COVID-19 Mortality Rates Declined, but Vary by Hospital

Jake Remaly

December 23, 2020

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

Mortality rates for inpatients with COVID-19 dropped significantly during the first 6 months of the pandemic, but outcomes depend on the hospital where patients receive care, new data show.

Dr David Asch

"[T]he characteristic that is most associated with poor or worsening hospital outcomes is high or increasing community case rates," write David A. Asch, MD, MBA, executive director of the Center for Health Care Innovation at the University of Pennsylvania in Philadelphia, and colleagues.

The relationship between COVID-19 mortality rates and local disease prevalence suggests that "hospitals do worse when they are burdened with cases and is consistent with imperatives to flatten the curve," the authors continue. "As case rates of COVID-19 increase across the nation, hospital mortality outcomes may worsen."

The researchers published their study online December 22 in JAMA Internal Medicine.

The quick and substantial improvement in survival "is a tribute in part to new science — for example, the science that revealed the benefits of dexamethasone," Asch told Medscape Medical News. "But it's also a tribute to the doctors and nurses in the hospitals who developed experience. It's a cliché to refer to them as heroes, but that is what they are. The science and the heroic experience continues on, and so I'm optimistic that we'll see even more improvement over time."

However, the data also indicate that "with lots of disease in the community, hospitals may have a harder time keeping patients alive," Asch said.  "And of course the reason this is bad news is that community level case rates are rising all over, and in some cases at rapid rates. With that rise, we might be giving back some of our past gains in survival — just as the vaccine is beginning to be distributed."

Examining Mortality Trends

The researchers analyzed administrative claims data from a large national health insurer. They included data from 38,517 adults who were admitted with COVID-19 to 955 US hospitals between January 1 and June 30 of this year. The investigators estimated hospitals' risk-standardized rate of 30-day in-hospital mortality or referral to hospice, adjusted for patient-level characteristics.

Overall, 3179 patients (8.25%) died, and 1433 patients (3.7%) were referred to hospice. Risk-standardized mortality or hospice referral rates for individual hospitals ranged from 5.7% to 24.7%. The average rate was 9.1% in the best-performing quintile, compared with 15.7% in the worst-performing quintile.

In a subset of 398 hospitals that had at least 10 patients admitted for COVID-19 during early (January 1 through April 30) and later periods (between May 1 and June 30), rates in all but one hospital improved, and 94% improved by at least 25%. The average risk-standardized event rate declined from 16.6% to 9.3%.

"That rate of relative improvement is striking and encouraging, but perhaps not surprising," Asch and coauthors write. "Early efforts at treating patients with COVID-19 were based on experience with previously known causes of severe respiratory illness. Later efforts could draw on experiences specific to SARS-CoV-2 infection."

For instance, doctors tried different inpatient management approaches, such as early vs late assisted ventilation, differences in oxygen flow, prone or supine positioning, and anticoagulation. "Those efforts varied in how systematically they were evaluated, but our results suggest that valuable experience was gained," the authors note.

In addition, variation between hospitals could reflect differences in quality or different admission thresholds, they continue.

The study provides "a reason for optimism that our healthcare system has improved in our ability to care for persons with COVID-19," write Leon Boudourakis, MD, MHS, and Amit Uppal, MD, in a related commentary. Boudourakis and Uppal are both affiliated with NYC Health + Hospitals in New York City and with SUNY Downstate and New York University School of Medicine, respectively.

Similar improvements in mortality rates have been reported in the United Kingdom and in a New York City hospital system, the editorialists note. The lower mortality rates may represent clinical, healthcare system, and epidemiologic trends.

"Since the first wave of serious COVID-19 cases, physicians have learned a great deal about the best ways to treat this serious infection," they say. "Steroids may decrease mortality in patients with respiratory failure. Remdesivir may shorten hospitalizations of patients with serious illness. Anticoagulation and prone positioning may help certain patients. Using noninvasive ventilation and high-flow oxygen therapy may spare subsets of patients from the harms of intubation, such as ventilator-induced lung injury."

Overwhelmed Hospitals

"Hospitals do not perform as well when they are overwhelmed," which may be a reason for the correlation between community prevalence and mortality rates, Boudourakis and Uppal suggested. "In particular, patients with a precarious respiratory status require expert, meticulous therapy to avoid intubation; those who undergo intubation or have kidney failure require nuanced and timely expert care with ventilatory adjustments and kidney replacement therapy, which are difficult to perform optimally when hospital capacity is strained."

Although the death rate has fallen to about 9% for hospitalized patients, "9% is still high," Asch said.

"Our results show that hospitals can't do it on their own," Asch said. "They need all of us to keep the community spread of the disease down. The right answer now is the right answer since the beginning of the pandemic: Keep your distance, wash your hands, and wear a mask."

Asch, Boudourakis, and Uppal have disclosed no relevant financial relationships. A study coauthor reported personal fees and grants from pharmaceutical companies outside the submitted work.

JAMA Intern Med. Published online December 22, 2020. Full text, Editorial

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: