Is a clinician's personal history of domestic violence associated with their clinical care of patients: a cross-sectional study.

OBJECTIVE: To investigate whether domestic violence (DV) impacts on health professionals' clinical care of DV survivor patients. DESIGN,
SETTING: Descriptive, cross-sectional study at an Australian tertiary maternity hospital. PARTICIPANTS: 471 participating female health professionals (45.0% response rate). OUTCOME MEASURES: Using logistic and linear regression, we examined whether health professionals' exposure to lifetime DV was associated with their clinical care on specific measures of training, attitudes, identification and intervention.
RESULTS: DV survivor health professionals report greater preparedness to intervene with survivor patients in a way that is consistent with ideal clinical care. This indicates that personal DV experience is not a barrier, and may be a facilitator, to clinical care of survivor patients.
CONCLUSIONS: Health professionals are at the front line of identifying and responding to patients who have experienced DV. These findings provide evidence that survivor health professionals may be a strength to the healthcare organisations in which they work since among the participants in this study, they appear to be doing more of the work seen as better clinical care of survivor patients. We discuss the need for greater workplace supports aimed at promoting safety and recovery from violence and strengthening clinical practice with patients. © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

Adjustment for potential confounders in regression rendering it distinct in this under-researched field; the inclusion of health professionals from all clinical backgrounds reflected in hospitals, and the recruitment of primary domestic violence (DV) health professional survivors.

The single recruitment site that prevents generalisation of the findings, and survey self-report and social desirability, which may have led to the under-reporting of DV.

While our 45.0% response rate is not ideal, considering the work demands of the nursing and medical participants in this study, and the representational participation of nurses, doctors and allied health professionals, we argue that our response rate is both acceptable and comparable to similar research.

Background

Domestic violence (DV), including intimate partner, family violence and sexual assault, are common traumas for Australian female nurses, doctors and allied health professionals.1 DV is a global public health issue, defined by WHO as ‘any behaviour within an intimate relationship that causes physical, psychological or sexual harm to those in that relationship’.2 It can encompass partner violence, child abuse or abuse by any member of a household.2 Throughout this paper, we use the term ‘DV’ to refer to violence by a partner or a family member and ‘survivor’ when referring to someone (health professional or patient) who has experienced DV.3 Women who have survived DV have poorer physical and psychological health, requiring more healthcare than non-abused women.4 Australian women’s lifetime prevalence of physical or sexual violence by an intimate partner is 25%, with 2.1% experiencing violence in the last 12 months.5 A recent study of 471 Australian female health professionals found that the prevalence of intimate partner violence was higher than in the general community, and lower than among unwell women attending a general practitioner, with a lifetime prevalence of 33.6%, while the 12-month prevalence was 11.5%.1 The lifetime prevalence of DV (violence by a partner and/or other family member) was 45.2%.1

The role of the health system and health professionals is to identify survivor patients and provide a timely, evidence-based response.6 There is mixed evidence about whether health professionals’ personal experiences of DV have an impact on the clinical care of their survivor patients.7–15 An extensive search of the academic literature identified four surveys about survivor health professionals’ clinical care of survivor patients.7 8 10 15 Two of these studies found that survivor health professionals performed more DV screening and raised DV with survivor patients more frequently during follow-up visits.7 8 However, the other two studies found no association between DV experience and clinical care.10 15 There were problems with three of these four studies.7 8 10 For example, two did not adjust for potentially confounding factors in their analysis,7 8 and the third, now nearly 20 years old, defined their survivor exposure group based on only two non-validated DV questions.10 The strongest research to date surveyed Swedish health professionals (n=588).15 After adjusting for professional background, experience and training, it found that care of survivor patients was not associated with personal experience of DV, however, DV training was positively associated with all aspects of care and knowledge.15 Another four studies about clinical care of survivor patients have been from the perspective of health professionals whose DV exposure was through family, friends or patients.9 11 13 14 We argue that the need for a more rigorous study is evident.

Methods

Aim, design and setting

The objective of this study was to address a gap in the available evidence about whether Australian health professional’s personal history of DV is associated with their clinical care of survivor patients. The research question at the outset of this project was: Is personal experience of DV associated with a health professional’s attitudes about DV survivor patients and the role of the health workplace; identification of survivor patients; comfort to discuss DV and clinical interventions with survivor patients? We hypothesised that, after adjusting for possible confounding background variables, compared with their non-abused peers, survivor health professionals would: (1) demonstrate more sensitive attitudes about survivors; (2) feel more comfortable discussing DV and sexual assault with their patients; (3) ask more patients about DV; (4) identify more survivors within a 6-month period and (5) provide more DV interventions to survivor patients, including DV referral. While not an initial focus of the study, the effect of training on clinical practice emerged as an interesting finding during the data analysis and was included in the results.

A full description of the study design, setting, participants and recruitment process has been reported previously in a paper about prevalence.1 In brief, we conducted an anonymous and voluntary cross-sectional survey of all health professionals in one Australian tertiary maternity hospital between 8 August and 31 December 2013. Participants were female health professionals (nurses, doctors and social workers) working with patients. An online survey link and encouragement to participate by the chief executive officer was distributed via email to all part-time/permanent clinical staff—nurse/midwives, doctors and allied health professionals. Staff were ineligible to participate if they were employed casually or did not work in a clinical capacity (ie, administration staff).

Data collection and measures

Exposure to DV encompassed 12 month and adult lifetime intimate partner violence and/or lifetime violence by a family member. Violence by an intimate partner was measured using the Composite Abuse Scale, a well-validated and widely used self-report measure of physically, sexually and emotionally abusive behaviours perpetrated by an intimate partner.16 This was measured by: scoring on the 12-month subscales, or two of the lifetime subscales, ‘Severe Combined Abuse’ or ‘Physical and Emotional Abuse’, or by scoring >7. Violence by a family member was measured by answering positively to either of two questions about lifetime physical, emotional and sexual abuse by a family member and witnessing parental abuse. Overall, 45.2% (212/471) of the female participants in this sample qualified for inclusion into the DV exposure group.1

The main predictor variable was exposure to DV. In a follow-up analysis, the predictor variables were DV training and demographics. The outcome variables were: attitudes (measured by Physician Readiness to Manage Intimate Partner Violence Survey,17 comfort discussing DV, DV inquiry and interventions after identifying a new DV case during a 6-month period (table 1). Adjustment for potential confounding variables was made a priori based on the literature, and included: age (40+ years),14 18 professional background (allied health),10 14 DV training (1+ days)9 10 15 and years of clinical experience (10+ years).14 15

Table 1

Variables included in analysis*

	Description
Independent variables
Exposure to DV†	30 CAS items measured 12 month and lifetime intimate partner violence and 2 family violence questions.
DV training‡§	6 items measured graduate and postgraduate DV training history (<8 hours‡/>8 hours).¶
Demographics**	3 items measured: age (<40 years/>40 years), professional background (medical/nursing/allied health) and years of clinical experience (<20 years/>20 years).
Dependent variables
Attitudes††	12 PREMIS items comprised two subscales; ‘Victim understanding’ (attitudes about survivors) and ‘Workplace issues’ (attitudes about the role of the workplace). Scoring via a 7-point Likert-type scale, with some items reverse scored due to intentional negative wording.
Comfort discussing DV§	4 items scored on a 5-point Likert-type scale measured comfort to discuss DV and sexual assault with patients (‘comfortable’/‘uncomfortable’).
	4 items scored on a 6-point Likert-type scale measured: ‘Did not avoid issue of DV’, ‘Did not find DV upsetting to talk about’, ‘Very aware of the issue’ and ‘Tried to go the extra mile with patients’ (‘agree’/‘disagree’). Some items reverse scored due to intentional negative wording.
DV inquiry§	1 item scored on a 5-point Likert-type scale measured frequency of asking all patients about DV (‘never’/‘ever’) during the previous 6 months.
	5 items measured identification of 1+ new patient survivor/s (‘0 new cases’/‘1+ new cases’) in the previous 6 months.
Interventions after identifying a new DV case§	10 items scored on a 5-point Likert-type scale measured: risk assessment, safety planning, case file documentation, use of clinical guideline, access of DV information to give to patients, clinical discussion at team meeting and with manager and DV referrals (‘never’/‘1–3+ times’) during the previous 6 months.
Variables used for adjustment§
Age	>40 years
Professional background	Allied health: social workers were the most common allied health professionals at this hospital and it was anticipated that they would likely have been in receipt of greater undergraduate and professional DV training.
DV training	>8 hours
Years of clinical experience	>10 years

*All items/measures were made into binary variables unless otherwise noted.

†Exposure to DV measured via CAS.16

‡Training also analysed as an outcome (dependent) variable.

§Bespoke item developed for the survey based on an extensive review of the literature.

¶Participants with no DV training were included in ‘<8 hours’.

**Demographic measures based on recruitment site specific data and Australian Institute of Health and Welfare.19

††Attitudes measured via PREMIS.17

CAS, Composite Abuse Scale; DV, domestic violence; PREMIS, Physician Readiness to Manage Intimate Partner Violence Survey.

Statistical analysis

Clinical interventions to identify and respond to DV were summarised using frequencies and percentages for categorical data and means and SD for ordinal data. Independent t-tests and X2 tests of comparison were used to compare mean scores. Linear regression compared differences in mean scores across exposure for attitude scores, while logistic regression was used for comfort asking about DV and clinical intervention variables. ORs, 95% CIs and p values were used to assess the likely size of the association between each clinical action and DV.

Data were analysed with STATA V.13.1.20

Patient and public involvement

No patients or the public were involved in developing the research question or outcome measures. Health professionals were involved, however, and they were informed by their clinical work with survivor patients. Health professionals contributed to the research questions and overall design of the study. Results of the study will be disseminated to participants via workplace newsletter items and staff public speaking forums at the recruitment site.

Results

Participant characteristics

The survey was sent to 1047 female health professional staff and 471 participated: 366 completed the survey electronically, while 105 returned a paper version, giving a response rate of 45.0%. Most participants were nurse/midwives, aged 30–60 years, had 10 or more years of experience, and were demographically representative of their non-participating peers (table 2). Survivor health professional participants (45.2%, 212/469) were significantly more likely to be aged 30–39 years and have an allied health background compared to participants who were not survivors.1

Table 2

Personal characteristics of participating health professionals

Characteristic	Total participants* (n=471) n (%)	No history of violence (n=257) n (%)	Lifetime domestic violence (n=212) n (%)	P value
Age (years)
<30	81 (17.2)	52 (20.2)	29 (13.7)	0.063
30–39	123 (26.2)	57 (22.2)	66 (31.1)	0.029
40–49	100 (21.3)	54 (21.0)	46 (21.7)	0.857
50–59	133 (28.3)	70 (27.2)	62 (29.2)	0.630
≥60	33 (7.0)	24 (9.3)	9 (4.2)	0.036
Health professional background
Nursing/midwifery	317 (67.5)	181 (70.7)	134 (63.2)	0.086
Medical	69 (14.7)	38 (14.8)	31 (14.6)	0.946
Allied health	61 (13.0)	21 (8.2)	40 (18.9)	0.001
Other†	23 (4.9)	16 (6.3)	7 (3.3)	0.148
Years of clinical experience
<5	70 (15.0)	39 (15.4)	31 (14.6)	0.826
5–9	67 (14.3)	35 (13.8)	32 (15.1)	0.687
10–19	119 (25.4)	62 (24.4)	57 (26.9)	0.542
20–29	99 (21.2)	53 (20.9)	45 (21.2)	0.924
≥30	113 (24.2)	65 (25.6)	47 (22.2)	0.390
Participants who supervise other staff	226 (48.2)	122 (47.8)	102 (48.1)	0.954
Adult intimate relationship (ever)‡	431 (92.9)	222 (88.1)	209 (98.6)	<0.01

*Denominators vary due to missing responses. Maximum missing data n=3 (0.6%).

†Health professionals working in a clinical role not already specified, that is, imaging, pharmacy.

‡33 participants were omitted from relationship questions because they had never been in a relationship.

Training and preparedness

Survivor health professionals were more likely to have received one or more days of DV training (adj OR 1.9, 95% CI 1.2 to 3.2) and to report more sensitive attitudes about DV survivors (adj. coef. 0.2, 95% CI 0.1 to 0.4) compared with their colleagues who had not experienced DV. Survivor health professionals were no more likely than others to find it upsetting to talk about DV with their patients (adj OR 0.8, 95% CI 0.5 to 1.1) (table 3). Irrespective of whether a health professional had experienced DV, having undertaken at least 1 day of DV training was positively associated with good clinical care, including identifying survivor patients (adj OR 9.6, 95% CI 5.0 to 18.8), risk assessment (adj OR 4.6, 95% CI 2.2 to 9.5), safety planning (adj OR 4.3, 95% CI 2.1 to 8.9) and referral (adj OR 2.1, 95% CI 1.0 to 4.1). This finding occurred even after adjustment for possible confounders (table 4). Univariate analysis suggested a positive association between hours of DV training and asking patients about the issue. The analysis also suggested that allied health professional participants (ie, social workers) were more likely to have had 1+ days of DV training and to have safety planned and referred survivor patients than the other professional groups (table 4).

Table 3

Health professional’s personal exposure to DV and their clinical practice

	All participants (n=471)*	Lifetime abuse by partner/family member		Unadjusted	Adjusted†	P values
		No abuse (n=257)	Abuse (n=212)
	n (%)			OR (95% CI)
Training (1+ days)	94 (20.1)	36 (14.1)	58 (27.4)	2.3 (1.4 to 3.6)	1.9 (1.2 to 3.2)	0.007
Preparedness for practice	Mean (SD)			Coef (95% CI)
Attitudes about survivors	5.1 (1.0)	4.9 (1.0)	5.3 (0.9)	0.3 (0.2 to 0.5)	0.2 (0.1 to 0.4)	0.009
Attitudes about the role of health services	4.4 (1.1)	4.3 (1.0)	4.4 (1.2)	0.1 (−0.1 to 0.3)	−0.1 (−0.3 to 0.1)	0.550
	n (%)			OR (95% CI)
Recent clinical practice‡	(n=422)	(n=226)	(n=194)
Comfort discussing DV	194 (46.0)	94 (41.6)	99 (51.0)	1.5 (1.0 to 2.2)	1.1 (0.7 to 1.7)	0.578
Comfort discussing sexual assault	165 (39.0)	77 (34.1)	87 (44.8)	1.6 (1.1 to 2.3)	1.2 (0.8 to 1.8)	0.455
Did not avoid issue of DV	254 (61.9)	93 (42.5)	62 (32.8)	1.5 (1.0 to 2.3)	1.3 (0.8 to 2.0)	0.232
Did not find upsetting to talk about	229 (55.8)	127 (56.0)	102 (54.0)	0.8 (0.6 to 1.2)	0.8 (0.5 to 1.1)	.186
Very aware of the issue	220 (54.3)	107 (50.0)	111 (59.0)	1.4 (1.0 to 2.1)	1.2 (0.8 to 1.8)	0.399
Tried to go the extra mile with patients	181 (44.5)	84 (38.7)	95 (50.3)	1.6 (1.1 to 2.4)	1.3 (0.9 to 2.0)	0.205
DV inquiry
Inquiry of 1+ patient/s	260 (61.6)	124 (54.9)	134 (69.1)	1.8 (1.2 to 2.7)	1.5 (1.0 to 2.3)	0.074
Identified 1+ new cases	193 (45.7)	91 (40.1)	101 (52.3)	1.6 (1.1 to 2.4)	1.3 (0.8 to 2.0)	0.263
Intervention/s with survivor patients§	(n=193)	(n=91)	(n=101)
Risk assessment	102 (53.7)	41 (46.1)	60 (60.0)	1.8 (1.0 to 3.1)	1.2 (0.6 to 2.4)	0.501
Safety planning	80 (41.7)	28 (31.1)	52 (51.5)	2.3 (1.3 to 4.2)	1.6 (0.8 to 3.2)	0.208
Case file documentation	139 (72.4)	63 (70.0)	75 (74.3)	1.2 (0.6 to 2.3)	1.1 (0.5 to 2.2)	0.786
Utilised DV clinical practice guideline	76 (40.0)	37 (41.1)	38 (38.4)	0.9 (0.5 to 1.6)	0.7 (0.4 to 1.4)	0.363
Accessed DV information	60 (31.4)	22 (24.4)	37 (37.0)	1.8 (1.0 to 3.4)	2.0 (1.0 to 4.0)	0.040
Discussed DV at a team meeting	125 (66.1)	56 (62.2)	68 (69.4)	1.4 (0.7 to 2.5)	1.2 (0.6 to 2.3)	.542
Discussed a DV case with manager	146 (76.4)	66 (74.2)	79 (78.2)	1.2 (0.6 to 2.4)	1.1 (0.5 to 2.3)	0.751
DV referrals
Internal hospital service	166 (86.0)	78 (85.7)	87 (86.1)	1.0 (0.5 to 2.3)	1.0 (0.4 to 2.5)	0.960
Community DV service	78 (40.6)	30 (33.3)	48 (47.5)	1.8 (1.0 to 3.3)	1.3 (0.7 to 2.7)	0.387

*Denominators vary due to missing values, maximum missing values n=19 (4.0%).

†Adjusted for age (40+ years), profession (social work), years of clinical experience (10+ years), training (1+ days).

‡During the last 6 months. 48 participants were excluded from the remaining analyses because they had not been in clinical practice.

§277 participants were excluded from analyses (229 participants who had not identified a new DV case and 48 participants not in clinical practice).

DV, domestic violence.

Table 4

The effect of training on clinical practice

	All participants (n=471)*	Length of training		Unadjusted	Adjusted†	P values
		<1 day (n=375)	1+ day (n=94)
	n (%)			OR (95% CI)
Demographics
Age
<40 years	204 (43.4)	169 (45.2)	33 (35.1)	0.7 (0.4 to 1.0)	0.6 (0.3 to 1.1)	0.090
>40+  years	266 (56.6)	205 (54.8)	61 (64.9)	1.5 (0.9 to 2.4)	1.8 (0.9 to 3.4)	0.090
Professional background
Nursing/midwifery	317 (67.5)	268 (71.7)	48 (51.1)	0.4 (0.3 to 0.6)	0.4 (0.2 to 0.6)	0.000
Medical	69 (14.7)	55 (14.7)	13 (13.8)	0.9 (0.5 to 1.8)	0.9 (0.5 to 1.8)	0.816
Allied health	61 (13.0)	31 (8.3)	30 (31.9)	5.2 (2.9 to 9.1)	5.3 (3.0 to 9.4)	0.000
Years of clinical experience
<20 years	256 (54.7)	207 (55.6)	47 (50.0)	0.8 (0.5 to 1.2)	0.6 (0.3 to 0.9)	0.028
>20+  years	212 (45.3)	165 (44.3)	47 (50.0)	1.2 (0.8 to 2.0)	1.7 (1.1 to 2.9)	0.028
Supervision of other staff	226 (48.2)	175 (46.9)	49 (52.1)	1.2 (0.8 to 1.9)	1.7 (1.0 to 2.9)	0.038
Preparedness for practice	Mean (SD)			Coef (95% CI)
Attitudes about survivors	5.1 (1.0)	4.9 (1.0)	5.8 (0.7)	0.9 (0.6 to 1.1)	0.8 (0.6 to 1.1)	0.000
Attitudes about the role of health services	4.4 (1.1)	4.2 (1.0)	5.0 (1.2)	0.8 (0.6 to 1.1)	0.7 (0.4 to 0.9)	0.000
	n (%)			OR (95% CI)
Recent clinical practice‡	(n=422)	(n=336)	(n=84)
Comfort discussing DV	194 (45.9)	125 (37.1)	68 (80.9)	7.2 (4.0 to 13.0)	6.4 (3.5 to 11.8)	0.000
Comfort discussing sexual assault	165 (39.0)	104 (30.9)	61 (72.6)	5.9 (3.5 to 10.1)	5.1 (2.9 to 8.9)	0.000
Did not avoid issue of DV	254 (61.9)	190 (58.5)	63 (75.9)	2.3 (1.3 to 3.9)	2.2 (1.2 to 3.9)	0.008
Did not find upsetting to talk about	229 (55.8)	173 (53.2)	54 (65.1)	1.6 (1.0 to 2,7)	1.6 (0.9 to 2.7)	0.095
Very aware of the issue	220 (54.3)	149 (46.4)	70 (85.4)	6.7 (3.5 to 12.9)	7.0 (3.5 to 13.7)	0.000
Tried to go the extra mile with patients	181 (44.5)	118 (36.5)	62 (75.6)	5.4 (3.1 to 9.3)	5.0 (2.8 to 8.9)	0.000
DV inquiry
Inquiry of 1+ patient/s	260 (61.6)	178 (53.0)	81 (96.4)	24.0 (7.4 to 77.4)	24.1 (7.3 to 78.8)	0.000
Identified 1+ new cases	193 (45.7)	121 (35.9)	71 (85.5)	10.6 (5.5 to 20.2)	9.6 (5.0 to 18.8)	0.000
Intervention/s with survivor patients§	(n=193)	(n=121)	(n=71)
Risk assessment	102 (53.7)	47 (39.5)	54 (77.1)	5.2 (2.6 to 10.1)	4.6 (2.2 to 9.5)	0.000
Safety planning	80 (41.7)	31 (25.8)	48 (67.6)	6.0 (3.1 to 11.4)	4.3 (2.1 to 8.9)	0.000
Case file documentation	139 (72.4)	76 (63.3)	62 (87.3)	4.0 (1.8 to 8.8)	3.4 (1.5 to 7.8)	0.004
Utilised DV clinical practice guideline	76 (40.0)	32 (26.9)	43 (61.4)	4.3 (2.3 to 8.1)	4.2 (2.1 to 8.3)	0.000
Accessed DV information	60 (31.4)	32 (26.7)	27 (38.6)	1.7 (0.9 to 3.2)	1.7 (0.9 to 3.4)	0.120
Discussed DV at a team meeting	125 (66.1)	69 (59.0)	55 (77.5)	2.4 (1.2 to 4.7)	2.4 (1.1 to 5.0)	0.019
Discussed a DV case with manager	146 (76.4)	82 (68.9)	63 (88.7)	3.5 (1.5 to 8.2)	3.3 (1.4 to 8.1)	0.007
DV referrals
Internal hospital service	166 (86.0)	97 (80.2)	68 (95.8)	5.6 (1.6 to 19.4)	6.4 (1.7 to 23.6)	0.005
Community DV service	78 (40.6)	35 (29.2)	42 (59.1)	3.5 (1.9 to 6.5)	2.1 (1.0 to 4.1)	0.042

*Denominators vary due to missing value. Maximum missing data n=3 (1.5%), unless otherwise specified.

†Adjusted for age (40 years and older), profession (social work) and years of clinical experience (10 or more years).

‡During the last 6 months, 48 participants were excluded from the remaining analyses because they had not been in clinical practice.

§277 participants were excluded from analyses (229 participants who had not identified a new DV case and 48 participants not in clinical practice).

DV, domestic violence.

Identifying survivor patients

In the unadjusted analysis, being a survivor health professional was associated with asking patients about DV during the previous 6 months and motivation ‘to go the extra mile’ with them. However, in the adjusted analysis a between-group difference did not remain, although the significance level for asking patients about DV was approaching 0.05 (adj OR 1.5, 95% CI 1.0 to 2.3, p=0.07) (table 3).

Clinical care

Of the 193 participants who identified a survivor patient in the last 6 months, the unadjusted results indicated that survivor health professionals were more likely than others to have provided DV information to patients, conducted risk assessments, safety plans and made referrals to services (table 3). However, in the adjusted analysis, the only association that remained was accessing DV information for patients (adj OR 2.0, 95% CI 1.0 to 4.0).

Discussion

These findings provide evidence that survivor health professionals may be doing more of the work seen as better clinical care of survivor patients than those without personal experience. Being a survivor health professional was significantly associated with uptake of DV training, more sensitive attitudes about survivors and a higher likelihood of having accessed DV information to give to survivor patients, which supports the hypothesis that survivor health professionals would demonstrate more sensitive attitudes about survivors compared with their non-abused peers. There was only partial support for the hypothesis that survivor health professionals would recall providing more DV interventions to survivor patients since the only significant association was having accessed more DV information for patients. However, the hypotheses that survivor health professionals would feel more comfortable discussing DV with their patients, ask more patients about DV, and identify more survivors within a 6-month period, were not supported after adjusting for age, years of experience and training. It is notable that survivor health professionals asked more patients about DV at a level approaching significance.

Strengths and limitations

Strengths of this study include adjustment for potential confounders in regression,7 8 11 13 14 the inclusion of health professionals from all clinical backgrounds reflected in hospitals7 8 10–14 and the recruitment of primary DV survivors.9 11 13 14 Limitations of this study include self-report and social desirability, which may have led to under-reporting of abuse, and the single recruitment site that prevents generalisability of findings.21 22 It is possible that DV survivors were more motivated to participate in the project than the other people,21 and we acknowledge the possibility that non-respondents may have differed from respondents in a way that affected our conclusions. Considerable attempts were made to address selection bias by active recruitment and strong encouragement to participate; a 45.0% response rate was achieved. Despite the sample limitations, considering the work demands of our participants and the representational participation of nurses, doctors and allied health professionals, we argue that our response rate is acceptable and comparable to similar research.7 8

The study in the context of other studies

The finding of an association between a health professional’s history of DV and aspects of clinical care of survivor patients echoes other research.7 14 A possible interaction between DV training, personal experience and clinical care has been suggested previously.9 However, the finding in this study of a relationship between a health professional’s history of DV and their participation in training is critical and new. This finding was surprising; we did not posit a hypothesis about survivors accessing more hours of professional training. We suggest that survivor health professionals may be more likely to attend training because they understand the issue, resultant impact on health and the need for timely responses, and/or they are seeking information or validation about their own experience.

The association between being a survivor health professional, holding more sensitive attitudes about survivors and providing DV information to patients is consistent with one previous study.14 This small study examined nurses’ thoughts, feelings and proposed actions in response to identifying survivor patients, finding an association between being a survivor nurse and having more sensitive, empathetic responses to survivor patients.14 Our study extends these findings since that analysis did not adjust for potential confounders and the exposure group included health professional participants with secondary exposure to DV through friends/family. We postulate that survivor health professionals may hold more sensitive attitudes about survivors and fewer misconceptions about DV because of empathy stemming from a shared trauma experience. Additionally, they may be more likely to access DV information for their patients because they believe that DV awareness is an important intervention in itself.

Implications

Given the association between being a survivor health professional and attendance at DV training, this should be regarded when developing and delivering DV training for health professionals.7 Such training could incorporate reflection, safety information, emotional health psychoeducation, referral, workplace support and promoting a safe and supportive healthcare workplace.15 23 More broadly, these findings provide evidence that survivor health professionals are an asset to the organisations in which they work since among the participants in this study, they appear to be doing more of the work seen as better clinical care of survivor patients. This finding rebukes the misconception that women who have experienced DV are enduringly vulnerable, a distortion which can encourage women to remain silent, especially at work, for fear of how they might be regarded if they speak up.24 This study presents an opportunity for health services to explore how the lived experience of DV for both their patient and staff survivors could inform and improve their service. A past critique of health and other ‘mainstream’ DV response services has been that they have not meaningfully consulted survivors.25 Listening to the experiences and needs of survivor health professionals may enhance the support those health professionals feel from their employer, strengthening their personal and professional capacity as they care for patients. There is evidence that accessing support for DV can result in meaningful change in survivors’ lives, including in their employment.18 We argue the need for greater workplace supports aimed at promoting safety and recovery from violence and strengthening clinical practice with patients. This requires organisational leadership, evidence-based response guidelines and resourced individuals to whom a disclosure can be made and who can provide varied levels of support (resource information, clinical debriefing, longer term emotional support).1 Trauma-informed care may provide a useful framework to guide the response of hospitals towards better supporting staff and patient DV survivors.26 A trauma-informed system is one in which all components have been organised with the understanding that trauma is a centralising influence in survivor’s lives, and organisational, operational and clinical practice should prioritise safety, control and the recovery trajectory.27 More research is required to better understand the impact of DV workplace supports for DV on health professional women’s well-being and clinical care. This study sheds light on the survivor experience, especially for women at work.

Conclusion

This research demonstrates that health professionals with a lived experience of DV attend more training aimed at improving clinical care of survivor patients, self-report more sensitive attitudes about survivors and access more DV information for patients after disclosure. This suggests that DV is not a barrier, and may be a facilitator, to clinical care of survivor patients. Healthcare workplaces should take account of this in their response to survivor health professionals, the development of DV training offered to staff, clinical care policies with patients and workplace supports.