Nine Lessons Learned From the COVID-19 Pandemic for Improving Hospital Care and Health Care Delivery

New York City (NYC) was the epicenter of the COVID-19 pandemic in the US in March 2020. A dense city of only 302 square miles, it has had 33 359 deaths and 109 192 hospitalizations due to COVID-19 as of June 15, 2021.¹ In guiding NYC Health + Hospitals, the largest municipal hospital system in the US, through the pandemic, we have learned a number of lessons. Although there is much to debate about the national public health response to COVID-19,² we focus on the lessons learned through COVID-19 that we believe have applicability for improving hospital care in the future. The growth of telehealth has been covered elsewhere.^3,4 We present 9 other lessons for improving hospital care and health care delivery.

Hospitals operate on tight financial margins with complex processes to match available beds and staff to the patient census. Even relatively small increases in the census, as can occur with a multicasualty accident, an unanticipated natural disaster, or a seasonal influenza outbreak, can result in a short-term crisis until the hospital can expand capacity. What occurred in NYC in March 2020 was just an extreme version of this. During a 6-week period, our 11 hospitals tripled our intensive care unit (ICU)–capable beds.⁵ For our hospitals, identifying appropriate space for the influx of patients was a huge challenge. We had more success outfitting additional areas of the hospital for inpatient care (eg, recovery room, “mothballed units”) than building field hospitals.⁵ The problem with field hospitals was replicating all the equipment and staffing needs of a hospital (eg, pharmacy, blood bank, radiology).

The challenges in identifying additional space paled in comparison to providing sufficient ICU nurses, respiratory therapists, intensivist physicians, and dialysis nurses.⁶ Canceling elective surgeries and discharging stable patients immediately provided a pool for staffing inpatient beds, but many diverted staff were unfamiliar with the areas where they were needed. We developed “just in time” training videos to familiarize clinicians with how to use ventilators and nursing competency modules to prepare nurses to assist in ICUs. Hospitals should have clear policies for providing emergency credentials to physicians not on their medical staff. In the extreme case of COVID-19, New York allowed the credentialing of out-of-state physicians with only proof of identity and a medical license. Under less extreme circumstances, reciprocity agreements between states for medical licenses and national databases with credentialing information for physicians⁷ would enable hospitals to more rapidly deploy physicians. Hospitals should develop relationships with regional centers where specialists, such as ICU intensivists, can use video monitoring to guide generalists in ICU care. This model could also work in routine situations when a patient with complicated care needs presents at a community hospital and is too unstable for transfer.

Going forward, all hospitals should have a detailed disaster plan that includes the following: what areas of the hospital to expand to and in what order (eg, recovery room first, ambulatory areas second), how to increase ability to care for incoming patients (eg, cancel routine surgeries and appointments), and how to gain immediate access to additional staff (eg, reassignment of staff to affected areas with appropriate training). The Table lists elements to include in a hospital disaster plan for dealing with an increased volume of patients or workforce shortages.

In the traditional hospital room, the patient is given a call bell to summon a nurse. During the pandemic, we minimized the number of times nurses entered patients’ rooms, yet still evaluated patients. Strategies included placing windows in walls, replacing wood doors with glass doors, and using communication and video devices in rooms. We intend to continue and expand these efforts. Indeed, video monitoring may prevent falls, a chronic problem in hospitals, by alerting nurses that patients are getting out of bed.⁸

Hospitals are required to have minimum air changes in patient rooms, with more frequent air changes required for isolation rooms. We quickly ran out of isolation rooms in the surge of the pandemic. We improvised using additional methods of decreasing potential transmission of COVID-19, such as placement of high-efficiency particulate air filters and UV lights.⁹ Other promising strategies for decreasing aerosolized transmission of disease are electronic (bipolar ionization) filtration and high fresh air exchange.¹⁰ Given the potential for transmission of airborne agents, it makes sense for hospitals to put more efforts into the quality of the air in the hospital, including in common areas, such as waiting rooms.

Frontline health care workers were labeled heroes for their efforts during the pandemic. Many struggled with the compliment because it implied they possessed superhuman strength. In walking into “hot” zones of sick patients with COVID-19, they experienced the very human emotions of vulnerability, fear of bringing the virus home to their families, and guilt at not being able to save all patients. Although the culture of service is a positive attribute of health care professionals, it can obscure the human needs for support, leading to burnout and misplaced emotions.

We provided stress management and resilience training, recharge rooms, peer support champion (wellness) rounds, and easy-to-access mental health resources.¹¹ We believe that these resources should remain permanently in place.

In January 2020, hearing reports from Wuhan, we isolated patients arriving from China who were presenting with fever and respiratory symptoms. Unfortunately, this instruction was wrong on several points. First, the triage nurses making these judgments were unmasked. Second, genetic mapping of the virus indicates that most of the cases in NYC were transmitted by people traveling from Europe. Third, many patients who did not have respiratory illnesses, including people who had traumatic motor vehicle injuries, turned out to also be infected with COVID-19.

Going forward, we intend to continue to wear masks in the hospital even if SARS-CoV-2 disappears (an unlikely possibility). Data on influenza rates in NYC suggest that the combination of mask wearing and social distancing dramatically decreased the percentage of outpatient visits for influenzalike illness from 5% to 1% at the peak of the season compared with the prior 5-year average.¹² Nationally, population data from 14 states of the US show the lowest cumulative rate of laboratory-confirmed influenza-associated hospitalizations since data collection began in 2005.¹³

Beyond this, the COVID-19 pandemic has challenged the hospital disease control dichotomy that respiratory spread occurs either through respiratory droplets (as with influenza) or aerosolized spread (as with measles). It is likely that SARS-CoV-2 is spread primarily through droplets but can also be airborne, requiring rethinking of respiratory infection control procedures and raising the question of whether it is appropriate to still have multioccupancy rooms in the hospital.

During the peaks of the COVID-19 pandemic, patients were unable to have visitors. To ameliorate this, we provided electronic tablets to allow patients to communicate with their families. An unexpected benefit was that patients used the tablets to communicate with friends and family who were not in the local area. Family conferences included faraway relatives in ways that would not have occurred prior to the pandemic. Tablets enabled us to increase participation because even people in the local area did not have to travel all at the same time to meet with the care team. A video option for including relatives in hospital care should remain postpandemic.

To cut costs, hospitals typically maintain minimal excess supplies. In addition, supplies are typically sourced from a limited number of vendors because consolidation of manufacturing decreases cost, especially in countries with low labor costs.

Unfortunately, these twin policies were a disaster during COVID-19. Hospital systems, including ours, ran out of supplies ranging from medications (eg, succinylcholine for intubation) to major equipment (eg, ventilators). We survived through substituting medications (eg, rocuronium for succinylcholine), placing stable patients on portable ventilators, and developing alternative supply chains.¹⁴ But these issues should not be resolved in crisis. Although individual hospitals cannot afford to maintain large excesses in supplies, regional caches with rotation of expiring medication should be established. Many countries have learned of the harms from no longer manufacturing certain medical equipment themselves. Ultimately, the federal government could subsidize sufficient manufacturing within the country to protect against severe shortages in crises.

The surge of patients with COVID-19 arriving in our emergency departments and ICUs forced multiple electronic health record process improvements that reduced the burden of order entry and documentation. These included order sets based on evolving science, smart phrases that can embed common physical examination findings or treatment plans into a note with a few keystrokes, and abbreviated nursing and physician notes.¹⁵ These electronic health record changes should continue after the pandemic to allow physicians and nurses to spend more time with patients and to reduce burnout.

COVID-19 exacerbated existing health disparities with racial and ethnic minority and low-income communities experiencing disproportionate infections, hospitalizations, and death.^16,17 The increased prevalence of hypertension, diabetes, and obesity among racial and ethnic minorities contributed to their greater morbidity and mortality. These conditions are best treated through primary care. Although the Emergency Medical Treatment and Active Labor Act ensures emergency care without regard to ability to pay, there is no similar program for ensuring primary and preventive care. To be fully accessible, primary care must be geographically near, be culturally competent, provide translation services for persons with limited English proficiency, and support persons with low literacy levels. If people are connected to a health care system, they are more likely to access care for both serious illness and day-to-day care. Increasingly, hospital systems are successfully partnering with community-based agencies and other resources to address the social determinants of health.¹⁸ These efforts should be expanded. In addition, hospital systems should ensure that their organization promotes antiracism, equity, and inclusion in their patient-care and employment policies and practices.

We can honor the people we lost to COVID-19 by learning from the experience in ways that could benefit all patients. We believe that the 9 lessons we have learned from the pandemic would improve hospital care and health care delivery in both day-to-day circumstances and future emergencies.

Corresponding Author: Mitchell H. Katz, MD, New York City Health + Hospitals, 125 Worth St, Room 514, New York, NY 10013 (mitchell.katz@nychhc.org).

Published Online: July 23, 2021. doi:10.1001/jamainternmed.2021.4237

Conflict of Interest Disclosures: None reported.