In longitudinal studies attrition may cause problems, since a selective drop out of respondents will bias research results. In mental health studies it is well known that respondents (patients) who suffer more severely from psychiatric disorders usually drop out earlier from the study. To minimize these problems, it is recommended to make efforts to keep the attrition as low as possible. In the NESDA study, the respondents who refused to visit the interview site (clinic) were either offered an interview at home or a telephone interview, thus creating a mixed-mode design. With the introduction of the mixed mode design we reduced attritions rates with about 10%. But since a reduction in attrition does not necessarily reduce bias, we will not only evaluate the response rate but also the response selectivity. From the perspective of a researcher it is interesting to know the implications of the mixed-mode design for their research results. Therefore we will investigate what the introduction of the mixed mode design means in terms of results of “an example analysis” by comparing the research results from the standard approach with those from the mixed mode approach. Finally, from the perspective of the survey organization, it is useful to predict the respondents’ choice of interview mode. Therefore we will try to model the respondents’ choice using socio-demographic variables, psychiatric characteristics and process variables such as the distance between residence and on-site interview location and evaluation of the baseline interview.