Skip to main content
Industry

Beyond the hype: Getting real about AI in health

[msce_cta layout=”image_center” align=”center” linktype=”blue” imageurl=”https://www.microsoft.com/en-us/industry/blog/wp-content/uploads/2018/07/Industry-Blog-AB-Test-Images_Health_R3-C.png” linkurl=”https://info.microsoft.com/ww-landing-Clinical-and-Operational-Analytics-eBook.html?lcid=en-us&wt.mc_id=AID723268_QSG_SCL_326314″ linkscreenreadertext=”Download the e-book now” linktext=”Download the e-book now” imageid=”30694″ ][/msce_cta]

With artificial intelligence (AI) still in early stages of adoption, there’s substantial hope among healthcare vendors and providers that it will make a significant impact. Yet, according to a recent survey by HIMSS Analytics, while healthcare providers believe AI will transform the industry, only 4% of those surveyed think their organization is prepared to take advantage of the power it holds. It’s no wonder that many providers are unsure about how AI can be applied in their organizations. With so much noise in the market, it’s challenging to know what’s viable in health today.

One area where AI moves away from theory and into making a real-world impact is predictive care guidance, the use of analytical solutions that draw on massive amounts of data to predict and improve patient outcomes. AI-driven predictive care guidance is being used today to make a difference and get results by providing real-time care guidance, predicting at-risk patients, and decreasing in-hospital infection rates, among other areas.

Preventing patient deterioration with early warnings and faster intervention

In a busy hospital, providers often supervise a diverse range of patients at various stages of acuity. As a result, it can be difficult to know which patients need attention at any given moment, and to prioritize care accordingly. By identifying and rating patterns in patient conditions and predicting an elevated risk of deterioration, AI helps providers intervene sooner, enabling them to avoid costly adverse events and adjust the patient’s care path as needed, ultimately improving outcomes.

As we begin to realize the benefits of applying AI, there’s evidence that this approach is saving lives. For example, Oschner Medical Center, in partnership with Microsoft and EHR leader Epic, launched an AI tool that assists care teams with on-the-spot guidance. By analyzing thousands of data points, the solution predicts potential patient deterioration and delivers timely, actionable “pre-code” alerts.

After a 90-day pilot with the system, Ochsner successfully reduced the hospital’s typical number of codes by 44%. Providers were empowered to prioritize care and treat patients more proactively, avoiding ICU admittance and mitigating rapid deterioration. They are now expanding use of the technology into more hospitals.

By leveraging data to identify potential deterioration, predictive care guidance enables caregivers to focus their efforts on preventative care, ultimately improving patient outcomes and care quality.

Predicting at-risk patients and reducing costs with enhanced chronic disease detection

As people live longer, chronic disease affects increasingly more patients worldwide. According to the Kaiser Family Foundation, chronic conditions account for nearly 75% of annual healthcare expenditure in the United States alone.2 Today’s healthcare system is designed and optimized for episodic care instead of long-term delivery, making chronic disease management a priority issue that could rupture the entire system. When faced with such a big burden, it’s easy to see how determining at-risk patients and enhancing preventative care represents a tremendous area of opportunity for AI. By predicting and lowering the need for acute care interventions for patients with chronic conditions, providers can decrease costs and minimize hospitalization.

Recently, data science leader and Microsoft partner KenSci partnered with the largest managed care provider in the United States to develop an award-winning3 platform to address cardiovascular diseases (CVDs). CVDs are the number one cause of death globally, representing approximately 31% of all deaths.4 KenSci applied machine learning to clinical data to identify patterns that indicated risk and provide predictive insights on care outcomes. Based on early results, the solution has been able to predict early onset of chronic heart failure and identify and stratify patients that may be at risk of readmission or need palliative services.

By helping providers detect and classify patients at risk for chronic conditions, physicians can tailor care plans based on a patient’s unique risk factors, turning AI into a powerful tool for treating disease and driving down costs.

Decreasing hospital infection rates and establishing best practices with actionable insight

Another way artificial intelligence is lowering costs and improving patient outcomes is by reducing hospital infection rates. The World Health Organization estimates hospital infection rates range between 5-12% globally. The costs associated with these infections in Europe alone are approximately $7 billion euros and 37,000 lives lost annually.5 Through sophisticated analysis of infection data, AI solutions uncover factors associated with higher infection risk, creating opportunities for mitigation. On the other end of the spectrum, AI-driven analysis also reveals factors associated with lower infection risk—paving the way for organizations to establish best practices.

Hospitals using AI-driven analysis to reduce infection rates are already seeing tangible results. For example, Epimed Solutions, a market leader in clinical and epidemiological information management, recently partnered with Rede D’Or in Brazil to develop a new approach to infection management in their intensive care units (ICUs). Rede D’Or was using an aging, on-premises analytics solution that didn’t offer real-time access to data. The new solution empowers clinicians to easily enter patient information at the point of care, which is then aggregated for near real-time access by the quality assurance office. With this data, they can generate reports that identify problems and compare quality indicators, enabling clinicians to intervene faster and reduce ICU infection rates.

Rede D’Or used this analysis to determine best practices in their top ICUs and implemented them throughout their network. As a result, hospital infection rates dropped 20%, mortality rates decreased, and care quality improved.

By making data easier to capture through mobile methods and applying advanced analytics to determine patterns, organizations can identify areas for improvement and evangelize best practices that keep patients free from infection.

Put artificial intelligence to work for your organization

Artificial intelligence holds transformative power to assist with predictive, preventative care guidance, sifting through thousands of data points to enable faster intervention, mitigate risk and deliver valuable insights. Learn more about how you can use AI to make the most of your data and improve patient outcomes.