[Evidence for casual household transmission of HIV: review and analysis]

All reported case-studies in which the author-described route of infection was deemed to be casual household contact were reviewed and analyzed, as well as all sample-based studies designed to estimate the frequency of casual household transmission. Several recurring biased mechanisms of interpretation were found in the studies, the use of which made it impossible to obtain evidence for casual household transmission. After correcting for these interpretations 8 documented cases of casual household transmission were found. Analysis of the sample-based studies indicated 10 cases of casual household transmission, which occurs at a risk of 0.4% per year of contact (95% confidence interval: 0.20 to 0.66).