Big Data's Impact On Health Outcomes

By Scott Westcott, Contributing Writer

As the amount of digital patient data grows exponentially, healthcare providers are seeking new methods of leveraging the power of Big Data to improve decision making and generate better patient outcomes.

The hype around Big Data is, well, really big these days in healthcare. The massive amount of information and data collected and stored is playing an increasingly important role in revealing insights and algorithms that can result in better patient outcomes.

The mainstream press has caught onto the story with U.S. News and World Report recently publishing, “How Big Data is Driving the Customization of Health Care.” The piece explores Big Data’s transformative potential in the healthcare space with authors writing, “In conjunction with mobility, Big Data is changing the way patients engage with their doctors and experience their treatment.”

Yet, while Big Data promises to be a gamechanger for caregivers and patients alike, many key challenges remain. First and foremost? How best to capture and assess data in manageable ways leading to greater insights that, in turn, translate to actionable results.

Experts say healthcare providers are in the early stages of interpreting and leveraging Big Data in ways that lead to actionable results. As providers recognize the potential of Big Data and become increasingly aware of growing compliance and consumer expectations, many are committing significant time, resources, talent, and money to harnessing data to create better efficiency and improved outcomes over time.

Among those organizations are the Austin Regional Clinic, a multispecialty group in Austin, TX, focusing on primary care, pediatrics, internal medicine, and 21 specialties, and Portland, OR-based Children’s Health Alliance, which works directly with health insurers on behalf of pediatricians on issues such as cost control, clinical quality improvement, and information management and sharing.

Participants:
Deborah Rumsey, Executive Director, Children’s Health Alliance, and Dr. Manish Naik, Medical CIO and Associate Chief of Internal Medicine at Austin Regional Clinic (ARC), shared their perspectives on how they are working to turn Big Data in to actionable intelligence.

Q: How are you currently extracting actionable intelligence from Big Data, and how are your findings and analyses driving action?

Naik: Today, we are testing the impact of collecting and acting on real-time feedback from patients across several locations using on-site tablets powered by Humm, which is an online feedback platform. We know that the best way to improve the ARC experience is to hear directly from our patients. Collecting faster feedback is allowing our clinic managers to make improvements, resolve issues, and take action on data immediately, as well as track trends over time. The program is still in the early stages, but the response rates and results to date have proven to be promising.

Rumsey: In the current environment, Big Data represents combined healthcare data from multiple sources — hospital and ambulatory EHR and ancillary clinical systems, claims, pharmacies, physician specialists, mental health, and community services — into a single database for analysis. This combined data set, overlaid with the appropriate analytics tools, can inform care at the population and patient level. In the future environment, we see Big Data as being used effectively to inform care at multiple levels — from public health to the individual provider. Big Data can inform providers about the health needs of their patients in order to deliver the right care at the right place and also prevent health problems based upon early identification of risk factors. If used appropriately at the point of care, Big Data can also assist with analytics to drive improvement in quality and patient experience of care, improve population health, and reduce costs.

Q: What role do you see Big Data playing in both the current and future healthcare environment?

Naik: With the widespread use of EHRs, we are collecting an everincreasing amount of discrete data on patients — examples include demographic information, clinical data, and financial data. Patients are also collecting more data than ever with the use of apps and smart devices. This data can eventually be used to make observations and predictions for health, both at an individual patient level and at a population health level. In the future, the hope is this data can help us deliver better care in a cost-effective manner for both individual patients and populations of patients.

Rumsey: The shift to manage risk, improve quality, and reduce costs cannot happen effectively without putting the data into the hands of the providers. The pediatricians in the Children’s Health Alliance have invested in a population management tool from Wellcentive to bring the multiple sources of data together into one centralized place to manage their population. This tool essentially empowers our providers and their care teams with information needed at the point of care delivery. Additionally, the analytics we are currently developing are provider-driven and designed to produce meaningful measures that can be utilized to better inform a physician of the patient’s holistic care needs. We believe that this change will happen much more quickly if the providers and their care teams have this type of comprehensive data to proactively understand and manage their population’s health. Armed with knowledge collected from other providers involved in the patient’s care, our physicians’ proactive care management can happen much more easily and timely, while reducing inefficiencies in the system.

Q: How are you leveraging Big Data to manage risk and gain efficiency, improve quality, and reduce costs?

Naik: Austin Regional Clinic implemented an EHR two years ago to help collect and integrate healthcare data. We also have a Patient Centered Medical Home that has its own database used for monitoring our performance with clinical quality measures and for conducting patient outreach. We have embedded nurse navigators at our clinics to help manage more complex patients that may be higher utilizers of healthcare services. At this time, much of our data is incomplete, and there is a lack of integration between clinical and financial data to help manage this risk. Most of our gains on managing cost of care have been through old-fashioned telephone outreach and identification of higher risk patients by our physicians. We are also able to use patient-level clinical data to do patient outreach to improve clinical quality metrics for preventative healthcare, such as breast cancer or colon cancer screening. Our EHR is also able to use patient-level data to send automated reminders to patients for some of these preventative health services through our patient portal. We are still working towards using the data at a population level to help us better manage risk and cost of care for populations of patients.

Rumsey: We have encountered several barriers in our attempt to effectively utilize Big Data. One example is the challenge of working with health plans and pharmacy benefit management companies to obtain claims and pharmacy data on behalf of our patient population. We continue to work with these organizations to demonstrate the value of sharing data with providers and the positive impact on patient care. In addition, we continue to work with our surrounding provider community to help them recognize that shared data improves everyone’s health outcomes. It’s an ongoing process.

Q: What results have you seen from your Big Data initiatives to date?

Naik: To date, we have seen improvements in diabetes care, management of high blood pressure, various cancer screenings, post-hospital care transitions, ER utilization, and end-of-life planning. For example, we have reduced all cause readmissions by 5 percent, and emergency department visits are down 10 percent. Our programs have helped the medical home staff and ARC in general fine-tune its processes and expand our abilities to help patients navigate and participate in their health care.

Rumsey: Putting data that is aggregated from outside sources into the hands of the providers and their care teams is adding tremendous value to our practices —- especially actionable data that goes beyond a rate or score. For example, we are currently linked to the state’s Emergency Department Information Exchange. This interfacing of our respective systems provides real-time information of our patients’ ED visits. Care teams can follow up with those patients to transition care, understand the reasons for the emergency visit, and determine if further office evaluation is necessary, often eliminating the need for an additional visit. As a result, our practices have a significantly lower ED visit rate than our comparative population.

Moreover, our initiatives driven by access to data are starting to yield significant results. Our immunization rate for two-year-olds is 86 percent compared to the state average of 67 percent. Our asthma patients have an 89 percent encounter rate, which results in better care as well as reduced inpatient and emergency room visits.

Q: What needs to be done with Big Data to extract more actionable intelligence from the data being captured?

Naik: Standardization of data feeds from third-party insurers and the development of standardized interfaces to allow us to absorb and analyze the data would be a very helpful step. A nationally standardized solution for sharing the data among all of the EHR vendors would also be very helpful. Currently, it is a challenge to share data electronically even with local hospitals, much less more distant sites where patients may have received care. In order to make valid conclusions, data needs to be complete and standardized. On the flip side, patient data, such as that being collected from patient devices and software, contains so much data that it could be overwhelming for clinicians and healthcare teams. This type of data needs more development to filter and deliver either relevant data or data only that is of specific interest to the question being asked by the healthcare provider.

Rumsey: We still have a lot to learn about utilizing Big Data. Current population management or quality reports often come from health plans that demonstrate how a provider performs on a certain metric identified by the plan. However, those reports do not provide sufficient information to inform the action necessary to improve care. We’ve been successful in managing our population through internal quality reports that allow our providers to drill down to the individual patient, which informs care delivery. We’re also working with our providers to develop reports and care alerts to focus on the areas deemed the most impactful to their specific population and to find easy ways to understand that data and turn it into meaningful action. Wellcentive is helping us co-develop the pediatric-specific protocols to cater the data to protocols that inform care and identify care gaps for the pediatric population. This specialized work will reach a population that has previously been underrepresented by the solutions in the marketplace.

Q: What aspects of healthcare will be most impacted by embracing Big Data analytics?

Naik: In theory, Big Data will help us better manage population health and cost of care, but it is a difficult prospect right now because of the lack of standardization and completeness of the data. Rumsey: My hope is that Big Data analytics, which will drive actionable data in the hands of the providers, will have a significant impact on the Triple Aim: Improve patient experience of care, improve population health, and reduce overall costs. As providers use this data to make more informed care decisions for their populations, not only will patients with complex medical conditions benefit, but also those identified patients at risk who require interventions before their condition becomes chronic will benefit. We also believe that sharing data beyond our practice network will enhance communication among all physicians and other care professionals caring for the same patient, including primary care, specialists, hospitals, mental and behavioral health, as well as community services such as schools and social services.