TABLE 5. Association between seroreversion and Borrelia burgdorferi antigen-specific IgG antibodies and annual rate of seroreversion by age group among participants of a follow-up to a cross-sectional survey of children and adolescents (KiGGS), Germany, 2003–2006 and 2014–2017 (n = 131).

Characteristics	Seropositive participants in KiGGS Baseline		Seroreversions		p valuea	Univariable analyses		Multivariable analysesb
	n	%	n	%		OR	95% CI	aOR		95% CI
Antigens (7 missing values)
OspC	24	19.4	8	16.3	0.49	0.72	0.27–1.80	0.55	0.16–1.72
VlsE	114	91.9	41	83.7	0.006	0.14	0.02–0.59	0.27	0.04–1.35
p39	49	39.5	11	22.4	< 0.01	0.28	0.12–0.62	0.57	0.20–1.60
p58	76	61.3	19	38.8	< 0.01	0.20	0.09–0.43	0.60	0.23–1.58
p83	39	31.5	5	10.2	< 0.01	0.14	0.04–0.36	0.36	0.10–1.17
DbpA	99	79.8	30	61.2	< 0.01	0.14	0.05–0.36	0.25	0.07–0.77
Number of antigens (7 missing values)
1	20	16.1	16	32.7	< 0.01	Reference
2	23	18.5	13	26.5		0.33	0.07–1.22	Not applicable
3	26	21.0	11	22.4		0.18	0.04–0.66
4	26	21.0	6	12.2		0.08	0.02–0.29
5	21	16.9	3	6.10		0.04	0.01–0.19
6	8	6.5	0	0.0		Not applicable
Age group (years)	n	%	n		p value	Annual rate of seroreversion		95% CI
1–2	2	1.5	1		0.10	Not applicable
3–6	29	22.1	18			5.70		5.06–8.52
7–10	32	24.4	14			4.01		2.57–5.74
11–13	29	22.1	12			3.81		1.27–3.81
14–17	39	29.8	11			2.56		0.69–2.78

aOR: adjusted odds ratio; CI: confidence interval; KiGGS: German Health Interview and Examination Survey for Children and Adolescents; OR: odds ratio.

^a To compare the proportion of seroreversions, we used the Pearson’s chi-squared test.

^b The multivariable regression model includes all antigen reactions.

Entries in bold represent statistically significant findings.