Available online at www.sciencedirect.com
ScienceDirect Comprehensive Psychiatry 55 (2014) 370 – 373 www.elsevier.com/locate/comppsych
Cognitive traits in inpatient adolescents with and without prior suicide attempts and non-suicidal self-injury Shlomit Tsafrir a , Elena Chubarov a , Gal Shoval a, b , Mali Levi a , Eitan Nahshoni b , Motti Ratmansky c , Abraham Weizman a, b, d , Gil Zalsman a, b, e,⁎ a
Child and Adolescent Division, Geha Mental Health Center, Petah Tikva, Israel b Sackler Faculty of Medicine, Tel Aviv University, Israel c Pain Rehabilitation Unit, Loewenstein Rehabilitation Hospital, Raanana, Israel d Research Unit, Geha Mental Health Center and Felsenstein Medical Research Center, Petah Tikva, Israel e Molecular Imaging and Psychopathology Division, Psychaitry DepartmentNY, USA
Abstract Objective: Establishing a psychiatric diagnosis and assessing suicidal tendencies is often a challenging task particularly in the early stages of an illness. Cognitive impairments characterize different psychiatric entities, but there is no known specific cognitive deficit profile that could help the clinician in achieving the diagnostic task. This study’s aim was to establish a cognitive profile and test its ability to differentiate psychiatric inpatient subgroups, in terms of suicidal risk and diagnosis. The sample constituted of 76 consecutive admissions to the psychiatric adolescent day-care unit, who were admitted for any diagnosis. Assessment included full psychiatric interview and cognitive evaluation, using the COGNISTAT test. Results: Of the 76 participants, 58% reported having suicidal ideation and 29% reported a prior attempted suicide. Subjects who had a prior suicide attempt had better orientation and attention scores in the COGNISTAT. Other cognitive domains did not differentiate between groups or diagnoses. Conclusion: These preliminary results suggest a significant association between specific cognitive characteristics and suicidal behavior in adolescents. Those cognitive characteristics might prove clinically useful in the assessment of suicide risk. Further study is needed to establish this association and generalize the conclusion to different populations. © 2014 Elsevier Inc. All rights reserved.
1. Introduction In the past it was a common notion that schizophrenia and bipolar affective disorder differ in regard to the cognitive impairment and natural history of the disease [1]. This dichotomy is probably inaccurate and affective disordered patients, similar to schizophrenic patients, has difficulties in daily occupational and social function, partly attributed to cognitive impairments [2].
Disclosure: All authors declare that they have no conflict of interest whatsoever with regard to this study. ⁎ Corresponding author. Child and Adolescent Division Geha Mental Health Center, Tel Aviv University 1 Helsinki St, P.O. Box 102 Petah Tikva 49100 Israel. Tel.: +972 3 925 8205; fax: +972 3 924 1041. E-mail address: [email protected] (G. Zalsman). 0010-440X/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.comppsych.2013.08.009
Those cognitive deficits characterize psychiatric population, and are of great influence on quality of life and daily functioning. A specific cognitive profile and its relevance to the psychiatric evaluation process is not well established [3,4]. Patients with schizophrenia suffer cognitive deficit early in the course of illness, the deficit is usually permanent, not affected by pharmacological intervention and is found to be general, but most prominently affect the memory, attention and executive functions [5]. Those deficits are correlated with complex prognostic outcome variables [6]. Patients who suffer from affective disorder have similar cognitive deficits, especially in the domains of attention, memory and learning, so that the cognitive profile might not differentiate those diagnoses [7,8].
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
371
Adolescent suicide is a troubling phenomenon with high psychiatric comorbidity rates and worldwide presence. Suicide is the second leading cause of death in children and adolescents in Israel with a frequency of 8:100,000, according to the data from the Israeli Ministry of Health. Early assessment of suicidality in adolescents is, therefore, critical. Specific cognitive dysfunction that distinguishes individuals who has a suicidal tendency is not well established, but particular deficits in the decision making process and attention are typically present [9]. Keilp et al. described cognitive deficits in memory and attention in depressed patients, and a specific, additional deficit in executive function in depressed patients who had a prior lethal suicide attempt [10]. The goals of the current study were to determine whether a cognitive profile could differentiate between different subgroups of psychiatric patients. The study focused on differentiating subgroups of patients in terms of suicidal tendencies and diagnostic classification, in a sample of adolescent inpatients.
psychiatric clinical situations [12,13]. Deficits in specific domains characterize specific patients populations, and is correlated with level of function [14,15]. The study was approved by the Geha Mental Health Center Review Board that exempted us from having an informed consent due to the retrospective file review nature of the study.
2. Methods
3. Results
2.1. Participants Seventy six adolescents consecutively hospitalized at the adolescent day-care unit, in a university affiliated mental health center, and were recruited for the current study. Diagnosis of all participants were based on DSM-IV criteria (American Psychiatric Association, 1994) [11] using data that had been collected during clinical interviews, daily observations and clinical team discussions. Subjects were excluded from the study if they did not undergo a cognitive evaluation during their hospitalization, for any reason. 2.2. Materials and procedures Adolescents admitted to the day care unit undergo a full psychiatric interview by a child and adolescent psychiatrist, and additional diagnostic and evaluation procedures. Medical records of adolescents who were admitted to the day care unit, and underwent cognitive evaluation as part of the assessment procedure, were scanned and socio-demographic, medical and psychiatric data was abstracted. Cognitive and functional evaluation is done routinely, by an occupational therapist, as a part of the assessment process. The assessment includes the COGNISTAT, a cognitive status assessment tool. The COGNISTAT assess five major domains of cognitive function: language, construction, memory, calculation and reasoning. Level of consciousness, orientation and attention are examined separately. The test generates a profile representing the abilities and limitations of the subject. The COGNISTAT protocol was standardized in normal functioning subjects, at different age groups, and it is used to assess cognitive function in a variety of neuro-
2.3. Statistical analysis SPSS version 17 was used in the statistical analysis. Student T test and one-way analysis of variance were performed as appropriate, to evaluate association between major diagnosis and suicidal risk (independent variables) with cognitive domains as measured by the COGNISTAT (dependant variable). All tests were two-tailed. Levene tests were conducted for equivalence of variance in the group compared, and when proved significant, equal variances were not assumed. Bonferroni adjustment was used to correct for multiple testing.
Seventy six adolescents admitted to the day care unit were recruited to the study. Their ages varied between 12 and 20 years, with an average age of 14.6. The male to female ratio was 39:37. Forty eight (63%) patients were admitted for the first time. The subjects had been diagnosed as suffering from a diverse variety of psychiatric disorders, according to the DSM-IV classification[16]. Twenty (26%) had a main diagnosis of schizophrenia, seventeen (22%) had a main diagnosis of major depressive episode, fifteen (20%) had a main diagnosis of borderline personality disorder and twenty four (31%) had another, less common, main diagnosis. In addition, some had two or three co-morbid disorders (data not shown). Forty four (58%) patients reported having suicidal ideation and twenty two (29%) reported a prior attempted suicide. Twenty three (30%) self-harmed; with no suicide intent. COGNISTAT was administered to all the patients, as part of the routine assessment procedure. The effect of level of function, defined as a COGNISTAT grade in each domain tested, on diagnosis, suicidal risk and non-suicidal self injury was tested statistically. There was no difference on any cognitive domain in the four diagnoses groups (data not shown). The difference between the comprehension domain scores was minimally, but significantly, better in the selfharmed subjects (Table 1). Orientation and attention domains had a slightly better score in the subjects who had a prior suicide attempt. This finding was also significant (Table 2). There was no significant difference on any cognitive domain in the subjects who reported having suicidal ideation (data not shown).
372
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
4. Discussion This study used COGNISTAT test to assess cognitive profile in 76 adolescent inpatients, suicidal and not suicidal. In this sample patients who reported a prior suicide attempt had better orientation and attention, compared to nonsuicidal patients. Patients, who self-harmed, with no suicidal intent, had slightly better scores in the orientation and comprehension domains. These patients had a cognitive profile similar to the patients who reported prior suicide attempt. This difference was smaller when looking at subjects who reported suicidal ideation vs. those who didn't and subjects who attempted suicide while hospitalized vs. non suicidal subjects in the COGNISTAT profile. We did not find an association between a DSM-IV diagnosis and a specific cognitive profile in the COGNISTAT. This finding highlights the importance of distinguishing between suicidal ideation and actual suicide risk[17,18]. The different characteristics of suicide attempters and suicide completers are well established in the scientific literature[19] and the population of ideators is probably a third different population. Behaviors of non-suicidal self injury and suicide attempt are found often in similar patient populations: female, often usage of non-violent suicide methods, and high frequency of specific personality traits, including impulsivity, hostility and aggressiveness [20]. When accounting for the diagnosis of borderline personality disorder aggressiveness seem to be the personality trait most valuable in assessing suicidal risk [21]. Subjects who self-harm have similar cognitive characteristics as subjects who commit low-lethality suicide attempts.
Table 1 Cognitive domains in COGNISTAT of adolescents with and without prior non suicidal self-injury. Non-suicidal Domain self-injury Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No
Orientation
Number of Mean Mean P value subjects difference
17 33 Attention 17 33 Comprehension 17 33 Repetition 17 33 Naming 17 33 Construction 17 33 Memory 17 33 Calculation 17 33 Similarities 17 33 Judgment 17 33
11.76 11.36 7.35 6.76 6 5.67 12.12 11.36 7.18 6.91 10.12 4.42 9.35 11.36 3 5.79 6.94 11.55 3.88 9.21
0.401
0.081
0.595
0.224
0.333
0.025
0.754
0.183
0.267
0.555
5.693
0.159
-2.011
0.613
-2.788
0.499
-4.604
0.409
-5.33
0.352
Table 2 Cognitive domains in COGNISTAT of adolescents with and without prior suicide attempt. Prior suicide Domain attempt
Number of subjects
Mean
Mean P value difference
Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No
15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35
11.87 11.34 7.73 6.63 5.93 5.71 12.27 11.34 7.27 6.89 4.93 6.97 9.53 11.17 3.27 5.51 7.07 11.23 4 8.86
0.524
0.019⁎
1.105
0.002⁎
0.219
0.148
0.924
0.113
0.381
0.416
-2.038
0.629
-1.638
0.69
-2.248
0.599
-4.162
0.471
-4.857
0.412
Orientation Attention Comprehension Repetition Naming Construction Memory Calculation Similarities Judgment
⁎ Significant after Bonferroni correction for multiple testing.
Subjects who attempt high-lethality suicide attempts has a different cognitive profile, perhaps similar to suicide completers[22]. Mann et al demonstrated that the choice of high lethality suicide method is a poor prognostic factor, and is correlated with lower long-term survival rates. In our sample of adolescent inpatients, suicidal patients had a slightly better cognitive function than non-suicidal patients. This finding should be interpreted with cautious, considering the whole sample was of psychiatric inpatients. Subjects who have better cognitive profile seem to have a wider range of affect and suffer from less negative symptoms and consequently experience more depression and suicidality. For instance, in a previous study on adolescent-onset schizophrenia we found that insight correlated positively with depression and suicidality [23]. On a different sample of adolescents we demonstrated that thoughts of suicide do not necessarily translate into an actual risk of suicidal behavior in adolescents with schizophrenia (unpublished data). The lack of association between specific diagnoses and COGNISTAT profile is consistent with the literature and in line with our hypothesis, that cognitive deficits are not specific, and could not differentiate various psychiatric populations. Moreover, a single cognitive evaluation is more likely to reflect a state rather than a trait. Therefore, an association with a particular psychiatric diagnosis was not expected. The major limitations of our study are the crosssectional retrospective design, the relatively small sample size, heterogeneous nature of the study population and lack of mentally healthy control group. Replicating it in a large sample from different setting in a prospective design is needed, in order to generalize the findings. A particular
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
limitation of the COGNISTAT is its lack of sensitivity or specificity for some cognitive domains [24,25]. Despite these limitations, this study has also two main strengths: First, all the described evaluations are part of a clinical routine. Thus the feasibility of using them is higher. Second, suicide is not studied here as a diagnosis, but rather as a dimension, which is congruent with contemporary perspectives.
Acknowledgment The authors extend their gratitude to the Judie and Marshall Polk Research Fund for Children at Risk for its financial support of this study. Authors also thank Mrs. Michaella Gerchak for editing services.
References [1] Hippius H, Muller N. The work of Emil Kraepelin and his research group in Munchen. Eur Arch Psychiatry Clin Neurosci 2008; 258(Suppl 2):3-11. [2] Green MF. Cognitive impairment and functional outcome in schizophrenia and bipolar disorder. J Clin Psychiatry 2006;67:e12. [3] Hill SK, Reilly JL, Harris MS, Rosen C, Marvin RW, Deleon O, et al. A comparison of neuropsychological dysfunction in first-episode psychosis patients with unipolar depression, bipolar disorder, and schizophrenia. Schizophr Res 2009;113:167-75. [4] Hill SK, Harris MS, Herbener ES, Pavuluri M, Sweeney JA. Neurocognitive allied phenotypes for schizophrenia and bipolar disorder. Schizophr Bull 2008;34:743-59. [5] Hawkins KA, Keefe RS, Christensen BK, Addington J, Woods SW, Callahan J, et al. Neuropsychological course in the prodrome and first episode of psychosis: findings from the PRIME North America Double Blind Treatment Study. Schizophr Res 2008;105:1-9. [6] Emsley R, Chiliza B, Schoeman R. Predictors of long-term outcome in schizophrenia. Curr Opin Psychiatry 2008;21:173-7. [7] Bora E, Yucel M, Pantelis C. Cognitive impairment in schizophrenia and affective psychoses: implications for DSM-V criteria and beyond. Schizophr Bull 2010;36:36-42. [8] Bora E, Yucel M, Pantelis C. Cognitive impairment in affective psychoses: a meta-analysis. Schizophr Bull 2010;36:112-25. [9] Keilp JG, Gorlyn M, Oquendo MA, Burke AK, Mann JJ. Attention deficit in depressed suicide attempters. Psychiatry Res 2008;159:7-17.
373
[10] Keilp JG, Sackeim HA, Brodsky BS, Oquendo MA, Malone KM, Mann JJ. Neuropsychological dysfunction in depressed suicide attempters. Am J Psychiatry 2001;158:735-41. [11] APA. Diagnostic and Statistical Manual of Mental Disorders (DSM) 4 ed. Washington D.C: American Psychiatric Association; 1994. [12] Engelhart C, Eisenstein N, Johnson V, Wolf J, Williamson J, Steitz D, et al. Factor structure of the Neurobehavioral Cognitive Status Exam (COGNISTAT) in healthy, and psychiatrically and neurologically impaired, elderly adults. Clin Neuropsychol 1999;13:109-11. [13] Schrimsher GW, Parker JD, Burke RS. Relation between cognitive testing performance and pattern of substance use in males at treatment entry. Clin Neuropsychol 2007;21:498-510. [14] Katz N, Tadmor I, Felzen B, Hartman-Maeir A. The Behavioural Assessment of the Dysexecutive Syndrome (BADS) in schizophrenia and its relation to functional outcomes. Neuropsychol Rehabil 2007; 17:192-205. [15] Niekawa N, Sakuraba Y, Uto H, Kumazawa Y, Matsuda O. Relationship between financial competence and cognitive function in patients with schizophrenia. Psychiatry Clin Neurosci 2007;61:455-61. [16] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association; 2000. [17] Mann JJ. A current perspective of suicide and attempted suicide. Ann Intern Med 2002;136:302-11. [18] Mann JJ, Waternaux C, Haas GL, Malone KM. Toward a clinical model of suicidal behavior in psychiatric patients. Am J Psychiatry 1999;156:181-9. [19] Oquendo MA, Carballo JJ, Rajouria N, Currier D, Tin A, Merville J, et al. Are high-lethality suicide attempters with bipolar disorder a distinct phenotype? Arch Suicide Res 2009;13:247-56. [20] Oquendo MA, Bongiovi-Garcia ME, Galfalvy H, Goldberg PH, Grunebaum MF, Burke AK, et al. Sex differences in clinical predictors of suicidal acts after major depression: a prospective study. Am J Psychiatry 2007;164:134-41. [21] Keilp JG, Gorlyn M, Oquendo MA, Brodsky B, Ellis SP, Stanley B, et al. Aggressiveness, not impulsiveness or hostility, distinguishes suicide attempters with major depression. Psychol Med 2006;36: 1779-88. [22] LeGris J, van Reekum R. The neuropsychological correlates of borderline personality disorder and suicidal behaviour. Can J Psychiatry 2006;51:131-42. [23] Horesh N, Zalsman G, Apter A. Internalized anger, self-control, and mastery experience in inpatient anorexic adolescents. J Psychosom Res 2000;49:247-53. [24] Blyth T, Scott A, Bond A, Paul E. A comparison of two assessments of high level cognitive communication disorders in mild traumatic brain injury. Brain Inj 2012;26:234-40. [25] Goold D, Vane DW. Evaluation of functionality after head injury in adolescents. J Trauma 2009;67:71-4.
ScienceDirect Comprehensive Psychiatry 55 (2014) 370 – 373 www.elsevier.com/locate/comppsych
Cognitive traits in inpatient adolescents with and without prior suicide attempts and non-suicidal self-injury Shlomit Tsafrir a , Elena Chubarov a , Gal Shoval a, b , Mali Levi a , Eitan Nahshoni b , Motti Ratmansky c , Abraham Weizman a, b, d , Gil Zalsman a, b, e,⁎ a
Child and Adolescent Division, Geha Mental Health Center, Petah Tikva, Israel b Sackler Faculty of Medicine, Tel Aviv University, Israel c Pain Rehabilitation Unit, Loewenstein Rehabilitation Hospital, Raanana, Israel d Research Unit, Geha Mental Health Center and Felsenstein Medical Research Center, Petah Tikva, Israel e Molecular Imaging and Psychopathology Division, Psychaitry DepartmentNY, USA
Abstract Objective: Establishing a psychiatric diagnosis and assessing suicidal tendencies is often a challenging task particularly in the early stages of an illness. Cognitive impairments characterize different psychiatric entities, but there is no known specific cognitive deficit profile that could help the clinician in achieving the diagnostic task. This study’s aim was to establish a cognitive profile and test its ability to differentiate psychiatric inpatient subgroups, in terms of suicidal risk and diagnosis. The sample constituted of 76 consecutive admissions to the psychiatric adolescent day-care unit, who were admitted for any diagnosis. Assessment included full psychiatric interview and cognitive evaluation, using the COGNISTAT test. Results: Of the 76 participants, 58% reported having suicidal ideation and 29% reported a prior attempted suicide. Subjects who had a prior suicide attempt had better orientation and attention scores in the COGNISTAT. Other cognitive domains did not differentiate between groups or diagnoses. Conclusion: These preliminary results suggest a significant association between specific cognitive characteristics and suicidal behavior in adolescents. Those cognitive characteristics might prove clinically useful in the assessment of suicide risk. Further study is needed to establish this association and generalize the conclusion to different populations. © 2014 Elsevier Inc. All rights reserved.
1. Introduction In the past it was a common notion that schizophrenia and bipolar affective disorder differ in regard to the cognitive impairment and natural history of the disease [1]. This dichotomy is probably inaccurate and affective disordered patients, similar to schizophrenic patients, has difficulties in daily occupational and social function, partly attributed to cognitive impairments [2].
Disclosure: All authors declare that they have no conflict of interest whatsoever with regard to this study. ⁎ Corresponding author. Child and Adolescent Division Geha Mental Health Center, Tel Aviv University 1 Helsinki St, P.O. Box 102 Petah Tikva 49100 Israel. Tel.: +972 3 925 8205; fax: +972 3 924 1041. E-mail address: [email protected] (G. Zalsman). 0010-440X/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.comppsych.2013.08.009
Those cognitive deficits characterize psychiatric population, and are of great influence on quality of life and daily functioning. A specific cognitive profile and its relevance to the psychiatric evaluation process is not well established [3,4]. Patients with schizophrenia suffer cognitive deficit early in the course of illness, the deficit is usually permanent, not affected by pharmacological intervention and is found to be general, but most prominently affect the memory, attention and executive functions [5]. Those deficits are correlated with complex prognostic outcome variables [6]. Patients who suffer from affective disorder have similar cognitive deficits, especially in the domains of attention, memory and learning, so that the cognitive profile might not differentiate those diagnoses [7,8].
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
371
Adolescent suicide is a troubling phenomenon with high psychiatric comorbidity rates and worldwide presence. Suicide is the second leading cause of death in children and adolescents in Israel with a frequency of 8:100,000, according to the data from the Israeli Ministry of Health. Early assessment of suicidality in adolescents is, therefore, critical. Specific cognitive dysfunction that distinguishes individuals who has a suicidal tendency is not well established, but particular deficits in the decision making process and attention are typically present [9]. Keilp et al. described cognitive deficits in memory and attention in depressed patients, and a specific, additional deficit in executive function in depressed patients who had a prior lethal suicide attempt [10]. The goals of the current study were to determine whether a cognitive profile could differentiate between different subgroups of psychiatric patients. The study focused on differentiating subgroups of patients in terms of suicidal tendencies and diagnostic classification, in a sample of adolescent inpatients.
psychiatric clinical situations [12,13]. Deficits in specific domains characterize specific patients populations, and is correlated with level of function [14,15]. The study was approved by the Geha Mental Health Center Review Board that exempted us from having an informed consent due to the retrospective file review nature of the study.
2. Methods
3. Results
2.1. Participants Seventy six adolescents consecutively hospitalized at the adolescent day-care unit, in a university affiliated mental health center, and were recruited for the current study. Diagnosis of all participants were based on DSM-IV criteria (American Psychiatric Association, 1994) [11] using data that had been collected during clinical interviews, daily observations and clinical team discussions. Subjects were excluded from the study if they did not undergo a cognitive evaluation during their hospitalization, for any reason. 2.2. Materials and procedures Adolescents admitted to the day care unit undergo a full psychiatric interview by a child and adolescent psychiatrist, and additional diagnostic and evaluation procedures. Medical records of adolescents who were admitted to the day care unit, and underwent cognitive evaluation as part of the assessment procedure, were scanned and socio-demographic, medical and psychiatric data was abstracted. Cognitive and functional evaluation is done routinely, by an occupational therapist, as a part of the assessment process. The assessment includes the COGNISTAT, a cognitive status assessment tool. The COGNISTAT assess five major domains of cognitive function: language, construction, memory, calculation and reasoning. Level of consciousness, orientation and attention are examined separately. The test generates a profile representing the abilities and limitations of the subject. The COGNISTAT protocol was standardized in normal functioning subjects, at different age groups, and it is used to assess cognitive function in a variety of neuro-
2.3. Statistical analysis SPSS version 17 was used in the statistical analysis. Student T test and one-way analysis of variance were performed as appropriate, to evaluate association between major diagnosis and suicidal risk (independent variables) with cognitive domains as measured by the COGNISTAT (dependant variable). All tests were two-tailed. Levene tests were conducted for equivalence of variance in the group compared, and when proved significant, equal variances were not assumed. Bonferroni adjustment was used to correct for multiple testing.
Seventy six adolescents admitted to the day care unit were recruited to the study. Their ages varied between 12 and 20 years, with an average age of 14.6. The male to female ratio was 39:37. Forty eight (63%) patients were admitted for the first time. The subjects had been diagnosed as suffering from a diverse variety of psychiatric disorders, according to the DSM-IV classification[16]. Twenty (26%) had a main diagnosis of schizophrenia, seventeen (22%) had a main diagnosis of major depressive episode, fifteen (20%) had a main diagnosis of borderline personality disorder and twenty four (31%) had another, less common, main diagnosis. In addition, some had two or three co-morbid disorders (data not shown). Forty four (58%) patients reported having suicidal ideation and twenty two (29%) reported a prior attempted suicide. Twenty three (30%) self-harmed; with no suicide intent. COGNISTAT was administered to all the patients, as part of the routine assessment procedure. The effect of level of function, defined as a COGNISTAT grade in each domain tested, on diagnosis, suicidal risk and non-suicidal self injury was tested statistically. There was no difference on any cognitive domain in the four diagnoses groups (data not shown). The difference between the comprehension domain scores was minimally, but significantly, better in the selfharmed subjects (Table 1). Orientation and attention domains had a slightly better score in the subjects who had a prior suicide attempt. This finding was also significant (Table 2). There was no significant difference on any cognitive domain in the subjects who reported having suicidal ideation (data not shown).
372
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
4. Discussion This study used COGNISTAT test to assess cognitive profile in 76 adolescent inpatients, suicidal and not suicidal. In this sample patients who reported a prior suicide attempt had better orientation and attention, compared to nonsuicidal patients. Patients, who self-harmed, with no suicidal intent, had slightly better scores in the orientation and comprehension domains. These patients had a cognitive profile similar to the patients who reported prior suicide attempt. This difference was smaller when looking at subjects who reported suicidal ideation vs. those who didn't and subjects who attempted suicide while hospitalized vs. non suicidal subjects in the COGNISTAT profile. We did not find an association between a DSM-IV diagnosis and a specific cognitive profile in the COGNISTAT. This finding highlights the importance of distinguishing between suicidal ideation and actual suicide risk[17,18]. The different characteristics of suicide attempters and suicide completers are well established in the scientific literature[19] and the population of ideators is probably a third different population. Behaviors of non-suicidal self injury and suicide attempt are found often in similar patient populations: female, often usage of non-violent suicide methods, and high frequency of specific personality traits, including impulsivity, hostility and aggressiveness [20]. When accounting for the diagnosis of borderline personality disorder aggressiveness seem to be the personality trait most valuable in assessing suicidal risk [21]. Subjects who self-harm have similar cognitive characteristics as subjects who commit low-lethality suicide attempts.
Table 1 Cognitive domains in COGNISTAT of adolescents with and without prior non suicidal self-injury. Non-suicidal Domain self-injury Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No
Orientation
Number of Mean Mean P value subjects difference
17 33 Attention 17 33 Comprehension 17 33 Repetition 17 33 Naming 17 33 Construction 17 33 Memory 17 33 Calculation 17 33 Similarities 17 33 Judgment 17 33
11.76 11.36 7.35 6.76 6 5.67 12.12 11.36 7.18 6.91 10.12 4.42 9.35 11.36 3 5.79 6.94 11.55 3.88 9.21
0.401
0.081
0.595
0.224
0.333
0.025
0.754
0.183
0.267
0.555
5.693
0.159
-2.011
0.613
-2.788
0.499
-4.604
0.409
-5.33
0.352
Table 2 Cognitive domains in COGNISTAT of adolescents with and without prior suicide attempt. Prior suicide Domain attempt
Number of subjects
Mean
Mean P value difference
Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No
15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35 15 35
11.87 11.34 7.73 6.63 5.93 5.71 12.27 11.34 7.27 6.89 4.93 6.97 9.53 11.17 3.27 5.51 7.07 11.23 4 8.86
0.524
0.019⁎
1.105
0.002⁎
0.219
0.148
0.924
0.113
0.381
0.416
-2.038
0.629
-1.638
0.69
-2.248
0.599
-4.162
0.471
-4.857
0.412
Orientation Attention Comprehension Repetition Naming Construction Memory Calculation Similarities Judgment
⁎ Significant after Bonferroni correction for multiple testing.
Subjects who attempt high-lethality suicide attempts has a different cognitive profile, perhaps similar to suicide completers[22]. Mann et al demonstrated that the choice of high lethality suicide method is a poor prognostic factor, and is correlated with lower long-term survival rates. In our sample of adolescent inpatients, suicidal patients had a slightly better cognitive function than non-suicidal patients. This finding should be interpreted with cautious, considering the whole sample was of psychiatric inpatients. Subjects who have better cognitive profile seem to have a wider range of affect and suffer from less negative symptoms and consequently experience more depression and suicidality. For instance, in a previous study on adolescent-onset schizophrenia we found that insight correlated positively with depression and suicidality [23]. On a different sample of adolescents we demonstrated that thoughts of suicide do not necessarily translate into an actual risk of suicidal behavior in adolescents with schizophrenia (unpublished data). The lack of association between specific diagnoses and COGNISTAT profile is consistent with the literature and in line with our hypothesis, that cognitive deficits are not specific, and could not differentiate various psychiatric populations. Moreover, a single cognitive evaluation is more likely to reflect a state rather than a trait. Therefore, an association with a particular psychiatric diagnosis was not expected. The major limitations of our study are the crosssectional retrospective design, the relatively small sample size, heterogeneous nature of the study population and lack of mentally healthy control group. Replicating it in a large sample from different setting in a prospective design is needed, in order to generalize the findings. A particular
S. Tsafrir et al. / Comprehensive Psychiatry 55 (2014) 370–373
limitation of the COGNISTAT is its lack of sensitivity or specificity for some cognitive domains [24,25]. Despite these limitations, this study has also two main strengths: First, all the described evaluations are part of a clinical routine. Thus the feasibility of using them is higher. Second, suicide is not studied here as a diagnosis, but rather as a dimension, which is congruent with contemporary perspectives.
Acknowledgment The authors extend their gratitude to the Judie and Marshall Polk Research Fund for Children at Risk for its financial support of this study. Authors also thank Mrs. Michaella Gerchak for editing services.
References [1] Hippius H, Muller N. The work of Emil Kraepelin and his research group in Munchen. Eur Arch Psychiatry Clin Neurosci 2008; 258(Suppl 2):3-11. [2] Green MF. Cognitive impairment and functional outcome in schizophrenia and bipolar disorder. J Clin Psychiatry 2006;67:e12. [3] Hill SK, Reilly JL, Harris MS, Rosen C, Marvin RW, Deleon O, et al. A comparison of neuropsychological dysfunction in first-episode psychosis patients with unipolar depression, bipolar disorder, and schizophrenia. Schizophr Res 2009;113:167-75. [4] Hill SK, Harris MS, Herbener ES, Pavuluri M, Sweeney JA. Neurocognitive allied phenotypes for schizophrenia and bipolar disorder. Schizophr Bull 2008;34:743-59. [5] Hawkins KA, Keefe RS, Christensen BK, Addington J, Woods SW, Callahan J, et al. Neuropsychological course in the prodrome and first episode of psychosis: findings from the PRIME North America Double Blind Treatment Study. Schizophr Res 2008;105:1-9. [6] Emsley R, Chiliza B, Schoeman R. Predictors of long-term outcome in schizophrenia. Curr Opin Psychiatry 2008;21:173-7. [7] Bora E, Yucel M, Pantelis C. Cognitive impairment in schizophrenia and affective psychoses: implications for DSM-V criteria and beyond. Schizophr Bull 2010;36:36-42. [8] Bora E, Yucel M, Pantelis C. Cognitive impairment in affective psychoses: a meta-analysis. Schizophr Bull 2010;36:112-25. [9] Keilp JG, Gorlyn M, Oquendo MA, Burke AK, Mann JJ. Attention deficit in depressed suicide attempters. Psychiatry Res 2008;159:7-17.
373
[10] Keilp JG, Sackeim HA, Brodsky BS, Oquendo MA, Malone KM, Mann JJ. Neuropsychological dysfunction in depressed suicide attempters. Am J Psychiatry 2001;158:735-41. [11] APA. Diagnostic and Statistical Manual of Mental Disorders (DSM) 4 ed. Washington D.C: American Psychiatric Association; 1994. [12] Engelhart C, Eisenstein N, Johnson V, Wolf J, Williamson J, Steitz D, et al. Factor structure of the Neurobehavioral Cognitive Status Exam (COGNISTAT) in healthy, and psychiatrically and neurologically impaired, elderly adults. Clin Neuropsychol 1999;13:109-11. [13] Schrimsher GW, Parker JD, Burke RS. Relation between cognitive testing performance and pattern of substance use in males at treatment entry. Clin Neuropsychol 2007;21:498-510. [14] Katz N, Tadmor I, Felzen B, Hartman-Maeir A. The Behavioural Assessment of the Dysexecutive Syndrome (BADS) in schizophrenia and its relation to functional outcomes. Neuropsychol Rehabil 2007; 17:192-205. [15] Niekawa N, Sakuraba Y, Uto H, Kumazawa Y, Matsuda O. Relationship between financial competence and cognitive function in patients with schizophrenia. Psychiatry Clin Neurosci 2007;61:455-61. [16] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association; 2000. [17] Mann JJ. A current perspective of suicide and attempted suicide. Ann Intern Med 2002;136:302-11. [18] Mann JJ, Waternaux C, Haas GL, Malone KM. Toward a clinical model of suicidal behavior in psychiatric patients. Am J Psychiatry 1999;156:181-9. [19] Oquendo MA, Carballo JJ, Rajouria N, Currier D, Tin A, Merville J, et al. Are high-lethality suicide attempters with bipolar disorder a distinct phenotype? Arch Suicide Res 2009;13:247-56. [20] Oquendo MA, Bongiovi-Garcia ME, Galfalvy H, Goldberg PH, Grunebaum MF, Burke AK, et al. Sex differences in clinical predictors of suicidal acts after major depression: a prospective study. Am J Psychiatry 2007;164:134-41. [21] Keilp JG, Gorlyn M, Oquendo MA, Brodsky B, Ellis SP, Stanley B, et al. Aggressiveness, not impulsiveness or hostility, distinguishes suicide attempters with major depression. Psychol Med 2006;36: 1779-88. [22] LeGris J, van Reekum R. The neuropsychological correlates of borderline personality disorder and suicidal behaviour. Can J Psychiatry 2006;51:131-42. [23] Horesh N, Zalsman G, Apter A. Internalized anger, self-control, and mastery experience in inpatient anorexic adolescents. J Psychosom Res 2000;49:247-53. [24] Blyth T, Scott A, Bond A, Paul E. A comparison of two assessments of high level cognitive communication disorders in mild traumatic brain injury. Brain Inj 2012;26:234-40. [25] Goold D, Vane DW. Evaluation of functionality after head injury in adolescents. J Trauma 2009;67:71-4.
