Industry Voices—3 strategies for data analytics experts to prepare for future challenges and opportunities

Healthcare data analytics professionals are in a perfect position to advance their organizations’ goals by helping make more informed decisions. Leveraging data, they pinpoint trends, assess performance, improve care delivery and predict future risk.

The value of analytics in healthcare is only expected to increase. Even as the discussion of big data and analytics has become ubiquitous, the market is projected to expand from an estimated $9.36 billion in 2017 to $34.16 billion by 2025, according to Allied Market Research.

As applications continue to grow, data analytics professionals—from chief information officers to data engineers and scientists—should be prepared to address challenges and seize future opportunities.

Below are three strategies that can help them prepare for the future.

Zero in on actionable insights to address specific goals

The industry has never had access to so much clinical, business, billing and public data and other data sources. And the ways to use the data are many and varied. Managing the sheer amount of information and its potential use cases can be both the best and the worst part of the job. Every data set can send an investigator on a trail to discover new insights, and it is easy to get distracted following an exciting possibility.

RELATED: Accenture: AI, blockchain will have 'transformational' impact in next 3 years

Defining outputs for analytics models to align precisely with the organization’s current goals, such as addressing a certain type of risk and working towards achieving immediately actionable insights present a good strategy in the world deluged with data.

Befriend uncertainty in its many forms

Uncertainty is unwieldy and stressful, and data analytics professionals have to address it in several areas.

To begin with, healthcare analytics differs from all other analytics fields in the level of uncertainty it has to account for. Compared to financial risk prediction, for example, predicting a patient journey and delivering actionable insights to improve outcomes is a much more complex task.

An individual taking out a loan ultimately is either going to pay it back each month or not: the outcome is clear. But if you are assessing a patient with several comorbidities, such as diabetes, hypertension and chronic obstructive pulmonary disease (COPD), there are multiple outcomes to consider, including emergency hospitalization, surgery, physician visits and medication adherence, to name just a few.

Then as the industry regulations change, so do the analytics models. The future of the Affordable Care Act will determine the type of risk that organizations will need to manage, which, in turn, will change how analytics models are designed.

RELATED: A shortage of data scientists could be holding back advances in healthcare

Analytics experts also need to factor in legislative inconsistencies between states—for example, when tracking quality measures. That needs to be done at the granularity of legislation, not at the aggregate population level.

Additionally, healthcare organizations must address ever-changing threats targeting data security and privacy. Information security teams are kept on their toes both by comprehensive privacy regulations and ever more sophisticated cyber-fraudsters.

Therefore, analytics models must also change in response to the internal and external drivers and pressures, which over time impact the availability, breadth and quality of data sources. 

Flexibility is the new strength

Consequently, the best approach to an uncertain, evolving, complex environment is built-in flexibility in the data analytics framework. In flexible, layered models, independent data points and attributes can be added to improve outputs and taken out if they are unavailable for use any longer.

RELATED: Use of predictive analytics is helping reduce costs at payers, providers: survey

Consider the latest trend of including non-healthcare-related data, such as socioeconomic attributes, into analytics models. Social determinants of health (SDOH), which organizations are now applying at a broader scale, offer the opportunity to view patients holistically with new insights about the potential barriers keeping them from realizing optimal results from care management and wellness plans. Integrating SDOH attributes, or scores, into predictive analytics adds another perspective to the model in an effort to help organizations better personalize care.

Keep your eyes on the prize

Data analytics professionals should strive to give their organizations the confidence to take on new risks by generating actionable insights from big data. Valuable insights, especially for the most complex patients, will allow for both accurate cost projections and earlier intervention for better outcomes.

It’s a very exciting time to be a part of healthcare transformation because we can do so many interesting things with data.

Victor Tavernini is vice president of product management and analytics at LexisNexis Risk Solutions, Health Care.