Using Information Technology To Improve COVID-19 Care At New York City Health + Hospitals

At the epicenter of New York City’s coronavirus disease 2019 (COVID-19) pandemic, New York City Health + Hospitals, the nation’s largest municipal health care system, found itself forced to rely on information technology to expedite and expand its volume of work in caring for large numbers of socioeconomically marginalized New Yorkers, who were disproportionately affected by COVID-19.¹ Just two years before COVID-19, NYC Health + Hospitals—a system serving more than 1.1 million patients annually through eleven acute care hospitals, a long-term care facility, a certified home care agency, and more than seventy ambulatory clinics—had operated on multiple different medical record, scheduling, financial, and data storage platforms, impeding maximal coordination, efficiency, and rapid-cycle learning. On March 1, 2020, the same day New York City reported its first case of COVID-19, the system turned on its unified electronic medical record (EMR) at its last remaining inpatient facility. The ways in which NYC Health + Hospitals used its newly inaugurated EMR to direct and facilitate pandemic response were indispensable in supporting tens of thousands of New Yorkers over the peak period of crisis, from March to May 2020.

As NYC Health + Hospitals braced for COVID-19’s impact as it rapidly spread across Europe and Asia, the system was armed with only a few anecdotes and minimal data from other locales on how to coordinate its response, particularly with regard to harnessing medical informatics. Given how rapidly COVID-19 clinical and epidemiological knowledge was evolving, it was clear that a clinically informed approach to informatics was necessary to help us adapt to rapidly changing needs on the front line. Accordingly, NYC Health + Hospitals integrated its informatics leadership and staff and established regular communications with system clinical response teams in emergency medicine, intensive care, internal medicine, and ambulatory care to best identify and respond to system needs in real time. Concurrently, clinical leadership recognized the value of informatics to help standardize practice across numerous front-line providers, including many volunteers and temporary staff who arrived as part of the pandemic response. The COVID-19 crisis served as a key catalyst for rapidly advancing health information technology adoption.

Overall, the NYC Health + Hospitals priorities for advancing its information technology system fell into five major categories: improving staff efficiency, standardizing clinical workup, informatics, improving the patient experience, and bridging information systems. We describe our approach and experience for each here.

Improving Staff Efficiency

As the number of patients with COVID-19 in emergency department (ED) and ambulatory care settings at NYC Health + Hospitals accelerated in the first days of the pandemic, the system deployed informatics tools to help rapidly evaluate and discharge low-acuity patients to free up resources for the moderately and severely ill. Note templates specific for COVID-19-related clinical presentations helped providers quickly capture essential clinical data while reducing the documentation burden by simplifying and automating as much of the note as possible. A semiautomated medical screening exam note was created by our Information Technology team for use by providers in triage to allow them to quickly evaluate low-acuity patients. The patient’s chief complaint from triage, vital signs, medication history, and problem list were all automatically populated into the note. A limited physical exam that did not require physical contact was presented as well, and the SmartNotes allowed the provider to select common symptoms such as cough, fever, and sore throat with a single click. The system generated patient education resources in the patient’s preferred language. These were manually updated frequently by system leadership to ensure that information remained accurate as institutional, local, and state responses evolved. In addition, in collaboration with our palliative care leadership, standardized language was agreed on to document medical futility in severe cases that were unlikely to improve with critical care. When the provider typed in “.dnrdni”, the “DotPhrase” of the agreed-on language would be inserted into the note. This both eased and improved documentation.

NYC Health + Hospitals also rapidly purchased more than seven hundred vital-sign monitors that were fully integrated into our EMR. Instead of having to do walking rounds to collect vital signs on patients, nursing staff could monitor up to thirty-six patients at a time at a central station as these data flowed directly into the EMR, freeing nursing staff from the need to manually transcribe vital signs. Staff were also given the ability to remotely call a video phone in a patient’s room to interact with the patient without exposing themselves unnecessarily, which also conserved personal protective equipment. This change was implemented quickly because leadership at the system and facility levels made clear early in the crisis that expenditures that would improve or extend our ability to provide care would be supported.

Standardizing Clinical Workup

If patients were presenting with severe enough symptoms to warrant further clinical testing, COVID-19 order sets for ED, medicine, and intensive care unit (ICU) were established within the EMR to help guide providers with targeted ordering. These order sets included preselected lab tests, radiology studies, and nursing orders, as well as other common orders such as oxygen delivery, that providers could include based on the needs of individual patients. These order sets were standardized across the system with multispecialty input to ensure that they were in line with national best practices and updated at the speed necessary to match this rapidly changing crisis.

Getting evidence-based guidelines to the front lines of care is challenging in the best of times. During this crisis, the order sets we established presented a vetted, default package of interventions to staff. As the evidence evolved, so did the order sets. They included biological markers of systemic inflammation useful in the management of COVID-19, such as D-dimer, ferritin, C-reactive protein, and lactate dehydrogenase, and patient positioning such as proning (lying on one’s stomach) was codified.

Informatics

Given the size and breadth of NYC Health + Hospitals, using the principles of information science to acquire and analyze heath data has been essential. Dashboards reflecting ED patient volume, ICU and medical floor availability, and staffing were rapidly created by Information Technology and our Office of Population Health as the urgency of the COVID-19 crisis overcame institutional inertia. These dashboards automatically pulled data from the integrated EMR and other data resources to present a unified single source of information. We matched capacity to bed demand across the entire enterprise, rather than at the individual hospital level. We transferred patients daily out of our hardest-hit hospitals to our other facilities. When the patients arrived at a new facility, their labs, radiology, notes, and vitals from the previous facility were all accessible. There was no need to sift through paper records. These measures helped with “level-loading” patients and staff members more equally throughout the system, alleviating patient surges and provider shortages at the most affected facilities.

A shared EMR informed by clinically active leadership made this work possible. The dashboards were not adapted to meet clinical needs, they were built in collaboration between data scientists and clinicians who also served within operations. In the crisis, enterprise definitions of what constituted ICU space—medical floor space capable of caring for critically ill patients with COVID-19—were agreed upon and historical local definitions set aside. The dashboard’s contents were reviewed daily on a morning call that included the facility chief medical officers, chief executive officers, and system leadership. This allowed for data-driven decision making and the rapid correction of inaccuracies.

At our peak we had more than twenty-five hundred patients with lab-confirmed active COVID-19 infection admitted throughout our system. This figure is an underestimate of the true volume, given lack of test sensitivity and certain populations such as “comfort measures only” patients who were not tested. We aggressively expanded bed space, increasing our medical floor space able to care for patients with COVID-19 by 50 percent (from two thousand to three thousand beds). An even greater demand was placed on our ICUs, where we increased capacity by 400 percent (from three hundred to fifteen hundred patients).

Facilities converted post-anesthesia care units and operating room space into new ICUs, and medical floor space was adjusted to allow them to accept ICU-level patients. Rehab space and pediatric floors were turned into adult medical floor units. The EMR was updated so that intensivists seeing an ICU patient in a post-anesthesia care unit or an adult hospitalist caring for a patient on a former pediatric unit would see the layout to which they were accustomed and not require retraining. This added capacity was reflected in the enterprisewide dashboards. The integrated EMR system enhanced our ability to standardize, share, and use information much more quickly than we could have under a system of paper records or an electronic system that was not fully standardized, allowing us to respond in real time to rapidly shifting demands.

To manage the increased volume of patients spread across new areas, we developed specialty-specific summary reports to increase efficiency. An ICU patient summary tailored to patients with COVID-19 allowed our intensivist to view and trend over time all of the important clinical variables, such as vital signs, recent lab tests, ventilator settings, and medications, on a single screen, and an infectious disease–focused summary tracked all suspected and confirmed COVID-19 cases within an entire hospital.

Improving The Patient Experience

Technology also helped NYC Health + Hospitals enhance the patient experience for patients in both hospital and outpatient settings. In an effort to limit the spread of COVID-19, NYC Health + Hospitals—similar to many other health systems—limited visitors in ED and inpatient areas. To bridge the social isolation created by this decision, including for some patients in their final moments of life, we configured purchased and donated tablet devices with video applications to connect patients with their loved ones. Tablets on rolling stands functioned across multiple rooms, and others were given to patients for the duration of their hospital stay. The meaningfulness and restoration achieved by connecting patients with their families also helped ease psychological distress among the clinical teams involved in their care.

Bridging Information Systems

In responding to the COVID-19 crisis, NYC Health + Hospitals helped lead broader, citywide efforts to transform the sharing of clinical information across institutions to enhance the quality of care. At the start of the crisis, a group of hospitals that cared for the majority of patients in the New York metropolitan area agreed to remove barriers to health information exchange. Before the pandemic, technological and institutional hurdles precluded providers in any one hospital system from accessing their patients’ records from other hospital systems. This frustrated patients, occasionally delayed care, and at times resulted in unnecessary testing or missed information. Leadership from across hospital systems citywide collectively agreed to remove barriers to this type of clinical information sharing. This process was made easier because most major NYC hospitals used Epic as their EMR. The technical barriers to information exchange had been overcome before the crisis, but the obvious patient benefit during the crisis allowed a reevaluation of the policy obstacles. Sites that did not use Epic also benefited from this process with the adoption of Carequality, a national framework allowing information exchange agnostic to EMR vendor. This option took a slightly more technical setup but had the same ultimate benefit. Beginning with the pandemic, providers were able to view patients’ clinical records seamlessly across most major health care institutions in the New York metropolitan area in a way that complies with appropriate regulations. Health Insurance Portability and Accountability Act (HIPAA) of 1996 and state regulations had long allowed information to be shared between providers during a patient visit, but out of an abundance of caution, written patient consent had historically been obtained at the point of care. This was a burden before the crisis but became untenable given the added volume of patients and potential for disease transmission from this physical process. Instead, patients are now informed of information exchange through the registration process and given the opportunity to opt out. As a result, the number of records received by NYC Health + Hospitals from other regional institutions almost tripled, going from 27,155 records the week of March 1 to 71,471 the week of May 3, aiding thousands of critical decisions.

In an example of citywide information exchange, one of our authors recently cared for a critically ill and nonverbal patient from a nursing home with a presentation suggestive of COVID-19 who had never previously received care at NYC Health + Hospitals and for whom there were no data in the NYC Health + Hospitals EMR. Data from another NYC health system, however, were available and provided the patient’s long list of comorbidities, including dementia, that had robbed her of her ability to communicate. There was also documentation making clear the patient’s goals of care in such a circumstance. Appropriately, when the patient died a few hours later, efforts were focused on her comfort instead of invasive procedures. The patient received the care she sought, staff were relieved of the psychological toll that accompanies declaring medical futility, and the family was able to speak with a physician who knew their loved one’s full medical history.

Conclusion

The ability of NYC Health + Hospitals to use its information technology effectively improved patient care delivery and laid the groundwork for a more optimized system response during a time of crisis. Speed in decision making was essential and required embedded communication with front-line staff involved in direct patient care. A number of barriers had to be overcome, but data were used effectively to address population and individual patient needs. Collectively, these innovations not only helped NYC Health + Hospitals weather one of the greatest health crises in its history but also positioned it organizationally to be able to use enterprise information technology in new and innovative ways in the period after the pandemic.

ACKNOWLEDGMENTS

An unedited version of this article was published online July 16, 2020, as a Fast Track Ahead Of Print article. That version is available in the online appendix. To access the appendix, click on the Details tab of the article online.

NOTE

• 1 Wadhera RK, Wadhera P, Gaba P, Figueroa JF, Joynt Maddox KE, Yeh RWet al. Variation in COVID-19 hospitalizations and deaths across New York City boroughs. JAMA. 2020;323(21):2192–5. Crossref, Medline, Google Scholar