How analytics helps at the point of care

The amount of patient and medical data generated by providers, payors, labs, researchers and connected devices has increased exponentially in recent years as healthcare organizations embraced digital technologies. That’s good news because more data can theoretically lead to better care and greater efficiency.

The unfortunate reality is that providers and other healthcare organizations easily can be overwhelmed by the firehose of raw data emanating from connected devices, lab tests, and patient visits to primary care providers and specialists.

To unlock the clinical and operational value of data from disparate sources, many healthcare organizations are turning to advanced analytics platforms that are able to detect patterns which can inform clinician treatment decisions and process improvements that reduce costs. This is critical as providers move toward value-based care (VBC) revenue models that reward outcomes and better cost and utilization control. Four important areas where analytics can drive VBC include: addressing Social Determinants of Health, understanding patient-generated data, bringing payor data to the point of care, and applying best evidence into practice.

Using Advanced Analytics to Address Social Determinants of Health

Physicians are able to provide better care when they understand all factors that impact the lives of their patients. This type of holistic view includes an awareness of a patient’s neighborhood and surrounding environment, their access to transportation, their ability to purchase healthy foods, their employment status as well as many other social determinants of health (SDOH). In fact, these social determinants have been found to be the most important indicators of health, and clinicians must have access to this information to improve health outcomes and address health disparities.

A 360-degree view of the patient is also an essential tool for providers who are moving to alternative payment models that reward better patient outcomes and lower the costs of care. In these new models of care delivery, SDOH data is required to understand the overall healthcare needs and risks within a defined population.

Fully leveraging SDOH for patients and providers requires the use of advanced analytics to help sift through volumes of data for actionable information. SDOH and advanced analytics combined can:

• Identify non-clinical factors that drive health inequities and impact outcomes
• Identify interventions that can ameliorate health disparities
• Close care gaps by addressing health behaviors
• Accelerate the transition to value-based care models

Making Sense of Personal Device Data

I have seen health systems build digital platforms that allow them to accept patient-generated health data. For example, many patients now rely on devices such as Apple Watches and electronic scales to monitor their own health and are excited to share their data and progress with a healthcare provider.

As you might imagine, the amount of raw data generated by these personal devices can be overwhelming! It is impossible for a primary care physician to wade through the large volumes of raw device data to find pertinent health information for multiple patients per day. Analytics tools are the answer; these tools must be used to assist providers in detecting patterns and trends that indicate a patient may have a health problem that needs to be addressed.

Understanding Payor Metrics and Using Payor Data

Many providers receive individualized reports about their performance from a number of different payor organizations. The problem is that each payor often has different reporting standards and the reports come in different formats and sources. For example, some payors may send patient information in a hard copy letter while others use an online portal.

Patients may change their jobs and health insurance providers often, which means that an individual’s information from the payor may not be up to date, may be formatted in disparate ways, and may even have different metric definitions. Many providers set aside these payor reports and metrics because they are so difficult to translate at the point of care.

A best practice in using payor data and reports is for health systems to negotiate for standard metrics that can be interpreted at a system level and to use advanced analytics to capture and integrate payor data at a system level to enhance their clinical data.

Keeping Up with Medical Research

Every week there are dozens of important research papers published in medical journals that include information that should be applied in a primary care practice. This is particularly true with precision medicine where new impactful research seems to be released almost daily. In this modern age of medicine, it’s a full-time job to keep up with and apply the latest evidence.

Decision support tools within electronic medical records (EMR) may assist with this process; however, these tools may take many months or even years to be updated with the latest evidence. Furthermore, decision support tools within the EMR may not help to quickly identify patients with specific conditions or gene variants who could benefit from emerging evidence about new therapies.

With advanced analytics tools, healthcare providers can improve the way that they integrate evidence into practice and care for their populations.

Applying Analytics Before, During, and After the Point of Care

Having relevant patient data at the point of care is indisputably critical to practicing effective medicine. The process for integrating analytics into the point of care can be challenging, though there are four approaches that must be used to be effective.

First, physicians require data for pre-visit planning. In addition to a list of all patients being seen at the practice that day, physicians need details of these patients’ care needs to ensure the practice has all of the necessary supplies available and have a plan in place for how care should be delivered that day.

Second, at the time of care delivery, clinicians need information at their fingertips to ensure that the patient is receiving all the care they need. This information is often delivered within the EMR and should mirror the information used in the pre-visit planning phase.

Third, the care team needs information after the visit to understand their effectiveness in managing the care plan and to identify any errors made during the patient’s visit. This information is needed quickly so that all members of the care team can reflect and better understand any factors that could have led to an error in care delivery.

Finally, the care team needs analytics that look at the entire patient population to understand who may need care or had a change in their condition and requires additional attention. The care team can then engage with patients to ensure that they are aware of any needed services and have access to a provider. For example, if analytics tools identify a patient with poorly controlled diabetes who has not been seen in the last nine months, a care team member can contact her to re-engage with her provider right away.

Now more than ever, there is more patient and medical data available to clinicians. Yet this data does little good if it can’t be accessed and understood by physicians and other team members before, during and after the point of care. Advanced analytics is the key to unlocking the clinical and operational value of this data.

_{About the Author}

_{Michael Dulin, MD, PhD, is currently chief medical officer for Gray Matter Analytics, as well as a professor at UNC Charlotte in the Department of Public Health Sciences where he directs the Academy for Population Health Innovation.}