Background:  Missing data are a challenge in any research setting. In hospice/palliative care, this can be a particular challenge because of attrition and variable ability to engage with the study. Although statistical methods can accommodate missing data, their validity depends on often untestable assumptions about why the data are missing. Experts recommend making every effort to minimise missing data in the first place.

Aim: To explore recommendations to reduce missing data in hospice/palliative care research.

Method: The Australian national Palliative Care Clinical Studies Collaborative conducted an online workshop on the design (of what) in hospice/palliative care to inform future studies. These deliberations included palliative care experts, biostatisticians, methodologists, and clinical trials collaborative members (which has randomised more than 2200 people to multi-site randomised clinical trials). Practical recommendations on minimising missing data in hospice/palliative care studies resulted from the discussion.

Result: The importance of taking active steps at the design and data‐collection stages to minimise missing data was seen as the key. Issues discussed centred on minimising the burden and inconvenience of data collection on the participants. Discussion included shortening the follow-up period for the primary outcome (although other data collection can continue after this point) in keeping with the hypothesis and available evidence, providing more than one mode of data collection (eg. face-to-face, telephone, and electronic) and minimising the amount of data collected by collecting only data with specific analytic plans. Other issues include the appropriate timing of endpoint assessments and careful selection of outcome measures avoiding those that are too long or difficult to measure. Training study staff in the importance of completing data collection and providing feedback is also deemed critical in minimising missing data.

Conclusion: Good design and conduct can help to minimise missing data. Whilst inevitable there are opportunities to reduce missing data.