Australian Institute of Criminology

Skip to content

The impact working alone has on officers successfully performing their duties

Research typically focuses on two aspects of single person patrols—the effectiveness and safety of the practice and police officer perceptions of single person patrols. In particular, single person patrols have great appeal for police management as they are viewed as being efficient and cost-effective, whereas two person patrols are considered preferable from a safety perspective (Johnson 1999; Wilson & Brewer 1991a). There is also debate about whether the practice of single person patrols is a false economy, an issue that is explored further below (Johnson 1999).

The following section continues to investigate the challenges involved in policing, with a specific focus on research related to single person patrols. This section reviews the available literature on single person patrols and summarises the main arguments used to justify or abolish single person patrols. Importantly, the lack of current research on the topic (with the majority being conducted in the early 1990s or earlier), has limited the ability to confidently apply the findings to the current policing practices.

Safety

In general, research on police working alone is limited, with mixed findings and opinions on its viability and effectiveness. It is widely recognised that police officers are faced with greater OH&S risks than those working in other occupations (eg Brandl & Stroshine 2003; Mayhew 2001). Being attacked or injured on the job by an offender is of primary concern for many officers. For example, in one American study, officers ranked being physically attacked as the third most significant stressor in their work, following killing someone in the line of duty and having a fellow officer killed (Violanti 1994). In addition, officers who were assaulted reported increased levels of alienation and decreased levels of job satisfaction and work-related support (McMurray 1990). Despite this, a US study showed that assaults on officers were a rare event when compared with accidents, including serious injury and death (Brandl & Stroshine 2003). Accidents were responsible for the most serious injuries sustained, the most medical treatment administered and the most days off work (Brandl & Stroshine 2003).

However, in 2009, Fridell et al. (2009) summarised the following eight workplace risk factors identified by the US National Institute for Occupational Health and Safety that are applicable to US police officers:

  • contact with the public;
  • having a mobile workplace such as a taxicab or police cruiser;
  • working with unstable or volatile persons in health care, social service, or criminal justice settings;
  • working alone or in small numbers;
  • working late at night or during early morning hours;
  • working in high-crime areas;
  • guarding valuable property or possessions; and
  • working in community-based settings.

Even though injuries from assaults are relatively rare, they can have significant long-term emotional and physical effects on victims. In Australia from 1830 to 1999, 187 officer homicides were recorded, with most through shooting (n=114), spearing (n=24; although none since 1933) and stabbing (n=12; Webster 1999). In Australia, approximately one officer per year is killed in the line of duty and many more are injured or assaulted (Mayhew 2001). Estimates have placed assaults on police officers in the range of 10 percent in Australia (Hastings 2007).

The most comprehensive and widely cited study comparing single person patrols and two person patrols was carried out in the United States by the San Diego Police Department and Police Foundation in 1976 (Boydstun, Sherry & Moelter 1977). The study concluded that with the exception of special situations, tactical assignments, need for field training and other temporary conditions, one person units should be the normal patrol unit in the city as single person patrols were safer and more economical than two person patrols. However, the conclusion of increased safety in single person patrols might be erroneously attributed to patrol mode (1 or 2 person) rather than the more likely scenario that single person patrols are naturally sent to more low-risk calls than two person patrols, thereby skewing the findings (Pruncken 1990). Pruncken (1990) contacted the San Diego Police Department to obtain a detailed update of the deployment policy 14 years after its implementation and found that the San Diego Police Department no longer used the methodology outlined in the 1976 study. This is because the police department reported experiencing the highest officer mortality rate of any major city in the United States following introduction of the policy. As a result, the San Diego Police Department deemed that it was safer to deploy two person units. The police department concluded that officer safety was the most important consideration, rather than cost considerations. Furthermore, they determined that single person patrols should be used only in low-risk patrol situations.

In addition, although single person patrols may be used for low priority/risk calls, even these calls can turn violent. In the United States, the FBI Uniform Crime Report indicates that from 1988 to 1997, 688 officers were killed in the line of duty. Three hundred and fifty of these occurred when officers were responding to ‘low priority’ calls. Of these, 298 officers (86%) were working in single person patrols (Thomas 1999). However, as these numbers are not expressed as rates based on the number of operation modes, it is impossible to draw any definitive conclusions regarding comparative safety between policing modes.

In the mid-1980s, researchers in the United States began ranking police calls for deployment on the basis of ‘danger ratios’ (Kaminski & Sorensen 1995). Danger ratios for a particular category of call are calculated by dividing the number of call-specific injuries by the number of calls for that activity. These types of call were then ranked on their respective ‘danger ratios.’ Uchida, Brooks and Kopers (1987) reported ‘legal intervention’ to be the most hazardous police activity in terms of risk of injury. Legal interventions include executing search and arrest warrants, transporting prisoners, conducting jail searches and backing up officers. Alcohol problems, domestic disturbances and other disturbances were classified as the second, third and fourth most dangerous police activity, respectively. This analysis contradicted the previously held notion that domestic disputes were the most dangerous deployment assignment for police (Kaminski & Sorensen 1995).

Another study, conducted in the United States in North Carolina by Hirschel, Dean and Lumb (1994) ranked the type of call based on the ratio of assault and injury to police. The 10 types of calls ranked from most likely to least likely to result in an officer assault were ‘mentally deranged’ (most likely), handling prisoners, other arrest, domestic disturbance, general disturbance, robbery, other, suspicious persons/circumstances, traffic and burglary (least likely). Offences not placed in any of the police activity category such as rape, damage to property and larceny, were allocated to ‘other arrests’. In a US study on predictors of police victimisation, it was found that officers are at greater risk of victimisation when they deal with an individual impaired by either drugs or alcohol, when they encounter the individual at night, when it is a police initiated encounter, when bystanders are present and when the officer knows the location is dangerous (Rabe-Hemp & Schuck 2007).

Assaults on police

As already indicated, the research evidence is mixed in relation to officer safety and police deployment policy. In examining officers’ risk of assault in the workplace, Fridell et al. (2009) identified several studies that found that the highest numbers of assaults on police in the United States are among officers who are alone at the time of assault. However, it has been suggested that this elevated number may be a reflection of the large proportion of officers assigned to single person patrols in the United States, rather than an increased risk by assignment type per se (Pate & Fridell 1993).

In 2008, of those US law enforcement officers assaulted in the line of duty, the largest percentage of officers assaulted (33%) were responding to disturbance calls (family quarrels, bar fights and so on), while the second highest percentage of officers assaulted were attempting other arrests (15%). Officers assaulted while handling, transporting, or maintaining custody of prisoners was the third most likely to result in officer assault (13%). Of the US law enforcement officers who were assaulted and were in one person vehicles, 60 percent were assisted by fellow officers and 40 percent were unassisted and alone (FBI 2009). Law enforcement officers assaulted alone and unassisted were handling a traffic pursuit/stop (31%), responding to a burglary in progress, or pursuing burglary suspects (29%; FBI 2009).

Pate and Fridell (1993) examined the level of victimisation in proportion to the patrol assignment (1 person patrol or 2 person patrol) across 56 large United States cities, controlling for crime rates and other community variables. They found higher levels of felonious killings among agencies that deployed a higher proportion of single person vehicle patrols. More recently, the US Uniform Crime Report (FBI 2009) provided information on law enforcement officers killed and assaulted in the line of duty. The report revealed that 62 percent of the law enforcement officers assaulted were assigned to one person vehicle patrols, 19 percent were assigned to two person vehicle patrols, five percent were assigned to detective/special assignment and 15 percent were assigned to ‘other’ assignment. Of the 154 law enforcement officers feloniously killed between 2000 and 2009, and who were alone and unassisted in single person patrols, 49 officers were involved in a traffic pursuit/stop, 31 in an ambush situation, 25 officers were investigating suspicious person/circumstance and 21 were in an arrest situation (FBI 2009).

Conversely, a US study found that police assigned to single person patrol vehicles were less likely to be killed while at work than those in two person patrols (Kaminski 2002). That is, fewer police homicides occur where single person patrols are deployed than where two person patrols are deployed. In addition, Australian and US researchers have found no statistical difference in safety between the two staffing modes, reporting that officers were assaulted the same number of times regardless of their assignment to single or two person patrols (del Carmen & Guevara 2003; Wilson & Brewer 1991a).

Clearly, the research evidence is mixed in relation to police safety and officer deployment policy. These results must be considered in the context of the different tasks that officer patrols respond to. Consideration must also be given to the applicability of the findings to other jurisdictions. Single person patrols are more likely to be deployed to low-risk calls, which may affect the rates of officer injury. It is also noteworthy that the likelihood of sustaining injury during the assault is statistically more likely for those patrolling alone compared with those patrolling in pairs (Wilson et al. 1990).

Resisting arrest

Resistance encountered by patrol officers is often factored into research examining the level of danger associated with police patrolling. Resistance can include verbal and/or physical actions. Research conducted by Wilson and Brewer (1991b) on patrol teams in South Australia and New South Wales reported that police encountered physical resistance in 16 percent of their patrol activities and in an additional 15 percent of taskings, officers encountered verbal argument with the potential to escalate to physical resistance. The likelihood of resistance varied with time of day, with increased resistance associated with the busier afternoon and night shifts, and the likelihood of greater numbers of civilians at these times (Wilson & Brewer 2001). Of the taskings that involved physical resistance, seven percent resulted in officers sustaining injury. The type of injury differed in severity, with most injuries involving simple abrasions or bruises (Wilson & Brewer 2001).

Australian research has indicated that although single person patrols were less likely to be dispatched to some taskings, the level of resistance officers encountered and the numbers of injuries sustained in single person patrols in general did not differ significantly from those encountered by two person patrols (Wilson & Brewer 1991b). The researchers argued that the level of resistance for officers on patrol was more dependent on the task being undertaken than on any other variable (Wilson & Brewer 1991b). By analysing offender resistance at police callouts, it was found that the level of resistance faced by police was not influenced by factors such as location (eg rural/urban), officer characteristics (eg age, gender, experience), jurisdiction, or the nature of patrol (ie 1 or 2 officers; Wilson & Brewer 1991a).

Regardless of whether one or two person patrols are in attendance, the potential for escalation from verbal resistance to physical violence is possible. For instance, Wilson and Brewer (1991b) found that officers in single person patrols appeared to be subjected to verbal abuse or approached the verbal abuse level in 13 of the 32 taskings an officer would have to attend. Even verbal abuse is of concern due to the potential of the situation escalating to violence (Wilson & Brewer 1991b). Two taskings—a request for urgent backup and a hotel brawl in progress—had an average level of resistance that was more severe than verbal abuse and argument and therefore would be advisable to have two person patrols for the sake of caution (Wilson & Brewer 1991b). As such, the authors suggested that time of day should be used to determine whether the taskings should be classified as either appropriate or inappropriate for single person patrol duties (Wilson & Brewer 1991b).

Other research has found that single person patrols are less likely to be involved in incidents of resistance than two person patrols (Boydstun, Sherry & Moelter 1977; Wilson & Brewer 2001). However, after raising this argument, Wilson and Brewer (1991b) also found no significant differences between patrolling modes and likelihood of contact with an offender. Johnson (1999) has speculated that two person patrols are more likely than single person patrols to use arrest as the method of resolving disputes (Johnson 1999). However, in an Australian study, it was found that in 28 of the 32 tasking activities the patrol mode, either single person or two person, did not affect the method of tasking resolution (Wilson & Brewer 1991b). Two of the four activities, hotel brawls and request for urgent backup, resulted in significantly more arrests and reports for two person patrols. Of the remaining two activities, the results were less clear cut. Single person patrols were more likely to arrest and warn more suspects than two person patrols when undertaking the ‘detainment of an intoxicated person’, whereas two person patrols were more likely to make an arrest, or indeed report that no offence had been detected for ‘domestic arguments’ (Wilson & Brewer 1991b: 22).

Use of force

Examination of use of force is also used in research as an indicator of officer safety, although the rate of incidents involving use of force are low in relation to the number of contacts with the community. A comparison of use of force incidents in New South Wales and South Australia for single person and two person patrols was conducted by Hastings (2007), with more use of force incidents occurring for two person patrols than single person patrols. The rate of reported use of force incidents by single person patrols remained relatively stable between 2000–01 and 2004–05, with approximately 0.02 reported incidents per 100 patrols, while rates for two person patrols ranged from 0.08 to 2.97 reports per 100 patrols. This finding was attributed to three explanations. These are:

  • single person patrols are less likely to be tasked to incidents considered to be high risk. This would intuitively result in two person patrols being involved in incidents that had a higher probability for the need to use force;
  • an officer working on their own may be more likely to try and avoid a physical confrontation, such as talking their way out of it, than two person patrol; and
  • an officer may be more reluctant to back down from a risky situation in front of their colleague than an officer working alone (Hastings 2007).

Hastings (2007) indicated that the mode of patrol is a less significant indicator of officer safety than the type of calls officers are deployed to. Hastings (2007) also concluded single person patrols were no less safe than two person patrols when safety controls such as rostering, policies, training and equipment are suitable.

Effectiveness and efficiency considerations

Deployment of officers in one and two person patrols varies both between and within jurisdictions in Australia and overseas (Rich 1984). Given finite funding and labour resources, operational effectiveness cannot be ignored. It has been suggested that single person patrols are more efficient than two person patrols because it is argued that two single person patrols can patrol twice as much area and be available for twice as many calls than one two person patrol car (Bureau of Justice Statistics in del Carmen & Guevara 2003). In addition, solo patrol cars can be implemented almost immediately at half the cost (Boydstun, Sherry & Moelter 1977), with no additional recruitment (‘A quick fix: One-man patrol cars’ The New York Times 24 September 1990 http://www.nytimes.com/1990/09/24/opinion/a-quick-fix-one-man-patrol-cars.html). However, these assessments of cost are foreign and were made over 20 years ago. No extensive or recent research was found on the cost and efficiency of single person patrols in Australia.

Wilson and Brewer (1992) noted increasing single person patrols potentially results in more cars on the road, which is likely to provide beneficial impacts in terms of increased police visibility, time spent on preventative patrol (and therefore the likelihood of detecting a crime in progress), the average area covered by patrol and the average length of patrol time. This must then be weighed against possible negative impacts, including the probability of effecting an arrest, job satisfaction and assaults and injuries to police officers. Despite these findings, Wilson (1990) found that the effects of increased police visibility, deterrence and crime detection as a result of the increase in number of cars on the road was so small that it justified only minor consideration. One difference between the two staffing modes was that two person patrols were reported to produce more traffic citations and citizen complaints (Wilson & Brewer 1991a).

In any patrolling evaluation, Wilson and Brewer (1992: 444) recommended incorporating factors such as:

...issues of effectiveness (type, quantity, and quality of patrol activity), efficiency (the relationship between patrol activity and cost), safety of patrol officers, and attitude and preference of the patrol officers for the various modes.

The increased efficiency of single person patrols has been questioned. A large investigation into one and two person patrols in Australia conducted in 1991 found that single person patrol staffing work levels were comparable to two person patrols (Wilson & Brewer 1991a). More specifically, the effectiveness of most police services (eg number of calls for service handled, total arrest rate) was not affected by the patrol staffing mode. Although two person patrols tended to handle incidents faster—which was also found earlier in the United States (see Chelst 1981)—overall activity levels were found to be comparable between one and two person staffing (Wilson & Brewer 1991a).

An argument raised against single person patrols is that they may be an inefficient use of resources as they are more likely to require backup than two person patrols and therefore could be considered a ‘false economy’ (Johnson 1999). On the other hand, it has been suggested there is little evidence to support this assertion because the vast majority of incidents require no more than one officer (Bailey 2008). Survey results from operational officers indicated that genuine emergencies occur relatively infrequently, with 47 percent of surveyed respondents not dealing with any during their last shift and a further 35 percent reporting that at most there were only one or two genuine emergencies during their shift (Bailey 2008). Furthermore, of the incidents attended, 35 percent of police officers indicated that none of the incidents required more than one officer, with an additional 46 percent claiming only one to three incidents needed two or more officers.

A British study into the impact of single and two person patrols pursuing burglars found no difference in the success of catching burglars between the two modes after differences in response times and stage of the burglary were taken into account (Blake & Coupe 2001). The report authors recommended greater deployment of single person patrols to increase the likelihood of catching burglars in the act as the nearest available unit could be closer to the burglary scene, resulting in decreased response times. Earlier, Kessler (1985) found that two single person patrol cars arrived at incident scenes faster than one two person patrol car and thereby increased efficiency in response times. One explanation for this finding was the effect of peer pressure among officers (Wilson & Brewer 1992).

Workplace attitudes to working in single person patrols

From the police officer’s point of view, opinion on single person patrols is mixed, but the majority of frontline officers interviewed in the United Kingdom did not like the concept (Prissell 2009). Particular concerns included that it was strenuous, a feeling of constantly having to ‘watch one’s own back’, not wanting to engage with potential threats late at night, and that they considered themselves easy targets (Prissell 2009). As well as possible apprehension about facing danger alone, there is also the more indirect concern after a dangerous incident of them not sharing the incident with another officer (Lindsey & Kelly 2004).

In the United States, a study was conducted on police officers’ perceptions of one and two person patrols regarding performance, applicability, effectiveness and safety issues (del Carmen & Guevara 2003). The researchers found that police believed the performance levels between one and two person patrols were comparable. However, two person patrols were considered preferable at night and in areas where police were not as trusted, were seen as more observant and were believed to respond faster to calls than single person patrols. They also found that officers did not consider that two single person patrols were twice as effective as one two person car, nor did they agree that they were more likely to be injured in a two person car than a one person car (del Carmen & Guevara 2003).

Research on Australian officers’ attitudes to single person patrol focusing on success, anxiety and patrol situation found a preference for two person patrols, a finding which is similar to research in the United Kingdom and United States (Brewer & Karp 1991). In Australia, even though most police officers (82%) surveyed preferred two person patrols for the 30 identified tasking situations described, the remaining 18 percent of officers indicated preferring single person patrols for an average of 10 identified situations (Wilson & Brewer 1991a). Prior experience of single person patrol was not associated with a higher frequency of single officer preference; however, officers with less than two year’s experience were associated with a greater preference for single person patrol (Wilson & Brewer 1991a).

Expectations of likely success has been used as an indirect measure of officer attitudes to patrol mode as it has been shown that this correlates with preference for working alone or in a group (Vancouver & Ilgen 1989). Officers with single person patrol experience indicated more patrol activities with high expectations of success and ‘normal’ anxiety levels. In other words, although officers expressed resistance to single person patrol modes when asked directly which patrol mode they preferred, they indicated a number of tasks they considered they could perform successfully within normal levels of anxiety, which single person patrols may be able to undertake (Brewer & Karp 1991).

In Australia, researchers have noted that officers prefer the two person patrol model, considering it to be both safer and more effective despite some research suggesting otherwise (see above). Furthermore, negative occupational outcomes can be expected among officers who perceive increased dangers resulting from single person patrols and who are required to undertake these patrols (Brewer & Karp 1991).

Anecdotally, an officer with experience working as a single person patrol indicated that single person patrol work could lead to a degree of loneliness, although this feeling usually subsides overtime (NT Police representative personal communication 4 Aug 2011). The officer raised concerns that when working alone an officer has to be well versed in legislation, general orders and police policies and procedures as officers did not have the luxury to ‘bounce’ ideas or seek assistance from a partner and sometimes radio or phone communication is not always a viable alternative. To overcome the boredom that can accompany working alone, the representative mentioned that ‘you will often find yourself doing more traffic apprehensions or attending as secondary vehicle to an incident and rendering assistance wherever possible’ (NT Police representative personal communication 4 Aug 2011). It was also suggested that single person patrols have the potential to negatively impact an officer in the following situations:

  • knowledge that some prosecutors will not proceed with matters in court if the defendant pleads not guilty as there is a perception that the judicial system is more likely to believe a defendant before a police officer;
  • uncertainty around knowing when help will arrive if an incident becomes violent; and
  • being unable to attend urgent jobs that could reduce the likelihood of harm to victims or people (NT Police representative personal communication 4 Aug 2011)

These are, of course, anecdotal accounts. Despite this, they offer a degree of insight into areas that have not been explored in available single person patrol literature and therefore warrant future exploration.

Single person patrols were trialled in certain parts of urban Sydney between February and July 1993 (Gibson 1995). A study of the trial found that officers who were on single person patrols reported increased job satisfaction and personal confidence (Gibson 1995). In Victoria, Parkinson (2010) reported that Traffic Management Unit and Regional Traffic Tasking Unit officers regularly patrolled alone and in many cases preferred to do so. Research has also indicated that members were happy to work in single person patrols for traffic duty as long as certain conditions were met (Hastings 2007). These included choosing to work in single person patrols primarily in daytime hours and that the decision to work alone was theirs and not decided for them. This would involve rostering sufficient officers on duty to enable two person patrols. In addition, it was noted that management/supervisor concerns about safety would take priority over an officer’s choice for working as a single person patrol.

As noted above, the perceived benefits of single person traffic patrols in Victoria include increasing the police unit’s productivity and the amount of road that can be covered (Hastings 2007). This patrolling mode was determined by the officer in charge of the shift and in regard to the preference of the officer.

Stress

Policing is recognised as one of the most stressful occupations (Sheena et al. 2005). A substantial amount of research over the least three decades has found police officers are at risk of physical health problems, burnout and psychological issues, as well as smoking, suicidal thoughts and alcohol abuse (Adams & Buck 2010). As well as experiencing many of the stressors normal to most workplaces (balancing work and family, excessive workload, lack of control and lack of support), police are also required to interact with hostile suspects and offenders and emotionally distressed victims (Adams & Buck 2010). As such, the minimisation of stressful situations for police is considered paramount. However, there is limited information on whether being deployed in single person patrols can add to this stress, or if it has no significant impact.

A recent study (Violanti et al. 2008) looked at psychological, physiological and subclinical measures of stress, disease and mental dysfunction among officers in Buffalo, New York Police Department in the United States, called the Buffalo Cardio-Metabolic Occupational Police Stress. The researchers identified that working solo at night without immediate backup could heighten stress (Violanti et al. 2008). However, no Australian police jurisdiction contacted for this report identified the deployment of single person patrols at night time.

In addition, facing unpredictable tasks can increase levels of stress for police officers dealing with operational situations, which is a key risk factor of psychological distress (Dollard, Winwood & Tuckey 2008). In 2005, PASA and the University of South Australia launched a project to examine resources, work demands and psychological wellbeing of frontline police officers in regional and metropolitan South Australia (Dollard, Winwood & Tuckey 2008). This study found that nearly 20 percent of constables and 19 percent of sergeants showed levels of psychological distress high enough to warrant support from a mental health professional. These rates were also twice as high as the national population. As well as serious health effects, psychological distress may also lead to stress compensation claims. Twelve percent of officers surveyed stated that they had put in a worker’s compensation claim for stress at least once (Dollard, Winwood & Tuckey 2008). Additional organisational and productivity consequences of these high levels of psychological distress were identified, including lower levels of job satisfaction and engagement, and increased risk for turnover. One of the key risk factors or demands associated with psychological distress was ‘stressfulness of operational incidents (Dollard, Winwood & Tuckey 2008). Despite this, there was no indication in this study of whether patrol mode contributed (either positively or negatively) to increasing the stressfulness of operational incidents and how officers in single person patrols cope with unpredictable tasks compared with other patrol modes.

Legislative requirements

Policing, like all professions, is subject to laws regarding workplace OH&S. The legislative requirement relating to duty of care for employers, although specified, remains unclear for actions and compliance required in relation to single person patrols (Association News 2007). Under the Workplace Health and Safety Act 1995, employer duties are in s 9, which states that the employer is to

ensure as far as reasonably practicable that the employee is, while at work, safe from injury and risks to health and, in particular, must

  1. Provide and maintain so far as is reasonably practicable–
    1. a safe working environment; and
    2. safe systems of work; and…
  2. provide any information, instruction, training and supervision reasonably necessary to ensure that each employee is safe from injury and risks to health.

There are limited examples of the application of legislation in relation to officer patrolling mode in Australia, however the issue was raised in NSW v Fahy 2007. In this case, an officer was claiming damages for developing Post Traumatic Stress Disorder after being left unassisted with a seriously injured man during the course of her duties. The NSW District Court awarded her damages for negligence

due to being unreasonably left in a traumatic situation without the support of a fellow police officer, and her injury was in consequence of a breach of duty by her employer to take reasonable care for her safety (NSW v Fahy 2007: np).

Despite this, the State successfully appealed to the High Court in relation to liability and the Court allowed the appeal. The case was appealed on the grounds that Ms Fahy had failed to establish that the State breached its duty of care and that it was not sufficient merely to allege that the State should have instructed police officers working in pairs, that they should whenever possible remain together and that they should provide psychological support to each other during traumatic incidents.

An example of the application of legislative requirements to single person patrols is illustrated by WA Police. In accordance with WA State legislation, s 19 of the Occupational Safety and Health Act (OSH) 1984 requires the employer to provide a safe place of work for employees. In relation to working alone, WorkSafe (part of the Department of Commerce, the Western Australian State Government responsible for the administration of the Occupational Safety and Health Act 1984, see http://www.commerce.wa.gov.au/WorkSafe/) released a Working Alone Guidance Note (WA DoC 2009). Although police officers are not specifically mentioned in regulation 3.3, the WA OH&S Regulations is used to provide guidance. Regulation 3.3 requires the employer to provide reliable means of communication for isolated employees (those working alone). As the current WA Police communications and phone system do not provide reliable means of communication for isolated employees, there is a need to develop a matrix of duties that are considered reasonably safe to be performed by a police officer working alone (Safety Officer WA Police Union personal communication 20 September 2010).

Guidelines and training

The overwhelming majority of training techniques and tactics used in police training are based on a two person patrol situation. Therefore, a need exists for the development of operational guidelines specific to each patrol mode, as it is insufficient to develop one set of guidelines to cover all modes in all circumstances (Wilson & Brewer 1991a). An example is apparent in Ireland, with health and safety regulations requiring Gardaí police officers to place a prisoner in the back of a car and to have a second police officer sitting directly behind the driver to ensure the officer driving is not attacked (‘Attacks on gardai spark demand to end one-man patrols’ Irish Independent 1 May 2008. http://www.highbeam.com/doc/1G1-178472475.html). These operational guidelines must incorporate information on taskings, environmental conditions and officer and offender characteristics. Therefore, in Ireland it is not possible for single person patrols to make arrests and drive back to the station (‘Attacks on gardai spark demand to end one-man patrols’ Irish Independent 1 May 2008. http://www.highbeam.com/doc/1G1-178472475.html 2008).