Electronic charting of transfusion medicine consults: implementation, challenges and opportunities

Abstract

Background

The Joint Commission lists improving staff communication (handoffs) as part of several National Safety Goals. In this study, we developed an electronic web-based charting system for clinical pathology handoffs, which primarily consist of transfusion medicine calls, and evaluated the advantages over a paper-based handwritten call log.

Materials and Methods

A secure online web browser application using Research Electronic Data Capture (REDCap) was designed to document on-call pathology resident consults. A year after implementation, an online survey was administered to our pathology residents in order to evaluate and compare the usability of the electronic application (e-consults) to the previous handwritten call log, which was a notebook where trainees hand wrote different components of the consult.

Results

The REDCap web-based application includes discrete fields for patients’ information, requesting physician contact, type of consult, action items for follow-up and faculty responses, as well as other information. These components have eventually progressed to be an online consult call catalog. With approximately 1079 consults per year, transfusion medicine-related calls account for ~90% of the encounters, while clinical chemistry, microbiology and immunology calls constitute the remainder. The overall response rate of the survey was 96% (29 of 30 participants). Of the 16 respondents who experienced both call log systems, 100% responded that REDCap was an improvement over the handwritten call log (P < 0·0001).

Conclusion

E-consult documentation entered into a web-based application was a user-friendly, secure clinical information access and effective handoff system as compared to a paper-based handwritten call log.

Background

Clinical pathology consultations have been identified as a critical activity in the field and encompass a wide variety of subspecialties including clinical chemistry, microbiology, haematology and transfusion medicine. Regarding transfusion medicine in particular, consults may include blood product management, transfusion reactions, factor product requests, haemapheresis procedures, and reporting of unexpected results and critical values. Unfortunately, clinical data associated with transfusion consults are often inadequately documented in electronic medical records (EMR), and some components may not even enter the EMR as they are usually documented on paper forms [1-3].

The consequences of undocumented or poorly documented transfusion data could have significant clinical consequences such as delays in transfusions, unnecessary or inappropriate transfusions, increases in operating room time and disruption in transition of care as a result of miscommunication [2,4]. Errors can occur at any point of the transfusion process and the stakes are high, because even seemingly minor transfusion errors have propensity to cause catastrophic morbidity or mortality [5,6]. Fatal errors are more likely to occur when there is a failure of appropriate monitoring or when there is unreliable tracking of the performed actions, which is often associated with handwritten records as they are not always readily available [1,7]. Paper-based records are less transportable, may contain illegible entries and are limited in data search and retrieval capacity [1]. The FDA actually recommends the use of electronic data systems in blood centres to trace every donation from collection to final administration of each component, including staff involved [6]. Moreover, medical centres that provide patient-related services including blood transfusions are HIPAA (Health Insurance Portability and Accountability Act) covered entities and thus must ensure and maintain the privacy of patients’ medical information during blood transfusion activities [6].

For more effective handoffs, better provider communication, and HIPAA-compliant documentation of transfusion medicine consults in an easily accessible method, health institutions need to work towards developing electronic, secured and accessible systems while eliminating, or at least reducing, the use of paper-based records. In fact, the National Transitions of Care Coalition (NTOCC) supports the application of a universal tool to facilitate direct communication between health care providers during patient transitions [8]. However, to date, studies evaluating the application of electronic documentation of transfusion medicine consults compared to paper records are limited [9]. We hypothesize that the development of an e-consult charting system will provide a comprehensive description and documentation of clinical pathology call consults as an easily accessible tool compared to paper charting. In this study, we aimed to develop a web-based application and evaluated the usability of e-charting over paper records in documentation of clinical pathology consults, most prominently transfusion medicine handoffs.

Materials and methods

General handoff structure

Pathology residents, covering day and night shifts, log each clinical pathology related call they receive from clinicians. Until the call is resolved, each consult is considered ongoing and would need to be handed off between shifts. At our institution, transition of care occurs twice a day between the day and night residents at 7:30 A.M. and 7:30 P.M. At the evening handoff, the day team is to contact the night resident to inform them which patients may have ongoing issues that need to be addressed at night. The night resident is also expected to log any new calls received at night. At morning handoff, the night resident contacts the day team to inform them of new calls received overnight. This basic handoff structure was used with the handwritten call log and the web-based electronic system. However, one notable difference is that ongoing calls are not only noted as such in the web-based system but are required to be referenced in an e-mail between shifts; this e-mail was not part of the handoff process during the time the handwritten log was in use.

Handwritten call log

Prior to the implementation of the web-based call log, the transfusion medicine residents would use a spiral bound notebook as a handwritten call log. This notebook was physically handed off between residents taking call to document the clinical pathology calls, the majority of which were transfusion medicine-related, they received during their shift from all facilities, which include the main university hospital, four allied hospitals including, the children’s hospital and the city community hospital. Calls mostly centred on transfusion services (e.g. transfusion reactions, haemapheresis requests, massive transfusion management, blood product management, coagulation factor approvals). The notebook was organized in seven columns that contained a limited number of fields with limited space for data entry. See Table 1 for the layout of the call log. Indeed, complex call entries were difficult to capture and follow due to lack of space. All entries were handwritten, and legibility was dependent on the given resident’s penmanship. Additionally, the notebook was worn from the constant exchange between residents. Frequently, there would be crumpled and/or torn pages, as well as pages that had fallen out and were placed back loosely between pages.

Table 1.

Layout of handwritten call log

Web-based application design and E-consult documentation

We designed a secure online web browser application using Research Electronic Data Capture (REDCap; https://www.project-redcap.org; powered by Vanderbilt, Nashville, TN) that allows for the development and management of online databases [10]. It offers methods for tracking data and user activity, and includes built-in project calendars, reporting tools, branching logic, file uploading and calculated fields. Although REDCap is used primarily for research data storage, its HIPAA-compliant security allowed for application in the clinical setting; furthermore, the database is password secure. Using this software with its unique infrastructure, we created an online call log for proper documentation of clinical pathology consults, the majority of which were transfusion related. To determine and characterize primary uses of the online call log, REDCap calls were counted and categorized by type of call.

Using the REDCap web-based application, we successfully created an online consult call catalog, with each call added being assigned a ‘Call ID’ (Fig. 1a). Consult data collection forms were constructed for the most commonly encountered call types: blood product management, transfusion reactions, haemapheresis request, coagulation factor request, critical laboratory values, referral test requests and other laboratory issues (Fig. 1b). In general, each consult report contains standard text fields including: requesting physician contact information, patient information (e.g. name, medical record number, and date of birth), pertinent call details, attending physician comments and call status (Fig. 1b, c). See Table 2 for standard layout of a basic call entry. Based on the call type selected, the report generates call-type specific text fields based on the REDCap system’s branching logic (Fig. 1d,e,f).

Fig. 1.

Screenshots of REDCap call entry data collection forms. Each call is assigned a new Call ID by adding new record. The call log is also searchable by different fields (e.g. Last name and MRN) (a). Call type field provides drop-down of most common types of calls (b). Each call entry, no matter the call type, contains characteristic patient demographic and call detail data entry fields (c). Based on the call type selected, branching logic of database customizes data entry fields to that specific call type (d, e, f).

Table 2.

Layout of standard REDCap collection data form for basic call entry

Field	Entry Options
Primary resident	Drop-down menu: All residents and fellows listed
Secondary resident	Drop-down menu: All residents and fellows listed
Attending	Drop-down menu: All clinical pathology faculty listed
Date and time	Drop-down calendar
Location	Select one: University Hospital Children's Hospital Community Hospital Subsidiary Hospitals Other
Call type	Drop-down menu: Blood Product Management Transfusion Reaction Factor Product Request Hemapheresis Request Massive Transfusion Other Transfusion Medicine Mislabeled Specimen Referral Testing Critical Value Another laboratory Issue
Provider and patient information
Provider name	Free text
Provider contact information	Free text
Patient last name	Free text
Patient first name	Free text
Patient middle initial	Free text
Medical record number	Free text
Date of birth	Drop-down calendar
Age	Free text
Gender	Select one: Male Female Unknown
Weight (kg)	Free text
Call information
Call details	Free text
Call status	Select one:Ongoing Resolved
Attending comments	Free text
For ongoing calls
Requires night/weekend involvement	Select one:Yes No
Pending tasks	Free-text
Date follow-up required	Drop-down calendar
Appropriate contacts	Free text
Form status
Complete?	Select one:Incomplete Unverified Complete Save & Exit button

Using REDCap, patients who still require follow-up can be found in the Ongoing Issues list. The day and night residents can easily review and update the consult form should they get called concerning those patients with pending issues. As calls are marked Resolved, the consult form is closed and is automatically moved off the Ongoing Issues list and is archived into the Resolved Issues list. Users are easily able to search REDCap call log by name, medical record number, date of birth and other patient identifiers independent of Ongoing or Resolved status. As such, the tool can be used while on service not just at transition of care. Of note, clinical pathology fellows and attendings are able to enter calls as well, or they can modify and provide feedback on calls entered by residents.

Study setting, survey design and application

One year after the implementation of the REDCap call log, we performed a cross-sectional survey evaluating the usability of the secure online web browser application for documenting and tracking clinical pathology consults electronically (e-consults). All clinical pathology residents and fellows were invited to participate in the study within our pathology and laboratory medicine department, which serves a tertiary care facility. As this was a programme evaluation and process improvement study, it was deemed exempt from Institutional Review Board submission.

A survey tool with closed and open-ended questions was designed for evaluating the usability of online consult documentation compared to the prior handwritten log for the clinical pathology calls, again, most of which were in support of the transfusion medicine service. The questionnaire was piloted by a clinical pathology faculty group involved in teaching residents. This piloting process aimed to ensure that the questions were not ambiguous and that residents would understand the concept of the survey. The survey consisted of 17 questions with a combination of Likert scale, yes/no questions and open-ended comments (Table 3). The survey was administered via SurveyMonkey (https://www.surveymonkey.com), an online survey software hosted by Emory University School of Medicine (Atlanta, Georgia, USA). All residents were invited to respond to questions concerning REDCap in order to evaluate the general perception of the online format as well as determine areas for improvement; however, only residents who had experience with both the handwritten call log and REDCap were able to respond to questions comparing the two systems. Responses were collected anonymously, and biweekly follow-up messages were sent to encourage survey completion and a robust response rate. Data collection was closed 6 weeks after the survey link was initially e-mailed.

Table 3.

Questionnaire reponses and analysis

Questionnaire Items	Answers		P value
	Yes	No
Do you use REDCap?	100	0·00%	<0·0001*
Do you find the reports useful (ongoing issues, ongoing issues brief, Attending reports, personal reports)?	86·21%	13·79%	<0·0001*
For which clinical services do you use REDCap?
Transfusion medicine	100·00%	0·00%	<0·0001*
Microbiology	42·31%	57·69%	0·4328
Clinical chemistry	64·29%	35·71%	0·1306
Immunology	12·50%	87·50%	0·0002*
Do you think REDCap is…
User-friendly (i·e·easy to navigate)?	82·76%	20·69%	0·0010*
Cumbersome (i·e· slow to navigate, too many fields to fill in)?	31·03%	68·97%	0·0411*
A convenient way to look up past patient encounters?	75·86%	27·59%	0·0106*
Improves patient care?	93·10%	6·90%	<0·0001*
A system that can create additionally work at times?	82·76%	17·24%	0·0004*
A good system for hand-off between night and day residents?	86·21%	13·79%	<0·0001*
A tool which could be useful to other pathology Programs?	89·66%	13·79%	<0·0001*
Do you think the handwritten call log was…
User-friendly (i·e· easy to navigate)?	60·00%	40·00%	0·4386
HIPAA compliant (secure)?	26·67%	73·33%	0·0707
A convenient way to look up past patient encounters?	13·33%	86·67%	0·0045*
A good system for hand-off between night and day residents?	20·00%	80·00%	0·0201*
Supplied sufficient information pertaining to calls (i·e· patient history, contact information, updates)?	33·33%	66·67%	0·1967
Did you ever leave the written call manual in an unsecured area?	31·25%	68·75%	0·1336
Overall, do you think REDCap is an improvement over the use of the handwritten call log?	100%	0·00%	<0·0001*

Advantages	Disadvantages
Promotes transparency between resident and attending	Requires a device
Allows attending to keep better track of decisions resident is making	People tend to be too verbose in their narratives·
Creates more comprehensive documentation of calls than paper log, which could be useful during retrospective review if needed	Lowers barrier for massive copy and paste offenses (copied content that adds no additional information)·
More secure (HIPAA compliant) and less cumbersome (don't have to carry a log book around)	Too much data must be entered into the fields which is slow and labor intensive·
Convenient for call updates and follow-up	More time consuming to enter information into REDCap in comparison to the paper log book

^*Indicates statistical signficance (P < 0·05)

Statistical analysis

Binary responses (yes vs. no) in the survey were analysed as percentages. A binomial test was employed to analyse the differences between the proportion of responses to every question in the survey, in addition to an overall proportion difference between yes and no using SAS® software (version 9.4 Cary, NC, USA); P < 0·05 was considered significant. Open-ended comments to learning objective questions were analysed only if they included measurable objectives. For observational data reporting, in cases where missing data in each category listed were greater than 10%, the data were excluded from the tables and observational analyses.

Results

Implementation of online consult call catalog

Since implementation of the REDCap call log, we have averaged approximately 1079 consults per year. Approximately 90% of calls have been directly related to transfusion medicine, with the remaining calls coming from other areas of clinical pathology (e.g. clinical chemistry, microbiology and haematology). Calls received came from hospitals covered by the clinical pathology residents including the main university hospital system (45%), the city community hospital (33%), the children’s hospital system (21%) and the smaller university associated hospitals (1%).

In terms of transfusion medicine-related calls, blood product management issues which consisted of calls related to special product requests, crossmatched/HLA-matched platelet requests, blood product triage and deviations from standard operating procedures constituted the majority of calls (32%). Transfusion reactions represented the second most common call type (18%), followed by haemapheresis related issues (16%). Factor product requests (e.g. prothrombin complex concentrates, recombinant factor VII, and antithrombin concentrate) and massive transfusion calls were the next most frequent encounters (10% and 8%, respectively). The remaining calls were classified as ‘other’ (16%).

Calls that were not directly related to transfusion (11% overall) consisted of referral testing (21%), critical values (13%), mislabelled specimens (7%) and other laboratory issues (59%). Although not all field entries were filled out, only ~4% of fields were left blank. Examples of fields left blank include: transfusion reaction classification (left blank on 210 call entries with no final diagnosis explicitly selected from the drop-down menu), blood product management type (left blank on 308 data collection forms) and factor product requested (left blank in 36 instances). Finally, of the total calls logged, only 138 (1·8%) had comments from an attending physician.

Survey responses and analyses

Thirty questionnaire survey links were e-mailed to participants. Respondents included residents at different postgraduate year (PGY-1, 2, 3 and 4) and clinical pathology fellows. The majority (79%) of the respondents were PGY-2, PGY-3 and PGY-4 residents. Of the 30 surveys sent out, 29 were completed, an overall response rate of 96%. Of the respondents who had experience using both call log systems, handwritten and online (n = 16/29), 100% agreed that REDCap was an improvement (i.e. preferred system) over the use of the handwritten call log, P < 0·0001. The overall proportion of yes responses were significantly higher than those of no (71% vs. 29%), P < 0·0001. Table 3 provides the questions of the survey and a summary of the responses. Table 3 also documents the advantages and disadvantages of the online call log as captured by the open-ended questions within the survey.

Most residents (86%) responded that they use the REDCap database daily or often when they are on a clinical pathology rotation or clinical pathology call. Residents were also asked to compare their use of the online call log to the previous handwritten call log. According to their responses, 56% of residents were more likely to actually record calls on the online call log vs. the handwritten call log. Also, 56% of residents looked up past encounters and updated current encounters/management plans much more often on the online call log compared to the handwritten call log. The increased usage of online clinical call log was due to its user-friendly format, indicated by almost 83% of those surveyed (Table 3).

Discussion

We demonstrated usability and advantages of e-charting in documenting clinical pathology consults, mostly related to transfusion medicine. Inter-departmental consultations are integral parts of clinical pathologists’ daily professional practice, and the transfusion medicine service is vitally important in performing both diagnostic and therapeutic patient care activities [11]. The documentation required at each stage of the transfusion process should be ‘user-friendly’ to encourage compliance by busy physicians and should have the potential to reduce errors by clinical staff moving between hospitals [12,13]. The online application of REDCap allowed better documentation compared to the handwritten call log.

Besides being cumbersome to share, not readily transportable, and difficult to read illegible entries, [1] we further demonstrated that handwritten documentation of transfusion medicine calls is not a HIPAA-compliant method as 31% of residents admitting to leaving the handwritten log in unsecured areas. Another study emphasized the limited access of paper records, demonstrating that 25% of the time paper charts may be missing when they are clinically needed for simultaneous use by another provider [1]. In our study, 68·97% of the responses documented that paper charting is slow to navigate and did not supply enough information pertaining to calls. Given the limitations of paper-based records, the application of e-charting/e-consult over paper records is considered preferable and easy to track. In addition to improved documentation and assessment, studies have shown that e-charting implementation in patient blood management supports appropriate practice of blood transfusion, enhances communication and helps promote standardized documentation of patient care [1]. Standardized constructions of electronic documents have the potential to enhance data mining, quality improvement and monitoring activities as well [1,14]. Using the simple search functions, residents can use the REDCap call log to refer to a patient’s previous encounters and investigate past management, which could help formulate a care plan for current and future encounters.

The application of healthcare information technology to transfusion medicine and haematology services is gaining popularity, as more institutions embrace appropriate patient blood product management and utilize their EMR to ‘hardwire’ appropriate transfusion practices [15-17]. A recent study by Cecchin et al. anonymously surveyed the effectiveness of e-consults in clinical haematology within a Veteran’s Affairs (VA) system and compared 302 e-consults to 305 face-to-face consults. Of the provider survey respondents, 100% were satisfied with e-consults; no negative outcomes were reported [15]. Comparably, our study results demonstrated user satisfaction with 82·76% of residents responding that the REDCap database was user-friendly, and 86·21% found it to be a very good system for handoff between night and day residents. This is critical because unanticipated medical errors often occur in ‘real time’ during nights and on weekends [18]. The proper transfer and receipt of patient information between different levels of patient care ensures continuity and promotes successful treatment plans [8].

Implementation of e-consulting and e-charting facilitate appropriate tracking of the patients’ care plan and improve patient safety through complete documentation of all performed activities, while highlighting what still needs to be done prior to closing the consult. At least one other pathology program has developed and published on a similar online call log for clinical pathology residents [9]. The University of Washington used Windows Access to build their own homegrown online call log to aid in documentation of consults with an aim of helping to improve patient care. Moreover, the authors point out that the online call log also serves to archive a residents’ experience and can thus be used to monitor the residents’ progress on benchmarks set forth by accrediting bodies, such as the six core competencies (patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism and systems-based practice) outlined by Accreditation Council for Graduate Medical Education (ACGME). The REDCap software is user-friendly, allowing for modifications that could be harnessed to address the core competencies or could be adapted by other institutions to create a call log to suit their specific needs.

Though there are numerous benefits to the REDCap database, it is not without its limitations. There is a challenge with compliance in filling out the data collection forms for each call in their entirety: not all fields were completed by our residents who left some fields blank, most notably the call type. Missing data fields hinder the search function of the call log as well as limit the potential use for research applications. With only 1·8% of call entries containing attending comments, increasing faculty participation is a future goal with hopes to improve patient care and capitalize on the educational facet of the program. We have also yet to explore using the available REDCap mobile application, which could improve the call log’s accessibility and documentation compliance. One final limitation is the annual fee associated with maintaining a REDCap database, a cost that might preclude its use at some institutions.

We conclude that with comprehensive e-consult documentation in REDCap, we were able to successfully incorporate transfusion medicine data into a web-based application and offer a user-friendly experience with more effective handoffs. The tool improved provider communication, with strong potential to improve patient care by securing medical information and safety. We believe, when feasible, e-consult documentation should be considered and implemented in medical systems to allow for better transition of care and management of transfusion medicine services.