Journal of Psychosomatic Research 77 (2014) 449–456

Contents lists available at ScienceDirect

Journal of Psychosomatic Research

Review

Perceptions of paramedic and emergency care workers of those who selfharm: A systematic review of the quantitative literature

Nigel Rees ⁎, Frances Rapport, Gareth Thomas, Ann John, Helen SnooksSwansea University, Pre-hospital Emergency Research Unit, Welsh Ambulance Services NHS Trust, United Kingdom

⁎ Corresponding author at: Welsh Ambulance Service NE-mail address: nigel.rees@wales.nhs.uk (N. Rees).

http://dx.doi.org/10.1016/j.jpsychores.2014.09.0060022-3999/Crown Copyright © 2014 Published by Elsevie

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 7 July 2014Received in revised form 4 September 2014Accepted 7 September 2014

Keywords:AmbulanceCareEmergency self harm paramedicPre hospital

Objective: The U.K. has one of the highest rates of self harm in Europe at 400 per 100,000 of population. Paramedicsand emergency staff may be the first professionals encountered, therefore understanding their views andapproaches to care is crucial. The aim of this study was to systematically review published quantitative literaturerelating to paramedic and emergency workers' perceptions and experiences of caring for people who self harm.Methods: CINAHL®, MEDLINE®, OVID ® and Psych INFO® databases were searched, PRISMA guidelines werefollowed, two researchers independently screened titles, abstracts and full papers against a priori eligibility criteria.Data synthesis was achieved by extracting and descriptively analysing study characteristics and findings.Results: 16 studies met inclusion criteria; one included ambulance staff, all used questionnaires. Training, policiesand guidelines improved staff knowledge and confidence in caring for peoplewho self harm. Limited access to train-

ingwas reported, ranging from75% to 90% of staff lacking any. Limited departmental guidelineswere also reported.Staff in acute settings exhibited increased feelings of negativity, becoming less positive closer to front line care. Re-cent studies report positive attitudes amongst emergency staff.Discussion:Despite guidelines indicating need for education and policies to guide staff in self harm care, there is lim-ited evidence of this happening in practice. The lack of literature including paramedics suggests a gap in our under-standing about care for self harm patients. This gapwarrants greater attention in order to improve care for patientswho self harm in their first point of contact.

Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Eligibility criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Data extraction and synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Outcome variables, instruments and data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Self harm training and education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451Experience of caring for people who self harm, perceptions and attitudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452Age of staff caring for those who self harm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452Attitudes towards self harm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452Setting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453Policies for dealing with people who self harm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453Gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454

HS Trust, Pre-hospital Emergency Research Unit, Lansdowne Hospital, Sanatorium Road, Cardiff CF11 8UL, United Kingdom.

r Inc. All rights reserved.

450 N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455

Introduction

Self harm is defined as “intentional self-poisoning or self- injury,irrespective of the nature of motivation or degree of suicidal intent”[1]. The U.K. has one of the highest rates of self harm in Europe at 400per 100,000 of population [2], which is increasing by 10% every threeyears [3], and is one of the five top causes of acute hospital admissions[1]. These figures may not represent the true scale of the problem, asself harm often involves attempting to inflict harm without the needformedical intervention, and as such it is often not recorded [4]. Indeed,only 10–20% of people who self harm present at hospital [5,6], meaningthat the true scale of self harm could be closer to 3 in 100 people [4].

Despite the scale of this problem in the population, many allege thatthe quality of care and attitudes fromhealth care staff are unsatisfactory[4,7–9]. Medical personnel are recognised as providing the most unsat-isfactory support [10], andpatients report negative and hostile reactionsfrom ambulance staff [11]. In a U.K. survey of those who self harm, 43%of people said they avoided emergency services because of previousnegative experiences [12]. They recalled being told they were wastingtime and resources, which compounded distress, leading to prematureself discharge and affecting their future help seeking behaviour.Crawford [13] found that those who discharged themselves from carebefore receiving a psycho social assessment were three times morelikely to repeat self harm.

Outcome variables, instrume

Records identified through database searching

n=975

RRsy

Records after duplicates

Records screened by title and abstract n=864

Full-text articles assessed for eligibility n=86

Studies included in review n=16

Total number of citations

Incl

uded

Elig

ibili

tySc

reen

ing

Iden

tifi

cati

on

Fig. 1. Inclusion and e

Many self harm patients' report that they find themselves ignoredby health professionals, not just because of negative attitudes to-wards them, but also because they are perceived as difficult to dealwith [14]. In hospital, staff reportedly distance themselves fromsuch patients, rationalising them as manipulative, attention-seeking and untreatable [15]. Yet such perceptions are grounded ineveryday inaccuracies and myths [16]. The effect on the patientwas noted by Nehls [17] who reported three themes derived from in-terviews with people who self harm. These included: living with alabel, living with self-destructive behaviour seen as being manipula-tive, and living with limited access to care. Above all, people who en-gage in self harmwant to be seen by empathetic health professionalswho are able to listen to them, be supportive and non-judgemental[1,18].

Despite paramedics and emergency staff often being the first pointof contact for thosewho self harm, few research studies have investigat-ed their care. The study from which this paper derives forms part of alarger project entitled paramedics' perceptions of caring for peoplewho self harm, which is using Evolved Grounded Theory Methodologyto explore and understand their care in order to inform policy andhelp tailor education. This paper concentrates on the systematicreview of the quantitative literature with the aim of investigating thecurrent level of knowledge about paramedics' and emergency workers'perceptions of caring for those who self harm.

nts and data analysis

ecords identified through other sources n=2ecords identified through previous stematic reviews

removed n=864

Records excluded n=778

Full-text articles excluded n=70

Reasons for exclusion:No relevance to emergency care n= 597No focus on perceptions of care n= 53Not relevant participants n=115Qualitative n=13

identified n=977

Qualitative methodologies: n=15Specialist mental health workers included n=5No relevance to pre hospital or emergency care n= 50

xclusion criteria.

Eligibility criteria:

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [20] were adhered to (Fig. 2).Papers were reviewed independently by two researchers screening their title, abstract and the full papers against the inclusion criteria (fig1). Feature Inclusion criteria Exclusion criteriaParticipants Paramedics, Ambulance Staff, Nurses and Doctors working in

the pre hospital or emergency settings.Other health and social care professionals not practicing in emergency or pre hospital care.Student clinicians, specialist mental health workers.Patients, service users

Focus of study

Research that explores, and generates new knowledge regarding, participants perceptions of care for those who self harm.Studies addressing attitudes, experiences, education or knowledge.Types of self harm. includes all those along the continuum from ideas, to self harm. and finally suicide completions.

Specific treatment and prevention studies will be excluded.Studies exploring attitudes to assisted suicide, and towards self-harm in certain specific populations (including learning disability, prisoners, and minority groups)

Type of Study

Published primary research using a recognised quantitative research: randomised, non randomised, observational studies

Narrative reviews, qualitative studies, systematic reviews, editorials, opinion pieces, non Englishlanguage papers, studies reported as posters or abstract onlyUnpublished studies Opinion papers

Date of publication

No date restrictions applied Not applicable

Language English Languages other than English

Fig. 2. PRISMA flow chart.

451N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

Methods

Search strategy

Electronic literature searches were conducted (June 2013) ofCINAHL®, MEDLINE®, OVID® and Psych INFO®, using a broad searchstrategy with the key words Self harm, Paramedic and Perceptions.MeSH terms and truncation were applied as follows: [self harm* ORsuicide*OR overdose* OR mental health*] AND [paramedic* OR nurse*OR doctor*] AND [emergency* OR pre hospital* OR ambulance*] AND[perceptions of care*]. No language or date restrictions were applied.References of included papers and grey literature such as guidelines,policy documents and conference proceedingswere also hand searched.

Eligibility criteria

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [19] were adhered to (Fig. 2). Paperswere reviewed independently by two researchers screening their title,abstract and the full papers against the inclusion criteria (Fig. 1).

Quality assessment

Quality was assessed using an adapted version of the Social Care In-stitute for Excellence Quality Assessment Tool [20] which was recentlyused in systematic reviews of care for people who self harm [21,22].Two review authors (NR and GT) independently assessed the qualityof papers, and disagreements were discussed at consensus meetings,until the two agreed on the final quality of all papers.

Data extraction and synthesis

Study characteristics and outcomes were re-coded using a pre-designed data abstraction form (Fig. 3). All studies meeting the inclu-sion criteriawere included in the review regardless of quality; however,in interpretation, more weight was given to studies of more robustdesign and relevance. The number of reports was so limited, and

differences between study design and measures so great, that combin-ing results with statistical testing for bias and heterogeneity was inap-propriate and meta-analysis was not undertaken. Rather, datasynthesis comprised data extraction and descriptive analysis.

Results

Eight hundred and sixty-four studies were screened for relevance. A random samplewas independently reviewed by NR and GT, with reasonable agreement at abstract stage(Kappa 0.649 sig. = 0.00) and full paper (Kappa 0 .783 sig. = .011); disagreementswere resolved by discussion until the final sample was agreed. Data were abstractedfrom 16 studies. One study included ambulance staff and this did not include paramedics.Participants included emergency doctors, nurses and ambulance staff. All studies reviewedused questionnaires, and scored an average of 9 out of 12 on quality assessment; this wascomparable to previous studies [22]. Studies were conducted in the UK (n= 5), Australia(n = 4), Ireland (n = 4), Taiwan (n = 1), and Finland (n = 2).

Outcome variables, instruments and data analysis

Participant demographic, descriptors and outcomemeasures varied considerably andincluded attitudes to self harm, age, gender, experience, and previous self harm training.There was also variation in measurement instruments; most studies used structuredquestionnaires, often self-developeddue to limited reliable andvalidmeasures specificallytargeting self-harm. The validity and reliability of these instruments was not alwaysdescribed. Six studies used validated measurement instruments, such as the Self-HarmAntipathy Scale [23], Attitudes towards Deliberate Self-Harm Questionnaire [23,24],Suicide Opinion Questionnaire [25,26] or Understanding Suicidal Patients Scale [27]. Asthis review included only quantitative studies, descriptive or inferential statistics wereused. The following themes were identified:

Self harm training and education

All studies reported a lack of training for health professionals caring for people whoself harm; previous training was reported by 18% to 22% of staff [24,28], 75% to 90% hadnot received any training [28–30]. Staff believed they lacked the necessary skills to carefor people who self harm, and perceived the need for further training in communication,and understanding of self harm, which they believed would give them more confidence[23,25,31]. When such programmes were attended, significant improvements werereported in knowledge, confidence and attitudes towards self harm [13,24,26,31,32].Three studies reported on the impact of education on professional development. Attitudesto self harm were more positive in nurses prepared through diplomas than hospitaltrained nurses [24,25], and medical staff indicated less of a perceived need for furthertraining about the impact of self harmonpatients than did nurses (p b 0.0001) [33]. Fried-man et al.. [31] administered a questionnaire to 117 Emergency Department (ED) staff,and whilst they felt self-laceration was an important problem, they felt unskilled in

Data abstraction formStudy Quality Number

of

subjects

Clinician Setting Methods Outcomes

Friedman et al (2006)

8 117 ED staff

88 nurses 29 doctors.

E/D UK Questionnaire Staff believed that self-laceration was an important problem but felt unskilled in managing. They were unsure of the

relationship between self-laceration and both mental illness and risk of suicide. They had previously received little training in self harm. In those without previous training, a longer period working in A&E was correlated with higher levels of anger

towards patients

Staff wished for further training and wanted a higher proportion of patients to be seen by specialist mental health services.A significant correlation between time worked in A&E and the degree to which they felt angry with the patients (P<.017).

Trend for staff with more A&E experience in to feel more inadequate (P <.11) and view patients as less mentally ill (P<.11)

Those who had not received training previously, and more years of experience felt more angry (p<.008),Untrained staff with more experience were more likely to feel that patients were not mentally ill (P< 024).

Mc Cann et al (2006)

6 43 ED Nurses E/D

Australia

Questionnaire Most nurses had received no educational preparation to care for patients who self harm.

Over 20% claimed the department either had no practice guidelines for self harm or they did not know of their existence; one-third who knew about them had not read them.

Older and more experienced nurses had more supportive attitudes than younger and less experienced nurses. (p,<0.04)Nurses who had attended in-service education on self harm had more positive attitudes than non-attendees (p<0.03)

Conlon & O’Tuathail (2012)

10 87 ED Nurses ED Ireland ‘Self -Harm

Antipathy Scale’ validated

questionnaire

Nurses show slightly negative antipathy overall, indicating positive attitudes towards self harm patients.

Attitudes were significantly different in accordance with a nurse’s age. Education and social judgment also contribute to the way nurses view, interact and make moral decisions regarding self

harm

The only statistically significant difference in overall antipathy was in participant’s age. The mean responses for the age group 21–40 were 98.45 (SD 21.23) and for the age group 41–60 was 89.91 (SD 16.70)

(p < 0.05).

Crawford et al (1998)

8 45 Nurses and Medical Staff

ED London UK

Non-randomised intervention study

comparing

psychosocial assessment of self

harm patients

before and after a one hour teaching

session

Following a one hour teaching session:

Sixteen (13%) of 125 sets of records before and 58 (46%) of 127 sets of records after the intervention were

judged to be adequate.

In the post intervention period, notes were more likely to be judged adequate when a pro forma was used as part of the assessment (52 of 66 with a pro forma and six of 61 without a pro forma, x2 = 60, p < 0.01).

To the question: patients you see who are always taking overdoses are less likely to kill themselves than those who have only tried once 29% agreed pre intervention and 9% post (p<0.01)

To the question: I feel I have the necessary skills to play my part in the assessment and treatment of deliberate self harm patients 52% agreed pre intervention and 75% post (p<0.01)

Egan et al (2012)

10 125 Nurses &

Doctors

ED Ireland Questionnaire Knowledge and confidence were significant contributors to perceived personal effectiveness in dealing with self harm

patients.

A positive relationship between knowledge and effectiveness (p<0.05). As knowledge increased, effectiveness increased. A positive relationship between confidence and effectiveness (p < 0.01). As confidence increased, effectiveness increased.

Effectiveness was significantly related to negativity ( p < 0.01). As effectiveness increased, negativity decreased.

A negative relationship between knowledge and negativity (p < 0.01). As knowledge increased, negativity decreased.

Sun et al (2011)

11 155 Casualty nurses ED Taiwan Questionnaire Nurses were found to hold positive attitudes to patients who self harm.

Nurses with higher levels of education achieved higher scores on positive attitude (p<0.048)

Nurses who had suicide care experience of nursing 1-10 patients held more positive attitudes than those who had cared for 21-30 patients (p<0.014)

Nurses who had no religion had more positive attitudes (p<0.025)

McAllister et al (2008)

9 27 Emergency nurses

Queensland Australia

Pre-test/post-test design

No significant improvement in the global measure of nurses’ professional self-concept Significant improvements (p < .05) from pre- to post-test for two subscales of the Perception of Nursing Scale

McAllister et al (2002a)

9 1008 Emergency nurses

23 public and 14

private ED

Queensland Australia

The Attitudes Towards

Deliberate Self-

Harm Questionnaire

(ADSHQ)

Four factors reflected nurses’ attitudes toward self harm:Perceived confidence in their assessment and referral skills

Ability to deal effectively with clients

Empathic approach

Ability to cope effectively with legal and hospital regulations that guide practice. 90% had no formal training in managing clients who self harm

Fig. 3. Data abstraction form.

452 N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

dealing with these patients. Knowledge and confidence were significant contributors toperceived personal effectiveness in dealing with people who self harm [29]. A positive re-lationship was found by Egan [29] between knowledge and effectiveness (p b 0.05); andbetween confidence and effectiveness p b 0.01 and the opposite relationship between ef-fectiveness and negativity (p b 0.01).

Experience of caring for people who self harm, perceptions and attitudes

McAllister [9] found that 96% of respondents in a study of seventy-one nursesworkingin two ED's in Australia reported professional experience of caring for people who selfharm, whilst 36% reported personal experience of self harm. Friedman et al. [31] howeverfound that incidence of self harmwas overestimated by UK ED staff; the mean estimate ofself harm presentations to one ED in a six month period was 117 cases, the actual numberwas 22 cases.

The effects of caring for people who self harm on professional attitudes was mixed,with several studies showing a positive relationship between level of experience of caringfor peoplewho self harmand unfavourable attitudes [25,31,34]. However,McAllister et al..(2002) [9], having administered the ‘Attitudes towards Deliberate Self-Harm Question-naire’ (ADSHQ) to nurses working in ED's in Australia, found significant correlations(p b 0.05) between years of professional caring experience and ADSHQ total score, andyears of experience.

Age of staff caring for those who self harm

Older caregivers had more empathy and understanding for patients who self harm.McCann et al.. [26] found that older andmore experienced nurses hadmore supportive at-titudes than younger and less experienced nurses (p b 0.04), older staff also displayedgreater empathy than younger staff and held more positive attitudes towards peoplewho self harm (p b 0.05) [23]. McLaughlin [30] administered an adapted Suicide OpinionQuestionnaire (SOQ) [35] to 200 nursesworking in ED's, and found thatmore older nursesthan younger nurses considered: those who tried to kill themselves did not really want todie (p b 0.05), those who threaten to commit suicide rarely do so (p b 0.05), and thosewho attempt to commit suicide often fail (p b 0.05), but disagreed that those who

attempted suicide using highly public places are more interested in getting attention(p b 0.05). In McCann et al.'s study [26] younger nurses were undecided in their responseto the statement that individualswho attempted suicidewere trying to get sympathy fromothers, whereas older nurses were more likely to disagree (p b 0.01).

The impact of self harm encounters on risk of future self harm or completed suicidewas raised in several studies [25,26,30,36]. 30% of emergency room staff in Suokas &Lonnqvist [36] disagreed that attitudes and behaviour of staff influenced recurrence of sui-cide attempts. McCann et al.. [26] reported that experienced nurses were more likely todisagree that individuals who threatened suicide rarely took their own lives (p b 0.01),whilst in McLaughlin [30] 71% of less experienced nurses and 92% of more experiencednurses agreed that most people who tried to kill themselves did not really want to die(p b 0 001), 53% of younger nurses and 65% of older nurses tended to agree that mostpeople who attempted suicide failed in their attempt (p b 0 05).

Attitudes towards self harm

Several studies reported positive attitudes towards people who self harm [23,24,26,27,30]. Sun et al.. [25] revealed how nurses displayed empathy towards those who selfharm, perceiving suicide attempts as people trying to communicate their pain, and keento seek attention. In studies reporting negative attitudes, the influencing factors presenteda complex picture. McAllister et al.. [9] found an ADSHQ score of 65.16 (SD = 4.38,range = 46–87), indicating generally negative attitude towards patients who self harmin ED nurses. However, ADSHQ identified four dimensions explaining variations in atti-tudes, which included perceived confidence in assessment and referral, ability to dealwith clients, legal and hospital regulations guiding practice, and an empathic approach.Amongst staff who had not been trained in self harm a longer period of ED working wascorrelated with higher levels of anger towards patients who self harm. Nurses reportedin Conlon & O'Tuathail [23] that patients who repeatedly presented following self harmevoked feelings of frustration and powerlessness and were often ignored ormarginalised.Sun et al.. [25] found that EDnurses found suicidal behaviour irritating, and inMcAllister etal.. [9] respondents weremore likely to agree that they felt helpless in dealing with peoplewho self harm.

Staff in large hospitals scored less empathy towards self harm clients (p<0.05)

Significant correlations were found between years of ED experience and ADSHQ total score (p<0.05), and years of ED

experience and an empathic approach towards self harm (p<0.01)Generally negative attitude towards clients who self harm.

Correlations between ED experience and ADSHQ score, years of ED experience and an empathic approach towards self

harm

Mc Alister (2002)

10 1008 Nurses ED

Queensland

, Australia

Questionnaires 90.1% had not had any formal training in self harm

96.3% indicated professional experience with people who self harm

36.1% indicated personal experience with people who self harm.46.3% indicated that neither a formal nor informal procedure existed in their place of work for managing self harm clients.

19% reported formal procedure existed, and 18.6% indicated that only an informal procedure was in place.

16% reported formal and informal procedures to manage self harm clients existed in their place of work82% reported no guidelines in place for assessing whether the client had determined alternative behaviours to self harm.

McCarthy & Gijbels (2010)

11 68 Nurses ED Ireland Descriptive and

correlational design Attitude

Towards

Deliberate Self-Harm

Questionnaire

(ADSHQ)

22%; had received education on self harm

Attitudes were more positive of nurses with either a diploma or postgraduate diploma than hospital trained nurses.Years of ED experience did not show statistically significant results

No correlation between confidence in assessment and gender, age, academic qualifications, years of ED experience, or a

history of self harm education.ED nurses held positive attitudes towards individuals presenting with self harm.

A statistically significant difference was found for individuals who had undertaken self harm education compared to those

who had not.

Turnbull & Chalder (1998)

8 51 37 nurses, 14

Doctors

ED London Questionnaire

before and after a

teaching package.

After implementation of a teaching package general knowledge improved. (P<0.03).

Attitudes did not become more favourable.

Ghodse (1978)

8 1248 Ambulance

Staff (15%), Nurses (17%),

Medical staff

(12%) other (2%)

62 Casualty

Units London

Questionnaire People who take an overdose accidentally are viewed more favourable than deliberate attempts (p<0.001)

Ambulance staff had a statistically significant more favourable attitude to accidental overdoses than any other professional group (p<0.001)

Ambulance staff expressed a very unfavourable attitude towards drug dependant patients (p<0.001)

Those with unfavourable attitudes towards suicidal patients had seen significantly more of these patients (p<0.01), and those who had very favourable attitudes had seen significantly less (p<0.01)

Significantly more males amongst those who expressed very unfavourable attitudes towards drug dependant and suicide

overdose patients (p<0.01)Significantly fewer nurses than ambulance staff held very unfavourable attitudes to drug dependant patients (<0.001)

Suokas, Suominen, Lonnqvist ( 2009)

6 100 Nurses &

Doctors

Helsinki Questionnaire. After 1 year of the consultation service, the understanding and willingness to care for attempted suicide patients, as

measured by the USP Scale, had not significantly changed: The mean score before the psychiatric consultation was 26.4 ± 6.6, and after 27.8 ± 8.0 (statistically: ns). Overall, there were no statistically significant differences in individual items of

the USP Scale.

Gender, age, profession, and length of work experience did not have a statistically significant effect on staff attitudes.

Suokas & Lonnqvist (1989)

6 184 Nurses &

Doctors

ED

comprising

of Emergency

room (ER),

Emergency ward (EW)

and ITU

Helsinki

Questionnaire. 72% of staff in ITU, 58% in EW and 25% in ER claimed they were as co-operative and sympathetic towards patients who

attempt suicide as other patients ((p<0.001)

62% of ITU, 46% EW and 30% ER staff were sympathetic and understanding towards patients who attempt suicide (p<0.01)

30% of ER staff disagree that attitudes and behaviour of staff has an influence on the recurrence of suicide attempts

((p<0.01)33% of ER staff disagree that a patient who makes recurrent attempts at suicide is at a great risk of succeeding

60% of EW, 54% ER, 16% of ITU staff admitted to letting their irritation show when treating a patient who had attempted

suicide (p<0..001)76% of ER, 40% EW and 23% of ITU staff agree that a patient who attempts suicide wastes staff time (p<0.001)

28% of ER staff agreed completely that patients who attempt suicide are misusing treatment facilities

51% of ER, 19% EW, 7% ITU (p<0.001) staff completely agreed that a patient who has attempted suicide takes up staff

time which is not available to those who are in greater need of help.Occupational groups did not differ in attitudes other than need for training: The majority of Physicians regarded themselves

as having sufficient training to treat patients who had attempted suicide, whereas nurses considered theirs insufficient.

Mackay & Barrowclough (2005)

6 180 Nurses & Doctors

4 A&E

depts.

Manchester

UK

Hypothetical vignettes followed

by questionnaire

Male staff expressed less sympathy (p<0.02), greater irritation (p<0.04) and frustration (p<0.01), less willingness to help (p<0.005) and personal optimism (p<0.03) for self harm than their female colleagues

There was no association found between experience, yet older age was associated with greater understanding (p<0.03)

Medical staff indicated less of a perceived need for further training on self harm than nurses (p<0.0001)

Mc Laughlin (1994)

12 142 Nurses Ireland ED Ireland Questionnaire Older nurses than younger nurses agreed that those who threaten to commit suicide rarely do so (P<0 05), that most people who tried to kill themselves did not really want to die (P<0 05) agreed that most people who attempt suicide fail m their

attempt (P<0 05), disagreed that those who attempt suicide using highly public places are more interested in getting

attention (P<0 05) More experienced nurses than less experienced nurses agreed that people who attempt suicide are trying to get sympathy

from others (P<0 01)

Fig. 3 (continued).

453N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

Evidencewas found to indicate that some staff perceive self harmas deserving a lowerpriority of care than those they perceive as purely ‘medical’ [30,34,36]. McLaughlin [30]reported in response to a hypothetical scenario, that all respondents felt a patientwith se-vere chest pain required urgent nursing intervention, whilst 18% of younger nurses and11% of older nurses felt that the overdosepatient did not. Suokas & Lonnqvist [36] reportedthat 76% of Emergency Room (ER) staff agreed that patients who attempted suicidewasted staff time, 28% of staff agreed these patients misuse treatment facilities, and 51%of staff agreed that these patients take up staff time that is sorely needed by others ingreater need of help. In Ghodse [34], ambulance staff had significantlymore favourable at-titudes to accidental overdoses than any other professional groupwhereas they expressedan unfavourable attitude towards drug-dependent patients.

Religious beliefs may also be a factor in attitudes to self harm care. In Sun et al.. [25]almost 50% revealed that they held religious beliefs; nurses with no religion had morepositive attitudes towards suicidal behaviour than those who followed a religion.

Setting

Staff in acute settings and large units reported higher levels of negativity towards peo-ple who self harm than those in non-acute settings [36]. Suokas & Lonnqvist [36] foundthat 72% of staff in ITU, 58% in the Emergency Ward (EW) and 25% in the EmergencyRoom (ER) claimed that they were as co-operative and sympathetic towards patientswho attempt suicide as towards other patients (p b 0.001). Sixty percent of EW, 54% ER,16% of ITU staff admitted to letting their irritation show when treating people who selfharm (p b 0.001). McAllister et al. [9] also found that staff in large hospitals scored

significantly lower on perceived ability to assess and refer self harm clients (p b 0.05)and were less empathetic towards self harm (p b 0.05).

Policies for dealing with people who self harm

Policies or guidelines for treating self harmwere reported to be lacking or underused.In McCann et al.. (2006) [26] 20% of nurses claimed the ED either had no practice guide-lines for self harm or if they did then nurses were unaware of their existence, though alarge minority (one-third) knew about them but had not read them. 46% of respondentsin McAllister (2002) [37] indicated neither formal nor informal procedures existed formanaging those who self harm, 82% reported having no guidelines in place for assessingself harm. McCarthy & Gijbels [24] also reported that 68% of ED staff agreed that hospitalpolicies impeded their ability to work effectively with self harm patients.

Gender

The proportion of female respondents ranged from 84% to 95% across studies. Twostudies found associations between gender and attitude towards self harm patients:Ghodse [34] found that significantly more males expressed unfavourable attitudes to-wardsdrugdependence and suicide overdose patients (p b 0.01).Mackay&Barrowclough[33] reported that male staff expressed less sympathy (p b 0.02), greater irritation(p b 0.04), increased frustration (p b 0.01), and a lack of willingness to help (p b 0.005)self harm patients than their female colleagues. However the picture was again mixed astwo studies failed to demonstrate any significant gender related findings [24,27].

454 N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

Discussion

All studies reviewed used questionnaires, and scored an average of 9out of 12 on quality assessment, and whilst this is comparable to previ-ous studies [22] such self reported data does not give us an indication ofwhat is actually happening in practice.

This review identified one study of the care provided by ambulancestaff [34] for patients who self harm. It was undertaken before the intro-duction of the professionalisation of paramedics in the UK as they wereaccepted to register with the Health Professions Council in 2001. Asparamedics are often the first professionals encountered by patientswho self harm understanding their views and approaches to caring forpeople who self harm is crucial as it may determine the quality of carethese patients receive. Staff in acute settings were found to exhibithigher levels of negativity towards self harm patients, and attitudesbecame less positive the closer to the frontline they worked [36]. Thiswas linked to situational factors such as competing priorities in a busyED, and patients being more difficult to deal with earlier in their care.Such findings may not directly transfer to the paramedic population,because whilst they may be at the front line and encountered early inthe care of patients who self harm, with a comparable workload toemergency staff, they usually have few competing priorities, caring forone patient at a time. A recent report by the All Party ParliamentaryGroup onMental Health [38] gives some insight into potential for differ-ing perceptions between emergency staff and paramedics. They foundthat some people who self harm felt theywere not treatedwith respect,listened to and stigmatised in the ED, whilst the same report includedambulance staff in their accounts of professionals who had supportedthem in a crisis being compassionate, supportive and understanding.

Guidelines clearly state that training is needed for paramedics andother emergency care professionals when caring for people who selfharm [1,7]. This systematic review has revealed scant evidence of theavailability or uptake of such training. The finding of several papers inthis review shows that attendance at such training improves staffknowledge and confidence and lowers negative attitudes towardsthose who self harm. The positive impact of education on attitudes[24,25,33] may merit further expansion and this should be clearly writ-ten into policy documents. Guidelines [7] also recommend that carepathways, protocols and local guidelines, are reviewed and revised.Despite this, studies report a lack of adoption or knowledge of suchguidelines [24,26,37] representing a limited influence of such guide-lines. The importance of this finding links with earlier assertions that,the awareness and availability of guidelines and policies will supportthe positive relationships between knowledge, confidence, perceivedpersonal effectiveness and decreasing negativity. Addressing the educa-tion needs and availability of guidance for staff has the potential topositively impact on care for people who self harm when they encoun-ter paramedic and emergency care staff, and is therefore a key recom-mendation of this study.

Staff with greater experience of care held more negative attitudesand more anger towards patients who self harm [25,34], however re-sultswere inconsistent, as studies reported a significant relationship be-tween the duration of clinical experience and attitudes to self harm [9],and significant correlations between years of ED experience and an em-pathetic approach towards self harm patients [30]. It is thereforeunclearwhether experience influences staff attitudes. Recent studies re-ported more positive attitudes from emergency staff which contrastswith a recent systematic review [22], and studies investigating thepatients' perspective [10,12,21,39]. A recent scoping review of attitudesof emergency care staff towards young people who self harm [39]recognised benign attitudes towards self harm in recently publishedstudies which reflects the mixed picture of our findings.

The findings of more positive attitudes towards self harm fromemergency staff is unclear, it may demonstrate the influence of educa-tion, policy, guidelines or media reporting, but this has yet to be exam-ined. Studies finding negative attitudes presented a more complex

picture than simply positive or negative [9,29], factors such as perceivedconfidence in assessment and referral, ability to deal effectively withclients, empathetic approach, and ability to cope effectively with legaland hospital regulations that guide practice, may be influencing thiscomplex picture. Staff who scored higher on these factors were morelikely to feel positive towards people who self harm and the care theyprovide. Self harm evokes feelings of irritation and anger [23,25,28,31,36], borne out of a sense of powerlessness, and frustrationwith patientswho self harm frequently returning to hospital. These perceptionsappear to manifest by self harm patients being given a lower prioritythan those with physical illness, and reduced entitlement to care,Cleaver [39] argues that staff downgrade the psychologicaldimensions of care when under pressure, prioritising physical care,which is less time consuming and resource intensive, particularly inthe ED.

The established link between self harm and completed suicides wasinvestigated by several studies in the review [13,25,26,30,36]. Emergen-cy staff acknowledged the risk of completed suicides in people who selfharm, which is called for in policy [7,18], however more experiencedstaff showed more awareness of this risk. One study found a quarterof suicides to be preceded by acts of self harm within the previousyear [40], therefore this awareness by experienced staff may be borneout of previous encounterswith such patients who self harm and subse-quently kill themselves. Such findings may be used to inform educationtargeting inexperienced staff highlighting the risk of suicide in peoplewho self harm.

Belief in a religion has been identified as a protective factor againstself harm and suicide [41–44] and this study found acknowledgementof this amongst staff. However, Sun et al.. [25] found a relationshipbetween religion and negative attitudes, which was also found byNorheim et al.. [45], where Christians agreed significantly less thatsuicide is acceptable than professionals with no religious background.Such a link may be due to historical religious attitudes portraying selfharm and suicide as sinful, however 42% of participants in Sun et al..[25] were Buddhists, where life comprises transmigration, and thesoul of those who complete suicide cannot be reincarnated [46]. Whistthis may be influencing attitudes to self harm it is not consistentlyobserved in studies [47]. Nevertheless, the role of religion in society,and its influence on professionals are complex and difficult to gainmeaningful insights into.

Females were over represented in comparison to males in allstudies; this is an important factor as two studies found females heldsignificantly more positive attitudes towards self harm than males [33,34]. These studies were dated, and there was a strong gender–role asso-ciation (male doctors and ambulance staff and female nurses). This maybe seen in Ghodse [34] where females had more favourable attitudes toself harm, at the time nursing was predominantly a female professionwhereas the opposite was the case for ambulance staff. More recentstudies fail to find significant differences in gender [24,27], once againsuch a mixed picture reveals the complexities of influences on percep-tions of care for people who self harm.

Limitations

The authors recognise that as perceptions of self harm care involvesa phenomenon occurring in the social world, understanding abstractedfrom quantitative studies will be an incomplete exploration of the liter-ature due to such complexities. The authors are clear to distinguish thepositivist position of quantitative literature from the interpretivist–con-structivist position of qualitative literature, which is explored through ametasynthesis of the qualitative literature along with an ongoingEvolved Grounded Theory. Studies in the present study used surveys,yet such self reported datamay not reflect actual perceptions, moreoverin evaluating the impact of education may teach the desired answers.The ongoing research considers the patients' perspective along withdeeper insights by interviews. The review may also be limited by its

455N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

search strategy and inclusion of only English language articles. It wasconsidered that this sample of staff would give the closest insight intoperceptions from staff providing care in the immediate aftermath ofself harm, mental health staff and non ED staff were excluded due tothe potential of specialist knowledge and non-acute setting influencingdiffering perceptions. In excluding these staff and settings several highquality papers were not entered into the study. There also exists poten-tial for publication bias, where there is selective reporting of completestudies with positive results. To limit these potential biases we hand-searched reference lists and also contacted first authors of all includedstudies to ask about unpublished material but this had a very low yield.

Conclusion

People who self harm often report that the care they receive fromhealth staff is unsatisfactory and yet paramedics and emergency staffare often the first professionals in contact with people who self harm.This study shows that some studies have been done into the attitudesof emergency nurses and doctors but very few have enquired intoambulance service staff and none have been carried out into modernprofessional paramedic care. The initial contact with patients followingself harm is a vital factor in these patients' acceptance or rejection ofcare, and further it may impact future self harm behaviours.

This study also found that the lack of departmental policy on selfharm was also an issue affecting perceived confidence amongst staff;neither limited formal nor informal procedures existed for managingpeople who self harm. This is of concern, as it may represent a failureof national guidelines to influence local and departmental policiesaimed to guide paramedic and emergency staff in their care for peoplewho self harm. This was also the case in relations to training in caringfor people who self harm; most staff reported never receiving suchtraining. Again this is of concern as national policy and guidelines advo-cate staff receive this training, yet the findings in this study point to apotential failure of such calls to translate into the adoption of trainingfor staff in practice.

Counter to findings from the patient perspective, there is evidencethat emergency staff increasingly hold positive attitudes towardspatients who self harm. However, the picture is more complex that sim-ply positive or negative. Attitudes are multi-dimensional, where selfharm can evoke feelings of irritation, anger, powerlessness, and frustra-tion in staff, and despite calls to receive the same priority and attentionas asthma or myocardial infarction, patients who self harm appear topresent a lower priority and legitimacy of care from staff. The perceptionsof care are therefore influenced by educational preparation of staff, expe-rience of caring for self harmpatients, age, religious belief, gender, setting,policies for dealingwith peoplewho self harm and, perceived confidence,assessment, referral and ability to cope with legal responsibilities.

Finally, it was found that staff in acute settings exhibit increased feel-ings of negativity, becoming less positive the closer to front line carewhich was linked to the busy workload in ED and competing priorities.Whist ED staff and paramedics share similarities in workload, few com-peting priorities present to paramedics who often deal with one patientat a time representing an important area which is under investigated. Aparamedicmay be the first health professional encountered by a patientfollowing self harm, and the lack of understanding of this encountermay be concealing the influence and potential contribution thatparamedics can make in the care of people who self harm. This studycontinues to evolve our understanding of care for people who selfharm, however further research is urgently required to explore theinfluence of these themes within the paramedic population.

Acknowledgements

This study is supported by Swansea University, the Pre HospitalEmergency Research Unit, Welsh Ambulance Services NHS Trust andFALCK Foundation Research Grant. No conflicts of interest exist.

References

[1] RCP. Royal College of Psychiatrists' report on self-harm, suicide, and risk. Availablefrom http://www.rcpsych.ac.uk/files/pdfversion/cr158.pdf; 2010. [Accessed on 02/03/2013].

[2] Horrocks J. Self poisoning and self injury in adults. Clin Med 2002;2:509–12.[3] NHS information Centre. NHS Information Centre report shows 10.6% increase in inten-

tional self harm hospital admissions in 3 years. Available from http://www.nhsconfed.org/Networks/MentalHealth/LatestNews/Archive/Pages/NHSInformationCentrereportshows106increaseinintentionalselfharmhospitaladmissions.aspx; 2011. [Accessed on10/10/2013].

[4] MIND. Understanding self harm and key facts and statistics. Available from www.mind.org.uk; 2014. [Accessed on 12/07/20012].

[5] Pages F, Arvers P, Hassler C, Choquet M. What are the characteristics of adolescenthospitalized suicide attempters? Eur Child Adolesc Psychiatry 2004;13:151–8.

[6] Ystgaard M, Reinholdt N P, Husby J, Mehlum L. Deliberate self harm in adolescents.Tidsskr Nor Laegeforen 2003;123:2241–5 [Norwegian].

[7] NICE. The short-term physical and psychological management and secondary pre-vention of self-harm in primary and secondary care. Available from www.nice.org.uk; 2004. [accessed on 2/01/2011].

[8] Harris J. Self harm: cutting the bad out of me. Qual Health Res 2000;10:164–73.[9] McAllister M, Creedy D, Moyle W, Farrugia C. Nurses' attitudes towards clients who

self harm. J Adv Nurs 2002;40:578–86.[10] Warm A, Murray C, Fox J. Who helps? Supporting people who self-harm. J Ment

Health 2002;11:12197130.[11] Mental health foundation. National inquiry into self-harm among young people:

truth hurts. London: Mental Health Foundation; 2006 [Available from: www.mentalhealth.org.uk [Accessed on 12/05/2013]].

[12] NCCMH. National Collaborating Centre for Mental Health Self-harm: cited in: RCP(2010). Self-harm, suicide and risk: helping people who self-harm. Final report ofa working group. College Report CR158 June 2010; 2004 [Available from: http://www.rcpsych.ac.uk/files/pdfversion/cr158.pdf [Accessed on 02/07/2014]].

[13] Crawford MJ, Turnbull G, Wessely S. Deliberate self harm assessment by accidentand emergency staff — an intervention study. J Accid Emerg Med 1998;15:189722.

[14] Cook S, Clancy C, Sanderson S. Self-harm and suicide: care, interventions and policy.Nurs Stand 2004;18:43–55.

[15] Vivekananda K. Integrating models for understanding self-injury. Psychother Aust2000;7:189725.

[16] Anderson M, Standen PJ, Noon JP. A social semiotic interpretation of suicidal behav-ior in young people. J Health Psychol 2005;10:317–31.

[17] Nehls N. Borderline personality disorder: the voice of patients. Res Nurs Health1999;22:28–51.

[18] RCP. Assessment following self-harm in adults: Council Report CR122. Availablefrom file:///C:/Users/bilbo/Documents/PHD/final%201/selh%20harm%20papers%23/systematic%20review%23/Assessment%20following%20SH%20CR122.pdf; 2004.[accessed on 04/02/2014].

[19] Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred reportingitems for systematic reviews and meta-analyses: the PRISMA Statement. PLoS Med2009;6:e1000097. http://dx.doi.org/10.1371/journal.pmed1000097.

[20] Social Care Institute for Excellence. Using qualitative research in systematic reviews:older people's views of hospital discharge. How knowledge works in social care.London: SCIE; 2006.

[21] Taylor T, Hawton K, Fortune S. Attitudes towards clinical services among peoplewhoself-harm; systematic review. Br J Psychiatr 2009;194:104–10.

[22] Saunders KEA, Hawton K, Fortune S, Farrell S. Attitudes and knowledge of clinicalstaff regarding people who self-harm: a systematic review. J Affect Disord 2012;139:205–16.

[23] ConlonM, O'Tuathail C. Measuring emergency department nurses' attitudes towardsdeliberate self-harm using the self-antipathy scale. Int Emerg Nurs 2012;20:3–13.

[24] McCarthy L, Gijbels H. An examination of emergency department nurses' attitudestowards deliberate self-harm in an Irish teaching hospital. Int Emerg Nurs 2010;18:29–35.

[25] Sun F, Long A, Boore J. The attitudes of casualty nurses in Taiwan to patients whohave attempted suicide. J Clin Nurs 2007;16:255–63.

[26] McCann T, Clark E, McConnachie S, Harvey I. Accident and emergency nurses' atti-tudes towards patients who self-harm. Accid Emerg Nurs 2006;14:4–10.

[27] Suokas J, Suominen K, Lonnqvist J. The attitudes of emergency staff towardattempted suicide patients: a comparative study before and after establishment ofa psychiatric consultation service. Crisis 2009;30:161–5.

[28] McAllisterM, Zimmer-GembeckM,MoyleW, Billett S.Working effectivelywith clientswho self-injure using a solution focused approach. Int Emerg Nurs 2008;16:272–9.

[29] Egan R, Kiran M, Sarma BA, Meena O'Neill BA. Factors influencing perceived effec-tiveness in dealing with self-harming patients in a sample of emergency departmentstaff. J Emerg Med 2012;43:1084–90.

[30] McLaughlin C. Casualty nurses' attitudes to attempted suicide. J Adv Nurs 1994;20:1111–8.

[31] Friedman T, Newton C, Coggan C, Hooley S, Patel R, OPickard M, et al. Predictors ofA&E staff attitudes to self-harm patients who use self laceration: influence of previ-ous training and experience. J Psychosom Res 2006;60:273–7.

[32] Turnbull G, Chalder T. Effects of education on attitudes to deliberate self-harm.Psychiatr Bull 1997;21:334–5.

[33] Mackay N, Barrowclough C. Accident and emergency staff's perceptions of deliberateself-harm: attributions, emotions and willingness to help. Br J Clin Psychol 2005;44:255–67.

[34] Ghodse A. The attitudes of casualty staff and ambulance personnel towards patientswho take drug overdoses. Soc Sci Med 1978;12:341–6.

456 N. Rees et al. / Journal of Psychosomatic Research 77 (2014) 449–456

[35] Domino G, Moore D, Westlake C, Gibson C. Attitudes towards suicide: a factoranalytic approach. J Clin Psychol 1982;3S:257–62.

[36] Suokas J, Lonnqvist J. Work stress has negative effects on the attitudes of emergencypersonnel towards patients who attempt suicide. Acta Psychiatr Scand 1989;79:474–80.

[37] McAllister M, Creedy D, Moyle W, Farrugia C. Study of Queensland emergency de-partment nurses' actions and formal and informal procedures for clients who selfharm. Int J Nurs Pract 2002;8:184–90.

[38] APPG. All Party Parliamentary Group onMental Health inquiry into parity of esteem.Available from http://www.rethink.org/media/967083/appg_written_evidence_-_emergency_care.pdf; 2014. [Accessed on 25/06/2014].

[39] Cleaver K. Attitudes of emergency care staff towards young people who self-harm: ascoping review. Int Emerg Nurs 2013. http://dx.doi.org/10.1016/j.ienj.2013.04.001.

[40] Owens D, Horrocks J, House A. Fatal and non-fatal repetition of self-harm. Br JPsychiatr 2002;180:193–9.

[41] Faruqui R, Afghan S. P03-445 — Suicide risk awareness in Pakistan: influence of re-ligious, cultural, legal, socio-economic and interpersonal factors. Eur Psychiatry2011;26:1615.

[42] Miller L, Wickramarante P, Gameroff MJ, SageM, Tenke CE,WeissmanMM. Religios-ity and major depression in adults at high risk: a ten-year prospective study. Am JPsychiatr 2012;169:89–94.

[43] Rasic D, Belik S, Elias B, Katz L, EnnsM, Sareen J. Spirituality, religion and suicidal be-havior in a nationally representative sample. J Affect Disord 2009;114:32–40.

[44] Oliffea JL, Ogrodniczukc JS, Bottorff JL, Johnsona JL, Hoyaka K. “You feel like you can'tlive anymore”: suicide from the perspectives of Canadian men who experience de-pression. Soc Sci Med 2012;74:506–14.

[45] Norheim B, Grimholt TK, Ekeberg O. Attitudes towards suicidal behaviour in outpa-tients clinics among mental health professionals in Oslo. BMC Psychiatry 2013;13:90.

[46] Tzeng W-C, Lipson JC. The cultural context of suicide stigma in Taiwan. Qual HealthRes 2004;14:345–58.

[47] Ouzouni C, Nakakis K. Nurses' attitudes towards attempted suicide. Health Sci J2013;7:119–34.