Blog

Kate Monica 


Using EHR data, researchers were able to predict short-term mortality for chemotherapy patients through machine learning.

Combining the power of EHR data with machine learning may be the key to more accurately predicting mortality among cancer patients undergoing chemotherapy.

This finding comes from a recent JAMA study by Elfiky et al. that explored the effectiveness of applying machine learning to EHR data to predict patient’s short term risk of death when they start chemotherapy.

While chemotherapy significantly lowers the risk of recurrence in early-stage cancers and can improve survival rates and symptoms in later-stage disease, the treatment is challenging and costly for patients.

“These patients experience burdensome symptoms without many of the potential benefits of chemotherapy,” wrote researchers in the report.

Researchers set out to find a way to more accurately predict mortality risk before administering chemotherapy treatment to ensure patients that undergo the stress and burden of treatment will also reap its benefits.

In the cohort study, researchers analyzed the EHR data of 26,946 patients starting 51,774 chemotherapy regimens at Dana Farber/Brigham and Women’s Cancer Center from January 1, 2004 to December 31, 2014. Researchers identified the date of death for patients by linking their health records to their Social Security data.

The team classified patients by primary cancer and presence of distant-stage disease using registry data codes for metastases. With this information, researchers attempted to accurately predict death within 30 days of starting chemotherapy with a machine learning model based on single-center EHR data.

Ultimately, the machine learning model was able to accurately predict mortality rates despite lacking genetic sequencing data, cancer-specific biomarkers, or any detailed information beyond EHR data. Specifically, patient EHR data used in the machine learning model including symptoms, comorbidities, prescribed medications, and diagnostic tests.

“Mortality estimates were accurate for chemotherapy regimens with palliative and curative intent, for patients with early- and distant-stage cancer, and for patients treated with clinical trial regimens introduced in years after the model was trained,” stated researchers.

Researchers emphasized that EHR data contains “surprising amounts of signal for predicting key outcomes in patients with cancer.”

In addition to proving accurate, the machine learning model developed by researchers would also only minimally increase administrative burden on clinicians. The machine learning algorithm would not require manual data input from clinicians.

Instead, the algorithm could pull directly from existing patient EHRs.

“Although our algorithm was developed using a single institution’s data, its inputs are available nearly everywhere with an EHR,” wrote researchers.

“In addition, no special infrastructure is required to pull these data from an institution’s data warehouse; in the same way that today’s EHR systems pull a rich set of data from a database to present it to clinicians, an algorithm could pull and process the same data in real time using the processing power on a desktop computer,” the team continued.

The team also suggested the machine learning algorithm could potentially be designed to support EHR integration. Healthcare organizations could integrate the algorithm directly into existing health IT systems. 

“Algorithmic predictions such as ours could be useful at several points along the care continuum,” wrote researchers. “They could provide accurate predictions of mortality risk to a clinician or foster shared decision making between the patient and clinician.”

By predicting short-term mortality for cancer patients, clinicians can identify patients who are unlikely to benefit from chemotherapy and instead may be better suited for early palliative care referral and advance care planning.

“For patients receiving systemic chemotherapy, an estimate of 30-day mortality risk may be a useful quality indicator of avoidable treatment-associated harm,” researchers concluded.

Leveraging EHR data to predict patient health outcomes may help providers to avoid clinical decisions that add unnecessary strain on patients for minimal benefit.

Kate Monica 


ECRI recently outlined three ways healthcare organizations can improve health IT use to reduce delayed, missed, and incorrect diagnoses.

Improving communication between providers, diagnostic testing and medication tracking, and documentation through health IT use can help to reduce delayed, missed, and incorrect diagnoses.

This set of safe health IT use recommendations was released as part of ECRI’s Partnership for Health IT Patient Safety collaborative, which was established in 2014. The multi-stakeholder partnership is open to participation from providers, health IT companies, professional organizations, and other industry insiders.

The partnership’s most recent workgroup focused on improving patient safety during diagnostic testing, test tracking, and medication changes. Chaired by Vanderbilt University professor of pediatrics and biomedical informatics professor Christoph Lehmann, MD, the group identified three strategies healthcare organizations should keep in mind when implementing health IT solutions.

In combination, these three strategies can be effective in closing the loop of patient diagnoses.

“The goal of this Partnership workgroup was to look for technology solutions that all stakeholders could implement to close the loop — the tools provided here will help to do just that,” said ECRI Institute Program Director Lorraine Possanza.

First, the workgroup emphasized the importance of developing and applying health IT solutions that will communicate the necessary information to the correct members of a patient’s care team at the right time.

“Improve the transmission of information using standards for the formatting of normal, critical, abnormal-noncritical, and abnormal results,” wrote members of the workgroup in the report.

Efficient and effective communication between testing facilities, pharmacies, providers, and patients can enhance patient care across care settings, ECRI suggested.

“Designing, testing, deploying, and implementing health IT solutions to improve these communication pathways has the potential to make closing the loop a seamless and elegant process, with all diagnostic results and medications communicated to the provider, the pharmacy, and the patient,” wrote the workgroup.

ECRI also emphasized the importance of tracking diagnostic results and medication changes.

“Tracking of diagnostic results and medication changes is a time-consuming, burdensome task, but necessary to ensure a closed loop,” noted report authors. “Identification of interruptions and potential failure points in the process is critical to find and react to failures to close the loop.”

The workgroup recommended healthcare organizations identify where health IT can be useful for resolving deficiencies and improving medication and test result tracking.

Using EHR functionality to track diagnostic testing and medication changes may also be helpful. Furthermore, implementing lab standards — such as LOINC — may help to automate accurate matching of results and ordered tests to close loops.

Finally, ECRI suggested healthcare organizations use health IT to link, acknowledge, and document the review of information and action taken.

“This step includes the actor reviewing and acknowledging or acting upon information,” clarified report authors.

Toward this end, the workgroup recommended healthcare organizations help to improve interoperability by integrating EHR systems and other health IT modules across the care continuum to facilitate communication and documentation across care settings.

“This is to facilitate communication and acknowledgment, including the use of application programming interfaces (APIs) to allow laboratory systems and hospitals to communicate, as well as the use of HL7 and fast healthcare interoperability resources (FHIR) to aggregate and merge patient data from separate data sources,” wrote the workgroup.

Developing EHR functionality such as diagnostic results notifications may help to promote communication, acknowledgement, and documentation of diagnostic testing, medication changes, and actions taken.

Implementing these health IT use practices and closing the loop will help to avoid delayed or missed diagnoses, which may lead to patient harm.

“These safe practice recommendations are a call to action,” maintained the workgroup. “Although the EHR and its technology components have the potential to facilitate timely follow-up across all healthcare settings, it may take regulatory efforts to make this possible.”

Jack Murtha


Digital behavioral interventions can cut down the kind of overprescribing that empowers antibiotic-resistant bacteria — superbugs — in a cost-effective manner, according to a new study.

The report, published this week in the Journal of General Internal Medicine, found that three behavioral economics interventions can reduce the number of inappropriate antibiotic prescriptions, increasing quality of life and decreasing antibiotic resistance and its associated costs.

Researchers from the University of Southern California (USC) Schaeffer Center for Health Policy and LA BioMed, a California-based innovation incubator, undertook the research to help curb a growing public health crisis. Antibiotic-resistant superbugs have infected more than 2 million people worldwide, in part because roughly half of American outpatient antibiotic prescriptions are unnecessary, according to the study.


“Healthcare needs more studies such as these, targeted to identify successful initiatives that are both cost-saving and life-saving,” said David Meyer, Ph.D., who leads LA BioMed.

For the study, researchers analyzed three interventions, including: 1) Suggested alternatives, which leverage digital clinical decision support tools to suggest treatments that don’t use antibiotics; 2) Accountable justification, which requires prescribers to justify their use of antibiotics in a patient’s electronic health record (EHR); and 3) Peer comparison, which entails sending emails to clinicians regarding how their prescribing rates stack up to their colleagues.

These three tactics had already proved effective in reducing the number of erroneous antibiotic prescriptions over an 18-month period.

So, USC and LA BioMed researchers designed a study using the 30-year Markov model, with inputs from the literature and U.S. Centers for Disease Control and Prevention surveillance data. They homed in on 45-year-old adults who had shown signs of acute respiratory infections, for which patients sometimes inappropriately receive antibiotic prescriptions.

Then providers received the three aforementioned digital tools, along with training on guidelines for treatments for acute respiratory infections.

Measuring discounted costs, quality-adjusted life years and cost-effectiveness, researchers found that the strategies were not only successful but also adept at cutting costs. The total cost of each training intervention was $178.21, while suggested alternatives ran $173.22, accountable justification ran $172.82 and peer comparison ran $172.52, according to the study. What’s more, the training group experienced 14.68 quality-adjusted life years, falling short of 14.73, 14.74 and 14.74 for the three groups that received behavioral economic interventions.

The study and others like it could have big consequences. Public health officials are growing increasingly concerned about the potential effects of antibiotic resistance. Some experts have claimed that the phenomenon could result in the next great outbreak. But, as this study suggests, part of the answer could lie in the EHR and a provider’s email account.

Jessica Kent


What are the best population health management strategies for addressing common chronic diseases?

Chronic diseases are among the most costly, prevalent, and avoidable ailments impacting population health.

Conditions such as diabetes, hypertension, and opioid addiction claim thousands of lives and billions of dollars each year. The Centers for Disease Control and Prevention (CDC) reports that chronic diseases account for seven of the top 10 causes of death in the US and consume 86 percent of the nation’s annual healthcare spending.

The increasing prevalence and rising costs of these conditions make chronic disease management one of healthcare’s most challenging and urgent endeavors.

Yet many healthcare professionals struggle to find the time, tools, and resources to meet the holistic needs of patients.

The close association between chronic disease and patients’ social determinants of health adds to the complexity of treating and preventing these disorders.

Providers must consider and address the conditions in which their patients live, work, and play, as well as their ability to exercise regularly and access healthy food, in order to effectively manage and deter chronic diseases.

To confront common chronic diseases and the many factors that contribute to them, stakeholders from across the healthcare continuum will need to develop population health strategies that will improve patient outcomes.

What are some of the most common chronic diseases affecting patients in the United States, and which population health strategies should healthcare stakeholders use to manage these conditions?

DIABETES

According to the CDC, over 29 million Americans are currently living with diabetes. Another 84 million are prediabetic, and even more may be undiagnosed and untreated. The condition also accounts for more than 20 percent of healthcare spending.

Diabetes risk is significantly tied to social and economic circumstances. It is more common among non-white populations, with black, Hispanic, and Native American populations experiencing the disease at much higher rates than whites.

Medication non-adherence is an issue that leads to additional complications for many diabetic patients, and it is also linked to non-clinical factors. A 2016 report from IMS Health found that nearly half of Medicare diabetic patients are unable to keep up with medication adherence due to limited financial resources, language barriers, and insufficient care access.

To increase medication adherence rates, providers can work to engage and educate patients about their medications by developing personalized adherence plans.  

Additionally, providers can coordinate with community, pharmacy, and public health resources to improve adherence rates.

A 2016 study demonstrated that medication adherence interventions that take place at  retail pharmacies can help patients stay on track with their therapies, reduce preventable hospitalizations, and reduce overall healthcare costs.

“Community pharmacists are uniquely positioned to help mitigate the high risk of medication discontinuation and improve adherence for patients initiating therapy because of their access to prescription refill information and frequent interactions with patients,” the study stated.

Managing diabetes goes beyond adhering to medications, however. Patients must also make healthy food and lifestyle choices and regularly check their glucose levels to maintain their health.

To ensure patients are on track with managing their diabetes, providers can engage patients with text messages and mHealth communications. Providers can use these tools to remind patients about upcoming appointments, ensure they are making healthier lifestyle choices, and keep them on track with blood sugar testing.

Healthcare payers can also employ this strategy and launch mHealth programs to improve diabetes treatment, as UnitedHealth Group recently did for its Medicaid Advantage members.

HYPERTENSION

One in every three American adults has hypertension, the CDC states. The condition is strongly correlated with other cardiovascular conditions, including heart disease and stroke, two of the leading causes of death in the US.

The condition is most commonly seen in non-Hispanic black males, and black individuals are twice as likely to die from the condition as whites are.

Improving hypertension rates will require a collaborative approach, according to the CDC.

“Using team-based care that includes the patient, primary care provider, and other health care providers is a recommended strategy to reduce and control blood pressure,” the organization notes.  

A number of health systems and community organizations have taken this approach, working to engage patients and deliver hypertension care directly to underserved populations across the country.

The University of Michigan Health System collaborated with Meijer pharmacies in 2016 to provide more accessible care to adults with hypertension, offering patients treatment and monitoring services in their own communities.

Additionally, researchers at Cedars-Sinai Medical Center recently enlistedover 50 barbershops in the LA area to offer blood pressure checks and pharmacist-led consultations to customers, aiming to enhance chronic disease management tools within the community.

The researchers found that hypertensive customers who met directly with pharmacists significantly lowered their blood pressure rates.

Organizations can also take an analytics-based approach to hypertension management, as Kaiser Permanente illustrated with its Hypertension Program Improvement Process.

Healthcare organizations can utilize clinical analytics and the EHR to create a registry of high-risk individuals who may benefit from lifestyle changes, as well as use clinical analytics algorithms to determine the best treatment methods for hypertension patients.

OPIOID ADDICTION

Opioid addiction is one of the nation’s biggest health crises. Providers can act as the first line of defense against opioid abuse, and organizations such as the FDA have considered addressing the epidemic with mandatory opioid education for all healthcare professionals.

Clinicians have a responsibility to recognize signs of opioid misuse in patients, prescribe alternate treatments, and prescribe opioids more judiciously to avoid long-term consequences.

Providers can utilize state Prescription Drug Monitoring Program (PDMP) data to determine if their patients may be abusing opioids. These programs have shown considerable promise in reducing unnecessary prescription rates and raising provider awareness about potential opioid misuse.

Opioid abuse is also significantly tied to social and economic circumstances, with those suffering from addiction often having deeply rooted social or mental health problems.

As a result, treating and managing this disease requires efforts not only from providers, but also from government officials and community organizations.  

Pennsylvania’s Opioid Data Dashboard, a government initiative to combat the opioid crisis, gives health officials, lawmakers, and the public access to real-time data to help identify trends for future community needs.

The dashboard also helps build predictive analytics models to deliver a comprehensive picture of the epidemic in Pennsylvania.

In addition, CMS recently released a document explaining how states can use telemedicine to treat Medicaid beneficiaries in rural or underserved areas struggling with opioid misuse.

The organization also recommended that states receive federal support to create shared electronic health plans between providers and patients, which would allow both parties to set goals for pain management regimens and counseling.

COPD AND ASTHMA

The CDC reports that nearly 15.7 million Americans have received a chronic obstructive pulmonary disease (COPD) diagnosis, while asthma affects about 25 million individuals in the US.

These chronic conditions cost the healthcare industry billions each year. They are also heavily associated with individuals’ environmental circumstances and are often exacerbated by exposure to air pollutants in the home and workplace.

Research has shown that public health officials can use EHR data from local hospitals to identify specific geographic areas where there is a high risk for asthma.

Once they have identified high-risk areas, public health officials can assess the air quality, and environmental inspectors can evaluate the hazards in the area.

Officials could also use EHR data to identify patients with severe asthma. By developing a registry of patients who are frequently admitted to the hospital for asthma, officials can flag those most in need of care coordination and individuals who might benefit from home visits.

Community care strategies that utilize sensor applications can significantly improve the health of patients with asthma and COPD, as well as identify the environmental factors that can affect patients’ quality of life.

A 2017 program in Louisville, Kentucky doubled the amount of symptom-free days for asthma and COPD patients by attaching a sensor directly to patients’ inhalers to track the number of puffs used per day, how many times patients experienced symptoms, and where they experienced those symptoms.

Nearly 82 percent of participants saw a decrease in inhaler use, while Louisville officials were able to identify high-risk areas and work to improve air quality in these places.

DEPRESSION AND OTHER MOOD DISORDERS

Between 2009 and 2012, depression affected 7.6 percent of Americans aged 12 and older. The mood disorder is more prevalent among minority and lower-income populations, and is also associated with higher rates of chronic disease.

Despite the correlation between mental illness and chronic conditions, only 30 percent of mentally ill patients are screened for chronic disease.

Integrated care delivery that considers a patient’s mental and physical health can significantly improve mental health outcomes and ensure these patients are receiving the care they need.

Organizations can place behavioral health and primary care providers in the same location to improve patient engagement, foster patient self-management, and address the social determinants of health.

Additionally, providers can use web-based risk assessment tools to stratify high-risk individuals and increase depression screenings for patients, particularly for those who are not often screened in traditional settings.

Providers can then deliver proactive, preventative care to these patients, and gather insights on the factors that most often contribute to depression and other mood disorders.

Chronic disease management is a challenging task that can be made easier by collaborative efforts from primary care providers, community organizations, and other healthcare stakeholders.

By working together to develop population health management strategies and manage and treat patients suffering from common chronic diseases, stakeholders can reduce and prevent the prevalence and cost of these conditions.


Tina Reed


An electronic medical record system is being credited with helping a public health system in Ohio reduce its opioid prescriptions for acute pain by more than 60% in the last 18 months.

Officials from Cleveland-based MetroHealth System said they also cut opioid prescriptions by 25% for chronic pain. In all, they estimate they cut opioid prescriptions by 3 million pills.

How'd they do it? Officials pointed to the alerts they set up in the EMR system.

In particular, those alerts for prescribers were set up to flag patients who may be at risk for addiction to guide them toward alternative drugs and lower doses. They also had an alert to add a prescription for the antidote drug Naloxone when prescribing opioids. That alert led to a 5,000% increase in Naloxone prescribing in the past three months.

Beyond the EMRs, officials said every provider that is licensed to prescribe narcotics is required to be trained in alternatives for pain relief and attend mandatory town hall meetings to identify processes and tools for safe opioid prescribing. The health system implemented a safe opioid prescribing simulation program to allow providers to practice difficult discussions with patients seeking opioids.

“We’ve been tackling the opioid epidemic for a long time. Not until recently, did we recognize that providers can do a lot more,” said Akram Boutros, M.D., president and CEO of MetroHealth.

The health system also opened a Pain & Healing Center, which includes alternative pain management therapies such as acupuncture, infusion therapy, reiki, pain management, neurology, psychology and psychiatry.

This comes a year after MetroHealth created an Office of Opioid Safety to focus on education, advocacy and treatment.

It's also part of a broader shift in MetroHealth's approach. In January, MetroHealth began transforming its campus to include far more green space, designing it as a "hospital in a park." At the time, officials said the change wasn't just about beautifying its campus but about incorporating more holistic healing strategies.

Frank Landman


Technology has changed communication and consumerism so fast that we barely notice it anymore. Less than a decade ago, most people couldn’t imagine things like smart homes and augmented reality. Now, they’re regular consumer products.

In healthcare, though, technology’s impact hasn’t been so subtle. As costs rise at nearly unbearable rates and patients become more involved in their care, new technologies play an increasingly vital role in helping healthcare organizations improve the quality of care they provide.

Patients, or the consumers in the healthcare industry, have benefited most from the industry’s technology race. From preventive treatments to virtual care and more effective disease management, these are just a few ways healthcare leaders can leverage tech to transform the industry:


1. Use data and machine learning to prevent catastrophe.

Data collection and machine learning have been two of the biggest disruptions in healthcare, and predictive analytics is the most notable reason why. By using AI to analyze data on patient and population health trends, providers can more accurately predict health crises and pinpoint their origins to prevent the spread of illness.

This tech was put into action last spring when Pariveda Solutions, a strategic services and information technology consulting company, and a Texas-based pediatric healthcare system used predictive analytics to help stop an infectious, hospital-borne disease that was spreading through the hospital at the time. They collected data on who cared for, saw and delivered medicine to the impacted children and tracked the movements of everyone on that list. In four weeks, predictive analytics determined the problem was central line-associated bloodstream infection (CLABSI), and the hospital was able to prevent it from spreading further

2. Make essential care more accessible with telemedicine, while decreasing data security risks.

For patients who experience mobility issues or who live in rural areas, telemedicine has provided a way to seek medical care more conveniently and given patients more choice in healthcare providers. They can schedule their appointments online with a laptop or mobile device, videoconference with providers to avoid unnecessary trips to the doctor’s office and even rely on internet-connected devices for remote health-related monitoring.

On the other hand, telemedicine may expose a healthcare organization and its patients to security risks involving HIPAA-protected information. Hackers and ransomware are rampant, so organizations must ensure their communications are protected. Fortunately, tech has stepped up to help with that as well, as evidenced by companies like Paubox, which specializes in a HIPAA-compliant email encryption service to protect against data breaches.

3. Provide more accurate, cost-effective treatment through electronic record keeping.

Since electronic medical records began to become standard in 2014, accuracy in medications and treatments throughout a patient’s care cycle has improved significantly. When patients move or must visit a specialist provider, EMRs are even more important in reducing errors. Computerized physician’s orders are easier to understand, patients’ compliance with those orders is easier to track and health conditions are being treated without wasting money on needless procedures.

According to an Institute of Medicine report, $210 billion is spent annually on unnecessary medical care each year. Fortunately, electronic patient records can play a role in reducing this waste. The positive effects of EMRs can be seen at Virginia Mason Medical Center, where an analysis of medical claims data found that high-cost treatments for conditions like back pain, headaches and sinus problems were being driven by expensive MRIs and CT scans, many of which were unneeded. Harvard Business Review reports that when the center embedded an evidence-based checklist for ordering advanced imaging into its EMRs, the use of these costly tests fell 25 percent.

4. Use video games and virtual reality to help with rehabilitation.

When video games become physically interactive with the Nintendo Wii, consumers realized that video games could actually help improve their physical health. Because the Wii’s controllers are operated with motion, games that involve movement — such as running, jumping, swinging a racket or rolling or throwing a ball — force players to get up and move to play. It didn’t take long for healthcare organizations to see the potential benefits of the Wii for everything from poststroke rehabilitation to promoting physical fitness in long-term care facilities.

Meanwhile, some in the medical field have envisioned benefits beyond getting exercise. A study published in the Journal of Geriatric Physical Therapy showed that bowling on the Wii helped elderly patients reduce their risks of falling and suffering an injury. Now, rehabilitation experts routinely use video games and virtual reality to help patients recover from serious injuries, cardiovascular disease, trauma and much more. The Creative Media and Behavioral Health Center at the University of Southern California, for instance, has found virtual reality and video games can help motivate patients who do at-home physical therapy.

Among industries increasing their tech use, healthcare has seen some of the most significant transformations from widely implementing advanced technology. The exciting thing is that technology continues to evolve, and with a firm grip on today’s leading advancements, the medical field is poised to transform even further.

Elizabeth Snell

The Healthcare Leadership Council’s roadmap addressing the opioid crisis states improved EHR integration regulatory requirements can help EPCS adoption.

Adopting electronic prescriptions for controlled substances (EPCS) through modern regulatory requirements, including improved EHR integration, is one critical solution to opioid abuse, according to the Healthcare Leadership Council’s (HLC) “Roadmap for Action.”

Having a national prescription drug monitoring program (PDMP) and better opioid stewardship and disposal were also key recommendations from the Council.

More than 70 healthcare organizations, including Surescripts and the National Association of Chain Drugstores have already announced their support of the roadmap.

“EPCS is just one tool in our arsenal to fight opioid abuse,” Surescripts said in a statement. “Robust, electronic medication history data is available nationwide across all care settings. Having an up-to-date view of a patient’s medication history at the point of prescribing empowers prescribers to make the best care decisions for their patients.”

HLC members include chief executives from numerous healthcare stakeholders, including but not limited to hospitals, health plans, pharmaceutical companies, medical device manufacturers, biotechnology firms, and health product distributors.


The council’s National Dialogue for Healthcare Innovation (NDHI) organized the Opioid Crisis Solutions Summit on May 14, 2018 in Washington, DC. The Roadmap was created based on discussions at the Summit.

The Roadmap highlighted five priority areas for healthcare leaders, policymakers, and regulators:

  • Improve healthcare system approaches to pain management
  • Improve healthcare system approaches to prevent opioid misuse
  • Expand access to evidence-based substance use disorder (SUD) treatment and behavioral health services
  • Promote improved care coordination through data access and analytics
  • Develop sustainable payment systems that support coordination and quality care

The Roadmap then discussed specific recommendations for healthcare leaders, policymakers, and regulators, with ten to 15 suggestions for each group.

The Drug Enforcement Administration (DEA) should modernize current regulatory requirements for EPCS. This could improve EHR integration and help reduce the extra cost and burden on healthcare providers.

CMS should “provide a secure electronic transmittal infrastructure that would facilitate electronic prior authorization in Medicare Advantage and Part D and move towards greater standardization and efficiency in the prior authorization process,” the Roadmap recommended.


Healthcare leaders should pledge to adopt e-prescribing for all controlled substances by 2020, and will also need to develop a plan for improving access to a range of evidence-based, non-opioid, opioid sparing, and non-pharmacological pain management therapies.

Healthcare leaders should also work to close the SUD treatment gap by working to increase access to appropriate in-person or telehealth SUD treatment and recovery services.

“Specific efforts could include organizational commitments to reducing care fragmentation, providing or incentivizing medication-assisted treatment (MAT) training in underserved areas, and investing in peer and recovery support workforce and services,” the Roadmap explained.

“While public policy has a vital role to play in removing barriers to advancements in care and empowering accelerated therapeutic innovation, private sector leadership is critical on every aspect of this issue, from improvements in pain management to data-driven proactive interventions to strengthened opioid stewardship,” Roadmap authors concluded.

“By building upon ongoing initiatives that are already yielding promising results, healthcare leaders can and will make a difference in stemming a crisis that has already claimed too many lives and damaged too many families and communities.”   


CMS released its own roadmap earlier this month, highlighting the need for interoperability and clinical data to help curb the current opioid crisis.

The agency explained that it would use health data to target prevention and treatment efforts and identify trends of fraud and abuse among patients.

“CMS is working to ensure that beneficiaries are not inadvertently put at risk of misuse by closely monitoring prescription opioid trends, strengthening controls at the time of opioid prescriptions, and encouraging healthcare providers to promote a range of safe and effective pain treatments, including alternatives to opioids,” CMS said in a June 2018 blog post. “We are also working on communications with beneficiaries to explain the risks of prescription opioids and how to safely dispose of them, so they are not misused by others.”

Leveraging health data to understand opioid use patterns across populations can be greatly beneficial, the agency explained. Additionally, that information can promote healthcare interoperability and health data exchange across the care continuum, helping monitor trends to assess the effectiveness of prevention and treatment solutions. 

“The roadmap is also a demonstration of CMS’ commitment to explore and offer viable options to address the crisis, to share the information we collect with other agencies and organizations, and to protect our beneficiaries and communities affected by the crisis,” CMS concluded.

Jessica Kent


A machine learning tool developed at Carnegie Mellon University uses big data from the electronic health record to accurately predict sepsis.

Researchers at Carnegie Mellon University’s (CMU) Heinz College are applying a machine learning algorithm to big data in the electronic health record (EHR) to more accurately predict sepsis, one of the most dangerous and insidious hospital acquired conditions.

The Sepsis Alliance reports that more than 1.7 million people in the US are diagnosed with sepsis annually. Of those affected by the condition, an estimated 270,000 die each year.

Sepsis is also the number one driver of hospital costs in the US, consumingmore than $27 billion annually. Many times, the infection is acquired in the community, and patients with complex comorbidities are often at the highest risk.

“The problem underlying sepsis is that it’s incredibly heterogeneous,” said Jeremy Weiss, MD, PhD, Assistant Professor of Health Informatics at CMU’s Heinz College, to HealthITAnalytics.com

“Anybody can get sepsis from a multitude of infections, at different sites, and with different comorbid profiles.”

Traditionally, providers identify high-risk sepsis patients by evaluating symptoms and medical histories.

However, Weiss and his team are working to improve the speed and accuracy of sepsis prediction.

Using machine learning and EHR data, Weiss has developed a method of accurately assigning risk scoresto patients, offering a way to catch sepsis earlier than is possible with standard processes.

“EHR data is very detailed. There’s a lot of time-stamped information,” Weiss said.

“A lot of classical analyses don’t get to capture that kind of information. With EHRs, where this data is automatically entered, we can look at the temporal progression of disease and update our risk models more adeptly.”

Weiss and his team utilize this time-stamped data to evaluate information such as blood tests, prescribed drugs, and blood pressure. This data is typically contained in the structured portion of the EHR, which they can access during a healthcare encounter and use to make real-time predictions.

“Data such as lab tests, prescribed medications, and vital signs will inform us when procedures were performed and the background set of diagnoses for the patients upon entry,” said Weiss.

To extract relevant patterns from structured EHR information, Weiss and his team are applying machine learning algorithms, which can analyze many different data points and assist in making accurate predictions, he said.

Weiss and his team are also exploring how clustering of similar cases can help to refine predictive analytics and get ahead of sepsis that may develop in future patients.  

“If we can identify clusters with very specific phenotypes, then we can tailor treatments to those subgroups,” said Weiss.

The Heinz College group is not the first to use machine learning algorithms and EHR data to accurately predict sepsis in patients.

Researchers at the University of Pennsylvania developed a machine learning tool that continuously monitored EHRs and identified patients headed for sepsis or septic shock a full 12 hours before the onset of the condition.

Additionally, researchers at North Carolina State University, in collaboration with Mayo Clinic and Christiana Care Health System in Delaware, have usedEHR data and machine learning to improve how the healthcare system identifies and treats patients with sepsis.

Weiss and his team hope to make similar strides in sepsis prediction and prevention.

“We’re trying to leverage much more information from the EHR to tailor simpler risk scores,” Weiss said.

“While those scores are a really good, quick check for what your general risk is, we’re really interested in having a much better risk estimate that will draw from a lot more signals.”

Bill Siwicki


Hospitals say the AI method is in use today automating things humans do well but don't have the time for – and there's much more to come.

Artificial intelligence is beginning to reshape healthcare and life sciences. And one application of AI, deep learning, is coming into its own.

Deep learning is a type of machine learning based on data representations rather than task-specific algorithms. Learning can be supervised, semi-supervised or unsupervised.

Much of the excitement around AI today is fundamental because of three ingredients: the development of algorithms that make artificial neural networks, the increasing supply of digital data that now can be created, and, critically, the "GPU" chip architecture – it stands for graphics processing unit – pioneered by vendor NVIDIA, said Mark Michalski, MD, executive director of the Massachusetts General Hospital and Brigham and Women's Hospital Center for Clinical Data Science.

"GPU chips are different than the CPU chips that run many of our computers today in that they solve many simple problems simultaneously, as opposed to one big problem at a time, like CPUs," Michalski explained. "It turns our brain's work in a similar way to GPUs, which is perhaps why GPU chips are so effective as tools for machine learning."

Anyone working in deep learning – the machine learning concept behind many recent AI advances – use GPUs. People have applied this technique to all kinds of data, including images, videos, audio recordings, text and more.

"One of the primary reasons we are building our own data center with dedicated GPUs is that we rely on medical data such as MRIs and CTs, which can be a lot of data, and is sensitive from a privacy perspective," Michalski explained. "We can feed the GPU information derived from medical records, like clinical notes, medical imaging data, pathology slides – just about everything we can learn from our records, we try to leverage."

GPUs and deep learning have big potential for how healthcare can manage and interpret imaging and other clinical data. Michalski is a radiologist by training, and in his field, there are many instances where one has to look at medical images and find abnormalities within images, characterize and measure those abnormalities, and describe a diagnosis based on those findings.

"A single radiologist may have to look through thousands of images a day," he said. "Having systems that can help point out an abnormality in a stack of normal images, and further, automatically measure that abnormality like a tumor or a dilated heart chamber, could be a huge productivity advantage for us."

And these advantages aren't just in radiology. Massachusetts General sees opportunities for helping support physicians throughout the hospital, everywhere from finding features in clinical reports to help oncologists identify ideal treatments for cancer patients to helping the operational leadership within the hospital identify opportunities to use scarce resources more effectively.

At the institutional level, executives and caregivers have been on the lookout for solutions to many of these problems for years, and there is broad hope that machine learning may help develop those solutions.

Consider this example of how the NVIDIA AI technology works: First there is a clinical challenge, such as identifying breast cancer cells on a pathology slide with machine learning. The first step in applying machine learning is to identify data that one can use to train a neural network. To do this, one needs sample data that has been labeled – that is, one needs someone to have pointed out where the cancer cells are in the slide and highlighted them appropriately.

"Sometimes to make the models work, we need lots of those examples," Michalski explained. "Getting enough data to make the AI work is often one of the key challenges. This contrasts with some other artificial intelligence applications, such as autonomous vehicles, where there are thousands upon thousands of images to draw conclusions from."

After the training data is secured, one selects the kind of neural network that one wants to use to build an AI model. This is the "brain" that one wants to teach to identify the cancer cells on a slide. Once that is done, one uses the GPUs to teach the brain and, after that, one can use that brain to identify the cancer cells, a process commonly called "inference."

"In healthcare, integrating the model into the workflow in a way that is helpful to the clinician sometimes is one of the more challenging parts," Michalski explained. "Having the brain isn't enough: It has to be inserted into a physician's process in the right way for it to be useful."

Today, Massachusetts General and Brigham and Women's Hospitals are using deep learning to automate the things that humans do well, but don't want to do or don't have the time to do. But  the hospitals are also starting to see that the techniques could have other big potential, such as picking up things in images and medical records that humans don't see very easily, which will be evolving in the coming years.

"We're lucky in medicine to have a lot of exciting advancements every day," Michalski said. "What makes AI so interesting is that it may be able to improve the lives of patients and physicians alike, all while potentially reducing costs. That's why I think a lot of us in the field believe AI and deep learning are going to be very important to the future of healthcare.

Kate Monica 


Researchers named EHR system customization as one of ten strategies that ease patient-centered medical home implementation.

EHR system customization can help to streamline the implementation of the patient-centered medical home (PCMH) model, according to new research published in the Journal of the American Board of Family Medicine (JABFM.)

PCMHs are designed to boost patient health outcomes at both the individual and population level while reducing healthcare organization costs and improving the patient experience. To achieve this aim, PCMHs prioritize enabling care coordination through EHR technology, health data exchange, data analytics, and population health management tools.

Researchers conducted interviews and focus group discussions with providers from 20 Patient-Centered Primary Care Homes (PCPCHs) in Oregon between 2015 and 2016 to learn about the challenges of PCMH implementation.


The team of researchers from the Oregon Health and Science University and Portland State University School of Public Health found that clinic leaders and staff from all 20 PCMHs reported similar challenges when implementing the care model. However, study participants used a variety of different strategies to overcome these obstacles.

Ultimately, researchers identified ten strategies that may be effective in overcoming common barriers to PCMH implementation, one of which was EHR system customization.

Clinic leaders reported that EHR use enabled provider communication and allowed clinicians to access specific health data about individual patients, patient populations, or other clinics.

“However, minute data entry differences could render a search useless if it was unable to capture all relevant data,” noted researchers in the report. “Clinics standardized data entry protocols; however, this did not address all issues of data analysis.”

Clinic leaders reported they had customized their EHR systems with additional tools to ensure the technology included all necessary data analytics capabilities to support the PCMH model.

“EHRs did not provide tools for data analysis, so clinics purchased third-party analysis tools, used basic Excel spreadsheets, or used existing staff to conduct analyses,” explained researchers. “In some cases, a staff member emerged as a data expert, shifting their role into a data management and analysis position.”

“For other clinics, the task was incorporated into each staff member’s role in addition to their existing responsibilities,” continued researchers. “Clinic staff frequently requested funding for data management and analysis staff or tools.”

In addition to EHR system customization, researchers also cited health data exchange as a useful strategy for ensuring continuity of care with providers at outside clinics, hospitals, specialty care facilities, and pharmacies.

“Many clinics were concerned that patients were not receiving continuity of care outside of the clinic via their pharmacists, specialists, and hospitals,” maintained researchers.

“Patients often did not give information to their providers, assuming that if one specialist knew, all providers knew,” the team continued. “Specialists also told clinics that they sent information via the EHR, not realizing that their EHR systems were incompatible.”

Providers made efforts to thoroughly follow through on patient referrals to overcome problems with continuity of care and health data exchange. Some providers formed partnerships with specialists to facilitate patient referrals, while others took time to remind patients and specialists of the myriad communication barriers that could delay referrals.

“Some interviewees reported additional challenges with specialists, and to a lesser degree, hospitals, neglecting to send patient records,” researchers wrote. “This was less of an issue with hospitals, as many had EHR access or formal agreements with the clinics.”

Some clinic leaders tracked and flagged specialists that neglected to engage in health data exchange. Furthermore, some providers threatened to stop referring patients to any specialists that refused to share patient EHRs.

Other strategies clinic leaders used to overcome challenges related to PCMH model implementation included integrating behavioral and mental health into primary care, incorporating screening, prevention, and disease management services, and preventing unnecessary emergency department visits.

Healthcare organizations interested in adopting the PCMH model can use these strategies to streamline the care model implementation process. As more healthcare organizations work to improve the patient experience, boost patient health outcomes, and cut costs, clinic leaders can look toward successful PCMHs as examples of how to achieve value-based care.

“What is important above all else is the dedication to continuous learning through implementation,” researchers concluded.

Jocelyn Kaiser


Who is your emergency contact? The answer to that question, standard in every doctor’s office, has now been used to predict the role of genes in hundreds of conditions, from diabetes to high cholesterol. A new study combined the emergency contact information of 2 million New Yorkers with their medical data to form family trees of heritability—all without ever looking at a patient’s DNA. The approach could be used by clinics widely to predict a person’s disease risks, if patients agree to let their data be used in this way.

“It’s an interesting idea that you can generate family structures across very large data sets” compiled by health care providers “and infer something about the shared basis of disease,” says cardiovascular disease genetics researcher Dan Roden of Vanderbilt University in Nashville, who was not involved with the study.

Hoping to explore the genetics of drug reactions, graduate student Fernanda Polubriaginof and others working in the lab of biomedical informatics researcher Nicholas Tatonetti at Columbia University wanted to determine whether patients at the school’s affiliated NewYork-Presbyterian Hospital were related. “It occurred to us there’s some data that every hospital routinely collects every time a person is admitted,” Tatonetti says: an emergency contact, who also often happens to be a blood relative.

His team pulled those contacts from electronic health records of patients who had given consent to use their information in research. To the scientists’ surprise, about one-third of the emergency contacts had also come to Columbia’s hospital for treatment. They then used the names, addresses, phone numbers, and relationships of these contacts to build 223,000 family trees connecting blood relatives. The biggest had more than 100 patients spread across four generations, Polubriaginof says.

The Columbia team and collaborators at the city’s Weill Cornell Medicine and Mount Sinai health systems also constructed family trees from thir records, bringing the total to 1.9 million patients with 7.4 million relationships. The patients were a diverse mix including Latinos, blacks, and whites.

The researchers then overlaid those trees with information on each individual’s health conditions, gleaned from billing codes and lab tests, and used the combined data to estimate the heritability of about 500 traits and diseases. For many conditions, such as glaucoma, the results matched previous estimates based on twin studies. For other conditions, the huge database may resolve conflicting results, the team reports today in Cell. For example, two small-scale studies have drawn different conclusions about the inherited risk for high cholesterol. The new study found that high levels of high-density lipoprotein, commonly considered the good kind of cholesterol, are 50% inherited, whereas high levels of low-density lipoprotein, the more dangerous kind, are 25% inherited.

Another 400 or so conditions in the paper hadn’t been studied much by geneticists. Sinus infections, for example, appeared to be 85% inherited, which matches up with anecdotal evidence that these infections run in families, Tatonetti says. (Here are the data sets for all 500 conditions.)

The method wasn’t perfect for so-called Mendelian disorders, diseases known to always occur if a person inherits a single copy of a flawed gene—in other words, they’re thought to be 100% inheritable. Sickle cell disease was 97% heritable according to the electronic records analysis, very close to what is expected. But for cystic fibrosis, another such condition, heritability was only 1%. That discrepancy could be explained by many factors that complicate the family tree analysis, including that families with the condition tend to be small, Polubriaginof says.

Family health histories are harder to collect from interviewing patients than you might think, says Roden, who also works with electronic health records. “This could be another way of producing the same information,” he says.

At the same time, the study could make people leery of sharing their emergency contact information, Roden suggests. The researchers stripped names and other identifying information from the patients’ health data after linking it to relatives’ records. But, “The idea that researchers are mining information about your family without letting you know, it does run the risk of alienating people. We have to be pretty careful to make sure the public stays a partner in efforts to grow these large data sets,” he says.

Because of privacy rules, the Columbia team doesn’t plan to share the family trees or disease risks with individual patients. However, Polubriaginof is using the overall results to suggest that Columbia’s physicians could do a better job of screening patients at risk for diabetes and celiac disease, an autoimmune disorder. And the New York City researchers aren’t alone: Academic health centers in Boston and Chicago, Illinois, are already using the team’s formula to trace their own patients’ family trees for research.


By Jeff Robbins, CEO and Founder, LiveData, Inc


The use of Systems of Record, such as EHRs, has been a key strategy for achieving the Institute for Healthcare Improvement Triple Aim: improved patient outcomes and increased patient satisfaction, with reduced costs.

With this in mind, the Health Information Technology for Economic and Clinical Health (HITECH) Act included requirements for meaningful use of EHRs, with significant compliance subsidies. Thanks to the HITECH Act, 83 percent of hospitals had some form of EHR by 2016. However, the promised benefits — no lost data, increased efficiency, and better patient experiences — have not always materialized. Some physicians and patients think healthcare is more inefficient than ever. To combat this, Systems of Engagement are being deployed to augment EHRs, simplify data collection, and share data in easy-to-consume visualizations, getting information to the right clinician at the right time to improve patient care.

Systems Of Engagement

System of Record/EHRs provide a single data source for patient records that a hospital can use for care planning and execution. In subsidizing EHR adoption, the government expected to see 1) improvements in accuracy and completeness of patient information, 2) better coordination of care, 3) secure access for patients to their health data to foster shared decision making, and 4) safer and lower-cost care. Healthcare faces unique challenges in implementing Systems of Record due to privacy and regulatory requirements. EHRs must support data entry requirements and usability features such as:

  • Every patient encounter must be recorded in a way that is effective, efficient, and accurate, and captures the audit trail for the clinician doing the data entry.
  • The Health Insurance Portability and Accountability Act of 1996 (HIPAA) strictly governs data security, entry, storage, and privacy.
  • Archiving data must ensure safety, redundancy, and retrieval for long periods of time.

Meeting these requirements demands significant investment in time and resources. The need for comprehensive, accurate patient data increases data entry time, while the sheer amount of data to store and retrieve requires a large outlay in server space, and more IT help to manage. HIPAA requirements also affect IT resource requirements and significantly complicate the kind of system interoperability that has produced efficiency improvements in other industries.

These issues are reflected in clinical attitudes toward EHRs. Many clinicians feel that EHRs slow down care delivery. This feeling is reinforced when organizations report declines in efficiency and financial losses. Wake Forest Baptist went from a $38.9 million operating gain to a $62.9 million operating loss in the year they implemented their EHR. Only $22 million of the swing was attributable to implementation costs.

EHRs do have a measurable positive effect on patient safety. In a 2012 survey, 88 percent of providers who had attested to Meaningful Use reported that their EHR produces clinical benefits for the practice and 75 percent reported that their EHR helps them deliver better patient care. A 2016 survey showed a 30 percent lower rate of adverse events for surgical patients at facilities using an EHR versus facilities without an HER. So how can healthcare organizations resolve legitimate issues that providers have with their Systems of Record, while maintaining and improving on the patient safety gains that EHR use has delivered?

That is where Systems of Engagement come in.

Systems Of Engagement

Systems of Engagement overlay and complement Systems of Record, allowing users to share and collaborate on mission critical information, in real time or over days and weeks. The term describes everything from email to instant messenger and social media to enterprise platforms for data integration, collaboration, and comprehensive analytics. For healthcare, Systems of Engagement can mitigate the regulatory and safety burdens on Systems of Record, allowing for more rapid testing and targeted performance improvement initiatives.

Healthcare Systems of Engagement have been slow to reach the market and to be adopted. Wariness among hospitals about further investment in software solutions is understandable, given the massive outlays involved in implementing EHRs and the lack of hard data for assessing the impact of new Systems of Engagement. To address this, software companies are partnering with analytics companies or incorporating analytics into their own tools. Analytics can give hospitals insight into the return on investment for Systems of Engagement, in dollars saved and quality improvements.

The recent focus on EHR implementation has also slowed software adoption. The HITECH Act mandated three separate states of attestation for Meaningful Use of EHRs, each more complex than the previous. Given the financial incentives and penalties associated with the Act, hospital IT departments have taken an “all hands on deck” approach to each Stage’s requirements. As Stage 3 requirements were only released in October 2015, many hospitals only now have the resources to consider software approaches to quality and utilization improvements.

Lastly, hospitals as institutions tend to be risk averse, while the apprentice-based system of medical, especially surgical, education favors tradition and preservation of the status quo. In this environment, already reeling from the upheaval of the EHR, an additional layer of software can be a tough sell, but will be needed to meet the challenges of modern healthcare.

In an industry where many still pine for the days of paper records and whiteboards, forward-thinking healthcare organizations will need to show the benefits that data-driven approaches can bring. Investing in Systems of Engagement that have a track record of success is the next step in securing comprehensive institutional buy-in to create a more efficient and effective healthcare world for patients and providers.




Dave Bennett


Because the data capacity for flash drives and secure digital memory card storage increased immensely over the past two decades, mobile phones have gone from simple handsets used for text messaging, phone calls, and the occasional game of Snake, to sophisticated devices that support high-fidelity music libraries, high-definition videos, high-resolution photos and elaborate video games.

Consumer expectation for data storage has also grown over those two decades. The rapid development of new technology—combined with a dramatic reduction in the cost of data storage—has brought to all sectors the possibilities of connection and communication. For healthcare, a wealth of data has revolutionized the way we approach patient care, ushering in an era in which the entire care cycle is fueled by analytics and decision support that improves overall quality and makes the lives of physicians easier.

Patient-as-consumer demands and expectations

The demand for increased data storage has been the driving force behind this rapid expanse of data technology, creating a positive developmental feedback loop. As the capacity for data has increased, so has the consumer appetite for it; consumers demanded greater and more practical data, which opened up previously unexplored avenues for the development of data-driven services. Because patients are also users of other rapidly expanding technologies, the healthcare industry’s goal is to keep up with the expectation for a robust user experience. 

For most industries, including the retail and financial industries, big data is now a tool used to predict consumer trends and provide service improvements, like increasingly targeted advertising, which uses things like social media “likes” and app activity to deliver a personalized experience directly to the individual.

Similarly, the healthcare industry is uniquely positioned to use the increasing visibility of personal data for practical healthcare and treatment solutions. 

Healthcare’s data boom

The increase in data storage capacity has enabled the leveraging of rich data sets of patient information. It started with the adoption of EHRs and EMRs—patient records that were seldom larger than a few megabytes of text and personal health information—and it continued with the need to incorporate things like images from X-rays or CT scans, which increased the need for, and potential applications of, granular patient information.

Today, the average patient generates close to 80 megabytes of data each year, including clinical and financial information, according to a May 2017 article by the New England Journal of Medicine. The next evolution of this technology will include a massive expansion in what can be stored and, in turn, how patient data can be leveraged.

How healthcare is expanding big data solutions

The amount of viable and useful data that can be stored and utilized for an individual patient has gone from megabytes into the realm of terabytes. From a practical standpoint, this has created new and unexpected challenges for clinical data storage—challenges being met by new innovations in cloud-based and on-site data storage solutions.

For example, cloud-based data solutions allow for reduced and flexible infrastructure costs, increased speed and agility, and increased scalability and availability of data. Many cloud providers also offer infrastructure that is compliant with various regulations and certifications, such as HIPAA and HITRUST.

With these new challenges, however, has also come enormous opportunity. The benefits to patients and consumers from this collection of data has opened the door to avenues of care never thought possible.

Some of the technology solutions gained from this increase in data capacity include:

– The development and empowerment of outpatient monitoring through in-home care and off-site patient communication.

– Real-time patient-to-provider updates and communication strengthened by the speed and security of digitized health records over traditional paper records.

– The incredible developments in machine learning as a result of the utilization of anonymized patient images and video data.

Population health management that can aggregate vast data sets to identify at-risk patients for proactive intervention.

– The potential for precision medicine enabled by genomic mapping of patient information.

This is just a handful of technology utilizing patient data that is already being used by healthcare organizations globally. Some of these technologies are in their third or fourth generation, others are only just emerging, and their ongoing development is yielding promising new possibilities.

Just as mobile phones evolved into today’s smartphones, both old and emerging technology continues to get better, more precise and more meaningful. In the early 1990s, as the data boom was only beginning, few could have imagined what would be possible today. Likewise, even with a far greater understanding of the reach and capabilities of big data, few today can likely imagine what will be possible tomorrow.

Evan Sweeney


Using data analytics layered on top of the EHR, the Mayo Clinic has turned the firehose of patient data into more of a trickle.

The renowned hospital system headquartered in Rochester, Minnesota, has used that approach to filter tens of thousands of data points down to 60 pieces of critical information that are displayed for ICU physicians in a visually appealing format. Using “ambient-intelligence” applications licensed by system’s venture capital arm, the approach gives physicians an extra hour each day that can be utilized for bedside care, three Mayo Clinic physicians wrote in Harvard Business Review.

“In fast-paced critical care units, where even small errors can have big consequences, this digital team member can overload physicians with information,” the authors wrote. “The sheer volume of data in EHRs creates a staggering challenge in complex environments such as intensive care units (ICUs) and emergency medicine departments.”

The initiative began with a series of interviews with 1,500 clinicians over a two-year period to understand which data points flowing through the EHR were particularly impactful.

Using that information, researchers built an EHR interface for ICU physicians that could filter out the unnecessary information and integrated a color-coordinated dashboard that emphasized important data points along with customized alerts.

Subsequent research shows mortality rates were cut in half among ICU patients treated after the implementation of the system. ICU stays also decreased by 50%.

The ICU has been a hotspot for data analytics as hospitals look to refine the data put in front of patients. For example, Dignity Health has developed clinical decision support software to send targeted alerts that could have sepsis, and Sutter Health has used AI to improve prescribing practices.

Meanwhile, the Food and Drug Administration is grappling with how to regulate clinical decision support software, leaving some organizations wondering how the agency will oversee ICU dashboards.

Sheri Stoltenberg


Five must-watch initiatives indicate a growing emphasis on patient-focused health IT innovation.

Now that providers have near universally adopted EHR systems across the country, what is the next frontier for healthcare technology? Based on some recently launched tools, health IT researchers and vendors are focused on health-data monitoring, health data access, and clinical decision-making as the industry catches up with consumerism. As such, these five initiatives serve as must-watch health IT developments, starting with Apple’s iPhone patient health information access.

Apple mHealth integration

2017 saw 85.5 million US iPhone users and likely many more since the release of the iPhone 8 and X. Thanks to the recent operating system update this year, iPhone users can now access their EHR patient portal data through the Health app from 12 health systems. The iPhone health data access is read-only at this point and limited to allergy, clinical vitals, health condition, immunization, lab result, medication and procedural information. Users access their data through individual portal credentials to eliminate patient matching concerns that plague multi-facility organizations. The initiative shows major progress in the battle for interoperability with the hope that more EHR vendors can partner with the tech giant.

Smart thermometer app for influenza

In another mobile health development, a smart thermometer application advances flu monitoring and prediction capabilities. A recent Center for Disease Control and Prevention (CDC) study revealed that among US influenza deaths from 2010 to 2016, nearly two-thirds of child fatalities occurred within seven days of developing symptoms. With the World Health Organization estimating three to five million cases of the flu annually, real-time tracking is in high demand for more proactive care. Utilizing a smart thermometer-connected app, researchers at the University of Iowa can now predict influenza spread up to two or three weeks in advance while tracking virus activity at both the population health and individual levels. This combats the several-week delay of CDC formal reporting with more efficient and rapid surveillance for households, so they can better anticipate and identify symptoms to initiate necessary treatment. 


Genomic EHR data and nursing

The next innovation dives deeper than disease monitoring to better understand and interpret an individual’s genetic health. A patient’s genomic information for diagnostic and therapeutic decision making has yet to be seamlessly integrated into today’s EHRs. The National Institute of Health (NIH) is actively working to push genomic nursing forward through its early adopter program utilizing Allscripts Sunrise and the 2bPrecise genomics and precision medicine solution. Moving beyond basic family history information, NIH funnels genetic testing results to the point of care with EHR genomic pedigree documentation. The NIH program aims to be a model for providers and vendors moving forward in using genetic information to take predictive analytics and preventive care to the next level.

Breast cancer decision-making tool

In another initiative toward informed patient care options, a Journal of Clinical Oncology study from the University of Michigan proves that the interactive iCanDecide breast cancer tool improves high-quality decision making for surgical treatment. iCanDecide focuses on knowledge building and value clarification with patients. The study consisted of 537 early-stage breast cancer patients across 22 surgical practices. The follow-up aim of the research is to integrate patient-facing decision tools into the clinical workflow to improve informed decision making that aligns with patient values.

Center for Connected Care telehealth hub

Lastly, as telemedicine programs proliferate across the country, Penn Medicine has created the Center for Connected Care to stand as one of the largest telehealth hubs. With 50 full-time staff, the center provides around-the-clock care to the health system’s patients as well as support to clinicians throughout Pennsylvania, New Jersey, and Delaware. Telehealth specialties include urgent care, chronically ill homecare and critically ill pregnancy services. In addition, Penn Medicine’s teletrauma program links specialty providers for immediate collaboration and decision-making in emergency situations when a patient cannot be moved. The expansive connected care initiative shows how telehealth will continue to help providers overcome patient travel obstacles and clinician network limitations.

These five mHealth, telehealth, genomic medicine and decision-making tool advancements serve as a sampling of the current industry progress beyond initial EHR implementation. They are also clear examples of how healthcare organizations and technology companies are recognizing the industry shift toward consumerism and active patient engagement in care decisions. While they are encouraging first steps, expanded interoperability and integration are necessary to fully enable informed value-based patient care and diagnosis for wider patient communities across the country.