ICP Effect of Acute Illness on Contact Patterns 2018

There are no data on changing contact patterns when ill in Africa, yet they could profoundly affect transmission of infections. We recorded named contacts (face-to-face conversations or touch) and opportunities for casual contact in a rural population in northern Malawi in three 24-hour periods: when attending clinic for an acute illness, the following day, and 2 weeks later, when well. Contact patterns for caregivers were also collected for participants aged <18 years. We interviewed 384 patients and 257 caregivers. Patients had an average of 16.8, 15.4 and 15.6 contacts on the three visits. Results for caregivers were similar. The day of the clinic visit the number of adult contacts and the proportion using taxis and minibuses were increased for all ages. Compared to the day after the clinic visit, participants seen on the same day of the week when well had a mean of 0.4 (95% CI -0.4, 1.2) extra contacts, with an increase of 0.8 (0.2, 1.3) for indoor contacts, 0.8 (0.0, 1.5) for contacts of =10 minutes, and 0.9 (0.2, 1.7) for contacts involving touch. Children tended to have more child contacts and fewer adult contacts when well, and adults tended to have more adult and fewer child contacts when well, but all differences were small (< 1 contact/day). When well participants were more likely to go to church, market and school, visit other houses and use public transport, than they were on the day after the clinic visit. Opportunities for casual contact were affected by illness more than contact with named individuals. Clinic visiting increased named and casual contacts, and being ill at home decreased such contacts. Illness increased dis-associative age mixing. These data should help refine models of infection spread.