Current routine immunizations for children aged ≤10 years in the United States in 2019 cover 14 vaccine-preventable diseases. We characterize the public-health impact of vaccination by providing updated estimates of disease incidence with and without universally recommended pediatric vaccines.
Prevaccine disease incidence was obtained from published data or calculated using annual case estimates from the prevaccine period and United States population estimates during the same period. Vaccine-era incidence was calculated as the average incidence over the most recent 5 years of available surveillance data or obtained from published estimates (if surveillance data were not available). We adjusted for underreporting and calculated the percent reduction in overall and age-specific incidence for each disease. We multiplied prevaccine and vaccine-era incidence rates by 2019 United States population estimates to calculate annual number of cases averted by vaccination.
Routine immunization reduced the incidence of all targeted diseases, leading to reductions in incidence ranging from 17% (influenza) to 100% (diphtheria, Haemophilus influenzae type b, measles, mumps, polio, and rubella). For the 2019 United States population of 328 million people, these reductions equate to >24 million cases of vaccine-preventable disease averted. Vaccine-era disease incidence estimates remained highest for influenza (13 412 per 100 000) and Streptococcus pneumoniae-related acute otitis media (2756 per 100 000).
Routine childhood immunization in the United States continues to yield considerable sustained reductions in incidence across all targeted diseases. Efforts to maintain and improve vaccination coverage are necessary to continue experiencing low incidence levels of vaccine-preventable diseases.
The United States childhood vaccination program has dramatically reduced morbidity, mortality, and disability for targeted diseases. Updated estimates of disease incidence and cases averted, reflecting changes in disease epidemiology, vaccine utilization, and vaccine recommendations (based on the 2017 to 2021 schedule), are needed.
The childhood vaccination program reduced the incidence of all targeted diseases—with reductions ranging from 17% (influenza) to 100% (diphtheria, Haemophilus influenzae type b, measles, mumps, polio, and rubella)—and averted >24 million disease cases for the 2019 United States population.
Childhood vaccination has dramatically reduced morbidity, mortality, and disability caused by vaccine-preventable diseases, with ∼21 million hospitalizations, 732 000 deaths, and 322 million cases of disease averted in the United States between 1994 and 2013.1 Among diseases targeted by vaccines recommended before 1980, 3—polio, measles, and rubella—have achieved elimination status as defined by the World Health Organization2 and 1—smallpox—has been eradicated.3 Diphtheria and tetanus have declined markedly in incidence with routine immunization and are well controlled,2 whereas the incidence of pertussis and mumps has declined when compared with prevaccine levels but still fluctuates given periodic outbreaks since vaccination was introduced.3 The public health burden of diseases targeted in the childhood immunization program between 1980 and 2005, including hepatitis A, hepatitis B, invasive Haemophilus influenzae type b (Hib), varicella, and invasive pneumococcal disease (IPD), has decreased by more than 80%3 ; reductions in related nontargeted diseases (eg, acute otitis media caused by Streptococcus pneumoniae) have also been observed.4 After 2005, the routine immunization schedule5 for United States children ≤10 years of age targeted additional pathogens, such as rotavirus and further pneumococcal serotypes.5
This study updates estimates of the reduction in overall and age-specific disease incidence associated with the routine childhood immunization program in the United States (based on the 2017 to 2021 vaccination schedule). This update incorporates changes in vaccine utilization rates and observed incidence of the targeted vaccine-preventable diseases since previous evaluations.3,6 The present analysis will be of interest to policy makers, public health decision makers, and modelers concerned with public health interventions to minimize the burden of vaccine-preventable diseases. A companion study evaluated the value of the childhood immunization program for the 2017 United States birth cohort.7
We estimated the epidemiologic impact of the United States routine childhood immunization program (ages ≤10 years) by calculating the percent reduction in overall and age-specific disease incidence rates for each disease targeted by the program. We multiplied the prevaccine and vaccine-era incidence rates (using age-specific data, where available) by 2019 United States population estimates,8 accounting for underreporting where necessary, to calculate the 2019 clinical disease burden with and without childhood immunization and to estimate the cases averted by vaccination. As in previous studies, we assumed that the difference between incidence rates during these periods was entirely attributable to the childhood immunization program.3,6
For the prevaccine period, we estimated disease incidence using published incidence estimates or calculated incidence using published annual case estimates and United States population data from the same period. For the vaccine era, we calculated incidence as the average incidence over the most recent 5 years of available surveillance data; we used published incidence estimates if surveillance data were not available. For both periods, we accounted for underreporting where necessary.
Prevaccine and Vaccine-Era Disease Estimates
Table 1 summarizes the prevaccine and vaccine-era disease incidence sources. Age-specific incidence data were used for all diseases except diphtheria, polio, tetanus, and rotavirus. Incidence of Hib and rotavirus was limited to ages <5 years and diphtheria to ages ≤10 years, given lack of data in older age groups in the prevaccine period and the fact that clinical burden was largely limited to those age groups in both periods. Incidence of measles, mumps, and rubella was included only up to age 40 years, as prevaccine incidence data in ages ≥40 years was unavailable. For pneumococcal pneumonia, pneumococcal acute otitis media (AOM), and rotavirus, resource use estimates (ie, hospitalizations, emergency department [ED] visits, and outpatient visits) are reported instead of incidence and disease cases because of limitations in the source data.
We obtained prevaccine diphtheria disease incidence for children aged ≤10 years from an economic evaluation by Zhou et al,9 which estimated incidence from a 1916 to 1919 survey of childhood vaccine-preventable diseases in 31 353 United States children and physician-reported data.10 We assumed the incidence reported by Zhou et al9 for ages 5 to 9 years uniformly applied to all children ≤10 years. We calculated vaccine-era incidence among children aged ≤10 years as the average value over the most recent 5 years (2014 to 2018) of available data from the Centers for Disease Control and Prevention (CDC) National Notifiable Disease Surveillance System (NNDSS) reports.11–15
We calculated prevaccine hepatitis A incidence using the average number of reported cases between 1990 and 1994 from the NNDSS16–20 divided by the 1994 United States population for each respective age group.21 We calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15 A systematic review and meta-analysis of underreporting of hepatitis A in nonendemic countries found that reported hepatitis A cases ranged from 4% to 97% of total estimated cases across 8 included studies, with a pooled proportion of 59%.22 As a result, an underreporting factor of 1.7 (1/59% = 1.7) was applied for prevaccine and vaccine-era estimates,22 which is similar to underreporting factors found in other studies.23
We estimated prevaccine hepatitis B incidence as the average number of reported cases between 1976 and 1980 from the NNDSS24–28 and calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15 The underreporting factor for hepatitis B (6.5) was obtained from a probabilistic model estimating underreporting of hepatitis A, B, and C.23
Haemophilus Influenzae Type b
We obtained prevaccine disease incidence for Hib for children aged <5 years for 1976 to 1984 from an economic analysis by Zhou et al.29 We calculated overall incidence by summing the incidence values reported separately for Hib-related meningitis, epiglottitis, bacteremia, pneumonia, cellulitis, arthritis, and other invasive diseases reported in Zhou et al.29 We calculated vaccine-era incidence among children aged <5 years as the average value over the most recent 5 years (2013–2017) of available data from CDC Active Bacterial Core (ABC) surveillance reports.30–34
For influenza, instead of using data from the period before influenza vaccines were routinely recommended, we estimated prevaccine incidence among children aged ≤10 years by using the number of cases and averted cases estimated by the CDC, assuming all averted cases would have occurred without vaccination.35–42 Specifically, we summed the number of reported cases to the cases averted by vaccination among children <5 years and children aged 5 to 10 years for 5 recent influenza seasons (2014–2015 to 2018–2019) and then divided the total number of cases by the number of children in the United States in each respective age group for the same period.8 An average incidence across the 5 years was then calculated for both age groups. For vaccine-era incidence, we used the same source and calculated the average incidence over the same 5 recent seasons (2014–2015 to 2018–2019). Our analyses did not account for the impact of adolescent and adult influenza vaccination or herd immunity in older age groups; therefore, incidence of influenza was restricted to ages ≤10 years, and we attributed all changes in incidence to vaccination in this age cohort.
Measles, Mumps, and Rubella
We estimated prevaccine pertussis incidence for birth to 10 years from 2 economic evaluations of diphtheria, tetanus, and acellular pertussis vaccine, which derived age-specific risk of pertussis from United States data in the 1920s and from Sweden in the 1980s.9,48 Prevaccine incidence for ages >10 years was calculated using the number of reported pertussis cases estimated by Roush and Murphy3 for ages >10 years during 1934 to 1943 (before the start of routine pertussis vaccination in the late 1940s) divided by the size of the United States population >10 years old over the same period.49,50 We calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15 An underreporting factor of 10 was applied in the prevaccine and vaccine eras (Table 1).51–56
For IPD, we calculated prevaccine disease incidence as the average value from the 1997 to 1999 ABC surveillance reports57–59 and calculated vaccine-era incidence as the average value from the 2013 to 2017 ABC surveillance reports.60–64
For pneumococcal pneumonia, we obtained prevaccine, age-specific, all-cause pneumonia hospitalization rates per 100 000 for the period 1997 to 199965 and all-cause outpatient visit rates per 100 000 for the period 1998 to 200066,67 (Table 1). For the vaccine era, we used the incidence of all-cause pneumonia from 2014 based on an analysis of a large convenience insurance claims dataset (MarketScan) multiplied by the percentage hospitalized or treated in an outpatient or ED setting taken from the same study.68 We multiplied the all-cause rates by the prevaccine69,70 and postvaccine71,72 percentage of all-cause pneumonia caused by pneumococcus (Table 1).
For pneumococcal AOM, we used prevaccine, age-specific incidence from 1997 to 1999 and vaccine-era incidence from 2012 to 2014 from a retrospective analysis of the National Ambulatory Medical Care Survey comparing ambulatory visit rates before the introduction of 7-valent and following 13-valent pneumococcal conjugate vaccine.4 We summed annual rates of physician office, hospital outpatient, and hospital ED visits to calculate a total annual ambulatory visit rate per 1000 children. To calculate pneumococcal AOM burden for each period, we multiplied all-cause rates by the percentage of AOM caused by pneumococcus in the prevaccine period (1995 to 2001) (44%) and vaccine era (2010 to 2016) (21%).73
For polio, we obtained the average number of paralytic poliomyelitis cases for the period 1951 to 1954 (before the introduction of the first polio vaccine in 1955) from Roush and Murphy3 . We divided the total number of cases by the average United States population size from 1951 to 1954 to estimate an overall incidence rate.49 Age-specific data were not available in the prevaccine period; therefore, the same incidence rate was used for all ages. A prevaccine underreporting factor of 2.1 was applied (Table 1).74,75 We calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15
We calculated prevaccine estimates of rotavirus-related burden among children aged <5 years using 1993 to 2002 data on the cumulative individual risk of event by age 59 months for events including hospitalizations, ED visits, and hospital or ambulatory outpatient visits.76 The median values were used to calculate annual probabilities of each type of rotavirus-related resource use. We further assumed rotavirus events were uniformly distributed from birth to age 5 years (Supplemental Table 3). In the vaccine era, we calculated rotavirus-related burden by multiplying prevaccine event rates by the estimated reduction in hospitalizations77 and reduction in ED and outpatient visits.78
We calculated prevaccine tetanus incidence based on the number of cases reported during 1947 to 1949 (before routine vaccination began in the late 1940s3 ) divided by the average size of the United States population during that same period.49 Data were not available by age in the prevaccine period; therefore, the same incidence rate was used across all ages in the model. We calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15
We calculated prevaccine varicella incidence using the average number of reported varicella cases between 1990 and 1994 (before vaccine introduction in 1995) from the NNDSS16–20 divided by the 1994 United States population for each respective age group.20,21 We calculated vaccine-era incidence as the average value over the most recent 5 years (2014 to 2018) of available data from the NNDSS.11–15 Underreporting factors of 22.2 and 10.4 were applied to prevaccine and vaccine-era incidence, respectively (Table 1).3,20
We report calculated incidence overall and by age for both the prevaccine and 2019 vaccine-era periods. We calculated the percent reduction in incidence overall and by age group for each disease by comparing the 2 periods. Using 2019 United States population estimates from the United States Census Bureau, we calculated the number of cases of each disease that would be expected in 2019 without and with the routine childhood immunization program and the number of cases of disease averted.
For infants (<1 year), prevaccine annual incidence per 100 000 was highest for pneumococcal AOM (49 324), influenza (18 903), measles (9200), and pertussis (4720) (Supplemental Tables 3–5). For young children (ages 1 to 4 years), as for infants, incidence in the prevaccine period was highest for pneumococcal AOM (15 004–49 324), influenza (18 903), measles (10 641–11 503), and pertussis (4720), as well as for varicella (4519). For school-aged children (ages 5–18 years), prevaccine incidence varied by age group but was highest for influenza (14 066), varicella (389–6480), pneumococcal AOM (4840), and pertussis (131–4720). For adults, prevaccine incidence was highest for pneumococcal pneumonia (29–1553), rubella (300), mumps (99–256), and pertussis (131).
After vaccines were introduced, incidence decreased for all diseases evaluated (Fig 1; Table 2). Incidence was reduced to less than 1 per 100 000 for 6 of the diseases: diphtheria, Hib, measles, polio, rubella, and tetanus. The incidence of mumps was reduced by >99% and varicella by 98%. The incidence of rotavirus-related hospitalizations among children aged <5 years was reduced by 91%; a lower reduction was observed for rotavirus-related ED visits (61%) and outpatient visits (45%). The incidence of pertussis was reduced by 91%, hepatitis A by 87%, hepatitis B by 86%, and IPD by 60%. Pneumococcal pneumonia hospitalization rates and outpatient visit rates decreased by 84% and 69%, respectively, and incidence of pneumococcal AOM decreased by 75%. The incidence of influenza among people aged <11 years was reduced by 17%.
For the 2019 United States population of 328 million people, the number of cases of each disease without and with the childhood immunization program and the estimated number of cases averted are shown in Table 2. In the vaccine era with routine immunization, the annual number of cases of disease was 0 for polio, <10 cases per year for diphtheria and rubella, and <100 cases per year for Hib and tetanus. Pneumococcal AOM and influenza represented the largest clinical burden annually (>1 000 000 cases per year), followed by pertussis, pneumococcal pneumonia, outpatient rotavirus gastroenteritis, and outpatient varicella (between 100 000 and 1 million cases per year).
Routine immunization was estimated to avert over 24 million cases of vaccine-preventable disease in 2019 across all age groups, ranging from approximately 1000 cases of tetanus averted to more than 4.2 million varicella cases averted (Table 2). Cases averted were greatest (>1 000 000) for influenza, measles, mumps, rubella, pertussis, varicella, and outpatient visits for pneumococcal AOM.
This analysis found that routine childhood immunization in the United States has continued to reduce the incidence of all targeted diseases. Landmark achievements have been the reduction in incidence of diphtheria, Hib, measles, polio, rubella, and tetanus to negligible levels (<1 case per 100 000 population annually); and >90% reduction in incidence for 10 diseases targeted by the routine childhood immunization program for children ≤10 years of age. These reductions equate to the prevention of over 24 million cases of disease for the 2019 US population.
Roush and Murphy3 evaluated the impact of routine childhood immunization on vaccine-preventable diseases for which recommendations were in place before 2005, using 2006 disease data. Our estimates were generally consistent with the previous results and other published studies,79 although we estimated a greater reduction in incidence of IPD (60% versus 34%) and of varicella (98% versus 85%). A potential explanation for these differences may be that our analysis used vaccine-era incidence from 2013 to 2017 for pneumococcal disease and from 2014 to 2018 for varicella, capturing the greater impact of the 13-valent pneumococcal conjugate vaccine (recommended in 2010 for infants) compared with the 7-valent pneumococcal conjugate vaccine and capturing the greater impact of 2-dose varicella vaccine compared with 1 dose (second dose added to recommendations in 2007).80
With sustained vaccine coverage at levels greater than 80% for most pediatric vaccines (with the exception of hepatitis A, rotavirus, and annual influenza vaccine), many vaccine-preventable diseases are now controlled as a public health problem or eliminated in the United States. However, despite significant impact of vaccines, continued risk from these vaccine-preventable diseases remains. When whole-cell pertussis vaccine was withdrawn in Sweden in 1979 because of concerns about safety and efficacy, incidence rates of pertussis similar to those observed in the prevaccine era returned in Sweden within a few years; after introduction of the diphtheria, tetanus, and acellular pertussis vaccine in 1996, incidence rates decreased markedly compared with the 1986 to 1995 10-year period.81,82 Similarly, despite elimination status being declared for measles in 2000, under-vaccination has led to continued measles outbreaks in the United States, jeopardizing elimination status for the disease.83–85 Diphtheria outbreaks continue to occur where vaccination rates are low, particularly in areas of social disruption, and are often associated with high rates of mortality.86,87 The most recent large outbreak occurred in Russia from 1990 to 1997, resulting in ∼115 000 cases and 3000 deaths across the population.88 These experiences underscore the importance of continued immunization in sustaining reductions in incidence of infectious diseases.
This analysis includes some limitations. First, consistent with previous studies,3,6 the analysis does not directly account for other public health measures (eg, better sanitation, healthcare access, and improved standards of care) that have been introduced over the past 70 years and likely contributed to the reduction in vaccine-preventable diseases. Furthermore, this analysis did not account for random error in the parameter estimates or account for the proportion of disease incidence reduction that may be attributed to adolescent and adult vaccines or to booster doses. As a result, the analysis may overestimate reductions in burden directly attributable to childhood immunization. Future analyses could address these methodological limitations using time-series analysis to identify and adjust for trends to explore the extent to which adolescent and adult vaccination programs, which have expanded since 2005,80,89,90 contribute to reduction in disease incidence.
Second, owing to limited data on differences among racial and ethnic groups, this analysis did not account for racial or ethnic disparities in vaccine coverage and incidence of vaccine-preventable diseases. Evaluating the public health impact of routine immunization among racial and ethnic groups is an important direction for future research. Moreover, this analysis was limited in scope to vaccine-preventable disease for vaccines included in the United States routine childhood immunization program for children ages ≤10 years. Expansion of this analysis to include vaccine-preventable diseases, such as meningococcus and human papillomavirus targeted by routine adolescent vaccines, is another potential area of future research.
Third, because annual incidence varies substantially from year to year for many vaccine-preventable diseases, we have calculated prevaccine and vaccine-era incidence as averages across multiple years, where data allowed. Despite our efforts to estimate average incidence values in both periods, significant epidemics or outbreaks occurred for some diseases that may not be reflected in the annual averages used in this analysis.91 For the vaccine era, data used to derive disease incidence were for years preceding the coronavirus disease 2019 (COVID-19) pandemic. There are multiple factors that may influence the impact of COVID-19 on the incidence of vaccine-preventable diseases. For example, behavior changes caused by nonpharmaceutical interventions, including lockdowns, face-covering use, and other social distancing measures may reduce the transmission of some diseases, while simultaneously causing disruptions to vaccine uptake and coverage for the pediatric population that may adversely impact the prevention of vaccine-preventable diseases.92–94 Future surveillance and survey data will help to understand the impact of the COVID-19 pandemic and other potential “shocks” to the immunization program on the transmission of other vaccine-preventable diseases.
Routine childhood immunization in the United States has continued to reduce the incidence of all targeted vaccine-preventable diseases. In the vaccine era, the incidence of diphtheria, Hib, measles, polio, rubella, and tetanus has been reduced to <1 per 100 000; across all targeted diseases, ∼24 million cases have been averted because of vaccination for the 2019 United States population. Routine immunization remains an effective public health intervention to avert disease; maintenance of high rates of vaccination coverage is necessary for sustained impact.
We thank Kate Lothman of RTI Health Solutions, who provided medical writing support for the development of this manuscript and whose services were funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co, Inc, Rahway, NJ, USA.
Ms Talbird, Mr Carrico, and Dr La conceptualized and designed the study, reviewed the literature, interpreted the results, and drafted the initial manuscript; Drs Chen, Nyaku, Carias, and Roberts conceptualized the study and provided input on the study design, secured funding, and interpreted the results; Dr Marshall interpreted the results; and all authors reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
COMPANION PAPERS: companions to this article can be found online at http://www.pediatrics.org/cgi/doi/10.1542/peds.2021-056007 and http://www.pediatrics.org/cgi/doi/10.1542/peds.2022-057831.
Dr La’s current affiliation is GSK, Philadelphia, Pennsylvania.
FUNDING: This study was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co, Inc, Rahway, NJ, USA.
CONFLICT OF INTEREST DISCLOSURES: Ms Talbird and Mr Carrico are employed by RTI Health Solutions, which received funding for the conduct of this study. Dr La was an employee of RTI Health Solutions when this study was conducted and is now an employee and shareholder in the GSK group of companies. Drs Chen, Carias, and Roberts are employees of Merck Sharp and Dohme LLC, a subsidiary of Merck & Co, Inc, Rahway, NJ, and are shareholders in Merck & Co, Inc. Rahway, NJ. Dr Nyaku was an employee of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co, Inc, Rahway, NJ and a shareholder in Merck & Co, Inc, Rahway, NJ when this study was conducted. Dr Marshall has been an investigator on clinical trials funded by GlaxoSmithKline, Merck, Pfizer, Sanofi Pasteur, and Seqirus, and he has received honoraria from these companies for service on advisory boards and/or nonbranded presentations.