An influenza season is defined by activity, or the percentage of specimens that test positive for the influenza virus, typically beginning in September, peaking around February, and ceasing by May of the following year in the northern hemisphere.1,2 We sought to update our previous analysis on the 2020-21 influenza season to explore influenza epidemiology during the ongoing COVID-19 pandemic.3
We analyzed the Milliman MedInsight® Emerging Experience research data set, studying September through May of the most recent three (2018-19, 2019-20, and 2020-21) influenza seasons. We confirmed our prior findings for the first three months of each season that influenza-related utilization rates, measured as office and emergency department visits and hospital inpatient admissions with a diagnosis code for influenza, were 96% lower for 2020-21 as compared to 2019-20 and 95% lower for 2020-21 as compared to 2018-19, regardless of payer type (see Figure 1).3
Figure 1: Influenza-related utilization (Office visit, emergency department visit, and hospital inpatient admission) per 1,000 members overall and by payer type for September through May of the most recent three influenza seasons
See Methods section for additional detail.
As reports from the U.S. Centers for Disease Control and Prevention (CDC) noted increased activity among those of younger age groups during what would have been the final months of the 2020-21 influenza season, we also explored influenza-related utilization by age group during the COVID-19 pandemic (April 2020 through June 2021).4 Similar to CDC, we observed a divergent increase in utilization for the youngest age group (0 to 4 years) beginning in March 2021 (see Figure 2).
Figure 2: Influenza-related utilization (Office visit, emergency department visit, and hospital inpatient admission) per 1,000 members overall and by age group for April 2020 through June 2021
See Methods section for additional detail.
Implications
An increase in influenza-related utilization among the youngest age group from March through June 2021 may indicate new evidence of shifting patterns in influenza virulence during the COVID-19 pandemic. Just as factors influenced by the COVID-19 pandemic likely disrupted the most recent influenza season, modification to these factors, such as relaxed transmission-reduction measures, may have allowed influenza activity to return to pre-COVID-19 pandemic levels at and beyond the typical end of an influenza season during 2020-21 for this age group.
Although we observed elevated influenza testing utilization for June 2021 as compared to June 2019, we are unable to discern from our analysis whether increased influenza burden drove the higher rates of testing (see Figure 3). Of note, panel tests, which included influenza and COVID-19, accounted for one-third of the testing overall in 2021 (20% for the youngest age group, 35% to 43% for all others; data not shown). Instead, the increase in influenza testing utilization may be indicative of increased surveillance of acute respiratory symptoms more generally, particularly among an age group for which adherence to transmission-reducing measures can be most burdensome and therefore challenging to maintain, and of a return to typical patterns in healthcare-seeking behavior.
Figure 3: Influenza tests per 1,000 members by payer type for March through June of 2019 and 2021 by age group
| 2019 | 2021 | |||||||
|---|---|---|---|---|---|---|---|---|
| AGE GROUP | MAR | APR | MAY | JUN | MAR | APR | MAY | JUN |
| Overall | 11.3 | 5.1 | 2.3 | 1.0 | 2.2 | 2.3 | 2.0 | 1.6 |
| 0–4 years | 37.6 | 19.7 | 8.9 | 4.4 | 6.7 | 7.4 | 8.1 | 7.8 |
| 5–17 years | 19.4 | 7.2 | 3.1 | 1.0 | 2.3 | 2.6 | 2.4 | 1.7 |
| 18–49 years | 9.4 | 4.4 | 1.9 | 0.9 | 1.8 | 2.3 | 1.8 | 1.4 |
| 50–64 years | 7.0 | 3.3 | 1.4 | 0.7 | 1.2 | 1.6 | 1.2 | 1.0 |
| 65 years and older | 5.8 | 2.9 | 1.3 | 0.6 | 2.2 | 1.0 | 0.9 | 0.7 |
| See Methods section for additional detail. | ||||||||
Historically, influenza virulence is lowest during the summer months (June through August in the northern hemisphere and December through March in the southern hemisphere). Though often categorized as temperature-sensitive given these patterns, influenza can circulate throughout the year in tropical and equatorial regions.1 While this may change over time such that influenza activity is detected regardless of month for non-tropical/equatorial regions, it is too early to know how evolutionary pressure, such as from COVID-19, has or will alter influenza virulence.
Younger age (under 4 years) and older age (over 64 years) are among the risk factors for complications due to infection from influenza. Fortunately, CDC reported only one pediatric death in the United States during the 2020-21 influenza season.4 As infection rates increase in the future, however, this number is likely to rise. Vaccination against influenza remains the most effective tool in reducing this morbidity and mortality associated with complications from infection.5,6 We found slight reductions in vaccination coverage for the period studied: vaccination rates for the 2020-21 influenza season were, on average, 18% lower than for 2019-20 and 21% lower than for 2018-19 (see Figure 4). The reductions occurred most for those age 50 years and older and least for those age 5 to 17 years.
Figure 4: Influenza tests per 1,000 members (%) by age group for September through May of the most recent three influenza seasons
| INFLUENZA SEASON | |||
|---|---|---|---|
| AGE GROUP | 2018-19 | 2019-20 | 2020-21 |
| Overall | 23.5% | 22.5% | 18.5% |
| 0–4 years | 44.1% | 44.8% | 37.9% |
| 5–17 years | 20.1% | 20.7% | 16.1% |
| 18–49 years | 15.7% | 16.6% | 15.2% |
| 50–64 years | 8.8% | 8.4% | 6.5% |
| 65 years and older | 48.6% | 44.3% | 38.0% |
| See Methods section for additional detail. | |||
National (CDC) and global (World Health Organization [WHO]) surveillance programs provide timely information on virulence, morbidity, and mortality arising from infection due to influenza.1,2 Healthcare claims can complement this information, including vaccination status, and allow for additional insight on specific population-based burden, such as avoidable healthcare use and expenditures.
Limitations
We analyzed data through June 2021, the most recent available. Claims from January 2021 on were adjusted using 2019 adjudication patterns by month, utilization category, and payer type. As noted, influenza utilization rates decreased significantly since the start of the COVID-19 pandemic as compared to prior. Further analysis may be required to evaluate the clinical significance of the differences by age group we observed as well as variation over time as the pandemic continues.
The data set analyzed represents a convenience sample of healthcare administrative claims data for more than 30 million covered lives occurring during the most recent three influenza seasons for insured individuals from all 50 U.S. states and may not be generalizable to all individuals with similar health insurance coverage nationally. Claims data can only detect encounters with healthcare professionals and therefore excludes influenza or influenza-like symptoms for which no services were sought. It is also possible that the payer and geographic distributions changed year to year. Finally, we did not measure influenza illness or vaccination for individuals who are uninsured.
Methods
The study population included individuals continuously enrolled for the six months prior to September 1 of each influenza season. Age was determined as of September 1 of each influenza season. Payer types were categorized as commercial—health maintenance organization (HMO), preferred provider organization (PPO), Patient Protection and Affordable Care Act (ACA), and other—with upwards of 2 million enrollees; Medicaid (HMO, PPO, other) with more than 1 million enrollees; Dual (Medicaid-Medicare dual eligibility) with 80,000 enrollees; Medicare Advantage with more than 300,000 enrollees; and Medicare fee-for-service (FFS) with around 400,000 enrollees.
Utilization included office visits, emergency department visits, and inpatient admissions with International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes for influenza of J09, J10, and J11 among the first 10 positions. Current Procedural Terminology (CPT) codes of 87610, 87275-6, 87400, 87501-3, 87631-3, 87636, 87637, and 87804 were used to identify diagnostic testing for influenza. Vaccination against influenza was identified by the presence of a CPT code of 90630, 90653-8, 90660-4, 90666-8, 90672-3, 90685-9, G0008, G9141-2, and/or Q2033-9; or International Classification of Diseases, Tenth Revision, Procedure Coding System (ICD-10-PCS) code of 3E01340 and/or 3E02340. Findings were not risk- or acuity-adjusted.
For additional information on MedInsight, please visit medinsight.milliman.com.
1WHO. Review of Global Influenza Circulation, Late 2019 to 2020, and the Impact of the COVID-19 Pandemic on Influenza Circulation. Retrieved March 27, 2022, from https://www.who.int/publications/i/item/who-wer-9625-241-264.
2CDC. Flu Season. Retrieved March 27, 2022, from https://www.cdc.gov/flu/about/season/flu-season.htm.
3Russo E., Pu, T., Williams, D.V. et al. (June 2021). The Mysterious Case of the Missing Influenza Season. Milliman Brief. Retrieved March 27, 2022, from https://www.milliman.com/en/insight/the-mysterious-case-of-the-missing-influenza-season.
4CDC. FluView Summary Ending on October 2, 2021. Retrieved March 27, 2022, from https://www.cdc.gov/flu/weekly/weeklyarchives2020-2021/week39.htm.
5CDC. Frequently Asked Influenza (Flu) Questions: 2021-2022 Season. Retrieved March 27, 2022, from https://www.cdc.gov/flu/season/faq-flu-season-2021-2022.htm.
6WHO. Global Influenza Programme. Retrieved March 27, 2022, from https://www.who.int/teams/global-influenza-programme/vaccines.