Fig. 2: Biting is heterogeneous and nonrandom.

A Individual observed bite counts (color) for each member (y) of each household (x) showing between- and within-household heterogeneity. Gray indicates individuals with no observed bites. Only people present for more than 3 months during the study period were included. B The observed distribution of bites is more unequal than random chance. Lorenz curves demonstrating cumulative share of nightly bites (y) across participants (x, sorted by ascending share of bites). Dashed line indicates the null of evenly distributed bites, dotted line the null of simulated (1000x) randomly-distributed bites, and solid line the observed distribution. Overall, 20% of people received 86% of observed bites. C Estimated monthly bites per person (from observed average nightly bites; color and number) by 4 risk factors (net use, P. falciparum infection status, gender, and age). n indicates the number of people included in the group and the bottom numbers are the range. D Biting rate ratios (points) and 95% confidence intervals (lines) for covariates in risk factor analysis based on 2055 person-nights. Adjusters include high transmission season, number of STR-typed mosquitoes in the household, number of household members, whether any household member tested positive for P. falciparum by a rapid diagnostic test in the previous month, and number of people in the sleeping space. Source data are provided as a Source Data file.