Secondary Uses of EHR Data: Three Ways to Improve Quality of Care - Cover

Secondary Uses of EHR Data: Three Ways to Improve Quality of Care

EHR software is a regularly talked-about tech in a physician’s tool belt. And with a 67% adoption rate, these systems are steadily making their way in healthcare facilities. However, implementing such software and using it on the primary level (collecting, storing, and sharing PHI to provide medical care) is just a piece of a much bigger puzzle.

To deliver value-based care, you need to pay particular attention to the secondary use of EHR data. This specifically includes unlawful activity detection and population health management (PHM): drug safety surveillance, disease risk prevention, mortality rate reduction, and more.

In this feature, we will delve into three cases of analyzing EHR data and describe the benefits hospitals will reap with its secondary use.

Stemming the tide of opioid abuse

In the United States, opioid misuse is increasing at an alarming rate. About 80% of the global opioid supply is prescribed and used in the USA. In 2012, two million Americans suffered from a substance abuse disorder, and four out of five new heroin users started out misusing prescription painkillers.

Fortunately, even such a desperate situation can be handled by dint of an EHR system. Here are three possible scenarios.

  • Integrating electronic prescribing of controlled substances (EPCS) with EHR software might be a way out. First, such a combination eliminates paperwork and prevents illegitimate opioid prescriptions from being processed. Second, this certified EHR tool helps reduce opioid abuse and discourage “doctor shopping” — a frequent change of physicians to find someone who is willing to prescribe painkillers. Although 88.1% of pharmacies are certified to receive EPCS transactions, only 20.2% of healthcare providers are actively prescribing painkillers with EHR and EPCS.

  • Prescription drug monitoring programs (PDMPs) are state-initiated electronic databases with PHI on using controlled substance prescriptions. And linking this data to medical software can help caregivers avoid inappropriate prescribing of pain medications as well as identify drug-seeking behavior.

  • Implementing a default option for a lower quantity of controlled substances in EHRs will increase compliance with opioid prescribing guidelines for acute pain. And there’s scientific proof: physicians from two Penn Medicine emergency departments prescribed fewer opioid substances when the EHR default was set to ten pills.

Including SDOH in EHRs to improve population health

Providers are increasingly waking up to the importance of social determinants of health (SDOH) and behavioral indices in public health management. The connection is simple: for example, a lack of food or housing can negatively influence patients with chronic diseases.

To address this social issue and have deeper insights into a patient’s conditions, there’s a need for collecting and including SDOH in EHR systems.

That will also smooth the way for creating optimized Medicaid payment models. Namely, the government will provide more incentives for the care of socially vulnerable people. That money can be used to help individuals find housing and better nutrition, as well as to build an IT infrastructure to link at-risk patients with doctors and community-based organizations.

To narrow the focus, the Institute of Medicine (IOM) stepped in with domain recommendations — social determinants and behavioral indices that should be incorporated into EHRs apart from tobacco and alcohol use, race/ethnicity, and residential address that are already being included in EHRs and custom PHM software.

These additional domains comprise education level, financial resource strain, physical activity, stress, social isolation, depression, intimate partner violence, and neighborhood median household income. Some of this data can be patient-generated, i.e. obtained from individuals via mHealth apps, wearable devices, SDOH documentation flowsheets, or electronic tablets in clinic waiting rooms.

A case in point: Partners HealthCare started to better manage chronic cardiac and weight-related conditions by allowing patients to upload PHI from their devices directly into EHRs.

According to a paper published in the Journal of the American Board of Family Medicine, incorporating IOM’s domains in EHRs can yield an array of benefits: better patient engagement, accelerated decision-making, enhanced care coordination with community resources, and faster referrals. Among other benefits are lower costs, reduced hospital admissions, and fewer health inequities.

Going mainstream with preventive medicine

Another application of EHR systems — beyond the primary use — comes in the form of preventive medicine. That means using built-in or third-party analytics to analyze mounds of data (EHR, historical, internet search, and patient-generated information) and forecast diseases. Here are some examples.


With a 40% mortality rate and $12.5 billion in care costs, sepsis is difficult to detect and distinguish from other conditions in its early stages. In turn, priceless time is lost before physicians can start immediate treatment.

Researchers from the University of California Davis are in the right direction to find a cure. They’re working on a sepsis detection algorithm that can use EHR data — blood pressure, respiratory rate, temperature, and white blood cell count — to give an early warning of sepsis.

Heart disease

Analysts from Kaiser Permanente’s Care Management Institute found out that implementing the IndiGO score (individualized clinical guidelines based on the Archimedes model of validated biomathematical simulations) might improve medical care in a meaningful fashion.

The IndiGO score uses information on human physiology, diseases, behavior, interventions, and national healthcare systems to forecast cardiovascular events.


A research published in BioMedCentral reveals that by sifting PHI (diagnoses, clinical histories, pharmacy data, and laboratory results) for signs of diabetes, EHR algorithms can detect more than 90% of sufferers and predict the date of the disease with 78.4% accuracy. Moreover, these mechanisms prevent a delayed diagnosis in 11% of cases, allowing clinicians to prescribe treatment earlier than ever before.

Kidney damage

An AI startup Medial EarlySign is also contributing to the development of preventive medicine.

Their AI-powered model combs through large amounts of EHR data — including demographics, laboratory test results, medications, and others — to spot diabetic patients who might be at risk of developing kidney disease within a year. And such information is becoming extremely valuable for facilitating early treatment of this illness.

Summing up the benefits

We’ve detailed just three ways you can make the most out of your health tech. Providers of healthcare software development services may raise the ante by offering other secondary use cases. And the outcomes you’ll reach with such advanced solutions will be more than tangible:


  • Improved patient outcomes
  • Refined decision-making
  • Revamped PHM
  • Fewer health inequities
  • Better chronic disease management
  • Reduced readmission rates
  • Minimized operational costs.


About the Author:

Yana Yelina is a Technology Writer at Oxagile, a New York-based software company that develops custom medical solutions. Her articles have been featured on Becker’s Hospital Review, Medical News, Datafloq, Healthcare Works Collective, Medgadget, to name a few. You can reach Yana at or connect via LinkedIn or Twitter.