Abstract Background Domestic and partner violence have considerable physical and mental consequences for the victims. The Belgian Health Interview Survey (HIS) 2013 previously evaluated a combination of two (indirect) related questions as an appropriate tool to assess domestic and partner violence in the general population. A more direct approach in the HIS 2018 will be compared with the HIS 2013 deduction method. Methods The HIS is a cross-sectional survey in a representative population sample. The more sensitive subjects like domestic and partner violence are part of the self-administered questionnaire. In 2013 the indicator related to domestic violence was derived from the question whether the perpetrators was a member of the household; for the indicator on partner violence also the household composition was taken into account. For the HIS 2018, the question has been adapted. It was specifically asked who the perpetrator was (i.e. unknown, colleague, acquaintance, friend, (ex-)partner, parents, children) from where easily these two indicators could be defined. Proportions are compared. Results In 2018, 1.0% of the population aged 15+ indicated being victim of domestic violence in the past 12 months, which is in line with the proportion found in 2013 (1.1%). However, for partner violence there is a difference in the proportions. In 2018, 0.6% of the population 18-74 years indicated being victim of partner violence in the past 12 months, which is only half of the proportion found in 2013 (1.3%). Conclusions For assessing domestic violence, defining perpetrators as members of the household or detailed questioning about the relationship towards the perpetrator produce similar results. For distinguishing partner violence, the estimations benefit from directly defining the relationship towards the perpetrator. Key messages Even when investigating sensitive topics, straightforward questions produce more accurate results. The problem of domestic violence and partner violence remains underreported and so it needs to ...