UBC Faculty Research and Publications

Transfers from intensive care unit to hospital ward: a multicentre textual analysis of physician progress… Brown, Kyla N; Leigh, Jeanna P; Kamran, Hasham; Bagshaw, Sean M; Fowler, Rob A; Dodek, Peter M; Turgeon, Alexis F; Forster, Alan J; Lamontagne, Francois; Soo, Andrea; Stelfox, Henry T Jan 28, 2018

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

Item Metadata


52383-13054_2018_Article_1941.pdf [ 584.76kB ]
JSON: 52383-1.0363333.json
JSON-LD: 52383-1.0363333-ld.json
RDF/XML (Pretty): 52383-1.0363333-rdf.xml
RDF/JSON: 52383-1.0363333-rdf.json
Turtle: 52383-1.0363333-turtle.txt
N-Triples: 52383-1.0363333-rdf-ntriples.txt
Original Record: 52383-1.0363333-source.json
Full Text

Full Text

RESEARCH Open AccessTransfers from intensive care unit tohospital ward: a multicentre textual analysisof physician progress notesKyla N. Brown1†, Jeanna Parsons Leigh1*†, Hasham Kamran1, Sean M. Bagshaw2, Rob A. Fowler3, Peter M. Dodek4,Alexis F. Turgeon5, Alan J. Forster6, Francois Lamontagne7, Andrea Soo1 and Henry T. Stelfox1AbstractBackground: Little is known about documentation during transitions of patient care between clinical specialties.Therefore, we examined the focus, structure and purpose of physician progress notes for patients transferred fromthe intensive care unit (ICU) to hospital ward to identify opportunities to improve communication breaks.Methods: This was a prospective cohort study in ten Canadian hospitals. We analyzed physician progress notes forconsenting adult patients transferred from a medical-surgical ICU to hospital ward. The number, length, legibilityand content of notes was counted and compared across care settings using mixed-effects linear regression modelsaccounting for clustering within hospitals. Qualitative content analyses were conducted on a stratified randomsample of 32 patients.Results: A total of 447 patient medical records that included 7052 progress notes (mean 2.1 notes/patient/day95% CI 1.9–2.3) were analyzed. Notes written by the ICU team were significantly longer than notes written by theward team (mean lines of text 21 vs. 15, p < 0.001). There was a discrepancy between documentation of patientissues in the last ICU and first ward notes; mean agreement of patient issues was 42% [95% CI 31–53%]. Qualitativeanalyses identified eight themes related to focus (central point – e.g., problem list), structure (organization, – e.g.,note-taking style), and purpose (intention – e.g., documentation of patient course) of the notes that varied acrossclinical specialties and physician seniority.Conclusions: Important gaps and variations in written documentation during transitions of patient care betweenICU and hospital ward physicians are common, and include discrepancies in documentation of patient information.Keywords: Provider communication, Documentation, Patient transfer, Intensive care unit, Progress notes, Hospital wardBackgroundTransitions of patient care are vulnerable periods inhealthcare delivery that expose patients to potentialbreakdowns in communication [1–3], medical errors [4],and adverse events [5, 6]. The transfer of a patient fromthe intensive care unit (ICU) to a hospital ward repre-sents an example of a common high-risk inter-specialtytransition of care where patients with complex life-threatening problems transition from the care of a crit-ical care medicine physician to a medical, surgical orprimary care physician. Ineffective handoffs can lead toapproximately 10% of adverse events in the ICU [7].Efforts to improve transitions of care have focused onpatient transfers during end-of-shift [2, 8, 9] or end-of-service [10–12] handoffs. However, there may be differ-ences in culture and clinical focus between providers indifferent service areas of a hospital [13–15], such as thetransfer of a patient from the intensive care unit (ICU)to a hospital ward.A report from the Canadian Institute for HealthInformation found that unintended harm occurs in oneof every 18 hospitalizations [16]. Communication issuesare a root cause of such adverse events [17]. Effectivetransitions of patient care, which have been previously* Correspondence: jeanna.parsonsleigh@albertahealthservices.ca†Equal contributors1Calgary, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 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.Brown et al. Critical Care  (2018) 22:19 DOI 10.1186/s13054-018-1941-0described as including continuous communication [10]and coordination [18] of patient needs across the health-care continuum, contribute to improved patient care[19, 20], safety [10, 21], and experience [21]. Themedical record is central to this process because it is therepository for documenting events [22, 23], currentissues [24], and services provided [25]. It is a source ofdurable information that healthcare providers and insome institutions, patients [26, 27] can access whenmaking clinical decisions [24, 28]. Physician progressnotes are a core component of the medical record[22, 29] and efforts to optimize documentation have in-cluded the development and implementation of standard-ized forms [22, 30], such as the subjective, objective,assessment, and plan (SOAP) model [31] for paper-basedrecords, or templates for electronic health records (EHRs)[32]. However, there is limited evidence about howinformation should be displayed in the medical record[33, 34], and little is known about documentation duringtransitions of patient care between different medicalspecialty teams [35, 36]. Moreover, a systematic review ofthe literature on patient handoff tools reported littlestandardization across handoff practices, including thestructural organization of tools currently in use [37],highlighting the need for an additional research focus onthe contextual nuances of handoffs and their effect onpatient-related outcomes [37]. Our goal was to contributeto this gap in the literature by describing current text-based communication practices during transfers of patientcare from ICU to a hospital ward, and identifyingopportunities to improve communication and ultimatelypatient safety.MethodsStudy designThis study was part of a multicentre prospective cohortstudy [38] describing transfers of care for adult patientswho were discharged from medical-surgical ICUs to hos-pital wards between July 2014 and January 2016 in tenhospitals across seven Canadian cities. The study hospi-tals were selected to ensure a diversity of institutionsincluding tertiary and community care, teaching, non-teaching, size of hospitals, English- and French-languagehospitals, and geographic representation. The HealthResearch Ethics Board at the coordinating center(University of Calgary REB13-0021) and at each of thestudy hospitals approved the study. We targeted enroll-ment of 50 consecutive consenting patients identified asready for transfer from the ICU to a hospital ward ateach study site (n = 500), but ended recruitment at 447patients due to time and resource limitations. Detaileddescriptions of the protocol for this study and the text-ual analyses used in this study have been published pre-viously [39, 40].Data collectionPaper medical records were collected for all enrolled pa-tients (see Additional file 1). Physician progress noteswithin each record were collected for up to ten consecu-tive calendar days depending on the length of hospitalstay: up to 2 days in ICU before transfer, the day oftransfer, and up to 7 days after transfer to the acceptinghospital ward. Notes were photocopied, de-identified,and assigned a unique identifier for analysis.Analytic approachNotes were analyzed to describe communication aroundthe time of transfer from the ICU to the accepting hos-pital ward. Two unique coding systems—descriptive andcontent analyses [39]—were developed to capture quan-titative and qualitative concepts of the notes, respect-ively. Through a preliminary analysis using open codingmethodology [41], overarching themes emerged todescribe: focus (i.e., central point), structure (i.e.,organization) [39], and purpose (i.e., intention) of text-based communication in physician progress notes. Acodebook (Appendix A) was developed by two reviewers(KB, JPL) [39] to systematically identify granular themesfor quantitative and qualitative concepts.Descriptive analysesDescriptive analyses included tabulation of basic descrip-tors (e.g., date), readability (e.g., legibility), the type of in-formation (e.g., patient history) and communicationdocumented (e.g., provider-provider, provider-family).Two reviewers (KB, HK) independently coded all med-ical records to capture these quantitative items [39].Inter-rater reliability for nominal quantitative codes wereevaluated on a random sample of 37 medical records(median kappa score 0.95, 95% confidence interval [CI]0.96–0.99). Notes written during the ICU stay werecompared to those written during the ward stay and be-tween medical and surgical patients using mixed-effectslinear regression models accounting for patients clus-tered within hospitals. Results were summarized usingmeans or mean percentages with 95% CIs. Analyses weredone using Stata version 14 (StataCorp LP, CollegeStation, TX, USA).Content analysesTo provide a more in-depth description of chartingacross institutions, content analyses were applied to apurposive, random sample of medical records from eachof the eight study hospitals where English was the lan-guage used in the medical record (n = 2 surgical patientsand n = 2 medical patients per hospital, totaling n = 32medical records, see Additional file 2) with the overallgoal of capturing similarities and differences in thefocus, structure, and purpose of each progress note. ABrown et al. Critical Care  (2018) 22:19 Page 2 of 8description of the inductive approach that led us tofocus on these overarching trends in the data has beenpublished elsewhere [39]. While our team was preparedto increase the number of charts in our sample if neces-sary, ongoing analysis revealed distinct recurring pat-terns in the data, with no new themes emerging prior tothe conclusion of our analysis on the initial sample of32. Emphasis was placed on variation across: (1) medicalversus surgical specialties, (2) hospitals, (3) levels ofphysician seniority, and (4) temporality. Moreover, weexamined the consistency of medical issues (i.e., problemlist) documented in the last note written by the ICUphysician team, and the first note written by the phys-ician team assuming care on the hospital ward. Thenumber of common issues between the two notes wasdivided by the total number of unique issues in bothnotes to obtain the measure of acceptance.ResultsFindings from quantitative descriptive analysesDocumentation, readability, and mode of communicationFrom the 447 medical records collected (370 written inEnglish), a total of 7052 individual physician progressnotes were identified and analyzed (Table 1). There wasa mean of 2.1 notes/patient/day [95% CI 1.9–2.3] duringthe 10-day period. Most notes were handwritten (97%[95% CI 89–100%]). None of the included hospitalsemployed an integrated electronic health record (EHR) forICU, transition, and ward notes. One ICU (Hospital H)used an electronic Word document template to type pro-gress notes that were subsequently printed on a separatepiece of paper for placement in the medical record. Of allnotes (handwritten and typed) per patient, 87% [95% CI78–96%] were legible (Appendix A), 95% [95% CI 92–98%] included a date, 49% [95% CI 40–58%] included atime, and 48% [95% CI 34–62%] included the identifica-tion of the writer (e.g., full name and/or pager number).Notes written by ICU team members were more frequent(mean/patient/day 2.4 vs. 1.8, p < 0.001) and longer (meannumber of lines of text 21 vs. 15, p < 0.001) than thosewritten by ward team members (Fig. 1). Most lines of textin notes described patient data (i.e., physician’s clinical as-sessment and test results), while fewer lines were dedi-cated to describing the patient’s history (i.e., purpose ofhospitalization, medical history) or clinical plan. Aminority of progress notes documented communicationbetween providers, providers-patients or -patient families.Similar documentation patterns were observed in thenotes of patients who had medical or surgical diagnoseswith a few exceptions. Compared to surgical patients,notes for medical patients were longer (mean number oflines of text 19 vs. 15 p < 0.001), and had more text aboutthe clinical plan (32% vs. 30%, p = 0.009) and aboutcommunication between providers or between providersand patients/family members (10% vs. 7%, p = 0.003)(see Additional files 3 and 4).Table 1 Quantitative descriptive analysis of physician progress notesMeasures Total (n = 7052) ICU stay (n = 1966) Transfer day (n = 1207) Ward stay (n = 3879) p valueaMean number of notes per day (per patient) 2.1 [1.9–2.3] 2.4 [2.2–2.6] 2.8 [2.6–3.0] 1.8 [1.6–2.0] p < 0.001Handwritten 97% [92–100%] 93% [87–100%] 93% [87–100%] 100% [94–100%] p < 0.001Legible 87% [78–96%] 86% [77–95%] 87% [78–96%] 87% [78–96%] p = 0.458Date included 95% [92–98%] 95% [92–98%] 96% [93–99%] 95% [92–98%] p = 0.778Time-stamped 49% [40–58%] 51% [41–61%] 54% [44–64%] 46% [36–56%] p = 0.006Signature/name 48% [34–62%] 48% [34–62%] 48% [34–62%] 47% [33–60%] p = 0.630Mean number of lines per note (per patient) 17.6 [15.9–19.3] 21.1 [19.1–23.1] 19.6 [17.6–21.6] 15.0 [13.1–17.0] p < 0.001Patient historyb 10% [7–13%] 11% [9–14%] 11% [8–13%] 7% [5–10%] p < 0.001Patient datab 59% [54–64%] 57% [52–62%] 58% [53–63%] 61% [56–66%] p < 0.001Patient care planb 31% [26–37%] 32% [26–37%] 32% [26–38%] 31% [26–37%] p = 0.701Communicationc 9% [7–10%] 7% [5–9%] 8% [6–10%] 9% [7–11%] p = 0.071Provider-providerd 56% [50–62%] 64% [56–73%] 62% [52–72%] 51% [44–59%] p = 0.012Provider-familyd 23% [16–29%] 23% [15–31%] 22% [12–31%] 22% [15–29%] p = 0.851Provider-patientd 38% [33–43%] 29% [20–38%] 36% [26–46%] 43% [35–50%] p = 0.007Data are presented as mean proportions (95% confidence intervals) unless otherwise indicated. Data include notes written by ICU and ward physicians categorizedaccording to patient location during the 10-day period. Indented variables are presented as distributionsaP values represent the comparison between the ICU stay and ward stay for each variablebPatient history (e.g., symptoms), data (e.g., laboratory results), and care plan (e.g., treatments prescribed) are presented as mean proportions (%) of the totalnumber of lines for each notecDocumentation of any communication between providers, between providers and family members, and between providers and patientsdProvider-provider, -family, -patient communication presented as mean proportions (%) of the total documented communication that is not mutually exclusiveBrown et al. Critical Care  (2018) 22:19 Page 3 of 8Themes identified from qualitative content analyses(n = 32)Of the 32 medical records included in the content ana-lyses, the majority of notes (73%, 95% CI 54–91%) withan identifiable writer were written by trainees (e.g., resi-dents, fellows).FocusPhysician progress notes were largely comprised of threemain areas of focus: documentation of a patient’songoing medical issues (i.e., problem list), physician sub-jective interpretation of a patient’s status and futurecourse of treatment (i.e., clinical impression), and a sum-mary of events that occurred within a specific timeframe (e.g., overnight).Problem listDiscrepancies in the documentation of patient’s problemsemerged as a key observation during the transition of carefrom ICU to hospital ward. Notes written by staffphysicians were more likely to document one or two mainissues of the patient. Trainee notes were more likely toinclude a numbered problem list and correspondingclinical plan (Additional file 5: Table S5). ICU and medicalteaching unit notes regularly included a head-to-toeassessment. Surgical specialty notes frequently includedinformation specific to an injury or disease.The mean number of issues reported in the last ICUphysician progress note was not significantly differentfrom the number documented in the first ward note (3.4vs. 2.9, p = 0.461). However, the mean agreement of issueslisted in the last ICU note and the first ward note was only42% [95% CI 31–53%] (i.e., there was discrepancy ofinformation in both notes). Similar findings were observedfor patients who had medical (42%) and surgical (42%)diagnoses, but varied substantially between hospitals(range 20–65%).Clinical impressionICU staff physician notes were more likely to documentprovider-patient and provider-family communicationthan trainee notes. Often, ICU and hospital ward staff notesoften included a description of the physician’s thoughtprocess (e.g., phrases such as “I think” and “I feel”), whichwas largely absent from trainee notes across all hospitals.Summary of eventsIn three of the 32 patient charts, no problem lists weredocumented for at least 48 hours after transfer from theICU. Delays in addressing items in the problem list werealso observed. For example, in the case of patient 030(Hospital D), neurology consultants documented thatthey would assess the patient once an electromyogram(EMG) was completed. The EMG was deferred over thepatient’s 7-day ward stay (Additional file 5: Table S5),resulting in a lack of consultation for this patient’sneurological issues.StructureAnalyses of the structure of physician progress noteslargely examined whether information was presented ina standardized or non-standardized manner. Two sub-themes emerged in the analysis of note structure: orderof information, and style and accessibility of notes.Fig. 1 Number of total lines of text in physician progress notes during transition of patient care. The bars represent the mean number of totallines of text in physician progress notes according to each day within the 10-day time period. 95% confidence intervals are displayed on each barBrown et al. Critical Care  (2018) 22:19 Page 4 of 8Order of informationStaff physician notes (ICU and ward) across all hospitalswere structured in a non-standardized format, but hadbroad similarities in the sequence of information docu-mented (i.e., patient history, exam(s)/interventions com-pleted, and clinical plan). Trainee notes in nine of theICUs and many most hospital wards followed a morestandardized format that resembled the SOAP notestructure [31]. One exception was that across all hospi-tals, surgical specialty notes were less structured andcontained less detailed information than other notes inthe sample (regardless of level of seniority) (Additionalfile 5: Table S5).Style and accessibilityGenerally, printed handwritten notes were more legiblethan cursive handwritten notes. Accessing sequentialevents in physician progress notes from the ICU thatused an electronic template (Hospital H) was difficult asthe typed template was on a separate piece of paper,inserted amongst pieces of paper with sequential hand-written notes, leading to non-chronologically orderednotes. ICU notes were typically documented in para-graph form whereas ward notes were largely docu-mented in bullet point form with few words per line innine of the ten hospitals. Notes from all hospitals in-cluded occurrences of questions posed by consultingphysicians visiting the ICU that had been answered inprevious notes (Additional file 5: Table S5).PurposeThe perceived purpose of physician progress notes (i.e.,documentation of patient course) emerged across threemain categories: preserving the patient’s story, decision-making, and documentation of communication.Preserving the patient’s storyAcross all hospitals, ICU notes and general medicalteaching units regularly retained important elements ofthe ‘patient’s story’ (i.e., information reporting patientstatus, relevant medical or social history, patterns thatemerged during care, and future-oriented care plan)[42, 43]. After transfer from ICU, the narrative of thepatient’s story changed as notes were shorter and lessstructured with contextual factors (e.g., relevant history)commonly absent. The lack of continuity regarding thepatient’s story was a key observation during the transitionof care from ICU to hospital ward. For example, patientC-023’s ICU note details a “complicated hospitalizationcourse” following ventral hernia repair that includedmultiple cardiac arrests and cardiac catheterizations, butin contrast, a ward note reported on wound and stapleremoval and the need for a cardiology follow-up(Additional file 5: Table S5).Details of future-oriented care planning (e.g., rehabili-tation needs) were more frequently documented in wardnotes than in ICU notes across all sites. In Hospital H, adesignated goals of care (i.e., resuscitative versus medicalversus comfort) section was auto-populated. Goals ofcare designations were documented in nine of 32 med-ical records (including those from Hospital H), usuallyaround the day of transfer, and specifically for patientswith recently changed goals of care, or for those thathad a recent meeting regarding prognosis.Documentation of decision-makingBoth ICU and ward notes routinely documented clinicalplans, but rarely provided a decision-making rationalefor plans requiring further justification (e.g., medicationsstarted and stopped, and clinical interventions appliedwithout explanation or outcomes) (Additional file 5:Table S5). For example, in the chart of patient 001(Hospital A), “family meeting to discuss prognosis…goalsof care” was documented, however, no follow-up notewas added to describe the nature or outcome of thisconversation. Absence of decision-making documenta-tion was a key observation during the transition of carefrom ICU to hospital ward.Documentation of communicationThe quantitative analyses showed that communication be-tween providers or between providers and patients/familymembers consisted of a small portion of physician progressnotes. When communication was documented, it mostoften appeared as directed provider-provider communica-tion (Additional file 5: Table S5). However, documentedresponses to directed provider-provider communicationwere rare, resulting in many instances of incompletetextual information exchange (Additional file 5: Table S5).DiscussionOur study provides a comprehensive description of text-based communication during the transition of patientcare between the ICU and hospital ward. Three key ob-servations emerged from our analyses:1. Discrepancies in the documentation of patientproblemsObservable differences were present in the problemlists documented by ICU and ward physicians inpatient medical records including a lack ofconsistency between the last ICU note and firstward note. Structure and stylistic differences in theproblem lists may impact continuity of patient care.For example, if care teams are used to sendingand receiving information in different ways(e.g., head-to-toe assessment versus disease focusedapproach), there is potential for loss of importantBrown et al. Critical Care  (2018) 22:19 Page 5 of 8patient information at the point of patient transferdue to the incompatibility of preferred forms ofcommunication. Differences in the continuity ofpatient problems at the point of transfer mayindicate that previous notes are not beingconsistently reviewed by physicians. Barring verbalexchanges to fill gaps in patient information, there isa risk of information loss from the one consistentform of durable patient information available at thepoint of transfer. One strategy to address thischallenge is to adapt the concept of medicationreconciliation [44] for patient problems at times oftransitions of care. Implementation of medicationreconciliation systems has greatly reduced errorrates and changes of medication orders [45].Applying this approach to the handover of patientproblems (e.g., by developing a comprehensiveprioritized list of problems during a period of patientcare, and describing whether the problems areactive, resolved and/or in need of future action atthe point of patient transfer) could pose as apotential solution for reducing the risk of informationloss and the potential impact this has on patient careand outcomes at the point of transfer. EHRs couldfacilitate such a strategy by tracking problems over time.2. Lack of continuity regarding the patient’s storyOur results indicate that continuity of “patient stories”[42, 43] was often not maintained during transitions ofcare. Variation in the focus and purpose of writtendocumentation between ICUs and hospital wards wereidentified as key barriers to maintaining continuity inthe patient’s story. In most cases, after transfer fromICU, less detail was used to describe patient historyand contextual factors that may have been implicatedin the patient’s initial admission to hospital.Consequently, patients who have ongoing issues thatare outside the clinical scope of focus of the physiciansassuming care may be at increased risk of these issuesnot being attended to [5, 6]. Providing patients andfamilies with concurrent access to their medical record[46] could potentially decrease information loss acrosscare settings and empower patients and families tofunction as a source of continuity in maintaining thepatient’s story [47]. In addition, a standardizedcommunication tool that synthesizes important patientinformation by documenting recurrent themes andforcing record keeping of essential aspects of care(e.g., goals of care) is needed to safeguard againstinformation loss during transitions of care.3. Absence of documentation in decision-makingRationale for clinical decisions was rarelydocumented in physician progress notes (i.e., howand why decisions were made), which may beattributed to workload and time available, or anattempt to streamline textual documentation [22].The absence of documented rationale for clinicaldecisions represents an important communicationgap since ICU patients routinely receive care frommany different providers and transitions of care arecommon. One strategy to address this gap would beto add a ‘why’ subheading to the ‘Plan’ portion ofthe SOAP model (i.e., SOAPy) to prompt notewriters to document the rationale for the planselected. EHRs could be effective tools forprompting rationale documentation.Furthermore, while EHRs have been shown to improveaccuracy and readability of progress notes through theautomatic population of routine information (e.g., date,time, signature) and the removal of penmanship errors[36] recent literature has demonstrated that this doesnot necessarily result in an effective capture of the pa-tient’s story [43, 48]. Furthermore, our data show thatclinical impressions and documentation of communica-tion were more difficult to identify at the ICU site thatutilized electronic notes due to the rigid structure andamount of information presented [36, 49]. Accessibilityof retrieving and documenting communication, clinicalimpression, and the patient’s story remains an importantissue that should be addressed in the design and devel-opment of any standardized communication tool (hand-written or electronic) that is utilized during transitionsof care.LimitationsOur study findings have some limitations. First, althoughcommon across Canada, the hospitals included exclu-sively used paper-based medical records for physicianprogress notes, and therefore it is unclear how ourresults would have been influenced by the inclusion ofEHR notes. The discrepancies in documentation ofproblem lists, decision-making and continuity of the pa-tient story are unlikely to be specific to paper-based re-cords. Second, for feasibility reasons, content analyseswere performed on English-language medical records.Third, we only examined notes that were written by phy-sicians. Future research should examine allied healthprofessional documentation for communication.ConclusionsDocumentation of patient problems, decision-making,and the patient’s story, emerged as central points ofdiscontinuity for patients transferred from ICU tohospital ward. This discontinuity can result in the loss ofBrown et al. Critical Care  (2018) 22:19 Page 6 of 8important information during transitions of patient careand ultimately impact the care patients receive. Keystrategies for improving information loss in the context ofvaried approaches to textual documentation (includingstrategies targeted at retaining a thick description of patientstatus, relevant medical and social history, patterns thatemerged during care, and future-oriented care plan) shouldbe implemented. Moreover, future research should investi-gate the implications of different approaches to note writ-ing on patient outcomes (i.e., incidence of adverse events)in an effort to define best practice. Defining standards forthe design and development of both handwritten and elec-tronic progress notes is a top priority for future research.Additional filesAdditional file 1: Number of patient medical records collected perstudy site. Total number of patient medical records and number ofphysician progress notes per study site included in this analysis. (DOC 31 kb)Additional file 2: Legend of patient and site identifiers. Patientidentifiers corresponding to the examples in the content analyses,distinguished by site and type of patient (surgical versus non-surgical).(DOC 31 kb)Additional file 3: Quantitative descriptive analysis of physician progressnotes for medical patients. Physician progress notes for medical patientscategorized according to patient location during the 10-day period.(DOC 45 kb)Additional file 4: Quantitative descriptive analysis of physician progressnotes for surgical patients. Physician progress notes for surgical patientscategorized according to patient location during the 10-day period.(DOC 45 kb)Additional file 5: Themes, subthemes and examples from contentanalysis. Textual examples for each theme and subtheme from thecontent analysis. (DOC 978 kb)AbbreviationsEHRs: Electronic health records; EMG: Electromyogram; ICU: Intensive care unit;REB: Research Ethics Board; SOAP: Subjective objective, assessment, and planAcknowledgementsNot applicable.FundingThis work was supported by the Canadian Frailty Network (CFN) (CORE 2013-12A).AFT is supported by a Canada Research Chair in Critical Care Neurologyand Trauma. SMB is supported by a Canada Research Chair in CriticalCare Nephrology.Availability of data and materialsThe datasets used and/or analyzed during the current study are availablefrom the corresponding author on reasonable request.Author’s contributionsKNB organized the data from the study sites, analyzed and interpreted thedata, as well as drafted the manuscript. JPL aided in interpretation of dataand drafted the manuscript. HK helped acquire data and contributed torevising the manuscript. SMB, RAF, PMD, AFT, AJF, and FL made substantialcontributions to conception and design of the project, oversaw theacquisition of data in their respective sites, and contributed to revising themanuscript. AS aided in data analysis and contributed to revising themanuscript. HTS made substantial contributions to conception and design ofthe project, oversaw the acquisition of data, and drafted the manuscript.All authors read and approved the final manuscript.Ethics approval and consent to participateAll participants provided written and informed consent. The Health ResearchEthics Board at the coordinating center (University of Calgary REB13-0021,which covers four Calgary-based sites) and at each of the study hospitalsapproved the study (UofA Pro00050646; UBS PHC Hi4-01667; Sunnbrook336-2014; QCH 201 40345-01H; Sherbrooke 14-172; Laval 2015-2171).Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher's NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Calgary, Canada. 2Edmonton, Canada. 3Toronto, Canada. 4Vancouver,Canada. 5Québec, Canada. 6Ottawa, Canada. 7Sherbrooke, Canada.Received: 17 October 2017 Accepted: 2 January 2018References1. Sutcliffe KM, Lewton E, Rosenthal MM. Communication failures: An insidiouscontributor to medical mishaps. Acad Med. 2004;79(2):186–94.2. Solet DJ, Norvell JM, Rutan GH, et al. Lost in translation: challenges andopportunities in physician-to-physician communication during patienthandoffs. Acad Med. 2005;80(12):1094–9.3. Greenberg CC, Regenbogen SE, Studdert DM, et al. Patterns ofcommunication breakdowns resulting in injury to surgical patients.J Am Coll Surg. 2007;204(4):533–40.4. Bell CM, Brener SS, Gunraj N, et al. Association of ICU or hospital admissionwith unintentional discontinuation of medications for chronic diseases.JAMA. 2011;306(8):840–7.5. Camiré E, Moyen E, Stelfox HT. Medication errors in critical care: Risk factors,prevention and disclosure. CMAJ. 2009;180(9):936–43.6. McAlister FA, Youngson E, Bakal JA, et al. Impact of physician continuity ondeath or urgent readmission after discharge among patients with heartfailure. CMAJ. 2013;185(14):E681–9.7. Abraham J, Nguyen V, Almoosa K, et al. Falling through the cracks:Information breakdowns in critical care handoff communication.AMIA Annu Symp Proc. 2011;2011:28–37.8. Horwitz L, Krumholz H, Green M, et al. Transfers of patient care betweenhouse staff on internal medicine wards: a national survey. Arch Intern Med.2006;166(11):1173–7.9. Landucci D, Gipe B. The art and science of the handoff: how hospitalistsshare data. Hospitalist. 1999;3(1):4.10. Apker J, Mallak LA, Gibson SC. Communicating in the "gray zone":perceptions about emergency physician-hospitalist handoffs and patientsafety. Acad Emerg Med. 2007;14(10):884–94.11. Kitch B, Cooper J, Zapol W, et al. Handoffs causing partient harm: Asurvey of medical and surgical house staff. Jt Comm J Qual Patient Saf.2008;34(10):563–70.12. Petersen L, Brennan T, O-Neil A, et al. Does housestaff discontinuity of careincrease the risk for preventable adverse events? Ann Intern Med.1994;121(11):866–72.13. Riesenberg LA, Leitzsch J, Massucci JL, et al. Residents' and attendingphysicians' handoffs: a systematic review of the literature. Acad Med.2009;84(12):1775–87.14. Horwitz L, Meredith T, Schuur J, et al. Dropping the baton: a qualitativeanalysis of failures during the transition from emergency department toinpatient care. Ann Emerg Med. 2009;53(6):701–710.e704.15. Pronovost PJ, Vohr E. Safe patients, smart hospitals. New York: HudsonStreet Press; 2010.16. Measuring patient harm in Canadian hospitals. In: Canadian Institute forHealth Information; Ottawa: CIHI; 2016.17. Zinn C. 14,000 preventable deaths in Australian hospitals. BMJ.1995;310:1487.Brown et al. Critical Care  (2018) 22:19 Page 7 of 818. Patterson E, Roth E, Woods D, et al. Handoff strategies in settings with highconsequences for failure: lessons for health care operations. Int J QualHealth Care. 2004;16(2):125–32.19. Gandi T. Fumbled handoffs: One dropped ball after another. Ann InternMed. 2005;142(5):352–8.20. Institute of Medicine. Crossing the quality chasm: A new health system forthe 21st century. Washington, DC: National Academy Press; 2001.21. Eggins S, Slade D. Communication in clinical handover: improving the safetyand quality of the patient experience. J Public Health Res. 2015;4(3):666.22. Koopman R, Steege L, Moore J, et al. Physician information needs andelectronic health records (EHRs): time to reengineer the clinic note.J Am Board Fam Med. 2015;28(3):316–23.23. Wood D. Documentation guidelines: evolution, future direction, andcompliance. Am J Med. 2001;110(4):332–4.24. Weed LL. Medical records that guide and teach. N Engl J Med.1968;278(12):652–7.25. Varelas P, Spanaki M, Hacein-Bey L. Documentation in medical recordsimproves after a neurointensivist's appointment. Neurocrit Care.2005;3(3):234–6.26. Wolff JL, Berger A, Clarke D, et al. Patients, care partners, and shared accessto the patient portal: online practices at an integrated health system.J Am Med Inform Assoc. 2016;23(6):1150–8.27. Walker J, Meltsner M, Delbanco T. US experience with doctors and patientssharing clinical notes. BMJ. 2015;350:g7785.28. Langewitz WA, Loeb Y, Nübling M, et al. From patient talk to physiciannotes - Comparing the content of medical interviews with medical recordsin a sample of outpatients in Internal Medicine. Patient Educ Couns.2009;76(3):336–40.29. Pollard SE, Neri PM, Wilcox AR, et al. How physicians document outpatientvisit notes in an electronic health record. Int J Med Inform. 2013;82(1):39–46.30. Carpenter I, Ram MB, Croft GP, et al. Medical records and record-keepingstandards. Clin Med. 2007;7(4):328–31.31. Kibble J, Hansen PA, Nelson L. Use of modified SOAP notes and peer-ledsmall-group discussion in a Medical Physiology course: addressing thehidden curriculum. Adv Physiol Educ. 2006;30(4):230–6.32. Mehta R, Radhakrishnan N, Warring C, et al. The use of evidence-based,problem-oriented templates as a clinical decision support in an inpatientelectronic health record system. App Clin Inform. 2016;17(3):790–802.33. Rosenbloom ST, Denny JC, Xu H, et al. Data from clinical notes: aperspective on the tension between structure and flexible documentation.JAMIA. 2011;18(2):181–6.34. Forsvik H, Voipio V, Lamminen J, et al. Literature review of patient recordstructures from the physician's perspective. J Med Syst. 2017;41(2):29.35. Walsh C, Siegler EL, Cheston E, et al. Provider-to-provider electroniccommunication in the era of meaningful use: a review of the evidence.J Hosp Med. 2013;8(10):589–97.36. Brown PJ, Marquard JL, Amster B, et al. What do physicians read (andignore) in electronic progress notes? App Clin Inform. 2015;5(2):430–44.37. Abraham J, Kannampallil T, Patel V. A systematic review of the literature onthe evaluation of handoff tools: Implications for research and practice.J Am Med Inform Assoc. 2014;21(1):154–62.38. Stelfox H, Leigh J, Dodek P, et al. A multi-center prospective cohort study ofpatient transfers from the intensive care unit to the hospital ward. IntensiveCare Med. 2017;43(10):1485–94.39. Parsons Leigh J, Brown K, Buchner D, et al. Protocol to describe the analysisof text-based communication in medical records for patients dischargedfrom intensive care to hospital ward. BMJ Open. 2016;6(7):1–8.40. Buchner D, Bagshaw SM, Dodek P, et al. Prospective cohort study protocolto describe the transfer of patients from intensive care units to hospitalwards. BMJ Open. 2015;5(7):1–8.41. Strauss A. Qualitative analysis for social scientists. Cambridge: CambridgeUniversity Press; 1987.42. Rankin JM. The rhetoric of patient and family centred care: an institutionalethnography into what actually happens. J Adv Nurs. 2014;71(3):526–34.43. Varpio L, Rashotte J, Day K, et al. The EHR and building the patient's story: Aqualitative investigation of how EHR use obstructs a vital clinical activity.Int J Med Inform. 2015;84(12):1019–28.44. Grimes TC, Duggan CA, Delaney TP, et al. Medication details documentedon hospital discharge: Cross-sectional observational study of factorsassociated with medication non-reconciliation: Medication detailsdocumented on hospital discharge in Ireland. Br J Clin Pharmacol.2011;71(3):449–57.45. Barnsteiner JH. Medication reconciliation. In: Hughes RG, editor. Patientsafety and quality: an evidence-based handbook for nurses. Rockville (MD):Agency for Healthcare Research and Quality; 2008.46. Delbanco T, Walker J, Bell SK, et al. Inviting patients to read their doctors'notes: a quasi-experimental study and a look ahead. Ann Intern Med.2012;157(1):461–70.47. Parsons Leigh J, Stelfox HT. Continuity of care for complex medical patients:How far do we go? Am J Respir Crit Care Med. 2017;195(1):1414–6.48. Roukema J, Los RK, Bleeker SE, et al. Paper versus computer: Feasibility of anelectronic medical record in general pediatrics. Pediatr. 2006;117:15–21.49. Clarke M, Belden J, Koopman R, et al. Information needs and informationseeking behaviour analysis of primary care physicians and nurses: aliterature review. Health Info Libr J. 2013;30(3):178–90.•  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:Brown et al. Critical Care  (2018) 22:19 Page 8 of 8


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