Knocking Down Data Barriers in Healthcare

Healthcare data is expanding at an exponential rate. Data compiled last year by Dell EMC shows health data grew an astounding 878% from 2016 to 2018, fueled in large part by the increasing use of telemedicine, wearable devices, and medical imaging. The COVID-19 pandemic has forced providers and patients to rely even more heavily on telemedicine, which has further accelerated the proliferation of healthcare data.

But all this patient data does little good if it can’t be accessed by providers when needed. Thus, it is critical that healthcare organizations can efficiently identify and aggregate patient data for use by providers to deliver quality care, improve outcomes, and reduce costs.

There are numerous clinical and operational benefits to sharing patient data. Having up-to-date information in their electronic health records (EHRs) gives providers an accurate picture of a patient’s current clinical profile and level of risk. Among other things, this helps ensure providers won’t prescribe contraindicated medications and makes them aware of “care gaps” such as missed annual screenings or failure by patients to fill prescriptions for chronic diseases such as hypertension.

Access to patient data also lets providers know if a patient has visited an emergency room or been admitted to a hospital. By having all relevant patient data, providers can avoid wasting money on services such as blood tests and imaging that already have been performed.

While this all sounds great in theory, the reality is there are many barriers to effectively sharing patient data. Historically, healthcare companies and disparate EHR systems have been “walled gardens” that either were unwilling to share data for proprietary reasons or were unable to communicate electronically with other systems, essentially trapping patient data in silos. HIPAA and privacy concerns have also impeded data sharing.

Data sharing opportunities and obstacles

Fortunately, there has been a real push in the past couple of years for healthcare companies to tear down their data walls and share information with other providers and systems. Some of this has been driven by regulatory changes. A rule passed in 2020 by the Office of the National Coordinator for Health IT (ONC), for example, mandates that patients have the ability not only to access their own health data, but to download and share it with third-party apps. This rule includes penalties for information blocking.

The evolving nature of the healthcare market itself has motivated stakeholders to share data. With so many applications making use of patient data, sharing has become essential for healthcare companies to stay relevant.

Despite these regulatory and commercial incentives, obstacles remain to effective data sharing. Inconsistent data quality and incompatible data are the biggest hurdles. Data compiled in one EHR may be incomplete or inaccurate. That might be due to mistakes or omissions at the point of data entry. In addition, EHRs may document information in different ways or employ different workflows.

In one EHR, there might be a small checkbox on a screen for a question about whether an elderly patient was asked if they are at risk of falling or have fallen recently. Another EHR, however, may not have a box for that information. Right away that creates problems for an analytics platform trying to determine how many elderly patients have had fall screenings, data that can affect a provider’s Medicare reimbursement. So, getting codified and universally usable data is a challenge for an analytics firm, which makes it difficult to draw meaningful conclusions about what’s actually happening within patient cohorts.

Solving patient data problems

Fortunately, inherent flaws in patient data can be minimized or eliminated through a set of data aggregation best practices designed to fill in gaps and enable interoperability.

The first step is collection of raw data. To provide the best care for patients, healthcare organizations need relevant data from multiple sources (inside and outside of their networks), including hospitals, urgent care facilities, private practices, health information exchanges (HIEs), retail pharmacies, labs, clinics, mobile apps, wearable devices, and insurance providers. Gathering this data – sometimes from hundreds of sources – is a formidable task, so healthcare organizations need a solution that is integrated with a large number of EHRs and other healthcare systems to optimize data collection.

The next step is normalization, in which two different healthcare organizations work together on a process to determine what information may be missing. This requires a partnership between systems that gives the receiving party transparency into data at the source to validate it. When data is found that is not entirely usable, it is put through a mapping process to convert it into a form that can be utilized. Part of the normalization process is updating and reconciling basic identifying data, such as confirming the patient’s most current address and contact information, correcting misspellings, and eliminating duplicate information.

The real value of normalization is it allows providers to see information they don’t know, whether it’s clinical data, lab data, claims data, or other patient-related information relevant to care, and whether it’s coming from a provider outside a network. The goal is to create a single, unified chart for a patient, or an Enterprise Master Patient Index (EMPI).

Normalization does more than help providers offer better care to patients. It improves the bottom line for healthcare organizations by identifying high-cost patients, thus enabling providers to better manage care for these patients. By uncovering gaps in care, providers can order tests and treatment for which they are able to receive full reimbursement while heading off potential health problems that could cost more to treat in the long run.

In addition to providing healthcare organizations with quality data for individual patients, data aggregation enhances population health management (PHM). Once data is aggregated, cleansed, and merged to create a single profile for each patient, algorithms are applied to categorize patients by cost, risk, condition, care gaps, and more. This broad view of patient population data gives provider organizations insights into health trends, allowing them to initiate mitigation efforts (as in the case of an infectious disease such as COVID-19) and plan resources.

The final step in the data aggregation process is making the cleansed and organized information available as an EMPI to healthcare organizations in an enterprise data warehouse (EDW).

Healthcare organizations have more patient data than ever to work with. Yet without the ability to collect, clean, and organize this data, providers and other healthcare stakeholders won’t be able to fully leverage this information, hurting patient and population care and costing money over time. Deploying a solution based on sound data aggregation practices can improve both health and financial outcomes.

Author Brian Russell is director of integration at Lightbeam Health Solutions, a population health management software company.