Using analytics to optimize care for a burgeoning patient population

image

Dr. Sean Frederick, Chief Medical Officer, Allscripts Population Health

The United States population stands at 323 million and has a net gain of one person every 12 seconds. And, as insurance coverage increases, more people than ever before are seeking health care services. Estimates show that 20 million people have gained health coverage because of the Affordable Care Act within the last few years, placing new demands on our healthcare system.

Unfortunately, the number of healthcare professionals is not increasing at the same rate as demand for services. It means we, as physicians, have to be more effective in how we manage our patients.

Analytics is a powerful tool for clinicians to use to navigate the rising tide of patients. We can customize analytics to provider, organization, geography, diagnosis or any number of attributes for a community of patients. Because, when we can key in on patients who are most at risk or who need the most help, we can align our priorities and resources to optimize care for them.

The evolution of analytics

Analytics in health care is becoming more sophisticated to manage the proliferation of patients, data and technologies. Thinking back to paper-based record systems, it was painstaking to extract, digitize and manipulate information to get any sort of population-based insights.

Today, that data is available in real time. Everything from vital signs to lab results to prescription information. We can analyze all of this information at a much faster rate than ever before, bringing enormous potential for predictive models and managing population health.

For example, analytics can make a difference in caring for people with diabetes, one of the world’s most prevalent and costly chronic conditions. Diabetes affects 422 million people worldwide, or one in 11 adults, according to World Health Organization estimates. Evidence-based best practices suggest there are strategies that can help patients manage their disease, but only if they are appropriately diagnosed and engaged with care providers.

One large health system looked at its entire patient population and to find people with a documented diagnosis of diabetes noted in the electronic health record (EHR). Using analytics, this organization then expanded its search by looking through the entire EHR for signs or symptoms commonly associated with diabetes, such as mediations or specific labs and their results to help identify patients with diabetic or pre-diabetic conditions – even in the absence of an official diabetes diagnosis. The team was shocked to find that 45 percent of its diabetic population would have gone unnoticed, if it had strictly relied on EHR documentation for identification.

Ongoing challenges of using analytics effectively

The example of diabetes points to one of the main challenges to using analytics effectively: unstructured data. Electronic systems store data with tremendous variability. For example, HbA1c levels (a test used to diagnosis and monitor diabetic patients) is stored in thousands of different ways when one looks across the spectrum of electronic systems. Visibility into the diagnosis represents the opportunity to better manage care. For diabetic patients that could mean making sure they have the right prescriptions, annual foot exams or eye exams, regular HbA1c checks and more.

Another challenge to using analytics effectively is determining when to include or exclude confidential data. Good governance practices can help address questions about what is appropriate to share when it comes to sensitive health data, such as psychological treatment, addiction history or information about adolescent patients. It’s important for physicians to have all relevant information when making clinical decisions, without jeopardizing patient privacy.

A third challenge is the “siloed” nature of data in health care. Clinicians need multiple data streams to feed a complete view of the patient – claims data, pharmacy data, remote monitoring data, EHR information, genomic data – all of these types of information can affect decisions at the point of care. But too often they are stored in different places and not combined in a readily accessible, meaningful way within the providers’ workflows.

The last challenge, and certainly not least, is resources. Having people with the right skill sets and adopting the right technology remain challenges for many organizations. This is especially true for smaller physician practices or hospitals without the resources of a large health system.

Analytics: not just for the highest risk patients

Health care has been very focused on using analytics to manage patients with chronic disease. Though they make up only 5 percent of the total patient population, people with chronic diseases comprise the highest percentage of cost. The rest of the population should get attention, too. It’s important to catch patients with rising risk, and keep those patients healthy.

Every person needs to regularly engage with healthcare providers. Even healthy people need to take actions – such as annual preventive screenings, immunizations, blood pressure checks, flu shots – to help reduce risk of acute episodes. Analytics can help caregivers engage this massive population at the right times with the right messages.

We are all working toward delivering smarter health care, and analytics is a powerful tool to achieve that goal. With aggregated, harmonized data at the point of care, clinicians can make better and faster decisions for improved outcomes.

Tags: , , , ,

Trackback from your site.

Leave a comment

You must be logged in to post a comment.

Get our eNewsletter

Email Marketing powered by StreamSend

Please follow & like us :)

RSS
Follow by Email
Facebook
Google+
https://us.hitleaders.news/using-analytics-to-optimize-care-for-a-burgeoning-patient-population/">
Twitter