What You Should Know:

– A new survey conducted by AHIMA reveals that 66% of health information (HI) professionals have experienced ongoing staffing shortages in their workplaces over the past two years. These shortages were notably prevalent in areas such as data quality, consumer health information, revenue cycle management, privacy, risk, and compliance, and data analytics.

– The report, “Health Information Workforce: Survey Results on Workforce Challenges and the Role of Emerging Technologies” also finds that despite workforce shortages, HI professionals see promise in artificial intelligence (AI) and machine learning (ML) technologies for alleviating some of the workforce burdens. However, this increased reliance on AI and ML also calls for upskilling within the profession.

Impact of Health Information Professionals Shortage

83% of respondents have witnessed an increase or persistence in unfilled HI positions within their organizations over the past year, indicating a pressing need for targeted interventions. The shortages in the HI profession have had far-reaching consequences, including reduced reimbursement, increased claims denials, lower patient data quality, and slower information releases. These issues have had a direct impact on healthcare quality and align with broader workforce trends in the healthcare sector.

The Promise of AI/ML Adoption

The survey indicated that 45% of respondents have adopted AI and ML in their departments, but this adoption comes with challenges, such as increased technical demands and the need for enhanced oversight. As a result, 75% of respondents consider upskilling the HI workforce as essential given the growing adoption of AI and ML tools.

Why It Matters

These findings from the survey are crucial in shaping the future management of patient health data and determining the necessary workforce to navigate emerging technologies. The Biden-Harris Administration and the US Congress are actively exploring the implications of AI and emerging technologies on the US workforce, and AHIMA intends to use these survey findings and recommendations to prepare the HI workforce through policy discussions, research, education, and training to ensure the secure management of patient health data.

Report Background/Methodology

AHIMA commissioned NORC at the University of Chicago to conduct the survey to examine the workforce challenges impacting HI professionals and assess the role of emerging and evolving technologies, such as AI and ML, in reshaping the HI workforce. AHIMA will use this information to improve data quality, increase productivity, and reduce administrative burden. With insights from 2,500 respondents, including AHIMA members and non-members, drawn from a vast pool of 35,000 in August 2023, the study spotlights the urgent need for action.

Healthcare providers are feeling the industry’s talent shortage in their administrative functions as well as clinical areas. One case in point: revenue cycle management. In a recent study, 63% of healthcare providers reported an inability to fill key revenue management roles. Reduced revenues and cash reserves are almost inevitable when providers don’t have sufficient staff to follow up on claims, manage appeals and help patients understand their financial responsibilities and options. 

Fortunately for providers, revenue cycle automation offers timely, cost-effective solutions to the talent shortage. Robotic process automation (RPA) is a proven technology for automating rote, repetitive processes that involve multiple steps and systems and/or substantial human and machine interaction. Using software bots, RPA essentially mimics the keystrokes of human operators. Many bots can be developed and deployed in a matter of weeks. They then work tirelessly and accurately, including during off-hours. At one institution, a software bot cleared a backlog of thousands of claims status checks in just a single weekend. It would have taken humans hundreds of days to accomplish that task. 

RPA is just the start. With software bots streamlining processes and improving data accuracy, the foundation is set for creating more sophisticated automation solutions built on AI and ML models. These can tackle more complex tasks that involve following business rules and making decisions based on the models’ data analysis.  These AI/ML solutions often are more expensive and take longer to implement than RPA. They are best suited for providers that already have standardized processes and cleaner data from their existing automation. 

Unlock revenue fast with robotic process automation

Automating RCM tasks frees up revenue management professionals to take on other, more complex activities, such as providing financial counseling during patient pre-registration activities. RPA solutions will also enable providers to improve revenue cycle productivity without adding additional employees. Many RCM tasks are excellent candidates for automation via software bots and RPA, including: 

Claims status checks. Bots can look up claims data and other information in payer portals, then update systems and even initiate next steps, eliminating these rote tasks and returning time to revenue professionals.

Automated patient pre-registration. RPA bots can link applications and systems together to automate more complex transactions and extract more value from them. Take a patient-facing, self-service registration portal. After a patient agrees to interact digitally with the provider, RPA bots can download patient registration requests; retrieve patient data from an electronic health record (EHR); then update the patient engagement system. The update can trigger the engagement system to send the patient a self-registration and payment link. When the patient completes those steps, the RPA bots can access the payment and patient demographic data and update the provider’s EHR. 

Digitally enabled prior authorization. Bots can easily retrieve patient data, insurance details, CPT codes, physician details, diagnosis codes and schedules from an EHR; flag cases requiring prior authorization; and submit them digitally to a payer portal. Then bots can update records with approved requests. They can also automatically route denied cases requiring additional information to the right clinicians, then refile them when updated. 

What about AI? 

RPA software bots essentially follow sets of rules. While a rule set can be complicated and involve several systems, software bots generally make preprogrammed if/then decisions. Outliers can be routed to finance professionals for follow-up. 

In contrast, automation solutions that incorporate AI and machine learning algorithms can evolve and eventually make autonomous decisions. Put very simply, an ML algorithm learns from the data sets to which it’s exposed, finding patterns and relationships. This makes ML potentially very powerful. ML algorithms can stratify patient accounts by a propensity to pay and automate financial assistance applications. That would reduce costs to collect while improving revenue realization. ML could also identify missing charges and help avoid revenue loss. Those applications, however, are complex. In general, the more advanced the technology, the more time and expense required to implement it. That’s not to take these options off the table. While they undoubtedly will play a role in coding and other tasks, ML and AI applications often are more than many providers need to solve immediate staffing and revenue realization issues. Healthcare organizations must carefully select which processes to automate to ensure the results meet their needs. 

Moving forward with automation

Providers must be clear about what they hope to achieve by automating their revenue cycles and realistic about the time and resources they have available to allocate to the project. The following steps can help guide decisions about which revenue cycle processes to tackle.

• Choose low-hanging opportunities first. Providers should build organizational automation experience before attempting more ambitious projects. The provider that solved its claims status check backlog with RPA initially applied the solution to claims from its largest payer. After succeeding there, the provider then expanded the initiative to claims from its other payers. 

• Choose opportunities that minimize IT involvement. Provider IT professionals often have many competing priorities. Developing RPA software bots requires minimal IT input. 

• Choose a vendor that understands healthcare revenue cycle services. Working with a skilled, experienced vendor helps minimize the time and input a provider’s revenue professionals must give to the engagement. Vendors with RPA and healthcare revenue expertise can build flexible bot frameworks so bots can be extended to other applications with minimal programming.

• Prioritize opportunities that tangibly improve patient and employee experiences. Improving the patient financial experience is a growing priority. Automating tedious, repetitive tasks reduces errors and frees revenue staff to work on more complex issues that deliver more value to patients and the organization. 

• Evaluate the return on investment. Most RPA projects should deliver a return 2 to 3 times greater than the investment. Reconsider projects that have lower anticipated returns or that indicate a long time to ROI. 

The societal changes that have shrunk healthcare’s labor pool are here to stay. Automating the revenue cycle will position providers to improve cash flow, enable their revenue professionals to work at the top of their abilities and offer patients the streamlined digital experiences they increasingly expect. Most importantly, providers will have more of the financial resources they need to focus on their true expertise, delivering patient care and improving outcomes.

About Noel A. Felipe, CRCR
With over 38 years of experience in healthcare accounts receivable management, Noel has a proven track record of developing progressive client-based solutions and building strong cross-functional teams to implement those solutions and maximize client results. At Firstsource Noel maintains direct account management responsibility for strategic clients and leads the development team for Firstsource’s digital collection and digital pre-service collection platforms.  Noel attended Miami Dade College, is a member of the American Association of Health Administration Management (AAHAM); has served two terms as president of the Florida Chapter of HFMA and was appointed to HFMA’s National Advisory Council for Revenue Cycle.

What You Should Know:

– CipherHealth, a provider of patient-centered engagement launches a new strategic initiative to leverage Google Cloud’s Vertex AI platform to revolutionize hospital operations and patient care.

– CipherHealth is building, training, deploying, and managing machine learning and AI models across products and across the spectrum of care.

Google Cloud’s Vertex AI to Accelerate Development of AI and Machine Learning Models

CipherHealth selected Google Cloud to facilitate a holistic approach to machine learning. Through Vertex AI, sitting on a foundation of secure, interoperable data, Cipherhealth’s data science team can fully construct, fine-tune, oversee and integrate machine learning models into various business processes. They are designed to cater to diverse modalities, such as free text (natural language), voice, vision, and tabular data. The entire initiative exists on CipherHealth’s HiTrust-certified Evolve Platform.

These AI solutions will aim to dramatically accelerate issue resolution, power sophisticated patient experiences, enhance automation, and lead with bi-directional communication. Ultimately, CipherHealth plans to use AI to create personalized care pathways for patients at all stages of care. The highly tuned model will engage, analyze, and adapt to the patient’s inputs to deliver timely insights to clinicians and an intuitive, empathetic experience for patients.

“We’re thrilled to leverage the full power of AI and machine learning to help healthcare leaders improve patient experience, staff satisfaction, and hospital operations,” said Nate Perry-Thistle, CipherHealth Chief Product & Technology Officer. “This important step will empower us to change the paradigm of patient engagement and help hospitals drive transformative growth.”

Functional balance is a critical health metric. One-quarter of seniors – about 36 million – will fall each year in the U.S. and more than 36,000 of them will die as a result. That translates into one fall every second of every day. The problem is growing, as fall-related deaths among seniors in the U.S. have more than doubled over the past 20 years, according to a research letter in JAMA Open.

It is also an expensive problem. According to a study in the Journal of the American Geriatric Society, non-fatal falls cost an estimated $50 billion in medical expenses, while fatal falls account for an estimated $754 million. 

As is often the case in today’s healthcare environment, advanced technologies powered by artificial intelligence (AI) and machine learning (ML) are emerging as a potential solution, helping to both prevent falls and accelerate recovery when falls lead to injury – particularly when they can be used in the comfort of the patient’s home.

Concerning Times for Seniors

The Centers for Disease Control and Prevention (CDC) considers falls to be a public health concern, and with good reason. One out of every four seniors fall each year and one out of every five falls causes an injury. Further, of those seniors reporting a fall, more than 45 percent fell two or more times. 

According to the Journal of the American Geriatric Society study, a significant percentage of healthcare expenditures among seniors in the U.S. can be attributed to falls, including approximately six percent of Medicare and eight percent of Medicaid expenditures. Falls among older adults also accounted for nearly 12 percent ($29 billion) of spending on home health services, long-term care facilities, and durable medical equipment.

The aging population suffers from multiple comorbidities and in many cases, they are simply not seen often enough by the proper clinician. Due to this fact, the frequency and severity of fall injuries have made prevention a priority. The more often a clinician assesses their patient, the earlier they can detect a problem, treat it, and improve the outcome. Reducing falls among older adults through strength and balance exercises and medication management could lead to a substantial reduction in healthcare spending. Further, multifactorial interventions, including screening and assessment of fall risk, which are often managed in clinical settings, have been shown to reduce falls by as much as 24 percent.

At-Home Prevention

Seniors who are at risk of falling and who have already been injured during a fall would benefit from new AI/ML-powered motion sensor technology designed for in-home. The latest medical-grade solutions monitor balance and movement and send critical information from the patient’s home back to the healthcare provider improving care management and outcomes.

Structured balance and gait tests are administered by the patient or caregiver to monitor for changes or trends. As a therapeutic tool, patients perform appropriate rehabilitative movements while AI/ML-enabled affixed motion sensors instantly and accurately “feedback” the movements using graphics and numerical display of information. This interactive biofeedback function brings an element of gamification to recovery and prevention, helping patients more accurately perceive changes in their neuromuscular activity for improved adherence and performance.

The patient’s data is sent back to the clinician in real-time, allowing them to monitor and actively ascertain progress against baseline metrics, without requiring an in-office follow-up evaluation – reserving limited staff resources for more critical patient care functions. It also delivers a range of motion, fatigability, and symmetry metrics for longitudinal monitoring.

Importantly, AI/ML-enabled sensor technology is also able to quantify previously unmeasurable parameters like pain, as well as objectify physical, surgical, pharmacological, and cognitive therapies. As a real-time biofeedback therapeutic tool, sensors also defocus patients’ attention on the pain and discomfort and redirect their focus on their functional performance. All of which contribute to lower care costs and improved patient outcomes.

Demonstrated Impact

At the Virginia Commonwealth University (VCU) Department of Neurosurgery, John Ward, MD, MSHA, is using such AI-enabled functional motion analysis to monitor patients’ gait and balance throughout the course of treatment for normal pressure hydrocephalus (NPH), a condition thought to occur in as much as one percent of the medical population. Caused by abnormal buildup of cerebrospinal fluid, NPH may affect normal movement and can lead to incontinence and memory issues. It is commonly treated by the placement of a ventricular shunt.

Dr. Ward first assesses a patient’s functional gait and balance pre-spinal tap to provide a baseline. This is followed by ongoing assessments to help guide any appropriate shunt adjustments and necessary treatment changes. Because results are reproducible, which is critical for longitudinal monitoring, they are both clinically reliable and contribute to broadening the understanding and treatment of NPH patients worldwide.

Transitioning this patient base to remote monitoring is essential. These patients are typically elderly, immobile, have difficulty getting to the healthcare facility, are at a greater risk for falling, and may have limited at-home oversight. Due to the nature of the disease, frequent shunt adjustment based on frequent monitoring of functional mobility is a big step forward in the treatment of NPH.

The outcomes realized by Dr. Ward’s NPH patients show why demand is increasing for clinically actionable and objective functional motion data to inform, demonstrate, and evaluate the short- and long-term efficacy of numerous diagnoses and treatments – demand that is now being met by coupling advanced AI/ML with mobile technologies to enable effective, real-time capture of motion data in clinical and real-world settings. HIPPA-compliant cloud-based analytics software then delivers that information back to providers as precise, accurate, and reproducible assessments, therapeutic decision support, and actionable data, stratifying risk, improving outcomes, and increasing the overall efficiency of the healthcare delivery system. 

“If the patient’s risk can be objectively assessed, then appropriate interventions can be performed to prevent falls and subsequent injury,” says Dr. Ward.

An AI/ML-Powered Solution

At a time when fall prevention is emerging as a leading public health concern, AI/ML-enabled technologies offer a promising solution. For seniors who need to improve their balance or recover from fall-related injuries, AI/ML-driven motion sensor technology speeds recovery and improves outcomes. It allows patients to actively and compliantly engage in highly effective in-home rehabilitation that is remotely managed by clinicians and other healthcare providers, improving balance, and reducing risks without sacrificing quality of care. 

About Dr. Frank Fornari

Dr. Fornari, co-founder of BioMech, brings more than 35 years of research, scientific and executive experience to this biotechnology group. Previously Dr. Fornari co-founded and was the CEO and medical director of Dominion Diagnostics, one of the country’s leading pharmaceutical monitoring laboratories and served there from 1997 to 2011. 

Prior to Dominion, Dr. Fornari was a research scientist and consultant to the pharmaceutical industry. He has extensive experience in basic scientific research, clinical medicine, toxicology, chemistry, teaching and drug development and has managed several academic, industrial and clinical laboratories and research facilities. Dr. Fornari has authored numerous scientific publications, has regularly presented at national scientific meetings, has been a member of many scientific societies, holds several patents and is a frequent speaker and industry expert in molecular genetics and pharmacology. 

What You Should Know: 

– Evidation has been awarded a contract by the Biomedical Advanced Research and Development Authority (BARDA), part of the Administration for Strategic Preparedness and Response within the U.S. Department of Health and Human Services (HHS), to fund the 2023 Influenza-like Illness (ILI) Detect & Protect Studies.

– This contract will also fund research to expand and better understand the potential impact of Evidation’s FluSmart program to encourage protective health behaviors. FluSmart pairs machine learning predictions with behavioral nudges, education, and personalized insights to help people better understand their risk of flu-like illness and take early action to help prevent transmission and severe disease.

Detect & Protect Studies

The Detect & Protect Studies will support the development of machine learning models to detect respiratory viral infections with data collected from consumer-grade wearable devices. This research will take place during the 2023-24 cold and flu season and bring together new devices and data types, serial PCR testing, and Evidation’s health measurement and engagement platform to create next-generation models for earlier and pre-symptomatic detection of COVID-19, influenza A and B, and respiratory syncytial virus (RSV).

The expanded partnership is the third BARDA award issued to Evidation to advance understanding and detection tools for respiratory infections. The first award funded the Home Testing of Respiratory Illness Study, a prospective flu monitoring study and benchmark dataset that used consumer wearables, self-reported symptoms, and PCR testing to understand and characterize respiratory virus onset. 

“Since our first study with BARDA in December of 2019, with the participation and consent of hundreds of thousands of Americans nationwide, we’ve learned how high-resolution data from consumer grade wearables–heart rate, sleep, physical activity and other more specific signals–can help identify when someone may be experiencing symptoms of flu-like illness,” said Christine Lemke, co-founder and CEO of Evidation. “We are excited to expand, validate, and improve our models, and find new ways to prompt people to take appropriate action when it matters most.”

Healthcare providers are feeling the industry’s talent shortage in their administrative functions as well as clinical areas. One case in point: revenue cycle management. In a recent study, 63% of healthcare providers reported an inability to fill key revenue management roles. Reduced revenues and cash reserves are almost inevitable when providers don’t have sufficient staff to follow up on claims, manage appeals and help patients understand their financial responsibilities and options. 

Fortunately for providers, revenue cycle automation offers timely, cost-effective solutions to the talent shortage. Robotic process automation (RPA) is proven technology for automating rote, repetitive processes that involve multiple steps and systems and/or substantial human and machine interaction. Using software bots, RPA essentially mimics the keystrokes of human operators. Many bots can be developed and deployed in a matter of weeks. They then work tirelessly and accurately, including during off-hours. At one institution, a software bot cleared a backlog of thousands of claims status checks in just a single weekend. It would have taken humans hundreds of days to accomplish that task. 

RPA is just the start. With software bots streamlining processes and improving data accuracy, the foundation is set for creating more sophisticated automation solutions built on AI and ML models. These can tackle more complex tasks that involve following business rules and making decisions based on the models’ data analysis.  These AI/ML solutions often are more expensive and take longer to implement than RPA. They are best suited for providers that already have standardized processes and cleaner data from their existing automation. 

Unlock revenue fast with robotic process automation

Automating RCM tasks frees up revenue management professionals to take on other, more complex activities, such as providing financial counseling during patient pre-registration activities. RPA solutions will also enable providers to improve revenue cycle productivity without adding additional employees. Many RCM tasks are excellent candidates for automation via software bots and RPA, including: 

Claims status checks. Bots can look up claims data and other information in payer portals, then update systems and even initiate next steps, eliminating these rote tasks and returning time to revenue professionals.

Automated patient pre-registration. RPA bots can link applications and systems together to automate more complex transactions and extract more value from them. Take a patient-facing, self-service registration portal. After a patient agrees to interact digitally with the provider, RPA bots can download patient registration requests; retrieve patient data from an electronic health record (EHR); then update the patient engagement system. The update can trigger the engagement system to send the patient a self-registration and payment link. When the patient completes those steps, the RPA bots can access the payment and patient demographic data and update the provider’s EHR. 

Digitally enabled prior authorization. Bots can easily retrieve patient data, insurance details, CPT codes, physician details, diagnosis codes and schedules from an EHR; flag cases requiring prior authorization; and submit them digitally to a payer portal. Then bots can update records with approved requests. They can also automatically route denied cases requiring additional information to the right clinicians, then refile them when updated. 

What about AI? 

RPA software bots essentially follow sets of rules. While a rule set can be complicated and involve several systems, software bots generally are making preprogrammed if/then decisions. Outliers can be routed to finance professionals for follow-up. 

In contrast, automation solutions that incorporate AI and machine learning algorithms can evolve and eventually make autonomous decisions. Put very simply, an ML algorithm learns from the data sets to which it’s exposed, finding patterns and relationships. This makes ML potentially very powerful. ML algorithms can stratify patient accounts by propensity to pay and automate financial assistance applications. That would reduce costs to collect while improving revenue realization. ML could also identify missing charges and help avoid revenue loss. Those applications, however, are complex. In general, the more advanced the technology, the more time and expense required to implement it. That’s not to take these options off the table. While they undoubtedly will play a role in coding and other tasks, ML and AI applications often are more than many providers need to solve immediate staffing and revenue realization issues. Healthcare organizations must carefully select which processes to automate to ensure the results meet their needs. 

Moving forward with automation

Providers must be clear about what they hope to achieve by automating their revenue cycles and realistic about the time and resources they have available to allocate to the project. The following steps can help guide decisions about which revenue cycle processes to tackle.

• Choose low-hanging opportunities first. Providers should build organizational automation experience before attempting more ambitious projects. The provider that solved its claims status check backlog with RPA initially applied the solution to claims from its largest payer. After succeeding there, the provider then expanded the initiative to claims from its other payers. 

• Choose opportunities that minimize IT involvement. Provider IT professionals often have many competing priorities. Developing RPA software bots requires minimal IT input. 

• Choose a vendor that understands healthcare revenue cycle services. Working with a skilled, experienced vendor helps minimize the time and input a provider’s revenue professionals must give to the engagement. Vendors with RPA and healthcare revenue expertise can build flexible bot frameworks so bots can be extended to other applications with minimal programming.

• Prioritize opportunities that tangibly improve patient and employee experiences. Improving the patient financial experience is a growing priority. Automating tedious, repetitive tasks reduces errors and frees revenue staff to work on more complex issues that deliver more value to patients and the organization. 

• Evaluate the return on investment. Most RPA projects should deliver a return 2 to 3 times greater than the investment. Reconsider projects that have lower anticipated returns or that indicate a long time to ROI. 

The societal changes that have shrunk healthcare’s labor pool are here to stay. Automating the revenue cycle will position providers to improve cash flow, enable their revenue professionals to work at the top of their abilities and offer patients the streamlined digital experiences they increasingly expect. Most importantly, providers will have more of the financial resources they need to focus on their true expertise, delivering patient care and improving outcomes.

About Noel A. Felipe, CRCR
With over 38 years of experience in healthcare accounts receivable management, Noel has a proven track record of developing progressive client-based solutions and building strong cross-functional teams to implement those solutions and maximize client results. At Firstsource, Noel maintains direct account management responsibility for strategic clients and leads the development team for Firstsource’s digital collection and digital pre-service collection platforms.  Noel attended Miami Dade College, is a member of the American Association of Health Administration Management (AAHAM); has served two terms as president of the Florida Chapter of HFMA and was appointed to HFMA’s National Advisory Council for Revenue Cycle.

The power and potential of artificial intelligence (AI) and machine learning (ML) in the healthcare industry are unparalleled. The use of automated technology systems like digital workflows, orchestration and storing of data digitally is helping to shape their application, enabling far greater personalization of treatment for patients and greater efficiency and speed of clinical trials.

As the amount of health data increases, the possibility of establishing a self-driving clinical trial is closer to reality – offering an incredible opportunity to drastically improve the capabilities of pharmacology, medical organizations and patient outcomes.

But where do we begin? What does this reality look like? And how can we ensure unparalleled accuracy? Let me walk you through it. 

Step 1: Crafting Optimal Study Design

The design of clinical research is fundamental to guarantee its success. Having well-developed and concise protocols is vital for successful trials, as inadequate designs can greatly increase costs, reduce efficacy, and threaten success. By utilizing AI and ML technology, research teams can easily create efficient study plans, doing away with manual design and leading to more rapid and precise settings and less potential for mistakes. In addition, AI can help determine the most suitable countries and research sites for a study while suggesting strategies to increase recruitment numbers and expedite study launches based on previously gathered data.

The integration of cutting-edge technologies helps stakeholders, such as regulatory agencies, insurers, and participants, identify the most optimal and beneficial research paths. While comprehensive validation processes allow designers to evaluate and, if necessary, modify the plan before launching the study.

Step 2: Streamlining Key Processes with Automation

Similarly to the role played in study design, AI and ML heavily assist in making site identification, patient recruitment, pharmacovigilance, clinical monitoring and early signal detection more beneficial and efficient for study organizers.  

When it comes to site identification and patient recruitment, one main step in clinical trial execution is finding the correct site that will yield the appropriate patients for the study at hand – a process that has become increasingly difficult as research becomes more specialized. Failure to conquer this step leads to longer timelines, increased expenses and a higher possibility of failure. But the integration of AI and ML can help reduce the risks associated with unsuccessful recruitment. Through mapping of patient populations and identifying the most likely sites to enroll the right people for the study, CROs/sponsors can have greater assurance that their recruitment efforts are having the greatest impact. Furthermore, added analytics investigate factors such as enrollment, safety, compliance and data quality, all of which must be tailored to the specific type of trial.

Not only that, but models can be trained with data from previous studies to predict which sites would be the most successful for a new project. This enables sponsors to open fewer sites, quicken the enrollment process, and lessen the risk of not having enough participants.

Step 3: Optimizing Quality Assurance Through Clinical Monitoring

Once a study has kicked off, multiple pressures follow. To guarantee accurate research is conducted, clinical trials must be continually monitored to identify and remove any potential risks related to patient safety, data accuracy, and adherence to protocol. Usually, this requires a lot of tedious manual work to evaluate risks at individual sites and create strategies to address them. The deployment of AI and ML; however, helps to lighten the workload by providing a way to analyze potential risk factors and develop predictive analytics that generate more meaningful clinical monitoring insights. Technologies can also be used to proactively spot sites that may struggle with performance and to predict which patients may be more likely to experience adverse events.

Step 4: Improving Safety and Overall Patient Care 

Apart from active patient and site monitoring, in pharmacovigilance (the process and science of monitoring the safety of medicines), large numbers of structured and unstructured data need processing to guarantee quality control and oversight. Technologies like optical character recognition (OCR), natural language processing (NLP) and deep neural networks are used to help format data easily. AI/ML are applied to automate laborious, manual processes such as translating and converting safety case records and adverse drug reaction (ADR) reports assisting evaluation and review of the effects of pharmaceuticals. These tools perform data analysis activities to identify potential adverse events, for example, by scanning conversations on social media and other websites, allowing research directors to improve the safety of patients while streamlining their workload.  

In addition, by utilizing the latest technology, algorithms are being developed to analyze medical data such as symptoms and treatments. Technology can assess a patient’s data quickly and flag any irregularities, prompting clinicians to take further action in a timelier manner. As a result, diagnosis is more proactive and effective, which is particularly beneficial in diseases like Alzheimer’s, which is usually only detected after it has progressed.

Transforming Clinical Research One Step at a Time 

All in all, the use of advanced technologies has revolutionized how we conduct research, expedite drug discoveries, diagnose, and treat patients – all while yielding more accurate and comprehensive decision making and more meaningful and precise outcomes. As we approach the reality of the self-driving clinical trial, we must focus on the elements that will make new technology integrations a success. 

In my opinion, this starts with establishing a robust infrastructure that supports trustworthy and unbiased data aided by protocols that meet global reach, security, and regulatory requirements. Vital to this idea is the ability to enable the processing and storage of a variety of content types (e.g., video, documents, audio, devices, and data) to provide users with a seamless experience. Additionally, it is essential for users to ensure ongoing monitoring and maintenance of the technology for reliable and unbiased results—as advanced technology is only as smart as the data it’s fed. 

About Gary Shorter

Gary Shorter is the Head of Artificial Intelligence at IQVIA Technologies, a global provider of advanced analytics, technology solutions, and clinical research services to the life sciences industry. Gary Shorter holds an MSc and has served as a global biostatistics lead for multiple compounds in clinical trials. His 25-plus years of experience allows him to bring the same level of quality and domain expertise to the realm of AI, to ensure that quality AI tools are built and validated to the rigor of regulatory agencies’ expectations. His recent products include Auto-eTMF and Auto-Translation specifically trained to clinical operations needs. 

What You Should Know: 

• AWS and Allen Institute for Brain Science are collaborating on a five-year project to create the largest open source database of brain cell data in the world named the Brain Knowledge Platform (BKP). 
• The strategic global collaboration for the U.S. National Institutes of Health (NIH) BRAIN Initiative Cell Atlas Network (BICAN) with participation from 17 other institutions around the globe including MIT and Princeton will pinpoint precisely why diseases like Alzheimer’s and Parkinson’s occur, to leading to advancements in the clinic.
• Allen Institute is evaluating generative AI services including Amazon Bedrock, to integrate foundation models into the platform. By incorporating generative AI, researchers can explore uncharted territories, simulate brain processes, and generate novel insights that may have remained undiscovered otherwise.

 Brain Knowledge Platform

One part of the brain knowledge platform work, led by Ed Lein, a senior investigator at the Allen Institute for Brain Science and a network of neuroscience researchers from 17 institutes across the world, will be to make a new map of the entire brain at cellular resolution. The other part of the effort, led by the Allen Institute’s head of data and technology, Shoaib Mufti, in collaboration with AWS, will be to use this brain map to create the largest open source database of brain cell data in the world. It will be the first of its kind to compile and standardize massive datasets on the structure and function of mammal brains.

The workhorse of the platform is single-cell genomics. Thanks to new technologies that measure the genes being used within individual brain cells, researchers can now better understand the brain’s cellular complexity and the genes that give cells their distinct functions. These highly detailed cell atlases will help researchers understand the origins of disease and, eventually, allow clinicians to pinpoint why diseases like Alzheimer’s and Parkinson’s occur.

“Centuries ago, we had colorful but rather crude maps of what people thought the Earth’s surface looked like. Cartographers knew roughly where the continents and islands were, and what the major geographic and political subdivisions were, but they were not very accurate,” said David Van Essen, a professor of neuroscience at Washington University in St. Louis, Missouri, and an Allen Institute research partner. “Recently, there’s been an explosion of information and better technology with satellite images and vastly more accurate and user-friendly navigation tools. What we aspire to do for the human brain is to get better and better maps with greater and greater detail, which will become powerful navigational tools for brain disorders.”

What You Should Know:

• Operational limitations—such as staffing shortages, variable patient flow, and poor visibility into resource availability—can make it difficult for healthcare organizations to effectively manage their capacity, leading to volatile workloads that see resources either under- or over-utilized. Some organizations are using capacity optimization management technology, which includes artificial intelligence (AI) and machine learning (ML), to collect data that can improve resource management, especially regarding the utilization of scarce assets like operating rooms, inpatient beds, and infusion centers.
• A new report – KLAS’ first to focus on capacity optimization management—shares what technology is being used and what outcomes it is driving for customers.

Analysing the Benefits Organizations Are Seeing Due to Capacity Optimization Management

Each year, KLAS interviews thousands of healthcare professionals about the IT solutions and services their organizations use. For this report, interviews were conducted over the last 12 months using KLAS ’ standard quantitative evaluation for healthcare software, which is composed of 16 numeric ratings questions and 4 yes/no questions, all weighted equally. Combined, the ratings for these questions make up the overall performance score, which is measured on a 100-point scale. The questions are organized into six customer experience pillars—culture, loyalty, operations, product, relationship, and value.

Capacity optimization management software utilizes data analytics, artificial intelligence (AI), and machine learning (ML) to enhance patient access, care delivery, and resource management, such as staff, beds, and operating rooms, in healthcare systems. This technology is applicable at both the department and organization levels, and current market solutions are available as standalone products that can integrate with patient flow, transfer center, and command center solutions, or as part of a comprehensive patient flow and transfer center suite. The provided chart outlines the validated solutions offered by various vendors and whether they are standalone products or part of a suite.

Key findings are as follows:

1. Strong Relationships from LeanTaaS Drive Value; Early Feedback on Qventus Also Shows Positive Impacts: LeanTaaS and Qventus offer standalone solutions for capacity optimization, and both are seen by respondents as driving strong outcomes and providing high value. Across LeanTaaS’ large customer base, respondents consistently report high satisfaction. They highlight the frequent, collaborative meetings they have with vendor representatives. The vendor’s guidance has allowed respondents to quickly adopt the solution and drive outcomes—e.g., increase patient throughput/capacity, decrease wait times in infusion centers, and improve block utilization in operating rooms (ORs). The limited number of Qventus respondents are mostly perioperative customers. They say Qventus understands their needs and effectively adapts to changing conditions, thus helping drive organizational efficiency. Specific outcomes include increased patient throughput/capacity and decreased length of stay; a few respondents cite the ability to maintain care levels despite having less staff. Some respondents experienced slow, resource-intensive implementations, and one respondent noted the need for improved change management practices.
2. EMR Vendor Epic Highlighted for Integration, Out-of-the-Box Functionality; Customers Want Additional Enhancements in the Future: Epic is the only EMR vendor in this study, and interviewed customers use the capacity optimization functionality via the following applications: Grand Central, Cadence, OpTime, MyChart, Cogito, Slicer Dicer, and Cognitive Computing. Multiple respondents mention Epic’s proficiency in integrating various data sources to provide a high-level view of an organization’s operational status. Further, the product is said to have strong out-of-the-box functionality that helps improve operational communication enterprise wide and increase visibility, leading to improved patient throughput and more balanced workloads.
3. TeleTracking & Care Logistics (Limited Data) Increase Patient Throughput; Implementation & Training Are Opportunities for Improvement: TeleTracking respondents report that increased visibility into their operations has led to actionable insights and greater efficiency. Reported outcomes include increased patient throughput/ capacity, improved productivity, and reduced resource waste. Many respondents also appreciate the accurate, customizable reports. The limited number of Care Logistics respondents all use Hospital Operating System, and these respondents say the vendor is customer focused; Care Logistics makes an effort to help customers successfully use the solution and drive outcomes by working within organization parameters.

In 2020, the Department of Justice estimated that fraudulent, wasteful, and abusive (FWA) billing practices account for more than $100 billion of the nation’s healthcare expenditures.¹ Today, the National Healthcare Anti-Fraud Association (NHCAA) conservatively estimates that healthcare FWA costs the nation about $68 billion annually, representing 3% of the nation’s $2.26 trillion in healthcare spending.²  FWA estimates from commercial health plans range as high as $230 billion annually, or 10% of total healthcare spending.³ 

This lost money is far from a concept or abstraction. Every dollar lost to fraudulent, wasteful or abusive billing hurts patients, honest providers, payors and governments. Third-party benefits providers often receive outsized blame for these costs, when in reality, fraud, waste, and abuse is extremely difficult to detect using conventional methods because providers submitting excessive or fraudulent billing claims are constantly changing their methods to avoid detection. 

Fortunately, new advances in artificial intelligence (AI) technology provide our industry a clear path forward to lowering healthcare costs by reducing excessive or overbilling in a way that rewards good providers and returns more dollars to patient care. By helping healthcare payors detect and prevent fraudulent, wasteful and abusive billing practices in greater quantities and before payments are made, it is estimated that up to $1 trillion in fraudulent, wasteful, and abusive costs can be eliminated from U.S. healthcare by 2030. It’s time to stop blaming benefits providers for spiraling costs and start addressing the technology that powers their day-to-day healthcare claims editing, audit and review systems. Here are the key concepts to consider. 

Static Claims Editing Systems Are Exploitable

Most healthcare benefit systems are based on a static, rules-based or use case-based technology that audit a very narrow set of criteria in determining whether a healthcare claim should be paid to the provider. While these systems do a good job of processing and paying billions of claims each year, their antiquated technology allows hundreds of billions of dollars in excessive bills or fraudulent bills to be paid. It’s not because the claims management companies don’t want to stop fraudulent and excessive billing, it’s because their technology can’t see the exploitation that’s occurring. 

Technology Has To See Provider Behaviors, Relationships and Outliers 

When cases are reviewed and adjudicated using traditional rules-based, use case-based and conventional AI methods, dynamic provider behaviors, relationships and outliers are hard to detect. You have to see a provider’s behavior around a claim and all claims and that provider’s relationships with other providers in order to detect fraudulent, wasteful and abusive billing at a significant level before claims are paid out. This means that this sophisticated, interdependent relationship between providers, a current claim form, historical claim forms, and all other providers in a network has to be able to be identified, analyzed and reported on in less than one second when a claim is submitted for payment. 

The Promise of Artificial Intelligence 

AI scares a lot of people, because it is hard to wrap your arms around what it is. Simply stated, one definition of artificial intelligence (AI) is technology that thinks and does what a human can do, but much faster. Even this simple description leaves out the benefit of unsupervised AI being able to identify an infinite number of ‘math problems’ that a human might not even know to look for in a data set. 

Early AI, and much of the conventional AI used in healthcare FWA detection and payment integrity work today, is not much more than a really advanced Excel spreadsheet. Much of the conventional AI operates using structured machine learning. This means that a machine is trained to perform an algorithm or series of algorithms that take an “if-then” approach to analyzing data. 

These structured machine-learning approaches are very helpful, but miss a lot of the dynamic trends, patterns and outliers that can be detected by advanced, unsupervised machine learning. To ‘see’ all of the FWA activity, you have to deploy unsupervised machine learning that identifies trends, patterns and outliers without being “told” specifically to go perform the task. This enables payors to see new fraud trends and patterns forming in near real-time that are indicators of behaviors and relationships that may be signs of excessive payments, over-payments, or even fraud. In short, you can see things and stop things from happening that you did not even know to tell your technology or staff to pursue. 

The Reality Of Integr8 AI^™ Technology In Stopping FWA 

Integr8 AI technology is a new generation of artificial intelligence that is patented for the detection of operational threats. The first application of the technology is to enable healthcare payors – commercial health plans, TPAs, CMS programs, etc. – to take a dynamic, provider-centric approach to processing, auditing and paying healthcare claims. This technology has proven to increase FWA detection by 2X to 10X in initial commercial use, all because it can “see” FWA activity that conventional technology can’t see. And Integr8 AI can see it in a way that does not slow down the claims editing, review and payment process. 

As one payment integrity executive said, “We need to be able to see the FWA activity that we all know is there. Current technology just doesn’t let us see the volume of FWA that next-generation Integr8 AI technology enables. The best part is that this type of technology operates on top of our current claims editing system. We don’t have to make new capital investments to make a big difference fast.”  

The Bottom Line for Benefits Providers

The battle against spiraling healthcare costs has important implications for every stakeholder in the healthcare value chain, but third-party benefits providers stand to benefit the most when fraudulent, wasteful and abusive costs are controlled. Today, almost a third of all insured Americans receive their health coverage through a third-party provider. Removing fraudulent, wasteful, and abusive costs helps benefits providers lower the cost of benefits for customers and their employees, automate and streamline operations, and increase bottom-line profitability. Technology, like Integr8 AI, enables the benefits to be quantified quickly and recognized almost immediately – regardless of what claims editing and adjudication system is being used. 

Now is the time for benefits providers to embrace sophisticated AI solutions for claims management, moving from a relatively static, claims-based model to a dynamic, provider-centric model. It’s time to take control in the fight against adaptable, malicious actors. That fight starts and ends with thinking about the technologies we have in place. 

About Theja Birur 
Theja Birur is the founder of 4L Data Intelligence and inventor of the patented Integr8 AI intelligence platform. She has 20 years of experience in analytics and artificial intelligence with most of that focused on solving payment and quality challenges for healthcare payers and public health agencies. Her career includes work in the government sector with the Ontario Ministry of Health in Canada, with IBM as a management consultant, and in the IBM Innovation Lab focused on analytics. Prior to founding 4L Data Intelligence, Theja worked as a consultant for the California State Compensation Insurance Fund where she was an Associate Director over Big Data and Data Warehouse functions.

What You Should Know:

• MultiPlan, a provider of data-enabled cost management, payment, and revenue integrity solutions for 700+ healthcare payors announced the acquisition of Benefits Science LLC (BST) for $160M in cash and stock.
• BST was founded in 2012 by a group of MIT-trained experts in data science, including Dimitris Bertsimas, Ph.D., who is recognized as an early pioneer in healthcare analytics and who serves as the company’s chief data scientist. Dr. Bertsimas will continue with BST post-closing. Today, BST’s machine learning algorithms and AI software help about 75,000 employers to predict future risk and manage health plan decisions.
• As part of the acquisition, BST will form the foundation of MultiPlan’s new Data & Decision Science service line by adding new decision analytics and software solutions that intersect with high customer demand and limited competitive offerings.

Acquisition Accelerates Launch of New Data & Decision Science Service Line 

The acquisition of BST strengthens MultiPlan’s foothold in large and fast-growing adjacent markets by unlocking the value of its significant and expanding claims flows for in-network commercial, Medicare Advantage and other government programs, property and casualty, and supplemental insurance markets.

Key products BST brings to MultiPlan’s new Data & Decision Science service line include:

• Price Transparency – a modern self-service software platform jointly developed with MultiPlan that provides prescriptive analytics and applications to help customers benchmark network performance, optimize network design, and improve competitive positioning. In less than a second, the solution can query over 500 billion records of machine-readable payor and provider pricing data now required by regulation to be made public. The solution will aggregate this vast contracted rate information and enrich it with MultiPlan’s extensive proprietary demographic and affiliate data on 1.3 million contracted providers, pricing technology, and deep clinical billing expertise.
• BenInsights – a modern software platform for employers and their consultants that quickly and accurately aggregates a plan’s data and provides highly flexible financial and clinical reporting and decision tools through a self-service software platform. BenInsights also integrates predictive risk modeling and prescriptive analytics and value-added services, such as benefit plan design and optimization.
• Risk Analytics & Insights – solutions that complement existing actuarial-based modeling by applying interpretable risk models, risk scoring, and prescriptive analytics for commercial and government health plans. Among other services, risk scoring can seamlessly attach to MultiPlan’s prepayment claims flows to help identify emergent risks by individual, group, or condition, and prescribe financial and clinical program enhancements across a plan sponsor’s organization.
• Other Market Solutions – a group of software solutions for supplemental insurance carriers and stop loss carriers, including digital claiming, digital underwriting, and targeted selling tools, that help improve plan performance and competitive positioning.

Financial Terms

Under the transaction agreement, MultiPlan will pay a consideration of $160M, comprised of $140.8M in cash and 21.6 million shares of MultiPlan common stock to acquire BST. Additionally, MultiPlan will establish a long-term incentive and retention program pursuant to which BST’s management team is eligible to receive target payments of $66M over three to five years, subject to ongoing service to MultiPlan and to adjustments based on performance relative to annual recurring revenue targets. BST is projected to generate revenues of approximately $16M with breakeven profitability in 2023. MultiPlan expects the acquired company to contribute over $100M of incremental annual revenues within the next several years and to approach corporate-level profit margins at scale.

The COVID-19 pandemic shined a spotlight on the urgent need to modernize the nation’s public health system. Despite success in rapidly developing vaccines, the unprecedented public health emergency also exposed significant gaps in U.S. public health infectious disease data collection and analysis methods which are critical for identifying behavioral risk factors and preventive actions.

The Problem

Unfortunately, inefficiency remains a hallmark of the U.S. public health surveillance system due to the following two lingering issues:

• Disparate data collection systems

The CDC receives data from all 50 states and more than 3,000 local jurisdictions and territories. Hospitals, providers, and laboratories use a variety of systems to collect this data which is then reported to state, city, and local public health agencies. The information is then shared with CDC and other federal agencies. In general, each city, county, and state decide what information is collected, as well as how and when it can be shared with CDC.

What’s more, many current systems rely on disease-specific monitoring and manual data entry, which substantially burdens federal data partners. State and local reports to CDC are often delayed because the systems and data are simply not interoperable.

• Antiquated data-sharing methods

While data is increasingly shared via automated, electronic exchanges, some data is still being sent by fax machines, excel spreadsheets, or even by phone. The CDC encourages standardization, but it lacks the authority to receive data directly without establishing a data use agreement with each state and local jurisdiction. 

As a result, the agency must manually clean the data before conducting the analyses needed to provide an aggregated picture of public health. It can take weeks or even months to share the data with public health authorities, providers, and the scientific community,

The key challenge: how to collect and share information more efficiently so that information turns into actionable insights that can shape important public health decisions?

The Progress

The good news is CDC is leading multiple initiatives to make our public health infrastructure more connected and resilient. The CDC’s Data Modernization Initiative (DMI), launched in 2020, is a multi-year, billion-dollar-plus program to modernize core data monitoring and surveillance infrastructure across the public health ecosystem with the goal of enabling faster, actionable insights to support better decision-making. The recently created Office of Public Health Data, Surveillance and Technology will support this effort.  

Four key actions for fully modernizing the public health data infrastructure, and expanding data collection and sharing are:    

1. Adopt a Scalable, Federated Data Mesh Infrastructure

Today’s network of siloed, disease-specific systems creates significant redundancies and inefficiencies. It cannot scale to support the level of data aggregation, access, and speed public health agencies need. 

A scalable, federated data mesh infrastructure would allow federal agencies to curate high volumes of rich, interoperable data across their ecosystems. They could then accelerate their aggregation and analysis, and in turn, their public warnings and outreach, which are critical for fast-moving threats such as infectious diseases. 

By decentralizing data repositories, a data mesh allows those who are most knowledgeable about their data to control it, namely the public health entities functioning as nodes in a network. Via the mesh, the CDC would engage with electronic health records (EHRs), lab reports, genomic sequencing information, immunization, and other records. State and local agencies would then similarly engage. With CDC defining mesh policies and managing the mesh, data can be ingested, cleaned, standardized, and provisioned for use. 

With such a decentralized information technology architecture, federal agencies could also integrate technology to facilitate HIPAA-compliant patient record matching. This could be achieved without creating bottlenecks typically associated with centralized reporting and dissemination. 

Powered by robust metadata, search features and a centralized data catalog, the mesh would enable authorized personnel to effectively find, access, aggregate, and analyze public health data. This information could also be merged to support the principal guidelines for sharing and managing data adopted by research institutions worldwide, known as the FAIR Principles (Findable, Accessible, Interoperable and Reusable).

2. Protect Privacy 

Protecting the confidentiality of patient health information must be a top priority when modernizing public health infrastructure. The data mesh described above can integrate privacy-preserving record linkage (PPRL) technology which allows for data to be linked across different data sets without exposing individuals’ personal information.

PPRL technology maintains HIPAA compliance while enabling the matching of identifiable patient data without compromising patient privacy and confidentiality. For example, PPRL employs hashing to convert variables such as names, birthdates, and addresses into encrypted tokens that preserve the original values.

Linking data at the patient level enables a comprehensive view of an individual’s health, allowing researchers to answer questions that would otherwise require extensive primary data collection or complex data use agreements.

By integrating PPRL with standardized Fast Healthcare Interoperability Resources (FHIR) data components, public health agencies would be able to ingest and collect data from multiple sources and feed it into scalable analytics and modeling tools.      

3. Expand Data Sources

Currently, limited  EHR and social determinants of health data (such as access to transportation, rates of chronic disease, food insecurity, and crime) are interoperable via the established standard – the United States Core Data for Interoperability (USCDI). This data should be augmented by structured health data which is currently siloed in other agency systems including:

• Geospatial data such as walkability and access to care
• Remote-sensing data, such as wastewater testing and satellite imagery
• Mobility data from smartphones, GPS, and sensors along highways 

By layering additional data from siloed health systems and non-health sources, public health agencies could enrich the baseline USCDI data to gain deep insights. Recent efforts demonstrate the value of multilayered data to track the spread of COVID-19 in wastewater samples across the country, understand the impact of social distancing during the pandemic, and predict obesity rates.     

While encouraging, however, these results are limited in scope. Real-time, actionable surveillance at scale is impossible because of the lack of interoperability across data sources. Alternate approaches that bring more data into public health models and simulations must be pursued.

By extending interoperability and connecting the universe of rich, relevant data, public health agencies could boost the accuracy of prevalence estimates, counter-balance biases in traditional data collection, effectively target control and prevention strategies, and better allocate resources.

4. Harness Intelligent Automation 

Modernizing surveillance systems without burdening the public health workforce is a major challenge.

Public health agencies at all levels face a dire shortage of workers, with roughly 44 percent considering leaving their jobs within the next five years. That’s why public health agencies should adopt intelligent automation tools.

Intelligent automation can significantly improve infectious disease reporting by automating the collection and transfer of relevant health information from EHRs. When a health worker records a particular symptom or disease case in a patient’s EHR, the system could automatically send the data directly to CDC, eliminating current administrative reporting burdens. Improvements in the EHR aren’t limited to public health use – intelligent automation systems can also enhance the care provided to patients and decision support provided to providers.

Intelligent automation systems could also scan and interpret lab reports and clinical notes to uncover disease cases that might otherwise elude health officials, and trigger reports to state and local authorities. Additionally, technology learns and adapts. Powered by artificial intelligence and machine learning, these systems can go beyond simple optical character recognition by leveraging natural language processing to understand context, reduce noise, and improve accuracy.

Conclusion   

With a more modernized data infrastructure, public health leaders will be better equipped to identify and contain outbreaks, understand disease burdens, guide policy changes, evaluate and improve prevention and control strategies, and target research investments. The bottom line: enhanced data collection and analysis capabilities are critical to improving our nation’s public health outcomes.

About Kenyon Crowley

Kenyon Crowley, PhD is the Health Analytics Lead for Accenture Federal Services. Dr. Crowley brings nearly twenty years of health information technology expertise to his role. In his role at Accenture Federal Services, Dr. Crowley will help to accelerate the responsible and ethical use of AI and other advanced analytics tools across the federal health sector to help improve the well-being of all people in the country.

What You Should Know:

• Researchers at the Icahn School of Medicine at Mount Sinai found that applying machine learning models to data collected passively from wearable devices can identify a patient’s degree of resilience and well-being. The study, published in JAMIA Open, supports the use of wearable devices, such as the Apple Watch, to monitor and assess psychological states remotely.
• The researchers note that mental health disorders account for 13 percent of the burden of global disease and that there are disparities in access to mental health care. Therefore, a better understanding of who is at psychological risk and improved means of tracking the impact of psychological interventions are needed. Wearable devices could provide an opportunity to improve access to mental health services for all people. “Wearables provide a means to continually collect information about an individual’s physical state. Our results provide insight into the feasibility of assessing psychological characteristics from this passively collected data,” said first author Robert P. Hirten, MD, Clinical Director, Hasso Plattner Institute for Digital Health at Mount Sinai. “To our knowledge, this is the first study to evaluate whether resilience, a key mental health feature, can be evaluated from devices such as the Apple Watch.”
• The study analyzed data from the Warrior Watch Study, which comprised 329 healthcare workers in New York City who wore an Apple Watch Series 4 or 5 and completed surveys on resilience, optimism, and emotional support. The metrics collected were predictive in identifying resilience or well-being states, supporting the further assessment of psychological characteristics from passively collected wearable data. The researchers intend to evaluate this technique in other patient populations to improve its applicability.
• In essence, the study highlights the potential for wearable devices and machine learning models to monitor and assess psychological states remotely, improving access to mental health services for all people. Further research is needed to refine the algorithm and improve its applicability in a range of physical and psychological disorders and diseases.

What You Should Know:

• Healthcare providers face the difficult challenge of coping with an ever-increasing workload while still providing high-quality patient care and trying to retain their staff. 
• As the CEO of CalmWave, Inc., Ophir Ronen is an expert in both patient outcomes and staff retention who understands the importance of leveraging AI technologies to reduce alarm fatigue and deliver more efficient quiet care. 

Delivering Efficient Care by Reducing Alarm Fatigue

Alarm fatigue is a critical issue that plagues the healthcare industry, particularly in intensive care units (ICUs) and critical care settings. It occurs when healthcare professionals become desensitized to the constant barrage of beeping alarms from medical equipment, which can lead to missed or delayed responses to life-threatening situations. The over-reliance on alarms often results in false alarms, creating a sense of mistrust and a potentially hazardous environment for patients and clinicians. This problem has been exacerbated by the increasing complexity of medical technology and the consequent proliferation of alarms.

 CalmWave, Inc., is an innovative solution that addresses the alarm fatigue problem. By leveraging artificial intelligence and machine learning, CalmWave, Inc., filters out unnecessary alarms and alerts healthcare providers only when it detects critical events, thereby enhancing patient safety and improving clinical outcomes. CalmWave, Inc.’s, technology is highly relevant as it enables clinicians to focus on the most critical patient needs and make more informed decisions, improving the overall quality of care delivered in healthcare settings.

In an interview with HIT Consultant, Mr. Ophir Ronen (CEO CalmWave, Inc.) talks about the importance of AI-driven solutions to alarm fatigue.

How can digital health help healthcare providers balance workloads while still delivering quality care? 

Ophir Ronen, CEO of CalmWave, Inc.: Digital health can revolutionize healthcare by optimizing operations health and streamlining workflows, allowing healthcare providers to balance workloads while maintaining high-quality care. One of the critical aspects of digital health is the ability to leverage and activate the vast amounts of data generated by various systems, such as electronic health records (EHR) and connected devices.

By aggregating medical data from multiple sources, including vital signs, EHR, and clinician attrition data, digital health solutions can help identify optimal clinical workloads. Artificial intelligence (AI) plays a crucial role in analyzing this data to generate objective measures to enhance staffing and workflow efficiency.

Furthermore, digital health can alleviate the burden of non-clinical tasks on healthcare providers by automating these processes. This automation allows clinicians to focus on patient care, ensuring the highest quality outcomes. By optimizing workloads and streamlining processes, digital health enables healthcare providers to work at the top of their licenses, benefiting patients and healthcare systems.

How can AI improve nurse retention and patient outcomes? 

Ronen: AI has the potential to significantly improve nurse retention and patient outcomes by addressing critical challenges faced by healthcare professionals, such as alarm fatigue and non-value-added work. By optimizing medical alarm systems and reducing non-actionable alarms, AI can create a more manageable work environment for nurses, increasing job satisfaction and reducing turnover.

In Intensive Care Units (ICUs) around the world, an overwhelming number of alarms (85-99%) are non-actionable, contributing to alarm fatigue, stress, and patient disturbance. AI can aggregate alarm data from various sources, such as pulse oximeters and blood pressure cuffs, to identify optimal thresholds that minimize non-actionable alarms. As a result, the clinical work environment improves, leading to higher nurse retention, and patients can rest more effectively, enhancing their recovery.

Moreover, AI can help alleviate the burden of non-value-added tasks that clinicians face daily. Nurses often spend a significant portion of their time on non-clinical work, which can contribute to burnout and attrition. By automating these tasks, AI allows nurses to focus on patient care, leading to better job satisfaction and patient outcomes.

An example of AI in action is the CalmWave Operations Health AI Platform. This platform analyzes the constant flow of data from vital signs monitors to provide objective measures of clinical workload. It identifies clinicians at risk of burnout, enabling healthcare leaders to make informed decisions and implement strategies to improve nurse retention.

Given the concerns about clinician burnout, what role does reducing alarm fatigue with data-driven insights have in combating burnout? 

Ronen: Reducing alarm fatigue through data-driven insights is essential in combating clinician burnout and improving patient care. As a significant contributor to stress and cognitive overload, alarm fatigue can negatively impact healthcare providers’ mental well-being and job satisfaction.

AI-based solutions, such as CalmWave, can identify the sources of non-actionable alarms and provide data-driven recommendations for clinicians to make informed decisions. By offering real-time insights, these AI platforms enable healthcare professionals to adjust alarm settings efficiently, ultimately reducing false alarms and creating a more manageable work environment.

Reducing alarm fatigue benefits not only clinicians but also patients. Quieter environments allow patients to rest more comfortably, contributing to faster healing and better overall outcomes. By addressing alarm fatigue, AI solutions can significantly enhance the quality of care and support healthcare providers in their mission to provide the best possible patient care.

What are some of the barriers to implementing AI-driven solutions aimed at combating alarm fatigue and clinician burnout in healthcare settings? 

Ronen: Implementing AI-driven solutions to combat alarm fatigue and clinician burnout in healthcare settings faces several barriers, including organizational complexity, risk aversion, cost, data privacy, and clinical risk concerns. Hospitals, being highly complex systems with multiple stakeholders and priorities, often need extensive testing and proven effectiveness to adopt new technologies.

Despite these challenges, AI-driven solutions like CalmWave’s Operations Health Platform can offer significant benefits in reducing costs, maintaining data security, and improving patient care. By providing objective measures of clinician workload, the platform helps to enhance nurse retention, reducing the financial burden of nurse attrition in hospitals.

CalmWave’s platform is also SOC2 Type II certified, ensuring that data remains secure and protected. Regarding patient care, the platform reduces non-actionable alarms, alleviating alarm fatigue for healthcare providers and creating a more conducive environment for patients to rest and recover.

Current healthcare systems are overwhelmed, understaffed, and under-resourced, making it difficult for clinicians and leadership to explore new solutions. Some AI implementations require lengthy integration processes and extensive training, adding to this challenge. However, CalmWave’s human-centric design philosophy focuses on minimizing implementation complexity, providing just-in-time training, and enabling clinicians to use the platform to optimize care readily. Overcoming these barriers and embracing AI-driven solutions like CalmWave can significantly enhance healthcare delivery and benefit both patients and providers.

What are other promising implementations of AI that can help alleviate clinician burnout? 

Ronen: Several promising AI implementations under development aim to alleviate clinician burnout. One such development involves integrating patient bedside monitoring equipment, allowing AI-driven platforms like CalmWave to analyze data more effectively and ensure better alarm management.

AI can also actively map individual alarms to specific incidents, generating alarms based on the overall incident rather than each individual alarm. This approach reduces alarm fatigue and cognitive overload, creating a more manageable work environment for healthcare professionals.

In addition to bedside monitoring integration, AI can help reduce paperwork burdens by automating documentation processes, freeing up more time for clinicians to focus on patient care. Increasing efficiency in administrative tasks further lowers the risk of burnout among healthcare providers.

AI can also assist in clinical decision support by analyzing large amounts of patient data and providing healthcare professionals with accurate and timely insights for informed decision-making. This enhancement of care quality reduces the cognitive load on clinicians and contributes to decreased burnout and improved job satisfaction.

What impact does ‘quiet care’ have on the patient? 

Ronen: Quiet care promises to revolutionize care for critically ill patients by providing a truly quiet, peaceful, and restful environment in the ICU, something that has been missing since the beginning of continuous patient monitoring. Clinicians once believed that they and their patients had no choice but to tolerate the noise from monitors, but CalmWave technology is changing that perspective.

Other patient populations, such as those in Neonatal Intensive Care Units and Pediatric Intensive Care Units, have already benefited from noise reduction efforts, recognizing the harm noxious stimuli can cause premature infants, newborns, and children. Studies on Labor & Delivery patients have shown that quieter, peaceful laboring environments lead to more relaxed mothers with lower blood pressure, less labor pain, and calmer babies.

Research also indicates that high noise levels, including alarm noise, in ICUs can negatively affect patients’ sleep quality and duration. Insufficient sleep can increase the risk of developing delirium, an altered mental state, with potential long-term effects. Reducing non-actionable noise in ICUs significantly improves patients’ ability to recover, as they can rest and heal properly. Quiet care allows clinicians to provide the best care to patients without distractions from non-actionable alarms, ultimately optimizing patient recovery and outcomes.

Though we seldom see their use in our modern world and, even then, only in fiction, there was a time when it was common for people to actually use things like crystal balls and divining rods to try to uncover unknown yet valuable information. As unbelievable as it may seem, soothsayers peered into crystal balls aiming to help seekers look into the future for guidance, while prospectors would rely earnestly on divining rods as they attempted to locate underground riches of water or oil.

While we may still entertain such images in some of the literature, TV, and movie fantasies we enjoy, in our modern professional world, we tend to entrust industry predictions to those with real, practical knowledge of the business landscape, because they trek, mine, and drill there regularly.

The world of medtech is no different, and for me and my team, connected care is the ground we travel, excavate, and explore on a daily basis. As we venture further into 2023, here’s our perspective on some of the connected care trends we expect to see on the road ahead, from digital therapeutics to remote patient monitoring and clinical trial management.

Precision engagement is an emerging development within digital therapeutics

One of the fast-growing categories within medicine today is digital therapeutics (DTx), which is the delivery of evidence-based treatment through digital solutions that help prevent, manage, or treat a disorder or disease. One recent report valued the global DTx market at $4.2 billion in 2021 and predicted it would expand at a compound annual growth rate of 26.1% between 2022 and 2030, with other estimates projecting even faster growth.

Within DTx, the emergence of precision engagement is a development that holds great promise, especially for chronic conditions where day-to-day choices and behaviors have a significant impact on health outcomes—conditions like diabetes, obesity, and hypertension.

While remote patient monitoring is clearly important for giving care teams visibility into the management of a patient’s condition in order to facilitate vital provider interventions, those living with chronic conditions requiring day-to-day management must also make dozens of additional decisions every day. But initiating provider interventions for all of these would simply not be possible nor even desirable. With diabetes, for example, these can range from food and exercise choices to the need to take medications or interact with a medical device, like a glucose monitor or an insulin pen or pump.

Enter precision engagement. Just as precision medicine can utilize a patient’s genetics or metabolic profile to uniquely fine-tune the dosing of a drug to an individual, precision engagement—with the help of AI and machine learning—can be used by digital health developers and physicians to program connected care platforms to issue electronic interventions or “nudges” that are uniquely tailored and helpful to the individual patient.

These digital nudges prompt a patient to take necessary actions throughout the day that are not only personalized to their needs but delivered in a way that is consistent with their lifestyle and preferences, leading to a better likelihood of engaging the patient and, ultimately, guiding them to better health outcomes. These digital interventions are known in behavioral medicine as just-in-time adaptive interventions or JITAI, and they are helping healthcare professionals use software to precisely engage the right patients with the right interventions at the right time.

With precision engagement, these solutions programmed into connected care platforms are able to digitally “learn” about an individual patient’s preferences from their responses to questions and from the daily decisions they make in their self-management as they engage with the platform’s corresponding app. This learning enables the software to personalize future digital nudges for the patient.

Precision engagement software might be used, then, to help identify the right moment of the day to generate a nudge, like suggesting the patient eat an apple or take a walk at a specific time of day because that’s when the individual is most receptive to such a suggestion.

Or, a digital nudge might involve time- or activity-triggered reminders, such as the need to take medication or to sync the patient’s medical device to the connected-care platform. It might even send the patient an encouraging message prompted by their reaching of a daily target, such as meeting a specific exercise goal.

Precision engagement can even tailor the type of communication used for nudges, from the use of a pop-up message or the suggestion of a video or article to the kind of voice used—maybe through empathy or even humor—to deliver the nudge. 

Precision engagement is one of the most exciting new developments within digital therapeutics, using digital health tools to deliver highly personalized, time-adaptive interventions in ways that lead to positive behavior change, extraordinary patient experiences, and improved health outcomes.

The need for greater RPM awareness is resulting in a measured pace of adoption

While necessity may have forced the issue for care teams during the pandemic regarding the adoption of telemedicine appointments, it turns out that remote patient monitoring (RPM) is still “one component of telehealth that has lagged,” according to the Medical Group Management Association (MGMA). In a Stat poll of 586 healthcare leaders taken by MGMA last year, the association found that 75% of medical practices had yet to offer RPM services.

Despite patients’ positive perspectives of RPM, demonstrated outcomes, payor recognition of RPM’s value, and the establishment of reimbursement mechanisms, the actual pace of RPM adoption has turned out to be more deliberate than these factors had originally led many to predict. In fact, in our work, we’ve found that a large part of preparing providers to make the actual leap to RPM adoption has really been a challenge of growing awareness.

For one thing, we’ve found that in the busy world of providing clinical care, some providers simply haven’t gotten a complete understanding of what RPM reimbursement looks like. So, we continue to chip away at the task of making sure our provider partners have the latest information.

And while some may have caught wind of RPM reimbursement, we’re coming across other providers who have the misconception that only Medicare reimburses for RPM. In reality, there are dozens of private payors covering RPM, with some reimbursing at even higher levels than Medicare. 

Another misconception we encounter among some providers is the mistaken belief that, to get reimbursement for RPM, they must implement every piece of it all at once, from getting patients set up and syncing their data to analyzing the data and providing patient consults. Not only is that not true, but the idea of such a weighty burden is partly why CMS has assigned unique CPT codes for discrete RPM activities. For many providers, implementing RPM is such a significant change management challenge that it actually makes the most sense for them to start small, which they can do by getting patients set up and focusing them on simply sharing their data remotely on a monthly basis. With that, providers can begin submitting for reimbursement, then build from there.

One of the most useful steps for providers unsure of where to begin is to find a reliable partner who specializes in RPM planning and implementation. Resources like AMA’s recently published 12-step RPM Playbook can help, as it covers every stage of establishing a fully operational RPM program.

Pandemic-induced use of decentralized clinical trials provided an up-close view of their efficiencies and is leading to increased adoption

Decentralized clinical trials (DCTs) are trials in which some or all study assessments are conducted at locations other than the investigator site via either tele-visits, mobile or local healthcare providers, local labs and imaging centers, home-delivered investigational products and/or mobile technologies. During the pandemic, when thousands of non-COVID trials—some 80%—were interrupted, virtual trial companies experienced an explosion in demand.

And if market projections are any indicator, demand for DCTs will continue to increase, with an analysis issued earlier this year projecting the global DCT market will grow from $6.1 billion in 2020 to nearly $16.3 billion in 2027.

While the need for social distancing that precipitated the sharp uptick in DCT demand may have subsided from its peak during the pandemic, it’s clear that continued demand for DCTs will be driven primarily by the efficiencies of the model that researchers witnessed first-hand during the pandemic.

One of the biggest advantages of DCTs is how they boost trial enrollment, as they often allow for patients to sign up and participate from home via remote monitoring. Remote participation opens trials and the benefits they provide to those living outside urban centers, which means the trend toward DCTs is also broadening the number and diversity of eligible enrollees.

DCTs can also reduce patient dropout rates and speed up study timelines, two of biggest challenges in life sciences R&D. And they help researchers realize significant cost savings from decreases in the number of physical trial sites and reductions in research staff and travel.

Driven by this wide range of efficiencies benefiting subjects, researchers, and study sponsors, it’s expected the demand for DCTs will continue to ramp this year and in the future.

Overall, we expect 2023 to be a year where our prospecting and development efforts in the connected care landscape will continue producing exciting advancements that will enable us to better support patients living with chronic conditions as well as the physicians and teams who care for them.

About Russ Johannesson

Russ Johannesson is Chief Executive Officer at Glooko, a leading provider of connected care, patient engagement, digital therapeutics, and clinical trial optimization. Deployed in over 30 countries and 8,000 clinical locations, Glooko’s mission is to improve the lives of people with chronic conditions by connecting them with their caregivers and equipping both with digital health technology for improved outcomes.