As a physician and long-time EHR user, I live and work in a world of discrete, narrative and non-narrative data. Notes that I or my colleagues have entered into the EHR, documents that have been scanned as image files or pdf documents into the document repository and discrete elements such as laboratory results. The data arrives from many different sources and systems and in multiple formats. It has to be parsed into a format that everyone understands and has been trained to act upon in an appropriate manner.
Rather than focus upon the technical elements of data interoperability and standards, the purpose of this article is focus on some of the practical implications and how small changes can create significant challenges in a multi-disciplinary medical practice for both clinicians and administrative staff.
- Semantic interoperability: Simply put, does the information transferred from one system to another have the same meaning to the clinician when it is presented in the EHR? The way that data is presented has the potential to negatively influence decision making if it is presented either in a format or location that is not logical or intuitive in terms of workflow. For example, if a diagnostic report is composed of both discrete elements (e.g. laboratory results) and narrative elements (e.g. interpretive comments), both types of data need to be presented in all appropriate views of the EHR. Because results may be viewed in table format or as discrete data embedded in notes or as a full report, how one uses the EHR becomes critical to the interpretation of the information. If a clinician’s standard workflow is to view the data in table format, it is important to indicate whether a narrative note is present. More complex is the fact that users can often create personalized views of data (dependent on the EHR system used). Managing quality control in these types of optimizations is crucial, but often a missing in the hustle and bustle of a busy practice. This is one simple example of an interoperability challenge that has nothing to do with the data, but rather the manner in which it is presented to the user in the EHR.
- Skills and training: While it may not appear to be immediately apparent, the computer skills of all staff members and knowledge of their EHR has a significant impact on data interpretation. From earlier surveys conducted by AmericanEHR, we know that many users received limited initial and follow-up training to use their EHR. These are highly complex systems that require ongoing maintenance and training. In a busy medical practice, in which the system appears to be operating ‘OK’, there is generally little desire to change anything. Patient care is the primary focus and frequently ongoing training takes a back-seat to daily operations. This is complicated by the fact that users have differing levels of computer skills and understanding of basic technology as well as the impact of staff turnover. When new clinicians or staff join a practice, who does the initial and ongoing training? Is it delegated to the most knowledgeable member or the individual who has the lowest workload at that point in time. Bad habits can easily be propagated throughout a practice simply through ignorance. If the EHR requires some personalization at the time that a new user is added in order to be consistent with standard office workflow, is there a quality assurance process to ensure that the user’s view of the data is same as others? Do users clearly understand where the data needs to be accessed and how it can be viewed in different formats in order to aid interpretation.
- EHR Version: One of the most frequent complaints of users is that their EHR does not have the necessary tools or capabilities to perform a certain function. In many of these cases, the user is either unaware that their EHR has the functionality and has not been trained to use the product properly, or may have an older version and needs to upgrade their product. In larger practices, version upgrades can be complex and can require servers and entire systems to be shut down for maintenance. As a result, sometimes upgrades are delayed because of potential disruption. Cloud-based (ASP) systems have a definite advantage as the software is updated centrally by the vendor with limited need to update technology at the user end.
The take home message is that the job is far from complete by simply ensuring that the ‘data’ can be moved from one EHR to another or from a feeder system to an EHR. The ultimate success of any EHR implementation is only partly about the technology. The human factors are equally important. Adequate support, training, succession planning and in-office expertise are critical in order to maintain smooth operation of any EHR system. The static paper chart has been replaced by a continuously evolving ‘living’ system that needs constant care and nurturing in order to operate optimally.