Decline in Receipt of Vaccines by Medicare Beneficiaries During the COVID-19 Pandemic

United States, 2020

Kai Hong, PhD; Fangjun Zhou, PhD; Yuping Tsai, PhD; Tara C. Jatlaoui, MD; Anna M. Acosta, MD; Kathleen L. Dooling, MD; Miwako Kobayashi, MD; Megan C. Lindley, MPH


Morbidity and Mortality Weekly Report. 2021;70(7):245-249. 

In This Article

Abstract and Introduction


On March 13, 2020, the United States declared a national emergency concerning the novel coronavirus disease 2019 (COVID-19) outbreak.[1] In response, many state and local governments issued shelter-in-place or stay-at-home orders, restricting nonessential activities outside residents' homes.[2] CDC initially issued guidance recommending postponing routine adult vaccinations, which was later revised to recommend continuing to administer routine adult vaccines.[3] In addition, factors such as disrupted operations of health care facilities and safety concerns regarding exposure to SARS-CoV-2, the virus that causes COVID-19, resulted in delay or avoidance of routine medical care,[4] likely further affecting delivery of routine adult vaccinations. Medicare enrollment and claims data of Parts A (hospital insurance), B (medical insurance), and D (prescription drug insurance) were examined to assess the change in receipt of routine adult vaccines during the pandemic. Weekly receipt of four vaccines (13-valent pneumococcal conjugate vaccine [PCV13], 23-valent pneumococcal polysaccharide vaccine [PPSV23], tetanus-diphtheria or tetanus-diphtheria-acellular pertussis vaccine [Td/Tdap], and recombinant zoster vaccine [RZV]) by Medicare beneficiaries aged ≥65 years during January 5–July 18, 2020, was compared with that during January 6–July 20, 2019, for the total study sample and by race and ethnicity. Overall, weekly administration rates of the four examined vaccines declined by up to 89% after the national emergency declaration in mid-March[1] compared with those during the corresponding period in 2019. During the first week following the national emergency declaration, the weekly vaccination rates were 25%–62% lower than those during the corresponding week in 2019. After reaching their nadirs of 70%–89% below 2019 rates in the second to third week of April 2020, weekly vaccination rates gradually began to recover through mid-July, but by the last study week were still lower than were those during the corresponding period in 2019, with the exception of PPSV23. Vaccination declined sharply for all vaccines studied, overall and across all racial and ethnic groups. While the pandemic continues, vaccination providers should emphasize to patients the importance of continuing to receive routine vaccinations and provide reassurance by explaining the procedures in place to ensure patient safety.[3]

Medicare enrollment and insurance claims data for beneficiaries enrolled in a fee-for-service plan during weeks 2–29 of 2019 (January 6–July 20) and 2020 (January 5–July 18) were obtained from the Centers for Medicare & Medicaid Services Chronic Conditions Warehouse.[5] PCV13 and PPSV23 were covered by Part B, which pays for services from health care providers, outpatient care, and some preventive services. Td/Tdap was covered by Part B if it was administered as part of medically necessary service because of an injury and was covered by Part D, which covers a range of prescription drugs, if it was administered as part of preventive care. RZV was covered by Part D. Weekly claims for vaccination were identified by procedure code or national drug code on claims with a service date within the week examined (measured Sunday through Saturday).* Because some of the weeks spanned 2 different months, the denominator populations for PCV13 and PPSV23 vaccination in each week were defined as Medicare beneficiaries who were continuously enrolled in Parts A and B in the claim month and the previous month, and were aged ≥65 years on the first day of that previous month. The denominator population for Td/Tdap and RZV were defined similarly, except that beneficiaries were continuously enrolled in Parts A, B, and D. Weekly rates of receipt of the examined vaccines were calculated as the percentage of the denominator population that received ≥1 dose of the corresponding vaccine during that week. The percentage change in vaccination rate in a week was calculated as the ratio of the rate in that week in 2020 to the rate in the corresponding week in 2019, minus 1. Descriptive statistical analyses were conducted for the total study sample and stratified by race and ethnicity (non-Hispanic White [White], non-Hispanic Black [Black], Hispanic or Latino (Hispanic), non-Hispanic Asian/Asian American/Pacific Islander [Asian], non-Hispanic other [Other]).§ All statistical analyses were conducted during September 12–15, 2020, using SAS (version 9.4; SAS Institute). This activity was reviewed by CDC and was conducted consistent with applicable federal law and CDC policy.

During the study period, the average denominator populations were 27,194,802 in 2019 and 26,916,993 in 2020 for PCV13 and PPSV23, and 18,752,789 in 2019 and 18,701,076 in 2020 for Td/Tdap and RZV. Among Medicare beneficiaries, weekly rates of vaccination for each of the four vaccines declined precipitously after the national emergency declaration, compared with the corresponding weeks in 2019 (Figure). During January 5–March 14, 2020, weekly percentages of Medicare beneficiaries vaccinated with PPSV23, Td/Tdap, and RZV were consistently higher than were those during the corresponding week in 2019. Weekly vaccination rates dropped sharply during the first week after the national emergency declaration on March 13, with declines ranging from 25% for PPSV23 to 62% for RZV. The largest declines in weekly vaccination rates occurred during April 5–11, 2020, for PCV13, PPSV23, and Td/Tdap and during April 12–18, 2020, for RZV, when weekly vaccination rates dropped by 70% for Td/Tdap to 89% for RZV. After reaching this nadir, vaccination rates began to recover gradually. At the end of the study period (week commencing July 12), the weekly vaccination rate for PPSV23 was 8% higher than that during the corresponding week in 2019, but weekly vaccination rates for other examined vaccines remained 24% (Td/Tdap) to 43% (RZV) lower.


Percentage of Medicare beneficiaries aged ≥65 years who received PCV13,* PPSV23, Td/Tdap, and RZV vaccines, by week§ — United States, January 6–July 20, 2019 and January 5–July 18, 2020
Source: Medicare enrollment and claims data from the Centers for Medicare & Medicaid Services Chronic Conditions Warehouse.
Abbreviations: ACIP = Advisory Committee on Immunization Practices; PCV13 = 13-valent pneumococcal conjugate vaccine; PPSV23 = 23-valent pneumococcal polysaccharide vaccine; RZV = recombinant zoster vaccine; Td/Tdap = tetanus-diphtheria or tetanus-diphtheria-acellular pertussis vaccine.
*ACIP voted to stop recommending routine PCV13 use among adults aged ≥65 years in June 2019.
In October 2017, RZV was approved by the Food and Drug Administration and recommended by ACIP preferentially over Zoster Vaccine Live for use in immunocompetent adults aged ≥50 years.
§Calculated as the percentage of the study sample of Medicare enrollees who received ≥1 dose of the corresponding vaccine during that week.
The starting date of the first examined week in 2019 is January 6.

Among Medicare enrollees in the study sample, an average (across all vaccines studied) of 85% were White, 7% were Black, 4% were Other racial and ethnic groups, and 2% each were Asian or Hispanic. Patterns of decline and recovery in vaccination among each racial/ethnic group were similar to the overall findings, but certain disparities in magnitude were observed (Table). By the most recently assessed week, the rate of PPSV23 receipt among White adults was 10% higher than that during the corresponding week in 2019, and the declines in PCV 13 and Td/Tdap vaccination (42% and 23%, respectively) were smallest among White adults. By comparison, the change in vaccination rates in the most recent week of the study period compared with the corresponding week in 2019 ranged from –57% (PCV13) to –9% (PPSV23) for Asian adults, −44% (PCV13) to –2% (PPSV23) for Black adults, −62% (PCV13) to –22% (PPSV23) for Hispanic adults, and –44% (RZV) to 3% (PPSV23) for Other race adults. The smallest decline in RZV vaccination rate (11%) during the most recent week was among Black adults.

*Procedure codes alone for PCV13 and PPSV23 and procedure codes plus national drug codes for Td/Tdap and RZV.
RZV was recommended by the Advisory Committee on Immunization Practices (ACIP) to be administered as a two-dose series. In the analysis a Medicare beneficiary was counted as receiving RZV if either the first or the second dose was administered and was counted twice if both doses were administered (in different weeks) during the study period.
§Race/ethnicity of beneficiary was reported in the Medicare enrollment database, which was obtained from the Social Security Administration's beneficiary record. American Indian or Alaska Native persons and persons whose race/ethnicity was unknown were grouped into Other.
45 C.F.R. part 46, 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.