Vaccine - (Page 6) M. Oviedo et al. / Vaccine 26 (2008) 6157–6164 Table 2 Estimated values for the negative binomial regression model fitted Modela Estimated model (model BN3) Estimate Intercept Sin Cos Year <12 years 12–18 years 19–39 years Vac Year × vac a b c d d 6161 Vaccination model (model BN5)b Estimate 76.99 0.21 0.27 −0.04 2.66 2.22 2.14 −1312 0.65 d Vaccination model (model BN6)c Estimated 76.62 0.21 0.27 −0.04 2.63 2.18 2.10 −1324 0.66 p <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.03 0.03 p <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.04 0.04 p <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.03 0.03 79.70 0.21 0.27 −0.04 2.66 2.22 2.14 −1053 0.53 All models are adjusted for year, age group, coverage vaccination, interaction year with coverage vaccination, cycling component and population as offset parameter. For the sensitivity analysis, vaccination coverage calculated in function of the population of each age group and year was used. For the sensitivity analysis, the variable coverage vaccination plus additional information on the vaccination coverage of risk groups and vaccinated children was used. Coefficients. Table 3 Prediction model for 2006 Ratea 12–18 years Observed rate Predicted by model BN0 Predicted by model BN1 Predicted by model BN2 Predicted by model BN3 Predicted by model BN4 a (95% CI) 12–18 years – (1.06–3.69) (1.26–3.17) (0.20–3.17) (0.29–2.67) (0–5.32) Global ratea 3.56 2.37 2.24 2.52 2.66 2.92 (95% CI) Global rate – (1.12–3.63) (1.32–3.16) (1.78–3.27) (1.88–3.44) (1.47–4.37) 1.75 2.37 2.22 0.53 1.48 1.77 Incidence rate per 100,000 person/years. Table 4 Sensitivity analyses Modela Yearly mean Incidence reduction by Year Model BN4 Model BN5b Model BN6b Model BN4.iterativec Mean Range 181.59 181.88 181.61 176.23 (168.60;195.59) −4.99 −4.99 −5.01 −4.81 (−4.16;−5.71) Vac −5.55 −5.43 −5.53 −5.36 (−4.82;−5.82) Year × vac 2.64 2.92 2.67 2.47 (0.78;2.87) Number of estimated cases according to the adjusted model. a All models are adjusted for year, age group, coverage vaccination, interaction year with coverage vaccination and population as offset parameter. b These models do not include the estimation of the cyclic component. c For Model BN4.iterative, the reported data for one year were extracted and the process was repeated for each year from 1992 to 2005 and the mean and range of the estimated coefficients were calculated. the estimated coefficients indicate that there was a reduction in the incidence over time (year) (p-value < 0.01), and that the introduction of vaccination (vac) was also significant (p-value = 0.04). The interaction between year of report and vaccination coverage (year × vac) was positively significant (p-value = 0.04), suggesting that the effect of vaccination was more important at the beginning (1999), than at the end (2005) of the study. The standardized residuals for model BN3 were independent among themselves and no pattern of correlation between them was observed; they are centred to zero and the variance was stable. Only two outliers were identified, corresponding to the 19–39 years and ≥40 years age groups in 1998, which corresponded to a foodborne outbreak [23]. The predictions provided by the fitted models were compared with the real cases (see Table 3). The predicted global rates were: Model BN0 2.4 (95% CI: 1.1–3.6), model BN1 1.3 (95% CI: 1.3–3.2), model BN2 2.5 (95% CI: 1.8–3.3), model BN3 2.7 (95% CI: 1.9.3–3.4) and model BN4 2.9 (95% CI: 1.5–4.4), while the observed rate (provisional data for 2006) was 3.6, a value contained in the CI predicted by models BN0 and BN4 which did not include the cycling component. 4.1. Sensitivity analysis Model BN4 was robust against different modifications in the vaccination coverage introduced in the BN5 and BN6 models (see Table 4). Model BN5, which used lower levels in the percentages of vaccination coverage, but the same covariates as model BN4, showed no significant differences in the estimated incidence levels. Similar results to model BN4 were reached with a similar statistical significance. Neither did model BN6, which incorporated the percentages of the risk groups vaccinated in 1995 and vaccinated children aged <3 years, show statistically significant differences. Table 4 shows the robustness of the model BN4 over time. The interactive model estimated a mean annual incidence of 176.23 cases/year (range 168.60–195.59 cases/year), and an estimated natural reduction in incidence of 4.81 cases/year (range
Table of Contents Feed for the Digital Edition of Vaccine Vaccine Vaccine - (Page Cover1) Vaccine - (Page 2) Vaccine - (Page 3) Vaccine - (Page 4) Vaccine - (Page 5) Vaccine - (Page 6) Vaccine - (Page 7) Vaccine - (Page 8) Vaccine - (Page 9)
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.