The impact of healthcare claims data analytics across the care continuum

In the ever-evolving healthcare landscape, data has emerged as a driving force behind groundbreaking discoveries and transformative advancements. Harnessing claims and patient-level data is essential to fully understand the patient journey. An extensive and complete history of the patient, along with a comprehensive view of health resource utilization, adds critical context to patient behaviors, diagnoses, treatments, and adherence. Understanding the entire patient journey requires a well-connected digital ecosystem to make smarter decisions to improve patient care.

Health claims data analytics can support the creation of actionable insights. The use of de-identified medical claims facilitates new scientific findings while protecting patient privacy. Data-driven analytics provide valuable insights for diverse patient populations that can drive essential change across the healthcare industry.

Benefits of healthcare claims data analytics

There are many tools that can be harnessed to conduct healthcare outcomes research. Healthcare claims data analytics paired with the integration of data from multiple sources, such as customer relationship management (CRM), patient registries, client portals, healthcare providers’ EHR systems, population health data, social determinants of health (SDOH), and social media can increase the value of the analytics.¹ Claims data can be linked to other data sources to answer complex questions about healthcare access, costs, utilization, and outcomes and uncover health disparities.

Methods for secure data linking

The de-identification of protected health information (PHI) enables researchers to derive valuable insights about population health, aiding healthcare decision-makers in tailoring patient care to achieve optimal outcomes. The HIPAA Privacy Rule only permits a covered entity or their business associates to de-identify data using specific standards and specifications. Covered entities must use either the expert determination method or the safe harbor method to remove PHI and deidentify the data for research purposes.

Safe Harbor

The safe harbor method mandates that covered entities and business associates remove all 18 PHI identifiers from data to prevent the data being linked back to an individual. These include obvious identifiers such as names and addresses, but safe harbor specifications also require the removal of all dates of finer granularity than year and any information on patient locations smaller than state. Once all identifiers are removed, HIPAA’s protections no longer apply. Nonetheless, this approach has one major drawback: it may strip too much valuable data, rendering the information less useful for certain research purposes such as time to event analyses or pinpointing geographic hotspots.

Expert Determination

Expert determination involves seeking guidance from a qualified subject expert to safeguard critical information and uphold patient privacy. Datasets de-identified using the expert determination method can in some cases include exact dates of service and geographic identifiers smaller than state, such as patient ZIP Code. This level of detail supports research involving treatment patterns and sequencing, the exact duration of treatments and outcomes, and geographic variation in treatment patterns and patient characteristics. However, there is no expiration date for re-evaluating the data and no one-size-fits-all solution for addressing privacy concerns.

While there are some limitations, the safe harbor and expert determination methods are valuable approaches for utilizing vital healthcare data while upholding patient privacy and adhering to HIPAA guidelines. The best method depends on the data available and the specific research questions. Inovalon’s experts can help researchers make the best decision regarding de-identification to accomplish the goals of the specific project.

Examples of claims data in studies

Research comparing patient populations

A recent valuable example is a long-term research partnership between Inovalon and Harvard Medical School. Using Inovalon’s comprehensive all-payer claims dataset, researchers were able to securely link to various datasets, including detailed socioeconomic data at the ZIP9-level with Medicare Advantage, 100% Medicare Fee-For-Service, and commercial insurance plan data. This unique longitudinal dataset allowed researchers to examine important differences between Medicare Advantage and Medicare Fee-for-Service members at the point beneficiaries are choosing which type of Medicare plan they enroll in, including important demographic information, healthcare utilization costs, and quality outcomes. The results will support plans, policymakers, and regulators in understanding the most effective and efficient models of care for the growing Medicare population.

Code-specific analyses

Researchers can evaluate the impact of patient SDOH through the use of Z codes, a set of psychosocial risk and economic determinant-related codes within the International Classification of Diseases, Tenth Revision, Clinical Modification. A study conducted by Inovalon analyzed data from 2017-2020 to investigate the use of SDOH Z codes among Medicare Advantage beneficiaries. The study found that while social risk factors significantly impact 80% of patient health outcomes,² fewer than 2% of claims have a documented Z code. These findings highlight the importance of educating health plans and providers about using Z codes to identify and address health disparities among underprivileged populations.

Researching rare diseases and novel treatments

A recent study investigated the outcomes and economic impact of Chimeric Antigen Receptor T-cell (CAR T) therapy in patients with Relapsed/Refractory Diffuse Large B-Cell Lymphoma. Using claims data from Inovalon’s primary source dataset, the study identified a specific group of patients and discovered that CAR T-cell therapy had a positive impact on treatment effectiveness. The researchers noted that CAR T-cell therapy was rarely used in older patients, especially those aged 75 and over, but showed positive benefit for those patients who are typically excluded from clinical trials.³

A related study of R/R DLBCL patients compared CAR T use with statistically matched patients who received the traditional standard of care, stem cell transplant. This study showed that CAR T had significant benefits in terms of reduced inpatient and emergency department utilization and total healthcare costs post- vs. pre-treatment, particularly for the subset of patients who received an allogeneic, as opposed to autologous, stem cell transplant.⁴ CAR T has the potential for reducing healthcare utilization and financial burden on patients, families, and the healthcare system. These studies underscore the importance of real-world data in advancing the use of novel yet potentially life-saving treatments.

Turn claims data into innovative insights with Data Solutions

Expert healthcare analysts can unlock valuable insights by utilizing real-world data to address important questions. Inovalon’s primary source dataset, combined with our industry expertise, can assist in uncovering answers to your most complex healthcare questions.

Ask our Data Solutions team how we can support your research needs.