1. Characteristics of participants by the type of BCGs
For the investigation of BCG scar formation, the data from a total of 5,713 children were used, collected from 121 day-care centers and kindergartens in 16 cities of South Korea. From the information collected for those children whose parents responded to the questionnaire, there were 2,995 male and 2,718 female children with a ratio 1.1 to 1. Among the 5,713, for 95.6% (n=5,462) the parent indicated that the child had been vaccinated with BCG.
The distribution of BCG scar status in the
Table 1 was categorized primarily by the data collected from scar inspection in the field. When an examiner recorded the child as 'no BCG scar' during the inspection but the parents had ticked vaccination with one of BCGs in the questionnaire, the answer from the parents was used. However, when an examiner recorded the child as 'no BCG scar' during the inspection and the parents marked 'not sure' for BCG status in the survey, then this was categorized as 'no scar' in the analysis.
For those children who had been inspected for BCG scar, 1,778 (31.1%) had been vaccinated with BCG Pasteur by the intradermal method and 3,684 (64.5%) with BCG Tokyo by the multipuncture method. From this analysis, 251 children (4.4%) had 'no BCG scar' with an increasing trend by age.
Among girls, 31.6% had been vaccinated with BCG Pasteur by the intradermal method (n=817) and 68.4% with the BCG Tokyo by the multipuncture method (n=1,769). Among boys, 15.9% had been vaccinated with BCG Pasteur by the intradermal method (n=361) and 84.1% with the BCG Tokyo by the multipuncture method (n=1,915).
The distribution of the type of BCGs varied by age, showing the proportions of BCG given with the intradermal method were from 45.3% to 23.7% (p<0.0001). It is noted that children less than 2 years old had relatively higher proportions of BCG Pasteur with the intradermal method than had been expected. Looking separately at the data collected from day-care centers and kindergartens, the proportion of BCG Pasteur vaccination with the intradermal method increased by age, indicating that the proportion of children vaccinated with BCG Tokyo with the multipuncture method was gradually increasing in recent years.
Analysis of the providers of BCG vaccination, in the case of children given BCG Pasteur/intradermal, 88.7% of responders said that their children had been vaccinated in public health care centers. On the other hand, the data showed that private clinics preferred to provide BCG Tokyo/multipuncture.
Interestingly, 7.4% of parents/guardians of BCG Tokyo vaccinees responded in the survey that their children had received the vaccine in public health care centers, where the BCG Tokyo vaccine is not provided. This may be due to recall bias or inaccuracy in recording, given that in the private sector vaccination is also provided from a public health center. Compared to the other private clinics, pediatrics clinics had a significantly lower proportion of vaccination with BCG Pasteur/intradermal as 11.9%.
To observe any association with the economic status of parents/guardians, information on monthly insurance payments in proportion to their incomes was collected in the survey. The data showed that those parents/guardians who paid the higher insurance payments (which indicates higher incomes) had the higher proportion of vaccination with BCG Tokyo/multipuncture (p<0.0001).
In terms of the regional preference of the type of BCGs, 35.0% of children living in Jeju had been vaccinated with BCG Tokyo/multipuncture while 75.5% of children living in Busan had been vaccinated with the same vaccination, showing that the big cities with higher income had a relatively high proportion of BCG Tokyo/multipuncture than other smaller cities. Since BCG Tokyo/multipuncture is more expensive than BCG Pasteur/intradermal which is free of charge, it seems children in bigger cities had more access to the more expensive BCG Tokyo vaccine.
2. BCG scar formation
Among those children who had been vaccinated with BCG Pasteur by the intradermal method at the age of 0 to 6 years (n=1,778), the overall median of the BCG scar size was 5.0 mm (interquartile range [IQR], 1~15). According to the non-parametric analysis in
Table 2, the data shows that the medians of scar size were significantly different by age (Kruskal-Wallis test, p<0.001), especially the older age group (4~6 years) had significantly higher scar sizes than the younger age group (0~3 years) (Wilcoxon rank sum test, p<0.001). There were 75 children (4.2%) who had scar sizes of more than 10 mm, which needs further investigation as to whether this was due to keloid formation or recording errors. From observation of the distribution in scar size, it is noted that there are distinct peaks at 5 and 10 mm (
Figure 1). It might be possible that examiners from public health care centers had 'digital preferences' when measuring the size of BCG scar at the inspection.
Among those children who had been vaccinated with BCG Tokyo/multipuncture, the number of BCG pin scars was recorded at ages of 0 to 6 years from the scar survey (n=3,684). For all the children, the median of pin scars was 16 (IQR, 12~16), the maximum number of pin scars was 36, and the minimum number was 1. According to the non-parametric analysis in
Table 2, the data shows that the medians of pin scar number were significantly different by age (Kruskal-Wallis test, p<0.001), especially the younger age group (0~3 years) had significantly higher pin scars than the older age group (4~6 years) (Wilcoxon rank sum test, p<0.001).