Journal of Public Health | Vol. 40, No. 2, p. e157–e163 | doi:10.

1093/pubmed/fdx049 | Advance Access Publication May 12, 2017

The involvement of alcohol in hospital-treated self-harm

and associated factors: findings from two national registries
Eve Griffin1, Ella Arensman1,2, Ivan J Perry2, Brendan Bonner3, Denise O’Hagan3,
Caroline Daly1, Paul Corcoran2,4
1
National Suicide Research Foundation, University College Cork, Cork, T12 XF62, Ireland

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

2
Department of Epidemiology and Public Health, University College Cork, Cork, T12 XF62, Ireland
3
Public Health Agency, Belfast, BT2 8BS, Northern Ireland
4
Department of Obstetrics and Gynaecology, University College Cork, Cork, T12 YE02, Ireland
Address correspondence to Eve Griffin, E-mail: evegriffi[email protected]

ABSTRACT

Background Alcohol is often involved in hospital-treated self-harm. Therefore it is important to establish the role of alcohol in self-harm as well
as to identify associated factors, in order to best inform service provision.

Methods Data on self-harm presentations to hospital emergency departments in Ireland and Northern Ireland from April 2012 to December
2013 were analysed. We calculated the prevalence of alcohol consumption in self-harm. Using Poisson regression models, we identified the
factors associated with having consumed alcohol at the time of a self-harm act.

Results Alcohol was present in 43% of all self-harm acts, and more common in Northern Ireland (50 versus 37%). The factors associated with
alcohol being involved were being male, aged between 25 and 64 years, and having engaged in a drug overdose or attempted drowning.
Presentations made out-of-hours were more likely to have alcohol present and this was more pronounced for females. Patients with alcohol on
board were also more likely to leave without having been seen by a clinician.

Conclusions This study has highlighted the prevalence of alcohol in self-harm presentations, and has identified factors associated with
presentations involving alcohol. Appropriate out-of-hours services in emergency departments for self-harm presentations could reduce the
proportion of presentations leaving without being seen by a clinician and facilitate improved outcomes for patients.

Keywords alcohol, emergency care, mental health

Introduction well as posing difficulties for medical staff in treating such

patients.6 Co-morbidity and dual diagnoses add further com-
The relationship between alcohol and suicidal behaviour is plexity. Therefore, it is important to establish a profile of
well established.1,2 A recent study found that premature, self-harm patients who present to emergency departments
alcohol-related deaths were nine times more likely among (EDs) following self-harm where alcohol is involved.
those who engage in self-harm compared to the general In this study, data from the National Self-Harm Registry
population.3 Furthermore, acute alcohol consumption is Ireland and the Northern Ireland Registry of Self-Harm are
linked to impulsive suicidal behaviours.4,5 One in ten self- utilized. We have previously established the incidence of
harm patients have an alcohol dependency and alcohol mis-
use is associated with between 23 and 36% of acts.6,7
Eve Griffin, Senior Postdoctoral Researcher
Alcohol is often consumed prior to or during an act of self-
Ella Arensman, Scientific Director
harm, with reported involvement ranging from 26 to 60%.7–9
Ivan J Perry, Professor of Public Health
Alcohol consumed during a self-harm act poses specific Brendan Bonner, Head of Health and Social Well-being Improvement (West)
challenges for the management and assessment of self-harm Denise O’Hagan, Consultant in Public Health Medicine
patients in acute hospital settings. Having alcohol on board Caroline Daly, PhD Researcher
may lead to delayed assessment following a self-harm act as Paul Corcoran, Senior Lecturer

© The Author 2017. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: [email protected] e157
e158 JOURNAL OF PUBLIC HEALTH

hospital-treated self-harm in these countries.8,9 This is the next care following ED attendance is recorded. The categor-
first paper to use data from two national self-harm registries ies in this variable are: admission to a general medical ward;
and the specific objectives were: admission to a psychiatric admission; patient refused admis-
sion/left without being seen by a clinician; discharged from
(i) to establish the extent to which alcohol is involved in ED following treatment.
hospital-treated self-harm in Ireland and Northern For the purposes of this study, alcohol involvement was
Ireland and defined as the intake of alcohol prior to or during the act of
(ii) to investigate what factors are associated with alcohol self-harm. The involvement of alcohol was recorded in a
being involved in self-harm acts. systematic way by the data registration officers, according to

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

the standard operating procedures. Alcohol involvement was
ascertained through hospital case notes—if it was recorded
Methods on registration or by the attending clinician, or if present on
Self-harm data toxicology reports.
Data on hospital-treated self-harm were obtained from both the
National Self-Harm Registry Ireland8 and the Northern Ireland Ethical approval
Registry of Self-Harm.9 All presentations due to self-harm The National Self-Harm Registry Ireland has ethical approval
made to acute hospital EDs in Ireland and Northern Ireland from the National Research Ethics Committee of the Faculty
between 1 April 2012 and 31 December 2013 were included. of Public Health Medicine. Ethical approval for the Northern
Both registries define self-harm as ‘an act with non-fatal out- Ireland Registry of Self-Harm has been granted by the Office
come in which an individual deliberately initiates a non-habitual for Research Ethics in Northern Ireland (ORECNI).
behaviour, that without intervention from others will cause self-
harm, or deliberately ingests a substance in excess of the pre-
Statistical analyses
scribed or generally recognized therapeutic dosage, and which is
aimed at realizing changes that the person desires via the actual As with common outcomes, odds-ratios provided by logistic
or expected physical consequences’.10 regression can be prone to misinterpretation as they may be
The National Self-Harm Registry Ireland is operated by far greater than risk ratios. We used Poisson regression to
the National Suicide Research Foundation and covers all identify the factors independently associated with alcohol
hospital EDs in Ireland. The Northern Ireland Registry of involvement in self-harm as it provides risk ratios—specifically
Self-Harm is operated by the Northern Ireland Public incidence-rate ratios (IRR). All self-harm presentations were
Health Agency. Since April 2012, all hospital EDs across included in the analysis and adjustment was made for the clus-
five Health and Social Care Trusts in Northern Ireland are tering in the data associated with multiple presentations by
included in the registry’s data collection. Data are collected self-harm repeaters.
by independently trained data registration officers who oper- We explored the association between alcohol involvement
ate according to the registries’ standardized operating proce- and the most frequently recorded methods of self-harm in
dures. Identifying cases of self-harm involve a combination the data: drug overdose (ICD-10 codes X60-64); self-cutting
of manually checking consecutive presentations to the EDs, (X78); poisoning (X66-X69); attempted hanging (X70) and
selecting potential cases on the basis of keyword searches attempted drowning (X71). The association between alcohol
and triage coding by hospital staff. Differences between hos- involvement and repetition of self-harm was also explored.
pital systems of recording emergency department presenta- In this instance, a repeat presentation indicates that an indi-
tions mean that flexibility in case finding is required. vidual had made a previous self-harm presentation within
the study period.
We explored whether associations differed by country and
Data items gender (i.e. interaction) due to previously reported differ-
Both registries have a core data set including the following ences in these variables.9,11 Because of the very large sample
variables: gender, date of birth, date and hour of attendance size some interactions reached statistical significance at P <
at hospital, method(s) of self-harm according to the 10th 0.05 but the difference in the magnitude of the associations
revision of the WHO’s International Classification of was very small. To ensure we confined the reporting to non-
Disease codes for intentional injury (X60-X84). Patient initi- trivial differences we applied a higher threshold whereby
als (in an encrypted format) and area of residence, coded to only interaction effects that were highly statistically signifi-
administrative area, are recorded. In addition recommended cant, i.e. at P < 0.001 level, were included in the multivariate
 THE INVOLVEMENT OF ALCOHOL IN HOSPITAL-TREATED SELF-HARM AND ASSOCIATED FACTORS e159

model. Country- and gender-specific associations which Whether alcohol was involved varied according to the primary
were statistically significant in the multivariate model are method of self-harm—being more common in presentations
reported. involving attempted drowning (1.27, 1.15–1.40), self-cutting
(1.08, 1.02–1.13) or intentional drug overdose (Ireland: 1.50,
Sensitivity analysis 1.37–1.64). The association with the latter was less apparent in
Whether alcohol was recorded as being involved in the self- Northern Ireland (1.08, 1.02–1.13). In contrast, presentations
harm presentation depended on relevant information being involving self-poisoning were less likely to have alcohol involved
present in the hospital notes. Such recording fluctuates by hos- (0.77, 0.65–0.91).
Time of presentation to the ED was associated with alco-

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

pital, meaning that associations between patient characteristics
and consuming alcohol vary depending on recording of infor- hol being involved and this marked variation was stronger
mation in each hospital. We tested whether the associations for women. In particular presentations made between mid-
were modified by the hospital’s level of recording of alcohol night and 9 am were most likely to involve alcohol in both
among self-harm presentations. This was done by including men (1.66, 1.58–1.74) and women (2.25, 2.14–2.37).
interaction effects in univariate Poisson regression models. Presentations made at the weekend were also more likely to
All analyses were conducted using Stata 13.0. involve alcohol (1.14, 1.11–1.17).
Self-harm patients brought to the ED by ambulance or by
emergency services were more likely to have consumed alco-
Results hol than those self-presenting (1.39, 1.33–1.44). Compared
Sample characteristics to presentations admitted to a general ward, patients admit-
ted to a psychiatric ward were less likely to have consumed
During the 21-month study period 24 513 individuals made a
alcohol (0.73, 0.65–0.81). Those leaving without being seen
total of 34 429 presentations due to self-harm (Table 1)—
or refusing admission had an elevated risk of having con-
19 831 (58%) were made to EDs in Ireland and 14 598 (42%)
sumed alcohol (1.11, 1.04–1.17). In Ireland, patients dis-
to EDs in Northern Ireland. Approximately 52% of the sam-
charged from the ED were less likely to have consumed
ple were female and 44% were aged between 25 and 44 years.
alcohol (0.86, 0.80–0.91). However the opposite was seen in
The most common method of self-harm recorded was inten-
Northern Ireland where patients discharged from the ED
tional drug overdose (71%). The only other common method
were more likely to have consumed alcohol (1.17, 1.12–1.22).
of self-harm was self-cutting, present in 23% of acts. Fewer
Alcohol being involved in a self-harm act varied depending
than one-third (29%) were repeat presentations and 31%
on whether it was a patient’s first presentation or a repeat pres-
were made by residents of urban areas.
entation. If it was a repeat presentation, men were more likely
to have consumed alcohol (1.08, 1.04–1.13), while female
Involvement of alcohol in self-harm acts
repeaters were less likely to have done so (0.88, 0.82–0.93).
Alcohol was involved in 14 637 (42.5%) presentations and more
so in Northern Ireland than in Ireland (49.6 versus 37.3%).
Men were more likely than women to have consumed alcohol Sensitivity analysis
(46.6 versus 38.8%), and alcohol involvement was most com- The association between each variable and alcohol being
mon among those aged between 45 and 64 years (52.7%). involved was examined for hospitals reporting below average
Alcohol was involved most often in cases of intentional drug levels of alcohol involvement and above average levels (in
overdose (42.5%) and attempted drowning (47.2%). Half of all their respective countries) to see if associations were
patients who left the ED without being seen by clinician/ dependent on the level of reporting of alcohol in hospitals.
refused admission had consumed alcohol (50.5%) (Table 2). There was little evidence to support this, as none of the
interaction effects were associated with P < 0.001. Thus the
Factors associated with alcohol involvement observed associations appear to be robust to the level of
reporting across the hospitals included in the study.
Table 2 also presents the results from the adjusted Poisson
regression. Risk of consuming alcohol as part of the self-
harm act was elevated for patients presenting to EDs in
Discussion
Northern Ireland (IRR = 1.35, 95% CI: 1.25–1.46). Women
were less likely to have consumed alcohol (0.87, 0.84–0.90). Main findings of this study
Patients aged between 25 and 64 years were most likely to The findings of this study have highlighted the role of alcohol
have alcohol on board (45–64 years: 1.56, 1.49–1.63). consumption in hospital-treated self-harm and associated factors.
e160 JOURNAL OF PUBLIC HEALTH

Table 1 Characteristics of self-harm presentations in study sample

Variable Ireland 19 831 (57.6%) Northern Ireland 14 598 (42.4%) All 34 329 (100%)

Gender
Male 9 082 (45.8%) 7 294 (50.0%) 16 376 (47.6%)
Female 10 749 (54.2%) 7 304 (50.0%) 18 053 (52.4%)
Age
<15 years 501 (2.5%) 265 (1.8%) 766 (2.2%)

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

15–24 years 5 733 (28.9%) 4 550 (31.2%) 10 283 (29.9%)
25–44 years 8 999 (45.4%) 6 020 (41.2%) 15 019 (43.6%)
45–64 years 4 108 (20.7%) 3 496 (24.0%) 7 604 (22.1%)
+65 years 490 (2.5%) 267 (1.8%) 757 (2.2%)
Alcohol involved
Yes 7 390 (37.3%) 7 247 (49.6%) 14 637 (42.5%)
Method of self-harma
Drug overdose 13 541 (68.3%) 10 820 (74.1%) 24 361 (70.8%)
Self-cutting 4 632 (23.4%) 3 424 (23.5%) 8 056 (23.4%)
Poisoning 416 (2.1%) 165 (1.1%) 581 (1.7%)
Attempted hanging 1 287 (6.5%) 557 (3.8%) 1 844 (5.4%)
Attempted drowning 594 (3.0%) 147 (1.0%) 741 (2.2%)
Repeat presentation
Yes 4 978 (25.1%) 4 938 (33.8%) 9 916 (28.8%)
City resident
Yes 5 886 (29.7%) 4 875 (33.4%) 10 761 (31.3%)

a
Note: Multiple methods of self-harm may be recorded for a single presentation, therefore the figures may sum to more than 100%.

While self-harm presentations made to hospital EDs in What this study adds
Northern Ireland were more likely to involve alcohol than We found that those who consumed alcohol prior to or dur-
those in Ireland, the factors associated with alcohol con- ing the self-harm act were more likely to present out-of-hours
sumption and self-harm were similar in both countries. and were also more likely to leave the ED without being seen
Those factors identified as being associated with alcohol by a clinician. In particular we found this effect greater among
consumption were being male, aged between 25 and 64 female patients, a result which has not previously been
years and having engaged intentional drug overdose, self- reported on. This highlights the challenge of providing appro-
cutting or attempted drowning. Time of presentation to priate and timely care and treatment for self-harm patients,
the hospital was strongly associated with alcohol consump- particularly out-of-hours. Alcohol intoxication at the time of
tion, with cases involving alcohol more likely to occur out- presentation to the ED provides challenges for the assess-
side of usual working hours and at the weekend. Self-harm ment of patients, and may prolong waiting time. The ED has
patients with alcohol on board were more likely to leave now become an important option for care for patients
the ED without being seen by a clinician. engaging in suicidal behaviour, yet the availability of mental
health teams varies across hospital sites and many do not
have 24/7 or on-call access, as advocated by the NICE
What is already known on this topic guidelines,13 even though most presentations occur out-of-
There is variation in the level of alcohol involvement in self- hours. An observational study in England found that the pro-
harm acts reported in previous studies.1,6,12 Our findings are portion of patients presenting with self-harm receiving an
lower than those reported in a recent English study,7 and con- assessment by the mental health services in the ED varied
sistent with previous findings from Irish data.8 A previous from 22 to 88%14 while a recent audit conducted by the
study of self-harm in Northern Ireland highlighted the influ- Royal College of Emergency Medicine found that few hospi-
ence of alcohol on the timing of self-harm presentations to tals were meeting the target of providing patients with an
the ED9 and its findings are supported here. assessment by mental health services within 1 h of referral.15
 THE INVOLVEMENT OF ALCOHOL IN HOSPITAL-TREATED SELF-HARM AND ASSOCIATED FACTORS e161

Table 2 Factors associated with alcohol being involved in self-harm

Variable Ireland Northern All Incidence rate 95% Confidence P-value

n (% alcohol) Ireland n (% alcohol) ratio interval
n (% alcohol)

Country (ref = Ireland) 7390 (37.3%) 7247 (49.6%) 14 637 (42.5%) 1.35 (1.25–1.46) <0.001
Gender (ref = Male) 3696 (40.7%) 3938 (54.0%) 7 634 (46.6%) 1.00 Ref –
Female 3694 (34.4%) 3309 (45.3%) 7 003 (38.8%) 0.87 (0.84–0.90) <0.001

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

Age (ref = 15–24) 1618 (28.2%) 1770 (38.9%) 3 388 (33.0%) 1.00 Ref –
<15 years 31 (6.2%) 23 (8.7%) 54 (7.1%) 0.29 (0.22–0.39) <0.001
25–44 years 3755 (41.7%) 3208 (53.3%) 6 963 (46.4%) 1.38 (1.32–1.44) <0.001
45–64 years 1853 (45.1%) 2156 (61.7%) 4 009 (52.7%) 1.56 (1.49–1.63) <0.001
65+ years 133 (27.1%) 90 (33.7%) 223 (29.5%) 0.94 (0.82–1.08) 0.411
Drug overdose in Irelanda 5590 (41.3%) – – 1.50 (1.39–1.61) <0.001
Drug overdose in Northern Irelanda – 5512 (50.9%) – 1.08 (1.02–1.15) 0.008
Self-cuttinga 1459 (31.5%) 1608 (47.0%) 3 067 (38.1%) 1.08 (1.02–1.13) 0.004
Poisoninga 99 (23.8%) 54 (32.7%) 153 (26.3%) 0.77 (0.65–0.91) 0.002
Attempted hanginga 392 (30.5%) 269 (48.3%) 661 (35.9%) 0.99 (0.92–1.07) 0.767
Attempted drowninga 262 (44.1%) 88 (59.9%) 350 (47.2%) 1.27 (1.15–1.40) <0.001
Time of attendance for men 917 (30.9%) 899 (40.4%) 1 816 (23.8%) 1.00 Ref –
(ref = 9 am to 6 pm)
6 pm to 12 am 1222 (40.1%) 1284 (52.8%) 2 506 (32.9%) 1.31 (1.25–1.37) <0.001
12 am to 9 am 1550 (50.8%) 1755 (66.5%) 3 305 (43.3%) 1.66 (1.58–1.74) <0.001
Time of attendance for women 857 (22.8%) 694 (28.9%) 1 551 (22.2%) 1.00 Ref –
(ref = 9 am to 6 pm)
6 pm to 12 am 1161 (31.9%) 988 (40.7%) 2 149 (30.7%) 1.78 (1.66–1.90) <0.001
12 am to 9 am 1670 (50.3%) 1627 (65.6%) 3 297 (47.1%) 2.25 (2.14–2.37) <0.001
Weekend presentationb 1321 (41.1%) 2626 (56.9%) 3 947 (50.4%) 1.14 (1.11–1.17) <0.001
Mode of arrival (ref = self-presented) 2107 (28.0%) 1418 (32.6%) 3 525 (29.7%) 1.00 Ref –
Ambulance/ emergency services 5119 (45.8%) 5818 (56.9%) 10 937 (51.1%) 1.39 (1.33–1.44) <0.001
Admission (ref = general admission) 3664 (36.7%) 1672 (20.8%) 6 222 (47.5%) 1.00 Ref –
Psychiatric admission 500 (26.6%) 152 (27.5%) 652 (26.8%) 0.73 (0.65–0.81) <0.001
Refused admission/ Left without being 1374 (46.3%) 153 (57.1%) 2 427 (50.5%) 1.11 (1.04–1.17) 0.001
seen
Discharged from ED (Ireland) 1852 (37.0%) – – 0.86 (0.80–0.91) <0.001
Discharged from ED (Northern Ireland) – 4370 (54.0%) – 1.17 (1.12–1.22) <0.001
Repeat presentation for menc 960 (42.7%) 1 376 (55.4%) 2 336 (49.4%) 1.08 (1.04–1.13) <0.001
Repeat presentation for womenc 820 (30.0%) 1 007 (41.1%) 1 827 (35.3%) 0.88 (0.82–0.93) <0.001
City residentd 2074 (35.2%) 2550 (52.3%) 4 624 (43.0%) 1.02 (0.99–1.05) 0.259

Note: Gender- and country-specific associations due to interaction are shown, where significant at P < 0.001. Analysis adjusted for the clustering in the
data associated with the inclusion of multiple presentations by self-harm repeaters.
a
The reference groups for methods variables are the cases involving other methods (e.g. for drug overdose, the reference group is cases not involving drug
overdose).
b
Cases presenting on weekdays (Monday–Friday) is the reference group.
c
Index act presentations is the reference group.
d
Presentations by residents of rural areas and towns is the reference group.

We found significantly higher levels of alcohol involve- alcohol drinking patterns within the population.16 We have pre-
ment in self-harm presentations in Northern Ireland. The viously found higher rates of self-harm in Northern Ireland9,17
observed difference in alcohol involvement in self-harm and population-based surveys have established a high inci-
between both countries may be related to differences in dence of DSM-IV disorders (particularly Post-Traumatic
e162 JOURNAL OF PUBLIC HEALTH

Stress Disorder) in the general population, which may nega- particular, the findings from this study have identified the
tively impact on mental health.18,19 In particular we found a factors distinguishing self-harm acts with alcohol present to
high proportion of alcohol involvement among residents of those without. Presentations by men and those made outside
Northern Ireland aged between 45 and 54 years (61%), a of normal working hours are strongly associated with alco-
group which have been exposed to the conflict in Northern hol. Furthermore, having alcohol on board is associated
Ireland. with leaving the ED without being seen by a clinician. These
The findings from this study have highlighted the role of findings underline the need for increased awareness of the
alcohol consumption in self-harm, associated with presenta- dual relationship between alcohol misuse and self-harming
tions occurring out of hours and at weekends, in particular behaviour among ED and mental health staff, as well as the

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

presentations leaving without treatment, as well as among importance of 24/7 access to mental health services for
male repeaters. These patterns confirm the need to ensure patients presenting with self-harm
that services are in place to provide adequate care for
patients with issues relating to co-existing alcohol misuse Limitations of this study
and self-harm, as outlined in the NICE guidelines for the A strength of this study is that it is based on data from two
management of self-harm.13 Given existing evidence linking national registries, with a large sample of presentations. The
alcohol misuse and alcohol-related disorders to self-harm, a standardized operation of both registries ensures compar-
range of services is needed to meet the needs of such ability of the data captured, and the overall consistency in
patients—from managing addiction issues associated with findings across both two jurisdictions is an indication of
self-harm through to addressing lower levels of alcohol mis- their robustness. A limitation of this study is related to the
use. Collaboration between mental health teams, addiction reporting of alcohol in self-harm presentations. Alcohol
services and other non-statutory agencies is recommended involvement at the time of the self-harm act was ascertained
where a patient presents with a dual diagnosis.13 Although via clinical notes in each hospital, therefore in most cases
this study did not examine recreational drug misuse it is there was no precise and objective information about the
likely that other forms of substance misuse are also related amount of alcohol consumed (e.g. blood alcohol content).
to self-harm and future research in that area would be use- As a result, the proportion of alcohol involvement varied
ful. It has previously been recommended that mental health slightly between hospitals. To this end, it is likely that per-
staff in the ED, as well as emergency responders, would centages reported here underestimate the extent of alcohol
benefit from training in the management of patients present- involvement in self-harm. Nevertheless, the sensitivity ana-
ing with alcohol intoxication.20 The Northern Ireland lysis conducted confirm that the associations established are
Suicide Prevention Strategy—Protect Life—and the Irish robust to this variation. Unlike previous studies, we did not
National Clinical Programme for the Management of Self- have information on existing diagnoses of alcohol-related
Harm Presentations to Emergency Departments both refer disorders, or information on existing alcohol dependency.
to national training programmes for mental health profes- Such information would help inform the development of
sionals and nurses relating to the care and management of services to meet the needs of this patient group.
self-harm in an ED setting.21,22
These findings also highlight the need to further strengthen
Acknowledgements
efforts to address alcohol consumption at a general population
level. In Northern Ireland, a framework for reducing alcohol The authors would like to thank all involved in the develop-
and drug related harm (2011–16) was published by the ment of the National Self-Harm Registry Ireland and the
Department for Health in 2011.23 In Ireland, a Public Health Northern Ireland Registry of Self-Harm. In particular we
Bill is being developed to target issues of alcohol misuse.24 would like to acknowledge the work of the data registration
Both identify the reduction of alcohol consumption at a general officers who collected the data as well as the data managers.
level, as well as targeting advertisement, labelling and pricing of
alcohol as key priorities and activities. We have previously high-
Funding
lighted increased self-harm presentations around public holi-
days11 and future work is warranted to explore the temporal This work was supported by the Irish Health Service
patterns of alcohol use in self-harm presentations reported Executive’s National Office for Suicide Prevention, funders
here. of the National Self-Harm Registry Ireland, and the Public
We have established the important role that alcohol plays Health Agency, funders of the Northern Ireland Registry of
in self-harm presentations made to hospital EDs. In Self-Harm.
 THE INVOLVEMENT OF ALCOHOL IN HOSPITAL-TREATED SELF-HARM AND ASSOCIATED FACTORS e163

References of Self-Harm in Primary and Secondary Care. London: National Institute

for Health and Care Excellence, 2004.
1 Cherpitel CJ, Borges GL, Wilcox HC. Acute alcohol use and sui- 14 Cooper J, Steeg S, Bennewith O et al. Are hospital services for self-
cidal behavior: a review of the literature. Alcohol Clin Exp Res 2004; harm getting better? An observational study examining manage-
28:18s–28s. ment, service provision and temporal trends in England. BMJ Open
2 Larkin C, Di Blasi Z, Arensman E. Risk factors for repetition of 2013;3:e003444.
self-harm: a systematic review of prospective hospital-based studies. 15 Royal College of Emergency Medicine. Mental Health in the ED
PloS One 2014;9:e84282. Clinical Audit 2014–15: National Report. London: The Royal College
3 Bergen H, Hawton K, Webb R et al. Alcohol-related mortality follow- of Emergency Medicine, 2015.
ing self-harm: a multicentre cohort study. JRSM Open 2014;5:1–11.

Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/e157/3819202 by guest on 28 September 2019

16 Ward M, McGee H, Morgan K et al. SLÁN 2007: Survey of Lifestyle,
4 Sher L, Zalsman G. Alcohol and adolescent suicide. Int J Adolesc Attitudes and Nutrition in Ireland. In: ‘One Island—One Lifestyle?’
Med Health 2005;7:197–203. Health and lifestyles in the Republic of Ireland and Northern Ireland:
5 Suominen K, Isometsä E, Henriksson M et al. Hopelessness, impul- Comparing the population surveys SLÁN 2007 and NIHSWS 2005,
siveness and intent among suicide attempters with major depression, Department of Health and Children. Dublin: The Stationery Office, 2009.
alcohol dependence, or both. Acta Psychiatr Scand 1997;96:142–49. 17 Griffin E, Corcoran P, Cassidy L et al. Characteristics of hospital-
6 Haw C, Hawton K, Casey D et al. Alcohol dependence, excessive treated intentional drug overdose in Ireland and Northern Ireland.
drinking and deliberate self-harm. Soc Psychiatry Psychiatr Epidemiol BMJ Open 2014;4:e00555.
2005;40:964–71. 18 Ferry F, Bunting B, Murphy S et al. Traumatic events and their rela-
7 Ness J, Hawton K, Bergen H et al. Alcohol use and misuse, self- tive PTSD burden in Northern Ireland: a consideration of the impact
harm and subsequent mortality: an epidemiological and longitudinal of the ‘Troubles’. Soc Psychiatry Psychiatr Epidemiol 2014;49:435–46.
study from the multicentre study of self-harm in England. Emerg 19 Bunting BP, Murphy SD, O’Neill SM et al. Lifetime prevalence of
Med J 2015;10:793–99. mental health disorders and delay in treatment following initial
8 Perry IJ, Corcoran P, Fitzgerald AP et al. The incidence and repeti- onset: evidence from the Northern Ireland Study of Health and
tion of hospital-treated deliberate self harm: findings from the Stress. Psychol Med 2012;42:1727–39.
world’s first national registry. PLoS One 2012;7:e31663. 20 World Health Organisation. Preventing Suicide: A Global Imperative.
9 Corcoran P, Griffin E, O’Carroll A et al. Hospital-treated deliberate Geneva: World Health Organisation, 2014.
self-harm in the western area of Northern Ireland. Crisis 2015;36:83–90. 21 Department of Health, Social Services and Public Safety. Protect Life—
10 Platt S, Bille-Brahe U, Kerkhof A et al. Parasuicide in Europe: the A Shared Vision: The Northern Ireland Suicide Prevention Strategy and Action
WHO/EURO multicentre study on parasuicide. Acta Psychiatr Scand Plan, 2006–11. Belfast, UK: Department of Health, Social Services
1992;85:97–104. and Public Safety, 2006.
11 Griffin E, Arensman E, Corcoran P et al National Self-Harm Registry 22 Health Service Executive. National Clinical Care Programme for the
Ireland Annual Report 2014. Cork: National Suicide Research Assessment and Management of Patients Presenting to Emergency Departments
Foundation, 2015. Following Self-Harm. Dublin: Health Service Executive, 2014.
12 Hawton K, Haw C, Casey D et al. Self-harm in Oxford, England: 23 Department of Health, Social Services and Public Safety. New Strategic
epidemiological and clinical trends, 1996–10. Soc Psychiatry Psychiatr Direction for Alcohol and Drugs. Phase 2 2011–16. Belfast, UK:
Epidemiol 2015;50:695–704. Department of Health, Social Services and Public Safety, 2011.
13 National Institute for Health and Care Excellence. Self-Harm: The 24 Department of Health. General Scheme of the Public Health (Alcohol) Bill
Short-term Physical and Psychological Management and Secondary Prevention 2015. Dublin: Department of Health, 2015.