integrating remote follow-up into electronic health records workflow
TRANSCRIPT
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/hlpt
Health Policy and Technology (2014) 3, 126–131
http://dx.doi.org/12211-8837 & 2014 F
☆This article was pfrom Jo Ann LeQuanpaid by St. Jude Meddescribed in this castudies used in theSt. Jude Medical or
nTel.: +1 970 297E-mail addresses
Integrating remote follow-up into electronichealth records workflow$
Lisa Diederichn, Timothy Johnson
Colorado Health Medical Group, Cardiology, 2121 E. Harmony Road, Fort Collins, Colorado 80528, USA
Available online 4 February 2014KEYWORDSRemote follow-up;Remote monitoring;EP clinic workflow;Electronic healthrecord;Clinic workflow;Pacing
0.1016/j.hlpt.2014ellowship of Postg
repared with medg of LeQ Medical,ical. St. Jude Medse study and asspaper. No com
another source wa6877.: LisaDiederich@u(L. Diederich).
AbstractBackground: The Colorado Health Medical Group, Cardiology (formerly Heart Center of theRockies) is a paperless clinic which follows about 4000 cardiac rhythm management devicepatients and relies heavily on advanced remote follow-up and remote monitoring features.Methods: Working with a device manufacturer and our own information technology team, wewere able to redesign our workflow procedures so that data downloaded from devices could beimported directly into our electronic health records with no need to manually populate anytemplates. We were able to import device data as discrete information, making records morereadily searchable.Results: With this revised workflow, we were able to transition from about 19 min per patientto about 3.5 min with no compromise in patient care. The new workflow allows device nurses tospend more time engaging with patients and reviewing clinical data and less time managingclerical tasks. Although not captured in this study, our improved workflow also helped usschedule all reimbursable remote and in-clinic follow-ups without fail.Conclusions: With technical support from the institution and the device company, deviceclinics can develop systems to import remote follow-up data directly into electronic medicalrecords allowing for more efficient workflows and time savings.& 2014 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.
.01.002raduate Medicine. Published by El
ical editorial and design helpAngleton, Texas, which wasical assisted in the processesisted in some of the timingpensation of any kind froms provided to the authors.
chealth.org,
Introduction
The demands of periodic follow-up of implantable cardiacrhythm management devices, such as pacemakers, implantablecardioverter-defibrillators (ICDs), and cardiac resynchronizationtherapy (CRT) devices with and without defibrillation, repre-sents a substantial workload for the healthcare teams caring fordevice patients. With increasing numbers of device patients,economic pressures on clinics to consistently do more with less,and highly complex new devices and therapeutic options,managing device patients has created substantial workflowproblems [1]. Up until recently, device follow-up required
sevier Ltd. All rights reserved.
127Integrating RFU data into EHR
regular in-clinic visits to interrogate the device, downloaddiagnostic reports, and adjust parameter settings, as needed.This required patients to report regularly to the clinic and,as these numbers increased, it placed a burden on theclinical staff.
Remote follow-up is defined here as the ability to conducta routine follow-up check of an implantable device by transmis-sion of device data to a secure website. Remote monitoring,defined as the ability of the device to self-report a remarkableevent (such as, but not limited to, atrial fibrillation orimpending end-of-service) to a secure website, advancedpatient safety. Clinical studies have confirmed that remotefollow-up and remote monitoring are safe and effective ways ofmanaging device patients [2,3]. Radiofrequency (RF) or “wire-less” devices further enhanced remote follow-up, becausewireless systems meant patients were not obligated to periodi-cally interact with a transmitter to relay information from theirimplantable device to the clinic's secure website. Wirelessremote monitoring allows for alerts about remarkable eventsto arrive at the clinic promptly, so that appropriate observationor intervention can occur without delay.
Remote follow-up and remote monitoring options haveindeed sped data into pacemaker clinics; however managementof these data has become a formidable challenge. Clinics usingpaper records might simply download the data, print out thereports, and add them to the patient's chart. This is a labor-intensive and time-consuming task. Clinics already using elec-tronic health records (EHRs) may still struggle with compat-ibility issues when transitioning to remote follow-up. Clinicsmay find it necessary to have all or part of this informationmanually input into the EHR. This is not only expensive andtedious, it can lead to data entry errors. Overcoming this hurdle—allowing the remote follow-up data to automatically migrateinto the EHR—will increasingly be a subject of future concernfor cardiac rhythm management device clinics around the world[4]. This article presents a case report from a large U.S.-baseddevice clinic that has achieved integration of remote follow-updata from secure websites or in-office device checks directlyinto EHRs, with minimal but clinically relevant interaction bythe staff. This integration has created a streamlined, econom-ical, and highly efficient workflow.
Background of the clinic
The Colorado Health Medical Group, Cardiology, formerly HeartCenter of the Rockies (Fort Collins, CO) first adopted remotecare of device patients in 2006 and today manages about 4000device patients, including patients from rural areas of Colorado,Wyoming, and Nebraska. The clinic treats patients with pace-makers, ICDs, and CRT systems from all manufacturers. Somepatients are “snowbirds,” who travel from the Rocky Mountainregion to Southern states in the winter. When remote care ofdevice patients was first introduced, patients were given theoption of participation. Today, the clinic automatically enrollsall device patients in remote follow-up. With some systems(Merlin@homeTM, St. Jude Medical, Sylmar, CA or CareLink™
Medtronic, Minneapolis, MN) the clinic is able to pair patientswith remote monitoring equipment at implant, so that they aredischarged from the hospital with the remote system. Remotefollow-up and monitoring have become the clinic's standard ofcare. The clinic uses Allscripts EHR.
Workflow
Original workflow
This workflow was developed for treating patients using thenewest-generation of devices with wireless remote monitor-ing from St. Jude Medical. The original workflow is depictedin Figure 1. A full-time equivalent (FTE), in our case aregistered nurse hired as a contract employee, was engagedto review the device clinic schedule in advance andmanually populate a proprietary template in the EHR. Thetemplate was an electronic form created by our clinic, intowhich the FTE entered information about the device,including programmed settings. This step was necessarybecause the copy-forward feature in the EHR did notprovide the needed information (customization issues).The template was created to integrate electronically intothe EHR. The initial concept of the template was to createan efficient pathway for device data to be incorporated intothe EHR, but the only way to populate the template was byhand. It took between nine and 10 min per patient to fill outthe template, depending on the device.
Next the patient's “red chart” was pulled. (We are apaperless clinic, but we still initiate sessions using a paper“red chart,” a quick reference tool with basic device datadeveloped for use in outreach clinics.) From a timing point, theEHR template may have been prepared minutes or days inadvance. The device nurse logs onto both the EHR and thedevice website. The device nurse downloads and reviews thefollow-up data and acknowledges the patient in the clinic'sschedule, setting up the next remote follow-up. This actiontypically takes fewer than 3 min. The device nurse thenpopulates the proprietary template with the transmitteddata—this takes three or 4 min. The transmission data are alsoarchived electronically on the device company site. In ourclinic, we contact our patients by telephone to tell themfollow-up was conducted. In most instances, the phone call isbrief and informs the patient that the device is operatingproperly. At this time, the device nurse also asks about thepatient's health and can respond to any questions the patienthas. If there is a problem that might necessitate the patient'scoming to the clinic, an appointment is made. In our workflowstudies, we timed several workflows, but the patient phone callis variable; it typically lasts a few minutes but can last longer.
The device nurse signs off on the report in the EHR and,in the lingo of our EHR system, “tasks” (sends a message to) thephysician with a report to sign. In our clinic, the physician mustsign off on all such reports by the close of the business day. Thedevice nurse then goes to the EHR and enters the diagnosis andbilling codes. When the physician sign-off is complete, thisgenerates a billing, and the “red chart” is returned to the fileroom. In other clinics, it may be that the device reports areprinted out, signed by the physician, and scanned back into thesystem; device reports may also be available in some systems asPDF files which can be attached to the EHR.
This original workflow timed out to be about 19 min perdevice follow-up. See Figure 1. This represented a con-siderable amount of time for a “remote follow-up” thatshould have offered us time-saving efficiencies. Further-more, our use of this method required us to hire an FTE tohelp manage the integration of remote follow-up data intoEHR, a necessary but expensive addition to our clinic.
Figure 1 Remote device interrogation workflow “before” the transition.
L. Diederich, T. Johnson128
New workflow
The fundamental change we made to our workflow wasto migrate the downloaded data from the remote websitedirectly into our EHR and avoid the costly, time-consuming stepof populating each template manually. To this end, our clinicworked with the Information Technology (IT) department andwith a cardiac rhythm management company (St. Jude Medical,Sylmar, CA) with a remote system (Merlin.net Patient CareNetwork or PCN) allowing for seamless entry of data directlyfrom the remote website into our EHR. The Merlin.net PCNsends an HL7 message to the interface engine of the EHRsystem. This generates a “task” to the staff for verification;once the “task” is verified, the interface engine populatesabout 125 relevant items first into the EHR and then into thenote. (Of the large amount of information the Merlin.net PCNsystem transmits, our clinic had selected 125 clinically relevantitems for the report; it would be possible to include more orfewer items). The electronic transfer of data eliminates thepossibility of user-input error and circumvents the need for theproprietary template. This creates the report that the devicenurse reviews and on which the physician signs off. A particularconcern in this step was that data had to be entered andretrievable as discrete data; this would enable the staff tosearch and retrieve records for patients with particular devicemodels or specific conditions, important priorities for the clinic.This overcame a previous problem at the clinic, namely that it
was sometimes difficult and time-consuming to retrieve allpatient records for, say, a specific device. See Figure 2.
In this new workflow, Medical Records pulls the red chart,and then the device nurse initiates the data transmissionfrom the device website and populates data electronicallyafter acknowledging the patient in the schedule. These dataare then exported to the EHR from the device website andare verified in the EHR system itself. The downloadedtransmission results are sent to both the EHR and thephysician. The device nurse reviews the pertinent data.The physician signs off on the patient transmission promptly,at the very latest by the close of business.
Similar to the old workflow, the device nurse then calls thepatient to confirm the remote follow-up, and updates theschedule on the device website. It remained a priority at ourclinic to have a staff member personally contact patients; itwould have been possible to have automated or semi-automated the phone call with this system. A scheduling stepoccurs; we set up the next remote follow-up at a 91-dayinterval and made the next “recall” or in-clinic appointment.
The new workflow takes about 3.5 min per patient, an81% time reduction from our old workflow of 19 min perpatient. Essential to our clinic, all of the key clinical steps inthe workflow remain preserved: data were reliably capturedand archived, both a device nurse and the physicianreviewed and signed off on the patient's follow-up report,the patient was personally contacted by phone, and the
Figure 2 New workflow has dramatically reduced time to 3.5 min for a device check, while minimizing the chance of user-inputerrors.
Table 1 Time in minutes to handle each step of theworkflow (per patient).
Template Transmission Total
St. Jude Medical 0 3.5 3.5Other 9.5 9.5 19
129Integrating RFU data into EHR
next follow-up was scheduled as the current one was billed.Furthermore, we minimized the opportunity for errors to beentered into our patients' chart through manual data entry,improved the searchability of our patient records, and savedabout 15 min per patient.
The transition to the new workflow became a clinic-wideproject with active participation by the entire team anddirect involvement with the manufacturer. There wasmutual education involved as our device nurses shared withthe IT team which device data were clinically relevant, andthe IT team communicated to clinicians how the new systemwould operate. With no existing models of this sort ofsystem to learn from, we collaborated with St. Jude Medicalto design the most efficient workflow.
The new workflow was readily embraced when it becameapparent how much faster and easier it was. Importantly, itdispensed with the clerical work of data entry and allowed ournursing staff to perform higher level work, namely to reviewclinical reports and engage with patients. Device nurses couldhandle more patients per day with no sacrifice in clinicalcare. Moreover, this new method has the potential to reducedata-entry errors and to make it easier to electronically searchand retrieve patient data.
Financial results
In 2011, St. Jude Medical remote device checks (n=796)composed 20.5% of all the device checks carried out at ourclinic, yet represented only 4.5% of our clinic's total
workflow. In calculating remote workflow time, the St. JudeMedical process was by far the fastest. See Table 1. Theimplications of this change on reimbursement are substan-tial. Calculating reimbursement per workflow hour (andbased on a registered nurse doing nothing but remotedownloads), St. Jude Medical accounted for $1216.87compared to $254.64 for non-St. Jude Medical systems.Since device nurses have many other tasks, this number istheoretical rather than actual, but it underscores thesubstantial value this new workflow represents.
Wireless remote systems also allow clinics to performall reimbursable device checks with minimal burden to theclinic or patient. Current reimbursement regulations allow forfour remote device checks annually, but quarterly devicechecks can be missed because of scheduling errors, lack ofpatient compliance, or other factors. Wireless remote systemsand an efficient workflow make it much easier to maintain thespecified follow-up schedule. Thus, there is no doubt afinancial benefit in terms of not missing device checks, butthat aspect was not calculated as part of this study.
L. Diederich, T. Johnson130
Discussion
In the 2010 TRUST study, ICD patients (n=1339) wererandomized to remote follow-up (BIOTRONIK, HomeMonitoring, Lake Oswego, OR) or conventional in-clinicfollow-up. Conventional patients, serving as the controlgroup, were followed with an in-office interrogation every3 months; remote follow-up patients were checked remo-tely every 3 months with an in-office follow-up at 3 and 15months only [2]. Over the course of the study, 46 patientsexperienced a total of 62 device-related problems. Deviceissues were significantly more frequently detected inremote than conventional patients (4.4% vs. 1.4%,p=0.004) and the average time of onset between theoccurrence of the problem and its evaluation was signifi-cantly shorter for remote than conventional patients(4.479.2 days vs. 23.6740.2 days, p=0.05). This studyshowed that remote follow-up conferred benefits to devicepatients. Other studies have demonstrated the value ofremote monitoring of unusual events [5,6], but the TRUSTstudy was limited specifically to remote follow-up.
Remote follow-up is not the same as an in-clinic visit wherethe healthcare team can observe the patient, update medicalhistory, and make changes to the device's programmablesettings. The “totally remote” patient does not exist. Remotefollow-up, at most, reduces the need for multiple in-clinic visitsa year. Remote follow-up cannot adjust device settings, but theATHENS study of device patients (n=3362) found that over4 months, only 22.8% of patients required parameter settingadjustments or a change in prescribed medication [7]. Mostdevice clinics know from their own experience that themajority of device follow-up visits do not result in clinicalintervention. When remote follow-up information reveals that achange in device settings or prescription adjustment is neces-sary, the patient can be brought to the clinic.
While the literature now supports the safety and efficacyof remote follow-up, it is important to determine if it isclinically efficient and if both patients and clinicians will besatisfied with the new procedures. Early adoption of remotefollow-up may have resulted in unexpected boluses of workas clinics struggled to manage electronic data porting intopaper systems or dealt with compatibility issues amongelectronic systems. The challenges facing clinics that are ator near the process of adopting EHR can seem overwhelm-ing. It has been our experience that despite the hurdles wefaced in first going paperless and then changing devicefollow-up systems, streamlining workflow, and integratingEHR to remote follow-up have the potential to streamlineworkflow with no compromise to patient care, safety,clinical confidence, or reimbursements.
There are obstacles to achieving this type of seamless,efficient workflow. Some of the challenges involve informa-tion technology, but such challenges can be met with properprioritization of resources. Other challenges involve patientadoption of remote follow-up. In this case study, patients ortheir families were given St. Jude Medical wireless remotemonitoring equipment at the point of discharge from thehospital, including a cellular adapter. It has been ourexperience that giving this equipment and instructions topatients at or near the point of implant causes them toregard remote monitoring as an essential part of living with
an implantable device. At our clinic, we provide all devicepatients with a cellular adapter (even if they currently havea landline), because the cellular adapter frees patients fromdependence on a landline for transmission.
Compliance is crucial. In the past, we arranged to haveremote monitoring equipment mailed to our device patientsweeks after implant, along with an instruction manual forsetting up the system; not surprisingly, our compliance rateswere much lower. Patients who get remote monitoringequipment in the mail, apart from the clinic, and weeksor months after hospital discharge tend to view remote careas an “option,” and may be intimidated by the unfamiliartechnology and unwilling to read through the manual tolearn how to use it. Today, we arrange that device patientsor their families receive the equipment in a “plug-and-play”state before they leave the hospital along with some short,simple instructions on how remote follow-up works. Byproviding the equipment and one-on-one training uponhospital discharge, patients seem to be more accepting ofremote care as an integral part of their new life as a devicepatient.
Studies indicate that device patients like remote follow-up/monitoring and recognize its benefits [8,9]. Newpatients at our clinic are informed that the remote systemis part of our standard of care.
Our staff was intensely involved in the re-invention ofour workflow and, in some ways, even led the effort tointegrate remote follow-up into EHR. It would be challen-ging to make such a significant change to workflow at theclinic without the full awareness and support of the entirestaff. In order to help set up our system, we relied on thehelp of the manufacturer, who pioneered ways in whichremote follow-up and electronic records can work together.
Remote follow-up is typically considered for ICD and CRTpatients, but it is our experience that pacemaker patientsmay need remote follow-up and monitoring as much or morethan patients with more sophisticated devices. In theATHENS study, the need for an intervention (device settingadjustment or prescription change) occurred in 22.8% ofdevice patients, but it should be noted that the need foraction varied by device: it occurred in 29.8% of CRT patientscompared to 22.8% of pacemaker and 18.6% of ICD patients[7]. These data suggest that pacemaker patients may be“sicker” than ICD patients, even if they have less techno-logically advanced devices. Our anecdotal observations arethat pacemaker patients can benefit from the highest levelsof care, including remote monitoring, since they often arethe patients who are at high risk for arrhythmias or otheradverse events.
Finally, it should be noted that there are many EHR systemsavailable today and compatibility across applications can beproblematic. While it may be anticipated that these electronicsystems will become better integrated in the future, it is notthe case today. An important goal for the future is to havebetter portability of EHR across systems.
Furthermore, our transition from the older to the newworkflow was facilitated by direct cooperation with collea-gues at St. Jude Medical and the technological featuresof their products. Without their direct involvement andthe capabilities of their systems, our transition would nothave been possible. Based on our experiences with St. Jude
131Integrating RFU data into EHR
Medical, we are now following Boston Scientific and Med-tronic devices in the same way.
Conclusion
In our clinic, we were able to integrate wireless deviceremote follow-up data directly into our EHR system,streamlining our workflow and reducing per-patient remotefollow-up time from about 19 min to about 3.5 min. Thistransition was accomplished with an in-house IT team and asolution that transmitted about 125 pieces of discrete data.This made our patient records electronically searchableand omitted the potential for data-entry error. The entireclinical team was involved in the transition, which hasreduced clerical workload and provided greater efficiencieswith no compromise to patient care.
This transition has created a “closed-loop workflow.” Thewireless device transmits daily to the clinic, resulting inhigher patient compliance; this generates more data intoour systems, but the new system can more readily importand manage these data. Thus, we have a higher volume ofdata without increased risk of errors and without increasedoverhead. The loop is closed because the wireless transmis-sion is automated through the EHR to proceed to instantadjudication to billing.
Acknowledgments
The authors acknowledge the assistance of St. Jude Medicalin creating the workflow diagrams. This paper was preparedwith editorial support from the Jo Ann LeQuang of LeQMedical, Angleton, Texas, whose services were paid bySt. Jude Medical.
References
[1] Young GD. Remote monitoring for implantable cardiac electro-nic devices. Heart Lung Circ 2012. (Epub 2012/04/24).
[2] Varma N, Epstein AE, Irimpen A, Schweikert R, Love C. Efficacyand safety of automatic remote monitoring for implantablecardioverter-defibrillator follow-up: the Lumos-T SafelyReduces Routine Office Device Follow-up (Trust) trial. Circula-tion 2010;122:325–32.
[3] Mabo P, Victor F, Bazin P, Ahres S, Babuty D, Da Costa A, et al.A randomized trial of long-term remote monitoring of pace-maker recipients (The COMPAS trial). Eur Heart J 2011. (Epub2011/12/01).
[4] de Cock CC, Elders J, van Hemel NM, van den Broek K, vanErven L, de Mol B, et al. Remote monitoring and follow-up ofcardiovascular implantable electronic devices in the Nether-lands: An expert consensus report of the Netherlands Society ofCardiology. Neth Heart J 2012;20(2):53–65. (Epub 2012/01/18).
[5] Spencker S, Coban N, Koch L, Schirdewan A, Muller D. Potentialrole of home monitoring to reduce inappropriate shocks inimplantable cardioverter-defibrillator patients due to lead fail-ure. Europace 2009;11(4):483–8. (Epub 2008/12/24).
[6] Crossley GH, Boyle A, Vitense H, Chang YF, Mead R. TheCONNECT (clinical evaluation of remote notification to reducetime to clinical decision) trial. J Am Coll Cardiol2011;57:1181–9.
[7] Mascioli G, Curnis A, Landolina M, Klersy C, Gelmini GP, Ruffa F.Actions elicited during scheduled and unscheduled in-hospitalfollow-up of cardiac devices: results of the ATHENS multicentreregistry. Europace 2011;13(12):1766–73. (Epub 2011/07/19).
[8] Ricci R, Morichelli L, Quarta L, Sassi A, Porfili A, Laudadio M,et al. Long-term patient acceptance of and satisfaction withimplanted device remote monitoring. Europace 2010;12(5):674–9.
[9] Gramegna L, Tomasi C, Gasparini G, Scaboro G, Zanon F,Boaretto G, et al. In-hospital follow-up of implantable cardio-verter defibrillator and pacemaker carriers: patients' inconve-nience and points of view. A four-hospital Italian survey.Europace 2012;14(3):345–50. (Epub 2011/11/15).