Use of Alcohol or Drugs

Use of alcohol or drugs was the most common variable reported, highlighting the need for preventive strategies such as encouraging sensible alcohol consumption and safe drug use. In a study of 15 MGEs in Australia, during summer months, the authors reported 90% of patient presentations occurred at events where alcohol was available.^{Reference Anikeeva, Arbon and Zeitz11} Alcohol and substance use were the most commonly identified variables in this review (n = 21/31; 67%).

Anikeeva, et al^{Reference Anikeeva, Arbon and Zeitz11} reviewed 15 MGEs in South Australia in which alcohol availability was categorized as either “alcohol available for purchase” or a “dry event.” In contrast, in most (n = 18/21; 86%) of the reviewed literature for the present study, the use of alcohol is not categorized in this way. Rather, the approach to how alcohol and drug use is captured is haphazard. The nonlinear study conducted by Arbon, et al argued that the accuracy of collecting alcohol or drug use data is limited to possible discrepancy between official and actual presence, and the effect of “loading” prior to the event.^{Reference Arbon, Bottema and Zeitz10} Additionally, Hawkins and Brice, studying the National Collegiate Athletic Association Final Four Semi-Final and National Championship games, found that the patients evaluated by Emergency Medical Services (EMS) had a rate of 70% alcohol use.^{Reference Hawkins and Brice12} Of note, intoxication, violence, and traumatic injuries occur through the use of drugs and alcohol at both sporting and music MGEs.^{Reference Ahmed and Memish13–Reference Hutton, Savage, Ranse, Finnell and Kub15}

Maladaptive and Adaptive Behaviors

Consequences to attendees at MGEs due to norm behavior changes of participants have been quantified through the Hawkins, et al study.^{Reference Hawkins and Brice12} The authors did a retrospective review of ambulance (EMS) and hospital emergency department records of patients injured as a result of risk-taking celebratory behavior from University of North Carolina (Chapel Hill, North Carolina USA) sports championship game. Characteristics of patients evaluated in relation to crowd behavior included assault and bonfire-associated burns.^{Reference Hawkins and Brice12} Another study reviewed used a two-part self-reporting survey of respondents from the school leavers’ celebrations outside of Perth, Australia which recorded the likelihood of negative consequences associated with alcohol and other drug use, as well as other risk factors, at the school leavers’ celebration. Consequences relating to individual behavior included emotional outburst, heated argument, physical aggression, unprotected sex, blackout, stolen private/public property, act of vandalism, and arrests for intoxicated behavior.^{Reference Hutton, Brown and Verdonk14} Sampsel, et al^{Reference Sampsel, Godbout, Leach, Taljaard and Calder36} published the first reports of MGEs associated with sexual assaults and found that sexual assaults peak at MGEs on specific holiday events, including New Year’s Eve, Canada Day, Halloween, and University Freshman’s week.^{Reference Sampsel, Godbout, Leach, Taljaard and Calder36}

In addition to sexual assaults, stampedes have been recorded as crowd behavior.^{Reference Ahmed and Memish13} Stampedes can occur as a result of unmanageable crowd density; however, crowd density is not necessarily documented as a stressor. Hopkins reported that if participants felt greater identification with fellow attendees, they then felt less crowded, stating the joy of crowds related to “collective self-realization” can be a potential health benefit for participants.^{Reference Hopkins and Reicher16}

Crowd Behavior

There are numerous theories of crowd behavior attempting to explain the tendencies of crowds; however, there are few practical applications to monitor and assess crowd behaviors in MGEs that have been discussed or examined. Of the 18 papers that cited crowd behavior as a key influence on MGE PPRs, nine provided a tool for collecting this data.

Patient presentation trends at 15 MGEs in South Australia had fieldworker’s record behavior on the crowd characteristics questionnaire.^{Reference Anikeeva, Arbon and Zeitz11} The crowd characteristics questionnaire recorded the number and proportion of patients presenting to in-site health care service by crowd characteristics variables, which included crowd density, male to female ratio, proportion of crowd seated or stationary, proportion of crowd in motion, proportion of crowd displaying cohesive behavior, and proportion of crowd wearing cohesive dress.^{Reference Anikeeva, Arbon and Zeitz11} Additionally, Hutton, et al’s^{Reference Hutton, Zeitz, Brown and Arbon5} study collected crowd behavior once every hour at five designated collection points using a tool adopted from the work of Zeitz, et al.^{Reference Zeitz, Tan, Grief, Couns and Zeitz7} The main form of data collected were descriptive field notes collected via observation categorized into behaviors: dancing, singing, sitting, resting, walking, talking, chatting, talking on a mobile, sending SMS, fighting, wrestling, videotaping, taking pictures, celebrating, or socializing.^{Reference Zeitz, Tan, Grief, Couns and Zeitz7,Reference Hutton, Munt, Zeitz, Cusack, Kako and Arbon8} Hutton, et al’s^{Reference Hutton, Zeitz, Brown and Arbon5} study evaluated crowd mood using descriptive tools developed by Pines and Maslach,^{Reference Pines and Maslach37} and crowd type using Berlonghi’s model,^{Reference Berlonghi38} applying a scoring system. Although the study was interpretive using participant observation, pre-event training was provided to ensure comprehensive and uniform data collection.

Crowd Culture

Crowd culture can include behaviors such as moshing and pushing and shoving. Hutton, et al documented social environmental factors at MGEs using Haddon’s matrix, an epidemiology tool used to determine potential injuries, through behaviors such as dancing or “moshing.”^{Reference Hutton, Savage, Ranse, Finnell and Kub15} Three studies used different methods to collect different variables to measure and/or identify crowd culture. For example, Milsten, et al^{Reference Milsten, Tennyson and Weisberg17} analyzed retrospective data from prehospital patient care reports to identify mosh-pit-related injuries at rock concerts, festivals, and electronic dance music events.^{Reference Milsten, Tennyson and Weisberg17} Hartman, et al used a scoring system that measured crowd intention as either “animated, intermediate, or calm,” but did not outline the merits behind classification.^{Reference Hartman, Williamson and Sojka18} Finally, Turris, et al listed classification examples of crowd behavior as: mood, activity levels, queuing, movement, behavior, predispositions, motivations, crowd movement, and flow.^{Reference Turris, Lund and Hutton19}

The Emergency Management Australia (Canberra, Australia) descriptors for “crowd types” include behavior (Table 3^{Reference Berlonghi38}).³⁹ Within this comprehensive review, the authors underline the lack of congruence with the use of terminology between “crowd behavior,” “crowd type,” “crowd management,” and “crowd mood.” Moreover, demographics of a crowd is used as an environmental descriptor of a “crowd type.”^{Reference Zeitz, Tan, Grief, Couns and Zeitz7} For example, the profile at a football game is typically male-dominated who are passionate and loyal fans. This crowd type makes predictable crowd behavior.^{Reference Challenger, Clegg, Robinson and Leigh40}

Table 3. Crowd Types/Behaviors^{Reference Berlonghi38}

Crowd Mood

The term “crowd mood” is a descriptor of crowd emotion and tone which has become an indicator of probable crowd behavior outcomes.^{Reference Hutton, Zeitz, Brown and Arbon5} Although crowd mood is less visible than crowd behavior, it can still be assessed and quantified. Pines and Maslach developed a practical matrix that classified the mood of crowd as passive, active, or energetic (Table 4).^{Reference Pines and Maslach37} The table scores the amount of verbal noise, physical movement, and overall audience participation into a numerical grading system.^{Reference Hutton, Zeitz, Brown and Arbon5}

Table 4. Crowd Mood Classifications^{Reference Zeitz, Tan, Grief, Couns and Zeitz7}

Table 5. Quality of Research

Hutton, et al adapted a psychosocial data collection tool from Zeitz, et al^{Reference Zeitz, Zeitz and Arbon41} and Emergency Management Australia³⁹ to group crowd mood as energetic, active, festive, passive, subdued, or delighted.^{Reference Hutton, Munt, Zeitz, Cusack, Kako and Arbon8} Despite Pines and Maslach’s^{Reference Pines and Maslach37} tool being successfully used in Hutton, et al’s study,^{Reference Hutton, Zeitz, Brown and Arbon5} Turris, et al^{Reference Turris, Lund and Hutton19} found that objective measures of crowd mood are not yet well-developed for use in the mass-gathering context; “Crowd mobility and density are sometimes used as proxies for crowd mood or indicators of risk (qualitative).”^{Reference Milsten, Tennyson and Weisberg17} Evidence suggests that crowd mood can be influenced by factors such as type of music^{Reference Friedman, Plocki and Likourezos20,Reference Steffen, Bouchama and Johansson21,Reference Robertson, Hutton and Brown42} or sporting rivalry.^{Reference Zeitz, Haghighi, Burstein and Williams22} Hutton, et al found that crowd mood at sporting events is influenced by the level of engagement with the event.^{Reference Hutton, Ranse and Munn6} For example, they found that spectators displayed a myriad of emotions, such as yelling, cheering, shouting, jumping up and down, and getting angry and upset.^{Reference Hutton, Brown and Verdonk14} The animated mood at sporting events and festivals was also exacerbated by the “conspicuous consumption” of large amounts of food and alcohol, which in turn can have adverse health effects.^{Reference Laing and Frost43} Further adverse circumstances can result in paranoia and mass hysteria leading to crowd crush and violence between attendees at MGEs.^{Reference Anikeeva, Arbon and Zeitz11} To further understand crowd mood, Ranse and Hutton^{Reference Ranse and Hutton3} included anxiety and psychiatric disorder under the mental health header in their patient data set and entry codes. Anxiety was observed as “Presentation related to anxiety or panic attack, not necessarily psychiatric in nature or substance related,” and psychiatric disorder was defined as “Psychiatric or mental health related presentations.”^{Reference Ranse and Hutton3} This remains the only documented evidence of mental distress and is an area that needs further attention in the MGE space.

Reason for Attending Event (Motivation)

The reason for attending an event has a strong impact on the bio-medical domain, as the individual’s motivation for attending and their behavior at an MGE can also be a factor in injury presentation.^{Reference Hutton, Munt, Zeitz, Cusack, Kako and Arbon8} Motivation for attending an event is a key influence on crowd behavior. However, due to visibility, Hutton, et al^{Reference Hutton, Zeitz, Brown and Arbon5} suggested crowd behavior should be the focus as the observable and measurable element. Reasons for attendance were categorized in the comprehensive review by Ranse, et al within individual demographics as either participant, spectator, official, or other.^{Reference Ranse and Hutton3} Hartman, et al scored crowd intentions data with a point value as animated, intermediate, or calm.^{Reference Hartman, Williamson and Sojka18} Hawkins, et al carried out a retrospective review of an MGE to celebrate the success of University of North Carolina men’s basketball championship.^{Reference Hawkins and Brice12} They found that participants’ behavior at celebratory riots ranged from non-threatening behavior to rioting, burning of personal property, and physical violence against other participants or bystanders.^{Reference Hawkins and Brice12} Although FitzGibbon, et al^{Reference FitzGibbon, Nable and Ayd23} and Friedman, et al^{Reference Friedman, Plocki and Likourezos20} did not specifically collect data on motivation for crowd attendance in their studies, Jaensch, et al^{Reference Jaensch, Whitehead, Prichard and Hutton44} and Lam, et al^{Reference Lam, Liang, Chikritzhs and Allsop24} found that a core part of motivation for festival-goers at outdoor music festivals was to use alcohol and other drugs in a group setting. Whereas participants’ motivations that go to religious festivals or pilgrimages are focused on abstinence and faith.^{Reference Hopkins and Reicher16} Like religious festivals, sporting events bring attendees together for a shared sense of belonging.^{Reference Hutton, Ranse and Munn6,Reference Ahmed and Memish13} However, many of these sport enthusiasts may be motivated to fit in with the crowd, which can lead to an over-consumption in alcohol.^{Reference Hutton, Ranse and Munn6}

Duration

Of the papers reviewed, nine included the event duration as a variable that was or should be collected at MGEs.^{Reference Hutton, Zeitz, Brown and Arbon5,Reference Hutton, Munt, Zeitz, Cusack, Kako and Arbon8,Reference Hartman, Williamson and Sojka18,Reference Turris, Lund and Hutton19,Reference Zeitz, Haghighi, Burstein and Williams22–Reference Lund, Turris, Amiri, Lewis and Carson26} Multiple-day events are generally associated with higher rates of medical use.^{Reference Alquthami and Pines25} Data from Hutton, et al’s pilot project^{Reference Hutton, Munt, Zeitz, Cusack, Kako and Arbon8} showed that there is a change in crowd behavior from the start of the event and the end of the event.^{Reference Hutton, Zeitz, Brown and Arbon5} For some of the research papers, duration was a fixed variable determined during data collection, which remained the same for all of the events.^{Reference Zeitz, Haghighi, Burstein and Williams22}

Crowd Demographics

Crabtree, et al’s^{Reference Crabtree, Mo and Ong27} study found that patients from rural/regional areas were twice as likely to require advanced treatment and review by health care professionals. Additional variables such as crowd interest/culture and individual motivation are not as frequently mentioned, but for specific MGEs, are an important aspect of data collection.