Dynamic prediction of hospital admission with medical claim data

We present an efficient model adapted for periodically updated data such as the monthly updated claim feed data released by CMS to predict the risk of hospitalization. In addition to processing big-volume periodically updated stream-like data, our model can capture event onset information and time-to-event information, incorporate time-varying features, provide insights of variable importance and have good prediction power. To the best of our knowledge, it is the first work combining sliding window technique with the random survival forest model. The model achieves remarkable performance and could be easily deployed to monitor patients in real time.

The full article can be downloaded below.  

Achieve Privacy-Preserving Priority Classification on Patient Health Data in Remote eHealthcare System

The Wireless Body Area Network (WBAN) has attracted considerable attention and become a promising approach to provide a 24-hour on-the-go healthcare service for users. However, it still faces many challenges on privacy of users’ sensitive personal information, confidentiality of healthcare center’s disease models. For this reason, many privacy-preserving schemes have been proposed in recent years. However, the efficiency and accuracy of those privacy-preserving schemes become a big issue to be solved. In this paper, we propose an efficient and privacy-preserving priority classification scheme, named PPC, for classifying patients’ encrypted data at the WBAN-gateway in a remote eHealthcare system. Specifically, to reduce the system latency, we design a non-interactive privacy-preserving priority classification algorithm, which allows the WBAN-gateway to conduct the privacy-preserving priority classification for the received users’ medical packets by itself and relay these packets according to their priorities (criticalities). Detailed security analysis shows that the PPC scheme can achieve the priority classification and packets relay without disclosing the privacy of the users’ personal information and confidentiality of the healthcare center’s disease models. In addition, the extensive experiments with an android app and two java server programs demonstrate its efficiency in terms of computational costs and communication overheads.

The full article can be downloaded below.  

Top Five Digital Health Technologies in 2019

Digital technologies are constantly evolving and finding new applications in healthcare, even while the industry is struggling with adoption and ‘digital transformation’. Each year new applications emerge, but the underlying technologies driving them remain the same. For 2019, we asked companies around the world one basic question: “Please indicate the key technology which you believe will have the most profound impact on the healthcare industry during 2019?” The results are not very surprising:

The full Forbes article can be viewed at this link.   

Rhode Island docs alarmed by subpoenas they link to EHRs

In recent days, process servers have delivered subpoenas charging doctors at a Rhode Island hospital with medical misconduct — mistakes reported by the clinicians themselves that didn’t injure any patients but were meant to draw attention to risks in their EHR system.

The episode has upset the doctors and their colleagues, who say the state health department’s punitive response could frustrate voluntary reporting of medical errors, reportedly the third leading cause of death in the United States. EHR safety researchers concur, saying that penalizing doctors for self-reported mistakes like this sends the wrong message.

“Anyone punishing individual providers for these events is punishing the wrong thing,” says Jason Adelman, chief patient safety officer at New York-Presbyterian Hospital. “These are system issues, not the provider being reckless. The focus should be on things like EHR usability and safety.”

The full Politico article can be viewed at this link.  

Data Analytics Platform Helps Hospitals Reduce Cost Of Patient Care

Rising pharmaceutical drug prices and declining reimbursements have made it difficult for US-based hospitals to prescribe medicines that their uninsured or under-insured patients can afford. Even with the government’s 340B program, which requires drug makers to discount some of their prescription medicines, hospitals struggle with rigorous regulatory requirements and complex procurement contracts, which make it hard for them to qualify for the discount program. It’s also hard for them to determine which therapies have the highest success rates and then identify which pharmacies have the necessary discounted drugs available.

Enter healthcare data analytics company Sentry Data Systems, which works with hundreds of hospitals and clinics “that serve some of the neediest and most vulnerable patient populations in the country,” says founder and CEO Travis Leonardi. “Our mission is to ensure that hospitals, physicians, and safety-net providers, not drug companies or pharmacies, remain in control of how to best care for their patients,” he says.

Sentry’s healthcare analytics software-as-a-service platform continuously captures terabytes of data containing patients’ protected healthcare information, and then categorizes that data into demographic profiles, conditions, treatments, and invoices. The company now pulls that data daily from the financial, clinical, and pharmacy systems of more than 11,000 US-based hospitals, health clinics, pharmacies, and specialty treatment centers, and then makes it available in its prebuilt analytical models for healthcare administrators to determine which type of treatment would be the most cost-effective for each patient’s specific condition. 

By analyzing patient populations across a range of variables, such as age, gender, and disease, “we’re helping hospitals make informed decisions based on real-world evidence, not anecdotal observations or scientific hypotheses,” says Dr. William Kirsh, Sentry’s chief medical officer. 

The full Forbes article can be read at this link.  

Opioid Prescribing and Physician Autonomy: A Quality of Care Perspective

The public health and social harms resulting from misuse of opioids, particularly substance use disorders related to prescription opioids, have been under intense scrutiny in recent years. Some individuals who receive prescription opioids also use heroin, which has additional risks due to unknown potency and adulteration. Center for Disease Control and Prevention (CDC) statistics indicate that 115 Americans die of an opioid overdose each day. Although the dominant media narrative assigns much of the blame to overprescribing or misprescribing by physicians, the news media less than 20 years ago frequently castigated physicians for failing to provide sufficient pain control and dismissed or ignored the possibility that inappropriate deployment of opioids could lead to addiction. As one article published in American Family Physician in 2000 stated: “Despite recent advances in the understanding of pain management, patients continue to suffer needlessly, primarily because of improper management and inadequate pain medication”. In 2001, a story appeared in the Chicago Tribune reporting that “[a] jury awarded $1.5 million to the family of an 85-year-old man who accused his doctor of not prescribing enough pain medication during his final days”.

Since then, the narrative has changed. Headlines such as “Who Is Responsible for the Pain-Pill Epidemic?” and “Doctors Increasingly Face Charges for Patient Overdoses” have led physicians in many cases to be extremely cautious in prescribing opioids and lawmakers to impose highly specific restrictions on treatment use of opioids. Legal actions seeking to impose liability on opioid manufacturers for their marketing practices have focused on the alleged effects of those marketing efforts on physician prescribing. Yet, nothing can relieve the responsibility of the physician as the party ultimately responsible for the decision to prescribe a controlled substance. That responsibility is grounded in professional obligations to treat patients appropriately and according to accepted standards of care, violation of which can lead to professional discipline, peer review actions, and quality-assurance measures. This article evaluates the effectiveness of recent legislative mandates and restrictions on opioid prescribing and proposes alternative frameworks for combatting and preventing harms caused by the misuse of prescribed opioids.

The full article can be downloaded below.  

2019 HEALTHCARE TRENDS FORECAST: THE BEGINNING OF A CONSUMER-DRIVEN REFORMATION 

The well-known proverb “may you live in interesting times” is considered by many to be a blessing, yet others believe it is actually an ancient curse. People’s interpretation tends to correspond with their own appetite for change, as well as their comfort (or lack of) in the face of ambiguity. Regardless of the adage’s origin and intent, many would agree that these are interesting times for healthcare. In fact, at HIMSS we believe healthcare is currently undergoing a period of reformation on a scope and scale that is unmatched.

A perfect storm of factors – including the shift toward value-based care, rising costs, health system consolidation, the approaching silver tsunami, regulatory pressures, increased consumerization, major technology players entering the market and the ever-expanding potential of digital health tools – are coalescing and fundamentally disrupting business models. Traditional healthcare institutions are in reactive mode. Upstarts are finding that healthcare is not as easily disrupted as industries like retail. For nearly everyone, there are more questions than answers.

At HIMSS, we believe it is our responsibility to help the industry make sense of these changes and tap into the promise and potential of information and technology. With that premise, we are introducing an annual forecast report. Bringing together insights from leadership across HIMSS and our subsidiaries, we’re aiming to shine a light around the corner and help illuminate the path to clinical and financial health.

Read on for our predictions for the industry in 2019.

The full article can be downloaded below.  

The Deadly Consequences Of Financial Incentives In Healthcare

For decades, elite business schools have touted the benefits of financial incentives to motivate sales teams, factory workers and rising executives. Results are mixed.

In medicine, financial incentives rarely achieve their intended goals. It’s not because they don’t work. As I tell students at the Stanford Graduate School of Business, it’s because they work too well.

Monetary rewards always change doctor behavior, but rarely achieve the outcomes desired—something my predecessor learned the hard way when he tried to increase patient satisfaction.

This article also provides alternative best practices for improving patient satisfaction that avoids many of the discussed pitfalls of financial incentives.  

Best Practices

• Transparency - Unblind the patient-satisfaction data for every physician and make the figures available for all to see. This allowed everyone to assess their own strengths and weaknesses.
• Education - We created educational programs to help physicians improve their satisfaction scores with personal coaching and development opportunities. 
• Convenient access to care - We made it easier for doctors to satisfy patients by offering convenient access to care through technology.

The full Forbes article can be viewed at this link.  

Filling the gaps in a patient’s medical data

MIT researchers have developed a model that can assimilate multiple types of a patient’s health data to help doctors make decisions with incomplete information.

The field of “predictive analytics” holds promise for many health care applications. Machine learning models can be trained to look for patterns in patient data to predict a patient’s risk for disease or dying in the ICU, to aid in sepsis care, or to design safer chemotherapy regimens.  

The process involves predicting variables of interest, such as disease risk, from known variables, such as symptoms, biometric data, lab tests, and body scans. However, that patient data can come from several different sources and is often incomplete. For example, it might include partial information from health surveys about physical and mental well-being, mixed with highly complex data comprising measurements of heart or brain function.

Using machine learning to analyze all available data could help doctors better diagnose and treat patients. But most models can’t handle the highly complex data. Others fail to capture the full scope of the relationships between different health variables, such as how breathing patterns help predict sleeping hours or pain levels.

In a paper being presented at the AAAI Conference on Artificial Intelligence next week, MIT researchers describe a single neural network that takes as input both simple and highly complex data. Using the known variables, the network can then fill in all the missing variables. Given data from, say, a patient’s electrocardiography (ECG) signal, which measures heart function, and self-reported fatigue level, the model can predict a patient’s pain level, which the patient might not remember or report correctly.

The full MIT News article can be viewed at this link.  

Translating cancer genomics into precision medicine with artificial intelligence: applications, challenges and future perspectives

In the field of cancer genomics, the broad availability of genetic information offered by next-generation sequencing technologies and rapid growth in biomedical publication has led to the advent of the big-data era. Integration of artificial intelligence (AI) approaches such as machine learning, deep learning, and natural language processing (NLP) to tackle the challenges of scalability and high dimensionality of data and to transform big data into clinically actionable knowledge is expanding and becoming the foundation of precision medicine. In this paper, we review the current status and future directions of AI application in cancer genomics within the context of workflows to integrate genomic analysis for precision cancer care. The existing solutions of AI and their limitations in cancer genetic testing and diagnostics such as variant calling and interpretation are critically analyzed. Publicly available tools or algorithms for key NLP technologies in the literature mining for evidencebased clinical recommendations are reviewed and compared. In addition, the present paper highlights the challenges to AI adoption in digital healthcare with regard to data requirements, algorithmic transparency, reproducibility, and real-world assessment, and discusses the importance of preparing patients and physicians for modern digitized healthcare. We believe that AI will remain the main driver to healthcare transformation toward precision medicine, yet the unprecedented challenges posed should be addressed to ensure safety and beneficial impact to healthcare.

The full article can be downloaded below.