Uded separatelyPLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.July 14,8 /Rift Valley Fever Risk Factors in MadagascarTable 2. Descriptive and univariate analyses for cattle and human seroprevalences. Characteristics Cattle Age 1 to 2 3 to 4 5 to 6 >7 Cattle density per sq. km < 9.7 9.7?4.3 14.3?9.1 > 19.1 Factor 1 Factor 2 Factor 3 Factor 4 Total cattle Human Habitat INK1117 supplement Gender Contact with ruminant Contact with raw milk Contact with fresh ruminant QAW039 site fluids Profession Urban Rural F M No Yes No Yes No Yes Butcher Farmers Health Contact with environment Others Age 18 to 26 26 to 37 37 to 46 > 46 Cattle density per sq. km < 6.3 6.3?1.7 11.7?2.0 > 22.0 Factor 1 Factor 2 Factor 3 Factor 4 Total human doi:10.1371/journal.pntd.0004827.t002 Positive 46 69 72 90 69 55 37 116 / / / / 277 9 150 50 109 103 56 140 19 158 1 1 95 1 9 53 30 35 40 54 51 51 28 29 / / / / 159 Total 353 422 361 296 359 357 362 354 / / / / 1432 150 1,529 851 828 1,209 470 1,576 103 1,675 5 6 755 19 52 847 455 423 361 440 450 420 389 420 / / / / 1,679 Seroprevalence [95 CI] 13.0 [9.7?7.0] 16.4 [13.0?0.2] 19.9 [15.9?4.4] 30.4 [25.2?6.0] 19.2 [15.3?3.7] 15.4 [11.8?9.6] 10.2 [7.3?3.8] 32.8 [27.9?7.9] / / / / 15.9 [14.0?7.8] 6.0 [2.8?1.1] 9.8 [8.4?1.4] 5.9 [4.4?.7] 13.2 [10.9?5.7] 8.5 [7.0?0.2] 11.9 [9.1?5.2] 8.9 [7.5?0.4] 18.4 [11.5?7.3] 9.4 [8.1?0.9] 20 [0.1?1.6] 16.7 [0.0?4.4] 12.6 [10.3?5.2] 5.3 [0.0?6.0] 17.3 [8.2?0.3] 6.3 [4.7?.1] 6.6 [4.5?.3] 8.3 [5.8?1.3] 11.1 [8.0?4.8] 12.3 [9.4?5.7] 11.3 [8.6?4.6] 12.1 [9.2?5.7] 7.2 [4.8?0.2] 6.9 [4.7?.8] / / / / 9.5 [8.1?1.0] p >0.20 p < 0. 01 p < 0. 2 p < 0.10 / p < 0.05 p < 0.05 p < 0.005 NA p < 0.005 p < 0.05 p < 0.001 p < 0.01 p >0.20 p < 0.10 p < 0.10 / p < 0.20 p < 0.001 Chi2 p<0.from Factor 1, Factor 2 and Factor 3 in both cattle and human multivariate models. The multicollinearity test did not detect any correlation between human related factors (VIF < 1.5). For cattle, the single selected model (weight 0.99; S1 Table) included age, cattle density and Factor 4 as explanatory variables (S1 Table and Table 3). Factor 4 and age had a significant positive effect on seroprevalence (estimation of fixed effect at 1.73 and 0.17 respectively; p<0.PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.July 14,9 /Rift Valley Fever Risk Factors in MadagascarTable 3. Results from the best cattle model. Variable Intercept Age Cattle density per sq. km / / < 6.3 6.3?1.7 11.7?2.0 > 22.0 Factor 4 NS = not significant doi:10.1371/journal.pntd.0004827.t003 / Estimate -2.34 0.17 Reference -0.24 -0.66 0.97 1.73 [-1.01?.54] [-1.61?.24] [0.30?.69] [0.96?.55] 95 CI [-3.02?1.72] [0.10?.23] / NS NS p < 0.01 p < 0.001 p < 0.001 / p-value /for both explanatory variables; Table 3). Areas with high cattle density (> 19.1 per sq. km) were at risk (p<0.01; OR = 2.6 95 CI [1.3?.4]; Table 3). According to AIC, seven models were considered as suitable for describing seroprevalence in humans and thus were analyzed using a multi-model inference approach (S1 Table). These models included age, Factor 2, Factor 3, Factor 4, gender, habitat, contact with raw milk, contact with fresh ruminant product, with live ruminant as explanatory variables (S1 Table and Table 4). Age, gender (male; OR = 2.3 95 CI [1.6?.3]) and Factor 4 had a significant positive effect on seroprevalence (p<0.001, p<0.05 and p<0.05 respectively; Table 4). Factor 2 had a significant negative effect on seroprevalence (p<0.05) whereas Factor 3 had a minor importance in this set of mod.Uded separatelyPLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.July 14,8 /Rift Valley Fever Risk Factors in MadagascarTable 2. Descriptive and univariate analyses for cattle and human seroprevalences. Characteristics Cattle Age 1 to 2 3 to 4 5 to 6 >7 Cattle density per sq. km < 9.7 9.7?4.3 14.3?9.1 > 19.1 Factor 1 Factor 2 Factor 3 Factor 4 Total cattle Human Habitat Gender Contact with ruminant Contact with raw milk Contact with fresh ruminant fluids Profession Urban Rural F M No Yes No Yes No Yes Butcher Farmers Health Contact with environment Others Age 18 to 26 26 to 37 37 to 46 > 46 Cattle density per sq. km < 6.3 6.3?1.7 11.7?2.0 > 22.0 Factor 1 Factor 2 Factor 3 Factor 4 Total human doi:10.1371/journal.pntd.0004827.t002 Positive 46 69 72 90 69 55 37 116 / / / / 277 9 150 50 109 103 56 140 19 158 1 1 95 1 9 53 30 35 40 54 51 51 28 29 / / / / 159 Total 353 422 361 296 359 357 362 354 / / / / 1432 150 1,529 851 828 1,209 470 1,576 103 1,675 5 6 755 19 52 847 455 423 361 440 450 420 389 420 / / / / 1,679 Seroprevalence [95 CI] 13.0 [9.7?7.0] 16.4 [13.0?0.2] 19.9 [15.9?4.4] 30.4 [25.2?6.0] 19.2 [15.3?3.7] 15.4 [11.8?9.6] 10.2 [7.3?3.8] 32.8 [27.9?7.9] / / / / 15.9 [14.0?7.8] 6.0 [2.8?1.1] 9.8 [8.4?1.4] 5.9 [4.4?.7] 13.2 [10.9?5.7] 8.5 [7.0?0.2] 11.9 [9.1?5.2] 8.9 [7.5?0.4] 18.4 [11.5?7.3] 9.4 [8.1?0.9] 20 [0.1?1.6] 16.7 [0.0?4.4] 12.6 [10.3?5.2] 5.3 [0.0?6.0] 17.3 [8.2?0.3] 6.3 [4.7?.1] 6.6 [4.5?.3] 8.3 [5.8?1.3] 11.1 [8.0?4.8] 12.3 [9.4?5.7] 11.3 [8.6?4.6] 12.1 [9.2?5.7] 7.2 [4.8?0.2] 6.9 [4.7?.8] / / / / 9.5 [8.1?1.0] p >0.20 p < 0. 01 p < 0. 2 p < 0.10 / p < 0.05 p < 0.05 p < 0.005 NA p < 0.005 p < 0.05 p < 0.001 p < 0.01 p >0.20 p < 0.10 p < 0.10 / p < 0.20 p < 0.001 Chi2 p<0.from Factor 1, Factor 2 and Factor 3 in both cattle and human multivariate models. The multicollinearity test did not detect any correlation between human related factors (VIF < 1.5). For cattle, the single selected model (weight 0.99; S1 Table) included age, cattle density and Factor 4 as explanatory variables (S1 Table and Table 3). Factor 4 and age had a significant positive effect on seroprevalence (estimation of fixed effect at 1.73 and 0.17 respectively; p<0.PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.July 14,9 /Rift Valley Fever Risk Factors in MadagascarTable 3. Results from the best cattle model. Variable Intercept Age Cattle density per sq. km / / < 6.3 6.3?1.7 11.7?2.0 > 22.0 Factor 4 NS = not significant doi:10.1371/journal.pntd.0004827.t003 / Estimate -2.34 0.17 Reference -0.24 -0.66 0.97 1.73 [-1.01?.54] [-1.61?.24] [0.30?.69] [0.96?.55] 95 CI [-3.02?1.72] [0.10?.23] / NS NS p < 0.01 p < 0.001 p < 0.001 / p-value /for both explanatory variables; Table 3). Areas with high cattle density (> 19.1 per sq. km) were at risk (p<0.01; OR = 2.6 95 CI [1.3?.4]; Table 3). According to AIC, seven models were considered as suitable for describing seroprevalence in humans and thus were analyzed using a multi-model inference approach (S1 Table). These models included age, Factor 2, Factor 3, Factor 4, gender, habitat, contact with raw milk, contact with fresh ruminant product, with live ruminant as explanatory variables (S1 Table and Table 4). Age, gender (male; OR = 2.3 95 CI [1.6?.3]) and Factor 4 had a significant positive effect on seroprevalence (p<0.001, p<0.05 and p<0.05 respectively; Table 4). Factor 2 had a significant negative effect on seroprevalence (p<0.05) whereas Factor 3 had a minor importance in this set of mod.