UBC Faculty Research and Publications

Commonalities and differences in the implementation of models of care for arthritis: key informant interviews… Cott, Cheryl A; Davis, Aileen M; Badley, Elizabeth M; Wong, Rosalind; Canizares, Mayilee; Li, Linda C; Jones, Allyson; Brooks, Sydney; Ahlwalia, Vandana; Hawker, Gillian; Jaglal, Susan; Landry, Michel; MacKay, Crystal; Mosher, Dianne Aug 19, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12913_2016_Article_1634.pdf [ 512.72kB ]
JSON: 52383-1.0366883.json
JSON-LD: 52383-1.0366883-ld.json
RDF/XML (Pretty): 52383-1.0366883-rdf.xml
RDF/JSON: 52383-1.0366883-rdf.json
Turtle: 52383-1.0366883-turtle.txt
N-Triples: 52383-1.0366883-rdf-ntriples.txt
Original Record: 52383-1.0366883-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessCommonalities and differences in theimplementation of models of care forarthritis: key informant interviews fromCanadaCheryl A. Cott1,2*, Aileen M. Davis1,2,3, Elizabeth M. Badley2,4, Rosalind Wong2, Mayilee Canizares2, Linda C. Li5,Allyson Jones6, Sydney Brooks7, Vandana Ahlwalia8, Gillian Hawker9, Susan Jaglal10, Michel Landry11,Crystal MacKay12 and Dianne Mosher13AbstractBackground: Timely access to effective treatments for arthritis is a priority at national, provincial and regional levelsin Canada due to population aging coupled with limited health human resources. Models of care for arthritis arebeing implemented across the country but mainly in local contexts, not from an evidence-informed policy orframework. The purpose of this study is to examine existing models of care for arthritis in Canada at the local levelin order to identify commonalities and differences in their implementation that could point to importantconsiderations for health policy and service delivery.Methods: Semi-structured key informant interviews were conducted with 70 program managers and/or careproviders in three Canadian provinces identified through purposive and snowball sampling followed by moredetailed examination of 6 models of care (two per province). Interviews were transcribed verbatim and analyzedthematically using a qualitative descriptive approach.Results: Two broad models of care were identified for Total Joint Replacement and Inflammatory Arthritis.Commonalities included lack of complete and appropriate referrals from primary care physicians and lack of healthhuman resources to meet local demands. Strategies included standardized referrals and centralized intake andtriage using non-specialist health care professionals. Differences included the nature of the care and follow-up, therole of the specialist, and location of service delivery.Conclusions: Current models of care are mainly focused on Total Joint Replacement and Inflammatory Arthritis.Given the increasing prevalence of arthritis and that published data report only a small proportion of currentservice delivery is specialist care; provision of timely, appropriate care requires development, implementation andevaluation of models of care across the continuum of care.Keywords: Models of care, Arthritis, Total joint replacementAbbreviations: DMARDs, Disease Modifying AntiRheumatic Drugs; OA, Osteoarthritis; TJR, Total Joint Replacement* Correspondence: cheryl.cott@utoronto.ca1Department of Physical Therapy, Faculty of Medicine, University of Toronto,Toronto, Canada2Arthritis Community Research & Evaluation Unit and Division of Health Careand Outcomes Research, Toronto Western Research Institute, UniversityHealth Network, Toronto, CanadaFull list of author information is available at the end of the article© 2016 Cott et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Cott et al. BMC Health Services Research  (2016) 16:415 DOI 10.1186/s12913-016-1634-9BackgroundThe management of arthritis is increasingly an issue forCanadian jurisdictions due to factors including, but notlimited to, the aging of the population and an increasingprevalence of chronic disease in a time of economic con-straint. The 2005 Summit on Standards for Arthritis Pre-vention and Care identified access to appropriate andtimely care as a right and a priority for people with arth-ritis [1]. Given the prevalence of arthritis (1 in 6Canadians now have some form of arthritis) and that theprevalence is expected to increase by more than 50 % by2020 [2] due to population aging and increasing rates ofobesity coupled with limited health human resources totreat people with arthritis and musculoskeletal condi-tions, access to care issues are likely to continue to dom-inate the health care agenda. The perceived imbalancebetween supply and demand has ignited the policy de-bate in Canada’s primarily publicly funded health systemregarding the best ways in which to provide timely ac-cess to health services for people with arthritis.Health policy and decision makers continue to lookfor alternative ways to optimize access to and delivery ofquality care. Professional organizations (e.g. CanadianOrthopaedic Association, Canadian Rheumatology Asso-ciation, Arthritis Health Professionals Association, Arth-ritis Alliance of Canada etc.) have identified models ofcare for arthritis as a priority and funding agencies suchas the Canadian Institutes of Health Research Instituteof Health Services [3] and Policy Research and TheArthritis Society [4] have stated research priorities in thedevelopment and evaluation of innovative models of carefor arthritis. Given these agendas, we need an under-standing of the landscape of existing models of care forarthritis such that future work can address gaps and op-portunities and models of care can be developed, imple-mented and evaluated to maximize care and outcomesfor people with arthritis.Important policy and clinical activity to enhance ac-cess is already underway across Canada. The identifica-tion of reduction of wait times for hip and kneereplacement as a priority in the 2004 Health Accord [5]and the transfer of federal dollars to the provinces to re-duce wait times created the impetus for the developmentof innovative models of care for arthritis to improve ac-cess and care for this patient group over the past decade.While some models of care for conservative manage-ment of arthritis have existed for decades, many new ap-proaches to improve timely care, particularly for peoplewith early inflammatory disease, also have developed,particularly since the advent of Disease Modifying Anti-Rheumatic Drugs (DMARDs). Yet, timely access to thesetreatments remains a priority at national, provincial, andregional levels, as both policy makers and the arthritishealth care community seek alternative, effective modelsof health care delivery. With the exception of hip andknee total joint replacement (TJR) which has taken aprovincial [6–8] and national approach [9], most modelsof care for arthritis have been implemented in local con-texts [10, 11].The term “models of care” is used with many differentmeanings encompassing condition-specific care to ser-vice delivery [12]. It has been defined as “an evidence-informed policy or framework that outlines the optimalmanner in which condition specific care should be madeavailable and delivered to consumers” [12]. This defin-ition and approach suggests a top-down approach to thedevelopment of models of care yet at the time of this re-search the range of models of care for arthritis inCanada have most often been developed at the locallevel [13]; however, these local contextual factors havenot been examined in-depth.Consideration of local contextual factors in studies ofhealth service delivery can include factors such as popu-lation health needs, geography, provider supply, technol-ogy and local and organizational policies [14]. Inaddition, local contextual factors can be influenced bynational, provincial and local policies, health care systemorganization, and payment and incentive systems [15].Some of these local contextual considerations may notbe amenable to traditional evaluative approaches, yetcould point to important contextual considerations withrespect to health policy and service delivery. The pur-pose of this study is to examine existing models of carefor arthritis in Canada at the local level in order to iden-tify contextual commonalities and differences in theirimplementation.MethodsSettingCanada has federally mandated publicly funded healthcare; however, the implementation of health care is aprovincial responsibility [16]. Further, the Canada HealthAct only stipulates universal coverage of physician andin-hospital health services [17]. Funding for community-based, non-physician services outside of hospitals is de-termined provincially.The study was conducted in three Canadian provinces:British Columbia; Alberta; and, Ontario. Ontario is thelargest province with a population of over thirteen mil-lion, whereas British Columbia and Alberta have moresimilar population sizes (approximately 4 million each).These provinces were chosen as: 1) they had significantactivity related to innovative approaches to arthritis care;2) they represent geo-political differences in delivery ofhealth care; 3) they all have vast areas with low popula-tion density; 4) they have differences in health humanresource availability for provision of arthritis care [18];5) the scope of practice varies (and is evolving) forCott et al. BMC Health Services Research  (2016) 16:415 Page 2 of 12physiotherapists, occupational therapists and pharma-cists who are often involved in arthritis care [19]; and, 6)service availability and coverage by provincial healthplans vary (e.g. Ontario has some community-based re-habilitation services specifically for persons with arthritisthat are paid for via provincial health care plans wherethere is currently no similar service in British Columbiaor Alberta). As such, the models in these provinces rep-resent a diverse base in which models of care for arth-ritis have been developed and implemented across thecountry.Study designWe used a two phase qualitative descriptive approach[20] beginning with: key informant interviews with pro-gram managers and/or care providers of arthritis servicedelivery; followed by more detailed examination of 6models of care (two in each province). The purpose ofthe first phase was to identify the scope of existingmodels of care. The purpose of the second phase was toexplore in more depth the strategies currently being uti-lized in various settings to address issues identified inthe first phase such as lack of complete and appropriatereferrals from primary care providers and lack of healthhuman resources to meet local demand, particularly inrural and remote areas.Type of participantsPhase 1: Key informant interviewsA purposive sample of key informants includedparticipants who represented various types of arthritismodels of care and/or who were known opinion leadersin arthritis care delivery. Recruitment occurredthrough: our research team members; existing contactsincluding polling national participants from ourworkshop Meeting the Challenges of Arthritis: ThinkTank on Extended Roles for Rehabilitation Professional[21] to ensure that we identified programs appropriatefor people with arthritis; and, through the use of asnowball technique whereby key informants who wereinterviewed identified other models and/or individualswho could inform the study. Key informants wereselected to ensure representation of professions,practice sectors, settings and geographic variation.Whenever possible, we interviewed more than onestakeholder from an identified model of care to ensuredepth of information and varying perspectives toconfirm findings. Data collection ceased once we wereno longer identifying any additional types of models.Phase 2: In-depth case studiesSampling for the case studies was done from themodels of care for arthritis identified in Phase 1. Ourinitial goal was to identify two cases in each province,one in a more urban setting and the other rural, for atotal of three rural and three urban. However, in oneprovince (British Columbia), models of care forarthritis in rural/remote areas are highly integratedwith centralized service delivery involving theprovince’s largest city, Vancouver. Therefore we chosetwo rural areas for the case studies as they eachincorporated the urban setting.Co-investigators from the respective provinces as wellas key informants from Phase 1 identified the keyplayers who would be relevant to provide theinformation we required for our in-depth interviews. Inaddition, a second interview was completed with thosekey informants from Phase 1 that held/described havinga decision-making role/position or who identifiedthemselves as key in the development of the model ofcare for arthritis in order to identify important teammembers/key informants in each case study.Data collection and analysisKey informants were interviewed over the telephone bya single interviewer (RW) using a semi-structured inter-view schedule developed in consultation with researchteam members (Additional files 1 and 2). Potential par-ticipants were sent a Consent Information Letter priorto the interview with verbal consent obtained prior toinitiating the telephone interview (Additional files 3 and4). At the beginning of each interview, participants wereasked to identify their job titles and roles, their educationalbackgrounds, their years of experience and the setting ofthe model. They then responded to a general questionabout how care was organized for people with arthritis intheir setting. Probes were used as necessary related to thetarget patient group, health care providers involved incare, and service access. On average, interviews were onehour. Memoing was conducted during the course of andat the end of the interview by the interviewer.Interviews were recorded, transcribed verbatim by aprofessional transcriptionist, and imported into NVivo 9software for analysis using a qualitative descriptive ap-proach [20]. The data were analyzed as follows: two re-search assistants (RAs) and two of the investigators(AMD, CC) independently coded three transcripts thenmet to discuss emerging themes and concepts anddevelop the coding scheme. All subsequent interviewswere independently coded and compared to ensureconsistency in coding between the two RAs. Any dis-crepancies in coding were resolved through discussion.The investigators and RAs met frequently to discussthe unfolding analysis. As the interviews progressed itbecame clear that there were two different sub-groupspresent in the identified models of care, even though insome smaller centres the health care professionals mayoverlap between them. These models of care addressedCott et al. BMC Health Services Research  (2016) 16:415 Page 3 of 12Total Joint Replacement (TJR) and Inflammatory Arth-ritis. Flow charts and team “maps” were developed foreach of the six case studies for TJR and InflammatoryArthritis from which we developed overall summarymaps for TJR and Inflammatory Arthritis. As the ana-lysis progressed, teleconferences were held with co-investigators from each province to ensure accuracy offindings. Research findings were then presented and dis-cussed in teleconferences or face-to-face meetings withthe entire research team. At no time did the researchteam attempt to compare effectiveness or utility of themodels of care either within or between provinces, ra-ther we focused on identifying commonalities and differ-ences in how different provinces addressed commonissues encountered.ResultsSeventy individual key informant interviews were con-ducted in Phase 1 between July 2010 and September2012 (22 in British Columbia; 20 in Alberta, and 28 inOntario). These key informant participants included 67clinicians with some having a director/manager or co-ordinator role and three additional directors who had nohealth care professional certification. Professional repre-sentation included 24 physiotherapists, 5 occupationaltherapists, 10 nurses/nurse practitioners, 1 each frompsychology and social work, 5 primary care physicians,13 rheumatologists, and 8 orthopaedic surgeons. Thekey informants had between 1 and 36 years of experi-ence related to arthritis management and care (average19.4 years). Of the 70 interviews, the key informants rep-resented 40 different locations of care delivery.In Phase 2, twenty-eight participants were inter-viewed in the six case studies (9 in British Columbia,10 in Alberta and 9 in Ontario). These interviewswere conducted between November 2011 and January2013. Of the 28 interviewees: 7 had a provincial pro-gram/health region management role; 2 a health re-gion professional practice role; 10 had an institution/program level management role; and, 9 were careproviders with a triage, coordination, system develop-ment role.The case studies with rural populations varied in size:Thunder Bay, Ontario, 154,067; Prince Rupert, BritishColumbia, 13, 052; Kootenay, British Columbia, 75,000;Edson/Hinton, Alberta, (two different communities butwithin one health region) each with a population of ap-proximately 9,000. London, the urban setting in Ontario,had a population of approximately 425,000 whereas theurban setting in Alberta (Calgary) had a population of1.4 million. All case study locations had a similar overallincidence of arthritis (approximately 20 %) except forthe urban area in Alberta (13 %) [22]. Overall Albertahas a younger population than Ontario or BritishColumbia, probably due to the oil industry, which mayexplain the lower prevalence of arthritis.As the data analysis progressed it became apparentthat although participants were talking about models ofcare for arthritis, they were actually talking about twosub-groups of patients: 1) patients with end stage osteo-arthritis (OA) being considered for TJR; and, 2) patientswith Inflammatory Arthritis. The models of care forthese groups differed, even though they may all be seenin the same setting and sometimes by the same healthcare professionals, depending on local contextual issuessuch as the availability of health care professionals andthe size of the community. We identified a number ofcommon factors that impacted implementation at thelocal level. These common factors affected models ofcare for both TJR and Inflammatory Arthritis but for dif-ferent reasons given the underlying disease trajectory ofthe two conditions.The results are presented as follows: first we describethe traditional models of care for TJR and InflammatoryArthritis; second, we identify commonalities in how thetraditional models of care for arthritis have been modi-fied in different contexts to address local needs; and,third, we identify differences in the models of care withrespect to TJR and Inflammatory Arthritis.Traditional model of care for arthritisWe identified an underlying traditional model of care forarthritis that was the basis upon which the innovationsand changes to service delivery were implemented bythe participants. In this traditional model of care, afterseeking care from a primary care physician, an individualis referred to a medical specialist (e.g. a rheumatologistor orthopaedic surgeon) for diagnosis, consultation andmanagement. The overall care of the patient is thentransferred back to the primary care physician with peri-odic review by the specialist.The traditional model of care for TJR involves the pri-mary care physician sending a referral to the surgeon.The surgeon then sees the patient and, if considered ap-propriate, schedules surgery. All patients are seen by thesurgeon whether or not they ultimately require surgery.It is estimated that using this model of care, only about20 % of the patients who end up seeing a surgeon actu-ally receive surgery [23]. Following surgery patients maybe discharged to subacute, inpatient rehabilitation orhome with or without home care or out-patient physio-therapy. Once discharged from hospital, if there are nopost-operative complications, they are typically followedby the surgeon 1–3 times over the year post-surgery.There are two main phases in the traditional model ofcare for TJR where wait times are an issue: from thetime of the referral from the primary care physician toactually having a surgical consult; and from the time ofCott et al. BMC Health Services Research  (2016) 16:415 Page 4 of 12the surgical consult to actually having the surgery. Theissues of time from surgical consult to surgery were ad-dressed by tackling inefficiencies in the system such asmore efficient use of consults with anesthesiology andinternal medicine, pre-operative home visits to facilitateacute hospital discharge, and better through put in oper-ating rooms (e.g. two operating rooms running simultan-eously with one surgeon). By the time we conducted ourinterviews, most of these issues had already been ad-dressed by the surgical wait time strategy of each prov-ince and the current focus of the models of care was onthe first phase: referral from the primary care physicianto surgical consult.The traditional model of care for Inflammatory Arthritisis similar to that for TJR in that patients are referred byprimary care physicians to the rheumatologist who initi-ates medical management that is continued by the pri-mary care physician with the patient followed by therheumatologist as needed. There are three major areaswhere there are likely delays in access to rheumatologycare: 1) symptom onset and assessment in primary care;2) first visit to a primary care physician and rheumatologyreferral; and, 3) time waiting to see the rheumatologist[24]. Any delay in seeing a rheumatologist is increasinglyan issue due to the need for aggressive early interventionwith DMARDs and biologic agents to improve clinicaloutcomes, functional status and quality of life [24].In contrast to TJR where the process of care culmi-nates in patients accessing surgery in a timely model, inthe traditional model of care for Inflammatory Arthritis,the focus is on getting the patients with the greatest acu-ity to the rheumatologist in a timely manner to initiatean appropriate treatment plan. The rheumatologist pro-vides the diagnosis and institutes appropriate medicalmanagement. Often these models of care are establishedin a disease specific context e.g. early rheumatoid arth-ritis, psoriatic arthritis, scleroderma, ankylosing spondyl-itis, lupus, complex osteoarthritis (OA). Hence, thesemodels of care for arthritis require referral to a rheuma-tologist and determination or confirmation of at least aprovisional diagnosis for access. In most cases, therheumatologist is the entry point to accessing teams ofother arthritis health care professionals.Commonalities in implementation of models of care forarthritisAlthough the overall purpose of the models of care forTJR and Inflammatory Arthritis differ due to differencesin disease trajectory, at the local level, common issuessuch as lack of complete and appropriate referrals fromprimary care physicians and lack of health human re-sources to meet local demand were identified as affect-ing the development and implementation of bothmodels of care.1. Lack of complete and appropriate referralsIn Canada, access to medical specialists generallyrequires a referral from a primary care physician.For both the models of care for TJR andInflammatory Arthritis issues of incomplete orinappropriate referrals from primary care physicianswere identified as having major impacts on theimplementation of efficient service delivery.For TJR, most of the models of care addressed theissue of appropriateness of the surgical referral by: 1)ensuring that the referral from the primary carephysician contained all the appropriate informationand that all the necessary tests, x-rays etc. had beencompleted prior to the patient seeing the surgeon;and 2) determining which of the referred patientswere potentially appropriate surgical candidatesprior to seeing the surgeon. According to oneinformant,“There were patients sitting on surgical consult waitlists that shouldn’t be there, that were far too early.But the GP (General practitioner) just didn’t knowwhat else to do with them.”Standardized referralsMany of the TJR models of care had implemented theuse of standardized referrals to address issues of incom-plete or inappropriate referrals from primary care physi-cians. Not only appropriateness as to whether patientswere surgical candidates but also which surgeon is ap-propriate (not all surgeons perform both hip and kneeTJRs). These standardized referrals are mainly com-pleted by primary care physicians, but some specialistsin Alberta also accept referrals from primary care nursesand physiotherapists.“There is a standardized referral form so that all theinformation is what is required by the people that arescreening the referral…diagnostic tests that arerequired both from a lab and a diagnostic imagingperspective are included.”As with the TJR model of care, the need to standardizeand screen referrals from primary care physicians to rheu-matologists for Inflammatory Arthritis was identified as akey priority. As with TJR, Inflammatory Arthritis paper re-ferrals are screened to ensure referrals are appropriate andinformation is complete for accurate prioritization, how-ever, the information requested on the standardizedrheumatology referral forms differ slightly from that forTJR. The rheumatology referral forms ask about inflam-matory markers, morning stiffness, swollen joints, andother systemic features that would particularly identifyCott et al. BMC Health Services Research  (2016) 16:415 Page 5 of 12those with new onset or inflammatory arthropathies re-quiring urgent consult by the rheumatologist. The goal isto gather adequate information so that when the rheuma-tologist sees the patient a diagnosis can be made andtreatment can be initiated in a timely fashion.Central intake/Pooled referralsIn addition to standardized referrals, many of the modelsof care had implemented central intake/pooled referralsystems whereby all referrals were sent to a central in-take where they were screened/triaged by 1) paperprioritization and/or 2) an in-person assessment.Paper prioritization screeningThe purpose of paper prioritization was to identify theprovisional diagnosis and ensure that all the necessaryinformation had been provided. If so, the referral wasthen sent to the appropriate practitioner, for example anorthopaedic surgeon or a rheumatologist.“If we can screen patients at the very beginning toknow which people are surgical and which are non-surgical then our ultimate goal would be to only sendpotential surgical candidates on to see the surgeon.What that does is it increases the productivity of thesurgeon because they’re really focused on providing thetype of care that is required.”In Inflammatory Arthritis models of care, referralsmay also be sorted to ensure patients are seen in themost appropriate program/clinic and in a timely fashion,so equally distributing workload and avoiding long waittimes. Examples include an early arthritis/interventionclinic, programs for specific diseases such as lupus, andfirst available rheumatology nurse practitioner’s or rheu-matologist’s clinic. Paper triage might be conducted bythe rheumatologist, a nurse, and/or by a physiotherapistwith specialty training. The referrals are reviewed,additional information including diagnostics requested,and prioritized based on urgency level. All referrals thatsuggest Inflammatory Arthritis are seen by therheumatologist.In-person screening assessmentsIn addition to paper prioritization, in-person screeningassessments prior to seeing the specialist may beperformed by a variety of health care professionals de-pending on the local context. For example, ExtendedRole/Extended Scope Practitioners (usually physiothera-pists or occupational therapists) are used in Ontariowhereas in Alberta, the role may be done by a retiredorthopaedic surgeon or a primary care physician with amusculoskeletal specialty. In British Columbia, some pri-mary care physicians have developed enhanced skillssuch that they can provide conservative management ortriage people who are candidates to the orthopaedic sur-geon (e.g. for foot and ankle surgery). In some larger set-tings in-person screening assessments are conducted byinterdisciplinary teams involving physiotherapy, occupa-tional therapy and nursing. Often these Extended Roletherapist or nurse-led clinics are held in parallel with theorthopaedic surgeon’s clinic so that potential surgicalcandidates can be seen by the surgeon on the same day.The value of screening is illustrated by the followingparticipant who was asked what determines whether apatient will see the musculoskeletal screener or theorthopaedic surgeon:“Well it’s availability of both, and it’s reliant onfunding as well…we were given some money to dosome screening which of course dropped our wait listto be seen substantially. And that funding is no longerthere. So what we do is we still continue to screenbecause really it makes our surgeons more efficient.”The purpose of the TJR screening assessments is todetermine if the referred patient is a candidate to beseen by the surgeon for consideration of surgery. If not,they are referred back to their primary care physicianwith instructions to return if their symptoms get worse,or to another, more appropriate health care professional.Patients triaged to conservative management are thosewhose needs are considered non-surgical who may re-quire some other form of treatment such as physiother-apy, weight loss, cortisone injections or bracing, or whomay be deemed too early for surgical intervention.If the patient is non-surgical now because they have somemedical problems…or they have a weight problem, thenwe would provide them with some recommendationsabout what they can do to get in better shape. Or to tryto relieve some of the aches and pains that they have withtheir joints. We would provide that information to thepatient and their family physician so that they knowwhat has been recommended. Then if the patient requiresfurther follow-up then they would come back to a centralclinic where that would be provided or they would becared for back in the community by their familyphysician.A few models of care try to work with these patients todevelop ‘personalized care plans’ with the intent of helpingpatients understand what to do and how to access localresources. For these non-surgical, non-inflammatory dis-ease patients, their ongoing care is now dependent on thefunding and services available in their own community.Although recognized as important for models of care forarthritis, this component is often not well developed.Cott et al. BMC Health Services Research  (2016) 16:415 Page 6 of 12..again I am talking about the surgical arthritispatients, not the patients with inflammatory arthritisor complex osteoarthritis that aren’t felt to be surgicalcandidates. That part of the model has never beenfully developed, although it was in the plans originally.In contrast, within Inflammatory Arthritis models ofcare, the focus of screening is to identify those patientswith the greatest acuity. In some settings in Ontario withExtended Role therapists, those referrals deemed as mostlikely to be non-inflammatory arthritis are triaged toThe Arthritis Society therapist/Advanced Clinician Prac-titioner in Arthritis Care. These therapists would seethese patients, provide an assessment, and develop acare plan eliminating the need for involvement of therheumatologist. In situations where Extended Role ther-apists are available, they are viewed as valuable adjunctsto rheumatology specialist care. As one rheumatologistsaid,“So you can use those individuals as enhancers orextenders of your own physical practice. And theseindividuals are becoming more and more common ina variety of community settings where there aretremendous pressures on demand for those fewerphysicians who are available. And this is a good thingbecause this provides expertise that can triage andallocate relatively scarce resources to the appropriatepractitioner.”2. Lack of Health Human Resources to meet localdemandThere are a number of system and local macroissues that affect access to health human resourcesat the local level. At the system level, there is theoverall shortage of specialists. At the local levelthese shortages are aggravated by geographic andpopulation issues. Canada is a very large country,geographically, with the majority of the populationclustered along the United States border. Althoughthere are many large urban settings in Canada, thereare still many Canadians who live in rural/remoteareas with little access to specialists such asorthopaedic surgeons and rheumatologists. The localsolutions that have evolved to address these issuesfocus on either: 1) getting the patients to thespecialist; or 2) getting the specialist to the patients.As one respondent said,We have a surgeon that travels to …once a month andhe does five hip replacements out there on the day thatwe go. And the patients that need to be seen by thesurgeon are seen in between his surgical cases. And I(Extended Role physiotherapist) do the post-operativefollow-up out there as well. And this spares (the pa-tients) three and a half hours of driving each way for a15 min follow-up appointment. We see anywhere from25 to 40 patients per day.Thus, in smaller communities, services are cyclicaland rely on the amount of time health care profes-sionals have and the availability of resources. Forexample, telemedicine might be useful to mitigatetravel, but health care professionals reported limitedtime to apply for funding for such programs, workout a program, devise policies and procedures, and tocoordinate with the specialist to enable the service tohappen.The province of British Columbia provides an ex-ample of the local context of geography and demo-graphics. British Columbia is a very mountainous,coastal province with a small population scatteredover a variety of towns and islands. As described earl-ier, models of care for arthritis in British Columbiaall involve centralized care in the city of Vancouver.Although there may be various reasons for this, oneis the impact of geography. All flights within theprovince go into and out of Vancouver therefore mak-ing it logical to send patients from smaller communi-ties to Vancouver as they would have to fly thereanyway to connect to other locations. Therefore, spe-cialist services in British Columbia are more central-ized than in the other two provinces.Differences in implementation of models of care forarthritisIt is when we shift to examine the actual care providedin TJR and Inflammatory Arthritis models of care thatdifferences between the two emerge, particularly interms of interventions provided, follow-up, role of thespecialist, involvement of other arthritis health care pro-fessionals and location of service delivery (see Table 1).1. Purpose of Models of Care for Total JointReplacement and Inflammatory ArthritisThe overall purpose of the TJR models of care is toget appropriate surgical candidates to the surgeonfor consult in a timely manner. Models of care forTJR are focused on an acute need with a single “fix”of surgical intervention. It is primarily a linear modelof care with the surgeon and surgery the endpoint ofcare with health care professionals workingsequentially with patients in a multidisciplinaryapproach to prepare patients for surgery. Specialistcare is end-loaded and confined to a time-limitedepisode of care beginning with the surgical consultand ending with post-operative rehabilitation andfollow-up.Cott et al. BMC Health Services Research  (2016) 16:415 Page 7 of 12“…once they’ve had their surgery they are triagedaccording to their specific needs. So the majority ofpatients…go home without any further care… anothergroup go home with home care; another group go homewith outpatient physiotherapy; another group go to ourfit program which is a short-term tune-up program formore complex patients. Then a select, smaller groupwould go to an inpatient rehab facility.”Unlike the TJR models of care, where the patient ac-cesses the specialist (orthopedic surgeon) for a specificlimited episode of care, in rheumatology, given thechronicity of Inflammatory Arthritis, patients need tocontinue to be seen and monitored by the rheumatolo-gist across the trajectory of their condition. Inflamma-tory Arthritis is characterized by episodes of acute flaresrequiring timely medical assessment and intervention.So the practice rosters of rheumatologists can be filledwith patients requiring follow-up and ongoing monitor-ing limiting the number of new patients that can beseen. Also, in contrast with TJR models of care whereaccessing the specialist can be seen as the end point, therheumatologist is the beginning of the InflammatoryArthritis model of care for diagnosis, initiation of med-ical management, and referral to the multidisciplinaryteam. Specialist care in Inflammatory Arthritis is frontloaded and continues indefinitely either with therheumatologist or in partnership with primary care phy-sicians or Extended Role therapists. As a rheumatologistexplains:“So if you have a severe rheumatic disease, you’regoing to be on my books forever. I’m going to see you atleast every 3 months ongoing. But if you have alimited issue, we’ll negotiate discharge back to primarycare. And in most situations we’re in co-care withprimary care.”Once seen by the rheumatologist, patients are typicallyreferred for multidisciplinary assessment and educationby team members such as nursing, physiotherapy andoccupational therapy at minimum, and sometimes adietician, pharmacist and/or social worker. All the In-flammatory Arthritis models of care have a multidiscip-linary, outpatient program that provides assessment,education and self-management. These programs rangefrom a 2 week, daily, intensive, personalized care pro-gram to occasional group education/clinic days. The goalof the majority of these programs is self-managementskill development delivered either within the program orthrough community resources.2. Follow-upFollow-up with TJR patients might be conducted bythe orthopaedic surgeon, family physician, therapistfrom The Arthritis Society, or an Extended Roletherapist. As one orthopaedic surgeon said,“We’re standardizing our follow-up protocol as welland we’re spreading out our follow-ups much morethan we had in the past. So they’re seen two to threetimes within the first year and then at the 2 yearmark, the 5 year mark, 10 year mark, and then, de-pending on how they’re doing beyond that, it’s a littlebit more individualized.”In contrast, follow-up is a key component of the In-flammatory Arthritis model of care. It can be deliveredby a variety of non-specialist health care professionals,in a variety of contexts such as, one on one, in person orin a group setting, or from a distance. Table 2 illustratesthe various types of follow-up provided depending on is-sues such as group size and geographic constraints.Inflammatory Arthritis models of care may include part-nerships between rheumatologists and other practitionerssuch as Extended Role therapists or Registered Nurses.“Because once the diagnosis is made and a plan is putinto place then there’s a fairly defined way of followingand monitoring these patients so that you don’t have tohave necessarily a physician involved at every stage.”In Alberta some rheumatologists have incorporateda Registered Nurse into the rheumatologist’s practice/drug monitoring program or clinic to follow-up pa-tients who are on medications/biologics. For example,Table 2 Types of follow-up in inflammatory models of careOne on one interaction Group interactionIn person Rheumatologist, PT, RN Multi-disciplinary refresher daysDistance Telemedicine, Telephone Internet clinics, chat roomsTable 1 Differences in TJR and inflammatory arthritis models ofcareTJR Inflammatory arthritisTriage Appropriatenessfor surgeryMedical AcuityIntervention Single Episode ofCare (Surgery)Multiple Episodes of Care forongoing medical management,educationFollow-up 1–2 times bysurgeonOngoing for medicalmanagement, educationRole ofspecialist inModel of CareSurgeon is theendpoint of theModel of CareRheumatologist is thebeginning (gatekeeper) of theModel of CareTypes ofteamworkSequential,multidisciplinaryConcurrent, interprofessionalLocation ofservice deliveryInpatient Outpatient/CommunityCott et al. BMC Health Services Research  (2016) 16:415 Page 8 of 12the pharmacovigilance program in Alberta for patientson biologic therapy records and monitors the long-term efficacy and adverse effects as well as measuresthe cost-effectiveness of biologics. The RegisteredNurses are responsible for all the reassessments thatare required to ensure renewal of the patients’ medi-cations. In addition, they have the skills to performintramuscular and intravenous injections and providepatients on such medications with instructions andteaching regarding these injections and/or arrange-ments to receive injections. Registered Nurses weredescribed as being a valuable addition to the practicebecause they take a holistic approach to patient careaddressing psychosocial issues and medication adher-ence problems that patients are experiencing. More-over, the presence of a Registered Nurse can reducethe rheumatologist’s caseload e.g., by seeing stable pa-tients at the 1-year mark instead of a visit with therheumatologist.“Basically what ends up happening is that if apatient’s stable and they need to be seen only every 6months, the rheumatologist will see them at 6 monthsand the pharmacovigilance nurse will see them at ayear, so it really cuts our routine follow-ups in half.”In Ontario, some rheumatologists have partnered withExtended Role therapists. The Extended Role therapistmay: 1) be given a subset of the rheumatologist’s case-load to follow, such as those who have established diag-noses, are stable, and on a chronic disease managementpathway; 2) monitor activity or response to medications(e.g. effectiveness and side effects) to determine whetherchanges in treatment are required; and, 3) assess andrecommend other resources that would help those indi-viduals manage their arthritis. The Extended Role ther-apist may perform these follow-up assessments on dayswhen the specialist is also available and/or in betweenthe specialist’s visit in order to book those requiring achange in treatment regime in an upcoming clinic. Thus,the rheumatologist may not see all patients for a full re-view since the Extended Role therapist follows thesestable patients independently but under supervision ofthe specialist. However, the rheumatologist is always in-volved with medication changes and other duties outsidethe Extended Role therapist’s scope of practice. At thetime of data collection for this study, training of Ex-tended Role therapists was only available in Ontario.DiscussionIn this paper we set out to identify commonalities anddifferences in existing models of care for arthritis inthree Canadian Provinces in order to identify contextualissues in the local implementation of these models.Canada provides a useful case given the combination ofhighly populated urban areas and underpopulated rural/remote areas that are difficult to access and requiremajor travelling of both patients and health care profes-sionals, all within publicly-funded settings.Within the broad umbrella of models of care for arth-ritis we found two distinct models of care – one for TJRand one for Inflammatory Arthritis. Common issues inboth models of care were lack of complete and appropri-ate referrals from primary care physicians and lack ofhealth human resources to meet local demand, particu-larly in rural and remote areas. Local strategies to ad-dress these issues include standardized referrals andcentral/pooled intake, triage utilizing non-specialist arth-ritis health care professionals, and use of telemedicineand travelling clinics to bring specialists and patients to-gether. Differences in the models of care for TJR and In-flammatory Arthritis included the rationale or reason fortriage, the nature of the care or intervention provided(surgical vs medical), the nature and extent of follow-up,the role of the specialist, types of teamwork, and the lo-cation of service delivery (urban vs rural/remote).Although the phrase “model of care for arthritis” isused in the literature, our findings indicate that there isnot a single model of care for arthritis given the differ-ences in disease trajectory and the availability of med-ical/surgical management for the various types ofarthritis. Just the clinical needs of the different types ofarthritis clearly lead to different models of care. Al-though the term “arthritis” is a common part of the Eng-lish and medical vernacular, there is much less publicawareness of the differing kinds of arthritis conditions(e.g. OA and Inflammatory Arthritis) and the differencesin their management. The juxtaposition of these issueswith demographic, geographic and health human re-sources issues, point to the need for health policy that issupportive of flexible, context-dependent, locally drivensolutions.Our findings illustrate how the focus of currentmodels of care for arthritis is on the early medical man-agement of Inflammatory Arthritis and TJR in the latestages of OA. Both represent areas in which there arecurrently accepted effective medical treatments (e.g.DMARDs and TJR) for the management of arthritis. Farless attention has been paid to the development ofmodels of care for early to mid-stage OA, which ac-counts for the majority of cases of “arthritis”. Thisburden-service gap is not unique to Canada but ispresent in most developed nations [25]. The focus onacute Inflammatory Arthritis and late OA is probablydue to the presence of medical technologies (drugs andsurgery) for those two conditions resulting in clear, pre-dictable pathways and targeted funding. In contrast,there is less effective medical management available toCott et al. BMC Health Services Research  (2016) 16:415 Page 9 of 12those with early to mid-stage OA. International modelsof care for OA typically use principles of chronic caremanagement such as care coordination, multidisciplin-ary team interventions and collaborative care planningwith persons with OA [25]. Even though TJR is only partof the overall management of OA, the models of care inthis study are focused on the terminal stages of the con-dition when surgery is indicated. We found few (if any)organized models of care for people with OA who arenot considered appropriate for surgery. Rather, thesepatients are dependent on haphazard availability ofcommunity-based services with unclear funding modelsand challenges in coordination and continuity betweenand amongst service providers. These findings highlightthe care gap for those with non-inflammatory, non-surgical arthritis, who represent the highest numbers ofpeople with arthritis. These findings will be presentedfurther in a later publication.Our findings indicate that reliance on primary careproviders as gatekeepers to specialist care can be prob-lematic given the growing and changing complexity ofhealthcare demand. Considerable time and resourceshave been expended in the existing models of care forarthritis we identified on improving the information re-ceived from primary care physicians in order to ensuretimely and appropriate access to specialist services. Re-search has shown that primary care physicians by andlarge are ill-equipped to diagnose and manage musculo-skeletal conditions such as early to mid-stage OA [24,26]. Yet there are other health care professionals withconsiderable expertise in the assessment and manage-ment of chronic musculoskeletal conditions.Our findings show how local need for health humanresources has led to innovative models of care for arth-ritis that incorporate the skills of other non-specialisthealth care professionals such as Registered Nurses andExtended Role therapists to help stream and triage thereferral process. However, only in Ontario are non-specialist health care professionals working in ExtendedRole roles in rheumatology and TJR settings. Researchhas shown models using extended role practitioners areacceptable to patients [27] and that a higher proportionof patients receive TJR when triaged to see the ortho-paedic surgeon [28–30]. Provincial legislation governsthe scope of practice of health care professionals in eachprovince and while Ontario has committed to ensuringthat health care professionals are working to their max-imum scope of practice and is working with regulatorybodies to amend legislation [19], British Columbia andAlberta have less such activity. Ongoing and projectedworkforce shortages [22, 31–34] will require thathealth care professionals work together in new waysto provide care to a growing number of people witharthritis.The differences in the utilization of non-specialistarthritis health care professionals in the TJR models ofcare as opposed to the Inflammatory Arthritis models ofcare points to how scope of practice and funding candrive the models. In Canada, universal health coverageunder Medicare only includes hospital-based servicesand physician services. Because TJR occurs in hospitalsettings, funding is available for non-specialist healthcare professionals in extended roles. In contrast,management of Inflammatory Arthritis is mainlycommunity-based and therefore funding for non-specialist health care professionals is tied to the rheuma-tologists’ billing, driving up costs and decreasing theavailability of scarce rheumatology resources by ensuringthat all patients must be seen by a specialist.The availability of health human resources with re-spect to specialist care for arthritis is an internationalissue. In a recent review, Badley and Davis [23] identifiedthe following issues as key to the ability to provide ne-cessary specialist care for arthritis both now and in thefuture: 1) inadequate availability of rheumatologists andorthopaedic surgeons to meet needs for care of projectednumbers of people with arthritis; and 2) the focus ofcurrent specialist care on Inflammatory Arthritis andTJR ignores the majority of the population with arthritiscreating a major gap in care delivery. Models of care forarthritis are required that maximize the appropriate useof specialists and other health care professionals with ex-tended skills and training.There are limitations to this paper that need to be ac-knowledged. It should be noted that this research wasintended to identify existing models of care for arthritisin three provinces, not every individual model of carefor arthritis in each province. As such, there may beexisting models of care for arthritis that we have notidentified that may address some of the issues identifiedabove. Some might argue that we did not find differentmodels of care for arthritis; rather, we found differentservice delivery locations that have modified the trad-itional primary care physician-specialist model of care.As such, our paper describes modifications to the exist-ing, traditional models of care, rather than uniquemodels of care for arthritis. This argument highlightsthe challenges with the term “models of care” that havebeen identified elsewhere [12]. Issues of terminologynotwithstanding, our findings demonstrate commonal-ities and differences in the ways that different locationsof service delivery have modified traditional models ofcare for persons with arthritis that have implications forhealth policy and service delivery.ConclusionsWe have identified numerous existing models of care inCanada for people with arthritis, the majority of whichCott et al. BMC Health Services Research  (2016) 16:415 Page 10 of 12are focused on TJR and Inflammatory Arthritis, withfew, if any, structured programs of care for persons withnon-inflammatory, non-surgical arthritis. There con-tinues to be barriers to receiving timely care and caregaps for some patient groups, particularly those forwhich there are not well-developed medical interven-tions with clear funding paths. Future work in develop-ing and implementing models of care for arthritis needsto consider the overall population with arthritis, the con-tinuum of care, and utilize the best components of theseoften locally-developed models to provide coordinated,equitable and effective care by the available professionaland community resources.Additional filesAdditional file 1: Semi-structured intial Interview Guides. (DOC 44 kb)Additional file 2: Semi-structured follow-up Interview Guides. (DOC 48 kb)Additional file 3: Consent Information Letters. (DOC 326 kb)Additional file 4: Scripts for obtaining consent. (DOC 45 kb)AcknowledgementsThe authors acknowledge the contributions of the other members of theModels of Care in Arthritis Team: Cy Frank, Sherry Barr, Louise Bergeron,Richard Birtwhistle, Jenny Lease.FundingCanadian Institutes of Health Research, Emerging Team Grant, Dec 2008 toNov 2013.Availability of data and materialsThe Qualitative data (interview transcripts) are not publicly available forreasons of confidentiality. Individual sites and or individuals might beidentifiable from the total story in the transcripts despite the redaction ofnames of individuals and sites.Authors’ contributionsCC, AM, and RW made substantial contributions to conception and design,acquisition of data, and analysis and interpretation of data as well as draftingand revising the manuscript. EB, MC, LL, AJ, SB, VA, GH, SJ, ML, CM and DMactively participated through teleconferences in the conception and designof the study and analysis of the data. All authors reviewed the manuscriptand provided revisions at various stages, have given final approval of theversion to be published, and agree to be accountable for all aspects of thework.Competing interestsThe authors declare that they have no competing interests.Consent for publicationNot Applicable.Ethics approval and consent to participateThis study was approved on June 28, 2010 by the University Health NetworkResearch Ethics Board in Toronto, Ontario, Canada (Protocol #MOCETG92253). Potential participants were sent a Consent InformationLetter. Verbal consent was obtained at the beginning of the telephoneinterview.Author details1Department of Physical Therapy, Faculty of Medicine, University of Toronto,Toronto, Canada. 2Arthritis Community Research & Evaluation Unit andDivision of Health Care and Outcomes Research, Toronto Western ResearchInstitute, University Health Network, Toronto, Canada. 3Institutes of HealthPolicy, Management and Evaluation and Rehabilitation Science, Faculty ofMedicine, University of Toronto, Toronto, Canada. 4Dalla Lana School ofPublic Health, Faculty of Medicine, University of Toronto, Toronto, Canada.5Department of Physical Therapy, Department of Medicine, University ofBritish Columbia and Arthritis Centre of Canada, Vancouver, BC, Canada.6Department of Physical Therapy, Faculty of Rehabilitation Medicine andSchool of Public Health, University of Alberta, Edmonton, AB, Canada.7Ontario Division of The Arthritis Society, Toronto, ON, Canada. 8OntarioRheumatology Association, Toronto, Canada. 9Women’s College Hospital andthe University of Toronto, Toronto, ON, Canada. 10Department of PhysicalTherapy, University of Toronto, Toronto, ON, Canada. 11Doctor of PhysicalTherapy Division, Department of Community and Family Medicine, DukeUniversity Medical Centre, Durham, NC, USA. 12Health Care and OutcomesResearch, Toronto Western Research Institute, University of Toronto, Toronto,ON, Canada. 13Arthritis Alliance Canada, Vancouver, BC, Canada.Received: 6 May 2015 Accepted: 4 August 2016References1. Alliance for the Canadian Arthritis Program. Arthritis isn’t a big deal……until you get it. Ask 4 million Canadians. Ottawa, Ontario, Canada: Reportfrom the Summit on Standards for Arthritis Prevention and Care; 2005.2. Badley EM, Wang PP. Arthritis and the aging population: projections ofarthritis prevalence in Canada, 1991–2031. J Rheumatol. 1998;25:138–44.3. Canadian Institutes of Health Research. Institute of Health Services andPolicy Research (IHSPR). http://www.cihr-irsc.gc.ca. Accessed 17 Aug 2016.4. The Arthritis Society. http://www.arthritis.ca.5. Health Canada. First Minister’s Meeting on the Future of Health Care 2004.A 10-year plan to strengthen health care. http://healthycanadians.gc.ca/health-system-systeme-sante/cards-cartes/collaboration/2004-meetingracontre-eng.php.6. OsteoArthritis Service Integration System (OASIS). Vancouver Coastal Health.http://oasis.vch.ca. Accessed 17 Aug 2016.7. Alberta Bone and Joint Institute. http://albertaboneandjoint.com. Accessed17 Aug 2016.8. Ontario Bone and Joint Health Network. http://boneandjointcanada.com.Accessed 17 Aug 2016.9. Bone and Joint Decade Canada. http://cihr.irsc.gc.ca/e/27032.html.10. MacKay C, Devitt R, Soever L, Badley EM. An exploration of comprehensiveinterdisciplinary models for arthritis. Working Report 2005–03. http://www.acreu.ca/pdf/pub5/05-03.pdf.11. Women’s College Hospital. Multidisciplinary Osteoarthritis Program.http://womenscollegehospital.ca/programs-and-services/centre-for-osteoporosis-and-bone-health. Accessed 17 Aug 2016.12. Briggs AM, Towler SCB, Speerin R, March LM. Models of care formusculoskeletal health in Australia: now more than ever to drive evidenceinto health policy and practice. Aust Health Rev. 2014;38:401–5.13. MacKay C, Veinot P, Badley EM. Characteristics of evolving models of carefor arthritis: a key informant study. BMC Health Serv Res. 2008;8:147.14. Mulvale G, Bourgeault IL. Finding the right mix: how do contextual factorsaffect collaborative mental health care in Ontario? Can Pub Pol’y. 2007;33(special supplement 1):S49–64.15. Stange KC, Glasgow RE. Considering and Reporting Important ContextualFactors in Research on the Patient-Centered Medical Home. Rockville, MD:Agency for Healthcare Research and Quality; 2013. ARHQ Publication No.13-0045-EF.16. Deber R, Gamble B. “What’s in, What’s Out”: stakeholders’ views about theboundaries of medicare. Healthc Q. 2004;7:4. doi:10.12927/hcq.2004.17235.17. Government of Canada. Canada Health Act, Bill C-3. 1984. Statutes ofCanada, 1984, 32–33 Elizabeth II (R.SZ.C. 1985, c. 6, R.S.C. 1989, c. C-6).18. Canadian Institute for Health Information (CIHI). Physiotherapists in Canada.2008. http://www.secure.cihi.ca/free_products/PT_report_2008_en.pdf.Accessed 17 Aug 2016.19. Davis AM, Cott C, Landry MD, Jones A, Li L, Lineker S, et al. Care for Peoplewith Arthritis: Health Human Resources. Models of Care in Arthritis (MOCA).Working Paper (10–03), 2010. http://www.modelsofcare.ca/resources.html#workingpaper_2010.20. Sandelowski M. Whatever happened to qualitative description? Res NursHealth. 2000;23:334–40.Cott et al. BMC Health Services Research  (2016) 16:415 Page 11 of 1221. MacKay C. Meeting the challenges of arthritis: Think tank on extended rolesfor rehabilitation professionals. Arthritis Community Research & EvaluationUnit (ACREU). Working Report 2007. http://www.modelsofcare.ca/resources.html#workingpaper_2007.22. Canadian Community Health Survey, Statistics Canada 2009/2010.23. Badley EM, Canizares M, Mahomed N, Veinot P, Davis AM. Provision oforthopaedic workforce and implications for access to orthopaedic servicesin Ontario. J Bone Joint Surg. 2011;93:863–70.24. Badley EM, Davis AM. Meeting the challenge of the ageing of thepopulation: issues in access to specialist care for arthritis. Best Pract Res ClinRheumatol. 2012;26:599–609.25. Speerin R, Slater H, Li L, Moore K, Chan M, Dreinhofer K, Ebeling PR, WillcockS, Briggs AM. Moving from evidence to practice: models of care for theprevention and management of musculoskeletal conditions. Best Pract ResClin Rheumatol. 2014;28:479–515.26. Glazier RH, Dalby DM, Badley EM, Hawker GA, Bell MJ, Buchbinder R, LinekerSC. Management of common musculoskeletal problems: a survey of Ontarioprimary care physicians. CMAJ. 1998;158:1037–43.27. Baker R, Sanderson-Mann J, Longworth S, Cox R, Gillies C. Randomisedcontrolled trial to compare GP-run orthopaedic clinics based in hospitaloutpatient departments and general practices. Br J Gen Pract. 2005;55:912–7.28. Aiken AB, Harrison MM, Atkinson M, Hope J. Easing the burden for jointreplacement wait times: the role of the expanded practice physiotherapist.Healthc Q. 2008;11:62–6.29. Aiken AB, Harrison MM, Hope J. Role of the advanced practicephysiotherapist in decreasing surgical wait times. Healthc Q. 2009;12:80–3.30. MacKay C, Davis AM, Mahomed N, Badley EM. Expanding roles inorthopaedic care: a comparison of physiotherapist and orthopaedic surgeonrecommendations for triage. J Eval Clin Pract. 2009;15:178–83.31. Shipton D, Badley EM, Mahomed NN. Critical shortage of orthopaedicservices in Ontario, Canada. J Bone Joint Surg Am. 2003;85:1710–5.32. Kur J. Rheumatologist demographics in British Columbia: a looming crisis.BC Medical Journal. 2011;53:128–31.33. Badley EM, Veinot P, Ansari H, MacKay C. Survey of Rheumatologists inOntario. Working Report 2008–03. http://www.acreu.ca/pdf/pub5/08-03.pdf2008.34. Iorio R, Robb WJ, Healy WL, Berry DJ, Hozack WJ, Kyle RF, et al. Orthopaedicsurgeon workforce and volume assessment for total hip and kneereplacement in the United States: preparing for an epidemic. J Bone JointSurg Am. 2008;90:1598–605.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Cott et al. BMC Health Services Research  (2016) 16:415 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items