Bend the cost curve: Learning more about your patients can lead to higher quality care
The following content is sponsored by Optum.
Providers can use advanced analytics to gain the knowledge they need to improve, manage and succeed in today's dynamic, value-based healthcare market.
As providers assume more risk, they're turning to population health management as a way to improve the quality and delivery of healthcare and control costs. As part of this strategy, they're expanding their chronic disease management programs into their communities and proactively monitoring and interacting with the populations they serve.
Underlying the success of these programs is the effective use of advanced analytics. With the help of sophisticated tools that scrutinize longitudinal claims and clinical data, for instance, providers are getting a more robust view of their patients with chronic diseases. They are identifying patients who haven’t been seen regularly — or whose health metrics are outside acceptable limits — and finding ways to more effectively manage them.
Predict the future: Are your patients at risk of being hospitalized?
Better managing patients with chronic diseases means having a fuller picture of their health — including predicting their risk for future complications and more accurately targeting interventions.
Chronic diseases such as diabetes, congestive heart failure, hypertension and chronic obstructive pulmonary disease are among the most common, costly and preventable of all health problems in the U.S., according to the CDC. Seven out of 10 deaths among Americans each year are from chronic diseases.
But much of the cost and harm to quality of life caused by chronic diseases can be mitigated. Organizations can prevent poor clinical outcomes and control costs by using advanced analytics to gain a deeper understanding of their patients and sort them by risk for hospitalization.
Using natural language processing to analyze echocardiogram results, for example, renders notes on ejection fractions structured and reportable, allowing physicians to better gauge the risk of their patients with congestive heart failure. That same technology can also enhance the capture of symptom and pulmonary function test results, extracting reportable data points. This can give physicians exceptionally sharp focus on the clinical status of patients with COPD.
The Mayo Clinic Health System is piloting a congestive heart failure predictive model that brings together in-depth clinical, diagnostic and demographic data to identify patients at highest risk for admission within the next six months. Mayo Clinic users review and export lists of these patients for outreach and coordination. They can then track the impact of this work by comparing benchmark hospitalization rates for the congestive heart failure population to the true outcomes for the coordinated population.
In addition to identifying and acting on gaps in care, organizations can use analytics to track clinical, operational and financial performance. Dashboard reports, for instance, can provide valuable insight into clinical performance, laying the groundwork for initiatives designed to promote physician practice of evidence-based medicine and drive improvements in quality, safety and efficiency. Gaining access to comprehensive longitudinal data can also help providers benchmark their practices against other practices across the country.
Stratify patients by risk to more effectively coordinate care
Segmenting a patient population lays the groundwork for devising effective care management and patient engagement programs. Many organizations have retooled their care management approach from a reactive model to one that is driven by predictive, proactive intervention and care.
Mid Hudson Medical Group in New York, for example, is using a clinical intelligence platform to identify high-risk patients with diabetes and focus its care coordination efforts to improve disease management.
After analyzing its data, Mid Hudson was able to identify which of its diabetic patients met criteria for proactive outreach. The group was able to single out patients whose HbA1c was greater than 7 percent at their last visit, or who had not been seen by a provider within the last 12 months. As a result, about a third of such patients were seen one or more times within the first eight months of the program. In this group of diabetics, one-third achieved an HbA1c level of less than 8 percent, and 60 percent of those with HbA1c higher than 9 percent became more intensively managed through more frequent visits with their primary care physician.
Yet another group, the Billings Clinic in Montana, has implemented a scalable approach to identify and track patients with hypertension. By applying analytics to its data, the clinic was easily able to find patients with hypertension based on clinical findings such as blood pressure readings. It was also able to stratify patients with hypertension into clinically relevant cohorts based on clinical findings, such as those consistent with kidney disease or diabetes. Billings was then able to further analyze these groups by clinical acuity, medication patterns or other process measures. The provider group went on to monitor the impact of its hypertension interventions over time and track its patients' control longitudinally. Physicians were also provided with reports that compared their results to their peers locally and at other leading practices throughout the U.S.
Use advanced analytics to prepare your organization for value-based care
Providers who use population health management principles to care for their patients with chronic conditions will be ahead of the curve as the industry continues its march toward value-based reimbursement. Leveraging advanced analytics to create more comprehensive risk profiles for patients with complicated illnesses will better position providers to make the transition from treating illness to managing health.