Identifying COVID-19 Vaccine Analytics to Prevent Hospitalizations, ED Visits

The full emergence of seasonal respiratory season is ongoing in the United States right now with spikes in incidence rates of influenza and COVID-19 spreading throughout the country.

Separately, but related to incidence rates, is the concept of data and analytics as they continue to be an evolving part of medicine and public health and how to address COVID-19. One of the specific areas being explored is vaccination statistics and how they can prevent the spread of severe disease and hospitalizations. A large, comprehensive study measured how widespread COVID-19 vaccination needed to be in order to provide protection against serious disease in adults by age groups. The study was published last year in The Lancet Regional Health - Americas.

From the study, the number needed to vaccinate (NNV) is a significant analytic they calculated.

“What that really represents is the number of people to be vaccinated that would reduce the number of hospitalizations or emergency department visits by 1,” explained study coauthor Shaun Grannis, MD, vice president, Data and Analytics, Regenstrief Institute. “So if in an unvaccinated population of 10 people, if there were typically 5 hospitalizations, then we look at that same population of 10 that were vaccinated and that the number of hospitalizations then decreases by 1 to 3. So, if we see a decrease among a given population, we're looking for that number that we need to [vaccinate]—that minimum number we need to treat to decrease the adverse outcome by 1. So we looked at number needed to [vaccinate] for hospitalizations. And we also looked at the number needed to vaccinate for emergency department (ED) visits as well.”

The Study Specifics

An analysis of real-world data from over 1.2 million patients from health systems in 4 geographically dispersed states—Indiana, Oregon, Texas and Utah— was conducted by the CDC's VISION Network. The study determined both the aforementioned NNV to prevent 1 COVID-19 associated hospitalization and the number needed to be vaccinated to prevent 1 COVID-19 associated emergency department visit.

Of the 1,285,032 patients included in the study, there were 938 hospitalizations and 2076 emergency department encounters. Looking at age groups, 555,729 (43.2%) patients were aged 18–49 years, 363,299 (28.3%) 50–64 years, and 366,004 (28.5%) ≥65 years.

Additionally, a majority of the people were female (n = 765,728, 59.6%), White (n = 990,224, 77.1%), and non-Hispanic (n = 1,063,964, 82.8%). In terms of vaccination, 37.2% of patients received a booster and 62.8% received only two doses.

“The median number needed to vaccinate to prevent 1 hospitalization was 205 people,” Grannis said. “So if you vaccinate 205 people, you'll prevent at least 1 hospitalization. The number needed to vaccinate [to prevent] 1 emergency department visit was 156.”

And in terms of understanding how this pertains to public health, Grannis says this data provides information to various people including decision makers who determine the number of vaccines needed to clinicians having one-on-one patient encounters and counseling them on the benefits of vaccination.

“This is a concrete number that helps providers, helps public health officials, and helps decision makers understand the effectiveness and the volume of vaccines required to have a particular impact,” Grannis said. “So for example, as a physician, I like to understand what the effectiveness of a particular intervention is. And so I like to be able to tell and communicate with my patients that by receiving this vaccine, you are reducing your risk for adverse outcomes. In fact, you're reducing your risk for hospitalization.”

Contagion spoke to Grannis about the significance of using NNV to extrapolate data needed in real-world settings and determining how they play a role in reducing hospitalizations and ED visits as well as other ancillary findings related to the study.


Reference
Adams K, Riddles JJ, Rowley EAK, et al. Number needed to vaccinate with a COVID-19 booster to prevent a COVID-19-associated hospitalization during SARS-CoV-2 Omicron BA.1 variant predominance, December 2021-February 2022, VISION Network: a retrospective cohort study. Lancet Reg Health Am. 2023;23:100530. doi:10.1016/j.lana.2023.100530