Personal experiences and attitudes towards intimate partner violence in healthcare providers in Guyana Vivienne Mitchella, Kendra P. Parekhb,*, Stephan Russb, Nicolas P. Forgeta,b and Seth W. Wrightb a
Georgetown Public Hospital Corporation, New Market Street, 314132, Georgetown, Guyana; bDepartment of Emergency Medicine, Vanderbilt University, 703 Oxford House, Nashville, TN, 37232 USA *Corresponding author: Tel: +1 615 936 0087; Fax: +1 615 936 1316; E-mail: [email protected]
Received 26 August 2013; revised 17 October 2013; accepted 18 October 2013 Background: Intimate partner violence (IPV) is prevalent throughout the world and is a devastating public health problem. Healthcare workers (HCWs) are tasked with treating victims of IPV but may be victims themselves. Guyana is a lower-middle income country in South America. This study sought to determine the knowledge and attitudes of Guyanese HCWs and their perceived barriers to providing care in addition to determining the prevalence of IPV victimization and perpetration among HCWs. Methods: HCWs at the only tertiary care hospital in the Guyana completed an anonymous survey that comprised 30 questions relating to IPV. Results: The survey was completed by 87.5% of eligible HCWs. Of the respondents, 81.8% were female, 49.9% had ever experienced abuse and 21% admitted to perpetrating violence. Multivariate analysis found that the age groups 31–40 years (OR 2.3, 95% CI 1.1–4.6) and 41–50 years (OR 2.3, 95% CI 1.2–4.7) had higher odds of accepting justification for physical violence, and so did nursing staff (OR 4.3, 95% CI 1.4–13.1). Overall, 29.9% of HCWs accepted justification for physical violence in at least one of the named scenarios. Conclusion: This study demonstrates a high prevalence of IPV among HCWs and identifies prevailing attitudes regarding IPV. This knowledge is essential in developing effective, appropriate training programs and identifies a need to address IPV among the healthcare workforce. Keywords: Domestic violence, Spouse abuse, Women’s health, Attitude of health personnel, Caribbean, Guyana
Introduction Intimate partner violence (IPV) is prevalent throughout the world and has been recognized as a public health problem with shortand long-term health consequences.1,2 The healthcare system is an integral component in the societal management of IPV. Most victims of IPV access the healthcare system at some point and it is an important portal into the social service, counseling, and judicial systems.3 Although healthcare workers (HCWs) often treat victims of IPV, HCWs may themselves be victims and/or perpetrators of violence, especially in societies where IPV is highly endemic. Community and society factors influence attitudes toward IPV and HCWs often share the prevailing cultural attitudes regarding IPV. For example, a study of physicians and nurses at a large university hospital in Turkey found that 69% of females and 84.7% of males accepted at least one justification for physical violence against a partner.4 Similar results have been found among health professionals in Serbia and nurses in rural South Africa.5,6 Guyana is a lower-middle income country with cultural and economic ties to other English-speaking Caribbean countries. It is a founding member and host country of the Caribbean
Community Secretariat. In general, women are in a disadvantaged position in society with decreased access to resources, less decision-making power, and higher rates of poverty.7 One particularly prevalent issue is IPV and there are almost daily media stories regarding instances of IPV.7,8 A recent retrospective study found that 16% of acute traumatic injuries in women presenting to the emergency department at the main public hospital were a result of IPV.9 The high prevalence of IPV has led to the development of a National Policy on Domestic Violence and various entities, including non-governmental organizations and the Guyana Ministry of Health, have partnered to reduce the prevalence.10 The training of HCWs is considered an essential part of this response. However, there is a paucity of data on which to base current training models for HCWs in Guyana or similar countries. Most IPV educational programs have been designed in higher-income countries. It is largely unknown how these educational programs function in lower-income countries and to be most effective, they likely require modification to address cultural norms and prevailing local attitudes. In order to provide key information for the development of IPV training programs for HCWs in Guyana and similar countries, this
# The Author 2013. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
273
ORIGINAL ARTICLE
Int Health 2013; 5: 273–279 doi:10.1093/inthealth/iht030 Advance Access publication 13 November 2013
V. Mitchell et al.
study sought to determine the knowledge and attitudes of HCWs regarding IPV and their perceived barriers to providing care for patients experiencing IPV. We also sought to determine the prevalence of IPV victimization and perpetration among current HCWs at the main public hospital.
Materials and methods Setting Guyana is an English-speaking country on the North Atlantic coast of South America. Its population of about 750 000 is ethnically diverse, with most individuals being of East Asian, African, or Amerindian heritage. As of 2010, the per capita gross domestic product was US$2996.11 Georgetown is the capital city and the main population center for the country. Georgetown Public Hospital Corporation (GPHC) is a government-funded teaching facility and is Guyana’s only tertiary care medical center.
Study design A cross-sectional survey of HCWs at GPHC was conducted during September 2011.
Study procedures Seven GPHC senior nurses and a social worker were selected via the Quality Control Office at GPHC to administer the survey. All surveyors underwent training by a sociologist and were supervised by a staff physician to ensure there was no undue influence on staff to participate in the study and that anonymity was maintained. Inclusion criteria were: age 18 years or older, current healthcare worker with the position of nursing staff (registered nurse, midwife, or nursing assistant) or physician staff (consultant physician, junior physician, or intern). Those selected to administer the survey or otherwise involved in carrying out the study were not eligible for participation. All eligible HCWs were approached during the workday and invited to voluntarily participate in the study. HCWs were explicitly informed that they did not have to participate and that all answers to the survey questions were anonymous. If the HCW agreed to participate, written informed consent was obtained. Participants were told that they did not have to complete items that made them feel uncomfortable and that they could stop the survey at any time. All surveys were completed anonymously by the participants on paper in a private space and then placed in sealed envelopes in a locked ballot box. The surveyors did not directly ask any survey questions. All HCWs were offered a brochure with IPV resources. The Guyana Ministry of Health Institutional Review Board approved the study protocol.
Survey instrument The survey instrument consisted of 30 questions and was developed following a literature review. It was then modified by local experts to ensure cultural acceptability and appropriateness. Prior to implementation, the instrument was pilot tested for comprehensibility and acceptability with a group of Guyanese HCWs. Basic demographic data were collected from each subject, including age (18–30, 31–40, 41–50, or .50 years), gender, job
274
title, years of health care experience (,5, 6–10, 11–15, 16–20, .20 years), and self-reported history of participation in an IPV training course. The instrument included items relating to seven domains (personal experiences, attitudes toward the justification of violence, attitudes toward the victim, attitudes toward screening for IPV, attitudes toward the outcome, knowledge, and barriers). A series of statements were presented and participants rated agreement using a 5-point Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’ for all of the items, with the exception of three questions relating to personal experiences that allowed for a ‘yes’ or ‘no’ answer and one open-ended question as described below. To assess personal experiences with IPV, an affirmative response to the question ‘Have you ever been emotionally or physically abused by your partner or someone important to you?’ or to the question ‘Within the last year, have you been hit, slapped, kicked, or otherwise physically hurt by someone?’ was considered a positive IPV screen. These two questions were derived from the Abuse Assessment Screen, a validated and simple IPV screening tool.12 A single question modified from the Multi-country National Demographic and Health Surveys program, ‘Have you ever hit, slapped, kicked, or done anything else physically abusive to an intimate partner?’ was used to determine if participants ever perpetrated physical violence against an intimate partner.13 A positive IPV screen or an affirmative answer to the perpetration question was considered a positive IPV experience screen. Attitudes towards the justification of violence were assessed with three questions adopted for local context from questions used in the Multi-country Demographic and Health Surveys program.1,13 These questions ask for agreement with scenarios that might justify a man beating a woman and range from trivial (‘spends too much time with friends’) to more socially/ morally serious (‘is unfaithful’). Each subject was classified as having a firm negative (answering ‘strongly disagree’ or ‘disagree’ to each of the three questions) or a positive attitude (answering ‘strongly agree,’ ‘agree,’ or ‘neutral’ to any of the three questions) toward the scenarios.14 Attitudes toward the victim, attitudes toward screening, knowledge, and barriers were assessed with four questions each while attitudes toward outcome were assessed with two questions. These questions were modified from surveys designed to examine HCWs’ attitudes and knowledge regarding IPV.15–17 There was a single open-ended question asking, ‘Under the Domestic Violence Act, what is an occupation order?’ An occupation order is part of legislation designed to protect victims of partner abuse in Guyana.18 A magistrate can grant an occupation order so that an IPV victim may live in the home even if it belongs to the abuser. The abuser is not permitted to live in the same home as the victim. Answers were dichotomized as correct or incorrect. An answer was considered correct if any element of an occupation order (e.g. ‘victim can remain in the home,’ ‘abuser can’t live in the home’) was included in the response. This openended question was considered one of the four questions within the knowledge domain.
Data analysis Statistical analysis was done using Stata 10.1 software (StataCorp LP, College Station, TX, USA). Logistic regression was used in a multivariate analysis to determine covariates associated with a positive attitude when queried about the justification of a man beating his
International Health
wife in any of three named scenarios (she argues with him, she spends too much time with friends, she is unfaithful). All covariates were entered into the logistic regression model with listwise deletion used to exclude observations with missing data. Odds ratios with 95% confidence intervals are reported along with the p-value. A p-value of ,0.05 was considered statistically significant.
Results There were an estimated 460 eligible staff (375 nursing staff and 85 physician staff) employed at GPHC during the study period excluding those involved in carrying out the study. Forty-five (41 nurses and 4 physicians) were on leave during the study period, leaving a total of 415 potential participants. A total of 363 HCWs consented and participated in the survey, representing 87.5% of those estimated to be eligible and present during the study period. Some participants declined to answer every question. Table 1 describes the demographic findings. There were 297 females, 48 males, and 18 unspecified. There were 283 nurses, 50 physicians and 30 unspecified. Most (256/283, 90.5%) of those identifying themselves as nursing staff were female and 46.0% (23/50) of those identified as physicians were female. Only 24/363 (6.6%) reported they had previously taken an IPV course. Table 1. Demographic characteristics of healthcare workers (n¼363) participating in a survey conducted to explore knowledge and attitudes in relation to intimate partner violence Characteristic Gender Female Male Not specified Age (years) 18–30 31–40 41–50 .50 Not specified Position Nursing staff Physician staff Not specified Healthcare experience (years) ,5 5–10 11–15 16–20 .20 Not specified Previous IPV training Yes No Not specified IPV: intimate partner violence.
Number (%)
297 (81.8) 48 (13.2) 18 (5.0) 196 (54.0) 61 (16.8) 61 (16.8) 30 (8.3) 15 (4.1) 283 (77.9) 50 (13.8) 30 (8.3) 171 (47.1) 45 (12.4) 29 (8.0) 36 (9.9) 47 (12.9) 35 (9.6) 24 (6.6) 308 (84.8) 31 (8.5)
Table 2 describes personal experiences of IPV by gender. More than half of the women and one-third of men had a positive IPV screen, meaning they had a history of having been a victim of IPV. In regard to perpetration, 21.0% (75/357) admitted to ever hitting, slapping, kicking, or doing anything else physically abusive to an intimate partner. Of those who admitted perpetration, 61/75 (81.3%) also had a positive IPV screen. Overall, 53.8% (192/357) had a positive IPV experience screen, having ever been a victim and/or ever perpetrated IPV. Table 3 shows the percentage of HCWs with positive attitudes regarding situations in which a man is justified in beating his wife/ woman. Overall, 29.9% (106/354) of HCWs had a positive attitude that a man is justified in beating his wife/woman in at least one of the given situations. Results of the multivariate logistic regression analysis can be found in Table 4.Those in the age groups 31–40 years (OR 2.3, 95% CI 1.1–4.6, p¼0.03) and 41–50 years (OR 2.3, 95% CI 1.2–4.7, p¼0.02) had increased odds of having a positive attitude towards justification for physical violence compared with those in the 18–30 years age group. Nursing staff also had increased odds (OR 4.3, 95% CI 1.4–13.1, p¼0.01) of having a positive attitude compared with physician staff, as did those with a history of having taken a training course in IPV (OR 3.0, 95% CI 1.1–8.3, p¼0.03). Being a victim of IPV (OR 1.3, 95% CI 0.7–2.2, p¼0.42) or admitting to having perpetrated IPV (OR 0.7, 95% CI 0.3–1.5, p¼0.38) was not associated with a positive attitude towards physical violence in the given scenarios. Attitudes toward the victim, screening, and outcomes are detailed according to gender in Table 5. Notably, 5.1% (18/355) agreed that if a person is hit by their partner they probably deserved it and 21.4% (75/350) agreed that dealing with IPV means interfering with the privacy of the family. In regard to attitude toward screening, 19.3% (67/348) thought it is of no use to screen for partner violence because facilities to aid victims are limited and 54.4% (196/360) agreed that most patients would feel offended if asked whether they were a victim of domestic violence. In regard to attitudes toward the outcomes, 68.3% (241/ 353) agreed that the police response to IPV complaints is usually inadequate and only 20.1% (70/349) agreed that the legal system adequately handles cases of IPV. Table 6 describes responses regarding HCW knowledge about IPV and perceived barriers to care. Regarding knowledge about IPV, only 33.2% (117/352) of HCWs agreed that they knew how to question patients about IPV and 44.8% (158/353) agreed they could readily recognize cases of domestic violence. Only 11.2% (39/347) agreed that they had sufficient training in the management of IPV. Only 6.1% (22/359) of the HCWs were able to correctly define an occupation order. None (0/50) of the physicians correctly defined an occupation order. With respect to the barriers to providing care to victims of IPV, 23.3% of respondents (81/348) agreed that they did not have enough time to discuss IPV with patients and 47.4% (166/350) that confidentiality is difficult to ensure when IPV is documented on the chart. Only 14.0% (48/344) agreed that there are adequate local training opportunities for staff.
Discussion Multiple studies in lower-income and higher-income countries have found that HCWs are personally affected by IPV. A study of
275
V. Mitchell et al.
Table 2. Healthcare workers’ personal experiences of intimate partner violencea Survey statement
Positive response n (%) Male
Have you ever been emotionally or physically abused by your partner or someone important to you? Within the last year have you been hit, slapped, kicked, or physically hurt by your partner? Have you ever hit, slapped, kicked or been physically abusive to an intimate partner? Positive IPV screenb Positive IPV experience screenc
Female
Not specified
13/45 (29) 149/293 (50.9) 12/18 (66.7)
Overall
174/356 (48.9)
4/45 (9)
24/294 (8.2)
4/18 (22.2)
32/357 (9.0)
7/45 (16)
63/294 (21.4)
5/18 (27.8)
75/357 (21.0)
15/45 (33) 151/294 (51.4) 12/18 (66.7) 16/45 (36) 163/294 (55.4) 13/18 (72.2)
178/357 (49.9) 192/357 (53.8)
a
Not all participants answered every question and/or gave complete demographic information. A positive response to one or both of the top two questions. c A positive response to any of the three questions. IPV: intimate partner violence. b
Table 3. Healthcare workers’ attitudes regarding justification of intimate partner violencea Positive attitude b n (%)
Survey statement
Male
A man is justified in beating his woman/wife when she argues with him A man is justified in beating his woman/wife when she spends too much time with friends A man is justified in beating his woman/wife when she is unfaithful Positive for at least one of the above scenarios a b
Not Specified
Overall
5/45 (11) 54/283 (19.1) 3/17 (17.6) 5/42 (12) 64/274 (23.4) 4/14 (28.6)
62/345 (18.0) 73/330 (22.1)
9/45 (20) 83/283 (29.3) 5/14 (35.7) 13/46 (28) 87/291 (29.9) 6/17 (35.3)
97/342 (28.4) 106/354 (29.9)
Not all participants answered every question and/or gave complete demographic information. Defined as answering ‘strongly agree’, ‘agree’, or ‘neutral’.
US medical students and faculty at a single institution found that 17% of females and 3% of males had ever experienced partner violence and a study of 1981 female nurses at three US hospitals and one geriatric center found that 25% had ever experienced IPV.19,20 Our study demonstrates that personal experiences with IPV are pervasive among HCWs at the main public teaching hospital in Guyana. Close to half (48.9%) of HCWs had experienced abuse in their lifetime with 9.0% experiencing ongoing abuse. Twenty-one percent admitted to perpetrating abuse. This was not completely unexpected as a study of South African rural community nurses found that 25 out of 35 (71.4%) female nurses had experienced abuse by a partner and six of eight (75%) male nurses had perpetrated abuse.6 At a tertiary care center in New Delhi, 65% of female nurses reported emotional violence and 43.3% physical violence.21 Our study reinforces the idea that although
276
Female
HCWs are tasked with treating IPV victims, they are often victims themselves. Despite the thought that HCWs would typically empathize with IPV victims, there is evidence that in high-prevalence areas, HCWs likely share prevailing cultural beliefs regarding IPV.4–6 Disturbingly, 29.9% of HCWs in this study felt that a man was justified in beating his wife in at least one of three named scenarios. This is not unique to Guyana and has been shown in other settings with a high prevalence of IPV.4,6,21,22 A study among Turkish physicians and nurses found that 69.0% of female and 84.7% of male HCWs accepted at least one offered scenario that would justify physical violence by a husband toward a wife.4 In a study of rural South African nurses, physical assault was universally accepted if there was infidelity on the part of the women.6 This was also seen in a study of female nurses in New Delhi, where
International Health
almost one-third felt infidelity was a justification for violence.21 Likewise 31.9% of nurses and nursing aides in Sa˜o Paulo, Brazil, who screened positive for IPV did not perceive their experiences
Table 4. Association of factors with a positive attitude toward justification of beating a woman/wife in one or more of the three scenarios Variable Gender Female Male Age (years) 18–30 31–40 41–50 .50 Position Physician staff Nursing staff Previous IPV training Ever an IPV victim Ever perpetrated IPV
OR
95% CI
p
Reference group 1.3
NA 0.5–3.5
NA NS
Reference group 2.3 2.3 1.5
NA 1.2–4.6 1.2–4.7 0.6–4.0
NA 0.03 0.02 NS
Reference group 4.3 3.0 1.3 0.7
NA 1.4–13.1 1.1–8.3 0.7–2.2 0.3–1.5
NA 0.01 0.03 NS NS
NA: not applicable; NS: non-significant; OR: odds ratio.
as abuse.22 Clearly, these attitudes must be addressed among HCWs in order for them to provide supportive, compassionate care to IPV victims. Multivariate analysis revealed that individuals in the age groups 31–40 years and 41–50 years had higher odds of accepting justification for physical violence compared to those aged 18–30 years. This finding by age group is not surprising, as changing cultural norms are likely to be reflected in differences within age groups. We did not find that those in the age group .50 years had significantly higher odds of accepting violence but it is likely that this because of the low numbers of workers in this age category. Nursing staff also had substantially higher odds of accepting physical violence compared to physician staff members on the multivariate analysis. It is unlikely that occupational experiences alone accounted for a difference of this magnitude. This finding was most likely attributable to uncontrolled confounders such as differences in education or socioeconomic status. It was surprising that having taken a previous training course on IPV was associated with positive attitudes towards justification of physical violence. There is no mandated IPV training in Guyana so it is unlikely that those with a history of pre-existing abusive behavior had been differently exposed to training programs. It is possible that this finding was a statistical anomaly, probably related to the small sample size of those having self-reported taken an IPV training course. It is, nevertheless, an example of why it is important to continue to evaluate the content and effectiveness of IPV training courses. Although most HCWs agreed with routine screening for IPV, almost 40% still did not agree with screening and 20% thought
Table 5. Healthcare workers’ attitudes toward intimate partner violence (IPV) victims, screening, and outcomesa Survey statement
Agree or strongly agree n (%) Male
Attitudes toward victims People can choose not to be victims. If a person is hit by their partner they probably deserve it Victims of abuse could leave the relationship if they wanted to Dealing with IPV means interfering with privacy of the family Attitudes toward screening We should routinely ask patients about IPV It is of no use to screen for partner violence since referral facilities to aid victims are limited Most patients would feel offended if asked whether they are victims of domestic violence Asking patients about domestic violence might lead to increased risk of violence after discharge home Attitudes toward outcome Health professionals can’t help IPV victims because they will return to the same social environment The police response to IPV complaints is usually inadequate a
Female
Not specified
Overall
32/48 (67) 217/291 (74.6) 14/16 (87.5) 2/47 (4) 14/292 (4.8) 2/16 (12.5) 27/46 (59) 206/293 (70.3) 11/16 (68.8) 10/47 (21) 61/288 (21.2) 4/15 (26.7)
263/355 (74.1) 18/355 (5.1) 244/355 (68.7) 75/350 (21.4)
30/48 (62) 181/287 (63.0) 6/47 (13) 59/286 (20.6)
7/15 (46.7) 2/15 (13.3)
218/350 (62.3) 67/348 (19.3
22/48 (46) 165/296 (55.7)
9/16 (56.3)
196/360 (54.4)
7/48 (15)
80/291 (27.5)
9/16 (56.3)
96/355 (27.0)
9/48 (19)
77/294 (26.2)
3/16 (18.8)
89/358 (24.9)
28/46 (61) 201/289 (69.6) 12/18 (66.7)
241/353 (68.3)
Not all participants answered every question and/or gave complete demographic information.
277
V. Mitchell et al.
Table 6. Healthcare workers’ knowledge of intimate partner violence (IPV) and perceived barriers to caring for IPV victimsa Survey statement
Agree or strongly agree n (%)
I know how to question patients about IPV I can readily recognize cases of domestic violence I feel I have sufficient training in the management of IPV I don’t have enough time to discuss IPV with patients There are adequate local IPV training opportunities for staff The legal system adequately handles cases of IPV Confidentiality is difficult to ensure when IPV is documented on the chart a
Female
Not specified
Overall
16/46 (35) 21/44 (48) 6/46 (13) 8/46 (17) 6/44 (14) 2/45 (4) 20/46 (43)
95/288 (33.0) 128/291 (44.0) 33/284 (11.6) 70/284 (24.6) 40/284 (14.1) 59/286 (20.6) 136/287 (47.4)
6/18 (33.3) 9/18 (50.0) 0/17 (0.0) 3/18 (16.7) 2/16 (12.5) 9/18 (50.0) 10/17 (58.8)
117/352 (33.2) 158/353 (44.8) 39/347 (11.2) 81/348 (23.3) 48/344 (14.0) 70/349 (20.1) 166/350 (47.4)
Not all participants answered every question and/or gave complete demographic information.
that screening was not helpful. This is especially concerning as previous studies in the USA have demonstrated that even when physicians have training in IPV screening, 50% still reported that they rarely or never screen their female patients.23 Despite data to the contrary, most HCWs thought that patients would feel offended if asked about IPV and that discussing IPV in the healthcare setting would lead to increased violence on discharge.24,25 Clearly, these are areas that IPV training programs need to address in order to increase knowledge and begin to change attitudes. Although many (44.8%) of the HCWs in this study felt they could readily recognize cases of IPV, they also felt that sufficient training was lacking. Knowledge of important local legislation was poor, as only 6% of workers could define an occupation order and no physicians could correctly define this. Barriers to care were significant and included inadequate police response and judicial support. Again, this is not unique to Guyana. A study of HCWs in Tanzania found that they perceived poor support in caring for victims of IPV and a study from Serbia also found a weak support network and lack of education as perceived barriers to providing care to IPV victims.5,26 In conclusion, this study conducted at the tertiary care center in the capital city of Guyana documents a high prevalence of IPV among HCWs. It also identifies some of the prevailing attitudes towards IPV victims, screening, and outcomes. This information is essential in designing relevant, effective training for HCWs and identifies a need to address IPV among the healthcare workforce.
Limitations Study participants filled out answers anonymously on paper and inserted their sealed questionnaire into a locked box. Participants were explicitly instructed to not answer any questions about which they felt uncomfortable. We found that almost 15% of the questionnaires had at least one unanswered question. The fact that many of the unanswered questions were demographic (e.g. age, job title, gender) suggests that some of the participants were concerned about confidentiality despite our attempts to ensure anonymity. Additionally, senior nurses were recruiting both junior and more senior HCWs to participate. Despite our safeguards, it is possible that some HCWs felt uncomfortable being
278
Male
approached by a more senior HCW. However, most respondents answered the questions regarding experiences with IPV, indicating an interest in supplying accurate information regarding this aspect of the study. Nevertheless, it is possible that our results are an underestimate of the true prevalence of IPV among HCWs because of the sensitive nature of the subject. Not all eligible employees participated in the study and it is possible that those who did not participate were different from those who were enrolled. Because of the anonymous nature of the survey we were not able to obtain demographic information regarding those who were not enrolled. Therefore, we could not determine if there were systematic differences between the populations. Our enrolment, however, was very high at over 85% of potentially eligible employees and it is unlikely that our results would have differed significantly if a higher proportion of employees had been enrolled. There are no universally accepted screening tools for determining the presence or absence of IPV.12 We chose to use a modified version of the Abuse Assessment Screen. While developed for use in pre-partum clinics, this screen has been used in varied populations, including in the emergency department setting, and has been found to have reasonable sensitivity and specificity.12,27 The instrument has not been specifically validated in Guyana, but the questions used are empirical and are likely to be valid in an English-speaking country such as Guyana. At the time of this study, there was no validated short screening tool to identify perpetrators of IPV, so a single question was used.28 Again, the empirical nature of this question is such that it is likely to be accurate in this setting, although it might not accurately identify those who were responding defensively to an event. A recent study describes a three-question scale to identify perpetrators or IPV, but this was not available at the time of our study.28 This study was conducted at the main tertiary-care referral center within Guyana. It is possible that our findings might not be generalizable to HCWs in smaller or more remote locations in Guyana. It is also possible that our findings might not be generalizable to other countries within South America or the Caribbean. However, the findings of this study, combined with others that have been done, all suggest that personal experiences with IPV are common among HCWs and that HCWs likely have attitudes in line with local cultural norms.
International Health
Authors’ contributions: VM participated in study design, data collection, onsite logistics and manuscript preparation. KPP participated in data analysis and interpretation and drafted the manuscript. SR participated in study design and data analysis and critically revised the manuscript. NPF participated in study design and data collection, onsite logistics, and revised the manuscript. SWW participated in study design, data analysis and interpretation, and critically revised the manuscript. All authors read and approved the final manuscript. KPP is the guarantor of the paper. Funding: The Guyana Ministry of Health in Georgetown, Guyana provided funding for logistical support. They were not involved in study design, data analysis or interpretation, or manuscript preparation. Competing interests: None declared Ethical approval: The Guyana Ministry of Health Institutional Review Board approved the study protocol. The Vanderbilt University Institutional Review Board reviewed the protocol and it was determined to not qualify as ‘human subject’ research (from the standpoint of Vanderbilt University) as no Vanderbilt personnel were involved in direct study procedures per 46.102(f)(2).
References 1 WHO multi-country study on women’s health and domestic violence against women: summary report of initial results on prevalence, health outcomes and women’s responses. Geneva: WHO; 2005. http:// www.who.int/gender/violence/who_multicountry_study/en/ [accessed 24 September 2012]. 2 Campbell JC. Health consequences of intimate partner violence. Lancet 2002;359:1331–6. 3 Tjaden P, Thoennes N. Extent, nature, and consequences of intimate partner violence: findings from the national violence against women survey. NCJ 181867. Washington DC: US Department of Justice, National Institute of Justice; 2000.
11 United Nations Statistics Division. Per capita GDP at current prices. http://data.un.org/Data.aspx?d=SNAAMA&f=grID%3A101%3BcurrID% 3AUSD%3BpcFlag%3A1#SNAAMA [accessed 22 September 2012]. 12 Rabin RF, Jennings JM, Campbell JC et al. Intimate partner violence screening tools. Am J Prev Med 2009;36:439–45. 13 Monitoring and Evaluation to Assess and Use Results Demographic and Health Surveys (Measure DHS). Profiling domestic violence: A multi-country study. 2004. http://measuredhs.com/publications/ publication-od31-other-documents.cfm [accessed 22 September 2012]. 14 Antai DE, Antai JB. Attitudes of women toward intimate partner violence: a study of rural women in Nigeria. Rural Remote Health 2008;8:996. 15 Bhandari M, Sprague S, Tornetta P et al. (Mis)perceptions about intimate partner violence in women presenting for orthopaedic care: a survey of Canadian orthopaedic surgeons. J Bone Joint Surg Am 2008;90:1590–7. 16 Roelens K, Verstraelen H, Van Egmond K et al. A knowledge, attitudes, and practice survey among obstetrician-gynaecologists on intimate partner violence in Flanders, Belgium. BMC Public Health 2006;6:238. 17 Gutmanis I, Beynon C, Tutty L et al. Factors influencing identification of and response to intimate partner violence: a survey of physicians and nurses. BMC Public Health 2007;7:12. 18 Help & Shelter. Domestic Violence Act of Guyana 1996. http://www. hands.org.gy/dv.html [accessed 12 September 2012]. 19 DeLahunta EA, Tulsky AA. Personal exposure of faculty and medical students to family violence. JAMA 1996;275:1903–6. 20 Bracken MI, Messing JT, Campbell JC et al. Intimate partner violence and abuse among female nurses and nursing personnel: prevalence and risk factors. Issues Ment Health Nurs 2010;31:137–48. 21 Sharma KK, Vatsa M. Domestic violence against nurses by their marital partners: a facility-based study at a tertiary care hospital. Indian J Community Med 2011;36:222–7.
4 Aksan HAD, Aksu F. The training needs of Turkish emergency department personnel regarding intimate partner violence. BMC Public Health 2007;7:350–9.
22 Oliveira AR, D’Oliviera AFPL. Gender-violence against the female nursing staff of a Brazilian hospital in Sao Paulo City. Rev Saude Publica 2008;42:868–76.
5 Djikanovic B, Celik H, Simic S et al. Health professionals’ perception of intimate partner violence against women in Serbia: Opportunities and barriers for response improvement. Patient Educ Couns 2010;80:88–93.
23 Williamson KJ, Coonrod DV, Bay RC et al. Screening for domestic violence: practice patterns, knowledge, and attitudes of physicians in Arizona. South Med J 2004;97:1049–54.
6 Kim J, Motsei M. ‘Women enjoy punishment’: attitudes and experiences of gender-based violence among PHC nurses in rural South Africa. Soc Sci Med 2002;54:1243–54.
24 Glass N, Dearwater S, Campbell J. Intimate partner violence screening and intervention: data from eleven Pennsylvania and California community hospital emergency departments. J Emerg Nurs 2001;27: 141–9.
7 US Agency for International Development. Gender assessment for USAID/Guyana. Arlington VA: DevTech Systems; 2003. http://pdf. usaid.gov/pdf_docs/PDACG104.pdf [accessed 24 September 2012].
25 Houry D, Kaslow NJ, Kemball RD et al. Does screening in the emergency department hurt or help victims of intimate partner violence? Ann Emerg Med 2008, 51:433–42.
8 Immigration and Refugee Board of Canada. Guyana: prevalence of domestic violence, availability of state protection, recourse and services available to victims. GUY102929.E. 2008. www.unhcr.org/ refworld/docid/492ac7c3c.html [accessed 24 September 2012].
26 Laisser RM, Lugina HI, Lindmark G et al. Striving to make a difference: health care worker experiences with intimate partner violence clients in Tanzania. Health Care Women Int 2009;30:64–78.
9 Parekh KP, Russ S, Amsalem D et al. Prevalence of intimate partner violence at a Guyanese emergency department. Int J Emerg Med 2012;5:23. 10 Ministry of Human Services and Social Security, Government of Guyana. National Policy on Domestic Violence 2008. www.sdnp.org. gy/16days/node/74 [accessed 25 September 2012].
27 Nelson HD, Bougatsos C, Blazina I. Screening women for intimate partner violence: a systematic review to update the US Preventive Services Task Force recommendation. Ann Int Med 2012;156:796–808. 28 Ernst AA, Weiss SJ, Morgan-Edwards S et al. Derivation and validation of a short emergency department screen tool for perpetrators of intimate partner violence: the perpetrator rapid scale (PERPS). J Emerg Med 2012;42:206–17.
279
Georgetown Public Hospital Corporation, New Market Street, 314132, Georgetown, Guyana; bDepartment of Emergency Medicine, Vanderbilt University, 703 Oxford House, Nashville, TN, 37232 USA *Corresponding author: Tel: +1 615 936 0087; Fax: +1 615 936 1316; E-mail: [email protected]
Received 26 August 2013; revised 17 October 2013; accepted 18 October 2013 Background: Intimate partner violence (IPV) is prevalent throughout the world and is a devastating public health problem. Healthcare workers (HCWs) are tasked with treating victims of IPV but may be victims themselves. Guyana is a lower-middle income country in South America. This study sought to determine the knowledge and attitudes of Guyanese HCWs and their perceived barriers to providing care in addition to determining the prevalence of IPV victimization and perpetration among HCWs. Methods: HCWs at the only tertiary care hospital in the Guyana completed an anonymous survey that comprised 30 questions relating to IPV. Results: The survey was completed by 87.5% of eligible HCWs. Of the respondents, 81.8% were female, 49.9% had ever experienced abuse and 21% admitted to perpetrating violence. Multivariate analysis found that the age groups 31–40 years (OR 2.3, 95% CI 1.1–4.6) and 41–50 years (OR 2.3, 95% CI 1.2–4.7) had higher odds of accepting justification for physical violence, and so did nursing staff (OR 4.3, 95% CI 1.4–13.1). Overall, 29.9% of HCWs accepted justification for physical violence in at least one of the named scenarios. Conclusion: This study demonstrates a high prevalence of IPV among HCWs and identifies prevailing attitudes regarding IPV. This knowledge is essential in developing effective, appropriate training programs and identifies a need to address IPV among the healthcare workforce. Keywords: Domestic violence, Spouse abuse, Women’s health, Attitude of health personnel, Caribbean, Guyana
Introduction Intimate partner violence (IPV) is prevalent throughout the world and has been recognized as a public health problem with shortand long-term health consequences.1,2 The healthcare system is an integral component in the societal management of IPV. Most victims of IPV access the healthcare system at some point and it is an important portal into the social service, counseling, and judicial systems.3 Although healthcare workers (HCWs) often treat victims of IPV, HCWs may themselves be victims and/or perpetrators of violence, especially in societies where IPV is highly endemic. Community and society factors influence attitudes toward IPV and HCWs often share the prevailing cultural attitudes regarding IPV. For example, a study of physicians and nurses at a large university hospital in Turkey found that 69% of females and 84.7% of males accepted at least one justification for physical violence against a partner.4 Similar results have been found among health professionals in Serbia and nurses in rural South Africa.5,6 Guyana is a lower-middle income country with cultural and economic ties to other English-speaking Caribbean countries. It is a founding member and host country of the Caribbean
Community Secretariat. In general, women are in a disadvantaged position in society with decreased access to resources, less decision-making power, and higher rates of poverty.7 One particularly prevalent issue is IPV and there are almost daily media stories regarding instances of IPV.7,8 A recent retrospective study found that 16% of acute traumatic injuries in women presenting to the emergency department at the main public hospital were a result of IPV.9 The high prevalence of IPV has led to the development of a National Policy on Domestic Violence and various entities, including non-governmental organizations and the Guyana Ministry of Health, have partnered to reduce the prevalence.10 The training of HCWs is considered an essential part of this response. However, there is a paucity of data on which to base current training models for HCWs in Guyana or similar countries. Most IPV educational programs have been designed in higher-income countries. It is largely unknown how these educational programs function in lower-income countries and to be most effective, they likely require modification to address cultural norms and prevailing local attitudes. In order to provide key information for the development of IPV training programs for HCWs in Guyana and similar countries, this
# The Author 2013. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
273
ORIGINAL ARTICLE
Int Health 2013; 5: 273–279 doi:10.1093/inthealth/iht030 Advance Access publication 13 November 2013
V. Mitchell et al.
study sought to determine the knowledge and attitudes of HCWs regarding IPV and their perceived barriers to providing care for patients experiencing IPV. We also sought to determine the prevalence of IPV victimization and perpetration among current HCWs at the main public hospital.
Materials and methods Setting Guyana is an English-speaking country on the North Atlantic coast of South America. Its population of about 750 000 is ethnically diverse, with most individuals being of East Asian, African, or Amerindian heritage. As of 2010, the per capita gross domestic product was US$2996.11 Georgetown is the capital city and the main population center for the country. Georgetown Public Hospital Corporation (GPHC) is a government-funded teaching facility and is Guyana’s only tertiary care medical center.
Study design A cross-sectional survey of HCWs at GPHC was conducted during September 2011.
Study procedures Seven GPHC senior nurses and a social worker were selected via the Quality Control Office at GPHC to administer the survey. All surveyors underwent training by a sociologist and were supervised by a staff physician to ensure there was no undue influence on staff to participate in the study and that anonymity was maintained. Inclusion criteria were: age 18 years or older, current healthcare worker with the position of nursing staff (registered nurse, midwife, or nursing assistant) or physician staff (consultant physician, junior physician, or intern). Those selected to administer the survey or otherwise involved in carrying out the study were not eligible for participation. All eligible HCWs were approached during the workday and invited to voluntarily participate in the study. HCWs were explicitly informed that they did not have to participate and that all answers to the survey questions were anonymous. If the HCW agreed to participate, written informed consent was obtained. Participants were told that they did not have to complete items that made them feel uncomfortable and that they could stop the survey at any time. All surveys were completed anonymously by the participants on paper in a private space and then placed in sealed envelopes in a locked ballot box. The surveyors did not directly ask any survey questions. All HCWs were offered a brochure with IPV resources. The Guyana Ministry of Health Institutional Review Board approved the study protocol.
Survey instrument The survey instrument consisted of 30 questions and was developed following a literature review. It was then modified by local experts to ensure cultural acceptability and appropriateness. Prior to implementation, the instrument was pilot tested for comprehensibility and acceptability with a group of Guyanese HCWs. Basic demographic data were collected from each subject, including age (18–30, 31–40, 41–50, or .50 years), gender, job
274
title, years of health care experience (,5, 6–10, 11–15, 16–20, .20 years), and self-reported history of participation in an IPV training course. The instrument included items relating to seven domains (personal experiences, attitudes toward the justification of violence, attitudes toward the victim, attitudes toward screening for IPV, attitudes toward the outcome, knowledge, and barriers). A series of statements were presented and participants rated agreement using a 5-point Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’ for all of the items, with the exception of three questions relating to personal experiences that allowed for a ‘yes’ or ‘no’ answer and one open-ended question as described below. To assess personal experiences with IPV, an affirmative response to the question ‘Have you ever been emotionally or physically abused by your partner or someone important to you?’ or to the question ‘Within the last year, have you been hit, slapped, kicked, or otherwise physically hurt by someone?’ was considered a positive IPV screen. These two questions were derived from the Abuse Assessment Screen, a validated and simple IPV screening tool.12 A single question modified from the Multi-country National Demographic and Health Surveys program, ‘Have you ever hit, slapped, kicked, or done anything else physically abusive to an intimate partner?’ was used to determine if participants ever perpetrated physical violence against an intimate partner.13 A positive IPV screen or an affirmative answer to the perpetration question was considered a positive IPV experience screen. Attitudes towards the justification of violence were assessed with three questions adopted for local context from questions used in the Multi-country Demographic and Health Surveys program.1,13 These questions ask for agreement with scenarios that might justify a man beating a woman and range from trivial (‘spends too much time with friends’) to more socially/ morally serious (‘is unfaithful’). Each subject was classified as having a firm negative (answering ‘strongly disagree’ or ‘disagree’ to each of the three questions) or a positive attitude (answering ‘strongly agree,’ ‘agree,’ or ‘neutral’ to any of the three questions) toward the scenarios.14 Attitudes toward the victim, attitudes toward screening, knowledge, and barriers were assessed with four questions each while attitudes toward outcome were assessed with two questions. These questions were modified from surveys designed to examine HCWs’ attitudes and knowledge regarding IPV.15–17 There was a single open-ended question asking, ‘Under the Domestic Violence Act, what is an occupation order?’ An occupation order is part of legislation designed to protect victims of partner abuse in Guyana.18 A magistrate can grant an occupation order so that an IPV victim may live in the home even if it belongs to the abuser. The abuser is not permitted to live in the same home as the victim. Answers were dichotomized as correct or incorrect. An answer was considered correct if any element of an occupation order (e.g. ‘victim can remain in the home,’ ‘abuser can’t live in the home’) was included in the response. This openended question was considered one of the four questions within the knowledge domain.
Data analysis Statistical analysis was done using Stata 10.1 software (StataCorp LP, College Station, TX, USA). Logistic regression was used in a multivariate analysis to determine covariates associated with a positive attitude when queried about the justification of a man beating his
International Health
wife in any of three named scenarios (she argues with him, she spends too much time with friends, she is unfaithful). All covariates were entered into the logistic regression model with listwise deletion used to exclude observations with missing data. Odds ratios with 95% confidence intervals are reported along with the p-value. A p-value of ,0.05 was considered statistically significant.
Results There were an estimated 460 eligible staff (375 nursing staff and 85 physician staff) employed at GPHC during the study period excluding those involved in carrying out the study. Forty-five (41 nurses and 4 physicians) were on leave during the study period, leaving a total of 415 potential participants. A total of 363 HCWs consented and participated in the survey, representing 87.5% of those estimated to be eligible and present during the study period. Some participants declined to answer every question. Table 1 describes the demographic findings. There were 297 females, 48 males, and 18 unspecified. There were 283 nurses, 50 physicians and 30 unspecified. Most (256/283, 90.5%) of those identifying themselves as nursing staff were female and 46.0% (23/50) of those identified as physicians were female. Only 24/363 (6.6%) reported they had previously taken an IPV course. Table 1. Demographic characteristics of healthcare workers (n¼363) participating in a survey conducted to explore knowledge and attitudes in relation to intimate partner violence Characteristic Gender Female Male Not specified Age (years) 18–30 31–40 41–50 .50 Not specified Position Nursing staff Physician staff Not specified Healthcare experience (years) ,5 5–10 11–15 16–20 .20 Not specified Previous IPV training Yes No Not specified IPV: intimate partner violence.
Number (%)
297 (81.8) 48 (13.2) 18 (5.0) 196 (54.0) 61 (16.8) 61 (16.8) 30 (8.3) 15 (4.1) 283 (77.9) 50 (13.8) 30 (8.3) 171 (47.1) 45 (12.4) 29 (8.0) 36 (9.9) 47 (12.9) 35 (9.6) 24 (6.6) 308 (84.8) 31 (8.5)
Table 2 describes personal experiences of IPV by gender. More than half of the women and one-third of men had a positive IPV screen, meaning they had a history of having been a victim of IPV. In regard to perpetration, 21.0% (75/357) admitted to ever hitting, slapping, kicking, or doing anything else physically abusive to an intimate partner. Of those who admitted perpetration, 61/75 (81.3%) also had a positive IPV screen. Overall, 53.8% (192/357) had a positive IPV experience screen, having ever been a victim and/or ever perpetrated IPV. Table 3 shows the percentage of HCWs with positive attitudes regarding situations in which a man is justified in beating his wife/ woman. Overall, 29.9% (106/354) of HCWs had a positive attitude that a man is justified in beating his wife/woman in at least one of the given situations. Results of the multivariate logistic regression analysis can be found in Table 4.Those in the age groups 31–40 years (OR 2.3, 95% CI 1.1–4.6, p¼0.03) and 41–50 years (OR 2.3, 95% CI 1.2–4.7, p¼0.02) had increased odds of having a positive attitude towards justification for physical violence compared with those in the 18–30 years age group. Nursing staff also had increased odds (OR 4.3, 95% CI 1.4–13.1, p¼0.01) of having a positive attitude compared with physician staff, as did those with a history of having taken a training course in IPV (OR 3.0, 95% CI 1.1–8.3, p¼0.03). Being a victim of IPV (OR 1.3, 95% CI 0.7–2.2, p¼0.42) or admitting to having perpetrated IPV (OR 0.7, 95% CI 0.3–1.5, p¼0.38) was not associated with a positive attitude towards physical violence in the given scenarios. Attitudes toward the victim, screening, and outcomes are detailed according to gender in Table 5. Notably, 5.1% (18/355) agreed that if a person is hit by their partner they probably deserved it and 21.4% (75/350) agreed that dealing with IPV means interfering with the privacy of the family. In regard to attitude toward screening, 19.3% (67/348) thought it is of no use to screen for partner violence because facilities to aid victims are limited and 54.4% (196/360) agreed that most patients would feel offended if asked whether they were a victim of domestic violence. In regard to attitudes toward the outcomes, 68.3% (241/ 353) agreed that the police response to IPV complaints is usually inadequate and only 20.1% (70/349) agreed that the legal system adequately handles cases of IPV. Table 6 describes responses regarding HCW knowledge about IPV and perceived barriers to care. Regarding knowledge about IPV, only 33.2% (117/352) of HCWs agreed that they knew how to question patients about IPV and 44.8% (158/353) agreed they could readily recognize cases of domestic violence. Only 11.2% (39/347) agreed that they had sufficient training in the management of IPV. Only 6.1% (22/359) of the HCWs were able to correctly define an occupation order. None (0/50) of the physicians correctly defined an occupation order. With respect to the barriers to providing care to victims of IPV, 23.3% of respondents (81/348) agreed that they did not have enough time to discuss IPV with patients and 47.4% (166/350) that confidentiality is difficult to ensure when IPV is documented on the chart. Only 14.0% (48/344) agreed that there are adequate local training opportunities for staff.
Discussion Multiple studies in lower-income and higher-income countries have found that HCWs are personally affected by IPV. A study of
275
V. Mitchell et al.
Table 2. Healthcare workers’ personal experiences of intimate partner violencea Survey statement
Positive response n (%) Male
Have you ever been emotionally or physically abused by your partner or someone important to you? Within the last year have you been hit, slapped, kicked, or physically hurt by your partner? Have you ever hit, slapped, kicked or been physically abusive to an intimate partner? Positive IPV screenb Positive IPV experience screenc
Female
Not specified
13/45 (29) 149/293 (50.9) 12/18 (66.7)
Overall
174/356 (48.9)
4/45 (9)
24/294 (8.2)
4/18 (22.2)
32/357 (9.0)
7/45 (16)
63/294 (21.4)
5/18 (27.8)
75/357 (21.0)
15/45 (33) 151/294 (51.4) 12/18 (66.7) 16/45 (36) 163/294 (55.4) 13/18 (72.2)
178/357 (49.9) 192/357 (53.8)
a
Not all participants answered every question and/or gave complete demographic information. A positive response to one or both of the top two questions. c A positive response to any of the three questions. IPV: intimate partner violence. b
Table 3. Healthcare workers’ attitudes regarding justification of intimate partner violencea Positive attitude b n (%)
Survey statement
Male
A man is justified in beating his woman/wife when she argues with him A man is justified in beating his woman/wife when she spends too much time with friends A man is justified in beating his woman/wife when she is unfaithful Positive for at least one of the above scenarios a b
Not Specified
Overall
5/45 (11) 54/283 (19.1) 3/17 (17.6) 5/42 (12) 64/274 (23.4) 4/14 (28.6)
62/345 (18.0) 73/330 (22.1)
9/45 (20) 83/283 (29.3) 5/14 (35.7) 13/46 (28) 87/291 (29.9) 6/17 (35.3)
97/342 (28.4) 106/354 (29.9)
Not all participants answered every question and/or gave complete demographic information. Defined as answering ‘strongly agree’, ‘agree’, or ‘neutral’.
US medical students and faculty at a single institution found that 17% of females and 3% of males had ever experienced partner violence and a study of 1981 female nurses at three US hospitals and one geriatric center found that 25% had ever experienced IPV.19,20 Our study demonstrates that personal experiences with IPV are pervasive among HCWs at the main public teaching hospital in Guyana. Close to half (48.9%) of HCWs had experienced abuse in their lifetime with 9.0% experiencing ongoing abuse. Twenty-one percent admitted to perpetrating abuse. This was not completely unexpected as a study of South African rural community nurses found that 25 out of 35 (71.4%) female nurses had experienced abuse by a partner and six of eight (75%) male nurses had perpetrated abuse.6 At a tertiary care center in New Delhi, 65% of female nurses reported emotional violence and 43.3% physical violence.21 Our study reinforces the idea that although
276
Female
HCWs are tasked with treating IPV victims, they are often victims themselves. Despite the thought that HCWs would typically empathize with IPV victims, there is evidence that in high-prevalence areas, HCWs likely share prevailing cultural beliefs regarding IPV.4–6 Disturbingly, 29.9% of HCWs in this study felt that a man was justified in beating his wife in at least one of three named scenarios. This is not unique to Guyana and has been shown in other settings with a high prevalence of IPV.4,6,21,22 A study among Turkish physicians and nurses found that 69.0% of female and 84.7% of male HCWs accepted at least one offered scenario that would justify physical violence by a husband toward a wife.4 In a study of rural South African nurses, physical assault was universally accepted if there was infidelity on the part of the women.6 This was also seen in a study of female nurses in New Delhi, where
International Health
almost one-third felt infidelity was a justification for violence.21 Likewise 31.9% of nurses and nursing aides in Sa˜o Paulo, Brazil, who screened positive for IPV did not perceive their experiences
Table 4. Association of factors with a positive attitude toward justification of beating a woman/wife in one or more of the three scenarios Variable Gender Female Male Age (years) 18–30 31–40 41–50 .50 Position Physician staff Nursing staff Previous IPV training Ever an IPV victim Ever perpetrated IPV
OR
95% CI
p
Reference group 1.3
NA 0.5–3.5
NA NS
Reference group 2.3 2.3 1.5
NA 1.2–4.6 1.2–4.7 0.6–4.0
NA 0.03 0.02 NS
Reference group 4.3 3.0 1.3 0.7
NA 1.4–13.1 1.1–8.3 0.7–2.2 0.3–1.5
NA 0.01 0.03 NS NS
NA: not applicable; NS: non-significant; OR: odds ratio.
as abuse.22 Clearly, these attitudes must be addressed among HCWs in order for them to provide supportive, compassionate care to IPV victims. Multivariate analysis revealed that individuals in the age groups 31–40 years and 41–50 years had higher odds of accepting justification for physical violence compared to those aged 18–30 years. This finding by age group is not surprising, as changing cultural norms are likely to be reflected in differences within age groups. We did not find that those in the age group .50 years had significantly higher odds of accepting violence but it is likely that this because of the low numbers of workers in this age category. Nursing staff also had substantially higher odds of accepting physical violence compared to physician staff members on the multivariate analysis. It is unlikely that occupational experiences alone accounted for a difference of this magnitude. This finding was most likely attributable to uncontrolled confounders such as differences in education or socioeconomic status. It was surprising that having taken a previous training course on IPV was associated with positive attitudes towards justification of physical violence. There is no mandated IPV training in Guyana so it is unlikely that those with a history of pre-existing abusive behavior had been differently exposed to training programs. It is possible that this finding was a statistical anomaly, probably related to the small sample size of those having self-reported taken an IPV training course. It is, nevertheless, an example of why it is important to continue to evaluate the content and effectiveness of IPV training courses. Although most HCWs agreed with routine screening for IPV, almost 40% still did not agree with screening and 20% thought
Table 5. Healthcare workers’ attitudes toward intimate partner violence (IPV) victims, screening, and outcomesa Survey statement
Agree or strongly agree n (%) Male
Attitudes toward victims People can choose not to be victims. If a person is hit by their partner they probably deserve it Victims of abuse could leave the relationship if they wanted to Dealing with IPV means interfering with privacy of the family Attitudes toward screening We should routinely ask patients about IPV It is of no use to screen for partner violence since referral facilities to aid victims are limited Most patients would feel offended if asked whether they are victims of domestic violence Asking patients about domestic violence might lead to increased risk of violence after discharge home Attitudes toward outcome Health professionals can’t help IPV victims because they will return to the same social environment The police response to IPV complaints is usually inadequate a
Female
Not specified
Overall
32/48 (67) 217/291 (74.6) 14/16 (87.5) 2/47 (4) 14/292 (4.8) 2/16 (12.5) 27/46 (59) 206/293 (70.3) 11/16 (68.8) 10/47 (21) 61/288 (21.2) 4/15 (26.7)
263/355 (74.1) 18/355 (5.1) 244/355 (68.7) 75/350 (21.4)
30/48 (62) 181/287 (63.0) 6/47 (13) 59/286 (20.6)
7/15 (46.7) 2/15 (13.3)
218/350 (62.3) 67/348 (19.3
22/48 (46) 165/296 (55.7)
9/16 (56.3)
196/360 (54.4)
7/48 (15)
80/291 (27.5)
9/16 (56.3)
96/355 (27.0)
9/48 (19)
77/294 (26.2)
3/16 (18.8)
89/358 (24.9)
28/46 (61) 201/289 (69.6) 12/18 (66.7)
241/353 (68.3)
Not all participants answered every question and/or gave complete demographic information.
277
V. Mitchell et al.
Table 6. Healthcare workers’ knowledge of intimate partner violence (IPV) and perceived barriers to caring for IPV victimsa Survey statement
Agree or strongly agree n (%)
I know how to question patients about IPV I can readily recognize cases of domestic violence I feel I have sufficient training in the management of IPV I don’t have enough time to discuss IPV with patients There are adequate local IPV training opportunities for staff The legal system adequately handles cases of IPV Confidentiality is difficult to ensure when IPV is documented on the chart a
Female
Not specified
Overall
16/46 (35) 21/44 (48) 6/46 (13) 8/46 (17) 6/44 (14) 2/45 (4) 20/46 (43)
95/288 (33.0) 128/291 (44.0) 33/284 (11.6) 70/284 (24.6) 40/284 (14.1) 59/286 (20.6) 136/287 (47.4)
6/18 (33.3) 9/18 (50.0) 0/17 (0.0) 3/18 (16.7) 2/16 (12.5) 9/18 (50.0) 10/17 (58.8)
117/352 (33.2) 158/353 (44.8) 39/347 (11.2) 81/348 (23.3) 48/344 (14.0) 70/349 (20.1) 166/350 (47.4)
Not all participants answered every question and/or gave complete demographic information.
that screening was not helpful. This is especially concerning as previous studies in the USA have demonstrated that even when physicians have training in IPV screening, 50% still reported that they rarely or never screen their female patients.23 Despite data to the contrary, most HCWs thought that patients would feel offended if asked about IPV and that discussing IPV in the healthcare setting would lead to increased violence on discharge.24,25 Clearly, these are areas that IPV training programs need to address in order to increase knowledge and begin to change attitudes. Although many (44.8%) of the HCWs in this study felt they could readily recognize cases of IPV, they also felt that sufficient training was lacking. Knowledge of important local legislation was poor, as only 6% of workers could define an occupation order and no physicians could correctly define this. Barriers to care were significant and included inadequate police response and judicial support. Again, this is not unique to Guyana. A study of HCWs in Tanzania found that they perceived poor support in caring for victims of IPV and a study from Serbia also found a weak support network and lack of education as perceived barriers to providing care to IPV victims.5,26 In conclusion, this study conducted at the tertiary care center in the capital city of Guyana documents a high prevalence of IPV among HCWs. It also identifies some of the prevailing attitudes towards IPV victims, screening, and outcomes. This information is essential in designing relevant, effective training for HCWs and identifies a need to address IPV among the healthcare workforce.
Limitations Study participants filled out answers anonymously on paper and inserted their sealed questionnaire into a locked box. Participants were explicitly instructed to not answer any questions about which they felt uncomfortable. We found that almost 15% of the questionnaires had at least one unanswered question. The fact that many of the unanswered questions were demographic (e.g. age, job title, gender) suggests that some of the participants were concerned about confidentiality despite our attempts to ensure anonymity. Additionally, senior nurses were recruiting both junior and more senior HCWs to participate. Despite our safeguards, it is possible that some HCWs felt uncomfortable being
278
Male
approached by a more senior HCW. However, most respondents answered the questions regarding experiences with IPV, indicating an interest in supplying accurate information regarding this aspect of the study. Nevertheless, it is possible that our results are an underestimate of the true prevalence of IPV among HCWs because of the sensitive nature of the subject. Not all eligible employees participated in the study and it is possible that those who did not participate were different from those who were enrolled. Because of the anonymous nature of the survey we were not able to obtain demographic information regarding those who were not enrolled. Therefore, we could not determine if there were systematic differences between the populations. Our enrolment, however, was very high at over 85% of potentially eligible employees and it is unlikely that our results would have differed significantly if a higher proportion of employees had been enrolled. There are no universally accepted screening tools for determining the presence or absence of IPV.12 We chose to use a modified version of the Abuse Assessment Screen. While developed for use in pre-partum clinics, this screen has been used in varied populations, including in the emergency department setting, and has been found to have reasonable sensitivity and specificity.12,27 The instrument has not been specifically validated in Guyana, but the questions used are empirical and are likely to be valid in an English-speaking country such as Guyana. At the time of this study, there was no validated short screening tool to identify perpetrators of IPV, so a single question was used.28 Again, the empirical nature of this question is such that it is likely to be accurate in this setting, although it might not accurately identify those who were responding defensively to an event. A recent study describes a three-question scale to identify perpetrators or IPV, but this was not available at the time of our study.28 This study was conducted at the main tertiary-care referral center within Guyana. It is possible that our findings might not be generalizable to HCWs in smaller or more remote locations in Guyana. It is also possible that our findings might not be generalizable to other countries within South America or the Caribbean. However, the findings of this study, combined with others that have been done, all suggest that personal experiences with IPV are common among HCWs and that HCWs likely have attitudes in line with local cultural norms.
International Health
Authors’ contributions: VM participated in study design, data collection, onsite logistics and manuscript preparation. KPP participated in data analysis and interpretation and drafted the manuscript. SR participated in study design and data analysis and critically revised the manuscript. NPF participated in study design and data collection, onsite logistics, and revised the manuscript. SWW participated in study design, data analysis and interpretation, and critically revised the manuscript. All authors read and approved the final manuscript. KPP is the guarantor of the paper. Funding: The Guyana Ministry of Health in Georgetown, Guyana provided funding for logistical support. They were not involved in study design, data analysis or interpretation, or manuscript preparation. Competing interests: None declared Ethical approval: The Guyana Ministry of Health Institutional Review Board approved the study protocol. The Vanderbilt University Institutional Review Board reviewed the protocol and it was determined to not qualify as ‘human subject’ research (from the standpoint of Vanderbilt University) as no Vanderbilt personnel were involved in direct study procedures per 46.102(f)(2).
References 1 WHO multi-country study on women’s health and domestic violence against women: summary report of initial results on prevalence, health outcomes and women’s responses. Geneva: WHO; 2005. http:// www.who.int/gender/violence/who_multicountry_study/en/ [accessed 24 September 2012]. 2 Campbell JC. Health consequences of intimate partner violence. Lancet 2002;359:1331–6. 3 Tjaden P, Thoennes N. Extent, nature, and consequences of intimate partner violence: findings from the national violence against women survey. NCJ 181867. Washington DC: US Department of Justice, National Institute of Justice; 2000.
11 United Nations Statistics Division. Per capita GDP at current prices. http://data.un.org/Data.aspx?d=SNAAMA&f=grID%3A101%3BcurrID% 3AUSD%3BpcFlag%3A1#SNAAMA [accessed 22 September 2012]. 12 Rabin RF, Jennings JM, Campbell JC et al. Intimate partner violence screening tools. Am J Prev Med 2009;36:439–45. 13 Monitoring and Evaluation to Assess and Use Results Demographic and Health Surveys (Measure DHS). Profiling domestic violence: A multi-country study. 2004. http://measuredhs.com/publications/ publication-od31-other-documents.cfm [accessed 22 September 2012]. 14 Antai DE, Antai JB. Attitudes of women toward intimate partner violence: a study of rural women in Nigeria. Rural Remote Health 2008;8:996. 15 Bhandari M, Sprague S, Tornetta P et al. (Mis)perceptions about intimate partner violence in women presenting for orthopaedic care: a survey of Canadian orthopaedic surgeons. J Bone Joint Surg Am 2008;90:1590–7. 16 Roelens K, Verstraelen H, Van Egmond K et al. A knowledge, attitudes, and practice survey among obstetrician-gynaecologists on intimate partner violence in Flanders, Belgium. BMC Public Health 2006;6:238. 17 Gutmanis I, Beynon C, Tutty L et al. Factors influencing identification of and response to intimate partner violence: a survey of physicians and nurses. BMC Public Health 2007;7:12. 18 Help & Shelter. Domestic Violence Act of Guyana 1996. http://www. hands.org.gy/dv.html [accessed 12 September 2012]. 19 DeLahunta EA, Tulsky AA. Personal exposure of faculty and medical students to family violence. JAMA 1996;275:1903–6. 20 Bracken MI, Messing JT, Campbell JC et al. Intimate partner violence and abuse among female nurses and nursing personnel: prevalence and risk factors. Issues Ment Health Nurs 2010;31:137–48. 21 Sharma KK, Vatsa M. Domestic violence against nurses by their marital partners: a facility-based study at a tertiary care hospital. Indian J Community Med 2011;36:222–7.
4 Aksan HAD, Aksu F. The training needs of Turkish emergency department personnel regarding intimate partner violence. BMC Public Health 2007;7:350–9.
22 Oliveira AR, D’Oliviera AFPL. Gender-violence against the female nursing staff of a Brazilian hospital in Sao Paulo City. Rev Saude Publica 2008;42:868–76.
5 Djikanovic B, Celik H, Simic S et al. Health professionals’ perception of intimate partner violence against women in Serbia: Opportunities and barriers for response improvement. Patient Educ Couns 2010;80:88–93.
23 Williamson KJ, Coonrod DV, Bay RC et al. Screening for domestic violence: practice patterns, knowledge, and attitudes of physicians in Arizona. South Med J 2004;97:1049–54.
6 Kim J, Motsei M. ‘Women enjoy punishment’: attitudes and experiences of gender-based violence among PHC nurses in rural South Africa. Soc Sci Med 2002;54:1243–54.
24 Glass N, Dearwater S, Campbell J. Intimate partner violence screening and intervention: data from eleven Pennsylvania and California community hospital emergency departments. J Emerg Nurs 2001;27: 141–9.
7 US Agency for International Development. Gender assessment for USAID/Guyana. Arlington VA: DevTech Systems; 2003. http://pdf. usaid.gov/pdf_docs/PDACG104.pdf [accessed 24 September 2012].
25 Houry D, Kaslow NJ, Kemball RD et al. Does screening in the emergency department hurt or help victims of intimate partner violence? Ann Emerg Med 2008, 51:433–42.
8 Immigration and Refugee Board of Canada. Guyana: prevalence of domestic violence, availability of state protection, recourse and services available to victims. GUY102929.E. 2008. www.unhcr.org/ refworld/docid/492ac7c3c.html [accessed 24 September 2012].
26 Laisser RM, Lugina HI, Lindmark G et al. Striving to make a difference: health care worker experiences with intimate partner violence clients in Tanzania. Health Care Women Int 2009;30:64–78.
9 Parekh KP, Russ S, Amsalem D et al. Prevalence of intimate partner violence at a Guyanese emergency department. Int J Emerg Med 2012;5:23. 10 Ministry of Human Services and Social Security, Government of Guyana. National Policy on Domestic Violence 2008. www.sdnp.org. gy/16days/node/74 [accessed 25 September 2012].
27 Nelson HD, Bougatsos C, Blazina I. Screening women for intimate partner violence: a systematic review to update the US Preventive Services Task Force recommendation. Ann Int Med 2012;156:796–808. 28 Ernst AA, Weiss SJ, Morgan-Edwards S et al. Derivation and validation of a short emergency department screen tool for perpetrators of intimate partner violence: the perpetrator rapid scale (PERPS). J Emerg Med 2012;42:206–17.
279
