Rethinking Medical Data Analysis

By Inga Shugalo

The pandemic has rendered many established healthcare processes and workflows inefficient. To fix the inefficiencies rapidly, many providers have had to speed up their digital transformation efforts by as much as 74%. The changes have powered a meaningful shift in the role of analytics in healthcare.

The broader focus

In the pre-pandemic times, medical data analysis was mostly concerned with optimizing resource allocation. However, with the advent of COVID-19, the applications of analytical tools in healthcare have broadened considerably.

Input for population health: On a larger scale, data analytics has allowed providers to identify infection hotspots and cross-link the localities with population health specifics. Now providers can detect not only at-risk patients but also rising-risk patients. Identifying such patients more efficiently may help providers reduce care costs and prevent unplanned hospitalizations.

Some providers have already tapped into managing rising-risk patient populations. For instance, Brigham and Women’s Hospital in Boston conducted research to detect early chronic obstructive pulmonary disease (COPD) risk factors in Hispanic patients. COPD generally reveals itself when patients are over 60 years old, and managing it at that point is a challenge. Having analyzed the results of the Hispanic Community Health Study/Study of Latinos, the team found asthma, smoking, and chronic sinusitis were among the key risk factors of early COPD. Addressing rising-risk patients, from there, involves raising awareness about the problem and offering viable patient education options to reach the targeted populations.

Personalized condition management: On a smaller scale, data analytics has contributed to significant results as well. Amidst the pandemic turmoil, providers have managed to supply safe remote patient monitoring (RPM) for chronic-condition patients through consumer-grade wearables and built-in analytical solutions that generate personalized insights. What’s more, many providers have run training sessions featuring RPM devices and analytics to help patients manage their data sets proactively.

Not all eligible patients have embraced the use of RPM. A recent American Heart Association study on blood pressure monitoring in U.S. adults delivered surprising results. About 80% of the participants named home-based blood pressure monitoring the best option versus an RPM device or on-premises kiosk. However, this doesn’t spell the end of RPM tools. In fact, industry leaders are now looking at the new RPM mode: device-free RPM. These solutions rely on communication via landline phone or text messages and employ advanced analytics for a range of conditions, including mental health issues.

X analytics in the game

The pandemic has propelled an unprecedented level of cooperation worldwide. Clinicians, medical researchers, and experts have openly shared their findings to contribute to crisis resolution. The number of available data sources has surged and diversified. Today’s data comes from a vast variety of structured and unstructured sources—research papers, news articles, clinical trials data, and social media posts, to name a few. New types of data call for new analytics, and this is where X analytics steps in.

According to Gartner, X analytics is the ability to leverage multisource structured and unstructured data regardless of its format—textual, audio, video, graphic, or other. X analytics can be combined with predictive and prescriptive analytics, which the industry already uses. However, analyzing the new data may require additional efforts—reconsidering the existing predictive, prescriptive, and other models, for example, as they may have become irrelevant or obsolete.

Providers are already employing X analytics in healthcare, at least to an extent. Today, experts use broad data collection to create full-scale patient profiles that reflect not only medical history but also socioeconomic and environmental factors. The NYU School of Global Public Health and Tandon School of Engineering have used this approach to apply social determinants of health to cardiovascular disease risk prediction models, which helped improve the models’ algorithmic precision.

Gathering real-time patient feedback

Patient feedback is the cornerstone of patient-centric care and its adjustable variables. But how can the industry make feedback collection simple for patients and clinicians?

Experts recommend enabling several channels, and here digital transformation has played its role. With diverse technologies launched across the continuum of care, providers have managed to gather patient feedback at a close-to-real-time pace via bedside tablets, kiosks, and text messages right after discharge. However, certain barriers have hampered the effort. Many patients had a lack of connectivity, insufficient knowledge of technology, or a lack of patient engagement.

Fortunately, healthcare experts have come up with a viable solution: recommending providers to team up with their tech vendors. The latter can gather patient feedback at each touchpoint, ensuring the technology works at all times and providing tech support to patients, thus taking care of the administrative side of patient satisfaction measurement. Meanwhile, the provider can fully concentrate on the clinical side.

What’s next?

Medical data analysis has leapt forward during the pandemic, transitioning from covering operational needs to enabling efficient populationwide predictions and personalized care. Thanks to digital transformation, patients will enjoy a connected ecosystem with multiple touchpoints that gather data sets for patient-specific insights. The future is bright for health data analytics, and providers need to tame health data in all its volume, diversity, and velocity to drive rapid multisource insights for highly individualized service.

Inga Shugalo is a healthcare industry analyst with Itransition.