By Pranam Ben, The Garage
The stethoscope is more than 200 years old. In that time, it has become an indispensable tool and the device most associated with physicians. Much more recently, electronic health records (EHRs) were introduced, and a short time later, data analytic platforms to help physicians make safe and effective clinical decisions using their rapidly accumulating electronic data.
Data analytics, however, are still not anywhere near as synonymous with healthcare as the stethoscope. That could soon change. As databases and analytic platforms accelerate and improve, specifically through automation, machine learning and artificial intelligence (AI), these systems will transcend decision-making support for individual patients and make much more far-reaching predictions about patient populations.
Accountable care organizations (ACOs) and other integrated health systems will benefit in several ways from these existing and anticipated technological advancements. Most significantly, providers will be able to better manage high-risk or near-risk patients, prevent adverse events that lead to emergency department visits and admissions, and improve clinical outcomes that lead to greater value-based care reimbursement.
Big Data Evolving Into Wide Data
Some organizations, especially ACOs and other integrated networks pursuing value-based population health management goals, are only now exploring all the possibilities of data analytics. As such, physicians, care managers and other providers may only be seeing a snapshot of clinical trends among high-risk patients. While this high-level insight is helpful, it often lacks actionable information. That means providers will need to dig further through their data to begin more targeted and more effective interventions, which adds to costs and contributes to care delays.
Sophisticated data analytics platforms, however, are delivering deeper, actionable information much faster thanks to advancing computer-processing speed and algorithm complexity. Big data, as it is commonly known, is also becoming wide, too, as the definition of what constitutes a patient record is expanding beyond EHRs. Genetic, lifestyle and socioeconomic data sets are now accessible or integrated with platforms to assist providers with more accurate and reliable insight.
The wideness of data is helping providers understand and manage social determinants of health (SDOH), which one study estimated contribute to as much as 80 percent of patient outcomes. What is even more crucial to provider organizations is that data analytics is delivering this deep insight into patients before these clinical and SDOH require a direct care intervention.
Rather, care managers can instead contact the patient to help them re-engage with their care plan, such as assisting with their care coordination, understanding their treatment plan and overcoming other adherence obstacles. In short, providers can prevent a near-risk patient from becoming high-risk and more quickly identify the outliers in populations who could potentially escalate care costs.
Artificial Intelligence Advancing Analytics
Predictive analytics such as these are used in healthcare organizations today, but it will become more common in the near future thanks to the introduction of AI and machine learning. Population health management platforms will be continually accessing clinical and other databases to identify trends, even when a care manager has not directly prompted the analysis. This constant machine learning would be initiated by the healthcare organization simply inputting its established care quality and cost goals into the AI-powered platform.
In what would take days or weeks of manual study, AI could accomplish in seconds or minutes. For example, the AI could analyze what clinical interactions or protocols within the organization have delivered the best outcomes among hundreds or thousands of patients with similar comorbidities, demographics and other characteristics. When commonalities are found, the analytics platform will deliver that insight in a very natural, conversational format to enhance, but not replace, the judgment of physicians and care managers.
The essential factor for time-pressed providers is that AI and machine learning will present them with information they want in an understandable and meaningful way so they can take action. In such scenarios, processes would be automated, reducing work and time for providers, but would also enable earlier outreach to avoid costly care and increase treatment plan adherence.
Making Data Analytics As Indispensable As The Stethoscope
Organizations that have only begun exploring how data analytics can improve outcomes among patient populations can accelerate adoption through promoting cultural change. Organizational leaders need to emphasize to clinicians and staff that patient care will become only more data-driven. They need to remind stakeholders that while analytics, AI and machine learning will never replace their physicians’ experience and expertise, they are essential to help providers perform at their best.
Likewise, organizations should seek experienced partners as they expand usage of data analytics. ACOs, hospitals and health systems have too many competing priorities to initiate building a population health management platform on their own. Partnering with companies that have already helped deliver performance improvements at similar integrated organizations across the country is more efficient, but also allows the organization to concentrate on its core focus, which is delivering safe, high-quality patient care.
Through this cultural and operational evolution, in a short time, data analytics will become as essential to physicians as the stethoscope, and far more important to the organization as whole.
About The Author
Pranam Ben is founder and CEO of The Garage.