TABLE 3.

Results of multivariate regression analysis for children aged 2 to 10 years

Explanatory variableResult of regression analysis for malaria outcome
Whether malaria was experiencedTime to first malaria episodeMalaria incidence
OR (95% CI)PHR (95% CI)PIRR (95% CI)P
Age (yr)0.78 (0.63-0.95)0.020.86 (0.80-0.94)<0.0010.91 (0.86-0.96)0.001
Gender2.04 (0.76-5.46)0.161.38 (0.94-2.03)0.101.03 (0.82-1.30)0.80
Ethnicity, Bambara0.23 (0.03-2.08)0.190.68 (0.34-1.34)0.261.12 (0.79-1.58)0.52
Ethnicity, Sarakole0.84 (0.08-8.96)0.891.00 (0.49-2.06)0.991.53 (1.07-2.18)0.02
Distance lived from clinic (km)0.30 (0.004-24.02)0.590.48 (0.07-3.23)0.450.77 (0.23-2.59)0.67
Bed-net usea0.70 (0.25-1.97)0.500.89 (0.59-1.35)0.590.96 (0.75-1.23)0.77
HbAS type0.23 (0.06-0.87)0.030.41 (0.21-0.79)0.010.43 (0.27-0.68)<0.001
HbAC type0.56 (0.16-1.95)0.360.81 (0.47-1.40)0.441.14 (0.80-1.63)0.47
Growth-inhibitory activityb0.50 (0.30-0.85)0.010.80 (0.63-1.02)0.070.93 (0.75-1.16)0.53
  • a Nightly bed-net use was self-reported at the end of the malaria season.

  • b For this analysis, growth-inhibitory activity was divided into deciles, and thus the regression coefficient for this variable represents the change in malaria risk with each 10% increase in growth-inhibitory activity, holding all other independent variables constant.