.33;5.32]
0.05
70.27 8.24 16.96 4.52
58 8 20 15
57.43 7.92 19.80 14.85
1 1.03 [0.44;2.44] 1.57 [0.89;2.75] 4.14 [1.99;8.61]
0.001
66.72 5.65 20.52 7.11
56 5 22 18
55.45 4.95 21.78 17.82
1 0.83 [0.28;2.49] 1.11 [0.63;1.98] 2.79 [1.44;5.41]
0.02
[1.06;3.28] [0.63;1.52] [0.76;2.01] [0.96;2.35] [1.30;3.32] [0.97;2.49]
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 3. Association of serious suicide attempts with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for serious suicide attempts with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EA ¼emotional abuse (white square); PA ¼ physical abuse (triangle); SA¼ sexual abuse (diamond).
and of attempting suicide earlier, more times, and with higher intent than those with only one or none of these conditions. The combination of physical abuse and PTSD also presented high odds ratios (OR 43) for a greater number of SAs, serious attempts and higher suicidal intent, which was not present with physical abuse or PTSD alone. Finally, the combination of sexual abuse and PTSD diagnosis was not associated with earlier SAs in our sample but it was associated with more attempts and specifically with serious attempts and more suicidal intent (see Figs. 2 and 3). These findings support previous studies that emphasized the relevance of the type and severity of childhood maltreatment to assess the risk of suicidal behaviors in later life (Lopez-Castroman et al., 2013). Indeed, the mental health outcomes and even the structural changes on the brain of abused children may depend on the nature of the abuse itself (Oquendo et al., 2013). Interestingly, “commission” trauma (emotional, physical or sexual abuse) conveyed more severe suicidal behaviors than “omission” trauma (emotional and physical neglect) in our study. If the subtypes of early life adversity define the severity of suicidal outcomes, then interventions, such
as the enhancement of perceived social support (Powers et al., 2009), may need to be personalized for victims of abuse. Childhood maltreatment may alter developmental processes related to cognitive and affective functioning (Mann et al., 1999; Shipman et al., 2005; Pechtel and Pizzagalli, 2010) and is frequently experienced as a recurrent trauma adopting different forms (Dong et al., 2004; Gilbert et al., 2009). It may also elicit a state of learned helplessness and conditioned fear, which is associated with traumarelated depression symptoms in animal models (Hammack et al., 2012), and an increased startle response independent of the presence of PTSD symptoms (Jovanovic et al., 2009). Thus, childhood trauma may facilitate the re-experience of trauma and PTSD symptoms through helplessness (Wiedemar et al., 2008), or other psychophysiological and personality variables (Orr et al., 2012). Actually, sexually abused children are often diagnosed with PTSD in their adult life (Coid et al., 2001; Davidson et al., 2004) and PTSD emerging from childhood trauma appears to be associated with multiple other forms of trauma during the lifetime (Cloitre et al., 2010). Therefore, subjects experiencing childhood abuse and later-life PTSD symptoms may be more
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
exposed to emotional difficulties, for instance in anger management and interpersonal relationships, and high reactivity (Springer et al., 2007; Cloitre et al., 2010; Perroud et al., 2010; van Harmelen et al., 2013). An excessive reactivity to life experiences together with diminished social support and inefficient emotional processing may explain the association with higher number of SAs, serious attempts and higher suicidal intent that we found among patients reporting childhood abuse and later-life PTSD symptoms. In this study we have examined a large cohort of consecutively admitted suicide attempters, whose features were thoroughly assessed. Moreover, the results were adjusted for other risk factors such as alcohol or substance use disorders. However, some limitations should be contemplated. The time lapse from childhood maltreatment may convey a recall bias, although evidence supports the use of this kind of retrospective reports (Hardt and Rutter, 2004), and this information was not confirmed by other sources. The characteristics of CTQ prevented us from considering other relevant variables of the childhood abuse, such as duration or role of the abuser. Other adverse experiences, different from childhood abuse, were not registered in our sample and therefore we could not explore suicidality in other types of early trauma. Finally, we could not analyze the temporal relationship between traumatic experiences, PTSD diagnoses and suicidal behavior. Although PTSD diagnoses always preceded the last suicide attempt, i.e. the moment when the subjects were assessed, we could not distinguish if PTSD diagnoses arose from childhood abuse or other trauma or if PTSD preceded the first SA. Such a distinction may have allowed us to improve our understanding of the relationship between these conditions. We studied here the features of suicidal behaviors after particularly traumatic life events, an issue that is often avoided in clinical interviews. Even among patients with PTSD or childhood abuse history there is frequently a tendency to deny or minimize their problems (Goodman et al., 2003; Davidson et al., 2004). However, PTSD and childhood abuse, particularly when combined, are associated with more severe suicidal behavior and higher suicidal intent. Future studies are needed to investigate the risk of completed suicide when this combination is present. Clinicians should remember that effective treatments are available (Cloitre et al., 2010), and investigate PTSD symptoms and childhood abuse history when dealing with suicidal patients.
Role of funding source Dr. Lopez-Castroman was supported by a FondaMental Foundation research grant. This study was financially supported by Programme Hospitalier de Recherche Clinique (CHU of Montpellier—PHRC UF 7653), and Agence Nationale NEURO2007 de la Recherche (NEURO2007—GENESIS). These institutions had no further role in the conception, design or redaction of the manuscript.
Conflict of interest None.
Acknowledgments Dr. Lopez-Castroman was supported by a FondaMental Foundation research grant (grant no. R11075FF). This study was financially supported by the Programme Hospitalier de Recherche Clinique (CHU of Montpellier—PHRC UF 7653), and the Agence Nationale de la Recherche (NEURO2007—GENESIS) (grant no. ANR-07NEURO-013-01). We thank Rosa Nunes and Catherine Genty for editorial assistance.
References Alonso, J., Angermeyer, M.C., Bernert, S., Bruffaerts, R., Brugha, T.S., Bryson, H., de Girolamo, G., Graaf, R., Demyttenaere, K., Gasquet, I., Haro, J.M., Katz, S.J., Kessler, R.C., Kovess, V., Lépine, J.P., Ormel, J., Polidori, G., Russo, l.J., Vilagut, G., Almansa, J., Arbabzadeh-Bouchez, S., Autonell, J., Bernal, M., Buist-Bouwman, M.A., Codony, M., Domingo-Salvany, A., Ferrer, M., Joo, S.S., Martínez-Alonso, M., Matschinger, H., Mazzi, F., Morgan, Z., Morosini, P., Palacín, C., Romera, B., Taub,
13
N., Vollebergh, W.A., ESEMeD/MHEDEA 2000 Investigators, European Study of the Epidemiology of Mental Disorders (ESEMeD) Project, 2004. Disability and quality of life impact of mental disorders in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr. Scand. Suppl. 420, 38–46. Asberg, M., Träskman, L., Thorén, P., 1976. 5-HIAA in the cerebrospinal fluid. A biochemical suicide predictor? Arch. Gen. Psychiatry 33, 1193–1197. Beck, A.T., Kovacs, M., Weissman, A., 1979. Assessment of suicidal intention: the scale for suicide ideation. J. Consult. Clin. Psychol. 47, 343–352. Bender, R., Lange, S., 2001. Adjusting for multiple testing—when and how? J. Clin. Epidemiol. 54, 343–349. Bender, T.W., Gordon, K.H., Bresin, K., Joiner, T.E., 2011. Impulsivity and suicidality: the mediating role of painful and provocative experiences. J. Affect. Disord. 129, 301–307. Bernstein, D.P., Fink, L., 1998. Childhood Trauma Questionnaire: A Retrospective Self-Report Manual. The Psychological Corporation, San Antonio, TX. Bolton, J.M., Robinson, J., 2010. Population-attributable fractions of Axis I and Axis II mental disorders for suicide attempts: findings from a representative sample of the adult, noninstitutionalized US population. Am. J. Public Health 100, 2473–2480. Brezo, J., Paris, J., Hébert, M., Vitaro, F., Tremblay, R., Turecki, G., 2008. Broad and narrow personality traits as markers of one-time and repeated suicide attempts: a population-based study. BMC Psychiatry 8, 15. Carpenter, L.L., Carvalho, J.P., Tyrka, A.R., Wier, L.M., Mello, A.F., Mello, M.F., Anderson, G.M., Wilkinson, C.W., Price, L.H., 2007. Decreased adrenocorticotropic hormone and cortisol responses to stress in healthy adults reporting significant childhood maltreatment. Biol. Psychiatry 62, 1080–1087. Cloitre, M., Stovall-McClough, K.C., Nooner, K., Zorbas, P., Cherry, S., Jackson, C.L., Gan, W., Petkova, E., 2010. Treatment for PTSD related to childhood abuse: a randomized controlled trial. Am. J. Psychiatry 167, 915–924. Coid, J., Petruckevitch, A., Feder, G., Chung, W., Richardson, J., Moorey, S., 2001. Relation between childhood sexual and physical abuse and risk of revictimisation in women: a cross-sectional survey. Lancet 358, 450–454. Davidson, J.R.T., Stein, D.J., Shalev, A.Y., Yehuda, R., 2004. Posttraumatic stress disorder: acquisition, recognition, course, and treatment. J. Neuropsychiatry Clin. Neurosci. 16, 135–147. Dong, M., Anda, R.F., Felitti, V.J., Dube, S.R., Williamson, D.F., Thompson, T.J., Loo, C.M., Giles, W.H., 2004. The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction. Child Abuse Negl. 28, 771–784. Gilbert, R., Widom, C.S., Browne, K., Fergusson, D., Webb, E., Janson, S., 2009. Burden and consequences of child maltreatment in high-income countries. Lancet 373, 68–81. Goodman, G.S., Ghetti, S., Quas, J.A., Edelstein, R.S., Alexander, K.W., Redlich, A.D., Cordon, I.M., Jones, D.P.H., 2003. A prospective study of memory for child sexual abuse: new findings relevant to the repressed-memory controversy. Psychol. Sci. 14, 113–118. Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56–62. Hammack, S.E., Cooper, M.A., Lezak, K.R., 2012. Overlapping neurobiology of learned helplessness and conditioned defeat: implications for PTSD and mood disorders. Neuropharmacology 62, 565–575. Hardt, J., Rutter, M., 2004. Validity of adult retrospective reports of adverse childhood experiences: review of the evidence. J. Child Psychol. Psychiatry 45, 260–273. Hauffa, R., Rief, W., Brähler, E., Martin, A., Mewes, R., Glaesmer, H., 2011. Lifetime traumatic experiences and posttraumatic stress disorder in the German population. J. Nerv. Ment. Disord. 199, 934–939. Jovanovic, T., Blanding, N.Q., Norrholm, S.D., Duncan, E., Bradley, B., Ressler, K.J., 2009. Childhood abuse is associated with increased startle reactivity in adulthood. Depress. Anxiety 26, 1018–1026. Kearney, C.A., Wechsler, A., Kaur, H., Lemos-Miller, A., 2009. Posttraumatic stress disorder in maltreated youth: a review of contemporary research and thought. Clin. Child Fam. Psychol. Rev. 13, 46–76. Kosten, T.A., Rounsaville, B.J., 1992. Sensitivity of psychiatric diagnosis based on the best estimate procedure. Am. J. Psychiatry 149, 1225–1227. Krysinska, K., Lester, D., 2010. Post-traumatic stress disorder and suicide risk: a systematic review. Arch. Suicide Res. 14, 1–23. Lopez-Castroman, J., Jaussent, I., Beziat, S., Genty, C., Olié, E., de Leon-Martinez, V., Baca-García, E., Malafosse, A., Courtet, P., Guillaume, S., 2012. Suicidal phenotypes associated with family history of suicidal behavior and early traumatic experiences. J. Affect. Disord. 142, 193–199. Lopez-Castroman, J., Melhem, N., Birmaher, B., Greenhill, L., Kolko, D., Stanley, B., Zelazny, J., Brodsky, B., García-Nieto, R., Burke, A.K., Mann, J.J., Brent, D.A., Oquendo, M.A., 2013. Early childhood sexual abuse increases suicidal intent. World Psychiatry 12, 149–154. Lopez-Castroman, J., Perez-Rodriguez Mde, L.M., Jaussent, I., Alegria, A.A., ArtesRodriguez, A., Freed, P., Guillaume, S., Jollant, F., Leiva-Murillo, J.M., Malafosse, A., Oquendo, M.A., de Prado-Cumplido, M., Saiz-Ruiz, J., Baca-Garcia, E., Courtet, P., European Research Consortium for Suicide (EURECA), 2011. Distinguishing the relevant features of frequent suicide attempters. J. Psychiatr. Res. 45, 619–625. Lu, W., Mueser, K.T., Rosenberg, S.D., Jankowski, M.K., 2008. Correlates of adverse childhood experiences among adults with severe mood disorders. Psychiatr. Serv. 59, 1018–1026. Mann, J.J., Waternaux, C., Haas, G.L., Malone, K.M., 1999. Toward a clinical model of suicidal behavior in psychiatric patients. Am. J. Psychiatry 156, 181–189. Oquendo, M., Brent, D.A., Birmaher, B., Greenhill, L., Kolko, D., Stanley, B., Zelazny, J., Burke, A.K., Firinciogullari, S., Ellis, S.P., Mann, J.J., 2005. Posttraumatic stress
14
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
disorder comorbid with major depression: factors mediating the association with suicidal behavior. Am. J. Psychiatry 162, 560–566. Oquendo, M.A., Miller, J.M., Sublette, M.E., 2013. Neuroanatomical correlates of childhood sexual abuse: identifying biological substrates for environmental effects on clinical phenotypes. Am. J. Psychiatry 170, 574–577. Orr, S.P., Lasko, N.B., Macklin, M.L., Pineles, S.L., Chang, Y., Pitman, R.K., 2012. Predicting post-trauma stress symptoms from pre-trauma psychophysiologic reactivity, personality traits and measures of psychopathology. Biol. Mood Anxiety Disord. 2, 8. Ozer, E.J., Best, S.R., Lipsey, T.L., Weiss, D.S., 2003. Predictors of posttraumatic stress disorder and symptoms in adults: a meta-analysis. Psychol. Bull. 129, 52–73. Pechtel, P., Pizzagalli, D.A., 2010. Effects of early life stress on cognitive and affective function: an integrated review of human literature. Psychopharmacology 214, 55–70. Perroud, N., Jaussent, I., Guillaume, S., Bellivier, F., Baud, P., Jollant, F., Leboyer, M., Lewis, C.M., Malafosse, A., Courtet, P., 2010. COMT but not serotonin-related genes modulates the influence of childhood abuse on anger traits. Genes Brain Behav. 9, 193–202. Powers, A., Ressler, K.J., Bradley, R.G., 2009. The protective role of friendship on the effects of childhood abuse and depression. Depress. Anxiety 26, 46–53. Preisig, M., Fenton, B.T., Matthey, M.L., Berney, A., Ferrero, F., 1999. Diagnostic interview for genetic studies (DIGS): inter-rater and test-retest reliability of the French version. Eur. Arch. Psychiatry Clin. Neurosci. 249, 174–179. Rothman, K.J., 1990. Statistics in nonrandomized studies. Epidemiology 1, 417–418. Savitz, D.A., Olshan, A.F., 1998. Describing data requires no adjustment for multiple comparisons: a reply from Savitz and Olshan. Am. J. Epidemiol. 147, 813–814 (discussion 815). Scott, K.M., Smith, D.R., Ellis, P.M., 2010. Prospectively ascertained child maltreatment and its association with DSM-IV mental disorders in young adults. Arch. Gen. Psychiatry 67, 712–719. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured
diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl 20), S22–S33 (quiz34–quiz57). Shipman, K., Edwards, A., Brown, A., Swisher, L., Jennings, E., 2005. Managing emotion in a maltreating context: a pilot study examining child neglect. Child Abuse Negl. 29, 1015–1029. Silverman, M.M., Berman, A.L., Sanddal, N.D., O’Carroll, P.W., Joiner, T.E., 2007. Rebuilding the tower of babel: a revised nomenclature for the study of suicide and suicidal behaviors. Part 1: background, rationale, and methodology. Suicide Life Threat. Behav. 37, 248–263. Slama, F., Courtet, P., Golmard, J.L., Mathieu, F., Guillaume, S., Yon, L., Jollant, F., Misson, H., Jaussent, I., Leboyer, M., Belliver, F., 2009. Admixture analysis of age at first suicide attempt. J. Psychiatr. Res. 43, 895–900. Springer, K.W., Sheridan, J., Kuo, D., Carnes, M., 2007. Long-term physical and mental health consequences of childhood physical abuse: results from a large population-based sample of men and women. Child Abuse Negl. 31, 517–530. Teicher, M.H., Samson, J.A., 2013. Childhood maltreatment and psychopathology: a case for ecophenotypic variants as clinically and neurobiologically distinct subtypes. Am. J. Psychiatry 170, 1114–1133. van Harmelen, A.L., van Tol, M.J., Demenescu, L.R., van der Wee, N.J.A., Veltman, D.J., Aleman, A., van Buchem, M.A., Spinhoven, P., Penninx, B.W.J.H., Elzinga, B.M., 2013. Enhanced amygdala reactivity to emotional faces in adults reporting childhood emotional maltreatment. Soc. Cogn. Affect. Neurosci. 8, 362–369. Wiedemar, L., Schmid, J.P., Müller, J., Wittmann, L., Schnyder, U., Saner, H., von Känel, R., 2008. Prevalence and predictors of posttraumatic stress disorder in patients with acute myocardial infarction. Heart Lung 37, 113–121. Wilcox, H.C., Storr, C.L., Breslau, N., 2009. Posttraumatic stress disorder and suicide attempts in a community sample of urban American young adults. Arch. Gen. Psychiatry 66, 305–311. Xie, P., Kranzler, H.R., Poling, J., Stein, M.B., Anton, R.F., Brady, K., Weiss, R.D., Farrer, L., Gelernter, J., 2009. Interactive effect of stressful life events and the serotonin transporter 5-HTTLPR genotype on posttraumatic stress disorder diagnosis in 2 independent populations. Arch. Gen. Psychiatry 66, 1201–1209.
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Posttraumatic Stress Disorder following childhood abuse increases the severity of suicide attempts Jorge Lopez-Castroman a,b,c,n, Isabelle Jaussent a, Severine Beziat a, Sebastien Guillaume a,c,d, Enrique Baca-Garcia b,e, Emilie Olié a,c,d, Philippe Courtet a,c,d a
Inserm U1061, Ho^pital La Colombiere, Montpellier, France IIS-Fundacion Jimenez Diaz, Department of Psychiatry, CIBERSAM, Madrid, Spain c Department of Emergency Psychiatry, CHRU Montpellier, France d University of Montpellier 1, Montpellier, France e Department of Psychiatry at the New York State Psychiatric Institute and Columbia University, NY, USA b
art ic l e i nf o
a b s t r a c t
Article history: Received 28 February 2014 Received in revised form 1 July 2014 Accepted 7 August 2014 Available online 15 August 2014
Objective: Posttraumatic Stress Disorder (PTSD) and childhood abuse are both consistently associated with a higher risk for suicide attempts. We hypothesize that among patients reporting childhood abuse, PTSD diagnoses are correlated with an increased severity of suicidal features. Method: We investigated 726 adult patients who had attempted suicide. These participants were assessed on lifetime clinical diagnoses and childhood abuse. The association of PTSD and childhood abuse dimensions with age at first suicide attempt, number of suicide attempts, violent attempts, serious attempts and suicide intent was studied. An adjusted multinomial logistic regression was performed to ascertain if childhood abuse and PTSD increased the severity of the suicidal behavior when combined. Results: Several types of childhood abuse (emotional, physical and sexual abuse) when combined with a lifetime diagnosis of PTSD showed an increased risk for more suicide attempts, serious attempts, and a higher level of suicidal intent compared with the absence of any or both risk factors. Conclusion: The combination of PTSD and childhood abuse should be investigated in clinical settings due to an augmented risk for more severe suicidal behavior. & 2014 Elsevier B.V. All rights reserved.
keywords: Posttraumatic Stress Disorders Suicidal behavior Childhood abuse
1. Introduction The development of Posttraumatic Stress Disorder (PTSD) depends on the type and severity of trauma exposure, as well as on individual vulnerability (Davidson et al., 2004). In Europe, the lifetime prevalence of PTSD has been estimated at 1.9% (Alonso et al., 2004), although rates may largely vary between countries (Hauffa et al., 2011). When present, PTSD has been repeatedly associated with a high suicidal risk (Krysinska and Lester, 2010). In fact, PTSD diagnoses, but not trauma exposure, were recently associated with suicide attempts (SA) in a community sample of urban American young adults (Wilcox et al., 2009). This association between PTSD and subsequent SAs appears to be independent of other mental disorders (Wilcox et al., 2009; Krysinska and Lester, 2010). Moreover, among mental disorders, PTSD has been estimated to be the fourth-ranked contributor to SAs with a 6.3% population-attributable fraction (Bolton and Robinson, 2010). This
n Corresponding author at: Inserm U1061, Hôpital La Colombiere, Pavillon 42, 39 Av Charles Flauhault, BP 34493, 34093 Montpellier, France. Tel. þ33 499614560. E-mail address: [email protected] (J. Lopez-Castroman).
http://dx.doi.org/10.1016/j.jad.2014.08.010 0165-0327/& 2014 Elsevier B.V. All rights reserved.
figure provides an approximation of the incidence of SAs that is directly linked to PTSD diagnoses. Childhood abuse is a prevalent public health problem that causes significant physical, psychological and societal consequences (Lu et al., 2008) and is also strongly associated with later SAs (Lopez-Castroman et al., 2012). In high-income countries, physical abuse may affect 4–16% of children, while 15–30% of girls and 5–15% of boys may suffer sexual abuse (Gilbert et al., 2009). Moreover, childhood abuse is a powerful predictor of future PTSD, in children and adolescents (Kearney et al., 2009) or adults (Ozer et al., 2003; Cloitre et al., 2010), and later-life revictimization, particularly among sexually-abused women (Coid et al., 2001). Several studies have investigated the link between childhood abuse and PTSD. The adjusted odds ratio for a lifetime diagnosis of PTSD following childhood maltreatment has been estimated to be 4.86 (Scott et al., 2010). Among individuals with PTSD, childhood abuse is associated with more comorbid mood disorders and more severe symptomatology (Teicher and Samson, 2013). Besides, a diagnosis of PTSD is more likely if the abused subject experiences another traumatic event as an adult (Xie et al., 2009). Although the mechanisms that link PTSD and childhood abuse with SAs are yet unknown, they may share some pathophysiological pathways.
8
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Autonomic and hypothalamic–pituitary–adrenal axis responses are altered in both conditions (Davidson et al., 2004; Carpenter et al., 2007) and may lead subjects experiencing childhood abuse and PTSD towards emotion dysregulation and interpersonal difficulties (Cloitre et al., 2010). Prior vulnerability traits, such as impulsive aggression, may be associated with the development of PTSD after a traumatic experience (Oquendo et al., 2005), and similar traits have been associated with childhood abuse and its transmission within families (Lopez-Castroman et al., 2012). Impulsivity might actually facilitate SAs through habituation to pain and fear, which impulsive individuals apparently experience more often (Bender et al., 2011). Therefore, an increased risk of suicide among abused subjects with subsequent PTSD can be expected. In this study we aim to investigate if the combination of childhood abuse and PTSD increases the severity of suicidal behaviors. We assessed a large sample of suicide attempters to ascertain a history of childhood abuse and a lifetime diagnosis of PTSD. We then examined the subtypes of childhood abuse, (i.e., emotional neglect, physical neglect, emotional abuse, physical abuse, sexual abuse) as well as the effect of PTSD as independent risk factors for more severe suicidal behaviors using five different indexes. We hypothesized that the occurrence of PTSD after childhood abuse would be associated with more severe suicidal behaviors during the lifetime of the participants, especially when reporting physical or sexual abuse.
2. Materials and methods 2.1. Participants Study participants were identified from a cohort of suicide attempters (n ¼1941), consecutively hospitalized and survivors of a current SA in a specialized unit of the Montpellier University Hospital. SAs were defined as self-injury behaviors with a nonzero level of suicidal intent (Silverman et al., 2007). Patients were aged 18 years old and over, French speaking, and had all four biological grandparents originating from Western European countries (for genetic purposes). All participants completed and returned a consent form. Overall, 726 patients had completed a diagnostic evaluation (including PTSD and major depression) and a questionnaire on childhood trauma. Excluded subjects presented a lower educational level and were less often diagnosed with nonPTSD anxiety disorders (po 0.05 for all comparisons), but did not statistically differ on the other socio-demographic and clinical variables. Trained psychiatrists or psychologists interviewed all patients. The local research ethics committee approved this study (CPP Sud Mediterranée IV, CHU Montpellier, France). 2.2. Assessment Patients were evaluated after remission of a potential mood episode (i.e. a current Hamilton Depression Rating Scale score below 15) (Hamilton, 1960). Either the French version of the Diagnostic Interview for Genetics Studies (DIGS) (Preisig et al., 1999) or the Mini International Neuropsychiatric Interview (MINI) (Sheehan et al., 1998) was used to obtain Axis I DSM-IV diagnoses. Lifetime diagnoses were determined using a best-estimate procedure: the psychiatrist in charge of the patient's care assigned the diagnosis based on MINI or DIGS interview, medical records and, when available, information from relatives (Kosten and Rounsaville, 1992). Lifetime PTSD diagnoses were assigned following MINI criteria for current or past PTSD. The assessment of the history of childhood trauma was performed using the short version of the
Childhood Trauma Questionnaire (CTQ) (Bernstein and Fink, 1998). It is a 28-item self-report questionnaire that investigates retrospectively five dimensions of child maltreatment: emotional abuse, emotional neglect, physical abuse, physical neglect and sexual abuse. Cut-off scores have been set for each type of trauma at four levels of maltreatment: None, Low, Moderate and Severe. The different cut-offs have been shown to have good specificity and sensitivity (Bernstein and Fink, 1998). Childhood trauma was considered only with moderate or severe scores. The suicide assessment procedure was based on the Columbia Suicide History Form (CSHF, Mann et al., 1999) and the Section O of the DIGS. The procedure is a semi-structured interview with validated questionnaires to collect information about sociodemographic features and characteristics of the suicide attempts. These questionnaires elicit in-depth information about lifetime suicide attempts, including suicidal methods and triggers, with questions such as “How many times have you tried to kill yourself?” (DIGS) or “When was the first time he/she ever made an attempt?” (CSHF). A suicide attempt was defined as violent, according to Asberg et al.'s (1976) criteria, when the method of suicide attempt was hanging, use of firearms, jumping from heights, several deep cuts, car crash, burning, gas poisoning, drowning, electrocution, or jumping under a train. Those suicide attempts that required intensive care interventions were considered serious. Age at first attempt was defined as the age at which the patient first made a SA. Age at first attempt was assessed by the interviewer and then blindly rated by an independent psychiatrist according to medical case notes and interviews. Cut-off for early age at first SA was set according to a mathematical modelization made in a previous study. In that study, the clinical picture and the history of childhood abuse differed between subjects younger or older than 26 years of age (Hamilton, 1960; Slama et al., 2009). The number of SAs for the analyses was also categorized using 1–2 and 42 as cut-off (Preisig et al., 1999; Lopez-Castroman et al., 2011). We further characterized suicidal behavior using the French version of the Suicide Intent Scale (SIS) (Beck et al., 1979; Sheehan et al., 1998), a 15-item semi-structured rating scale yielding a global score that indicates the severity of the suicidal intent. For each subject, only the highest SIS score of all previous SA was analyzed. Two subscales of the SIS were studied: expected lethality and planning (Brezo et al., 2008). The cut-off for the analyses of SIS scores was established using the higher tertile (SIS score r19 or 4 19). 2.3. Statistical analysis Associations between suicidal indexes (age at first SA, number of SA, SIS score, serious suicide attempt and violent suicide attempt) and subject characteristics, PTSD diagnosis and CTQ dimensions were quantified with odds ratios (OR) and their 95% Confidence Intervals (CI). Sociodemographic, and clinical variables associated with outcome variables (at p o0.05) were included in logistic regression models to estimate adjusted ORs for PTSD diagnosis and CTQ dimensions. When appropriate (i.e., when PTSD diagnosis or CTQ dimensions were significantly associated with the suicidal indexes), the interaction terms were tested using Wald χ2 tests given by the logistic regression model. If the interaction was not significant, we studied the additive effects of PTSD diagnosis and CTQ dimensions dividing the sample into 4 groups for each outcome variable: 1) subjects who reported both the childhood trauma and PTSD; 2) subjects who reported the childhood trauma but were not diagnosed with PTSD; 3) PTSD subjects not reporting childhood trauma; and 4) subjects who denied both risk factors. Significance level was set at p o0.05. Given the exploratory nature of our study, multiple test adjustments were not made (Rothman, 1990; Savitz
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
and Olshan, 1998; Bender and Lange, 2001). Analyses were performed using SAS statistical software (version 9.2; SAS Inc, Cary, North Carolina).
3. Results 3.1. Sample description Median age of the sample was 39.6 years (range: 18.0–83.4). Most of the sample was composed of women (n ¼540, 74.4%) with intermediate or high educational level (n ¼560, 79.5%). Less than one-third of the sample made more than two suicide attempts (28.5%) and these were rarely violent (11.0%) or severe (14.0%). Approximately half of the sample (46.5%) attempted suicide for the first time by the age of 26. Family history of suicidal behavior was common (38.8%). PTSD was diagnosed in 14.2% of the sample (n ¼103). The most frequent lifetime diagnoses were anxiety disorders (78.3%) and major depression (70.1%), followed by bipolar disorder (26.9%). Finally, a large majority of patients reported at least one type of childhood abuse according to the CTQ (68.9%, n¼ 500). One in every three subjects reported 3 or more different types of abuse (30.3%, n ¼ 220). The most frequent type of childhood abuse was emotional neglect (45.3%, n ¼329) followed by emotional abuse (42.0%, n¼ 305). Physical abuse was reported by 168 subjects (23.2%), sexual abuse by 212 subjects (29.2%) and physical neglect by 203 subjects (28.0%). Among subjects with lifetime PTSD, 88.3% (n ¼91) reported at least one type of childhood abuse and 58.2% reported three or more types of abuse (n¼ 60). The repetition of SAs (42) was associated with young age at the first SA ( o26 years; chi2 ¼40.3, df ¼1, p o0.0001), high suicidal intent (SIS score 419; chi2 ¼7.3, df ¼1, p ¼0.007), serious SAs (chi2 ¼44.8; df ¼1; po 0.0001) and violent SAs (chi2 ¼5.0; df ¼ 1; p¼ 0.025). A high suicidal intent was also associated with a history of serious SAs (chi2 ¼23.7; df ¼1; p o0.0001) or violent SAs (chi2 ¼3.7; df ¼1; p ¼0.053). No other significant associations were found between the suicidal indexes. In order to facilitate their presentation, henceforward all results will be adjusted by variables significantly associated with at least one outcome variable (age at first SA, number of SA, SIS score, serious SA, violent SA): sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder. Detailed results appear in Table 1. 3.2. Age at first SA Sexual abuse was associated with a young age at first SA (o26 years). Subjects with PTSD or emotional abuse were also younger at the time of their first SA, but these findings were borderline significant after adjustment. No other CTQ dimension was significantly associated with age at first SA and we found no interactions between CTQ dimensions and PTSD regarding age at first SA (data not shown). Sexual abuse without lifetime PTSD was associated with a younger age at first SA (OR¼1.67; 95% CI¼ 1.11–2.49). Similar, but non-significant, adjusted OR for early onset of SA were obtained in patients with PTSD but no sexual abuse (OR¼2.02; 95% CI¼0.99–4.12) and patients with a combination of both risk factors (OR¼1.70; 95% CI¼ 0.94–3.06) (Table 2). 3.3. Number of SAs PTSD and all types of childhood abuse, with the exception of physical neglect, were associated with a greater number of SAs (42 SAs). As CTQ dimensions and PTSD did not interact with respect to the number of SAs (data not shown), we studied the additive effects of their combination. Compared with subjects reporting no PTSD or
9
Table 1 Description of the sample (n¼ 726). Variables
n
%
Sex (female) Age median (range) Educational level (secondary or higher) Lifetime smoking Suicide attempt characteristics Violent suicide attempt Serious suicide attempt Number of suicide attempts median (range) Number of suicide attempts 42 SIS total median (range) SIS total 419 Age at first suicide attempt median (range) Age at first suicide attempt r 26 Family history of suicidal behavior Family history of suicide Lifetime diagnoses Posttraumatic Stress Disorder (PTSD) Major depression Bipolar disorder Anxiety disorders Alcohol use disorders Substance use disorders Cranial trauma Eating disorder Childhood Trauma Questionnaire CTQ - emotional neglect CTQ - physical neglect CTQ - emotional abuse CTQ - physical abuse CTQ - sexual abuse No abuse 1 abuse subtype 2 abuse subtype 3 abuse subtype 4 abuse subtype 5 abuse subtype
540 39.6 560 490
74.4 (18.0–83.4) 79.5 68.9
80 101 2 201 16 195 28 340 270 130
11.0 14.0 (1–21) 28.5 (0–30) 27.5 (7–74) 46.5 38.8 19.0
103 509 194 509 186 117 91 126
14.2 70.1 26.9 78.3 25.7 16.1 12.8 17.5
329 203 305 168 212 226 161 119 104 74 42
45.3 28.0 42.0 23.1 29.2 31.1 22.2 16.4 14.3 10.2 5.8
emotional neglect, patients with PTSD only (OR¼ 2.39; 95% CI¼1.17–4.86), emotional neglect only (OR¼1.72; 95% CI¼1.18– 2.52) or a combination of both (OR¼ 2.09; 95% CI¼1.13–3.87) were more likely to make a greater number of SAs in the adjusted analyses. Regarding emotional abuse, increased odds of making more SAs were found among subjects with emotional abuse but no PTSD (OR¼ 1.57; 95% CI¼ 1.07–2.30) or with a combination of both (OR¼2.56; 95% CI¼1.44–4.56). Subjects with PTSD and physical abuse combined were also more likely to make a greater number of SAs (OR¼3.43; 95% CI¼1.73–6.79) than those without risk factors, but not those with either physical abuse or PTSD alone. Finally, sexual abuse alone (OR¼1.96; 95% CI¼ 1.29–2.97) or in combination with lifetime PTSD (OR¼2.22; 95% CI¼1.24–3.96) was also associated with a greater number of SAs compared to the absence of these risk factors (Table 3, Fig. 1).
3.4. Suicide Intent Scale High Suicide Intent Scale scores were associated with lifetime PTSD diagnosis and with emotional, physical and sexual abuse. PTSD and these CTQ dimensions did not interact with regards to the suicidal intent (data not shown). After adjustment, PTSD diagnosis alone or emotional, physical or sexual abuses alone were not significantly associated with an increased suicidal intent compared with the absence of both risk factors. However, when coexisting with a PTSD diagnosis, all three types of abuse were significantly associated with increased suicidal intent: emotional (OR¼ 3.74; 95% CI¼2.08–6.72), physical (OR¼3.70; 95% CI¼1.85–7.38) and sexual (OR¼3.40; 95% CI¼1.89–6.13) (Table 4, Fig. 2).
10
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Table 2 Age at first suicide attempt compared by PTSD diagnosis and dimensions of the CTQ. Age at first attempt
426 N¼ 386
Variables
n
%
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
41 173 106 144 85 90
10.62 44.82 27.46 37.31 22.02 23.32
62 155 96 160
226 16 119 25 280 16 65 25
a
r 26 N ¼ 336 OR [95% CI]a
p-valuea
18.45 46.13 28.57 47.62
1.58 1.02 1.07 1.33
0.05 0.90
82 121
24.40 36.01
1.17 [0.80;1.69] 1.57 [1.10;2.24]
58.55 4.15 30.83 6.48
157 19 117 43
46.73 5.65 34.82 12.80
1 1.52 [0.72;3.19] 1.28 [0.90;1.81] 1.89 [1.05;3.40]
0.12
72.54 4.15 16.84 6.48
190 25 84 37
56.55 7.44 25.00 11.01
1 2.02 [0.99;4.12] 1.67 [1.11;2.49] 1.70 [0.94;3.06]
0.02
[0.99;2.51] [0.74;1.40] [0.75; 1.53] [0.96;1.84]
0.72 0.08 0.42 0.01
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Table 3 Number of suicide attempts compared by PTSD diagnosis and dimensions of the CTQ. Number of suicide attempts
SA¼ 1–2 N ¼ 515
Variables
n
%
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional neglect No–No PTSD–no emotional neglect No PTSD–emotional neglect Yes–Yes Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
59 215 144 190 104 121
11.46 41.75 27.96 36.89 20.19 23.50
43 113 57 112 64 90
275 25 181 34
53.40 4.85 35.15 6.60
300 25 156 34
a
SA42 N ¼205
OR [95% CI]a
p-valuea
20.98 55.12 27.80 54.63 31.22 43.90
1.70 1.60 0.87 1.70 1.68 1.90
0.03 0.008 0.49 0.003 0.009 0.0006
75 17 87 26
36.59 8.29 42.44 12.68
1 2.39 [1.17;4.86] 1.72 [1.18;2.52] 2.09 [1.13;3.87]
0.005
58.25 4.85 30.29 6.60
84 9 78 34
40.98 4.39 38.05 16.59
1 1.32 [0.58;3.01] 1.57 [1.07;2.30] 2.56 [1.44;4.56]
0.007
371 40 85 19
72.04 7.77 16.50 3.69
122 19 40 24
59.51 9.27 19.51 11.71
1 1.10 [0.58;2.06] 1.30 [0.83;2.05] 3.43 [1.73;6.79]
0.005
369 25 87 34
71.65 4.85 16.89 6.60
100 15 62 28
48.78 7.32 30.24 13.66
1 1.92 [0.93;3.99] 1.96 [1.29;2.97] 2.22 [1.24;3.96]
0.002
[1.07;2.72] [1.13;2.26] [0.59;1.29] [1.20;2.42] [1.14;2.47] [1.31;2.73]
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 1. Association of number of suicide attempts with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for greater number of suicide attempts ( 4 2) with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EN¼ emotional neglect (dark square); EA¼ emotional abuse (white square); PA ¼ physical abuse (triangle); SA¼sexual abuse (diamond).
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
11
Table 4 Suicide Intent Scale score compared by PTSD diagnosis and dimensions of the CTQ. Suicide Intent Scale score
r19 N ¼ 513
Variables
n
%
PTSD lifetime (Yes vs No) Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD–no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
60 222 135 202 107 137
11.70 43.27 26.32 39.38 20.86 26.71
40 94 63 94 54 67
286 25 167 35
55.75 4.87 32.55 6.82
366 40 87 20 347 29 106 31
a
419 N ¼195 OR [95% CI]a
p-valuea
20.51 48.21 32.31 48.21 27.69 34.36
2.27 1.25 1.45 1.49 1.51 1.47
[1.42;3.64] [0.89;1.77] [1.00;2.11] [1.05;2.12] [1.03;2.24] [1.01;2.15]
0.0006 0.20 0.05 0.02 0.04 0.05
93 8 62 32
47.69 4.10 31.79 16.41
1 1.00 [0.43;2.33] 1.15 [0.78;1.70] 3.74 [2.08;6.72]
0.0002
71.35 7.80 16.96 3.90
121 20 34 20
62.05 10.26 17.44 10.26
1 1.75 [0.96;3.19] 1.22 [0.77;1.92] 3.70 [1.85;7.38]
0.002
67.64 5.65 20.66 6.04
117 11 38 29
60.00 5.64 19.49 14.87
1 1.22 [0.58;2.59] 1.04 [0.66;1.63] 3.40 [1.89;6.13]
0.0006
n
%
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 2. Association of high suicide intent with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for higher suicidal intent ( 419) with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EA ¼ emotional abuse (white square); PA ¼physical abuse (triangle); SA¼ sexual abuse (diamond).
3.5. Violent suicide attempt
4. Discussion
Violent suicide attempters were not associated with lifetime PTSD diagnosis or any CTQ dimension.
PTSD may be caused by childhood abuse or by other traumatic experiences, and both are powerful risk factors for suicidal behaviors. In this study we have tried to understand the relationship between these two exposures among suicide attempters by investigating the putative effect of their combination in a number of suicidal indexes. Some dimensions of childhood abuse were particularly associated with severity features of the suicide attempts when coexisting with a lifetime PTSD. Emotional neglect, either alone or in combination with lifetime PTSD, was associated with more SAs when compared with subjects without these conditions. However, no other significant association was found between emotional neglect or physical neglect and PTSD with regard to the severity of suicidal behavior. On the other hand, subjects with emotional abuse and PTSD presented higher odds of making serious attempts
3.6. Serious suicide attempt After adjustment, serious suicide attempts were still associated with lifetime PTSD and physical abuse but significant associations with sexual abuse and emotional abuse were lost. PTSD and these CTQ dimensions did not interact with regards to severe suicide attempts (data not shown). Subjects reporting PTSD and emotional abuse (OR ¼2.66; 95% CI ¼ 1.33–5.32), physical abuse (OR ¼4.14; 95% CI ¼ 1.99–8.61) or sexual abuse (OR ¼2.79; 95% CI ¼ 1.44–5.41) were more likely to make serious suicide attempts than subjects presenting only one or none of these risk factors (Table 5, Fig. 3).
12
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
Table 5 Serious suicide attempters compared with non-serious attempters by PTSD diagnosis and dimensions of the CTQ. Serious suicide attempts
No N¼ 619
Variables
n
%
n
%
PTSD lifetime Emotional neglect (Yes vs No) Physical neglect (Yes vs No) Emotional abuse (Yes vs No) Physical abuse (Yes vs No) Sexual abuse (Yes vs No) Combination of PTSD–emotional abuse No–No PTSD– no emotional abuse No PTSD–emotional abuse Yes–Yes Combination of PTSD–physical abuse No–No PTSD–no physical abuse No PTSD–physical abuse Yes–Yes Combination of PTSD–sexual abuse No–No PTSD–no sexual abuse No PTSD–sexual abuse Yes–Yes
79 279 170 249 133 171
12.76 45.07 27.46 40.23 21.49 27.63
23 48 32 53 35 40
340 30 200 49
54.93 4.85 32.31 7.92
435 51 105 28 413 35 127 44
a
Yes N¼ 101 OR [95% CI]a
p-valuea
22.77 47.52 31.68 52.48 34.65 39.60
1.86 0.98 1.24 1.50 2.08 1.55
0.03 0.92 0.38 0.07 0.002 0.06
43 5 35 18
42.57 4.95 34.65 17.82
1 1.27 [0.46;3.50] 1.30 [0.79;2.15] 2.66 [1.33;5.32]
0.05
70.27 8.24 16.96 4.52
58 8 20 15
57.43 7.92 19.80 14.85
1 1.03 [0.44;2.44] 1.57 [0.89;2.75] 4.14 [1.99;8.61]
0.001
66.72 5.65 20.52 7.11
56 5 22 18
55.45 4.95 21.78 17.82
1 0.83 [0.28;2.49] 1.11 [0.63;1.98] 2.79 [1.44;5.41]
0.02
[1.06;3.28] [0.63;1.52] [0.76;2.01] [0.96;2.35] [1.30;3.32] [0.97;2.49]
Adjustment for sex, educational level, smoking, eating disorders, alcohol or substance use disorders, major depression and bipolar disorder.
Fig. 3. Association of serious suicide attempts with combinations of childhood abuse and PTSD. Legend: Odds Ratios and 95% Confidence Intervals for serious suicide attempts with different combinations of the Childhood Trauma Questionnaire dimensions and the diagnosis of Posttraumatic Stress Disorder (PTSD). The horizontal line represents the reference group, composed by subjects that had no PTSD and did not report the corresponding type of childhood abuse. Shapes indicate childhood abuse subtype. EA ¼emotional abuse (white square); PA ¼ physical abuse (triangle); SA¼ sexual abuse (diamond).
and of attempting suicide earlier, more times, and with higher intent than those with only one or none of these conditions. The combination of physical abuse and PTSD also presented high odds ratios (OR 43) for a greater number of SAs, serious attempts and higher suicidal intent, which was not present with physical abuse or PTSD alone. Finally, the combination of sexual abuse and PTSD diagnosis was not associated with earlier SAs in our sample but it was associated with more attempts and specifically with serious attempts and more suicidal intent (see Figs. 2 and 3). These findings support previous studies that emphasized the relevance of the type and severity of childhood maltreatment to assess the risk of suicidal behaviors in later life (Lopez-Castroman et al., 2013). Indeed, the mental health outcomes and even the structural changes on the brain of abused children may depend on the nature of the abuse itself (Oquendo et al., 2013). Interestingly, “commission” trauma (emotional, physical or sexual abuse) conveyed more severe suicidal behaviors than “omission” trauma (emotional and physical neglect) in our study. If the subtypes of early life adversity define the severity of suicidal outcomes, then interventions, such
as the enhancement of perceived social support (Powers et al., 2009), may need to be personalized for victims of abuse. Childhood maltreatment may alter developmental processes related to cognitive and affective functioning (Mann et al., 1999; Shipman et al., 2005; Pechtel and Pizzagalli, 2010) and is frequently experienced as a recurrent trauma adopting different forms (Dong et al., 2004; Gilbert et al., 2009). It may also elicit a state of learned helplessness and conditioned fear, which is associated with traumarelated depression symptoms in animal models (Hammack et al., 2012), and an increased startle response independent of the presence of PTSD symptoms (Jovanovic et al., 2009). Thus, childhood trauma may facilitate the re-experience of trauma and PTSD symptoms through helplessness (Wiedemar et al., 2008), or other psychophysiological and personality variables (Orr et al., 2012). Actually, sexually abused children are often diagnosed with PTSD in their adult life (Coid et al., 2001; Davidson et al., 2004) and PTSD emerging from childhood trauma appears to be associated with multiple other forms of trauma during the lifetime (Cloitre et al., 2010). Therefore, subjects experiencing childhood abuse and later-life PTSD symptoms may be more
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
exposed to emotional difficulties, for instance in anger management and interpersonal relationships, and high reactivity (Springer et al., 2007; Cloitre et al., 2010; Perroud et al., 2010; van Harmelen et al., 2013). An excessive reactivity to life experiences together with diminished social support and inefficient emotional processing may explain the association with higher number of SAs, serious attempts and higher suicidal intent that we found among patients reporting childhood abuse and later-life PTSD symptoms. In this study we have examined a large cohort of consecutively admitted suicide attempters, whose features were thoroughly assessed. Moreover, the results were adjusted for other risk factors such as alcohol or substance use disorders. However, some limitations should be contemplated. The time lapse from childhood maltreatment may convey a recall bias, although evidence supports the use of this kind of retrospective reports (Hardt and Rutter, 2004), and this information was not confirmed by other sources. The characteristics of CTQ prevented us from considering other relevant variables of the childhood abuse, such as duration or role of the abuser. Other adverse experiences, different from childhood abuse, were not registered in our sample and therefore we could not explore suicidality in other types of early trauma. Finally, we could not analyze the temporal relationship between traumatic experiences, PTSD diagnoses and suicidal behavior. Although PTSD diagnoses always preceded the last suicide attempt, i.e. the moment when the subjects were assessed, we could not distinguish if PTSD diagnoses arose from childhood abuse or other trauma or if PTSD preceded the first SA. Such a distinction may have allowed us to improve our understanding of the relationship between these conditions. We studied here the features of suicidal behaviors after particularly traumatic life events, an issue that is often avoided in clinical interviews. Even among patients with PTSD or childhood abuse history there is frequently a tendency to deny or minimize their problems (Goodman et al., 2003; Davidson et al., 2004). However, PTSD and childhood abuse, particularly when combined, are associated with more severe suicidal behavior and higher suicidal intent. Future studies are needed to investigate the risk of completed suicide when this combination is present. Clinicians should remember that effective treatments are available (Cloitre et al., 2010), and investigate PTSD symptoms and childhood abuse history when dealing with suicidal patients.
Role of funding source Dr. Lopez-Castroman was supported by a FondaMental Foundation research grant. This study was financially supported by Programme Hospitalier de Recherche Clinique (CHU of Montpellier—PHRC UF 7653), and Agence Nationale NEURO2007 de la Recherche (NEURO2007—GENESIS). These institutions had no further role in the conception, design or redaction of the manuscript.
Conflict of interest None.
Acknowledgments Dr. Lopez-Castroman was supported by a FondaMental Foundation research grant (grant no. R11075FF). This study was financially supported by the Programme Hospitalier de Recherche Clinique (CHU of Montpellier—PHRC UF 7653), and the Agence Nationale de la Recherche (NEURO2007—GENESIS) (grant no. ANR-07NEURO-013-01). We thank Rosa Nunes and Catherine Genty for editorial assistance.
References Alonso, J., Angermeyer, M.C., Bernert, S., Bruffaerts, R., Brugha, T.S., Bryson, H., de Girolamo, G., Graaf, R., Demyttenaere, K., Gasquet, I., Haro, J.M., Katz, S.J., Kessler, R.C., Kovess, V., Lépine, J.P., Ormel, J., Polidori, G., Russo, l.J., Vilagut, G., Almansa, J., Arbabzadeh-Bouchez, S., Autonell, J., Bernal, M., Buist-Bouwman, M.A., Codony, M., Domingo-Salvany, A., Ferrer, M., Joo, S.S., Martínez-Alonso, M., Matschinger, H., Mazzi, F., Morgan, Z., Morosini, P., Palacín, C., Romera, B., Taub,
13
N., Vollebergh, W.A., ESEMeD/MHEDEA 2000 Investigators, European Study of the Epidemiology of Mental Disorders (ESEMeD) Project, 2004. Disability and quality of life impact of mental disorders in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr. Scand. Suppl. 420, 38–46. Asberg, M., Träskman, L., Thorén, P., 1976. 5-HIAA in the cerebrospinal fluid. A biochemical suicide predictor? Arch. Gen. Psychiatry 33, 1193–1197. Beck, A.T., Kovacs, M., Weissman, A., 1979. Assessment of suicidal intention: the scale for suicide ideation. J. Consult. Clin. Psychol. 47, 343–352. Bender, R., Lange, S., 2001. Adjusting for multiple testing—when and how? J. Clin. Epidemiol. 54, 343–349. Bender, T.W., Gordon, K.H., Bresin, K., Joiner, T.E., 2011. Impulsivity and suicidality: the mediating role of painful and provocative experiences. J. Affect. Disord. 129, 301–307. Bernstein, D.P., Fink, L., 1998. Childhood Trauma Questionnaire: A Retrospective Self-Report Manual. The Psychological Corporation, San Antonio, TX. Bolton, J.M., Robinson, J., 2010. Population-attributable fractions of Axis I and Axis II mental disorders for suicide attempts: findings from a representative sample of the adult, noninstitutionalized US population. Am. J. Public Health 100, 2473–2480. Brezo, J., Paris, J., Hébert, M., Vitaro, F., Tremblay, R., Turecki, G., 2008. Broad and narrow personality traits as markers of one-time and repeated suicide attempts: a population-based study. BMC Psychiatry 8, 15. Carpenter, L.L., Carvalho, J.P., Tyrka, A.R., Wier, L.M., Mello, A.F., Mello, M.F., Anderson, G.M., Wilkinson, C.W., Price, L.H., 2007. Decreased adrenocorticotropic hormone and cortisol responses to stress in healthy adults reporting significant childhood maltreatment. Biol. Psychiatry 62, 1080–1087. Cloitre, M., Stovall-McClough, K.C., Nooner, K., Zorbas, P., Cherry, S., Jackson, C.L., Gan, W., Petkova, E., 2010. Treatment for PTSD related to childhood abuse: a randomized controlled trial. Am. J. Psychiatry 167, 915–924. Coid, J., Petruckevitch, A., Feder, G., Chung, W., Richardson, J., Moorey, S., 2001. Relation between childhood sexual and physical abuse and risk of revictimisation in women: a cross-sectional survey. Lancet 358, 450–454. Davidson, J.R.T., Stein, D.J., Shalev, A.Y., Yehuda, R., 2004. Posttraumatic stress disorder: acquisition, recognition, course, and treatment. J. Neuropsychiatry Clin. Neurosci. 16, 135–147. Dong, M., Anda, R.F., Felitti, V.J., Dube, S.R., Williamson, D.F., Thompson, T.J., Loo, C.M., Giles, W.H., 2004. The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction. Child Abuse Negl. 28, 771–784. Gilbert, R., Widom, C.S., Browne, K., Fergusson, D., Webb, E., Janson, S., 2009. Burden and consequences of child maltreatment in high-income countries. Lancet 373, 68–81. Goodman, G.S., Ghetti, S., Quas, J.A., Edelstein, R.S., Alexander, K.W., Redlich, A.D., Cordon, I.M., Jones, D.P.H., 2003. A prospective study of memory for child sexual abuse: new findings relevant to the repressed-memory controversy. Psychol. Sci. 14, 113–118. Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56–62. Hammack, S.E., Cooper, M.A., Lezak, K.R., 2012. Overlapping neurobiology of learned helplessness and conditioned defeat: implications for PTSD and mood disorders. Neuropharmacology 62, 565–575. Hardt, J., Rutter, M., 2004. Validity of adult retrospective reports of adverse childhood experiences: review of the evidence. J. Child Psychol. Psychiatry 45, 260–273. Hauffa, R., Rief, W., Brähler, E., Martin, A., Mewes, R., Glaesmer, H., 2011. Lifetime traumatic experiences and posttraumatic stress disorder in the German population. J. Nerv. Ment. Disord. 199, 934–939. Jovanovic, T., Blanding, N.Q., Norrholm, S.D., Duncan, E., Bradley, B., Ressler, K.J., 2009. Childhood abuse is associated with increased startle reactivity in adulthood. Depress. Anxiety 26, 1018–1026. Kearney, C.A., Wechsler, A., Kaur, H., Lemos-Miller, A., 2009. Posttraumatic stress disorder in maltreated youth: a review of contemporary research and thought. Clin. Child Fam. Psychol. Rev. 13, 46–76. Kosten, T.A., Rounsaville, B.J., 1992. Sensitivity of psychiatric diagnosis based on the best estimate procedure. Am. J. Psychiatry 149, 1225–1227. Krysinska, K., Lester, D., 2010. Post-traumatic stress disorder and suicide risk: a systematic review. Arch. Suicide Res. 14, 1–23. Lopez-Castroman, J., Jaussent, I., Beziat, S., Genty, C., Olié, E., de Leon-Martinez, V., Baca-García, E., Malafosse, A., Courtet, P., Guillaume, S., 2012. Suicidal phenotypes associated with family history of suicidal behavior and early traumatic experiences. J. Affect. Disord. 142, 193–199. Lopez-Castroman, J., Melhem, N., Birmaher, B., Greenhill, L., Kolko, D., Stanley, B., Zelazny, J., Brodsky, B., García-Nieto, R., Burke, A.K., Mann, J.J., Brent, D.A., Oquendo, M.A., 2013. Early childhood sexual abuse increases suicidal intent. World Psychiatry 12, 149–154. Lopez-Castroman, J., Perez-Rodriguez Mde, L.M., Jaussent, I., Alegria, A.A., ArtesRodriguez, A., Freed, P., Guillaume, S., Jollant, F., Leiva-Murillo, J.M., Malafosse, A., Oquendo, M.A., de Prado-Cumplido, M., Saiz-Ruiz, J., Baca-Garcia, E., Courtet, P., European Research Consortium for Suicide (EURECA), 2011. Distinguishing the relevant features of frequent suicide attempters. J. Psychiatr. Res. 45, 619–625. Lu, W., Mueser, K.T., Rosenberg, S.D., Jankowski, M.K., 2008. Correlates of adverse childhood experiences among adults with severe mood disorders. Psychiatr. Serv. 59, 1018–1026. Mann, J.J., Waternaux, C., Haas, G.L., Malone, K.M., 1999. Toward a clinical model of suicidal behavior in psychiatric patients. Am. J. Psychiatry 156, 181–189. Oquendo, M., Brent, D.A., Birmaher, B., Greenhill, L., Kolko, D., Stanley, B., Zelazny, J., Burke, A.K., Firinciogullari, S., Ellis, S.P., Mann, J.J., 2005. Posttraumatic stress
14
J. Lopez-Castroman et al. / Journal of Affective Disorders 170 (2015) 7–14
disorder comorbid with major depression: factors mediating the association with suicidal behavior. Am. J. Psychiatry 162, 560–566. Oquendo, M.A., Miller, J.M., Sublette, M.E., 2013. Neuroanatomical correlates of childhood sexual abuse: identifying biological substrates for environmental effects on clinical phenotypes. Am. J. Psychiatry 170, 574–577. Orr, S.P., Lasko, N.B., Macklin, M.L., Pineles, S.L., Chang, Y., Pitman, R.K., 2012. Predicting post-trauma stress symptoms from pre-trauma psychophysiologic reactivity, personality traits and measures of psychopathology. Biol. Mood Anxiety Disord. 2, 8. Ozer, E.J., Best, S.R., Lipsey, T.L., Weiss, D.S., 2003. Predictors of posttraumatic stress disorder and symptoms in adults: a meta-analysis. Psychol. Bull. 129, 52–73. Pechtel, P., Pizzagalli, D.A., 2010. Effects of early life stress on cognitive and affective function: an integrated review of human literature. Psychopharmacology 214, 55–70. Perroud, N., Jaussent, I., Guillaume, S., Bellivier, F., Baud, P., Jollant, F., Leboyer, M., Lewis, C.M., Malafosse, A., Courtet, P., 2010. COMT but not serotonin-related genes modulates the influence of childhood abuse on anger traits. Genes Brain Behav. 9, 193–202. Powers, A., Ressler, K.J., Bradley, R.G., 2009. The protective role of friendship on the effects of childhood abuse and depression. Depress. Anxiety 26, 46–53. Preisig, M., Fenton, B.T., Matthey, M.L., Berney, A., Ferrero, F., 1999. Diagnostic interview for genetic studies (DIGS): inter-rater and test-retest reliability of the French version. Eur. Arch. Psychiatry Clin. Neurosci. 249, 174–179. Rothman, K.J., 1990. Statistics in nonrandomized studies. Epidemiology 1, 417–418. Savitz, D.A., Olshan, A.F., 1998. Describing data requires no adjustment for multiple comparisons: a reply from Savitz and Olshan. Am. J. Epidemiol. 147, 813–814 (discussion 815). Scott, K.M., Smith, D.R., Ellis, P.M., 2010. Prospectively ascertained child maltreatment and its association with DSM-IV mental disorders in young adults. Arch. Gen. Psychiatry 67, 712–719. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured
diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl 20), S22–S33 (quiz34–quiz57). Shipman, K., Edwards, A., Brown, A., Swisher, L., Jennings, E., 2005. Managing emotion in a maltreating context: a pilot study examining child neglect. Child Abuse Negl. 29, 1015–1029. Silverman, M.M., Berman, A.L., Sanddal, N.D., O’Carroll, P.W., Joiner, T.E., 2007. Rebuilding the tower of babel: a revised nomenclature for the study of suicide and suicidal behaviors. Part 1: background, rationale, and methodology. Suicide Life Threat. Behav. 37, 248–263. Slama, F., Courtet, P., Golmard, J.L., Mathieu, F., Guillaume, S., Yon, L., Jollant, F., Misson, H., Jaussent, I., Leboyer, M., Belliver, F., 2009. Admixture analysis of age at first suicide attempt. J. Psychiatr. Res. 43, 895–900. Springer, K.W., Sheridan, J., Kuo, D., Carnes, M., 2007. Long-term physical and mental health consequences of childhood physical abuse: results from a large population-based sample of men and women. Child Abuse Negl. 31, 517–530. Teicher, M.H., Samson, J.A., 2013. Childhood maltreatment and psychopathology: a case for ecophenotypic variants as clinically and neurobiologically distinct subtypes. Am. J. Psychiatry 170, 1114–1133. van Harmelen, A.L., van Tol, M.J., Demenescu, L.R., van der Wee, N.J.A., Veltman, D.J., Aleman, A., van Buchem, M.A., Spinhoven, P., Penninx, B.W.J.H., Elzinga, B.M., 2013. Enhanced amygdala reactivity to emotional faces in adults reporting childhood emotional maltreatment. Soc. Cogn. Affect. Neurosci. 8, 362–369. Wiedemar, L., Schmid, J.P., Müller, J., Wittmann, L., Schnyder, U., Saner, H., von Känel, R., 2008. Prevalence and predictors of posttraumatic stress disorder in patients with acute myocardial infarction. Heart Lung 37, 113–121. Wilcox, H.C., Storr, C.L., Breslau, N., 2009. Posttraumatic stress disorder and suicide attempts in a community sample of urban American young adults. Arch. Gen. Psychiatry 66, 305–311. Xie, P., Kranzler, H.R., Poling, J., Stein, M.B., Anton, R.F., Brady, K., Weiss, R.D., Farrer, L., Gelernter, J., 2009. Interactive effect of stressful life events and the serotonin transporter 5-HTTLPR genotype on posttraumatic stress disorder diagnosis in 2 independent populations. Arch. Gen. Psychiatry 66, 1201–1209.