Received: 20 January 2017 Revised: 30 April 2017 Accepted: 2 May 2017

DOI: 10.1002/hup.2606

SPECIAL ISSUE ON NOVEL PSYCHOACTIVE SUBSTANCES

The risk of violence associated with novel psychoactive

substance misuse in patients presenting to acute mental health
services
Abu Shafi1,2 | Paul Gallagher1 | Neil Stewart1 | Giovanni Martinotti2 | Ornella Corazza2

1
Highgate Mental Health Centre, Camden and
Islington NHS Foundation Trust, London, UK
2
College Lane Campus, University of
Hertfordshire, Hatfield, UK
Correspondence
Dr. Abu Shafi, Highgate Mental Health Centre,
Camden and Islington NHS Foundation Trust,
Dartmouth Park Hill, London N19 5EX, UK.
Email: abu.shafi@nhs.net
Abstract
Novel psychoactive substance (NPS) availability is increasing at a phenomenal rate and so is their
associated misuse. Currently, the harms associated with NPS misuse are unclear. This report
presents the initial findings of a continuing study looking into the impact of NPS misuse on
admissions to an acute mental health facility in London. Violence preadmission and violence
during admission have been found to be significant in NPS misusers compared to non‐NPS
substance misusers. Our findings have important consequences for acute mental health,
emergency, and public health services. NPS misusers require specific assessment and risk
management to reduce the risk of violence towards others.

KEY WORDS

harm, legal highs, novel psychoactive substances, synthetic cannabinoids, violence

1 | INTRODUCTION (European Monitoring Centre for Drugs and Drug Addiction

Novel psychoactive substance (NPS) misuse is emerging as a serious
threat to the mental and physical health of users.(Gray, Bressington,
Hughes, & Ivanecka, 2016) NPS are associated with a greater risk of
harm and a rise in communicable disease rates.(Stanley, Mogford,
Lawrence, & Lawrie, 2016).
The United Nations Office for Drugs and Crime (UNODC)
defines NPS as “substances of abuse, either in a pure form or a
preparation, that are not controlled by the 1961 Single Convention
on Narcotic Drugs or the 1971 Convention on Psychotropic
Substances, but which may pose a public health threat” (UNODC
Early Warning Advisory (EWA) on New Psychoactive Substances
(NPS), n.d.). The UNODC has reported an unprecedented rise in
the number of NPS, identifying more than 600 by December 2015,
and declared their use as a significant risk to public health and a
challenge to drug policy (United Nations Office on Drugs and Crime
(UNODC), n.d.).
A similar trend has been documented by the European Monitoring
Centre for Drugs and Drugs Addiction (EMCDDA), and they report
that the most common NPS available are synthetic cannabinoids
(European Monitoring Centre for Drugs and Drug Addiction
(EMCDDA), n.d.‐a). An EMCDDA alert in 2016 highlighted 13 deaths
in Europe associated with the synthetic cannabinoid MDMB‐CHMICA
(EMCDDA), n.d.‐b).
In the United Kingdom, the two main legislative statutes regulating
the availability of drugs, the Medicines Act 1968 and the Misuse of
Drugs Act 1971, have been recently supported by the enactment of
the Psychoactive Substances Act 2016, which came into force on
26th May 2016. This new legislation defines a psychoactive substance
as “any substance which is capable of producing a psychoactive effect
in a person who consumes it … by stimulating or depressing the
person's central nervous system, it affects the person's mental
functioning or emotional state” (Home Office, n.d.). Until this date,
NPS had been readily available to buy from certain high street shops
and commonly known as “legal highs.”
NPS misuse is a growing concern for mental health services.
Patients with a mental health diagnosis have an increased rate of
comorbid substance abuse compared with the general population
(Crawford, 1996). Studies have revealed similar observations with
NPS misusers (Martinotti, Lupi, Acciavatti, et al., 2014; Toftdahl,
Nordentoft, & Hjorthoj, 2015). Mental health patients with comorbid
substance misuse have poorer clinical outcomes, pose a greater finan-
cial burden for providers (Dickey & Azeni, 1996), and have a higher
number of admissions than those who do not have substance misuse
problems (Bartels, Drake, & Wallach, 1995). General substance misuse
has been found to be associated with the development of mental

Hum Psychopharmacol Clin Exp. 2017;32:e2606. wileyonlinelibrary.com/journal/hup Copyright © 2017 John Wiley & Sons, Ltd. 1 of 4
2 of 4 SHAFI ET AL.

health signs and symptoms in susceptible users (Brady & Sinha, 2007), TABLE 1 Demographics of sample
and NPS misuse has been specifically associated with the presentation No NPS
of severe mental, behavioural, and physical signs and symptoms Total sample NPS use use
Variable (442) (58) (384)
(Zanda, Fadda, Chiamulera, Fratta, & Fattore, 2016).
At the end of 2015, there was growing anecdotal evidence at the Age (years) 43.7 (15.4) 36.2 (9.8) 44.9 (15.8)

Camden and Islington NHS Foundation Trust that substance misusers Gender (male) 244 43 201

were being admitted with a clinical presentation of acute psychosis, A&E presentation 256 32 224

thought disorder, disturbed perception, agitation, and aggression that Police involvement 184 29 155

were far more extreme and severe in comparison than observed Sectioned 191 30 161

previously. Patients were admitting to the use of “legal highs.” Initial Opiate use 55 18 37

observations were reported (Shafi, Gallagher, Stewart, & Metastasio, Cocaine use 99 29 70

n.d.). A research group was created in January 2016 to compare and Amphetamine use 44 22 22

analyse the effect of NPS use with non‐NPS substance misuse on Cannabis use 112 21 91

the mental health of patients being admitted with deterioration in their Alcohol use 160 32 128

mental state. The aim of this group was to study the particular Polysubstance misuse 38 15 43

characteristics of NPS misusers and to explore if there were similarities Previous violence 164 41 123

and differences between them and non‐NPS misusers. Forensic history 145 35 110
Substance misuse 44 14 30
Psychosis 228 33 195
2 | METHODS Mania 54 4 50
Depression 44 3 41
All admitted patients were initially screened for substance misuse. The Anxiety disorder 19 2 17
admission notes were then further data‐mined from the electronic
Personality disorder 57 9 48
medical record by five researchers (PG, AB, HC, TC‐J, and KM) and
Dual diagnosis 43 14 29
their results checked by another researcher (AS). A database was PICU 50 20 30
created and admission criteria for analysis selected, including process Seclusion 28 8 20
of admission, current diagnosis, use of recreational substances,
Rapid tranquilisation 70 26 44
investigations, treatment, violence preadmission, violence during
Antipsychotic 347 45 302
admission, length of stay, and readmission. We report on the database Mood Stabiliser 64 11 53
results obtained from March to September 2016.
Antidepressant 108 9 99
Benzodiazepine 167 38 129

3 | RESULTS Violence preadmission 125 46 79

Violence during admission 107 46 61

Statistical analysis was completed using Stata IC version 12.1. Aside Admission length (days) 40.8 (32.7) 50.4 (33.5) 39.4 (32.4)

from admission length, there were no missing data. Where the patient Readmission 47 8 39

had not been discharged within 90 days of admission; admission length

was considered as 90 days. Primary outcomes of violence preadmis-
sion, violence during admission, and subsequent readmission were
analysed using logistic regression. Admission length was analysed
using linear regression.
Further statistical analysis was conducted to investigate the rela-
tionship between all data collected and the outcome variables to check
for confounding.
The sample comprised 442 admissions; 244 (55.2%) of these were
male. The mean age was 43.7 years, standard deviation 15.4 years. Of
these admissions, 256 (57.9%) presented to accident and emergency
departments prior to admission, 184 (41.6%) admissions involved
police at some point in the process, and 191 (43.2%) were admitted
under a section of the Mental Health Act (MHA). In this sample, 58
(13.1%) had documented use of NPS. Of note, 43 members of the
sample had both a substance misuse diagnosis and a mental health
diagnosis, referred here as a dual diagnosis (Table 1).
The population using NPS differed significantly from those who
were not. In this sample, they were notably younger (p < .001) and
more likely to be male (OR 1.61, 95% CI [1.18, 2.20]). They were also
more likely to be admitted with police involvement, although this asso-
ciation was less significant (p = .162, OR 1.48, 95% CI [0.85, 2.57]).
Similarly, they were more likely to be admitted under section of the
MHA (p = .162, OR 1.48, 95% CI [0.85, 2.58]; Table 2).
The use of all illicit substances was more common in the group
using NPS, with reported amphetamine use being most strongly asso-
ciated (p < .001). Similarly, opiate and cocaine use were associated with
NPS use (p < .001). Alcohol and cannabis use were less strongly asso-
ciated with NPS use. The most common category of NPS misuse
reported in our sample study was synthetic cannabinoids (Spice and
Black Mamba, 91%), followed by synthetic cathinones (7%).
Diagnoses more common amongst NPS users were substance mis-
use disorders (p < .001), psychosis (p = .475), and personality disorder
(p = .564). Diagnoses less common included mania (p = .176), depres-
sion (p = .189), and anxiety disorders (p = .711). Additionally, NPS
use was associated with previous violence or aggression in the risk
assessment and forensic history.
On initial analysis, reported use of NPS was strongly associated
with violence preadmission (OR 14.7, 95% CI [7.43, 29.08]) and
SHAFI ET AL. 3 of 4

TABLE 2 Associations with NPS misuse similar. Use of any illicit substance or alcohol was also associated with
Variable Probability Effect size violence during admission. These factors were incorporated into the
logistic regression. After adjustment for these factors, the strength of
Age (years) <.001 −8.66 (−12.86 to −4.46)
association was again reduced but remained significant: OR 15.98,
Admission length (days) .016 11.05 (2.05 to 20.05)
95% CI [7.05, 36.20]. There was no statistically significant interaction
Gender (male) .003 1.61 (1.18 to 2.20)
between the factors evaluated.
A&E presentation .167 0.88 (0.50 to 1.53)
Admission length was noted to be significantly associated with
Police involvement .162 1.48 (0.85 to 2.57)
both use of the MHA during admission and violence (both during or
Sectioned .162 1.48 (0.85 to 2.58)
prior to admission). Patients under section or those who had been
Opiate use <.001 4.22 (2.20 to 8.10)
violent had considerably longer admissions. Use of any illicit
Cocaine use <.001 4.47 (2.51 to 7.96)
substances by themselves did not have any significant association;
Amphetamine use <.001 10.05 (5.08 to 19.91)
with the exception of amphetamines, illicit substances and alcohol
Cannabis use .043 1.82 (1.02 to 3.28)
demonstrated a reduction in admission length, albeit nonsignificant.
Alcohol use .002 2.46 (1.41 to 4.31)
Diagnostic category had a notable effect on admission length:
Polysubstance misuse <.001 4.26 (2.39 to 7.59)
Psychosis and mania diagnoses were associated significantly with a
Violence preadmission <.001 14.70 (7.43 to 29.08)
longer admission. Comparatively, depression, anxiety, and personality
Violence during admission <.001 20.23 (10.13 to 40.41)
disorders were associated with significant reduction in admission
Previous violence <.001 5.10 (2.78 to 9.33)
length. After adjustment for violence prior to admission, there was
Forensic history <.001 3.78 (2.13 to 6.68)
no longer a significant relationship between use of NPS and admission
PICU <.001 6.19 (3.21 to 11.95)
length, although the effect of longer admissions remained present
Seclusion .017 2.90 (1.21 to 6.94)
(effect size 6.07, 95% CI [−3.90, 16.03]). Notably, there was significant
Rapid tranquilisation <.001 6.28 (3.43 to 11.50)
interaction between violence prior to admission and reported NPS use,
Antipsychotic .855 0.94 (0.48 to 1.83)
with the presence of both significantly prolonging admission (effect
Mood stabiliser .300 1.46 (0.71 to 3.00)
size 28.45, 95% CI [6.50, 50.40]).
Antidepressant .094 0.53 (0.25 to 1.12)
Forty‐seven of the admissions examined resulted in a subsequent
Benzodiazepine <.001 3.76 (2.10 to 6.72)
admission within the follow‐up period. Factors influencing readmission
Substance misuse <.001 3.67 (1.81 to 7.45)
at 90 days were examined. Notably few factors showed a strong
Psychosis .475 1.23 (0.70 to 2.13)
association with readmission. Use of MHA during the admission
Mania .176 0.48 (0.17 to 1.39)
(p = .012, OR 0.41, 95% CI [0.21, 0.82]) and presence of a psychotic
Depression .189 0.45 (0.13 to 1.49)
diagnosis (p = .028, OR 0.49, 95% CI [0.26 to 0.92]) were both
Anxiety disorder .711 0.75 (0.17 to 3.35)
associated with reduced likelihood of readmission. Comparatively,
Personality disorder .564 1.26 (0.58 to 2.72)
presence of dual diagnosis (p = .025, OR 2.51, 95% CI [1.12, 5.64])
Dual diagnosis <.001 3.89 (1.91 to 7.93)
and personality disorder (p = .002, OR 3.08, 95% CI [1.51, 6.29]) were
Readmission .404 1.41 (0.63 to 3.20)
associated with a greater instance of readmission.

violence during admission (OR 20.23, 95% CI [10.13, 40.41]). There

was a significant association with admission length, with NPS use
associated with a longer duration of admission (p = .016, effect size
11.05, 95% CI [2.05, 20.05]). Although readmission occurred more
than would be expected within the NPS group, this was not statistically
significant (p = .404, OR 1.41, 95% CI [0.63, 3.20]),
Further analysis of other factors associated with violence prior to
admission included younger age, male gender, and a prior history of
violence. Use of any illicit substances (including alcohol) was
significantly associated with violence, with reported use of more than
one substance or presence of a dual diagnosis being more strongly
associated. After incorporating these factors into the logistic regres-
sion, the relationship between reported NPS and violence prior to
admission remained statistically significant, although the strength of
association was reduced: OR 8.67, 95% CI [4.06, 18.51]. There was
no statistically significant interaction between the differing variable.
Analysis of the factors associated with violence during admission
included younger age and previous violence. Previous violence was
more strongly associated with violence during admission than prior
to admission. Notably, the association with a dual diagnosis remained
4 | DISCUSSION
The findings reflect that mental health care alongside emergency and
public health services are facing an increasing and unprecedented risk
of violence from NPS misusers compared to non‐NPS substance
misusers. This link between NPS misuse and violence requires further
research; one hypothesis when examining synthetic cannabinoids is
that the nature in which these NPS bind as full agonists at the
cannabinoid receptors compared to the partial agonist action of
cannabis causes a stronger and more prolonged effect, leading to an
increase in violence. Furthermore, synthetic cannabinoids do not
contain the molecule Cannabidiol, which is present in cannabis and
has neuroprotective and antioxidant properties, and this may lead to
an increase in violence. This increased risk of violence is leading to
enormous challenges on how to manage NPS misusers safely as well
as creating safe working environments for health care professionals.
Targeted risk assessment and management strategies are required to
reduce the risk of violence from NPS misusers during the admission
process. Members of the public are also increasingly at risk, especially
4 of 4
carers and relatives of NPS misusers. The prison services have
reported a rise in mortality associated with NPS misuse (Prisons and
Probation Ombudsman Independent Investigations, 2016).
The findings suggest NPS users having a longer length of stay in
hospital and a higher readmission rate compared to non‐NPS
substance misusers. This has implications for bed utilisation and
service delivery within an already strained resource environment and
limited mental health beds, especially intensive care beds that are
reserved for those who present with the most severe deterioration
of mental health, which NPS misusers are. Resource planning will be
critical in the future if NPS users are found to be consistently utilising
beds for longer inpatient stays and being readmitted more quickly.
With limited acute mental health beds available, and a rise in NPS users
presenting with acute mental deterioration, care pathways will need to
be identified to manage these patients appropriately and safely.
The study also suggests that admission assessments need to be
improved due to the current lack of knowledge and understanding of
NPS misuse. This has led to the development of a 10‐question NPS
risk assessment and management tool to support health care
professionals with their assessments and risk planning. This tool is
currently being validated, but initial feedback has been very positive.
Health care professionals require more education and support in the
management of NPS misuse as this was frequently reported to our
research group.
The findings reported are, to the best of our knowledge,
amongst the first to be published on this particular subject, and it
is our opinion that they should be shared and disseminated as early
and widely as possible to alert professionals to the current risks
posed by NPS misuse and to encourage similar studies and collabo-
ration in the future.
REFER ENC ES
Bartels, S. J., Drake, R. E., & Wallach, M. A. (1995). Long‐term course of
substance use disorders among patients with severe mental illness.
Psychiatric Services, 46, 248–251.
Brady, K. T., & Sinha, R. (2007). Co‐occurring mental and substance use
disorders: The neurobiological effects of chronic stress, focus. Am
Psychiatr Publ, 5, 229–239.
Crawford, V. (1996). Comorbidity of substance misuse and psychiatric
disorders. Current Opinion in Psychiatry, 9, 231–234.
Dickey, B., & Azeni, H. (1996). Persons with dual diagnoses of substance
abuse and major mental illness: Their excess costs of psychiatric care.
American Journal of Public Health, 86, 973–977.
SHAFI ET AL.
European Monitoring Centre for Drugs and Drug Addiction (EMCDDA).
Health responses to new psychoactive substances 2016. http://www.
emcdda.europa.eu/system/files/publications/2812/TD0216555ENN.
pdf (accessed 20th Sep 2016).
European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) EU
early warning system. Alert 8 February 2016. http://www.emcdda.
europa.eu/system/files/publications/2873/2016.4528_WEB.pdf
(accessed 20th Sep 2016).
Gray, R., Bressington, D., Hughes, E., & Ivanecka, A. (2016 Mar). A system-
atic review of the effects of novel psychoactive substances ‘legal highs’
on people with severe mental illness. Journal of Psychiatric and Mental
Health Nursing, 1.
Home Office. Psychoactive substances bill. http://www.legislation.gov.uk/
ukpga/2016/2/contents/enacted/data.htm (accessed 20 Sep 2016).
Martinotti, G., Lupi, M., & Acciavatti, T. (2014). Novel psychoactive
substances in young adults with and without psychiatric comorbidities.
Bio Med Res Int, 2014, 815424.
Prisons and Probation Ombudsman Independent Investigations. 2016
Sep 23. http://www.ppo.gov.uk/wp‐content/uploads/2016/09/PPO‐
amends‐NPS‐prison‐death‐figure‐news‐release.pdf. (accessed 10th
October 2016)
Shafi, A., Gallagher, P., Stewart, N., & Metastasio, A. Psychoactive Sub-
stances Act 2016‐unknown unknowns. E‐letter to the editor. The
British Journal of Psychiatry.
Stanley, J. L., Mogford, D. V., Lawrence, R. J., & Lawrie, S. M. (2016 May 1).
Use of novel psychoactive substances by inpatients on general adult
psychiatric wards. BMJ Open, 6(5), e009430.
Toftdahl, N. G., Nordentoft, M., & Hjorthoj, C. (2015). Prevalence of sub-
stance use disorders in psychiatric patients: A nationwide Danish
population‐ based study. Social Psychiatry and Psychiatric Epidemiology,
51, 125–140 https://doi.org/10.1007/s00127‐015‐1104‐4.
United Nations Office on Drugs and Crime (UNODC). Global Smart Update
2016. http://www.unodc.org/documents/scientific/Global‐SMART‐
Update‐2016‐vol‐16.pdf (accessed 16 Sep 2016).
UNODC Early Warning Advisory (EWA) on New Psychoactive Substances
(NPS) https://www.unodc.org/LSS/Page/NPS (accessed 16 Sep 2016).
Zanda, M. T., Fadda, P., Chiamulera, C., Fratta, W., & Fattore, L. (2016 Apr).
Methoxetamine, a novel psychoactive substance with serious adverse
pharmacological effects: A review of case reports and preclinical
findings. Behavioural Pharmacology, 28.
How to cite this article: Shafi A, Gallagher P, Stewart N,
Martinotti G, Corazza O. The risk of violence associated with
novel psychoactive substance misuse in patients presenting
to acute mental health services. Hum Psychopharmacol Clin
Exp. 2017;32:e2606. https://doi.org/10.1002/hup.2606
Copyright of Human Psychopharmacology: Clinical & Experimental is the property of John
Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to
a listserv without the copyright holder's express written permission. However, users may
print, download, or email articles for individual use.