Family Practice Advance Access originally published online on February 3, 2006
Family Practice 2006 23(3):273-278; doi:10.1093/fampra/cmi119
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The response of general practitioners to the threat of violence in their practices: results from a qualitative study
a Discipline of General Practice, School of Medical Practice and Population Health, Faculty of Health, University of Newcastle Callaghan, NSW 2308
b Centre for Clinical Epidemiology and Biostatistics, School of Medical Practice and Population Health, Faculty of Health, University of Newcastle Level 3 DMB, Royal Newcastle Hospital, Newcastle 2300
c School of Social Sciences, Southern Cross University Lismore Campus, PO Box 157, Lismore 2480, Australia
Correspondence to Dr P Magin, Email: parker.magin{at}newcastle.edu.au
Received 20 June 2005; Accepted 28 December 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
Background. Violence directed towards GPs has been recognized as a significant problem in the UK. In Australian urban general practice, no study has previously examined this topic.
Objective. The objective of this study was to investigate the responses of Australian urban GPs to experiences of violence and to perceptions of risk of violence.
Methods. Design: A qualitative study of data collected from two sources—focus group discussions and qualitative questionnaire responses. Focus group discussions were audiotaped and transcribed. Questionnaires offered the opportunity for respondents to make qualitative comments. The focus group transcripts and qualitative questionnaire responses were coded independently by members of the research team and subjected to thematic analysis. Setting: Three urban Divisions of General Practice in New South Wales, Australia. Subjects: Focus groups were conducted with male and female GPs comprising a range of ages, socio-economic practice catchments and practice structures. Questionnaires were distributed to all GPs in the three divisions.
Results. The GPs in this study perceived themselves as being at significant risk of occupational violence. Despite responses to violence being largely ad hoc and uncoordinated, a coherent schema of GPs' responses to the threat of violence is apparent in the data. This has been characterized as encompassing primary, secondary and tertiary responses, and reflects a continuum of proactive to reactive responses.
Conclusion. The findings will have implications for further research and for policy in the area.
Keywords. Family practice, violence.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
Violence directed towards GPs and their staff is an area which has been recognized as being a significant occupational health issue.1 The British National Health Service has instituted a policy of ‘zero tolerance’ to violence by patients towards medical and nursing staff.2
Quantitative research, mainly questionnaire studies from the UK, have established a relatively high prevalence of violence in the work experiences of GPs.3–6 Qualitative research has the potential to further elucidate the effect of work-based violence on GPs and their medical practice. Though qualitative research in this area is sparse, it has been demonstrated that British GPs display a complex set of responses to experiences of violence in their practice and these often do not conform to ‘zero tolerance’ policies.3
In Britain, GPs practising in inner city and urban estates were at significantly higher risk of violence than those in rural areas.3 The present study aims to investigate the responses of Australian urban GPs to experiences of violence and to perceptions of risk of violence.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
Data collection consisted of focus group interviews and qualitative questionnaire responses. Four focus group interviews with a total of 18 GP participants and a total duration of over 9.5 h were conducted. After recruiting for prospective participants through three urban Divisions of General Practice (Divisions are geographically-based organizational structures in Australian general practice), we used purposive sampling to include GPs from a range of urban areas and with a range of characteristics. We sampled male and female GPs; doctors established and those relatively new to general practice; and GPs practising within different socio-economic patient catchments. The three participating Divisions of General Practice encompassed capital city and regional city urban areas.
The focus groups followed a theme list drafted by the research team with themes developed from both previous literature and the study aim. However, the discussions were informant-led as much as possible and as data collection progressed, additional items were added as new areas of interest emerged. At the outset of each focus group there was consideration and discussion of a recognized definition of occupational violence3,5 (incorporating physical violence and non-physical forms of violence such as verbal abuse or threat). Each focus group was conducted by two researchers—one with a general practice and one with a social science background.
Focus group discussions were recorded to audiotape with participants' consent and then transcribed to word file. Data analysis was cumulative and concurrent throughout the data collection period. The research team engaged in a systematic search for alternative themes and rival explanations of the data. All researchers were engaged in separately coding and interpreting transcripts. The research team then met to discuss rival explanations. These were then assessed against the weight of evidence looking for patterns and conclusions fitting the preponderance of data. This process concluded when all researchers were happy with the chosen interpretations and conclusions. Analysis was also strengthened by searching for negative cases in code and theme development. No substantive negative cases were found.
Further qualitative data was obtained by offering GPs an opportunity, during completion of a written questionnaire on violence in general practice, to provide further qualitative comments regarding their experiences and perceptions of violence. The questionnaires were sent to all GPs practising in three New South Wales urban divisions of general practice. Of the 528 subjects who completed the survey (49% of the sample), 154 also offered qualitative responses. These responses were coded and analysed in the same way as the focus group transcripts. Compared with the 374 questionnaire respondents who didn't provide written qualitative responses, those who did were comparable on demographic measures except that they were older (54 years versus 50 years, P < 0.01). In terms of experiences of eight types of violence surveyed, there were no significant differences except that the providers of qualitative responses were less likely to have experienced one type of violence: ‘threats’ (P = 0.04).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
The GPs in this study described a diverse range of (often idiosyncratic) strategies for dealing with violence in their clinical practice. Despite the heterogeneous and often ad hoc nature of these measures, they could be seen as conforming to a schema of primary, secondary and tertiary strategies (Table 1). The strategies could be further conceptualized along a trajectory of proactive and reactive responses to violence.
|
Primary strategies
‘Primary’ strategies were measures designed to avoid the occurrence of conflict, aggression and violence. These strategies are identified here as consisting primarily of either a restriction of practice or discouragement of violence.
Restriction of practice. Limitation or restriction of practice was the most prominent strategy cited by the GPs in this study. GPs confined their clinical practice to situations or patients with which they felt safest. This could be seen to operate at ‘population’ and ‘individual’ levels.
Population level measures involved a geographic, temporal or demographic basis. The most prominent geographic factor was seen to be surgery versus home or domiciliary visits. As the GPs in the focus groups perceived home visits to be intrinsically the more dangerous location for consultations, a risk reduction strategy for some of the participants was to eschew home visits:
‘The policy at my practice is to provide home visits in exceptional circumstances only (e.g. Elderly people who cannot get to the surgery due to their generally ill health) [...] our policy reflects our concerns about the possibility of violence being greater if we provided clinical care away from the surgery.’ (GP128, male.)If domiciliary visits were to be performed, a further geographic restriction of practice was to avoid home visits in certain lower socio-economic areas perceived to be more dangerous. Housing Commission estates were seen to be particularly high risk.
‘I mean, I guess if you felt that something was a very high risk, you'd just have to use your common sense and ah, refuse to actually go to that location and try to figure out a way around that.’ (GP17, Focus group 4, male.)The temporal factor in practice restriction was predicated on the perception that after-hours consultations (in the surgery and, especially, on domiciliary visits) were higher risk than in-hours consultations. As a result many of the participating GPs had withdrawn from or severely limited their after-hours work:
‘I worked in a housing commission area and um, there were a lot of very frightening places that I went to visit, even in the day time. I wasn't going there in the night. So, I've set the boundaries a long way back.’ (GP12, Focus group 3, female.)A further means of restricting practice (categorized here as being on ‘demographic’ grounds) was to avoid certain categories of patients. A consistent theme in demographic restriction of practice was the perceived dangerousness of drug-seeking patients (patients seeking opiates or benzodiazepines):‘I will never do after hours calls because of the very real threat of violence. I have deliberately sought work in a very safe environment so as to minimize anxiety.’ (GP127, male.)
‘Um, with the practice that I'm in, we're very active discouraging people who are looking for drugs and I think that has decreased a hell of a lot the potential violence. You just don't seem to face a lot of these issues that we're talking about.’ (GP8, Focus group 2, female.)An effective method of discouraging attendance of particular demographic groups (deemed to be at potentially higher risk of violence) was felt to be financial disincentive, as can be seen by these responses:
‘Our practice too has been very strongly discouraging druggies, basically by putting the price up. Yeh, very simple and since that's happened, we've had much less aggressive behaviour.’ (GP10, Focus group 2, female.)But the obverse of this strategy was recognized to be a shifting of risk—an increased risk of violence at other, bulk-billing or mixed-billing practices:
‘Violence is on the increase as more practices start charging all patients a full consultation fee. This deters addicts but [...] practices that offer mixed billing are becoming targets for these not nice people.’ (GP57, female.)
A further means of restricting practice that could be seen to be essentially ‘demographic’ was to ‘close the books’. By no longer accepting new patients, practitioners could restrict access to their practice by unknown patients who, they felt, might carry a higher risk of violence than that of their existing practice profile. Even in practices which, had not closed their books, home visits to unfamiliar patients were generally avoided:
‘Having limited my practice to people I know or people who are referred by people I know, um, it puts me in a good position but I did it purposely in order to protect myself.’ (GP3, Focus group 1, female.)‘Now I would know the patient before doing a house call, so I feel pretty safe.’ (GP48, female.)
Individual level strategies of ‘vetting’ individual patients for potential dangerousness and ‘blacklisting’, ‘blackballing’ or banning of patients who had previously transgressed the practice's boundaries of violence or aggression were common. Exclusion was on the basis of appearance, familiarity to the doctor concerned and past history of violence (or of an individual's medical or psychiatric history that would indicate a higher risk of violence).
‘[S]he [the secretary] actually reads my mind and she knows exactly what I want in terms of picking people to see, we tend to agree almost 100%.’ (GP3, Focus group 1, male.)‘We also have a list of people who under no circumstances will they be given appointments and no appointments will be given to their relatives.’ (GP57, female.)
Discouraging violence. A range of strategies could be seen as discouraging violence in potentially violent situations. These included, avoiding practitioner isolation; companions and destination documentation; and health professional and police support.
Avoiding Practitioner Isolation: The presence of other staff, patients or bystanders was seen as a discouraging violence and being a potential support in the event of violence. Respondents often structured their consulting to incorporate this:
‘Our practice is very busy and there is always a lot of people around which inhibits some violent and aggressive behaviours.’ (GP88, female.)
Companions and Destination Documentation: Outside the surgery, rather than attend potentially dangerous situations alone, the presence of other professionals or of a companion was seen as discouraging violence and providing support. A strategy employed by several of the female GPs, when on an after-hours call, was to be accompanied (usually by a spouse) on call-outs. This strategy was formalized in some after-hours arrangements by having a security guard accompany the doctor on calls:
‘And even then if it's in the middle of the night, my husband will come with me. He won't let me go and he won't come inside the house but he'll just say if you're not out in 5 minutes, I'm coming in.’ (GP5, Focus group 2, female.)‘There is a policy to call security guard for home visits...I believe it is a brilliant policy and a brilliant response to concerns raised over this issue. I would not work in the service without the security measure.’ (GP94, female.)
Health Professional and Police Support: If a call-out was felt to be potentially dangerous several of the focus group participants reported that they would enlist professional ‘back up’ (e.g. the community mental health team) or the police, rather than informal supports, to accompany them on the call.
Secondary strategies
A further level of response could be seen as being ‘secondary’ measures—strategies to prevent aggression or lesser levels of violence escalating into more serious, physical violence. These strategies were further delineated into the categories of (1) personal; and (2) environmental measures.
Personal measures. Personal measures to prevent the escalation of conflict or aggression, refers here to GP's practice modification, behaviours or interactions with patients. Included here are procedural measures such as the doctor attending promptly to patients displaying signs of agitation or aggression in the waiting room.
There was agreement among the participants that the key strategy lies in the use of interpersonal skills and negotiation techniques to prevent aggression escalating and to defuse tensions:
‘I'm going to give them more time. Straight away I'm going to do more listening. Um, try and let them have the floor for a little while and I'm also going to try and keep calm, as calm as possible myself.’ (GP18, Focus group 4, male.)
Another strategy that was felt by the GPs to be less satisfactory, but sometimes unavoidable was acceding to patient demands in the face of aggression or threats, or implied threats, of violence. This was usually, though not exclusively, related to demands for drugs of addiction:
‘A male 6'2 tall, 30 yrs old was asking for a prescription for Serepax. I was alone at night doing a locum. The patient said that if he didn't get his Serepax he becomes very aggressive. I said ‘How many would you like?’ I reasoned that the number one priority was to care for myself.’ (GP105, male.)
Physical retreat from the situation was seen as a means of defusing an episode of escalating aggression and an obvious option in such a situation, especially for female practitioners:
‘I guess as a female my tendency is to try and get out of somewhere where I would feel threatened, never to stand and confront so I would remove myself.’ (GP11, Focus group 3, female.)
Environmental measures. Environmental measures refer to modifications in the built environment of general practice. Structural measures within the GP consulting room, receptionist counter and waiting room were seen as the essential elements in preventing access of potentially violent patients to doctors and staff and in permitting escape from danger.
Physical boundaries, is one such measure, as illustrated by the following quote:
‘We've got a very high counter that nobody can sort of like see behind.’ (GP5, Focus Group 2, female.)Escape routes, were also referred to:
‘[I]n our practice ah, we have a a whole series of um, anti violence measures. Um, the first every ah, every room has two doors. [...] Um, so every room has an exit. We also have an exit from the surgery that is not visible to the patients.’ (GP10, Focus Group 2, female)
Tertiary strategies
Tertiary strategies were measures to deal with established violence. It should be noted that though for the most part the primary and secondary strategies outlined above were ad hoc—the result of informal decisions made by individual practitioners rather than widespread policy approaches—there were a number of measures or themes that were common or near-universal among the focus group respondents. Considered approaches to established violence, however, were not as prevalent. They are detailed below as self defence, alarm systems and unorthodox measures.
Idiosyncratic measures. Self-defence: The participants in this study did not have any particular plans for physically responding to violence apart from one male who had taken up Tae-kwon-do to better deal with any assault by patients.
Alarm systems: More developed measures were outlined by one GP for summoning aid in the event of patient violence in her practice:
‘Then we have a duress alarm which rings police with a siren ah, and we have two methods of activating that at the front desk and then we have a paging system in our rooms where we have a page button on our phones.’ (GP10, Focus group 2, female.)
Unorthodox measures: Other strategies mentioned were less orthodox, e.g. brandishing a Jehovah's Witness bible at an assailant, buying a fake gun to respond to violence, and one participant said:
‘I would just fake faint or a heart attack or something and I would hope that the person would feel that they had caused a heart attack or in some way harmed me and then leave.’ (GP3, Focus Group 1, male.)
These unorthodox strategies, perhaps, reflected a sense of GPs' helplessness in the face of the threat of violence:
‘Undoubtedly we are primates struggling to handle too large a troop. Where do I get the capsicum spray?’ (GP33, male.)
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
There was much variability in the comprehensiveness and sophistication of GPs' responses to violence in this study, and an acknowledgement that much of the approach to potential violence was ad hoc and uncoordinated. Nevertheless, a number of levels of intervention were identifiable—characterized here as primary, secondary and tertiary.
The primary and secondary strategies were far more prevalent than the tertiary strategies. There are a number of reasons why this might be so. Firstly, despite many of the respondents having experienced assault or serious threats at some time in their career, episodes of physical violence (as in a previous qualitative study)3 were perceived to be relatively rare. As noted, strategies were seldom the result of considered or coordinated planning. They were more likely to be post hoc responses to experiences—and these experiences were more likely to be verbal abuse and explicit or implicit threats rather than physical violence.
A further reason may be the culture and value system of general practice. GPs both in this and in previous research3 have expressed the view that violence by their patients may represent a failing on their (the GPs') part—a lack of professional expertise or competence. Respondents in this study were keen to be ‘pro-active’, rather than reactive—especially in the area of interpersonal and negotiation skills which were seen as attributes of the experienced and competent practitioner.
The most prevalent strategy identified by GPs in this study was restriction of practice—geographic, demographic and temporal. It is interesting to note that limitation of practice was largely concerned with avoiding taking on the care of patients who were drug seeking or had illicit drug use problems or with patients from low socio-economic areas. There was little suggestion that these GPs attempted to, or felt they would like to, restrict access to their practice of patients with mental health problems. This was despite there being a perception among the focus group participants that mental health patients were a more dangerous (and much more unpredictable) demographic than illicit drug users. This is consistent with the findings of Elston et al.3 that London GPs considered a diagnosis of mental illness rendered the transgressor less culpable for their behaviour and afforded them a degree of toleration. The perception of practitioners in the current study that drug-seeking addicted patients were responsible for their own actions, while mentally ill patients may not be, mirrored the dichotomy in Elston et al.'s study of ‘victims’ and ‘villains’, and medicalization versus criminalization.
While not as prominent as in this study, the restriction of practice in GPs' responses to violence has been noted in previous UK studies.3,5–8 The differing structures of service provision between the UK and Australia can be seen to greatly influence the scope and consequences of this restriction. Australia, like many other countries but unlike the UK, The Netherlands and Denmark,9,10 does not have registration of patients with a GP or practice and Australian GPs do not have specific contractual obligations to provide particular services to patients. Billing practices for consultations in Australian general practice comprise a mix of government-subsidised and patient-borne expense. Price disincentives and selective denial of access to care (most egregiously, ‘vetting’), as were found in this study, can be implemented by Australian GPs without the constraints imposed by the structure and contractual arrangements of the UK NHS.3,6 And, unlike the UK,3 there is no mechanism for finding alternative care for a patient excluded from a practice. Thus, Australian practitioner responses to violence may contribute to inequality of access to primary health care and a drift to a two-tier health system.
As previous studies on occupational violence have been almost exclusively conducted within the UK NHS, the impact on service provision and access to primary care in other countries (with general practice organisation more akin to that of Australia, or with mixed capitation and fee-for-service such as New Zealand11), may previously not have been fully appreciated.
The qualitative nature of this study (and triangulation of written and focus group responses) afforded a rich and nuanced appreciation of a complex and emotionally challenging area. Despite the purposive sampling of a range of GP and practice demographics, and despite thematic saturation having been achieved in the focus groups (and the agreement with triangulation from a large number of written responses), the relatively small number of focus group participants renders this an exploratory study requiring replication in larger studies. Furthermore, quantitative studies to assess the prevalence of restrictions to general practice in response to fears of violence, and of the flow-on effect of such limitations on other services (e.g. accident and emergency services) and on patients, are indicated.
The findings will, however, have implications for further research and health policy. There has been much debate in Australia and internationally regarding the need to improve access to after-hours primary care and the best means to achieve this.12–15 What has not been appreciated in this debate is the degree to which fear of violence is impacting upon GPs' retreat from after-hours care and the possibility that systems-level solutions to safety concerns may attenuate these fears and improve GP participation in after-hours care.
Furthermore, the restrictions on practice that we have chosen to categorize as geographical and demographic could, especially in health systems without the UK NHS's guarantee of access, be seen to entail serious compromise of equity of access to primary care. This is emphasized by a de facto means of achieving demographic and geographic restriction employed by some practices in this study: increasing fees beyond the capacity or willingness of some groups to pay.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
A schema for characterizing GPs' responses to violence as primary, secondary or tertiary has been proposed. The findings will have implications for further research and for policy, especially in health systems without general practice registration.
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
Ethics: Ethics approval for this project was obtained from the University of Newcastle Human Research Ethics Committee. Approval No. H-534-0303.
Funding: The study was supported by a National Health and Medical Research Council grant.
Conflict of interest: The authors have no conflict of interest.
|
Notes |
|---|
Magin P, Adams J, Ireland M, Joy E, Heaney S and Darab S. The response of general practitioners to the threat of violence in their practices: results from a qualitative study. Family Practice 2006; 23: 273–278
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionDeclarationReferences |
|---|
1 Wright NM, Dixon CA, Tompkins NE. (2003) Managing violence in primary care: an evidence-based approach. Br J Gen Pract 53:557–562.[Medline]
2 Department of Health. NHS Zero Tolerance Zone: We Don't Have to Take This. Resource Pack 1999. (The Stationery Office, London).
3 Elston MA, Gabe G, Denney D, Lee R, O'Beirne M. (2002) Violence against doctors: a medical(ised) problem? The case of National Health Service general practitioners. Sociol Health Illn 24:575–598.[CrossRef]
4 Hobbs FD. (1991) Violence in general practice: a survey of general practitioners' views. BMJ 302:329–332.
5 General practitioners—violence against general practitioners and their staff. In: British Medical Association; 2003. Available at: http://www.bma.org.uk.
6 Health and Policy Economic Research Unit BMA. Violence at work: the experience of GPs in Northern Ireland. 2003.
7 D'Urso P and Hobbs R. (1989) Aggression and the general practitioner. BMJ 298:97–98 Available at: http://www.bma.org.uk/ap.nsf/Content/violenceNI.
8 Hobbs FD. (1994) General practitioners' changes to practice due to aggression at work. Fam Pract 11:75–79.
9 Sandvik H and Cho H-J. (2002) Attitudes to family practice registration programmes. Survey of Korean and Norwegian family doctors. Fam Pract 19:72–76.
10 Van Weel C. (2004) How general practice is funded in the Netherlands. Med J Aust 181:110–111.[Medline]
11 Tomlin A and Martin I. (2003) Registered and unregistered patients in general practice: implications for health organisations. N Z Med J 116:U684.[Medline]
12 Alexander C and Fraser J. (2004) Occupational violence in an Australian healthcare setting: implications for managers. J Healthc Manag 49:377–392.[Web of Science][Medline]
13 McKinley RK and Roberts C. (2001) Patient satisfaction with out of hours primary medical care. Qual Health Care 10:23–8.
14 Lattimer V, Smith H, Hungin P, Glasper A, George S. (1996) Future provision of out of hours primary medical care: a survey with two general practitioner research networks. BMJ 312:352–356.
15 Lattimer V, George S, Thompson F, Thomas E, Mullee M, Turnbull J, et al. (1998) Safety and effectiveness of nurse telephone consultation in out of hours primary care: randomised controlled trial. The South Wiltshire Out of Hours Project (SWOOP) Group. BMJ 317:1054–1059.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. Miedema, J. Easley, P. Fortin, R. Hamilton, and S. Tatemichi Disrespect, harassment, and abuse: All in a day's work for family physicians Can Fam Physician, March 1, 2009; 55(3): 279 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Magin, J. Adams, E. Joy, M. Ireland, S. Heaney, and S. Darab Violence in general practice: Perceptions of cause and implications for safety Can Fam Physician, September 1, 2008; 54(9): 1278 - 1284. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||