Purpose The EQ-5D-3L and 5L are widely used generic preference-based instruments, which are psychometrically sound with the general population, but little is known about the instruments' feasibility in the elderly. Therefore, this systematic review summarises the available literature with regard to the feasibility properties of the instruments in the elderly population. Methods We conducted a systematic search in PubMed, PsycInfo and EuroQol databases using pre-specified vocabulary and inclusion/exclusion criteria to identify publications until November 2020. Study characteristics and outcomes referring to the feasibility of the EQ-5D-3L and 5L in the elderly were extracted, if all study participants were at least 65+ years. Results We identified 17 studies reporting feasibility outcomes based on four criteria: missing values, completion rates, completion time and broad qualitative statements referring to the completion. Missing values per dimension ranged from 0 to 10.7%, although being mostly below 7%. The completion rate was around 90% or better, whereas the EQ VAS rating was missing from 2.3 to 25.3% of the respondents. Only two of the included studies examined the EQ-5D-5L; 15 studies reported on the EQ-5D-3L. Conclusion Comparing our findings against the general population from published literature, we find that feasibility outcomes in older age groups are just below that of younger populations. Furthermore, older respondents have a higher propensity of requiring assistance or even an interviewer-based approach. Nonetheless, the reviewed literature indicates that the EQ-5D-3L still has good feasibility properties and, hence, is highly applicable in older respondents. However, further research is needed to explore feasibility properties of the EQ-5D-5L in this population.