- Email: [email protected]
Psychotropics and risk of violent crime – Authors' reply
Correspondence
We declare no competing interests.
*Michael Dettling, Ion Anghelescu [email protected] Department of Psychiatry and Psychotherapy, University Clinic Charite, Berlin 14050, Germany (MD); and Clinic Dr Fontheim, Liebenburg, Germany (IA) 1
2
3
4
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. Malhi GS, Chengappa KN, Gershon S, Ghaemi SN. Atypical mood stabilizers: a new role for neuroleptics? Bipolar Disord 2011; 13: 583–86. van der Werf M, Hanssen M, Köhler S, et al. Systematic review and collaborative recalculation of 133 693 incident cases of schizophrenia. Psychol Med 2014; 44: 9–16. Sherazi R, McKeon P, McDonough M, Daly I, Kennedy N. What’s new? The clinical epidemiology of bipolar I disorder. Harv Rev Psychiatry 2006; 14: 273–84.
risk of violence and crime rates, leading to an inherent selection bias in the study sample? 2 Moreover, type and phase of illness were not mentioned, which are both known to affect the risk of violent behaviour and subsequent crime rates.3 Sufficient evidence exists to suggest that substance misuse complicates and confounds aggressive behaviour across the range of mental disorders.4 The authors did mention rates of substance misuse, but their possible confounding effects deserve more discussion. The paradoxical issue of benzodiazepine both inducing and reducing agitation in patients was also left unaddressed. In their study, Fazel and colleagues also showed additional benefit of the use of an antipsychotic rather than a mood stabiliser for reduction of violence and crime rates. However, they did not mention whether the patients were on single or multiple antipsychotics, which could affect study outcomes and results interpretation. Finally, their claim of the study’s superiority compared with previous works should be taken with caution. We feel that the authors could have substantially increased the validity of their findings by asking a few questions from patients or their immediate caregivers related to compliance, domestic violence, effect of hospitalisation, and substance misuse. We declare no competing interests.
*Mohit Varshney, Rishab Gupta [email protected] Department of Psychiatry, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, Delhi 110029, India 1
2
Seema Fazel and colleagues1 deserve credit for their epidemiological study. However, some issues merit further discussion. Were non-compliant patients excluded from withinindividual analyses, thus missing patients who might have had severe psychopathology and, hence, a high www.thelancet.com Vol 384 December 6, 2014
3
4
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. Hodgins S, Riaz M. Violence and phases of illness: differential risk and predictors. Eur Psychiatry 2011; 26: 518–24. Bobes J, Fillat O, Arango C. Violence among schizophrenia out-patients compliant with medication: prevalence and associated factors. Acta Psychiatr Scand 2009; 119: 218–25. Dolan M, O’Malley K, McGregor K. The role of psychopathic traits and substance abuse in predicting violent victimization in patients with schizophrenia spectrum disorders. Personal Ment Health 2013; 7: 28–38.
Authors’ reply We are grateful to Michael Dettling and Ion Anghelescu for raising clinically relevant questions about the drugs that we investigated in our study.1 We selected antipsychotics and mood stabilisers, but not hypnotics (including benzodiazepines) because we focused on long-term outcomes associated with widely prescribed psychiatric drugs, whereas hypnotics are typically prescribed for short periods. We studied antipsychotics and mood stabilisers together for the primary analyses because their coadministration is common, at least in schizophrenia-spectrum disorders. Additionally, we investigated drugs separately according to standard drug classes, rather than to their potential effects. Thus, second-generation antipsychotics with possible mood-stabilising effects were investigated as antipsychotics. As shown in table 3 of our study,1 both first-generation and second-generation antipsychotics were associated with substantial reductions in violent crime in bipolar disorder (hazard ratio 0·52, 95% CI 0·39–0·92). The sex ratio of our sample shows that we used a national patient register with full coverage for outpatients and inpatients. Nevertheless, although the authors of the epidemiological investigation2 cited by Michael Dettling and Ion Anghelescu do not report prevalence ratios for schizophrenia, the female to male incidence rate ratios for the age groups <20 years (0·53, 95% CI 0·41–0·69) and 20–29 years (0·47, 0·41–0·54) are similar to the female to male prevalence ratios in our sample (0·51, 0·49–0·53), which consisted mostly of those aged 20–29 years. In bipolar disorder, the prevalence ratio in our sample was 1·74 (1·68–1·80), an over-representation of women consistent with register-based studies in other countries, such as Denmark.3 Regarding Mohit Varshney and Rishab Gupta’s question about non-compliant patients exclusion—
Robert Brook/Science Photo Library
outbursts. We would like to expand the discussion with some clinical and drug-based remarks. First, to group drugs into antipsychotics and mood stabilisers, or hypnotics, seems uncommon for clinicians because hypnotics represent normal comedication for both antipsychotics and mood stabilisers, at least for acute psychotic and aggressive states. Do any data derived from this study exclude this context? Second, some atypical antipsychotics have mood-stabilising effects.2 Were any data collected from bipolar patients using only atypical antipsychotics as mood stabilisers? Third, the sex distribution in the study for both schizophrenia and bipolar disorders differs obviously from established sex distributions known for these diseases.3,4 Do Fazel and colleagues have an explanation for this finding? To better understand this difference, could this sex discrepancy be seen in the type of violent crime patients diagnosed with schizophrenia and bipolar disorder have been arrested for?
2025
Correspondence
CDC/Science Photo Library
these individuals were central to our study design, which included investigation of rates of violent crime when individuals were taking drugs compared with when they were not. In relation to multiple antipsychotic use, prevalence was low in our sample (0·92%) during the study duration. Other factors could mediate the associations that we report, and we agree that this fact warrants further investigation with different approaches. The type and phase of illness could be relevant in schizophrenia4 and, possibly, bipolar disorder. 5 However, inclusion of sensitive markers and use of caregivers as additional informants do not lend themselves to population-based studies. These studies are necessary to investigate outcomes such as violent crime, suicide, and premature mortality because they need substantially larger sample sizes and longer follow up than have been possible in most randomised controlled trials so far.6
For the ptbnet see http://www. tb-net.org/index.php/ptbnet
NL has received a speaker’s honorarium from Lundbeck in June, 2013. All other authors declare no competing interests.
*Seena Fazel, Johan Zetterqvist, Henrik Larsson, Niklas Långström, Paul Lichtenstein [email protected] Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK (SF); and Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden (JZ, HL, NL, PL) 1
2
3
4
5
6
2026
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. van der Werf M, Hanssen M, Köhler S, et al. Systematic review and collaborative recalculation of 133 693 incident cases of schizophrenia. Psychol Med 2014; 44: 9–16. Kessing LV. Gender differences in the phenomenology of bipolar disorder. Bipolar Disord 2004; 6: 421–25. Witt K, van Dorn R, Fazel S. Risk factors for violence in psychosis: systematic review and meta-regression analysis of 110 studies. PloS One 2013; 8: e55942. Swann AC, Lijffijt M, Lane SD, Kjome KL, Steinberg JL, Moeller FG. Criminal conviction, impulsivity, and course of illness in bipolar disorder. Bipolar Disord 2011; 13: 173–81. Leucht S, Tardy M, Komossa K, et al. Antipsychotic drugs versus placebo for relapse prevention in schizophrenia: a systematic review and meta-analysis. Lancet 2012; 379: 2063–71.
Shortage of purified protein derivative for tuberculosis testing We wish to highlight a crucial shortage of purified protein derivative (PPD), the test substance used for tuberculin skin testing, which is substantially affecting capabilities and practices for tuberculosis screening in several countries across Europe. Screening to identify individuals with latent tuberculosis infection who might benefit from preventive therapy is a key strategy of tuberculosis control programmes in most countries with low tuberculosis prevalence.1 Children are at particular risk if screening is not implemented, because of an increased rate of progression from latent tuberculosis infection to active tuberculosis without preventive therapy compared with adults.2 In June, 2014, a member of the Paediatric Tuberculosis Network European Trials Group (ptbnet, composed of clinicians and researchers with a special interest in paediatric tuberculosis), first highlighted issues with the supply of PPD in Austria to our network. Within 48 h, 12 other members from eight European countries reported similar problems, prompting us to collect more data across the network via a short questionnaire. 35 physicians from 23 European countries contributed data. 21 (60%) based in 14 different countries reported a shortage of PPD. Most (17) physicians reporting a shortage were using RT 23 (Statens Serum Institute Denmark); fewer reported shortages of Tubertest (Sanofi Pasteur, n=2), and PPD Tuberculin (BulBio, n=1; St Petersburg Institute of Vaccines and Sera, n=1). 13 (37%) physicians reported changes in practices resulting from the PPD shortage, such as restriction of remaining supplies to patients
at greatest risk, or replacement of tuberculin skin testing with interferon-gamma release assay testing, with substantial implications for costs and logistics and potentially inferior sensitivity in young children.3 Worldwide, only a small number of manufacturers produce PPD. No communication from PPD manufacturers alerted the community to this shortfall. We therefore believe that monitoring of PPD production by a supranational agency is needed to ensure that supplies meet clinical demand for a test reagent that remains crucial to public health. We declare no competing interests.
Marc Tebruegge, Matthias Bogyi, Antoni Soriano-Arandes, *Beate Kampmann on behalf of the Paediatric Tuberculosis Network European Trials Group [email protected] Department of Paediatric Infectious Diseases, University Hospital Southampton NHS Foundation Trust (MT); Department of Paediatrics, Wilhelminenspital, Vienna, Austria (MB); Programa Especial de Malalties Infeccioses Vall d’HebronDrassanes, Barcelona, Spain (AS-A); and Academic Department of Paediatrics, St. Mary’s Campus, Imperial College London, London W2 1PG, UK (BK) 1
2
3
Pai M, Denkinger CM, Kik SV, et al. Gamma interferon release assays for detection of Mycobacterium tuberculosis infection. Clin Microbiol Res 2014; 2: 3–20. Newton SM, Brent AJ, Anderson S, Whittaker E, Kampmann B. Paediatric tuberculosis. Lancet Infect Dis 2008; 8: 498–510. Connell TG, Tebruegge M, Ritz N, Bryant P, Curtis N. The potential danger of a solely interferon-gamma release assay-based approach to testing for latent Mycobacterium tuberculosis infection in children. Thorax 2011; 6: 263–64.
www.thelancet.com Vol 384 December 6, 2014
We declare no competing interests.
*Michael Dettling, Ion Anghelescu [email protected] Department of Psychiatry and Psychotherapy, University Clinic Charite, Berlin 14050, Germany (MD); and Clinic Dr Fontheim, Liebenburg, Germany (IA) 1
2
3
4
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. Malhi GS, Chengappa KN, Gershon S, Ghaemi SN. Atypical mood stabilizers: a new role for neuroleptics? Bipolar Disord 2011; 13: 583–86. van der Werf M, Hanssen M, Köhler S, et al. Systematic review and collaborative recalculation of 133 693 incident cases of schizophrenia. Psychol Med 2014; 44: 9–16. Sherazi R, McKeon P, McDonough M, Daly I, Kennedy N. What’s new? The clinical epidemiology of bipolar I disorder. Harv Rev Psychiatry 2006; 14: 273–84.
risk of violence and crime rates, leading to an inherent selection bias in the study sample? 2 Moreover, type and phase of illness were not mentioned, which are both known to affect the risk of violent behaviour and subsequent crime rates.3 Sufficient evidence exists to suggest that substance misuse complicates and confounds aggressive behaviour across the range of mental disorders.4 The authors did mention rates of substance misuse, but their possible confounding effects deserve more discussion. The paradoxical issue of benzodiazepine both inducing and reducing agitation in patients was also left unaddressed. In their study, Fazel and colleagues also showed additional benefit of the use of an antipsychotic rather than a mood stabiliser for reduction of violence and crime rates. However, they did not mention whether the patients were on single or multiple antipsychotics, which could affect study outcomes and results interpretation. Finally, their claim of the study’s superiority compared with previous works should be taken with caution. We feel that the authors could have substantially increased the validity of their findings by asking a few questions from patients or their immediate caregivers related to compliance, domestic violence, effect of hospitalisation, and substance misuse. We declare no competing interests.
*Mohit Varshney, Rishab Gupta [email protected] Department of Psychiatry, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, Delhi 110029, India 1
2
Seema Fazel and colleagues1 deserve credit for their epidemiological study. However, some issues merit further discussion. Were non-compliant patients excluded from withinindividual analyses, thus missing patients who might have had severe psychopathology and, hence, a high www.thelancet.com Vol 384 December 6, 2014
3
4
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. Hodgins S, Riaz M. Violence and phases of illness: differential risk and predictors. Eur Psychiatry 2011; 26: 518–24. Bobes J, Fillat O, Arango C. Violence among schizophrenia out-patients compliant with medication: prevalence and associated factors. Acta Psychiatr Scand 2009; 119: 218–25. Dolan M, O’Malley K, McGregor K. The role of psychopathic traits and substance abuse in predicting violent victimization in patients with schizophrenia spectrum disorders. Personal Ment Health 2013; 7: 28–38.
Authors’ reply We are grateful to Michael Dettling and Ion Anghelescu for raising clinically relevant questions about the drugs that we investigated in our study.1 We selected antipsychotics and mood stabilisers, but not hypnotics (including benzodiazepines) because we focused on long-term outcomes associated with widely prescribed psychiatric drugs, whereas hypnotics are typically prescribed for short periods. We studied antipsychotics and mood stabilisers together for the primary analyses because their coadministration is common, at least in schizophrenia-spectrum disorders. Additionally, we investigated drugs separately according to standard drug classes, rather than to their potential effects. Thus, second-generation antipsychotics with possible mood-stabilising effects were investigated as antipsychotics. As shown in table 3 of our study,1 both first-generation and second-generation antipsychotics were associated with substantial reductions in violent crime in bipolar disorder (hazard ratio 0·52, 95% CI 0·39–0·92). The sex ratio of our sample shows that we used a national patient register with full coverage for outpatients and inpatients. Nevertheless, although the authors of the epidemiological investigation2 cited by Michael Dettling and Ion Anghelescu do not report prevalence ratios for schizophrenia, the female to male incidence rate ratios for the age groups <20 years (0·53, 95% CI 0·41–0·69) and 20–29 years (0·47, 0·41–0·54) are similar to the female to male prevalence ratios in our sample (0·51, 0·49–0·53), which consisted mostly of those aged 20–29 years. In bipolar disorder, the prevalence ratio in our sample was 1·74 (1·68–1·80), an over-representation of women consistent with register-based studies in other countries, such as Denmark.3 Regarding Mohit Varshney and Rishab Gupta’s question about non-compliant patients exclusion—
Robert Brook/Science Photo Library
outbursts. We would like to expand the discussion with some clinical and drug-based remarks. First, to group drugs into antipsychotics and mood stabilisers, or hypnotics, seems uncommon for clinicians because hypnotics represent normal comedication for both antipsychotics and mood stabilisers, at least for acute psychotic and aggressive states. Do any data derived from this study exclude this context? Second, some atypical antipsychotics have mood-stabilising effects.2 Were any data collected from bipolar patients using only atypical antipsychotics as mood stabilisers? Third, the sex distribution in the study for both schizophrenia and bipolar disorders differs obviously from established sex distributions known for these diseases.3,4 Do Fazel and colleagues have an explanation for this finding? To better understand this difference, could this sex discrepancy be seen in the type of violent crime patients diagnosed with schizophrenia and bipolar disorder have been arrested for?
2025
Correspondence
CDC/Science Photo Library
these individuals were central to our study design, which included investigation of rates of violent crime when individuals were taking drugs compared with when they were not. In relation to multiple antipsychotic use, prevalence was low in our sample (0·92%) during the study duration. Other factors could mediate the associations that we report, and we agree that this fact warrants further investigation with different approaches. The type and phase of illness could be relevant in schizophrenia4 and, possibly, bipolar disorder. 5 However, inclusion of sensitive markers and use of caregivers as additional informants do not lend themselves to population-based studies. These studies are necessary to investigate outcomes such as violent crime, suicide, and premature mortality because they need substantially larger sample sizes and longer follow up than have been possible in most randomised controlled trials so far.6
For the ptbnet see http://www. tb-net.org/index.php/ptbnet
NL has received a speaker’s honorarium from Lundbeck in June, 2013. All other authors declare no competing interests.
*Seena Fazel, Johan Zetterqvist, Henrik Larsson, Niklas Långström, Paul Lichtenstein [email protected] Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK (SF); and Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden (JZ, HL, NL, PL) 1
2
3
4
5
6
2026
Fazel S, Zetterqvist J, Larsson H, Långström N, Lichtenstein P. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014; 384: 1206–14. van der Werf M, Hanssen M, Köhler S, et al. Systematic review and collaborative recalculation of 133 693 incident cases of schizophrenia. Psychol Med 2014; 44: 9–16. Kessing LV. Gender differences in the phenomenology of bipolar disorder. Bipolar Disord 2004; 6: 421–25. Witt K, van Dorn R, Fazel S. Risk factors for violence in psychosis: systematic review and meta-regression analysis of 110 studies. PloS One 2013; 8: e55942. Swann AC, Lijffijt M, Lane SD, Kjome KL, Steinberg JL, Moeller FG. Criminal conviction, impulsivity, and course of illness in bipolar disorder. Bipolar Disord 2011; 13: 173–81. Leucht S, Tardy M, Komossa K, et al. Antipsychotic drugs versus placebo for relapse prevention in schizophrenia: a systematic review and meta-analysis. Lancet 2012; 379: 2063–71.
Shortage of purified protein derivative for tuberculosis testing We wish to highlight a crucial shortage of purified protein derivative (PPD), the test substance used for tuberculin skin testing, which is substantially affecting capabilities and practices for tuberculosis screening in several countries across Europe. Screening to identify individuals with latent tuberculosis infection who might benefit from preventive therapy is a key strategy of tuberculosis control programmes in most countries with low tuberculosis prevalence.1 Children are at particular risk if screening is not implemented, because of an increased rate of progression from latent tuberculosis infection to active tuberculosis without preventive therapy compared with adults.2 In June, 2014, a member of the Paediatric Tuberculosis Network European Trials Group (ptbnet, composed of clinicians and researchers with a special interest in paediatric tuberculosis), first highlighted issues with the supply of PPD in Austria to our network. Within 48 h, 12 other members from eight European countries reported similar problems, prompting us to collect more data across the network via a short questionnaire. 35 physicians from 23 European countries contributed data. 21 (60%) based in 14 different countries reported a shortage of PPD. Most (17) physicians reporting a shortage were using RT 23 (Statens Serum Institute Denmark); fewer reported shortages of Tubertest (Sanofi Pasteur, n=2), and PPD Tuberculin (BulBio, n=1; St Petersburg Institute of Vaccines and Sera, n=1). 13 (37%) physicians reported changes in practices resulting from the PPD shortage, such as restriction of remaining supplies to patients
at greatest risk, or replacement of tuberculin skin testing with interferon-gamma release assay testing, with substantial implications for costs and logistics and potentially inferior sensitivity in young children.3 Worldwide, only a small number of manufacturers produce PPD. No communication from PPD manufacturers alerted the community to this shortfall. We therefore believe that monitoring of PPD production by a supranational agency is needed to ensure that supplies meet clinical demand for a test reagent that remains crucial to public health. We declare no competing interests.
Marc Tebruegge, Matthias Bogyi, Antoni Soriano-Arandes, *Beate Kampmann on behalf of the Paediatric Tuberculosis Network European Trials Group [email protected] Department of Paediatric Infectious Diseases, University Hospital Southampton NHS Foundation Trust (MT); Department of Paediatrics, Wilhelminenspital, Vienna, Austria (MB); Programa Especial de Malalties Infeccioses Vall d’HebronDrassanes, Barcelona, Spain (AS-A); and Academic Department of Paediatrics, St. Mary’s Campus, Imperial College London, London W2 1PG, UK (BK) 1
2
3
Pai M, Denkinger CM, Kik SV, et al. Gamma interferon release assays for detection of Mycobacterium tuberculosis infection. Clin Microbiol Res 2014; 2: 3–20. Newton SM, Brent AJ, Anderson S, Whittaker E, Kampmann B. Paediatric tuberculosis. Lancet Infect Dis 2008; 8: 498–510. Connell TG, Tebruegge M, Ritz N, Bryant P, Curtis N. The potential danger of a solely interferon-gamma release assay-based approach to testing for latent Mycobacterium tuberculosis infection in children. Thorax 2011; 6: 263–64.
www.thelancet.com Vol 384 December 6, 2014