E-Liability and e-Documentation.

Abstract

Electronic medical records (EMRs) and paper medical records are profoundly different. The age-old saying pertaining to nursing documentation and liability in the paper world was, “If it wasn’t documented, it wasn’t done.” This statement, of course, was and is not accurate. Although comprehensive and complete documentation may be the goal, the reality is that nursing care is sometimes performed and not documented, sometimes for legitimate reasons.Today in the electronic world, there are lots of built-in documentation choices. If a nurse is not familiar with or cannot readily recall where all potentially relevant choices are, will she or he be held accountable for something not documented? This challenge is accentuated when an excessive number of rarely used or unused built-in choices require nurses to wade through numerous screens and volumes of content to find what they need. Perhaps the built-in choices are not words or terms the clinician would normally use or do not fit the patient or situation. The issue is more problematic when free-text options have been reduced or eliminated. Making matters worse are frequent changes to screens, content, and content location resulting from updates, upgrades, safety issues, new quality metrics, and/or optimization efforts. This requires nurses to unlearn how they have been documenting and create new documentation workflows. All of these create an electronic environment that is not ideal for supporting nursing documentation or practice.Equally important to appreciate is that unlike the paper record the EMR is more than a documentation tool. It is a decision support tool with increasing amounts and complexity of clinical decision support (CDS) functionality. A simple example is the often-seen EMR alerts. Alerts can be construed to represent the standard of care and, when overridden, especially without a documented reason, deviation from the standard of care and clinical judgment can be retrospectively questioned.1 However, EMR alerts can be based on erroneous information. For example, in a recent study,2 researchers found that approximately 95% of patients with the most commonly reported allergy, to penicillin, did not actually have the allergy that was reported in the EMR. How do nurses know when alerts are based on accurate information, and what is the standard of care for responding to an alert, if there is one?Although prescribed, forced-choice, or built-in documentation is not routinely considered CDS functionality, it should be considered such. In a recent evaluation of an EMR, researchers found that prescribed documentation elements for pressure ulcers included elements that were inconsistent with the definition of pressure ulcers, thereby offering nurses erroneous information and inaccurate choices for documenting pressure ulcers.3 With CDS functionality in EMRs come new risks and accountabilities that are only beginning to be understood.Despite the challenges, one practice that carries through from the paper world to the electronic world is the heavy reliance on documentation for risk management and malpractice litigation. The increasing involvement of EMRs in litigation parallels the widespread implementation of EMRs in hospitals and clinics and illuminates the differences in how patient data and information are entered, stored, and retrieved in the medical record. This Technology Today column highlights the unique risks, or e-liability, posed by EMRs that should be considered by nurses, managers, and informaticists. These risks are important because legal precedents surrounding the use of EMRs are developing and evolving.It is simple to appreciate that EMRs and paper medical records are different, but the implications of these differences are far reaching, especially when it comes to illustrating retrospectively how care was provided. Paper records are more static, easier to chronicle in relation to documented times, and make changes in content easier to discern, with the “patient’s story” more visible and in the clinicians’ words. On the other hand, electronic documentation is considerably more dynamic.Electronic versions of the same EMR from the same vendor in the same version typically differ from one organization to the next because organizations customize their EMRs. Even within the same EMR in one health care organization, how clinicians interact with the EHR can differ in multiple ways. For example, documentation can occur in real time or later, sometimes not capturing the actual time when nursing care was provided. Another example is that the same content can be input by different clinicians in different areas of the EMR, which limits nurses’ ability to analyze changes in the patient’s condition or responses to care over time. The clinicians’ own words are difficult to find, with forced documentation choices, “smart phrases,” and limited opportunities for narrative input. And in most cases, the view of the EMR is different from one profession to another, meaning that what nurses see in an EMR is different from what physicians, pharmacists and others see. As if this were not enough variation, the layout, navigation, content, and functionalities change over time with ongoing EMR updates, upgrades, optimizations, addition of new documentation requirements, responses to safety issues, and more.In March 2015, The Joint Commission (TJC) released a sentinel event alert related to the safe use of health information technology (IT).4 They identified 120 sentinel events related to health IT between January 1, 2010, and June 30, 2013, categorizing them into 8 sociotechnical dimensions (listed in the Table in order of frequency). Although TJC listed ergonomics and usability issues at 33%, it can be argued that 92% of the factors listed in the Table are usability issues because workflow, communication, clinical content, training, processes, and software design all fall within the realm of usability.The dimension of “people” cited by TJC is contrary to principles of EMR usability.5 The usability principle of “error” refers to the prevention of use errors as opposed to user errors. This distinction is important when identifying solutions. Citing users or “people” as the cause of health-IT issues is reminiscent of the age of health care quality that focused on human error versus process error. It is important to focus on technology error, as good usability helps users prevent and recover from errors when using the EMR.5Another concern is the description of “people” relating to training and failure to follow established processes. Training need is inversely proportional to good usability of health IT including EMRs, meaning the more training that is required the worse the usability. Keep in mind that training comes in many forms including the ongoing need for superusers, end-user training, 1-on-1 training, at-the-elbow just-in-time support, sandboxes and playgrounds, routine updates, tip sheets and pocket guides, clinical help desks and remote support, user groups and listservs, and scribes.6 All of these are indicators of the need to improve usability and reduce the need for training. An EMR is usable if it is easy to learn, efficient to use, and useful to nurses and their practice.5A recent study7 on EMR-related events in medical malpractice claims reported the existence and harm incurred when using this technology. Using a claims database, researchers examined cases from January 1, 2012, through December 31, 2104, finding 248 cases (< 1% of total cases) involving errors associated with EMRs in ambulatory care settings (146 cases), inpatient settings (102 cases), and emergency department settings (25 cases). The researchers further classified the etiology as being either system-related issues involving technology, design, and security or user-related issues. Ninety percent of the cases involved medication errors (31%), diagnosis errors (28%), treatment errors (31%), and other errors (10%). More than 80% of the cases in each setting—ambulatory, inpatient and emergency department—resulted in medium or high severity of patient harm. An example identified by researchers as being EMR-related, and readily appreciated by EMR users, included test results and patient evaluations documented in multiple locations of the EMR creating difficulty in identifying overall decline, resulting in the death of a patient.Electronic documentation has benefits, such as the oft-cited improvements in legibility and clarity of physician orders. However, capturing the so-called patient’s story, the patient’s unique health care experience as well as current status, is significantly more difficult in the electronic world as it exists today. Although summary screens may be available, clinicians can be left to ponder where in the EMR additional data that they currently need to make informed decisions might be. Audit trails can demonstrate where they looked, but it can be difficult to retrospectively produce the patient’s story and what clinicians were thinking when care is called into question, accentuating concerns for e-liability.Following TJC-reported data on sentinel events related to health IT outlined in the Table, usability testing of EMRs should be a top priority for reducing e-liability risk. Usability testing should be based on sound usability principles and testing methods as well as analyses of the multiple dimensions of usability that support the complex, nonlinear practice of nursing previously discussed in another Technology Today column.8 Equally important, strategies to improve EMR usability should be based on sound design principles that match the usability issue and desired practice goal(s).If not already in place, organizations should establish and sustain an e-liability committee until the EMR and other clinical information systems reach a mature, less problematic state. Consider bringing together a diverse team of nurse experts to work together for ongoing evaluation and learning about EMRs. Include staff nurses, informaticists with design and usability expertise, risk managers, health information management professionals, infection prevention nurses, as well as quality and safety experts. Nurse managers are key in staying connected to the evolving practice of nursing in the digital age and their ability to routinely communicate findings on e-liability to health care leaders.If health care organizations have experienced e-liability or malpractice cases involving the EMR, perform a postanalysis as it pertains to questions and comments about the EMR exposed throughout the process. Identify what questions were asked and what issues surfaced pertaining to the use of the EMR by nurses. It is important to share lessons learned through publications and presentations because health care is relatively young in the era of widespread EMR use and EMRs continue to evolve. Learning organizations are imperative, as is a learning nursing profession.There should be a reliable system for staff nurses to report EMR issues, near misses, and actual incidents. To gain a more comprehensive picture, nursing reports should be combined and analyzed with other data being collected about the EMR and nursing practice such as data from clinical help desks, medication errors, CDS overrides, amount of unplanned and planned downtime, and more. A significant amount of resources has been and continues to be invested in EMRs. Nurses’ voices in evaluating how well the technology is working for them and implementing improvements is important in reducing e-liability and maximizing the return on investment.EMRs hold significantly more data than paper records, most likely involving missing and inaccurate data. This multifaceted application has numerous menus and screens, documentation elements, and silos as well as increasing amounts of CDS functionality. All of these create more complexity and liability risks.1 This Technology Today column is not intended to be a comprehensive overview of e-liability related to e-documentation but rather a stimulus for dialogue and actions to evaluate and improve the documentation and decision support environment being provided to practicing nurses in EMRs.