UBC Faculty Research and Publications

The Older Persons’ Transitions in Care (OPTIC) study: pilot testing of the transition tracking tool Reid, Robert C; Cummings, Garnet E; Cooper, Sarah L; Abel, Stephanie L; Bissell, Laura J; Estabrooks, Carole A; Rowe, Brian H; Wagg, Adrian; Norton, Peter G; Ertel, Mike; Cummings, Greta G Dec 14, 2013

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

Item Metadata


52383-12913_2013_Article_2934.pdf [ 281.59kB ]
JSON: 52383-1.0221297.json
JSON-LD: 52383-1.0221297-ld.json
RDF/XML (Pretty): 52383-1.0221297-rdf.xml
RDF/JSON: 52383-1.0221297-rdf.json
Turtle: 52383-1.0221297-turtle.txt
N-Triples: 52383-1.0221297-rdf-ntriples.txt
Original Record: 52383-1.0221297-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessThe Older Persons’ Transitions in Care (OPTIC)study: pilot testing of the transition tracking toolRobert Colin Reid1, Garnet E Cummings2, Sarah L Cooper3, Stephanie L Abel3, Laura J Bissell1,Carole A Estabrooks3, Brian H Rowe2, Adrian Wagg4, Peter G Norton5, Mike Ertel6 and Greta G Cummings3*AbstractBackground: OPTIC is a mixed method Partnership for Health System Improvement (http://www.cihr-irsc.gc.ca/e/34348.html) study focused on improving care for nursing home (NH) residents who are transferred to and fromemergency departments (EDs) via emergency medical services (EMS). In the pilot study we tested feasibility ofconcurrently collecting individual resident data during transitions across settings using the Transition TrackingTool (T3).Methods: The pilot study tracked 54 residents transferred from NHs to one of two EDs in two western Canadianprovinces over a three month period. The T3 is an electronic data collection tool developed for this study to recorddata relevant to describing and determining success of transitions in care. It comprises 800+ data elementsincluding resident characteristics, reasons and precipitating factors for transfer, advance directives, familyinvolvement, healthcare services provided, disposition decisions, and dates/times and timing.Results: Residents were elderly (mean age = 87.1 years) and the majority were female (61.8%). Feasibility ofcollecting data from multiple sources across two research sites was established. We identified resources andrequirements to access and retrieve specific data elements in various settings to manage data collection processesand allocate research staff resources. We present preliminary data from NH, EMS, and ED settings.Conclusions: While most research in this area has focused on a unidirectional process of patient progression fromone care setting to another, this study established feasibility of collecting detailed data from beginning to end of atransition across multiple settings and in multiple directions.Keywords: Transfers, Transitional care, Transition tracking, Nursing home, Emergency departmentThe Older Persons’ Transitions in Care (OPTIC)study: pilot test resultsBackgroundDespite the growing number of studies addressing prob-lems in caring for older adults transferred between in-stitutions at times of urgent need, progress in qualityimprovement during transitions has been hindered bychallenges in measurement and attribution [1]. Currentstudies that assess outcomes of transitions of care havefocused on patient readmission rates [2-4], adverseevents [5,6] and the Care Transition Measure [7,8].These outcomes do not address the complexity of tran-sitions as they do not evaluate the entire transitionprocess; rather, they evaluate the transition from onecare setting to another. None of these studies evaluatedwhether a successful outcome was achieved in the transi-tion process from the perspective of residents or multiplestakeholders.In a recent report from the Canadian Institute forHealth Information [9], 10% of all seniors’ (75+ yearsof age) admissions to acute care were nursing home(NH) residents, who overall had the longest waits foradmission and highest lengths of stay and levels of re-admission after discharge, making these admissionsresource intensive. Much of the knowledge about useof emergency departments (EDs) by NH residents,* Correspondence: gretac@ualberta.ca3Faculty of Nursing, University of Alberta, 5-110 Edmonton Clinical HealthAcademy, 11405 87 Avenue, Edmonton, Alberta T6G 1C9, CanadaFull list of author information is available at the end of the article© 2013 Reid et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Reid et al. BMC Health Services Research 2013, 13:515http://www.biomedcentral.com/1472-6963/13/515however, arises from retrospective record reviews anda focus on effects of residents on the ED [10]. NorthAmerican studies have reported a yearly transition in-cidence ranging from 23-60% from NH to ED [11-13].A Canadian study that reported a 60% transition ratefound of those 60%, 30% were admitted to hospital[12]. While the majority of NH residents sent to hos-pital for medical care return to their NHs, manyexperience multiple transfers among care settings andproviders [14]. For example, Callahan et al. [15] foundthat ‘compound’ transitions occurred in 20% of transi-tions for dementia patients, each of which presented anew risk for communication errors, duplication of ser-vices, medical errors, and provision of care in conflictwith the individual’s or family’s goals of care. Indeed,more generally transfers among the NH population arefraught with errors, inefficiency, suboptimal care andunmet care needs [16-19]. Therefore, studies are neededwhere a transition can be evaluated as a continuousprocess, corrected through system change, and includesperspectives of sending, interim and receiving pro-viders, residents, and their families [17,20]. The deve-lopment and use of well-designed tools to trackprocesses, events and communications throughout atransition will provide data and advance understandingto make recommendations for change to improveoutcomes and quality.The OPTIC study is a three-year observational studyfunded by a Canadian Institutes of Health Research(CIHR) Partnerships for Health System Improvement(PHSI) grant (http://www.cihr-irsc.gc.ca/e/34348.html).This funding model requires participation and engage-ment by researchers and healthcare decision-makersfrom the genesis of ideas driving the project, to funding,and knowledge utilization to effect system change. Thisstudy examines the care that residents (aged 65 and older)receive when transferred from NHs via an emergency callto emergency medical services (EMS), to EDs and back.The objectives of the overall OPTIC study are to analyzeall transfers between NHs and two EDs in two citiesover a one-year period, to improve care, minimize com-plications, reduce stress on residents, families, staff andresources, improve management of resident transfers,and to develop a tool to measure transition success. Inorder to analyze transitions, the OPTIC team devel-oped the Transition Tracking Tool (T3) [21], whichfacilitates detailed concurrent tracking of case-relateddata for transitions by individual NH residents begin-ning with the decision to transfer from the NH to theED and ending with the resident’s return to the NH ordeath. The purpose of the OPTIC pilot study was totest feasibility of concurrently collecting individualresident data during transitions across these threesettings using the T3. We had four process objectives(#1-4) and one outcome objective (#5) to assess thefeasibility of:1. Recruiting facilities/service providers in each ofthree study settings;2. Enrolling participants;3. Accessing and extracting data elements in eachsetting from patient care records;4. Determining necessary revisions to the T3 tool, and5. Describing the sample of transitionsMethodsThe OPTIC study protocol has previously been reported[21]. It includes the development of the T3 by OPTICresearchers and decision-makers, the OPTIC conceptualframework of the transition tracking process across threecare settings (NH to ED via EMS and then return to NHvia inter-facility transport services) and definitions of keyterms in the T3. This transition process mapping fromNH to ED was done to guide the study design, data collec-tion, and analyses [21].Briefly, the overall OPTIC study has three phases occur-ring over 42 months (November 2009 to April 2013). InPhase 1, qualitative data were collected from three stake-holder groups (residents and families, frontline healthcareproviders, and managers/administrators) from three set-tings in each province (n = 71 participants) [22]. Our in-tent was to develop indicators of successful transitions[21]. In Phase 2, we undertook a one-year data collec-tion phase using multiple tools including the T3 with aminimum target of 400 residents experiencing caretransitions. This phase began with a three-month pilotof the T3 reported here. In Phase 3, we will completeanalysis, interpretation, and knowledge translation ofthe study.SettingThe OPTIC pilot study was conducted in Kelowna,British Columbia (BC), and Edmonton, Alberta (AB).Kelowna’s 2011 population was 117,312, which repre-sents a percentage increase of 9.6% since 2006 which ishigher than the national average (5.9%) for this timeperiod [23]. Kelowna’s population is among the oldestin the country with 19.1% aged 65+, compared to therelatively young population of Edmonton (11.7%) andCanada as a whole at 14.8% in 2011 [23]. Organizationof healthcare services also differs between regions; 13NHs served by a single ED in Kelowna, and 37 NHsserved by seven EDs in Edmonton (see detailed descrip-tion in [22]). These contrasts allowed for assessment offeasibility of collecting pilot data using the T3 in differenthealthcare settings and populations.Reid et al. BMC Health Services Research 2013, 13:515 Page 2 of 11http://www.biomedcentral.com/1472-6963/13/515SamplePurposive, convenience samples were drawn from NHfacilities.Facility recruitmentIn AB, all 37 NHs, one of the seven EDs, EMS and aninter-facility transfer service (IFTS) were approached toparticipate. Twenty five NHs agreed to participate inthe OPTIC pilot, of which seven were voluntary, eightwere public, and 10 were private. The AB ED, EMS andIFTS were publically owned and operated. In BC, all 13NHs, ED, BC Ambulance Service (BCAS) and IFTSwere approached to participate. Of the 12 participatingNHs in BC, four were publically owned and operated,seven were private and one was voluntary. The BC IFTSwas privately operated, while the ED and BCAS werepublically owned and operated.Resident enrolmentAll residents aged 65 and older transferred via EMS fromparticipating NHs to a participating ED were eligible forinclusion. We targeted the first 25 complete cases in eachprovince or all cases completed between April to June,2011 (whichever came first) to assess the feasibility of datacollection tools and procedures.EthicsA majority of Canadian NH residents (approximately 60%[9]) have some level of cognitive impairment, renderingmany incapable of consenting to participate on their ownbehalf [24]. Consistent with Tri-Council policy, a waiverof consent was obtained to enroll all NH residents ex-periencing emergency transfers [21]. In BC, a full waiverof consent was obtained from the regional health au-thority for transferred residents in nine of the 12 par-ticipating NHs. For the remaining three NHs, whilewaiver of consent was granted to collect data at thehospital, operational approval required that residents(with a Cognitive Performance Scale score of 0–2) ortheir family members (if the resident had a CognitivePerformance Scale ≥3) [25] provide written consentprior to research staff accessing their NH care record.In AB, all participating NHs granted full waiver of con-sent. Ethics approval was obtained from the Universityof Alberta Health Research Ethics Board (HREB B:Pro00010666; Pro00017240) for AB and Interior HealthResearch Office and Research Ethics (UBCO BREB:2010–017) as well as the University of British ColumbiaOkanagan Behavioural Research Ethics Board (UBCOBREB: H10-00127) for BC.MeasureThe T3 is an electronic data collection tool created to ob-tain case-related data about residents to track processes[21], events and communications among healthcare pro-viders [26] during their transition. The T3 is comprised ofthe following elements [21]:NH: Demographic and medical, reasons for transfer,decision and timing of transfer, documentation thataccompanied the resident during transition, andassistive technologies and devices (ATDs) (e.g., eyeglasses, cane, hearing aids);EMS: Canadian Triage and Acuity Scale (CTAS) scores[27] documentation received from NH, and preparedor received for ED, and timing information duringtransfer (notification and actual transfer times, arrivalat ED);ED: Timing information, from arrival at ED toassessment by nurse and doctor, consultation,diagnostic tests, chief complaint(s), reason(s) foradmission, and overall length of ED stay (sub-dividedinto admission to inpatient bed and discharged fromED). We did not record all details during inpatientstays or for those transferred to a non-NH setting.However we did continue to track individuals oncedischarged from the inpatient unit.Disposition: Resident’s location following transfer to ED(admission to inpatient care, return to original NH,transfer to another NH, or death);Discharge from ED/Inpatient IFTS/EMS: Quality ofcommunication between ED/Inpatient and IFTS,documentation sent during transfer, times ofnotification and actual transfer;Return to NH: Medical follow-up information,ED/Inpatient data, documentation sent during transferfrom ED/Inpatient via IFTS/EMS, update of resident’smedical list and patient care recommendations, andclinical assessment.Data collection procedureFor eligible residents, OPTIC staff collected transitioninformation via on-site access to records (paper-basedhealth records, electronic health records and patient careplans) at NHs and EDs using an electronic T3 version.This electronic version was programmed by Nooro On-line Research (https://nooro.com) for desktop and wire-less device (iPAD or laptop) application. Data wereentered directly into wireless devices in care settingsand automatically uploaded to a secure database. Thissignificantly reduced workload and data entry error aspaper copies were not used, eliminating the need fordata re-entry into an electronic database. The electronicapplication was tested extensively prior to field use.Data collection issues and procedure standardizationwere managed at regular team meetings among researchstaff and investigators from both provinces.Reid et al. BMC Health Services Research 2013, 13:515 Page 3 of 11http://www.biomedcentral.com/1472-6963/13/515AnalysisData were analyzed using IBM SPSS Statistics 20 (SPSSInc., Chicago, IL). Percentages are reported for categor-ical data; medians, means and standard deviations forcontinuous data. Data are reported for both provincescombined. While we plan to compare differences inresident characteristics, transitions and outcomes acrossboth provinces, the pilot sample size was too small tomake meaningful comparisons. Variation in the numberof missing cases by item resulted in variation in denom-inators. The relevant denominator is therefore reportedfor each data point. Missing data patterns were analysedby preparing a missing data report (% missing) for first-order questions (no skip patterns or secondary data) toidentify data elements that required further work andcases to be excluded if more than 50% of data was notretrieved across all settings.ResultsResident sample descriptionOf 114 cases identified during the 3 month pilot period(AB = 85, BC = 29), 54 transitions (AB, N = 28; BC, N = 26)had data from each setting (NH, EMS, ED) and wereincluded in the pilot study. In Alberta, 57 cases wereidentified in the ED but either did not meet the inclu-sion criteria (from assisted living facilities; facilities forwhich we had not yet received operational approval); or,we could not retrieve data from each setting in the tran-sition. Resident transitions included were from 10 ABNHs (range 1-6 cases per NH) and 8 BC NHs (1–8).A majority of the 54 residents were female (61.8%)and mean age was 87.1 years (SD = 6.9) ranging from71 to 100 years. Residents frequently had multiplehealth challenges ranging from 0 to 7 per resident, with47.2% having four or more. The most commonly re-corded impairments were vision (59.6%), cognition(55.8%), activities of daily living (50.0%) and mobility(48.1%). Nearly two-thirds of residents had legal/proxysubstitute family decision-makers (64.8%). Almost alltransferred residents (>95%) had next of kin identifiedin the resident’s NH chart. Demographic data werecomplete and accessible.Process objective 1: recruiting facilities/service providersin each of three study settingsA total of 25 NHs in AB (73%) and 12 NHs in BC (92%)agreed to participate in the pilot study. Several AB NHswere still in the process of providing operational approvalwhen the pilot study began. Both AB and BC EDs agreedto participate and provide data collection assistance. EMSin AB and IFTS in both provinces agreed to participate,while BCAS declined.Process objective 2: enrolling participantsStrategies to enroll residents differed by province. In BC,research staff provided NH managers/delegates with anotification script and designated phone number to callto leave a message on a password protected voicemailwhen a resident (aged 65 or older) was sent via ambulancefor an emergency transfer to the ED. In AB, the initialnotification process involved EMS notifying research staffabout a transfer from participating NHs to ED. EMS waschosen for two reasons - they were going to the NH andknew which ED the resident went to. In both provincesmost data elements were collected from resident healthrecords. The main sources of data were NH charts, EMSpatient care record (PCR), ED chart and inpatient recordsas applicable. Some data were accessed electronically, forexample, the Emergency Department Information System(EDIS) in AB.Process objective 3: accessing and extracting dataelements in each setting from patient care recordsMissing dataEach study site and data element had varying degrees ofmissing data. The most frequently missing data elementswere documentation of Assistive Technologies and De-vices (ATD - e.g., eye glasses, canes, dentures) accompany-ing the resident from one setting to the next. Whetherthese were sent with residents was rarely recorded in anyOPTIC study site. Information related to the decision totransfer including trigger events and involvement inconversations to transfer was more complete (e.g., 7.4%missing for trigger events). Information on times be-tween events during transitions was not complete (e.g.,24.1% missing for time of arrival at ED to being seen by anurse). ED data were generally most complete of all studysites (e.g., no missing cases for consultations, diagnosticstests; 14.8% missing for final diagnosis). Documentationdata were least accessible in originating NHs (NH1)(26.0% missing) and on return to NH post-transition(NH2) (33.3% missing). EMS data from NH1 to ED(EMS1) (13.0% missing) were most often accessible aswere ED data (5.6% missing). Data were missing mostoften for the resident’s return trip from ED to NH2 viaIFTS (46.3%). This is likely due to a problem with dataelements missed in programming the T3 electronic datacollection form. Feasibility of data collection was thusvariable and subject to alterations in data collectionprotocol.Time to complete the T3Assuming ideal circumstances, where data were immedi-ately available to research staff, T3 completion time wouldhave been 2–3 hours. However, actual completion time ofan individual transition from start to finish routinely tookmuch longer and varied widely between cases. There wereReid et al. BMC Health Services Research 2013, 13:515 Page 4 of 11http://www.biomedcentral.com/1472-6963/13/515several reasons for this. First, complexity of resident tran-sitions led to significant differences in time for completiondepending on the resident’s medical needs. For example, ittook longer to track transfers when a resident was admit-ted to an inpatient unit rather than discharged back toNH. Second, OPTIC staff tracked multiple transfers atonce, complicated in AB because not all NH operationalapprovals had yet been secured. This resulted in retainingcases enrolled with ED operational approval until NHoperational approval to collect data was received. Third, anumber of factors made it difficult to access residents’charts, e.g. if research staff were interrupted or had diffi-culty accessing hospital medical records or NHs charts,resulting in multiple trips to hospitals or NHs to collect alldata. Fourth, logistics were an important consideration inAB due to the large number of NHs providing operationalapproval and their geographic spread across the city. Itwas important to coordinate data collection for transfersfrom one NH or NHs located in the same vicinity tooptimize efficiency. Thus, data collection for some trans-fers was delayed until it was possible to attend the NH.Process objective 4: determining necessary revisions tothe T3The pilot study allowed the OPTIC team to identify andrectify “glitches” in the T3 such as missing “other”, “notapplicable” and/or “not recorded” options, and incorrectskip pattern questions. Regarding specificity of time ele-ments, we provided the opportunity for data collectors toindicate whether an event occurred during the day (0700–1500), evening (1500–2300), or night (2300–0700) shiftwhen exact times were not recorded. Additional field notesections were added to allow research staff to provide con-textual notes where appropriate, and a data dictionary wasdeveloped to define each term. No data elements wereadded or deleted. Due to the comprehensive nature ofdata collection and frequent travel to ED and NH sites,additional research assistants were contracted in eachprovince. Two additional iPads were acquired and config-ured to allow a secure online server to store bothcomplete and incomplete files. All transition tracking datawere stored on secure servers with security configuredsuch that each case could be accessed by one OPTIC staffmember at a time. For example, one data collector coulddownload a case while in the ED to enter data on the resi-dent’s admission. Once that case was uploaded to the ser-ver, another data collector in the NH could download theupdated case on the resident’s return and add additionalinformation.The initial process of EMS identifying residents beingtransferred to the ED in Alberta and then notifying re-search staff did not work. The process was revised to haveresearch staff access the Emergency Department Informa-tion System (EDIS) to identify new eligible transfers. InBC while the initial identification process worked, signifi-cant staff support was required to remind NH staff aboutthe study, and simultaneously inquire about transfers thathad occurred.Due to the large number of participating facilities andstaff at each site, research staff attempted to locate andestablish a working relationship with a contact person(s)at each site. When these relationships were developed,data collection was more efficient and complete as con-tact persons provided OPTIC research staff with accessto charts and other required information. Each study siteused a different charting system; even NH health recordsvaried by facility. During the pilot, data collectors learnedwhere to look in residents’ charts for specific data ele-ments in the T3.Outcomes objective 5: describing the sample oftransitionsThe resident at the NH (NH1): the beginning of a transitionA trigger event – either a change in resident’s conditionand/or an acute event – caused one or more individualsto be consulted about possible transfer, with one individualmaking the final decision to initiate transfer (see Table 1).Falls causing injury were the most common trigger event(30.9% of all trigger events), followed by a change inphysical condition (14.7%), and nausea/vomiting/diarrhea(11.8%). Between one and three trigger events were re-corded for each resident. Trigger event data were availableand accessible for 50 of the 54 residents.NH staff most frequently involved in discussions con-cerning decision to transfer were Registered Nurses (RN;38.1%), the physician of record (23.8%), Licensed PracticalNurses (LPN; 19.0%) family/friend caregiver (11.9%), andresidents (2.4%). The resident’s physician most often madethe final decision to transfer (33.3%), followed by RNs(25.0%), family/friend caregiver (18.8%), LPNs (14.6%),and residents themselves (4.2%).Emergency medical services: from NH to ED (EMS1)Transitions occurred each day of the week, with almosthalf on Mondays (24.5%) and Fridays (20.4%). Transferstook place most often on day shift (0700–1500, 53.1%),evening shift (between 1500–2300, 38.8%) and night shift(2300–0700, 10.2%). Figure 1 shows the distribution ofED transfers by day of week and by shift.Emergency departmentSelect timing variables following arrival at ED are pre-sented in Table 2. Median time between arrival and beingseen by a nurse was 70 minutes (N = 41), and to be exam-ined by an ED physician was 73 minutes from arrival atED (N = 28). Median time for the decision on disposition(arrival at ED to decision to admit to hospital or dischargeto original NH (N= 28)) was 394 minutes. Median lengthReid et al. BMC Health Services Research 2013, 13:515 Page 5 of 11http://www.biomedcentral.com/1472-6963/13/515of stay in the ED for residents not admitted as inpatientswas 468 minutes (N = 18), compared to 529 minutes forresidents admitted to inpatient units (N = 17). For thelatter, median time between arrival at ED and return toNH was 5 days, 6.5 hours (N = 15).Prior to ED admission, all residents were assessed usingCTAS to determine urgency for treatment. The scale scoreranges from 1 to 5 (5 = lowest priority for immediate treat-ment, 1 = life-saving resuscitation is needed). In BC, CTASscores were assigned by RNs; in AB, CTAS scoreswere assigned by EMS personnel or RNs (84.0%, 16%,respectively). Most residents (74.0%) were scored 3, whichindicated a non-life threatening condition requiring imme-diate action.Upon admission to ED, consultations, diagnostic testsand medical procedures (Table 3) were completed. Fewerthan half of residents (40.7%, n = 22) received specialtyconsultative services and 10 (18.5%) received consultationsfrom two or three services. Internal Medicine was themost common (25.9%), followed by orthopaedics (14.8%),gastroenterology and gerontology (11.1% each). All resi-dents had one or more diagnostic tests performed(mean = 2.4 tests). The three most common diagnostictests conducted were lab work (27.1% of the total testsperformed), X-rays (25.6%), and urinalysis (20.9%). Al-most half of transferred residents underwent medicalprocedures (46.3%).Final ED diagnosis was recorded for 85.2% residents.Fractures were most common (26.1%), followed by falls re-lated injuries (17.4%) and respiratory conditions (10.9%).Delirium was infrequently recorded during transitions. Inthree cases, the chief complaint on arrival to ED was re-corded as follows: bizarre behaviour (n = 1) with a finaldiagnosis of query delirium, and confusion (n = 2) withfinal diagnoses of dementia and iron toxicity respectively.We collected data on disposition decision and actual dis-position of resident. Fewer than half (43.4%) of residentswere admitted as inpatients, while 54.7% were dischargedback to their originating NH. One person was sent to adifferent NH. Three residents (5.7%) died while in in-patient care.Transport back to NHData were available for 44 residents. IFTS was the mostcommon mode of transport back to a NH (86.4%), withthe remainder by EMS, EMS Patient Transfer or family/friend caregiver.Tracking of Assistive Technologies and Devices (ATD)One component of the T3 was designed to capturewhether or not nine specific ATD were recorded as ac-companying the resident through each transition stage,including: glasses, dentures, hearing aids, medications,healthcare card, cane/walker clothing, slippers, and an“other” category to capture items not on this list. Over-all, ATDs were rarely recorded throughout the transitionprocess.Tracking accompanying documentationAppropriate transfer of critical documentation was trackedacross transition settings; however, the number of caseswhere data were available differed by study site. Frequen-cies of resident documentation being sent across two ormore stages of a transition are presented in Table 4, whichin turn can be used to identify sources of information gapsTable 1 Events and decisions leading to transitionsTrigger event(s) N (%)Falls 21 (30.9)Fractures and other fall related injuries 4 (5.9)Hip/pelvis/leg pain 1 (1.5)Change in physical condition 10 (14.7)Nausea/vomiting/diarrhea 8 (11.8)Change in mental status 5 (7.4)Shortness of breath 3 (4.4)Family/friend caregiver request 3 (4.4)Chest Pain 3 (4.4)Urinary symptoms 3 (4.4)Change in behaviour 2 (2.9)Respiratory symptoms 2 (2.9)Other 4 (5.9)# of trigger events 68Valid cases/missing 50/4Who was involved in decision to transfer? (Check all that apply)Registered Nurse 32 (38.1)Physician of record 20 (23.8)LPN 16 (19.0)Family/friend caregiver 10 (11.9)Resident 2 (2.4)Nurse practitioner 2 (2.4)Physiotherapist 1 (1.2)Healthcare Aide (HCA) 1 (1.2)# involved 84Valid cases/missing 50/4Who made the final decision to transfer?Physician of record for resident 16 (33.3)Registered Nurse 12 (25.0)Family/Friend 9 (18.8)Licenced Practical Nurse 7 (14.6)Resident 2 (4.2)Nurse practitioner 2 (4.2)Valid cases/missing 48/6Reid et al. BMC Health Services Research 2013, 13:515 Page 6 of 11http://www.biomedcentral.com/1472-6963/13/515across care settings. Medication lists were recorded as sentmost often in the transfer to ED rather than on return toNH. Records of allergies, do not resuscitate (DNR) orders,patient care plans and advance directives were passedfrom the originating NH through all settings back to theNH.The ED summary, inpatient summary, transfer record,lab results/orders, patient follow-up and others were notcommonly recorded or found in the resident’s NH chartupon return. All documentation types were missingmost often for the return trip via IFTS/EMS to the NH(NH2).DiscussionColeman defined ‘transitional care’ as “a set of actionsdesigned to ensure the coordination and continuity ofhealthcare as patients transfer between different locationsor levels of care in the same location” [28]. While most re-search in this area has focused on the progression of pa-tients from one care setting to another as a unidirectionalprocess, the reality is that transitions occur across multiplesettings and in multiple directions [5,10,29]. Despite thechallenges identified during the course of this pilot study,concurrent case tracking during transitions was feasible.Few NH to ED transition studies have been conductedin Canada (e.g. see [11,16,30]) and much of the literatureis limited to one part of the system – the ED. Few focuson pre- and post-hospital transfer services. Each care set-ting tends to behave in a “silo” manner, resulting in diffi-culty with relationship formation and limits to access tovital patient information which can compromise cross-sitecare coordination [31]. This pilot study is unique in that itfollowed residents throughout their transition experiencesacross all care settings and captured in-depth data aboutthe transition process not elsewhere available. An advan-tage of this approach is the opportunity to identify gaps incare as the resident moves through the different organiza-tions that comprise the system. Previous research has indi-cated that a significant information gap often existsbetween NHs and the ED [32-34]. The T3 was designed torecord the flow of all relevant documentation during thecourse of a transfer, allowing for the identification of thesetting within which these gaps occurred. With one excep-tion (discussed below) the T3 was successful in trackingthe flow of documentation through each setting.Consistent with the literature, the T3 pilot study foundthat the most common information gap in documentationsent between facilities were: assistive technologies andFigure 1 Transfer day of the week and shift (day, evening and night).Table 2 Time for resident to be seen from arrival time at the ED (hours: minutes)Time from arrival at the ED to N MDN Mean (SD)Being seen by a nurse 41 1:10 3:54 (10:53)Being seen by a physician (history and examination details recorded) 28 1:13 1:40 (1:08)Decision to admit or return to NH 28 6:34 11:00 (20:31)Actual admission to the inpatient unit 17 8:49 14:08 (10:36)Actual transfer back to NH (admitted to inpatient unit) (total time spent in ED and hospital) 15 126:29 196:01 (170:25)Actual transfer back to NH (not admitted) (total time spent in ED) 18 7:38 10:54 (10:53)Reid et al. BMC Health Services Research 2013, 13:515 Page 7 of 11http://www.biomedcentral.com/1472-6963/13/515devices [32], laboratory test results [34] and cognitivefunction documentation [32-34]. These three studies[32-34] identified information gaps at single settings inthe transition, whereas the T3 is able to identify infor-mation gaps that occur at each setting across the entiretransition. The pilot study clearly illustrates the incon-sistency of the content sent across settings and has thepotential to help identify essential elements requiredacross the entire transition as well as informationdeemed essential for each setting. To date, no knownstudies have identified essential information across theentire transition [32,34].In both provinces, however, effective tracking of resi-dents across settings necessitated site-specific strategiesand strong relationships between research staff and faci-lities staff. The Partnerships in Health System Improve-ment model of conducting research (http://www.cihr-irsc.gc.ca/e/34348.html) requires close cooperation betweenresearchers and decision-makers from the genesis of theidea through to meaningful knowledge translation of theresults. Relationship building and maintenance are criticalto successful data collection from planning to implemen-tation [35]. Consistent with recommendations from theresearch literature (e.g. see [36]), regular face-to-face con-sultation sessions, meetings and other forms of communi-cation formed the basis of these relationships. It is difficultto imagine the successful trialing of the T3 withoutsuch in-depth involvement and combined sense of pur-pose by both researchers and decision-makers.The T3 tracked assistive technology and devices; how-ever, tracking of ATDs was difficult due to inconsistentdocumentation in the residents’ care records at all pointsof the transfer. In a study by Hammel et al. stakeholdersincluding patients and providers identified ATD use asdecreasing demands on others for time, assistance, energy,mitigating safety risk and conducting activities of dailyliving and that this benefit far exceeded just functionalindependence [37]. Poor tracking and provision ofATDs during transitions has been a source of ED pa-tient dissatisfaction and has accounted for up to a thirdof patient complaints [38]. In extreme cases, the loss ofdentures (one ATD) in a NH has led to a documentedadverse drug event [39]. Although this pilot study wasnot measuring cost, both the prohibitive cost of re-placing these items for some and the difficulty or im-possibility of replacing them in a population withdementia require further investigation. There are alsoimplications for quality of life in the event that ATDsare not available to residents during their transitionand beyond.Any conclusions based upon outcome data in this pilotstudy (objective 5) are limited by the small sample size.Due to the dynamic nature of the process, some datacould only be gathered a period of time after dischargeTable 3 Assessments, care and services while in theED (n = 54)CTAS score N (%)I 0 (0.0)II 7 (14.0)III 37 (74%)IV 5 (10.0)V 1 (2.0)Valid cases/missing 50/4Service(s) providedConsultation(s) 22 (40.7)Diagnostic test(s) 54 (100.0)Medical procedure(s) 25 (46.3)Valid cases (for each service) 54Consultation(s)Internal medicine 7 (25.9)Orthopaedics 4 (14.4)Gastroenterology 3 (11.1)Gerontology 3 (11.1)Other 10 (37.0)Total consultations 27Valid cases (for each consultation type) /Missing(no consultation)22/32Diagnostic test(s)Lab work 35 (27.8)X-rays 33 (26.2)Urine 27 (21.4)Electrocardiogram 13 (10.3)CT scan 9 (7.1)Ultrasound 3 (2.4)Radiometer 2 (1.6)Other 7 (5.6)Total consultations /Valid cases (for each test type) 126/54ED Final DiagnosisFractures (hip/pelvis, limb) 12 (26.1)Falls related injuries 8 (17.4)Respiratory (respiratory failure, pneumonia, COPD) 5 (10.9)Urinary-related illness 4 (8.7)Altered mental status (dementia, confusion) 4 (8.7)Gastro-intestinal (GI bleed, bowel obstruction) 3 (6.5)CVA (stroke) 3 (6.5)Cardiac (cardiac arrest, chest pain, CHF) 2 (4.3)Abnormal blood work 2 (4.3)Other 3 (6.5)Valid cases/missing 46/8Reid et al. BMC Health Services Research 2013, 13:515 Page 8 of 11http://www.biomedcentral.com/1472-6963/13/515from hospital. It is possible that not all information (e.g.,communication, documentation, etc.) transferred withthe resident was kept in the resident’s care record or wasrecorded in the chart at the time the research staffaccessed residents’ records.ConclusionsOlder adults transferred between NHs and EDs representa group of patients at risk for errors and adverse medicaloutcomes. While collecting data on transitions from aNH to the ED and back is complex, it can be achieved.Overall, the T3 provided valuable and detailed informa-tion about transitions in care for elderly patients trans-ferred from NH to EDs in two Canadian provinces.Further research will expand the sample size, providedetailed documentation of the transition issues facingelderly NH patients and provide recommendations forreducing gaps in care for this vulnerable population.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsRCR participated in the design of the study, was provincial lead for theBC component of the study and was lead author on the writing of themanuscript. GEC was lead on the T3 phase of the OPTIC project, wasEMS lead and participated in the design of the study and editing of themanuscript. SLC was second writer for the manuscript, prepared figures,provided editing and was responsible for preparation of the manuscriptfor publication submission. SA carried out data collection. SA and LJBcarried out data collection. CAE participated in the design of the study,original conception of study, worked with GGC, PGN and RCR for fundingof the study and editing of the manuscript. BHR provided access to ED fordata collection as the AB ED lead for the project, participated in design ofthe study and editing of the manuscript. AW participated in design of theTable 4 Documentation accompanying resident at each stageStage of transitionaNH1 EMS1 ED EMS2 NH2Documentation N (%) N (%) N (%) N (%) N (%)Medication list 33 (82.5) 47 (100.0) 48 (94.1) c 18 (50.0)Patient summary and transfer Information 29 (72.5) 42 (89.4) 41 (80.4) c 6 (16.7)Record of allergy 27 (67.5) 38 (80.9) 39 (76.5) c 8 (22.2)DNR Order 23 (57.5) 31 (66.0) 34 (66.7) c 6 (16.7)Advance directive 17 (42.5) 19 (40.4) 17 (33.3) c 2 (5.5)Patient Care Plan 11 (27.5) 11 (23.4) 17 (33.3) c 5 (13.9)Resident Data 8 (20.0) 18 (38.3) 16 (31.4) c 1 (2.8)Physician Orders & Notes 3 (7.5) 7 (14.9) 5 (9.8) c 1 (2.8)Resident Clinical Data 2 (5.0) 4 (8.5) 8 (15.7) c 3 (8.3)List of Diagnoses 3 (7.5) 5 (10.6) 6 (11.8) cPCR form b b 44 (86.3) c 3 (8.3)ED summary b b b 19 (65.5) 14 (38.9)Inpatient summary b b b 8 (27.6) 8 (22.2)Transfer record b b b 7 (24.1) 8 (22.2)Lab results/orders b b b c 9 (25.0)ED Nurses’ Notes b b b c 8 (22.2)Patient follow up b b b c 3 (8.3)Inpatient forms b b b c 3 (8.3)Consultations b b b c 3 (8.3)Inpatient Nurses’ Notes b b b c 3 (8.3)Follow up appointments b b b c 2 (5.5)OR documentation b b b c 3 (8.3)Other 2 (5.0) 1 (2.1) 4 (7.8) c 0 (0.0)Valid cases (for each type of documentation) 40 47 51 29 36Missing 14 7 3 22d 15daNH1 = originating nursing home; EMS1 = emergency medical services to ED; ED = emergency department; EMS2 = transportation service from ED; NH2 = return tonursing home.bnot applicable.cnot recorded in the T3.d3 residents died in inpatient care.Reid et al. BMC Health Services Research 2013, 13:515 Page 9 of 11http://www.biomedcentral.com/1472-6963/13/515study. PGN participated in the design of the study, original conception ofthe study and worked with GGC, CAE and RCR to secure funding for thestudy. ME provided access to ED data collection as the BC ED lead for theproject. ME also reviewed an early version of the manuscript. GGC is thenominated principal investigator for the OPTIC research program, providedleadership and coordination of the AB group, led its design and proposaldevelopment, and helped to draft the manuscript. All authors read andapproved the final manuscript.AcknowledgementsThis work was supported by the Canadian Institutes of Health Research (CIHRgrant# CIHR PHE 101863), the Michael Smith Foundation for Health Research(MSFHR), Alberta Heritage Foundation for Medical Research (AHFMR), AlbertaHealth Services (AHS), Interior Health Authority, Kelowna, British Columbia(IHA). A partnership made possible by the Partnership for Health SystemImprovement Program offered by the Canadian Institutes of Health Research.We also acknowledge funding from the following organizations: Universityof Alberta Hospital Foundation; and the BC Network for Aging Research(BCNAR). Dr. Greta Cummings holds a Centennial Professorship at theUniversity of Alberta; Dr. Estabrooks holds a Tier I Canada Research Chair inKnowledge Translation from CIHR; Dr. Rowe holds a Tier I Canada ResearchChair in Evidence-based Emergency Medicine from CIHR. These funding bodiesdid not have a role in design, collection, analysis or interpretation of data.We are grateful for the active partnership and full contribution of theresearch team of decision-makers and researchers.Principal investigatorsGreta G. Cummings, Nominated Principal Applicant, University of AlbertaCarole A. Estabrooks, Principal Applicant, University of AlbertaPeter G. Norton, Principal Applicant, University of CalgaryR. Colin Reid, Principal Applicant, University of British Columbia OkanaganCo-investigatorsJoan Bottorff, Co-Applicant, University of British Columbia OkanaganGarnet E. Cummings, Co-Applicant, University of AlbertaNorah Keating, Co-Applicant, University of AlbertaJacques S. Lee, Co-Applicant,Sunnybrook Health Sciences CentreMeredith Lilly, Co-Applicant, University of British ColumbiaCandace Nykiforuk, Co-Applicant, University of AlbertaBelinda Parke, Co-Applicant, University of AlbertaCarole Robinson, Co-Applicant, University of British Columbia OkanaganBrian H. Rowe, Co-Applicant, University of AlbertaAdrian Wagg, Co-Applicant, University College HospitalPrincipal decision-makersJoanne Konnert, Chief Operating Officer, Interior HealthGlenda Coleman-Miller, Vice President, University of Alberta HospitalDecision-makersCarol Anderson, Executive Director, Edmonton Zone, Alberta Health ServicesCindy Crane, Manager, Interior Health, Kelowna General HospitalMichael Ertel, Chief and Medical Director, Interior Health Kelowna GeneralHospitalKaren Latoszek, Senior Manager, Alberta Health ServicesTracy Buffam, Client Services Manager, Good Samaritan Southgate CareCenterCorinne Schalm, Director, Access & Innovation, Continuing Care, AlbertaHealthCindy Regier, Director Residential Services, South OkanaganSunil Sookram, Medical Director EMS, Alberta Health ServicesAdditional membersFaye Burch, Director Residential Services, Cottonwoods Care CentreSarah L. Cooper, Graduate Student, Research AssistantAuthor details1School of Health and Exercise Sciences, University of British Columbia’sOkanagan campus, 3333 University Way, ART, Kelowna, British Columbia V1V1V7, Canada. 2Department of Emergency Medicine, Faculty of Medicine andDentistry and School of Public Health, University of Alberta, Edmonton,Alberta, Canada. 3Faculty of Nursing, University of Alberta, 5-110 EdmontonClinical Health Academy, 11405 87 Avenue, Edmonton, Alberta T6G 1C9,Canada. 4Division of Geriatric Medicine, Department of Medicine, Faculty ofMedicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada.5Department of Family Medicine, Faculty of Medicine, University of Calgary,Calgary, Alberta, Canada. 6Kelowna General Hospital, Interior HealthAuthority, Kelowna, British Columbia, Canada.Received: 31 May 2013 Accepted: 9 December 2013Published: 14 December 2013References1. Boling PA: Care transitions and home health care. Clin Geriat Med 2009,25(1):135–148.2. Scott IA: Preventing the rebound: Improving care transition in hospitaldischarge processes. Aust. Health Rev 2010, 34(4):445–451.3. Mor V, Besdine RW: Policy options to improve discharge planning andreduce rehospitalization. JAMA 2011, 305(3):302–303.4. Gruneir A, Bronskill S, Bell C, Gill S, Schull M, Ma X, Anderson G, Rochon PA:Recent health care transitions and emergency department use bychronic long term care residents: A population-based cohort study. J AmMed Dir Assoc 2012, 13(3):202–206.5. Tsilimingras D, Bates DW: Addressing postdischarge adverse events:a neglected area. J Qual Patient Saf 2008, 34(2):85–97.6. Hastings SN, Oddone EZ, Fillenbaum G, Sloane RJ, Schmader KE: Frequencyand predictors of adverse health outcomes in older medicarebeneficiaries discharged from the emergency department. Med Care2008, 46(8):771–777.7. Coleman EA, Smith JD, Frank JC, Eilertsen TB, Thiare JN, Kramer AM:Development and testing of a measure designed to assess the qualityof care transitions. Int J Integr Care 2002, 2:e02.8. Parry C, Mahoney E, Chalmers SA, Coleman EA: Assessing the quality oftransitional care: further applications of the care transitions measure.Med Care 2008, 46(3):317–322.9. Canadian Institute for Health Information: Patient pathways: Transfers fromcontinuing care to acute care. Ottawa, Ontario: CIHI; 2009.10. McCloskey R, van den Hoonaard D: Nursing home residents in emergencydepartments: A Foucauldian analysis. J Adv Nurs 2007, 59(2):186–194.11. Gruneir A, Bell CM, Bronskill SE, Schull M, Anderson GM, Rochon PA: Frequencyand pattern of emergency department visits by long-term care residents–a population-based study. J Am Geriat Soc 2010, 58(3):510–517.12. McCloskey R: Use of one ER by nursing home residents in NewBrunswick. Geriat Today 2004, 7:53–58.13. Castle NG, Mor V: Hospitalization of nursing home residents: A review ofthe literature, 1980–1995. Med Care Res Rev 1996, 53(2):123–148.14. Malone M, Danto-Nocton E: Improving the hospital care of nursing homeresidents. Ann of Long-Term Care 2004, 12:42–49.15. Callahan CM, Arling G, Tu W, Rosenman MB, Counsell SR, Stump TE, HendrieHC: Transitions in care for older adults with and without dementia. J AmGeriat Soc 2012, 60(5):813–820.16. McCloskey RM: A qualitative study on the transfer of residents between anursing home and an emergency department. J Am Geriat Soc 2011,59(4):717–724.17. Coleman EA: Falling through the cracks: Challenges and opportunities forimproving transitional care for persons with continuous complex careneeds. J Am Geriatr Soc 2003, 51(4):549–555.18. Crilly J, Chaboyer W, Wallis M: Continuity of care for acutely unwell olderadults from nursing homes. Scand J Caring Sci 2006, 20(2):122–134.19. Naylor MD: Transitional care of older adults. Ann Rev Nurs Res 2002,20:127–147.20. Kelly NA, Mahoney DF, Bonner A, O’Malley T: Use of a TransitionalMinimum Data Set (TMDS) to improve communication between nursinghome and emergency department providers. J Am Med Dir Assoc 2012,13(1):85–89.21. Cummings GG, Reid RC, Estabrooks CA, Norton PG, Cummings GE, RoweBH, Abel SL, Bissell L, Bottorff JL, Robinson CA, et al: Older Persons’Transitions in Care (OPTIC): A study protocol. BMC Geriatr 2012, 12:75.22. Robinson CA, Bottorff JL, Lilly MB, Reid C, Abel S, Lo M, Cummings GG:Stakeholder perspectives on transitions of nursing home residents tohospital emergency departments and back in two Canadian provinces.J. Aging Stud 2012, 26:419–427.Reid et al. BMC Health Services Research 2013, 13:515 Page 10 of 11http://www.biomedcentral.com/1472-6963/13/51523. Statistics Canada: Edmonton, Alberta (Code 4811061), Canada (Code 01)and Kelowna, British Columbia (Code 5935010) (table). Census Profile,2011 Census. 2012. http://www12.statcan.gc.ca/census-recensement/2011/dp-pd/prof/index.cfm?Lang=E.24. Alzheimer Society of Canada: Rising Tide: The impact of dementia onCanadian society. Toronto, ON; 2010. http://www.alzheimer.ca/en/Get-involved/Raise-your-voice/Rising-Tide.25. Morris JN, Fries BE, Mehr DR, Hawes C, Phillips C, Mor V, Lipsitz LA: MDSCognitive Performance Scale. J Gerontol 1994, 49(4):M174–M182.26. Gillespie SM, Gleason LJ, Karuza J, Shah MN: Health care providers’opinions on communication between nursing homes and emergencydepartments. J Am Med Dir Assoc 2010, 11(3):204–210.27. Beveridge R: CAEP issues. The Canadian Triage and Acuity Scale:A new and critical element in health care reform. Canadian Associationof Emergency Physicians. J Emerg Med 1998, 16(3):507–511.28. Coleman EA, Berenson RA: Lost in transition: Challenges andopportunities for improving the quality of transitional care. Ann InternMed 2004, 141(7):533–536.29. Davis MM, Devoe M, Kansagara D, Nicolaidis C, Englander H: “Did I do asbest as the system would let me?” Healthcare professional views onhospital to home care transitions. J Gen Intern Med 2012,27(12):1649–1656.30. McCloskey RM: A conceptual framework for understandinginterorganizational relationships between nursing homes andemergency departments: Examples from the Canadian setting.Policy, Polit Nurs Pract 2009, 10(4):285–294.31. American Medical Directors Association: Transitions of care in the long-termcare continuum clinical practice guideline. Columbia, MD: AMDA; 2010.32. Cwinn MA, Forster AA, Cwinn AJ, Hebert G, Calder L, Stiell IG: Prevalence ofinformation gaps for seniors transferred from nursing homes to theemergency department. J Emerg Med 2009, 11(5):462–471.33. Gaddis GM: Elder care transfer forms. Acad Emerg Med 2005, 12(2):160–161.34. Stiell A, Forster AJ, Stiell IG, van Walraven C: Prevalence of informationgaps in the emergency department and the effect on patient outcomes.CMAJ 2003, 169(10):1023–1028.35. Hinkin T, Holtom BC, Klag M: Collaborative research: Developing mutuallybeneficial relationships between researchers and organizations. OrganDyn 2007, 36(1):105–118.36. Lomas J: Connecting research and policy. J Policy Res 2000, 1(1):140–144.37. Hammel J, Southall K, Jutai J, Finlayson M, Kashindi G, Fok D: Evaluatinguse and outcomes of mobility technology: A multiple stakeholderanalysis. Disabil Rehabil Assist Technol J 2012, 00:1–11.38. VanderVeen L, Ritz M: Customer satisfaction: A practical approach forhospitals. J Healthc Qual 1996, 18(2):10–15.39. Field TS, Mazor KM, Briesacher B, Debellis KR, Gurwitz JH: Adverse drugevents resulting from patient errors in older adults. J Am Geriat Soc 2007,55(2):271–276.doi:10.1186/1472-6963-13-515Cite this article as: Reid et al.: The Older Persons’ Transitions in Care(OPTIC) study: pilot testing of the transition tracking tool. BMC HealthServices Research 2013 13:515.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitReid et al. BMC Health Services Research 2013, 13:515 Page 11 of 11http://www.biomedcentral.com/1472-6963/13/515


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