Value-Based Care and Staff Shortages: Can AI Help?

The slow-but-steady move to value-based care (VBC) has challenged healthcare organizations to improve patient outcomes, but long-standing personnel shortages have made this challenging. At first glance, it seems obvious that finding and recruiting more qualified workers is the only answer. After all, more doctors and nurses means better care for patients . . . right? 

For years, shortages in qualified nurses and clinicians have been predicted and confirmed. Many inventive regulatory and procedural efforts to ease the burden are being considered, like allowing non-physician personnel to handle clinical tasks, removing unnecessary licensing barriers, and staffing strategies. Some healthcare organizations use a Kaufman Hall system to predict staffing through all shifts based on beds, acuity, etc. Yet personnel shortages continue to be a problem. While these strategies may help operationally, they don’t necessarily improve the quality of patient care. 


Value-based care became a major movement after the 2010 Affordable Care Act, which spurred the Centers for Medicare & Medicaid Services (CMS) to implement several value-based programs, like the Hospital Value-Based Purchasing and Hospital Acquired Conditions Reduction Programs. According to CMS, “Value-based programs reward health care providers with incentive payments for the quality of care they give to people with Medicare.”

The idea is to provide better treatment for patients, resulting in better outcomes at a reduced cost. CMS incentivizes providers to move away from a fee-for-service model to one that achieves desirable results, like reduced hospital readmissions and fewer hospital acquired conditions. To do this, providers must look at a patient holistically and take multiple factors into account, like mental health, lifestyle, environment, and other social determinants of health (SDoH).

The rise of digital healthcare technologies has led many organizations to consider whether these advances can help with personnel shortages while preserving quality of patient care. 

In fact, today’s technologies make it possible to address those shortages—through reduced workloads—and to not only maintain, but elevate quality of care. Proactive innovation can help healthcare organizations reach these goals, by using artificial intelligence (AI) and predictive analytics for diagnosis and clinical decision-making.


“By augmenting human performance, AI has the potential to markedly improve productivity, efficiency, workflow, accuracy and speed, both for [physicians] and for patients.” – Eric Topol, MD, director and founder of Scripps Research Translational Institute.


Rapid, accurate diagnosis is crucial to improving patient outcomes and reducing staff workloads. AI is rapidly changing how diagnosis happens by augmenting clinicians’ medical and patient-specific knowledge. Doctors and nurses are only human; and while the human brain is able to store extensive amounts of data, recalling that data is more difficult. IBM Watson for Health stores and reviews information from diverse sources all over the world, like medical journals, textbooks, case studies, and treatment and response plans. This technology can store more data and recall it more quickly than any human. 

AI can also automate patient data-gathering. Arterys Cardio DLTM uses AI imaging to provide “automated, editable ventricle segmentations,” helping clinicians identify patient heart problems by providing more data with less manual work. Automated diagnosis support has also arrived on the scene. Stanford University’s CheXNeXt tool reliably screens chest X-rays for 14 different pathologies. While qualified personnel still have to review the findings, this pre-screening saves time and increases accuracy. 

And the FDA just approved an echocardiogram software made by Caption Health that enables non-specialists to take ultrasound pictures of the heart. Delegating tasks out of the workload of highly specialized and certified personnel is ideal for addressing staffing shortages—as well as burnout rates.
Clinical decision-making naturally flows from diagnosis. A recent Health Data Management survey showed AI is already playing a significant role in this area. 53% of survey respondents indicated that their organizations are using AI and predictive analytics to improve clinical decision-making. An EBioMedicine project team used a proprietary program to identify patients at risk of developing multi-drug resistant infections by examining electronic health records (EHRs) and evaluating them in light of massive amounts of clinical trial and study data. The algorithms predicted whether patients would contract one of two specific infections (or neither). The team reported a 95% accuracy rating.


“Artificial intelligence and predictive analytics allow doctors to choose a treatment regimen most likely to produce favorable outcomes based on the specific patient’s risk factors.”


Proactive, early detection of vulnerable patients is another way AI and analytics are making a difference. Penn Medicine implemented a platform called Palliative Connect to analyze patients’ EHRs and predict whether they would need and benefit from palliative care. This changed their approach from reactive, based on clinician and patient requests, to proactive, based on data and risk-factors. “Patients and families rated the program favorably as a satisfying experience, and . . . Front-line clinicians said they felt it was an acceptable intervention and it improved the care of their patients.” 

These technologies can also give healthcare organizations a high-level view into clinical best practices and variations in care. Texas Children’s Hospital began using an enterprise data warehouse (EDW) to aggregate and analyze care data to improve practice flows. The result was a 35% reduction in hospital-acquired conditions (HACs). Healthcare analytics can examine a wide variety of factors, including SDoH, to quickly provide clinicians, staff, and executives summary reports showing patterns and correlations. This can lead to the formation (or reformation) of best-practice care flows that both relieve personnel workload, staffing changes like using more navigators, advanced level providers and improve patient outcomes.

Artificial intelligence and predictive analytics have deep implications to the business of healthcare. Increasing speed and accuracy of diagnosis, analyzing best practices, and proactively identifying at-risk patients lowers costs and increases revenue by reducing unnecessary tests and ineffective treatment. Decreasing time spent on research and manual tasks relieves workload, giving all care team members more time for patient care. All of those things add up to improved patient experience, quality of care, and health outcomes.

Healthcare organizations can only thrive by proactively innovating—pursuing ways to put digital technology to use and implementing in a way that serves everyone. AI and analytics for better diagnosis and clinical decision-making are a crucial part of digital transformation.