At the time, the 2009 HITECH Act and the introduction of the Meaningful Use program represented a momentous step in healthcare’s digital transformation. With billions of dollars in government investment and incentives in nearly two years’ time, over 77 percent of hospitals had reached Stage 3 of the Electronic Medical Record Adoption Model (EMRAM), or gone even further, pulling millions of data points into newly-established EMRs.
Over the next 10 years, as EMRs grew into a broader and more collaborative snapshot of a patient’s health, the government attempted to streamline digitization regulations to bolster adoption even more, and by 2019 — despite some challenges — healthcare data was generally flowing more freely. However, by 2020, as a result of the pandemic and health tech’s record boom, many hospitals were essentially drinking from a data firehose.
Amid the ongoing deluge of data, health systems are simultaneously dealing with other institutional challenges related to the COVID-19 pandemic, including increased patient volume, physician burnout, an ongoing nursing shortage, the burden of ensuring cybersecurity and an overwhelming digital health ecosystem. Healthcare data analytics can and should serve to alleviate some of these operational challenges. In order to do so, here are four characteristics that data should achieve in order to be truly actionable and valuable.
1. Accessible — While the collection of more data points from patients (clinical, demographic, etc.) and other financial or operational sources (staffing, claims, business analytics, etc.) is theoretically good, it is of little-to-no value if it is not retrievable, downloadable, and searchable. Collecting or subsequently organizing datasets in standardized formats is the first step towards making it more accessible in other ways.
2. Digestible — Depending on a project’s scope, goals, timeline, stakeholders, or any number of factors, teams may require a range of specificity of the data. Quality data analytics tools should be usable by employees at various levels within the health system, from support staff to the most high-level decision-makers. Rather than investing in specialty training or dedicated data science teams to make sense of the increasing amount of information hospitals are collecting, the onus has really shifted to the market to create tools that meet healthcare professionals where they are with more visual and digestible data tools. Although today’s EMRs have made a giant step towards improved data accessibility, more can be done to centralize and democratize data for non-clinical hospital staff and ultimately to patients themselves.
3. Understandable — Although data may be technically accessible and visually appealing, without improved data literacy, even the largest and most digitally-transformed hospitals may not actually be benefiting from their data. Critical to Garter’s definition of data literacy as “the ability to read, write and communicate data in context,” is also “the ability to describe the use case, application and resulting value.” These deficiencies are unfortunately especially prevalent in the healthcare community. Further, a recent survey from Philips found that 35 percent of younger healthcare professionals are unsure how to use patient data and analytics to inform care. It is clear that training and upskilling may be required to ensure every employee is competent in their data use.
4. Timely — In addition to being accessible and understandable, data also has to be timely in order to be actionable. Healthcare is already known for its lack of timeliness when it comes to patient interactions and operational efficiency. Delayed operational data can paint an inaccurate representation of a situation to decision-makers, potentially leading to bad decision making that can cost time, money, reputational damage, decreased employee satisfaction and a lower quality of patient care.
Between the collection of increasingly more patient data through EMRs and other digital health tools over the last decade, in addition to new data streams as a result of COVID, health systems are overwhelmed. While data flow is a critical first step, more needs to be done to improve accessibility, visualization, and data literacy in order to make it truly actionable on a large scale. Fortunately, the analytics tools that healthcare workers need exist, and prioritizing their implementation will result in a meaningful improvement to hospital operations, finance and, ultimately, patient care.
About Jon-Michael P. Smith
Jon-Michael P. Smith is the Head of Healthcare & Life Science Analytics – Data Integration at Qlik. He is an experienced senior sales leader and certified medical coder with expertise in IoT, Big Data, real-world evidence, clinical trials, health services research, patient-reported outcomes (PROs), clinical informatics, and patient-centered care. He has decades of experience working with global health delivery systems, pharmaceutical clinical trial sponsors, CRO’s, biotech and med-device organizations.