Ambulance Use Is Associated With Higher Self-rated Illness Seriousness: User Attitudes and Perceptions

Transcription

1 ORIGINAL RESEARCH CONTRIBUTION Ambulance Use Is Associated With Higher Self-rated Illness Seriousness: User Attitudes and Perceptions Ghasem (Sam) Toloo, PhD, Gerry J. FitzGerald, MD, FACEM, FRACMA, Peter J. Aitken, MBBS, FACEM, EMDM, MClinEd, Joseph Y. S. Ting, MBBS, MSc (Lond), PGDipEpi, DipLSHTM, Kirsten McKenzie, PhD, Joanna Rego, MPH, and Emma Enraght-Moony, PGBSc Abstract Objectives: The objective was to study the role and effect of patients perceptions on reasons for using ambulance services in Queensland, Australia. Methods: A cross-sectional survey was conducted of patients (n =911) presenting via ambulance or selftransport at eight public hospital emergency departments (EDs). The survey included perceived illness severity, attitudes toward ambulance, and reasons for using ambulance. A theoretical framework was developed to inform this study. Results: Ambulance users had significantly higher self-rated perceived seriousness, urgency, and pain than self-transports. They were also more likely to agree that ambulance services are for everyone to use, regardless of the severity of their conditions. In compared to self-transports, likelihood of using an ambulance increased by 26% for every unit increase in perceived seriousness; and patients who had not used an ambulance in the 6 months prior to the survey were 66% less likely to arrive by ambulance. Patients who had presented via ambulance stated they considered the urgency (87%) or severity (84%) of their conditions as reasons for calling the ambulance. Other reasons included requiring special care (76%), getting higher priority at the ED (34%), not having a car (34%), and financial concerns (17%). Conclusions: Understanding patients perceptions is essential in explaining their actions and developing safe and effective health promotion programs. Individuals use ambulances for various reasons and justifications according to their beliefs, attitudes, and sociodemographic conditions. Policies to reduce and manage demand for such services need to address both general opinions and specific attitudes toward emergency health services to be effective. ACADEMIC EMERGENCY MEDICINE 2013; 20: by the Society for Academic Emergency Medicine Growth in demand for, and use of, ambulance services is a global problem. 1,2 Queensland, with a population of 4.5 million in 2010, is the third largest state in Australia. 3 It has the highest per-capita usage of ambulance services (169 per 1,000 persons in 2009/2010), which is growing at an annual From the School of Public Health and Social Work, Queensland University of Technology (ST, GJF, JR), Brisbane, QLD; the Emergency Department, The Townsville Hospital, and School of Public Health, Tropical Medicine and Rehabilitation Sciences, James Cook University (PJA), Townsville, QLD; Mater Health Services, Careflight Medical Services Qld, and University of Queensland Medical School (JT), Brisbane, QLD; the Centre for Accident Research and Road Safety Queensland, School of Psychology and Counseling, and National Centre for Health Information Research and Training, School of Public Health, Queensland University of Technology (KM), Brisbane, QLD; and the Clinical Performance & Service Improvement Unit, Queensland Ambulance Service (EE), Brisbane, QLD, Australia. Received August 17, 2012; revision received October 29, and December 13, 2012; accepted December 14, This paper is based on a research project on demand for emergency health services in Queensland funded by an Australian Research Council (ARC) Linkage grant in partnership with Queensland Ambulance Service (QAS). EEM is an employee and Director of Clinical Performance and Service Improvement Unit at Queensland Ambulance Service, which partly funds the research project in partnership with an Australian Research Council (ARC) Linkage grant. The authors have no potential conflicts of interest to disclose. Supervising Editor: Manish N. Shah, MD, MPH. Address for correspondence and reprints: Sam Toloo, PhD; sam.toloo@qut.edu.au. ISSN by the Society for Academic Emergency Medicine 576 PII ISSN doi: /acem.12149

2 ACADEMIC EMERGENCY MEDICINE June 2013, Vol. 20, No rate of 4.4%, 4 more than twice the rate of population growth (2.1%) during the same period. 5 The literature on drivers of demand for ambulance transport points to a multitude of factors. Clark et al. 6 found that ambulance arrivals in Australia were higher among the elderly and patients with higher-acuity illness or mental, neurologic, or trauma conditions. A study of 930 elderly patients at an urban medical academic center in the United States found the most common reasons for ambulance usage were immobility (33%), illness (22%), request by others (21%), instruction from health care provider (10%), and lack of transportation (10%). 7 Other more recent reviews 8 11 have suggested that escalating demand for ambulance and emergency health services in Australia may be explained by a range of individual and health-related factors, such as illness acuity, accessibility, and availability of other forms of transport; socioeconomic status, social support, and distance from hospital; systemwide elements, such as pricing and quality of service; societal factors, such as population growth and aging and rising community expectations; and greater awareness of, and increases in, the burden of disease. Some studies have identified inappropriate use, in which patients may not meet clinical criteria to be considered as emergency or urgent cases, to be the main driver for increasing demand for emergency health services Many of these studies rely on retrospective analysis by health professionals to determine if a presentation was inappropriate. None sought patients views as to why they used ambulance services. There is also a clear discrepancy between health staff and patient perceptions. In one study, emergency physicians agreed with the transport method in 68% of users and 92% of nonusers (j = 0.61, p < 0.01). 24 What appears to the patient as an emergency or urgent condition does not necessarily correspond with a specialist s view of urgency. Conceptual Framework A conceptual framework based on the literature and health-seeking behavior theories was developed to guide this study. 4 Accordingly, the decision to use the ambulance service occurs as a function of reasoning and rational choice, by weighing potential threats of the health condition against benefits of and barriers to the action. These immediate factors are further influenced by personal and attitudinal characteristics, such as trust in the efficiency and effectiveness of the system, perceived availability and accessibility of resources, previous experience or cues to action, and self-efficacy or belief in one s ability to control the situation. These personal and subjective factors may in turn be affected by the individual s place in the community, as determined by his or her lifestyle, social support, ethnicity, religion, and other sociodemographic characteristics such as age, sex, and socioeconomic status Figure 1 illustrates a summary of the developed model. For the purposes of this article, and using the conceptual framework, we focus on immediate factors affecting the ambulance use decision, and hypothesize that arrival by ambulance is likely to be associated with: 1. Higher perception of the severity of the condition (perceived severity); 2. Higher concern about the costs of alternative means of transport (perceived costs); 3. Higher expectation of receiving better care (perceived benefits); and INDEPENDENT FACTORS MODERATING FACTORS OUTCOME VARIABLES Social & Network Support Information, instrumental, emotional, esteem, material Perceived Costs & Benefits Self -efficacy Socio-demographics o og cs Age, sex, socioeconomic status, marital & living status, ethnicity General HealthStatus Perceived Acuteness Seriousness, urgency, pain Ambulance Use Arrival method, frequent use Health Beliefs & Preferences Health beliefs, trust in system, preferences, habits, values Cues to Action Previous experience, health awareness campaigns Direct effects Interaction effects Figure 1. Theoretical model of demand for ambulance services. Note: One-sided arrows indicate causal relations. Two-sided arrows suggest that the concepts are inter-related, so that presence or experience of one phenomenon affects the other and vice versa.

3 578 Toloo et al. AMBULANCE USE AND SELF - RATED ILLNESS SERIOUSNESS 4. Previous use of the service (cues to action). This article seeks to fill a gap in knowledge by providing an understanding of patients perspectives toward, and reasons for using, ambulance services in Queensland. This study is part of a larger research program investigating factors and drivers of demand for ambulance and hospital emergency department (ED) services in Queensland, Australia. METHODS Study Design and Population This was a cross-sectional survey. The survey was administered within public hospital EDs, between March and May 2011 (autumn season), while patients were waiting for or were under treatment. All Australian citizens and permanent residents are entitled to universal free access to public hospitals and ambulance services. Prehospital emergency services including response, treatment, and road transport are provided by Queensland Ambulance Services (QAS). In 2010 and 2011, QAS responded to 801,000 service requests with 735,000 patients, of whom 91.8% were transported. 28 Nearly 27% of patients attending major public hospital EDs arrive by ambulance (data not published). Approval to conduct the research project was granted by the Queensland University of Technology s Human Research Ethics Committee (QUT-HREC). Consequent clearances were granted for obtaining and analyzing ED data by Queensland Health Research Ethics Committee (QHREC; HREC/09/QHC/26) and QUT-HREC ( ). The Prince Charles Hospital s Ethics Committee, on behalf of QHREC (HREC/10/ QPCH/98), Mater Health Services Human Research Ethics Committee (1621AC), and QUT-HREC ( ) provided approvals for multisite research and collection of data from patients. Survey Content and Administration We operationalized the hypotheses of the conceptual model into questions and scales to form a structured survey questionnaire. Separate survey versions were developed for adult patients (self-complete) and for children (to be completed by parents/guardians) with wordings adjusted to ensure the appropriateness of language for the survey respondent groups. For the purposes of this article: 1. Perceived severity was measured using three questions with responses scored on a scale of 1 to 10, with 1 representing the lowest score (refer to Data Supplement S1, question 1, available as supporting information in the online version of this paper). 2. Perceptions of ambulance transport or use were addressed by levels of agreement, with eight statements about ambulance services using a five-point Likert type scale (Data Supplement S1, question 2). 3. Reasons for using ambulance was addressed by asking if any of 10 reasons were considered, and to what extent, when they decided to call the service (Data Supplement S1, question 3). Additional reasons could also be given, with the few provided able to be coded back to the initial options. Pilot testing was conducted with 45 adult patients and 21 parents at three of the Brisbane-based sample hospitals during December Questions and items measuring study constructs and concepts were tested for face validity and internal consistency, and adjustments were made accordingly. The interviewers were graduates or final-year undergraduate students of social science, psychology, public health, or nursing and had experience in data collection. Each received 2 days of training. Data collection took place between 08:00 and 22:00 hours on at least two weekdays and one weekend day. The university s Occupational Health and Safety rules required the interviewers to be deployed in pairs and no data collection to be performed overnight. All patients who presented to the ED during data collection periods were eligible to participate. To partly compensate for the inability to collect data overnight, patients still under treatment from overnight were included. Non English-speaking patients were allowed help from companions to interpret for them, and interviewers assisted with completion of questionnaires for patients unable to read or write. Interviewers returned later to patients who were sleeping or in conditions that prevented participation. Patients being transferred, admitted, or discharged or leaving without treatment were given questionnaires with stamped envelopes to return the forms, but were not followed up, as we did not have access to identifiable information. Patients considered dangerous were not included. The parent or guardian s consent was obtained for patients under 18 years old. Sampling. At the time of the survey, 29 major public hospital EDs from across Queensland used the Emergency Department Information System (EDIS), representing 75% of all patients attending the state s public EDs. These were grouped into three geographical regions (major cities, inner regional, and outer regional/ remote) according to Australian Institute of Health and Welfare classification. 29 Thirteen EDs were in major cities, 11 were in inner regional areas, and five were in outer regional and remote locations, receiving 52, 31, and 17% of patients, respectively. Eight EDs were then randomly selected. Four were located in major cities, two in inner regional areas, and two in outer regional or remote areas, receiving 54, 24, and 22% of the patients, respectively. A sample size quota was determined for each ED based on their 2009/2010 presentations. A regression analysis was conducted to detect changes in emergency health services use within regional groupings. The results were entered in Power Sample software 30 (Vanderbilt University, Nashville, TN) with 80% power and alpha of A sample size of 859 was calculated with an intraclass correlation coefficient of and design effect factor of This number was increased to a minimum 900 respondents to account for possible incomplete questionnaires. Data Analysis Data analysis was done using PASW Statistics (IBM, SPSS Inc., Armonk, NY) and MS-Office Excel 2007 (Microsoft Corp., Redmond, WA). An exploratory

4 ACADEMIC EMERGENCY MEDICINE June 2013, Vol. 20, No principal component factor analysis was conducted to extract common themes among items measuring ambulance perceptions. Internal consistency among the extracted factors was measured using Chronbach s alpha. The means procedure and F-tests were used to calculate the differences for dependent interval variables (e.g., age) within subcategories of nominal variables (e.g., sex). Chi-square tests were used to test the associations between categorical variables. Multinomial logistic regression was used to calculate the independent effect of perceptual variables on ambulance use. RESULTS From a total of 1,608 patients who presented to EDs during the data collection shifts, 1,361 (85%) were able to be approached, from whom 911 valid questionnaires were collected (response rate = 67%) consisting of 687 adult and 224 parent questionnaires. The main reasons provided for nonparticipation included being too unwell (n =41), not having time, discharged, or busy with kids (n =17); mental health or drunk (n =10); under age without accompanying parents (n =4); and non-english speaker (n =3). Of the total 911 patients surveyed, 24% (n =223) arrived by ambulance, and 68% (n =619) used their own or public transport. The remaining respondents arrived by other means, such as police and community services, or the question was not answered. The ratio of ambulance to nonambulance transports in the survey closely resembles the statewide figure of 27%. For the purposes of this article, only ambulance and self-transports are included in the analysis. Perceived Severity Table 1 shows that patients arriving by ambulance were significantly more likely to perceive their conditions as serious, urgent, and painful, than self-transports. Perceived seriousness and urgency were strongly correlated (Spearman s rho = 0.79), while pain score was only moderately correlated with the former two questions (0.43 and 0.37, respectively). General Perceptions of Ambulance Service The factor analysis with both unrotated and rotated (Varimax) matrices extracted two factors with Eigenvalues greater than 1. Items a d (Data Supplement S1, question 2) were strongly loaded on the first factor and drew upon the belief that everyone is entitled to use ambulance services regardless of the severity of their conditions. Items e and f were highly loaded on the second factor and reflected the belief that ambulance services were for emergency situations only. The other two items were poorly associated with any of these factors. The Chronbach s test returned high internal consistency (a = 0.75) among items a d ( ambulance for all ), but was weak (a = 0.58) for the items on the second factor. Table 2 shows the overall attitudes and perceptions of the participants towards usage of ambulance services. While overall a majority of the respondents (62%) disagreed with the notion that everybody is entitled to use an ambulance if they feel unwell or cannot afford other means (i.e., ambulance for all), the ambulance arrivals were more inclined to agree with this indicator than self-transports. The majority of patients (87%) agreed that ambulances should be used only in emergency situations. However, fewer (70%) agreed that the use of ambulance in nonemergency circumstances was misuse of the service. Method of arrival was not significantly associated with either of these perceptions. Nearly half of the respondents were of the view that patients arriving by ambulance get a higher priority at the ED, but one-quarter disagreed, with no significant difference in responses between ambulance arrivals and self-transports. Almost half (46%) thought that people would still use the ambulance even if they had to pay for it, with ambulance arrivals significantly more likely to favor this option. Previous Experience Ambulance arrivals were significantly more likely to have used it before. One-third of the ambulance arrivals reported they had used ambulance at least one other time in the 6 months prior to the survey, compared to 13% of the self-transports (p < 0.01). Regression Analysis In a multivariable analysis, perceived seriousness was the only factor to remain significantly associated with ambulance use in the adjusted analyses (Table 3). Compared to self-transports, likelihood of using an ambulance increased by 32% for every unit increase in perceived seriousness. Previous experience using an ambulance in the 6 months prior to the survey was not independently and significantly associated with repeat use within that period. Reasons for Using Ambulance Urgency (87%) and severity (84%) were the most often considered reasons for calling an ambulance, followed by the need for better and safer care (over 70% each; Table 4). Approximately one-third of respondents thought they would receive higher priority in the ED Table 1 Perceived Seriousness, Urgency, and Pain by Mode of Arrival Ambulance Self F (p-value) Mean SD 95% CI N Mean SD 95% CI N Perceived seriousness (<0.01) Perceived urgency (<0.01) Pain score (<0.05)

5 580 Toloo et al. AMBULANCE USE AND SELF - RATED ILLNESS SERIOUSNESS Table 2 Perception of Ambulance Services by Mode of Arrival Arrival Method (%) Item Agreement Level Ambulance Self Chi-square p-value (a-d) Ambulance for all* Strongly agree/agree <0.01 Neutral Disagree/strongly disagree N e) People should call the ambulance only Strongly agree/agree if it s an emergency or urgent situation. Neutral Disagree/strongly disagree f) Using an ambulance for nonemergency conditions is a misuse of the system. g) Patients get a higher priority in the hospital if they arrive by ambulance. h) People would still use an ambulance even if they had to pay an extra fee. *See Data Supplement S1 for details. N Strongly agree/agree Neutral Disagree/strongly disagree N Strongly agree/agree Neutral Disagree/strongly disagree N Strongly agree/agree <0.01 Neutral Disagree/strongly disagree N Table 3 Multinomial Logistic Regression Results for Ambulance Use Factor Values ORs 95% CI p-value Perceived <0.15 seriousness Perceived NS urgency Pain score NS Previous Once* <0.01 ambulance Twice or more* 1.00 use Ambulance Strongly agree/ NS for all agree Neutral NS Disagree/strongly 1.00 Ambulance fee ok disagree Strongly agree/ <0.01 agree Neutral NS Disagree/strongly 1.00 disagree NS = not significant. *For patients who had arrived by ambulance this was included in the calculations. due to their arrival by ambulance. Additionally, nearly one-third of patients stated a lack of car or someone to drive them as reasons for requesting the ambulance. The perception of free transport and unaffordability of alternative transport were each mentioned by less than 20% of respondents as being factors that were considered in their decision processes. Further bivariate tests were conducted to examine whether attitudes varied significantly by perceived indicators of severity, prior use of ambulance, and sociodemographic characteristics. However, due to low variance in most items, the differences were not statistically significant. DISCUSSION To the best of our knowledge, this is the first study to examine patients reasons and perceptions for using ambulances in Australia. We tested patients perceptions of illness severity, opinions of ambulance services, and previous use of ambulance services. The multivariate analysis of these factors showed that only perceived seriousness of the illness was significantly associated, with 32% higher use of an ambulance compared to selftransports. Patients who had arrived by ambulance added extra explanations about their reasons for using the service. Between 84 and 87% of the cases mentioned the severity and urgency of the condition as reasons for calling an ambulance. Our study further adds to previous knowledge 7,24,31 that the principal reason for using an ambulance is the patient s perception of the condition s severity and urgency, combined with the belief that the patient would receive proper care. This finding challenges discussions about inappropriate or unnecessary use of acute health services, including both EDs and ambulances. Studies that examine the patient s decision to seek urgent health care from the perspective of post hoc definitive diagnostic outcomes 12,13,16 23 do not take patient concerns into account. It is the patient who is faced with making the decision to seek assistance, and he or she is not generally in possession of sufficient professional clinical assessment skills to enable fully informed judgments; thus the patient has to call it as they see it. There may also be significant disagreement

6 ACADEMIC EMERGENCY MEDICINE June 2013, Vol. 20, No Table 4 Reasons for Using Ambulance Stated by Patients Who Arrived by Ambulance Item Not Considered (%) Considered to Some Extent (%) Considered to a Great Extent (%) Total (N) The patient needed immediate (urgent) care Patient s condition was too severe to use other means of transport It was safer to come by ambulance The patient required special care during transport Other options were not as suitable as the ambulance for the patient s problem We thought the patient would get higher priority in the hospital if arrived by ambulance There wasn t a car available to drive the patient to the hospital Nobody was there to drive the patient to the hospital Because it did not cost us an extra fee to use ambulance We couldn t afford to pay a taxi to bring the patient to the hospital between patients reasons and physicians views on the appropriateness of patients decisions. 24 The theoretical framework and previous research suggest that various psychological and sociodemographic factors likely affect the attitudes and perceptions of patients and caregivers. Indicators of low social support, such as living alone or being single, have been reported as factors increasing ambulance usage. 10,32 However, the mechanism of this is not well known. A longitudinal study of 4,724 Dutch respondents showed that lower social support was associated with poorer self-assessed health status and unhealthy lifestyle. 33 Having access to a greater pool of material and nonmaterial resources and support may also provide individuals with better control over the situation (self-efficacy), and more options to consider at a time of acute sickness, rather than relying heavily on professionals and ambulance transport. These perceptions and resources can in turn vary according to the individual s sociodemographic circumstances and access to alternative resources. Around 20% of ambulance users stated that they considered the costs of alternative transport as a barrier, and one-third mentioned not having a car as a reason for requesting an ambulance. Material barriers (e.g., not having a car or enough money) to access affordable care are likely to be exacerbated for people with lower levels of social support and create a vicious circle within which low socioeconomic status, poverty, and inequality restrict people s access to care. A recent microsimulation study in Australia found that if unemployed people and those living in regional and remote areas had the same level of access to general practitioners as those in urban areas, there would be an increase of 14% to 20% in general practitioner visits. 34 Part of this need for primary health care is likely to be diverted to public EDs and hence generates a need to use ambulances for some patients. Our study also found information and evidence that can have implications for demand management strategies. First, a considerable number of those who used ambulances did so even though they did not consider their conditions as urgent or severe (12.6 and 15.8%, respectively). Similarly, over one-quarter of the ambulance users used them despite knowing their conditions did not require special care. This may contribute to a proportion of unnecessary workload for ambulance services. Further analyses by underlying factors such as sociodemographic status of these patients may provide better insight into understanding this issue. Second, about half of the respondents believed ambulance arrivals would get higher priority at the ED. Onethird of the ambulance arrivals also confirmed that they considered this to an extent as a reason to use the ambulance. Public education campaigns to emphasize that all patients attending EDs are triaged according to the urgency of their conditions and not by their modes of arrival, as well as consistent practice of triaging ambulance and nonambulance patients in EDs, may help address this. Third, ambulance users were more likely than nonusers to agree that people would still use ambulances if they had to pay for the service. This suggests that the decision to use an ambulance is made on the basis of need and not cost for the majority of cases. However, 18% of ambulance users mentioned they used the service because it did not cost them an extra fee; one in three because they did not have a car and around 18% because they could not afford a taxi. Any future plan to introduce user charges needs to consider the consequences of such a policy on those who cannot access health care without using the ambulance services. LIMITATIONS These findings must be interpreted within the limitations of this study. First, this was a cross-sectional study and some of the questions, including perceived severity items, were measured retrospectively. However, the possibility of recall bias may be limited as data collection took place soon after patient arrival to the hospital. Second, despite measures taken to increase participation, sampling procedures and data collection restrictions may have imposed some selection bias. Although

7 582 Toloo et al. AMBULANCE USE AND SELF - RATED ILLNESS SERIOUSNESS high-urgency patients may have been underrepresented in this study, the main focus of debate is the patients perceived as inappropriate from an emergency health service perspective. There is no question or argument about why patients in high-acuity categories attend an ED or use an ambulance. Understanding the patients views in the lower-acuity and nonurgent categories is essential and addressed by this study. We also did not ask nonambulance users why they did not use ambulances. However, the general attitudes and opinions about ambulance services overall may help provide this information. Finally, the study was conducted within public hospital EDs in Queensland and does not cover communitywide perceptions, nor can it be generalized to other settings. However, the high response rate to our survey of 67%, and similarity between the proportion of ambulance arrivals between our survey (24%) and the ED data (27%), are likely to reduce selection bias and provide a reasonable degree of confidence in the findings. CONCLUSIONS The main reasons for ambulance use in this study relate to perceived seriousness of the illness (severity and urgency) by patients or their care givers. This challenges discussions about inappropriate use of acute health services, including ED and ambulance, based on retrospective analysis by health professionals without considering the patient perspective. This research should be repeated in other states and countries to identify both common patterns and psychological, demographic, and socioeconomic characteristics that result in differences. By understanding those factors we may be better placed to identify strategies that may reduce growth in demand. These studies also need to be conducted among people who decided not to use the ambulance service and among the wider community with a view to determining factors affecting their decision-making processes. The authors acknowledge the assistance and support received during the research process, in particular from Prof. Vivienne Tippett, Ms. Megan Robinson, and Ms. Denielle Bailey (QUT); Mr. Russell Bowles and Mr. David Eeles (QAS); Dr. Kevin Chu and Dr. David Ward (Queensland Health); and emergency departments that participated in the data collection. Jamie Quinn of the Australian Centre for Pre-hospital Research, QAS, read and commented on the paper s various drafts. References 1. Larkin GL, Claassen CA, Pelletier AJ, Camargo CA Jr. National study of ambulance transports to United States emergency departments: importance of mental health problems. Prehosp Disaster Med. 2006; 21: Ohshige K, Tochikubo O. A descriptive study on the trend of ambulance utilization in an aging society, Yokohama, Japan. Yokohama Med Bull. 2003; Australian Bureau of Statistics. Australian Demographic Statistics, Population by Age and Sex, Australian States and Territories. Available at: Page/3201.0Jun%202010?OpenDocument. Accessed Mar 20, Toloo S, FitzGerald G, Aitken P, Ting J, Tippett V, Chu K. Emergency health services: demand and service delivery models. Monograph 1: Literature Review and Activity Trends. Brisbane: Queensland University of Technology, ABS. Australian Social Trends, Cat. No Canberra: Australian Bureau of Statistics; Australian Social Trends. Available at: abs.gov.au/ausstats/abs@.nsf/allprimarymainfeatures/f157bcd66301bdc3ca2577ac00157b0a? opendocument. Accessed Mar 20, Clark MJ, Purdie J, FitzGerald GJ, Bischoff NG, O Rourke PK. Predictors of demand for emergency prehospital care: an Australian study. Prehosp Disaster Med. 1999; 14: Shah MN, Glushak C, Karrison TG, Mulliken R. Predictors of emergency medical services utilization by elders. Acad Emerg Med. 2003; 10: Queensland Ambulance Services. Audit Report. Brisbane: Queensland Ambulance Services; Available at: default.asp. Accessed Mar 20, Lowthian JA, Jolley DJ, Curtis AJ, et al. The challenges of population ageing: accelerating demand for emergency ambulance services by older patients, Med J Aust. 2011; Lowthian J, Cameron PA, Stoelwinder JU, et al. Increasing utilisation of emergency ambulances. Aust Health Rev. 2011; 35: Australian Institute for Primary Care. Factors in Ambulance Demand: Options for Funding and Forecasting. Consolidated Report. Melbourne, Australia: La Trobe University, Zachariah BS. The problem of ambulance misuse: whose problem is it, anyway? Acad Emerg Med. 1999; 6: Richards JR, Ferrall SJ. Inappropriate use of emergency medical services transport: comparison of provider and patient perspectives. Acad Emerg Med. 1999; 6: Billittier AJ, Moscati R, Janicke D, Lerner EB, Seymour J, Olsson D. A multisite survey of factors contributing to medically unnecessary ambulance transports. Acad Emerg Med. 1996;3: Brown E, Sindelar J. The emergent problem of ambulance misuse. Ann Emerg Med. 1993;22: Little GF, Barton D. Inappropriate use of the ambulance service. Eur J Emerg Med. 1998; Palazzo FF, Warner OJ, Harron M, Sadana A. Misuse of the London ambulance services: how much and why? J Accid Emerg Med. 1998; Kawakami C, Ohshige K, Kubota K, Tochikubo O. Influence of socioeconomic factors on medically unnecessary ambulance calls. BMC Health Serv Res. 2007; Ting JY, Chang AM. Path analysis modeling indicates free transport increases ambulance use for minor indications. Prehosp Emerg Care. 2006; 10:

8 ACADEMIC EMERGENCY MEDICINE June 2013, Vol. 20, No Ohshige K. Reduction in ambulance transports during a public awareness campaign for appropriate ambulance use. Acad Emerg Med. 2008; 15: Ohshige K, Kawakami C, Kubota K, Tochikubo O. A contingent valuation study of the appropriate user price for ambulance service. Acad Emerg Med. 2005; 12: Volans AP. Use and abuse of the ambulance service. Prehosp Immed Care. 1998; Snooks H, Wrigley H, George S, Thomas E, Smith H, Glasper A. Appropriateness of use of emergency ambulances. J Accid Emerg Med. 1998; Jacob SL, Jacoby J, Heller M, Stoltzfus J. Patient and physician perspectives on ambulance utilization. Prehosp Emerg Care. 2008; 12: Conner M, Norman P (eds). Predicting Health Behavior: Research and Practice with Social Cognition Models. 2nd ed. New York, NY: Open University Press, Glanz K, Rimer BK, Viswanath K (eds). Health Behavior and Health Education: Theory, Research and Practice. 4th ed. San Francisco, CA: Jossey- Bass, Weinstein ND, Rothman AJ, Sutton SR. Stage theories of health behavior: conceptual and methodological issues. Health Psychol. 1998; 17: Council of Ambulance Authorities Annual Report: The Council of Ambulance Authorities, Available at: Accessed Mar 20, Australian Institute of Health and Welfare. Australian Hospital Statistics Canberra: Australian Institute of Health and Welfare, Available at: aspx?id= Accessed Mar 20, Dupont WD, Plummer WD. PS: Power and Sample Size Calculations. Available at: Accessed Mar 20, Shah SM, Cook DG. Socio-economic determinants of casualty and NHS Direct use. J Public Health. 2008; 30: Snooks H, Porter A, Button L, et al. What factors predict 999 use amongst unscheduled care service users? [abstract]. Emerg Med J. 2009; 26:e Croezen S, J Picavet HS, Haveman-Nies A, Verschuren WM, de Groot LC, van t Veer P. Do positive or negative experiences of social support relate to current and future health? Results from the Doetinchem Cohort Study. BMC Public Health. 2012; 12: e Schofield DJ, Shrestha RN, Callander EJ. Access to general practitioner services amongst underserved Australians: a microsimulation study. Hum Resour Health. 2012; 10:e1. doi / Supporting Information The following supporting information is available in the online version of this paper: Data S1. Appendix: survey items.