Clinical Challenges and Opportunities With Current Electronic Health Records

Practicing Oncologists' Perspective

Debra Patt; Philip Stella; Linda Bosserman


J Oncol Pract. 2018;14(10):577-579. 

Oncology clinicians have five main roles: diagnosis, care planning and management, physical examination, procedures, and the development and use of evidence to continuously improve health outcomes. As we work to incorporate rapidly expanding and increasingly expensive diagnostic and therapeutic options into individual care plans for patients living longer, our fundamental clinician roles have increasing multistakeholder interest and oversight. The electronic health record (EHR) needs to facilitate physician roles while empowering individualized precision medicine, incorporating patient preferences and payer requirements, as well as providing insights into clinical, financial, research, and quality health outcomes.

Our current EHR systems, however, were primarily built as billing systems that ensure regulatory compliance. With expanding clinical needs, the systems have added features to meet many other operational objectives. Although these enhancements to the EHR have increased functionality and usefulness of the record, they have also caused inadvertent delays in care and increasingly meaningless clinician workloads and have challenged team-based care aimed at freeing clinicians to think and engage with patients to develop and manage optimal care plans and health outcomes.[1] The impact of these increasing mental and physical demands on clinicians has been a marked increase in burnout by oncologists.[2]

American Medical Association collaborations report that 80% of burnout is due to organizational factors, with the major contributors being high workloads, workflow inefficiencies, increased time spent in documentation, loss of meaning in work, social isolation at work, and the cultural shift from health values to corporate values. For now, each of these six areas is negatively affected by our current EHRs.[3,4]

Taking it a step further, the American Medical Association programs divide burnout solutions into three major areas: culture of wellness, efficiency of practice, and personal resilience. The issues of practice efficiency cover the need to "improve workflow efficiency and maximize the power of team-based care" as well as the need to "reduce the clerical burden and tame the EHR"[5] Additional time spent with the EHR is considered the wrong work, because it prevents clinicians from deeply thinking about and interacting with their patients, networking, studying or researching optimal care plans, and coordinating care that can most improve the health outcome of those they are passionate about serving.

With systems designed largely by engineers and the reality of huge financial penalties for not complying with regulations initially developed to limit practice volumes, how do we get it right? To evaluate our EHR information systems, we can measure their benefit effects in system functionality, performance, and security; content; service characteristics; satisfaction; and net benefits to care.[6] Many of these functionalities have developed in oncology-specific EHR platforms to enhance the benefits of the system, although with variable efficacy.

System benefit effects include functionality enhancements like order entry and decision support; performance characteristics such as remote accessibility, reliability, and system response time; and security. System enhancements that are oncology specific include clinical pathways embedded as clinical decision support systems within the EHR that are linked to orders and partnered with dosing, supportive care protocols, and patient education. Other clinical decision support system tools include clinical trial identification software and predictive analytics populated with information already within the health record to assist clinicians in making real-time decisions about patient care. Functional links to therapy choices that can populate therapy choices and manage inventory will offer operational efficiencies to practices and reduce errors. System enhancements can also reach to the patient from interoperable EHRs to patient portals or personal devices for meaningful engagement, such as patient-reported outcomes instruments, disease-specific education, and interval triage management.

Content enhancements have also grown in importance for oncology as content knowledge has changed dramatically over time, with a need for sequential details on cancer types, stage, biomarkers, and new treatments as therapy decisions are based on increasing complexity. Point-of-care medical reference information such as UpToDate or guidelines embedded within the EHR benefits clinicians and patients to make access to knowledge for decision making more immediate.

Service-related issues, such as speed, updates, ease of data entry, availability of informative reports and dashboards, programming, training, and support, can all improve the impact of an EHR system in the cancer clinic. User satisfaction, including user competency with systems, user friendliness, ease of learning the system, and how well the EHR supports patient care, can influence the effects of a system on practice.

Other net benefits may include quality or the measure of quality, safety, other health outcomes, efficiency, improved health system management capability, care coordination, and financial benefit. With the repeal of the sustainable growth rate formula and advent of the Medicare Access and CHIP Reauthorization Act, the quality payment programs including growth of alternative payment models for commercial and public payers have made these aspects increasingly more important.

As we seek to increase the functionality of EHRs in oncology in this way, we fail when we only consider the engineering software solutions; we must consider design engineering, clinical process workflow, and the care of humans as we continue to grow the functionality of these systems. We require careful consideration and a recognition that the data needs for value-based care need to supplant the volume-control data and regulatory needs of the past. Without changing workflows to engage patients and staff in data collection and documentation, we will continue to push physician burnout, which has significant workforce costs while negatively affecting patient care. We need to better partner the promise of care delivery that EHRs can facilitate to improve efficiency in clinical care.

Our current EHR problems boil down to six problems:

Data Entry: We need to standardize accurate, discrete entry of data that populate useful fields that prompt and guide care plan development over time with trackable outcomes that are meaningful. We need workflow changes and patient engagement to minimize physician data entry, as exemplified by the lack of staging and molecular data in 50% of the charts that CancerLinQ downloads. Most of these data are in synoptic path reports or dictated notes, but until we automate and standardize data entry by staff, we will not have data for all the downstream needs, such as authorization, decision support, treatment planning, and outcome reporting.

Discrete Data Elements: We need to agree on a list of standardized data elements and entry sites in every EHR. Critical data, such as a patient's stage, tumor features, therapy details, rationale for pathway choice, response, progression, and toxicity data, could then be collected and presented in ways that are easily retrievable and used for care givers, patients, administrators, and payers throughout the course of disease. We need to agree on terminology for sequential episodes of disease and care to prompt later-line therapies whose outcomes depend on previous therapies and responses. Given the limits of natural language processing, this would lessen burnout by freeing doctors to think about and care for patients while allowing systems to better understand the growing complexities of care and their costs and quality with real-world metrics of value.

Reports and Dashboards: We need data presented in informative reports and actionable dashboards to support visit notes, survivorship care plans, and OCM care plans. Quality metrics required by OCM and the Medicare Access and CHIP Reauthorization Act should be reported from discrete EHR data. Standardized patient-reported outcomes can trigger documentation and clinician processes to address toxicities while reducing hospital and emergency room visits.[7] Even CMMI, which has all the charge data, recognizes that clinical data are required to accurately depict a patient population. Registries have been developed to obtain this information, but without defined discrete data entry into EHRs, they are populated by hand at considerable expense and often without real-time feedback to the clinicians, who could use the data to enhance their care plans.

Workflow Processes: Although process change is hard work, it is required to empower physician roles. Fear of regulatory noncompliance has inhibited expanded team-based data entry. A major negative EHR impact revolves around the interpretation of doctor's orders. In the past, clinicians instructed staff to have a patient get a test or schedule a visit, especially when patients called between visits. Staff generated a requisition with the clinician's authorized signature or made an appointment. In the current EHRs, physicians are now told to stop their work, go into the EHR, find the patient, and personally enter and sign each order before staff can act on it. This is a major waste of clinician time and a barrier to efficient team-based care authorized by standard operating procedures. In addition, EHR enhancements and training are critical to keep up to date but remain expensive and time consuming, especially when users of the same EHR cannot easily pool build teams and share upgrade costs.

Interoperability: we need to be able to push data meaningfully between systems that face multiple interoperability challenges today.

Design Engineering: Not only do we need systems with heightened functionality but also we need heightened functionality with improved clinical process workflow. Software engineers and the clinical team, including administrators, patients, and medical staff, together need to solve for requirements collaboratively to improve efficiency and reduce burnout.

In our 25 to 35 years of practice, a good example of technology making clinicians' lives easier is the PACs radiology programs. Many oncologists would not remember finding and fishing through x-ray packets to understand and show patients the details of their scan. We need our EHR systems to partner with us like PACs systems do. And, of course, we can only imagine an EHR system that serves us as well as Amazon, whose ordering and delivery platform has empowered our lives.

We have a horizon that will disenfranchise inefficient medical homes and models of care. As blockchain technology will allow patient information to flow more seamlessly, optimally interfaced EHR systems will continue to improve the quality of care. We need to agree on data standards, improved data entry/interface population standards, analytic reporting, and integrated quality and clinical research processes as well as mobile tele-oncology offerings that let the EHR empower our clinician workforce to better partner with patients and their care teams to achieve enhanced health outcomes regardless of where a patient lives.