Description of vaccination coverage and hesitancy obtained by epidemiological survey of children born in 2017-2018, in Belo Horizonte and Sete Lagoas, Minas Gerais, Brazil

ABSTRACT Objective To describe vaccination coverage and hesitation for the basic children’s schedule in Belo Horizonte and Sete Lagoas, Minas Gerais state, Brazil. Methods Population-based epidemiological surveys performed from 2020 to 2022, which estimated vaccine coverage by type of immunobiological product and full schedule (valid and ministered doses), according to socioeconomic strata; and reasons for vaccination hesitancy. Results Overall coverage with valid doses and vaccination hesitancy for at least one vaccine were, respectively, 50.2% (95%CI 44.1;56.2) and 1.6% (95%CI 0.9;2.7), in Belo Horizonte (n = 1,866), and 64.9% (95%CI 56.9;72.1) and 1.0% (95%CI 0.3;2.8), in Sete Lagoas (n = 451), with differences between socioeconomic strata. Fear of severe reactions was the main reason for vaccination hesitancy. Conclusion Coverage was identified as being below recommended levels for most vaccines. Disinformation should be combated in order to avoid vaccination hesitancy. There is a pressing need to recover coverages, considering public health service access and socioeconomic disparities.


INTRODUCTION
3][4] In particular, the BCG vaccine (bacillus Calmette-Guérin) showed a 10% drop in the period. 5[8] With the aim of supporting control and prevention actions, epidemiological surveillance aims to obtain accurate and timely information about coverages. Within this context, the National Vaccination Coverage Survey 2020 (INCV 2020) aimed to compute coverage in a more realistic way, and this study aimed to describe vaccination coverage and hesitancy for the basic childhood schedule based on the INCV surveys carried out in Belo Horizonte and Sete Lagoas.

Study design
This is a population-based survey, carried out in the cities of Belo Horizonte and Sete Lagoas, between September 2020 and March 2022.The study is part of the INCV 2020 carried out in the Brazilian state capitals, Federal District and in 12 cities in the interior region of Brazil with more than 100,000 inhabitants. 3,10

Background
In 2020, Belo Horizonte had an estimated resident population of 2,521,564 inhabitants, 5.2% (130,707) of whom were children born into the 2017 and 2018 cohorts, and a birth rate of 10.42 live births per 1,000 inhabitants.According to the PNI Information System, in 2018 there were 191 vaccination rooms that were either public, private or both.In addition to the state capital Belo Horizonte, we chose the city of Sete Lagoas f rom among municipalities located outside the Belo Horizonte metropolitan region with more than 100,000 inhabitants because its coverage is one of the lowest.In 2020 the municipality of Sete Lagoas had an estimated resident population of 241,835 inhabitants, 5.9% (14,167) of whom children born into the 2017-2018 cohorts, and a birth rate of 10.70 live births per 1,000 inhabitants.Taking both the public and private health sectors, 28 establishments administered vaccines in 2021, according to data from the Sete Lagoas City Health Department. 10

Participants
The target population was comprised of 59,957 live births in Belo Horizonte and 5,261 live births in Sete Lagoas from the 2017-2018 birth cohorts.

Sampling
The sampling procedure was carried out in multiple stages.The stratif ied sample, according to socioeconomic strata, was formed by clusters with selection in two stages (random selection of census tracts and households).
In each municipality, the socioeconomic strata were defined by ordering the census sectors according to the average income of those responsible for each household, the proportion of literate child guardians and income greater than or equal to 20 minimum wages.The census tracts were the primary units of analysis.
Subsequently, clusters were defined with a minimum number of children to be selected randomly, in order to reach the desired sample size.The children were located based on the geographic coordinates of the addresses available on the Live Birth Information System and, when necessary, by active searching in the clusters.Details of sampling in each municipality have already been described in previous publications. 3,10

Variables of interest
The full basic schedule included the set of vaccines to be administered up to 24 months of life and which are included in the Ministry of Health's basic childhood vaccination schedule: BCG, hepatitis B, 5-in1, inactivated poliovirus vaccine (IPV), 10-valent pneumococcal conjugate, human rotavirus, meningococcal C conjugate, yellow fever.The full schedule at 24 months includes, in addition to the basic schedule vaccines, MMR (measles, mumps and rubella), hepatitis A, chickenpox and attenuated oral poliovirus vaccine (OPV), DTP booster (diphtheria, pertussis and tetanus), meningococcal C and pneumococcal vaccine. 3ellow fever vaccine was not included because its having been introduced or not into the basic schedule varied between states.
T h e e x p o s u r e v a r i a b l e s w e r e sociodemographic, maternal (reproductive), household, family consumption, and child vaccination data, in addition to reasons for vaccination hesitancy, difficulties encountered,

ORIGINAL ARTICLE
Vaccination coverage and hesitancy in Minas Gerais, Brazil and guardians' perceptions about vaccines. 3As different compositions of vaccines are used to protect against the same diseases and as their use is different in the public and private sectors, in these situations, the administration dates of these vaccines were standardized in a variable related to each vaccine on the PNI schedule, such as 5-in-1 vaccine = 5-in-1 + hexavalent + acellular.Details of the procedure for each vaccine are described in the national survey technical report. 10ta source/measurement Coverage, using the administrative method, which represents the proportion of the target population vaccinated, was obtained by dividing the number of administered doses of a vaccine by the target population, multiplied by 100. 11In the survey, coverage was calculated based on the administration dates of vaccines recorded on vaccination cards, with vaccination schedules being calculated for administered doses (number of recorded doses of each vaccine) and valid/timely doses (considering the time they were administered in relation to date of birth and intervals between doses). 3These data were obtained from photographs of the vaccination cards, which were interpreted and transcribed by professionals with vaccination room experience.Children without a vaccination card were considered unvaccinated after an unsuccessful search for their record on the PNI Information System.Further information was obtained via the questionnaire answered by the person responsible for each child.Details of the field work, data collection and transcription, as well as problems noted and potential biases, have already been presented in previous publications. 3,10

Sociodemographic characteristics of the families:
-Family consumption level: defined according to cutoff points of the 2019 Brazilian Economic

Maternal characteristics:
-Schooling (incomplete elementary, complete elementary or incomplete high school, complete high school or incomplete higher education, complete higher education or above); -Age group (< 20 years, 20 -34 years, ≥ 35 years), race/skin color (White, mixed race, Black, Asian, Indigenous), paid work (yes/ no), lives with a partner (yes/no), number of children alive.
Parental perception regarding statements about vaccines was assessed using a Likert scale, with answer scores ranging f rom 1 (totally disagree) to 5 (totally agree).The score obtained was later regrouped into the following categories: I totally or partially disagree, I am indifferent, I totally or partially agree. 10Regarding vaccination hesitancy of those responsible for the child, agreements or disagreements in relation to the following statements were considered: vaccines are not important; does not trust vaccines provided by the government; does not believe that vaccines are important for children's health; Vaccinations are not important for neighborhood children; there is no need for vaccines for diseases that no longer exist; vaccines cause severe reactions.

Statistical methods
The descriptive analysis of coverages was carried out separately for Belo Horizonte and Sete Lagoas, by calculating summary statistical measures (means and proportions) and building graphs of point and interval estimates of coverages prevalence and other characteristics of the study population, considering the complex sampling plans, measurement weights and subsequent calibration of population samples. 3,10The results were interpreted based on the coverage targets established by the Ministry of Health: 90%, for BCG and rotavirus; 95%, for hepatitis B, meningococcal C, 5-in-1, pneumococcal, poliomyelitis, hepatitis A, MMR; and 100% for DPT.
Estimates of the socioeconomic and demographic characteristics of the survey families, mothers and children were presented only for the highest and lowest income strata (A and D), due to similarities in coverage prevalence levels between the intermediate strata.The analyses were carried out using R statistical software version 4.3.2. 13

Ethical considerations
The

RESULTS
We studied data on 1,866 children from Belo Horizonte (no losses), 470 from socioeconomic stratum A, 458 f rom stratum B, 469 f rom stratum C and 469 from stratum D. Of the total sample, 99.2% had a vaccination card.By strata, 99.9%, 98.5%, 99.0% and 99.4% of children, respectively, from strata A, B, C and D, had a vaccination card.63.5% of children in stratum A and 9.3% in stratum D were vaccinated in private vaccination services.
We studied data on 451 children from Sete Lagoas (no losses), 99.9% of whom had a vaccination card, with little variation between strata.Percentage of use of private vaccination services was 53.7% for stratum A, and 19.0% for stratum D.
Table 1 shows the socioeconomic and demographic characteristics of the survey families, mothers and children, by highest and lowest extreme income strata (A and D) in Belo Horizonte and Sete Lagoas.In Belo Horizonte, the percentage of children from families with high and medium consumption was 70.1% in stratum A, while those with low and very low consumption accounted for 87.8% in stratum D. In Sete Lagoas, the proportion of families with high and medium consumption in stratum A (53.1%) was lower than that found for Belo Horizonte.
In Belo Horizonte, 53.4% of families in stratum A had monthly income of more than BRL 3000, while in Sete Lagoas, income was below this amount for 81.9% of stratum D families.In Belo Horizonte and Sete Lagoas, respectively, 15.0% and 6.1% of mothers in socioeconomic stratum D had higher education or above, while 76.5% and 62.4% of mothers in stratum A had higher education or above.15.1% of families in Belo Horizonte, and 23.8% in Sete Lagoas, were Bolsa Família program beneficiaries.

ORIGINAL ARTICLE
Vaccination coverage and hesitancy in Minas Gerais, Brazil Figure 1 shows coverages point estimates (%) of the full schedule at 24 months of age, doses administered and valid doses, both total and according to population characteristics, in both municipalities.
Table 2 shows coverage of doses administered and valid doses for the immunobiological products assessed.In Belo Horizonte, coverage varied f rom 75.7% for the pneumococcal vaccine booster, to 90.3% for the first dose (D1) of 5-in-1 vaccine, with regard to valid doses.
Figure 2 shows coverage of the vaccines (administered and valid doses) on the vaccination schedule recommended by the Ministry of Health for children aged up to 24 months, in Belo Horizonte and Sete Lagoas.Coverage was lower in stratum A, although coverages variability was greater in stratum D in both municipalities.In Belo Horizonte, valid doses coverage was lower than that of doses administered for the pneumococcal vaccine booster, being higher in stratum A. In Sete Lagoas, there was a greater difference between doses administered and valid doses for the DPT booster in the general population, and for the pneumococcal booster in stratum A.
Among the reasons for vaccination hesitancy, not vaccinating children, even when taking them for vaccination at a health center, was reported by 18 Figure 3 shows the frequency of the main reasons for vaccination hesitancy, with emphasis on non-vaccination due to fear of severe reactions, which was reported by 18.0% (95%CI 14.2;22.4) of those responsible for child vaccination in Belo Horizonte, and by 23.1% (95%CI 17.1;30.3),in Sete Lagoas.Vaccination hesitancy due to the idea that it is unnecessary to vaccinate against diseases that no longer

ORIGINAL ARTICLE
Vaccination coverage and hesitancy in Minas Gerais, Brazil   Mistrust gov.vac.: does not trust vaccines provided by the government; not important: does not believe that vaccines are important for children's health; not needed in neighborhood: vaccines are not important for the children in the neighborhood; no disease: there is no need for vaccines for diseases that no longer exist; severe reactions: vaccines cause severe reactions.

DISCUSSION
In general, coverage was below recommended levels, with significant differences between the socioeconomic strata of the municipalities.In Belo Horizonte, all vaccines were below target, whereby rotavirus second dose and DPT and meningococcal C boosters had the lowest coverage in terms of doses administered, while rotavirus second dose and poliomyelitis and DPT boosters having the lowest coverage with valid doses.In Sete Lagoas, the MMR second dose and pneumococcal, meningococcal C and DPT boosters had the lowest coverage.5-in-1 third dose, yellow fever, MMR second dose, chickenpox and meningococcal C and DPT boosters were below target.
A considerable percentage of children's guardians reported difficulties in accessing vaccination at the right time at their health centers.Other reasons for vaccination hesitancy cited were non-vaccination for fear of adverse events or the belief that it was unnecessary to vaccinate their children against diseases that no longer exist.
The results suggest good integration between immunization health services in relation to vaccination control for children, represented by the low proportion of children without vaccination cards.This may be related to strengthening linkage and trust in SUS health services, as well as social programs, such as the Bolsa Família program, requiring children to be vaccinated, a fact that generates greater awareness of the importance of vaccination in low-income communities and, consequently, higher coverages. 14,15cioeconomic and demographic inequalities in the target population of municipalities can have an impact on children's health, thus affecting coverages.Children in stratum D are more likely to belong to families with low income and low maternal education, as well as a greater number of children, making access to health services difficult and compromising fulf ilment of the vaccination schedule.The lowest coverage occurred in the least vulnerable stratum (stratum A), despite there having been changes in social programs in the period, and the socioeconomic strata having been defined based on the 2010 Census.Furthermore, studies have shown that awareness the importance of vaccination is higher in low income communities. 14,15Therefore, greater attention should be paid to stratum A, although also guaranteeing equitable access to vaccines in other strata, in order to protect the health and well-being of children. 16verage was higher in households with four or more dwellers, which may be explained by greater linkage with primary health services or coverage by the Family Health Strategy.In Sete Lagoas, coverage was lower among children whose mothers only had elementary education, possibly due to less access to information and education about the importance of immunization.
Access can be enhanced through primary health care and the proximity of families to health services, as demonstrated by a systematic review in European countries and Australia.That 2019 study shows that structural and organizational aspects of health care systems for young children are important for equity in vaccine acceptance. 17ere was greater vaccination hesitancy in stratum A in Belo Horizonte, possibly due to greater access to information and misinformation, such as fake news and rumors on digital media.Diff iculty in accessing a primary health care center was the reason reported by families who were most vulnerable.Fear of severe reactions demands that information be disseminated about the real risks of vaccines, reducing misconceptions and promoting greater adherence. 18In this context, digital media amplify anti-vaccine discourse, with objections related to adverse events and minimization of disease severity. 19global overview of systematic reviews on barriers to childhood vaccination identified 573 descriptions, categorized into six broad categories: (1) access, (2) clinical or health system barriers, (3) concerns and beliefs, (4) perceptions and experiences of health, (5) knowledge and information and (6) social or family influence.These reasons appeared in the INCV 2020, requiring reflection on strategies to change this scenario, such as awareness campaigns about the importance of vaccination, focused on strata with lower coverages, training of health professionals, improving access, with special attention to socially vulnerable families. 20other systematic review assessed parents' perceptions, showing that they considered mandatory immunization schedules to be a violation of their rights, and did not like schedules that offered financial incentives for vaccination.On the other hand, some parents felt that schedules limiting school access for unvaccinated children gave them peace of mind. 21r study described information on vaccination coverage and hesitancy, in addition to official records, using information from public and private services, including information on unvaccinated children.Although it may have had an impact on the child immunization process, we did not assess the effects of the COVID-19 pandemic. 3,10A limitation of this study is that it was not possible to assess significant differences between the two municipalities studied, since the sample was not designed for this purpose.Both municipalities showed low coverages and inequalities between socioeconomic strata, pointing to the need to recover high coverages levels, prioritizing vaccines with coverage below recommended levels in all socioeconomic strata, considering greater access and health education.

Figure 1 -
Figure 1 -Coverage (%) of the full immunization schedule at 24 months, according to socioeconomic and demographic characteristics reported by the children's guardians, in Belo Horizonte (n = 1,866) and Sete Lagoas (n = 451), National Vaccination Coverage Survey, Brazil, 2020

Figure 3 -
Figure 3 -Reasons for vaccination hesitancy reported by the guardians of children up to 24 months old in Belo Horizonte (n = 1,866) and Sete Lagoas (n = 451), both total and by socioeconomic strata A and D, National Vaccination Coverage Survey, Brazil, 2020