females; subsequent models were compared to this model using change in log likelihood relative to change in degrees of freedom and Akaike's Information Criterion (AIC). A model that constrained male and female parameters to be equal did not provide a good fit to the data and so parameters were estimated separately by gender in subsequent models. Next, models were estimated that tested the following hypotheses regarding sources of variance in risk for AUD and remission: (i) no environmental factors shared by twins influence risk for AUD or remission (AE model for AUD and remission), (ii) influences on remission are distinct from influences on AUD (i.e. the phenotypes share no genetic or environmental variance), (iii) there are no genetic influences on remission (CE model for remission), and (iv) there are no genetic or shared environmental influences on remission (E model for remission). The fit statistics from these models were used to identify the best-fitting models for males and females in the final model. All models were adjusted for age using a binary variable representing individuals aged ≥30 years because these individuals were more likely to be remitted from two or more [<30 years: 35.5%; ≥30 years: 44.0%; odds ratio (OR) 1.4,
