The proliferation of diagnostics and screening tests in the United States has generated massive amounts of patient data. To effectively store and manage these mounting volumes of data, the US health systems have invested over $20 billion in electronic health records (EHR systems) over the past decade. Based on the data from our healthcare market intelligence, over 85% of healthcare providers and physicians in the United States had access to EHR systems by the end of 2017. Although electronic medical records are poised to revolutionize patient care by putting patient information at the doctor’s fingertips, not all physicians are content with its implementation. Physicians often find themselves caught up in data entry with limited time for patient interactions. Moreover, electronic health records are currently far from being the panacea of patient safety and operational efficiency that they were expected to be. However, as electronic health records system matures, we can expect the future of electronic health records to live up to their potential in the long run.
Future of electronic health records : Key roadblocks to overcome
Prior to the implementation of electronic health records in the US, healthcare providers and patients were unable to effectively follow patient in time and space. Furthermore, such siloed systems added on to the hospital expenditure and wastage of physician time, restricting the provision for a better quality of care. EHR systems have the potential to mobilize hospital’s resources and enhance the overall patient outcome. But a major challenge of EHR systems is that various vendors had separately developed systems with different data formatting, making it difficult to share patient records between hospitals, physicians and external testing labs. This also makes it challenging to use data collected by patient monitoring devices. The Fast Healthcare Interoperability Resources (FHIR) draft standard is trying to develop a standard for storing and transmitting data across healthcare organizations. And this is now being widely accepted by vendors of electronic health records systems. The recent regulations proposed by the IS government health insurance plan might soon make FHIR mandatory for electronic health records.
Reducing physician’s data entry work
Natural language processing is a great way to reduce the time spent on data entry work for physicians and facilitates them to allot more time to patients. However, this comes with its own set of challenges. Firstly, the clarity of the output will largely depend on the physician’s way with words. This could hamper the reliability of the report. Secondly, with such technology, patient privacy is at stake. Although countries like the US and UK have strong medical data privacy regulations, especially in the case of data transmission, we still come across data breach instances. Such breaches can lead to healthcare fraud especially in the case of medical insurance.
Managing data entry errors
One of the key advantages of electronic health records was believed to be its ability to reduce errors, misplacement, and oversights in healthcare records. Also, in many cases, the illegibility of physicians’ handwriting was found to be the cause of several errors. Although the introduction of electronic medical records has reduced instances of medical errors considerably, the chances of errors cannot be completely ruled out. Digitalization could also pave for new opportunities for medical error. Drug safety authorities have confirmed that human-computer interactions were responsible for over 50% of problems in laboratory testing. This could drastically hamper patient care by causing inconvenience, errors or delays in diagnosis. Although electronic health records systems are designed to alert clinicians in case of mistakes, often these alerts are shrugged off as a ‘computer error’ and this in some cases could prove to be fatal for patients. In order to reduce the risk of alert fatigue causing similar mistakes in the future, it is advisable to switch off the most frequently ignored alerts.