**2. Results**

A total of 1800 samples, 879 for the years 2016–2017 and 921 for the years 2018–2019, were tested. The median age of all subjects whose serum samples were included in the study was 51 years, 50 years (age range 20–93 years) and 51 years (age range 20–94 years) for those sampled in 2016–2017 and in 2018–2019, respectively.

Collected samples were stratified by year of collection (2016–2017 and 2018–2019) and further stratified by sex and age group (20–60 and >60 years old). IgG borderline and positive samples identified by ELISA, immunofluorescence assay (IFA) and microneutralization (MN)/plaque reduction neutralization (PRN) assays in different years of collection by age group are reported in Table 1.

Out of the 879 samples collected in 2016–2017, 17 (1.9%, 95% confidence interval (CI) 1.2–3.1) samples were ELISA IgG positive or borderline. Comparisons of ELISA IgG positive results with sex (*p* = 0.83) and age groups (*p* = 0.69) did not show any statistical significance. Twelve samples (1.4%, 95% CI 0.8–2.4) were confirmed IgG positive by IFA. IFA IgG positive results were also not statistically associated with sex (*p* = 0.85) and age groups (*p* = 0.71). Six samples were positive by MN and PRN assays, thus showing a total prevalence of 0.7% (95% CI 0.3–1.5) of samples with neutralizing antibodies. MN/PRN positive results lack any statistical significance with sex (*p* = 0.87) and age groups (*p* = 0.74).

Out of the 879 samples collected in 2016–2017, 92 were also tested by ELISA IgM. Four samples were found positive (4.35%, 95% CI 1.36–11.0), one of which was also positive for ELISA IgG, IFA IgG, and neutralizing antibodies.

Out of the 921 samples collected in 2018–2019, eight (0.9%, 95% CI 0.4–1.7) samples were positive by ELISA IgG. Comparisons of ELISA IgG positive results with sex (*p* = 0.18) and age groups (*p* = 0.71) did not yield any statistical significance. Six samples (0.65%, 95% CI 0.3–1.45) were confirmed IgG positive by IFA. IFA IgG positive results were also not statistically associated with sex (*p* = 0.23) and age group (*p* = 0.77). Five samples were positive by MN and PRN for a total prevalence of 0.5% (95% CI 0.2–1.3). MN/PRN positive results lack any statistical significance by comparison with sex (*p* = 0.26) and age groups (*p* = 0.79).

Table 2 shows a summary of results with the characteristics of subjects who showed neutralizing antibodies to WNV. The median age was 57 years (age range 30–91 years).

In the univariate logistic regression model, the independent variables, sex and age group, did not show, consistently, statistically significant associations with IgG positive and borderline results for both 2016–2017 and 2018–2019 years of collection. In the multivariate logistic regression model, the independent variables confirmed the lack of association with IgG positive or borderline results (Table 3).


1.2–3.1)

0.8–2.4)

0.3–1.5)

95% CI, 95% confidence interval; IFA, immunofluorescence assay; MN, micro neutralization; PRN, plaque reduction neutralization.

0.4–1.7)

0.3–1.45)

0.2–1.3)

**Table 1.** ELISA IgG, IFA IgG, and MN/PRN borderline and positive samples by years of collection and age group. Data on sex and age group of collected and positive


**Table 2.** Information of subjects (years of collection, age and sex) and serologic results (ELISA, IFA, MN and PRN titer) of the samples showing WNV neutralizing antibodies by MN/PRN assays.

> IFA, immunofluorescence assay; MN, micro neutralization; PRN, plaque reduction neutralization.

**Table 3.** Results for the univariate logistic regression model and the multivariate logistic regression model by independent variables (sex and age group).


OR, odd ratio; 95% CI, 95% confidence interval.
