One of the primary goals of household studies of infectious disease transmission is to estimate the household secondary attack rate (SAR), the probability of direct transmission from an index case A to a susceptible household member B during A's infectious period. In a household with m susceptibles and a single index case, the number of secondary infections is often treated as a binomial ( m, p) random variable where p is the SAR. This assumes that all subsequent infections in the household are transmitted directly from the index case. Because a given transmission chain of length k from A to B has probability p k, it is thought that chains of length k > 1 can be ignored when p is small. However, the number of transmission chains of length k from A to B can be large, so the total risk of infection through any chain of length k can be much greater than pk. In simulations, we show that estimation of the SAR using a binomial model is biased upward and produces confidence intervals with poor coverage probabilities. Accurate point and interval estimates of the SAR can be obtained using chain binomial models or pairwise survival analysis. To extend the method of analyzing household studies infectious disease transmission data we proposed accelerated failure time models with covariates. Kenah [19] showed that parametric survival analysis could be used to handle dependent happenings in infectious disease transmission data by taking ordered pairs of individuals, not individuals, as the units of analysis. The failure time in this approach is the contact interval, which is the time from the onset of infectiousness in an individual i to infectious contact from i to individual j, where an infectious contact is sufficient to infect j if he or she is susceptible. The contact interval distribution shows how infectiousness changes over time in infected individuals and can be used to calculate the transmission probability as a function of time exposed. These methods assumed the same contact interval distribution in all infectious-susceptible pairs. However, many important questions in infectious disease epidemiology involve the effects of covariates (e.g., age or vaccination status) on the risk of transmission. Here, we generalize pairwise survival analysis in two ways: First, we introduce a pairwise accelerated failure time model in which the rate parameter of the contact interval distribution depends on covariates associated with infectiousness in i and susceptibility in j. Second, we show how internal infections (between individuals under observation) and external infections (from environmental or community sources outside the observed population) can be handled simultaneously. In a series of simulations, we show that these methods produce valid point and interval estimates and that the ability to account for external infections can be critical to consistent estimation. Finally, we use these methods to analyze household surveillance data from Los Angeles County during the 2009 influenza A(H1N1) pandemic.