|

Criminiseriali.it

See discussions, stats, and author profiles for this publication at: ARTICLE in INTERNATIONAL JOURNAL OF LAW AND PSYCHIATRY · APRIL 2013
Impact Factor: 1.19 · DOI: 10.1016/j.ijlp.2013.04.017 · Source: PubMed
4 AUTHORS, INCLUDING:
67 PUBLICATIONS 176 CITATIONS
2 PUBLICATIONS 0 CITATIONS
Hope and Healing Center & Institute
87 PUBLICATIONS 4,082 CITATIONS
Available from: Matthew S Stanford Retrieved on: 23 June 2015

Contents lists available at International Journal of Law and Psychiatry Pharmacotherapy of impulsive aggression: A quality comparison of controlled studies Alan R. Felthous ,Sarah L. Lake Brian K. Rundle Matthew S. Stanford a Division of Forensic Psychiatry, Department of Neurology and Psychiatry, Saint Louis University School of Medicine, Saint Louis, MO 63104, United Statesb Department of Psychology & Neuroscience, Baylor University, Waco, TX 76798, United States Available online 30 April 2013 The present study assessed the quality of pharmacotherapy trials to treat impulsive aggressive behavior. While asearch of the literature found 55 peer-reviewed published studies on the pharmacotherapy of aggression, only 23 met criteria for inclusion in the quality analysis. To be included in this review, the study must have had at Anti-impulsive aggression agents (AIAAs) least one comparison group to control for placebo effects. The study must have also adequately defined and diag- nosed the presence of impulsive aggression or intermittent explosive disorder. The primary reason studies were Impulsive aggression excluded from the quality analysis was that impulsive aggression was not specifically defined as the behavior being treated (25 of 32, 78%). The results of the quality analysis found that higher quality studies (n = 10; 45%) were characterized by a clear definition of impulsive aggression; specific criteria for what constitutes an impulsiveaggressive act; the exclusion of participants with neurological disorders, serious mental disorders, and/or low IQ;and information concerning the serum levels of the medication being investigated. A significant weakness found inthe literature is the paucity of high quality studies accessing the efficacy of pharmacological agents other than an-ticonvulsants for the treatment of impulsive aggression.
2013 Elsevier Ltd. All rights reserved.
patients Research on aggression and violence has consistently recognized Aggressive behavior is one of the most prevalent concerns in the clin- two subtypes of aggressive behavior: an impulsive type and a premeditat- ical setting and presents as a symptom of numerous psychiatric and neuropsychiatric disorders. In addition, aggressive behavior without co- Impulsive aggression is typically described as an emotionally charged morbid psychopathology may present as a significant problem in clinical aggressive response characterized by a loss of behavioral control ). This type of aggressive display has alsobeen referred to in the literature as affective or reactive ⁎ Corresponding author. Tel.: +1 314 977 4825.
E-mail address: (A.R. Felthous).
(aggression. Specifically, 0160-2527/$ – see front matter 2013 Elsevier Ltd. All rights reserved.
A.R. Felthous et al. / International Journal of Law and Psychiatry 36 (2013) 258–263 described it as "a hair trigger response to a stimulus that results in an ag- Noting the inconsistent findings in anti-aggression drug trials with itated state and culminates in an aggressive act; during the agitated phenytoin in particular, and state…information processing appears to be inefficient." identified the following failures in quality No single medication is FDA approved for the treatment of impulsive of investigations that may have led to the discrepant results: failure aggression, yet a number of drugs representing a variety of classes of to exclude subjects with psychiatric or neurological disorders that medication are used for this purpose. Because results of studies on were identified with objective criteria, or subjects who were taking drug efficacy are so inconsistent, it has been difficult to arrive at firm other medications; failure to control for intelligence levels and differ- conclusions about the efficacy of individual drugs. Also recognized, ent drug serum levels; and reliance on only self-report or subjective however, is the variable quality of such studies. When comparing re- observation rather than "well defined behavioral increases of aggres- sults, judging the quality of each study is essential. Discrepant results sive acts as criteria measures" (). The present in demonstrated drug efficacy may be explained by differences in systematic review examines these and other parameters of quality in study design or quality. To date, reviews of antiaggressive drug trial drug trials in order to distinguish studies of higher quality, regardless have either addressed drug efficacy with little attention given to quality of whether the study results support efficacy of the drug(s) being of the studies, or the reviews have been concerned with only one partic- studied in the treatment of impulsive aggression.
ular class of drugs such as anticonvulsants/mood stabilizers (e.g., In addition to the factors identified by Barratt as lacking in drug trials for the treatment of impulsive aggression, this review includes other Perhaps one of the most frequent flaws in otherwise reasonably common measures of quality in drug trials, especially for psychotropic designed studies and reviews of the pharmacotherapy of clinical ag- drugs. Through the winnowing of high quality research from less rigor- gression is the failure to recognize that subtypes of aggression exist ous studies, the best evidence for drug efficacy may emerge with more (e.g., impulsive aggression). A number of book chapters and scientific consistent findings that support sound clinical decisions in selecting articles on the pharmacotherapy of aggression either fail to specify therapeutic agents.
the type of aggression being treated, or if the aggression is qualified Huwiler-Müntener developed a 30 item instrument, the 1996 as impulsive, an adequate definition is not provided. Consolidated Standards of Reporting Trials (CONSORT) statement identified this as a major problem in the scientific ) that was designed to literature on the pharmacotherapy of aggression. This has continued assess the quality of drug trials in general ( to be an oft repeated error in both individual studies and study Other scales have been developed to measure quality of drug trials Inadequate diagnostic criteria and imprecise diagnoses have long including a 2001 version of the CONSORT confounded research on the treatment of mental disorders. The need for sufficient objective criteria for the diagnosis of mental disorders The original 1996 CONSORT has been was the motivation for the development of the utilized by several journals apparently with improved quality of drug for common but serious mental disorders, the research diagnos- trial reporting . The 1996 CONSORT has been tic criteria (), and the polythetic advanced as useful in assessment of methodological quality in particular diagnostic method which dramatically transformed the Diagnostic and Statistical Manual in its Third Edition For the present comparison, we developed a quality review instru- Within this latter taxonomy, intermittent explosive ment that like the 1996 CONSORT focuses on quality. This includes disorder (IED) first appeared critical parameters of the 1996 CONSORT, but also items identified giving clearer and more objective criteria than emotionally unstable by and as items that personality of the first DSM (); are needed to assess for the quality of drug trials in treatment of explosive personality of DSM II impulsive aggression in particular.
or emotionally unstable personality disorder, impulsive type Classes of medications, which have been used to treat impulsive and borderline type of the International Classification of Diseases (ICD aggression and IED, and have tested for efficacy with controlled trials 10, While IED was carried over include: anticonvulsants, mood stabilizers, selective serotonin reup- subsequent editions of the DSM take inhibitors, and beta adrenergic blockers. This review examined impulsive aggression was increasingly the quality of individual trials of specific agents within each of these defined and recognized apart from but taking into account the research categories, for only those controlled studies with an adequate diagno- supporting IED (e.g., sis of impulsive aggression/IED, and rated the quality of each study One reason for the continued recognition of impulsive aggression based upon quality parameters — most of which are used to assess as a separate but overlapping condition with IED is that in earlier edi- the general quality of drug trials. Additionally, this review looked tions of the DSM (gen- for study recognition of potential biological markers that appear to eralized impulsivity was an exclusionary criterion for IED. However, respond to drug treatment conterminously with subsidence in aggres- generalized impulsivity can and often does coexist with impulsive sive behavior.
outbursts of physical aggression In the currentedition, DSM-IV-TR (gener- alized impulsivity is no longer an exclusionary criterion.
A second and continuing reason for preserving the diagnosis of im- A concern for drug trial studies in general and for reviewers of such pulsive aggression is the recognition of its common co-morbidity with studies is the tendency to include only those studies with positive results.
borderline, antisocial, and other personality disorders ( This bias towards reporting positive results with a net effect of Type I er- ), whereas the DSM criteria for IED discourage making the diag- rors can exist with the individual investigators, the drug industry that fi- nosis where the same phenomenon occurs with a personality disorder nances drug trials, the peer review and editorial processes in selecting ). Even with the diagnostic reports for publication and in reviews of drug trial studies. Also known options of IED, impulsive aggression, and 20 years after as the "file drawer problem" this bias in the science caveat about the importance of correct characterization and publication of drug trials has been especially studied and discussed of the aggressive behavior being examined, studies and reviews on in the context of peer review and editorial selection of drug trial reports the pharmacotherapy of aggression fail to adequately define the for publication ( condition for which a drug is given in an experimental trial.
A.R. Felthous et al. / International Journal of Law and Psychiatry 36 (2013) 258–263 the quality comparison. The primary reason studies were not includ- view, we made a point of including all studies of threshold relevance and ed in the quality comparison was that they did not specifically assess quality regardless of the positivity or negativity of study results.
impulsive aggression but evaluated aggression in general ().
4.2. Interrater reliability 3.1. Selection of drug trial studies Comparison of the quality scores for the two independent raters found that they were highly consistent with an intraclass correlation Using MEDLINE and PsycINFO, the authors attempted to identify all controlled studies in the English language that tested for drug effi-cacy in the pharmacotherapy of aggression. The present article did 4.3. Assessment of quality not include studies in the pharmacotherapy of acute agitation on anemergency basis, but rather the ongoing administration of a medica- One of the 23 studies included in the quality comparison tion in order to prevent or reduce the frequency and intensity of ) was a follow-up study of another future acts of aggression. To be included in this review, the study (, thus the two were treated as a single study. The must have at least one control or comparison group to control for pla- average quality score for the 22 studies was 41.3 (SD 6.6). The median cebo effect. The study must have adequately defined and diagnosed and mode were both 42.0. Anticonvulsants were by far the most the presence of impulsive aggression by approximating the definition common pharmacological agents studied in the treatment of impulsive of impulsive aggression provided by apply- aggression, comprising 14 of the 22 studies. Quality scores for the anti- ing the criteria for IED from one of the DSMs since the third edition, or convulsant studies are presented in while quality scores for the a (reasoned) modification of such criteria.
remaining studies are presented in Ten studies had a quality scores above the median and were clas- 3.2. Assessment of quality: Checklist sified as higher quality while the remaining 12 studies were classifiedas lower quality. Statistical comparison between higher and lower A 19-item quality checklist was developed by the authors using quality studies on the checklist items were done using t-tests and items identified by , and the 14 items for are reported in .
which at least three (out of five) judges in usedto rate the quality of reports of randomized clinical trials. All items used to assess the quality of drug trials in the treatment of impulsiveaggression are listed in The present study assessed the quality of pharmacotherapy trials Although some quality measures are clearly either fully present or to treat impulsive aggressive behavior. While a search of the litera- fully absent, others can exist in various degrees, for example, the de- ture found 55 published studies on the pharmacotherapy of aggres- scription of inclusion and exclusion criteria. Therefore weighted points sion, only 23 met criteria for inclusion in the quality analysis. The were given to each item, with 0 = the item is absent from the study, primary reason studies were excluded from the analysis was that im- 1 = the item is present, 2 = the item is present to a satisfactory de- pulsive aggression was not specifically defined as the behavior being gree, or 3 = the item is present to a highly satisfactory degree. Dichot- treated (25 of 32, 78%). This demonstrates a significant general weak- omous quality measures (e.g., random allocation) which were either ness in the aggression pharmacotherapy literature.
present or absent were scored 0 = the item is absent or 3 = the item As described earlier, physical aggressive behavior has traditionally is present. Total scores on the quality checklist can range from 0 to 57.
been classified into two distinct subtypes ((a) an emotionally charged, uncontrolled type of aggressive 3.3. Assessment of quality: Process display (impulsive, hostile, affective, and reactive) or (b) a planned,controlled, unemotional aggressive act (premeditated, instrumental, Two doctoral graduate students in psychology, both co-authors of predatory, and proactive). Significant differences in neurochemistry, psy- this article, were first familiarized with the checklist items. Each me- chophysiology, cognitive functioning and response to treatment/inter- thodically examined each study independently and rated the individ- vention have been shown between these aggressive subtypes ual items. After each study was rated, the results of the two ratings ). This makes the careful characterization were compared for consistency and the scores averaged for a total of aggressive behavior imperative when assessing treatment effective- score on each item.
ness. Presently, a majority of the pharmacotherapy of aggression litera-ture appears to have ignored this distinction making the determination 3.4. Analysis of results of efficacy for many medications in the treatment of aggressive behaviordifficult if not impossible.
Interrater reliability was calculated for the scores of the two inde- The results of the quality analysis found that higher quality pharma- pendent raters using an intraclass correlation for consistency. Based cotherapy studies are characterized by a clear definition of impulsive ag- upon the median score of all items used to assess study quality, gression, specific criteria for what constitutes an impulsive aggressive each study was rated as higher (above the median) or lower (at the act, the exclusion of participants with neurological disorders, serious median and below) quality. Finally, statistical analyses were applied mental disorders and/or low IQ, and information concerning the serum to higher and lower quality studies to determine on which criteriathey differed.
Reasons studies were excluded from quality comparison.
4.1. Selection of drug trial studies Not specifically impulsive aggression Literature searches for studies assessing drug efficacy in the phar- macotherapy of aggression found 55 peer-reviewed publications. A Retrospective chart review Single administration review of those studies found that 23 met criteria for inclusion in A.R. Felthous et al. / International Journal of Law and Psychiatry 36 (2013) 258–263 Quality of anticonvulsant trials in the treatment of impulsive aggression.
Comparison of quality checklist items between higher and lower quality studies.
Quality checklist items Measures of aggression a) Impulsive aggression adequately defined b) Behavioral measures of aggressive acts c) Criteria for impulsive aggressive acts d) Intent to treat aggression Recruitment of participants a) Clear inclusion criteria (5–12 years old) b) Exclusion criteria 1. Neurological disorders 2. Serious DSM Axis I mental disorders 4. Other psychotropic medications a) Clear hypothesis and objectives (10–18 years old) a) Clear/validated outcomes b) Description of withdrawals and dropouts c) Adequate follow-up e) Serum levels reported Bolded p-values indicate items which were significantly associated with the higher Forensic patients quality studies.
a Denotes dichotomous item.
levels of the medication being investigated. Higher and lower quality characterized as higher quality (). It is clear from the literature studies both showed weaknesses in the reporting of power, following- that several anticonvulsant agents are effective in the treatment of im- up participants after the study ended, and the inclusion of biological pulsive aggressive outbursts markers as outcome measures These results call into question Unfortunately, this cannot be said for most other pharmacologi- the reliability of the data from those studies characterized as lower qual- cal agents. Only a limited few have been assessed and most of those ity and demonstrate the need for a standardized approach to the assess- studies were characterized as lower quality in the present analysis ment of pharmacological agents in the treatment of impulsive aggressive ). It is hoped that this study will serve as a call to aggression researchers to redouble their efforts in the investigation of pharmaco- A significant problem in the literature is the paucity of studies logical agents such as selective-serotonin reuptake inhibitors, lithium, accessing the efficacy of pharmacological agents other than anticon- and atypical antipsychotics for the treatment of impulsive aggressive vulsants for the treatment of impulsive aggression. Anticonvulsant agents have been widely researched with a majority of those studies Most reviews of drugs used to treat aggressive behavior do not me- thodically evaluate the quality of drug trials. An exception is themeta-analysis by . The aim of this meta-analysis was to assess the evidence for efficacy of mood sta- Quality of drug trials in the treatment of impulsive aggression.
bilizers in reducing aggression that was repetitive or impulsive. The in- vestigators cast a broad net and identified 52 drug trials for treating aggression, but only eight reports, representing ten studies, contained enough information to use in the meta-analysis. Using the Jadad Scale ), to assess for risk of bias, eight had (5–12 years old) Jadad scores of three or higher. From their analysis, only three of these studies included intention to treat analysis. Evidence supported efficacy (5–12 years old) for carbamazepine/oxcarbazepine, phenytoin and lithium, but not all mood stabilizers. Levetiracetam and valproate were shown to be not In comparing the meta-analysis by Jones and colleagues with the present systematic review, some differences should be noted. The focus of the Jones meta-analysis was drug efficacy, but using some mea- sures of quality to measure bias. The emphasis of the present review was comparing the quality of the drug trials regardless of positive or negative outcome. Parameters were intended to assess methodological quality, not reporting quality which is more the concern of the JadadScale. The present study sought studies that convincingly targeted im- pulsive aggression, adequately defined and diagnosed. Recurrent ag- (8–11 years old) gression that is not also impulsive would not have qualified. The present study was open to the trial of adequate quality of any drug, not just mood stabilizers.
A.R. Felthous et al. / International Journal of Law and Psychiatry 36 (2013) 258–263 The present study, though targeting impulsive aggression, did not Callaham, M. L., Wears, R. L., Weber, E. J., Barton, C., & Young, G. (1998). exclude any demographic groups. Thus, it was not limited to drug tri- als in adult subjects, but also in children. The reader must bear in Campbell, M., Adams, P. B., Small, A. M., Kafantaris, V., Silva, R. R., Shell, J., et al. (1995).
mind the possibility of children responding differently than adults to some drugs. In addition, children are more likely to have a primary diagnosis of attention deficit hyperactivity disorder which can include Campbell, M., Small, A. M., Green, W. H., Jennings, S. J., Perry, R., Bennett, W. G., et al.
impulsive aggression but includes other symptoms as well. Similarly, some adults who are treated for impulsive aggression may actually have adult ADHD and their other symptoms of ADHD may not be Chalmers, T. C., Smith, H., Jr., Blackburn, B., Silverman, B., Schroeder, B., Reitman, D., registered in the study. Future investigations can disentangle pure impulsive aggression from that associated with ADHD and then test Coccaro, E. F., & Kavoussi, R. J. (1997). for differential responses to presumed anti-aggressive drugs and Coccaro, E. F., Lee, R. J., & Kavoussi, R. J. (2009). Cueva, J. E., Overall, J. E., Small, A. M., Armenteros, J. L., Perry, R., & Campbell, M. (1996).
Critical to improving pharmacotherapy of clinical aggression is satis- factory methodologic quality. This requires not only accurate and unbi- Dickerson, R. (1990). ased measuring but also an adequate definition and diagnosis of the Dodge, K. A., & Coie, J. D. (1998). condition that is to be measured. With compelling, replicated research that impulsive but not premeditated aggression is amenable to pharma- cotherapy (e.g., the first step in research design is to Donovan, S. J., Stewart, J. W., Nunes, E. V., Quitkin, F. M., Parides, M., Daniel, W., et al.
distinguish impulsive aggression and determining how its intensity and frequency is to be monitored. Other parameters of quality are basically those used to assess quality of trials of drugs in general, but with some Feighner, J. P., Robins, E., Guze, S. B., Woodruf, R. A., Jr., Winokur, G., & Munoz, R.
considerations specific to the study of aggression.
Most of the higher quality studies (i.e., those with a score above the Felthous, A. R., Bryant, S. G., Wingerter, C. B., & Barratt, E. (1991). median) were anticonvulsants. All of these well tested anticonvulsants with the exception of levetiracetam were shown to be efficacious.
Foster, H. G., Hillbrand, M., & Chi, C. C. (1989). Evidence from studies of satisfactory methodologic quality supports the efficacy of phenytoin, carbamazepine, valproate/divalproex in treating impulsive aggression. Of the drug trials involving non-anticonvulsant Gardner, D. L., & Cowdry, R. W. (1986). drugs, the quality of the trials with lithium and fluoxetine are most satisfactory, and the individual trials of these drugs shows them to be Greendyke, R. M., & Kanter, D. R. (1986). The present results suggest that the reliability of a majority of the Greenwald, A. G. (1975). pharmacotherapy of aggression studies presently in the literature is questionable. The authors suggest that three methods are desperately Hollander, E., Swann, A. C., Coccaro, E. F., Jiang, P., & Smith, T. B. (2005). needed in future pharmacotherapy studies of aggression to overcome the present shortfall: careful characterization of aggressive behavior Hollander, E., Tracy, K. A., Swann, A. C., Coccaro, E. F., McElroy, S. L., Wozniak, P., et al.
into subtypes, the use of a standard approach to the assessment of treatment efficacy, and the investigation of a wide range of pharma- Houston, R. J., & Stanford, M. S. (2006). cological agents.
Houston, R. J., Stanford, M. S., Villemarette-Pittman, N. R., Conklin, S. M., & Helfritz, L. E.
Huwiler-Müntener, K. (2000). American Psychiatric Association (1952). American Psychiatric Association (1968). American Psychiatric Association (1980). Huwiler-Müntener, K., Jüni, P., Junker, C., & Egger, M. (2002). American Psychiatric Association (1987). Jadad, A. R., Moore, A., Carroll, D., Jenkinson, C., Reynolds, J. M., Gavaghan, D. J., et al.
American Psychiatric Association (1994). American Psychiatric Association (2000). Jones, R. M., Arlidge, J., Gilham, S., Reagu, S., van den Bree, M., & Taylor, P. J. (2011). Amery, B., Minichiello, M. D., & Brown, G. L. (1984). Lee, R., Kavoussi, R. J., & Coccaro, E. F. (2008a). Barratt, E. S., Kent, T. A., Bryant, S. G., & Felthous, A. R. (1991). Leiberich, P., Nickel, M. K., Tritt, K., & Pedrosa Gil, F. (2008). Barratt, E. S., & Slaughter, L. (1998). Barratt, E. S., Stanford, M. S., Felthous, A. R., & Kent, T. A. (1997). Mahoney, M. J. (1977). Mattes, J. A. (2005). Begg, C., Cho, M., Eastwood, S., Horton, R., Moher, D., Olkin, I., et al. (1996). Mattes, J. A. (2008). A.R. Felthous et al. / International Journal of Law and Psychiatry 36 (2013) 258–263 McEllistrem, J. E. (2004). Spitzer, R. L., Endicott, J., & Robins, E. (1975). Stanford, M. S., Anderson, N. E., Lake, S. L., & Baldridge, R. M. (2009). Moher, D., Jones, A., LePage, L., & COSORT Group (2001a). Stanford, M. S., Helfritz, L. E., Conklin, S. M., Villemarette-Pittman, N. R., Greve, K. W., Moher, D., Schulz, K. F., Altman, D. G., & CONSORT Group (2001b). Adams, D., et al. (2005). Stanford, M. S., Houston, R. J., Mathias, C. W., Greve, K. W., Villemarette-Pittman, N. R., Olson, C. M., Rennie, D., Cook, D., Dickersin, K., Flanagin, A., Hogan, J. W., et al. (2002).
& Adams, D. (2001). Stanford, M. S., Houston, R. J., Mathias, C. W., Villemarette-Pittman, N. R., Helfritz, L. E., & Raine, A., Meloy, J. R., Bihrle, S., Stoddard, J., LaCasse, L., & Buchsbaum, M. S. (1998). Conklin, S. M. (2003). Sterling, T. D. (1959). Rosenthal, R. (1979). Tritt, K., Nickel, C., Lahmann, C., Leiberich, P. K., Rother, W. K., Loew, T. H., et al. (2005).
Sackett, D. L. (1979). Sheard, M. H., Marini, J. L., Bridges, C. I., & Wagner, E. (1976). World Health Organization (1991). Smith, M. L. (1980).

Source: http://www.criminiseriali.it/documenti/Felthous%20et%20al.,%202013.pdf