Electronic Health Reporter

By Ophir Ronen, CEO, CalmWave.

Hospitals routinely collect vast amounts of data, including information about patients’ health, care delivery, and organizational performance. This data could theoretically be utilized to drive huge improvements in health outcomes and operational efficiency.

Rather than this massive amount of health data being an advantage, it’s most often considered a burden as there are inconsistencies in documentation, aggregation methods, organization of, display of, and most importantly, how it’s used.

There’s also a lack of resources required to effectively manage this overwhelming amount of information. The data gap not only leads to lost opportunities to improve healthcare but is a major contributor to some of healthcare’s current biggest issues like burnout and staffing.  Hospitals are always striving to better leverage healthcare data. Revised processes that reduce manual inputs, eliminate redundancy, and include central EHR systems are a few goals that come to mind. However, some of the biggest wins will only be gained once we tap into more advanced tools that leverage artificial intelligence (AI).

With so much medical information already available, including large complex data sets, sophisticated AI systems are just what the doctor ordered to get these data sets organized and in use.   To unlock data, hospitals must incorporate more data science and use artificial intelligence (AI) methods to operationalize their learnings.

Data science is an umbrella term for statistical techniques, design techniques, and development methods. It involves pre-processing analysis, prediction, and visualization, whereas AI is the implementation of a predictive model to foresee events. Advanced data science and AI can not only help organize all of this information, but also generate trends and insights so that hospitals can deliver more precise care, identify operational hazards, and create a more optimized approach for managing their current workload.

Many Sources of Information 

The problem, as it stands, is too much information from disparate sources and in different formats. Hospital data comes from a variety of places, such as electronic health records, administrative systems, insurance providers, patient-submitted forms, local HR systems, hospital medical devices, and, remote monitoring services. 

This information also comes in various forms, including structured digital data, physical documents, photographs, charts, and more. The volume and diversity of data make it difficult to store in conventional databases and format it for use across multiple frameworks. 

More Data, More Problems?

Humans are complex beings. And when caring for a patient in the hospital, every bit of information that can be collated can be helpful to provide the best possible care for that patient.  The first step is making sure the information is (digitally) captured. Paper/manual processes are prone to human error. 

The more electronic devices and EHRs evolve, the more accurate and consistent data that is produced. The next step is organizing this data in a way that is usable for clinicians in an actionable way. Cognitive overload is a common complaint among caregivers. More information isn’t always better. 

This is where human-centered design is key. Leveraging a combination of data science to process big data, along with modern intuitive user interfaces, are what’s needed to bring new levels of adoption and activation of available hospital data. The insights that intuitive AI and data science can provide can transition the healthcare industry to a more objective, data-driven, operationally-efficient care model.  

Imagine a world where supervisors and administrators have data at their fingertips when making important staffing and clinical decisions. Imagine a world where every second of every day, the data ‘traffic’ can be used to provide an objective, consistent way to measure the operational load on the hospital, the department, or even the clinician.

A Holistic View Leads to Better Outcomes. 

Perhaps the greatest use case for AI in health care stems from its ability to ingest and aggregate large quantities of data from numerous sources, giving providers a highly detailed view of a patient’s health journey and current state. With this information at their disposal, providers can paint a much more holistic view of the patient. 

A single vital sign or medication response is hardly enough to provide a comprehensive assessment to identify early warning signs of serious conditions and apply preventative measures before the condition develops. Physicians in particular can also deliver more personalized treatment plans when they have a more comprehensive view of a patient.

Moreover, AI can review trends in historical data to identify patterns that are hard for the human brain to quickly identify. These patterns enable advanced AI-powered solutions s to deliver insights to help predict an individual’s risk of developing certain conditions or suffering a health emergency. Hospitals are increasingly incorporating early warning systems that monitor multiple parameters (vital signs, patient status) to generate alerts for rapid response teams, with the idea of improving patient care. 

The more automated the data inputs, the more sophisticated and more predictive these tools can become. These same signals and data can even help clinicians optimize existing vital signs equipment for each patient’s unique condition.   Additionally, alarm fatigue is a known issue in the hospital environment, and there are healthcare analytics companies using existing vital signs data along with other data to create a virtuous cycle to remediate alarms and improve operational efficiency.   

Greater efficiency and working conditions with AI

Once data is collected and organized with AI, the potential benefits are limitless. This is the case across practically every industry looking to drive performance, and healthcare is no exception. However, healthcare is complicated. What are the performance goals in healthcare? More profit? More patients treated? Greater efficiency? Growth? Stability? Patient satisfaction? Shorter length of stay? Hospitals, unfortunately, have many stakeholders to consider, and therefore can’t ignore any of these metrics. Data is key to driving results in every one of them.   And AI is the enabling technology. Here’s how AI can help: 

Re-Admissions and Length of Stay: This data is readily available, and is slowly becoming a key measure of hospital operational efficiency. Hospitals can improve patient care even further by using AI to collect data related to patient admissions, length of stay (LOS), and re-admissions as well as the number of critically ill patients requiring life-saving care during different time periods. This can help hospitals determine the exact points at which operations and individual workflows become congested or disrupted. 

Seasonality: Patient care undergoes many fluctuations. Some of them are cyclical, driven by the seasons, day of week, or time of day. But data has shown that other, more sporadic events, like local sporting events or special surgeries, trigger massive shifts in patient care needs. Tracking this data on a macro scale, in concert with key clinical metrics, has the opportunity to improve hospital strategies for care delivery. For instance, studies have shown that if surgeries were shifted beyond the early morning, clinical resources could be better distributed. Another emergency department moved their clinical team to on-call, because the department was empty during the 4 hrs before, during, and after the local football game. These insights may be anecdotally understood by providers, but it’s the data that allows it to be operationalized.

Employee Health: Hospitals can also review the aforementioned metrics to identify the busiest and slowest times of day, track the sickest of patients requiring the most care, and can understand the overall impact on providers. This has the opportunity to not just help orchestrate staffing in a more efficient and equitable fashion, but can also be used as a tool to improve employee satisfaction. Burnout is a major issue plaguing healthcare, and any opportunity to better measure employee health and operational efficiency can only be good. This not only provides transparency for the organization, but it empowers the administration with the data they need to make more informed decisions.  

A new era in medical data

Data is already flowing at high volumes throughout the healthcare system. But it’s one of the biggest assets NOT being utilized to its full advantage. And many hospitals are struggling to figure out how to unlock the potential of this data. Only those that invest in the staff and technology backed by data science and AI-powered solutions will be successful in the long run. 

Thanks to AI, hospitals finally have the means to put their wealth of valuable data to use. 

While these developments do signify change and a major cultural shift, most staff members would happily accept a calmer, more transparent, and more equitable working environment that is data-driven and ultimately puts both patients and providers at the forefront.