Trials that have a more heterogeneous study population are more generalisable to clinical practice and potentially have more external validi