Review article: People who present on multiple occasions to emergency departments

REVIEW ARTICLE

Review article: People who present on multipleoccasions to emergency departments emm_1449 532..540

Katherine Nelson,1 Margaret Connor,1 Cynthia Wensley,1 Cheryle Moss,3 Margaret Pack1 andTania Hussey2

1Graduate School of Nursing, Midwifery and Health, Victoria University of Wellington, 2WellingtonHospital Emergency Department, Capital and Coast District Health Board, Wellington, and 3School ofNursing and Midwifery, Faculty of Medicine, Nursing and Health Sciences, Monash University,Melbourne, Victoria, Australia

Abstract

Research about people who present on multiple occasions to the ED began in the 1980s.Despite this, little is known of their journey as patients. Understanding ED use as a journeycan help clinicians improve how they meet the needs of this patient group. Multiple termswere used to identify research on the use of the ED. Papers were included if they had aprimary focus on multiple presentations by the general ED population. Integrative reviewmethods were used to extract findings related to the patients’ journeys. The findingsconfirm a sequence of events and processes that provide an outline of the journey throughthe experience of people who present on multiple occasions. The journey concerns people’sdecisions to present and re-present to the ED, their assessments on arrival, dilemmas oftreatment, outcomes of care and long-term health outcomes. This patient group often havehigh and complex health needs, engage extensively with other health services and havepoor long-term health outcomes. The issue of multiple presentations to the ED is complexand ongoing because of the morbidity of the people concerned, the preference of patients toattend, the purpose of the ED and preparation and role of the personnel and the difficultieswith continuity of care. The provision of care for people who present on multiple occasionscan be improved within the ED and health services generally through a better understand-ing of presentations.

Key words: decision-making, hospital emergency service, utilization.

The journey of people who present to EDs multipletimes is an important concept. Such a concept givescoherence to the experience of how these peopleoften live with complex health circumstances. Thejourney provides a lens on their vulnerability anddemonstrates the importance of clinicians understand-ing the interwoven pattern of patient perceptions,

decision-making and health needs, which result in theprocess of becoming a regular attendee of the ED.

Since the 1980s1,2 researchers have sought to describethe multiple attendances to the ED from different per-spectives and to find solutions to reduce the heavy useof the ED by some people. Given this 30 year timeframewe considered it timely to review and integrate this

Correspondence: Dr Katherine Nelson, Graduate School of Nursing, Midwifery and Health, Victoria University of Wellington, PO Box7625, Newtown, Wellington 6242, New Zealand. Email: [email protected]

Katherine Nelson, PhD, RN, Senior Lecturer; Margaret Connor, PhD, RN, GSNMH, Research Fellow; Cynthia Wensley, MHSc, RN, GSNMH,Lecturer; Cheryle Moss, PhD, RN, Director (Community Engagement), Associate Professor Nursing, Research and Practice Development; MargaretPack, PhD, RSW, Lecturer; Tania Hussey, RN, Emergency Nurse.

doi: 10.1111/j.1742-6723.2011.01449.xEmergency Medicine Australasia (2011) 23, 532–540

© 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

literature. The present paper reports our findings in thehope that it will lead to improvements in the health careand subsequently the health status of people who attendthe ED on multiple occasions.

Although the literature regularly refers to frequentusers, attenders or visitors a decision was made to referto the population as people who have multiple presen-tations (PMPs). The words people and presentationsreflect that this is not about ‘users’ of a service, ‘attend-ers’ or ‘visitors’ but about people going to the ED attimes of need.

Review methods

The search involved Medline, CINAHL, Social ScienceAbstracts, PsychINFO and Cochrane databases usingthe MeSH and subject terms ‘emergency service’,‘Hospital/trends’, ‘hospital/utilization’ and ‘frequentusers’ or ‘heavy users’ or ‘repeat users’ or ‘re-presentation’. The search was limited to humans andthe English language up to April 2010. Papers wereincluded if they had a primary focus on multiple pre-sentations by the general ED population and wereexcluded if they were about specific age groups or diag-nostic populations, opinion pieces or literature reviews.Reference lists were reviewed for additional papers.Each paper was read and data extracted in relation toPMPs. The summarized data were critiqued and inte-grated to establish what was known about peoples’ EDjourney. The majority of the studies were observationaland their quality was accessed using the STROBE cri-teria.3 As this review utilized an integrative processpapers are only referred to with respect to particularcontributions not to the whole contribution theirresearch made.

Findings

The research, conducted in 11 countries, had a state ornational focus, a singular or group of EDs focus orconcerned a subset of PMPs (Table 1). The countrieshave differently funded health systems. Studies weredesigned to describe the population demographicallyand clinically at the time of presentation and in thelonger term, and to understand the nature of attendance.Some EDs in which the research was conducted weregeneralist whereas others excluded some groups.

The STROBE assessment found most studies wererobust and provided descriptions of definitions, data

extraction and analysis techniques. Some articles lackeda detailed description of the ED, missing data were notalways accounted for and funding sources were notnamed. Some studies reported use of more than one EDbut did not take this into account in the analyses andfew reported the impact of known deaths on utilization.

What constituted multiple use varied. Most studieshad a pre-set definition of �4 presentations over12 months, the choice of which was not always justified.A few used a statistical approach9,24 and some relied onnurses to identify people who they considered mademultiple presentations.25,26 More recent studies focusedon �10–12 presentations over 12 months. Studies thatexamined utilization in fewer than 12 months used time-frames between 1 and 6 months, and those looking atthe long-term impact used between 2 and 10 years.

Generally studies reported that with increasing pre-sentations there were increases in mean age, male atten-dance, the acuteness of the triage score and out-of-hoursattendance. A mixed picture of general health, pasthistory, presenting complaints and ED discharge diag-noses was evident. Many studies that collected datadirectly from PMPs excluded those who were seriouslyunwell or hospitalized,1,25,27–29 or used specialist servicesat time of presentation,30,31 or were unable to speakEnglish1,25,28,29 or Swedish.27 Therefore, what is knownfrom patient perspectives is mainly from a subset ofPMPs. For the purposes of the present paper the journeyintegrates the multiple presentations as one narrativecommencing from the decision to go to the ED throughto the life time outcomes of being a PMP.

Person’s decision to attend the ED

Several factors influenced PMPs’ decisions to attend theED. People experienced symptoms that they attemptedto self-manage30 or sought help from primary health-care (PHC) physicians.28 They often interpreted theirsymptoms (most often physical in nature) as life-threatening and therefore saw the ED as the appropriateplace to attend.28,30 Associated with the perceived threatto their life certain PMPs reported high levels of exis-tential distress25,30,32 and others experienced stress.29

Existential distress included fear of dying, loss ofcontrol, powerless in life, isolation and being physicallyunsafe. This distress was sometimes associated withtraumatic events such as past injuries and deaths offamily and friends.30 The need for ED attendance wasalso linked to PMPs’ pre-existing health.33,34 Routes to

Multiple presentations to the ED

533© 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

Table 1. Author, year, country, design and contribution to the ED journey framework

Author Year Country Design (data source) No PMPs Contribution to journey framework• Person’s decision to attend ED• Assessment and triage• Dilemmas of treatment• Outcomes of presentations• Solutions to minimize presentations• Long-term health outcomes

State and national studiesChan and Ovens4 2002 Canada Cohort (ED & GP records) 6 239 Assessment; Treatment; OutcomesJelinek et al.5 2008 Australia Cohort population study (ED records) 22 453 Assessment; Outcomes; SolutionsFuda and Immekus6 2006 USA Observational (all service records) 64 064 Assessment; TreatmentZuckerman and Shen7 2004 USA Survey (telephone) 3 135 Treatment; Long-termHunt et al.8 2006 USA Survey (telephone) 3.5 million Treatment

Specific ED studiesLocker et al.9 2007 UK Descriptive (ED records) 2 764 Decisions; Assessment; OutcomesMurphy et al.10 1999 Ireland Cross-sectional cohort (ED records) NS Decisions; Assessment; OutcomesHansagi et al.11 1991 Sweden Mixed method (all service records) 545 Decisions; Assessment; TreatmentBlank et al.12 2005 USA Descriptive (ED records) 3 667 Decisions; Assessment; TreatmentSalazar et al.13 2005 Spain Descriptive (ED records) 86 Decisions; Assessment; OutcomesMoore et al.14 2009 UK Descriptive (ED records) 2 951 Decisions; Assessment; OutcomesRuger et al.15 2004 USA Cross-sectional (ED records) 962 Assessment; Outcomes; Long-termHelliwell et al.16 2001 NZ Descriptive (ED records) 86 Assessment; OutcomesShiber et al.17 2009 USA Cohort (ED records) 49 Assessment; OutcomesKennedy and Ardagh18 2004 NZ Cohort (all service records) 77 Assessment; Treatment; Outcomes;

Long-termMandelberg et al.19 2000 USA Cross-sectional cohort (ED records) NS Assessment; Long-termGunnarsdottir and Rafnsson20 2005 Iceland Cohort (mortality records) 281 Assessment; Long-termDent et al.21 2003 Australia Descriptive (all service records) 500 Assessment; Outcomes; Solutions;

Long-termHansagi et al.22 2001 Sweden Descriptive (all service records) 2 147 Assessment; TreatmentHuang et al.23 2008 Taiwan Descriptive (all service records) 260 Treatment; Long-term

Multiple presenter subsetsJacoby and Jones1 1982 USA Descriptive (interviews) 30 Decisions; AssessmentAndrén and Rosenqvist2 1985 Sweden Quazi-experimental (ED records, interviews) 232 SolutionsRask et al.24 1998 USA Cohort (prospective; all service records) 351 DecisionsMalone25 1996 USA Ethnography (observation, interviews) 46 Decisions; Assessment; TreatmentPope et al.26 2000 Canada Pre-post design (ED records) 58 SolutionsAndren27 1988 Sweden Case control (questionnaire; service records 248 Assessment; TreatmentLucas and Sanford28 1998 USA Descriptive (face-to-face survey) 134 Decisions; OutcomesSandoval et al.29 2010 USA Comparative (interviews) 69 Decisions; AssessmentOlsson and Hansagi30 2001 Sweden Qualitative (interviews) 10 Assessment; OutcomesSun et al.31 2003 USA Cross-sectional (survey & interviews) 2 333 Decisions; OutcomesMalone32 1998 USA Ethnography (observation, discussion,

interviews)46 Decisions; Assessment; Treatment

Huang et al.33 2003 Taiwan Descriptive (interviews) 200 DecisionsAndren and Rosenqvist34 1987 Sweden Cohort (ED records; interview) 232 DecisionsKne et al.35 1998 USA Cohort (prospective review) 76 Decisions; Assessment; SolutionsMehl-Madrona36 2008 USA Descriptive (interviews) 200 Assessment; TreatmentMilbrett et al.37 2009 USA Descriptive correlational (ED records) 201 Assessment; Treatment; Outcomes;

Long-termShumway et al.38 2008 USA RCT (interviews, economic costs) 252 Assessment; SolutionsHansagi et al.39 1990 Sweden Cohort (ED records, interviews, mortality

register)530 Assessment; Long-term

Byrne et al.40 2003 Ireland Cross-sectional (interviews, GPquestionnaire)

100 Treatment; Outcomes

Hansagi et al.41 2008 Sweden RCT (service records, PHC questionnaires) 1 799 SolutionsOkin et al.42 2000 USA Pre-post design (all service records, cost

data)53 Solutions; Long-term

Lee and Davenport43 2006 USA Quazi-experimental (ED records) 50 SolutionsMichelen et al.44 2006 USA Quazi-experimental (ED records) 711 SolutionsSpillane et al.45 1997 USA RCT (service & intervention records) 70 SolutionsPhillips et al.46 2006 Australia Pre-post design (all service records) 60 Solutions

NS, not significant; PMP, people who have multiple presentations.

K Nelson et al.

534 © 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

the ED varied, with one study finding up to 56% use anambulance9 whereas another reported over 80% madetheir own way.10

Lucas and Sanford28 reported that 58% of 134 PMPsperceived they needed immediate action for their health.Only 27% reported difficulties in attending a PHC facil-ity. The preference for ED might reflect PMPs attitudestowards EDs.11 Some people chose the ED because oftheir established association with the service29 orbecause they had no PHC provider.35 A US study foundthat patients did not want to replace their ED presenta-tion with a clinic appointment.31

People often chose the ED because of access issues.Jacoby and Jones1 termed this choice as one of ‘expedi-ency and immediacy’. This takes account of the ED’sgeographical proximity, free care and the lack of acces-sibility of a PHC physician either geographically,10,29

hours or appointments available,1,10,28 cost28 or havingno telephone.24 One US study31 reported limited finan-cial resources as a driver in using the ED, whereasanother US study12 led the researchers to questionwhether publicly funded insurance poses access prob-lems in PHC. However, in Spain where the cost of PHCis free to patients they seldom used it prior to goingto the ED.13 Research from Ireland10 found increasesin self-referral attendance between 17.00 hours and09.00 hours, and in Britain14 the majority of PMPs pre-sented when general practice was probably closed. Themore visits people made, the more likely they were tocome out of hours.

Assessment and triage

ED data indicated that PMPs presented with a range ofurgency.14,15 Moore et al.14 found the higher the numberof presentations the higher the acuity of the presenta-tions. A few PMPs acknowledged that ED staff mightnot judge their condition as they did.30 Evidence sup-ported their acknowledgment as in several instancesclinicians labelled them as non-urgent,1 inappropriate30

or non-emergency,10 which infers they had minor clini-cal concerns.

Although staff refrained from being ‘familiar’ andtended to judge the worthy from the unworthy of care,patients reported the importance of non-judgementalcaring coupled with clinicians knowing them as reduc-ing their stress.25,30,32 An ED culture of ambivalent atti-tudes towards PMPs triaged as non-urgent and/or notappropriate was sometimes apparent.25 Embedded inthis culture was a view that PMPs interfere with the

‘real’ business of the ED.36 Malone25 found that PMPsrecognized as people, including being called by name,developed relationships that led to a sense of a ‘family-like’ context.

Although assessment in the ED generally led to therecording of symptoms, past history and either a provi-sional or definite diagnosis, the literature indicated amixed picture on whether these activities routinely hap-pened for PMPs. Apart from studies that reported oninterventions to address the issue of multiple presenta-tion, only one study16 reported systems in place denot-ing different procedures for PMPs.

Studies that examined clinical records found gaps inassessments and care. For example, PMPs’ past historywas not always investigated17 and sometimes there werelimited notes on family background and psychosocialhistories.18 Shiber et al.17 found that diagnoses made onone presentation were not always recorded as comorbidi-ties in subsequent presentations and questioned whetherclinicians looked at patients’ previous ED records. Pre-senting symptoms were often associated with a physicalor mental health condition. Symptoms included abdomi-nal, dental or back pain, chest pain, headache, breathingdifficulties, respiratory distress, dizziness and fever.12,13,37

Reasons for presenting had some international variation,for example, sickle cell anaemia is only reported in USstudies,12,19 and cancer-related presentations werecommon in Taiwan where the hospice movement isrelatively new.38 PMPs often presented with differentcomplaints at each visit37 and often had poor generalhealth.38 Research from several countries identifiedPMPs as ill people with multiple diagnoses who experi-enced high morbidity.11,17,18,20,21,25,35,39 Chronic conditionsincluding congestive heart failure, chronic obstructiveairways disease, asthma, diabetes and hypertensionwere often present.

Underlying psychiatric and/or psychological issueshave been associated with PMPs since early publica-tions. Although some research revealed PMPs had pre-viously undiagnosed mental health issues at the time ofusing the ED,9,16,18,21,27,32,35 others reported PMPs hadknown disorders.4,29,36 An Australian study5 exploringcharacteristics of PMPs observed that the presence ofmental health disorders increased with frequency ofattendance. Consistent with this Australian research aNew Zealand study reported that those with psychologi-cal and addiction conditions were more likely to con-tinue to regularly attend over time than those withphysical conditions.18 Furthermore, suicidal tendencieswere sometimes present among people with psychologi-cal and psychiatric conditions.30,39



Many PMPs experienced multiple issues, which werenot necessarily explored or were recorded as ‘non-specific’diagnoses and ill-defined conditions.5,22,36 These included:‘procedure not carried out’, ‘unspecified reason’ and‘unplanned discharges’. Such evidence indicated dilem-mas for both PMPs and ED personnel. A Swedish studyfound that in the absence of a clear medical diagnosis,patients tended to attach their own meaning to theirsymptoms as a way of dealing with their situations.30

Dilemmas of treatment

Although numbers of people with complex issuesreturned time and again to the ED some clinicians did notbelieve they were qualified to deal with them.25,32 Thepresenting symptoms were responded to but there wasoften a failure to capture complexity within the timeperiod of a rapid response system. The identification ofan array of social problems in Malone’s study25 led tostaff conceptualizing the ED as an ‘assembly line’ withthe aim of moving patients through as quickly as pos-sible leading towards discharge. This agenda, accordingto Malone, produced frustration among staff and meantthat clinicians were more likely to ‘miss subtle problems’and ‘focus more on control than care’. In one Swedishstudy psychosocial issues were present in the records butthere was no evidence of any plan to address them.11

Continuity within ED from presentation to presenta-tion was compromised because of the episodic contactPMPs had with a service and because patients mostlysaw different clinicians at each presentation,4 attended adifferent ED6,22 or presented with a different problem.37

PMPs also attended other health services fre-quently4,7,23,27,40 including having multiple hospitaladmissions making continuity of care between the EDand other providers cumbersome. Hunt et al.8 found that65% of PMPs also attended outpatient clinics (OP) �5times. Although it is not known if these were from EDreferrals, it is known that ED physicians were theprimary referrers (61%) to specialists for a group whoseregular PHC provider only referred 9%.4 Althoughinvolving other providers might increase ED use,27 theirinvolvement can also lead to new protocols to improvecare pathways for some groups.12

Outcomes of presentations

Admission to hospital was a common outcome foracutely ill PMPs5,9,17,18,28,31 with some studies reporting

more than 25% of presentations leading to admission. Afew PMPs had died on route to or in ED.5,15 Those whowere less ill were usually discharged home, transferredto other units or to after-hours PHC clinics.5,13 Researchwas unclear about the overall proportion referred backto PHC. Of 1200 randomly selected presentations in aUS study37 people were discharged home on 96% ofoccasions. This might reflect the lower acuity andadmission rate of US ED populations compared withother jurisdictions. It was also reported that as the fre-quency of presentations increased ED clinicians wereless likely to refer a patient to PHC and more likely torefer them to OPs.10 Referrals to specialists, often in OPclinics, were reported.4,13,14,30 Of significance is that notall those who were referred to or had OP appointmentsattended these.21 Research from patient perspectives30

found that those who were diagnosed as suicidal andreferred to a psychiatrist only had short-term engage-ment, which did not result in a change in their need tocontinue to present to the ED. PMPs, especially thosewith low triage scores, sometimes ‘walked out’ of the EDprior to treatment.17,21,37,40 The average stay in the ED byPMPs was reported as high as 5 h.9

Solutions to minimize peoplepresenting multiple times to the ED

The researched solutions were mainly ED cliniciandriven and raise further insights into the dilemmas ofhow best to care for PMPs. Dent et al.21 in specificallyscrutinizing a sample for diversion back to PHC found itdifficult to define who might be best managed in PHC.Compounding the notion of redirecting PMPs to PHCwere the results of an Australian study that found theseriousness and acuteness of presentations were outsideof expected PHC delivery with an inpatient stay beingregularly required.5 It was also recognized that PHC andED clinicians did not necessarily have the skills or timeto attend to the complex needs of PMPs.21

Two main types of interventions were reported: oneconcerned information sharing, the other case manage-ment (CM). Although information sharing between PHCand EDs had no impact in reducing use, clinicians foundit useful.41 CM took many forms and demonstratedreductions,2,26,42–44 no change45 and increases46 in admis-sions and was found to be cost-effective.42 For example,a US study43 that piloted the use of CM to work withPMPs who had presented three times within a monthreported a 7.4% reduction in visits between the pre and

K Nelson et al.


post period. CM also demonstrated changes in PMPshousing,42 access to other health services26,44 and tosocial services.2,26 Pope et al.26 reported the success of apilot targeting PMPs identified as ‘difficult’ to managebecause of their chronic or complex medical condition orbecause of their drug seeking, violent or abusive behav-iour. Okin et al.42 found that those who were no longerhomeless as a result of the intervention had lowerhealth-care costs than those who remained homeless.These solutions, which operated inside and outside ofthe ED, took account of context and clientele andinvolved the cooperation of ED, PHC, other practitionersand social agencies. A study following PMPs reportedreductions in presentations following interventionssuch as haemodialysis and kidney transplants.35

Long-term health outcomes

Studies reported that the population of PMPs changedannually.18,19 Although a proportion continued to makemultiple presentations, others no longer presented or hadfewer presentations and a new group of PMPs emerged.This phenomenon might relate to the higher rates ofmorbidity and deaths at an early age for PMPs.7,18,20,39,42

These samples include people with physical or mentalillness and/or alcohol and drug abuse and numbers diedfrom deteriorations of their conditions. These reportsindicated that PMPs are seriously ill people although notalways in an acute phase of their conditions.

Discussion

Creating the multiple presentations of people to the EDas a journey brings a new view of what is going on forthese people. The reasons for the presentations are mul-tifactorial and require an integrated and coordinatedresponse to deal with a variety of interrelated medicaland psychosocial needs. By highlighting the complexneeds of PMPs the present paper has identified the needfor clinicians across the health sector to rethink theirresponses and where necessary extend their skills tobetter meet the heath needs of these people.

In piecing together constituents of the journey a lackof overall coordination of care is apparent. Althoughsome plans of care were in place little was reportedabout the in-between time of visits and what role, if any,EDs have in ensuring documentation of what is occur-ring during these periods. The literature gave theimpression that each presentation was triaged and

assessed in isolation as opposed to being connected withearlier health events. What was not clear was whetherpeople were representing with ongoing illnesses or forunrelated health events. More research incorporatingthe patient voice about their health needs and issues,particularly about their circumstances and decision-making around attending the ED, would inform under-standing of the process of becoming, remaining and/orshifting from the status of PMP. In the meantime, oneway to rectify this gap in knowledge is having cliniciansdocument the in-between times at each presentation andcreate a narrative of a PMP’s journey.

The use of triage scores to evaluate the appropriatenessof the need to have presented to the ED understates theseriousness and complexity of many PMPs’ health.Although many studies reported large numbers who wereacutely seriously ill, there was a general emphasis on thosewho had non-urgent triage scores. There was evidencethat ED clinicians responded well to acute care needs.However, it was apparent that clinicians often had dif-ficulties with understanding and addressing the complexcircumstances that led PMPs to present to the ED.

Solutions researched have focused on the individualPMPs in setting up ways, usually some sort of CMscheme, to attend to their complex issues with the aim ofreducing presentations to the ED. These solutions wereinstigated by clinicians to address what they believe wasgoing on for the patient group. The assumption inherentin most solutions was that better support, oversight andintegration will fix the problem and shift them into amore independent and functional space. Most of thisresearch was on pilot innovations, which involved smallpopulations and was costly in terms of resources andtime used in setting up. Personnel generally embarked oninnovative projects with enthusiasm and commitment.However, the cost-effectiveness of these solutions in thelonger term where staff turnover has occurred or theintervention has become routine care is not known.Involving health professions especially those who listento patient narratives such as psychiatry liaison andmental health staff, social workers and nurses in theinnovative endeavours could add new insights to thepilot projects and provide evidence and impetus tosupport initiatives becoming mainstream services.

The development of innovative approaches shouldbuild on working with individual PMPs, which has beenfound to be successful in the CM research and alsoembody what Swedish researchers27 recommend as along-term solution that addresses the misery and exis-tential distress that PMPs experience. Group therapy,behaviour therapy and social skill training are identified



by these authors and both ED personnel and GPs areseen as the facilitators of this therapy. For the develop-ment of such innovative approaches to have a chance ofsuccess any clinician taking on the new role would needsupport to develop the new skills. In some situationssuch therapies might be instigated within psychiatricservices. However, getting the patient to attend sessionsoutside ED might be difficult. In EDs not suited tohaving clinicians as therapists it will be important thatclinicians liaise with the appointed therapists andbecome skilled in helping patients use their new skills.Providing these supports would assist in capturingPMPs moments of readiness to learn.

Programmes did not always take account of theimmediacy and expediency issues involved in the deci-sion to attend the ED. Also given that many PMPs werefinancially deprived, socially isolated and unemployedor on social security, there is a need for the wider struc-tural and systemic issues to be considered to improvehealth outcomes for this group.42 The provision of staffeducation on the likely structural and systemic issues inthe community an ED serves would increase awarenessof these influences on health circumstances.

Given the limitations in sampling of several PMPstudies, research that takes a random sample of allPMPs is needed to obtain an appreciation of the widerissues of this group and their preference for attendingthe ED. Multiple representations to EDs will continueuntil we better understand this preference and whyreferrals to address health issues that people bring tothe ED are unsuccessful.

Study limitations

Searching only three databases and reviewing referencelists might have resulted in relevant studies such asdissertations being missed. The decision to excludepapers reporting on specific populations might under-state the information available on PMPs. Given thedocumented differences in definitions and the differ-ences in the health systems involved in the research thisreview has almost certainly analysed data from a het-erogeneous group. Conclusions therefore might not begeneralizable to all PMPs.

Conclusion

This review informs the description of the journeythrough the ED as experienced by PMPs, the clinicians

who attend them and the impacts and consequences ofthis journey. The study indicates PMPs have multiplehealth needs, are often acutely seriously unwell, havehigh health service use and generally poor long-termhealth outcomes. The literature points to several inter-woven threads that indicate that multiple representa-tions to the ED is a complex ongoing issue. Thesethreads include the morbidity of PMPs including theirpsychosocial issues, the purpose of ED and preparationand role of personnel, patient preferences and accessto health care and other resources. There is limitedresearch into solutions and how to embed successfulones into the health system is not well understood. Togain increasing knowledge of the journey through theED future research must address accounts of the patientperspective of living with complex illnesses, in differentcontexts and from different age groups.

Acknowledgements

The research for this article was partially funded by theTertiary Education Commission through a Strategy toAdvance Research (STARS) grant.

Author contributions

KN reviewed articles and was lead author of paper. MCreviewed articles and contributed to main drafting ofpaper. CW contributed to paper development and finalpaper. CM contributed to paper development andfinal paper. MP reviewed articles and contributed tofinal paper. TH contributed to final paper.

Competing interests

None declared.

Accepted 12 May 2011

References

1. Jacoby LE, Jones SL. Factors associated with ED use by ‘repeater’and ‘nonrepeater’ patients. J. Emerg. Nurs. 1982; 8: 243–7.

2. Andrén KG, Rosenqvist U. Heavy users of an emergency depart-ment: psycho-social and medical characteristics, other health carecontacts and the effects of a hospital social worker intervention.Soc. Sci. Med. 1985; 21: 761–70.

3. Vandenbroucke JP, von Elm E, Altman D et al. Strengthening thereporting of observational studies in epidemiology (STROBE):

K Nelson et al.


explanation and elaboration. Ann. Intern. Med. 2007; 147:W-163–W-94.

4. Chan BT, Ovens HJ. Frequent users of emergency departments.Do they also use family physicians’ services? Can. Fam. Physi-cian 2002; 48: 1654–60.

5. Jelinek GA, Jiwa M, Gibson NP, Lynch A. Frequent attenders atemergency departments: a linked-data population study of adultpatients. Med. J. Aust. 2008; 189: 552–6.

6. Fuda KK, Immekus R. Frequent users of Massachusetts emer-gency departments: a statewide analysis. Ann. Emerg. Med.2006; 48: 9–16.

7. Zuckerman S, Shen YC. Characteristics of occasional and fre-quent emergency department users: do insurance coverage andaccess to care matter? Med. Care 2004; 42: 176–82.

8. Hunt KA, Weber EJ, Showstack JA, Colby DC, Callaham ML.Characteristics of frequent users of emergency departments.Ann. Emerg. Med. 2006; 48: 1–8.

9. Locker TE, Baston S, Mason SM, Nicholl J. Defining frequent useof an urban emergency department. Emerg. Med. J. 2007; 24:398–401.

10. Murphy A, Leonard C, Plunkett PK et al. Characteristics ofattenders and their attendences at an urban accident and emer-gency department over a one year period. J. Accid. Emerg. Med.1999; 16: 425–7.

11. Hansagi H, Edhag O, Allebeck P. High consumers of health carein emergency units: how to improve their quality of care. Qual.Assur. Health Care 1991; 3: 51–62.

12. Blank FS, Li H, Henneman PL et al. A descriptive studyof heavy emergency department users at an academicemergency department reveals heavy ED users have betteraccess to care than average users. J. Emerg. Nurs. 2005; 31:139–44.

13. Salazar A, Bardes I, Juan A, Olona N, Sabido M, Corbella X. Highmortality rates from medical problems of frequent emergencydepartment users at a university hospital tertiary care centre.Eur. J. Emerg. Med. 2005; 12: 2–5.

14. Moore L, Deehan A, Seed P, Jones R. Characteristics of frequentattenders in an emergency department: analysis of 1-year atten-dence data. Emerg. Med. J. 2009; 26: 263–7.

15. Ruger JP, Richter CJ, Spitznagel EL, Lewis LM. Analysis of costs,length of stay, and utilization of emergency department servicesby frequent users: implications for health policy. Acad. Emerg.Med. 2004; 11: 1311–17.

16. Helliwell P, Hider P, Ardargh M. Frequent attenders atChristchurch Hospital’s emergency department. N.Z. Med. J.2001; 114: 160–1.

17. Shiber JR, Longley MB, Brewer KL. Hyper-use of the ED. Am. J.Emerg. Med. 2009; 27: 588–94.

18. Kennedy D, Ardagh M. Frequent attenders at Christchurch Hos-pital’s Emergency Department: a 4-year study of attendance pat-terns. N.Z. Med. J. 2004; 117: U871.

19. Mandelberg JH, Kuhn RE, Kohn MA. Epidemiologic analysis ofan urban, public emergency department’s frequent users. Acad.Emerg. Med. 2000; 7: 637–46.

20. Gunnarsdottir OS, Rafnsson V. Mortality of the users of a hos-pital emergency department. Emerg. Med. J. 2006; 23: 269–73.

21. Dent A, Phillips GA, Chenhall AJ, McGregor LR. The heaviestrepeat users of an inner city emergency department arenot general practice patients. Emerg. Med. J. 2003; 15: 322–9.

22. Hansagi H, Olsson M, Sjoberg S, Tomson Y, Goransson S. Fre-quent use of the hospital emergency department is indicative ofhigh use of other health care services. Ann. Emerg. Med. 2001;37: 561–7.

23. Huang J, Weng R, Lai C, Hu J. Exploring medical utilizationpatterns of emergency department users. J. Formos. Med. Assoc.2008; 107: 119–28.

24. Rask KJ, Williams MV, Mc Nagny SA, Parker RM, Baker DW.Ambulatory health care use by patients in a public hospitalemergency department. J. Gen. Intern. Med. 1998; 13: 614–20.

25. Malone RE. Almost ‘like family’: emergency nurses and ‘frequentflyers’. J. Emerg. Nurs. 1996; 22: 176–83.

26. Pope D, Fernandes CM, Bouthillette F, Etherington J. Frequentusers of the emergency department: a program to improve careand reduce visits. CMAJ 2000; 162: 1017–20.

27. Andren KG. A study of the relationship between social network,perceived ill health and utilization of emergency care. Scan. J.Soc. Med. 1988; 16: 87–93.

28. Lucas RH, Sanford SM. An analysis of frequent users of emer-gency care at an urban university hospital. Ann. Emerg. Med.1998; 32: 563–8.

29. Sandoval E, Simth S, Walter J et al. A comparison of frequent andinfrequent visitors to an urban emergency department. J. Emerg.Med. 2007; 9: 1–7.

30. Olsson M, Hansagi H. Repeated use of the emergency depart-ment: qualitative study of the patient’s perspective. Emerg. Med.J. 2001; 18: 430–4.

31. Sun BC, Burstin HR, Brennan TA. Predictors and outcomes offrequent emergency department users. Acad. Emerg. Med. 2003;10: 320–8.

32. Malone RE. Whither the Almshouse? Overutilization and the roleof the emergency department. J. Health Polit. Policy Law 1998; 23:797–832.

33. Huang JA, Tsai WC, Chen YC, Hu WH, Yang DY. Factors asso-ciated with frequent use of emergency services in a medicalcenter. J. Formos. Med. Assoc. 2003; 102: 222–8.

34. Andren KG, Rosenqvist U. Heavy users of an emergency depart-ment – a two year follow-up study. Soc. Sci. Med. 1987; 25:825–31.

35. Kne T, Young R, Spillane L. Frequent ED users: patterns of useover time. Am. J. Emerg. Med. 1998; 16: 648–52.

36. Mehl-Madrona LE. Prevalence of psychiatric diagnoses amongfrequent users of rural emergency medical services. Can. J. RuralMed. 2008; 13: 22–30.

37. Milbrett P, Halm M, Paul S. Characteristics and predictors offrequent utilization of emergency services. J. Emerg. Nurs. 2009;35: 191–8.

38. Shumway M, Boccellari A, O’Brien K, Okin R. Cost-effectivenessof clinical case management for ED frequent users: resultsof a randomized trial. Am. J. Emerg. Med. 2008; 26: 155–64.

39. Hansagi H, Allebeck P, Edhag O, Magnusson G. Frequency ofemergency department attendances as a predictor of mortality:



nine-year follow-up of a population-based cohort. J. Public HealthMed. 1990; 12: 39–44.

40. Byrne M, Murphy AW, Plunkett PK, McGee HM, Murray A,Bury G. Frequent attenders to an emergency department: a studyof primary health care use, medical profile and psychosocialcharacteristics. Ann. Emerg. Med. 2003; 41: 309–18.

41. Hansagi H, Olsson M, Hussain A, Ohlen G. Is informationsharing between the emergency department and primary careuseful to the care of frequent emergency department users? Eur.J. Emerg. Med. 2008; 15: 34–9.

42. Okin RL, Boccellari A, Azocar F et al. The effects of clinical casemanagement on hospital service use among ED frequent users.Am. J. Emerg. Med. 2000; 18: 603–8.

43. Lee KH, Davenport L. Can case management interventionsreduce the number of emergency department visits by frequentusers? Health Care Manag. (Frederick) 2006; 25: 155–9.

44. Michelen W, Martinez J, Lee A, Wheeler DP. Reducing frequentflyer ermergency department visits. J. Health Care Poor Under-served 2006; 17: 59–69.

45. Spillane LL, Lumb EW, Cobaugh DJ, Wilcox SR, Clark JS,Schneider SM. Frequent users of the emergency department: canwe intervene? Acad. Emerg. Med. 1997; 4: 574–80.

46. Phillips GA, Brophy DS, Weiland TJ, Chenhall AJ, Dent AW. Theeffect of multidisciplinary case management on selectedourcomes for frequent attenders at an emergency department.Med. J. Aust. 2006; 184: 602–6.

K Nelson et al.


Documents

Review article: People who present on multiple occasions to emergency departments