Remotely collected physical performance measures could improve inclusion of under-served groups in clinical research as well as enabling continuation of research in pandemic conditions. It is unclear whether remote collection is feasible and acceptable to older patients, or whether results are comparable to face-to-face measures. We conducted a systematic review according to a prespecified protocol. We included studies with mean participant age ≥ 60 years, with no language restriction. Studies examining the gait speed, Short Physical Performance Battery, distance walk tests, grip strength, Tinetti score, Berg balance test, sit-to-stand test and timed up and go were included. Reports of feasibility, acceptability, correlation between remote and face-to-face assessments and absolute differences between remote and face-to-face assessments were sought. Data were synthesised using Synthesis Without Meta-analysis methodology; 30 analyses from 17 publications were included. Study size ranged from 10 to 300 participants, with a mean age ranging from 61 to >80 years. Studies included a broad range of participants and conditions. Most studies had a moderate or high risk of bias. Only two studies undertook assessment of acceptability or feasibility, reporting good results. Correlation between face-to-face and remote measures was variable across studies, with no measure showing consistently good correlation. Only nine studies examined the accuracy of remote measures; in six studies, accuracy was rated as good (<5% mean difference between face-to-face and remote measures). There is a lack of robust evidence that remote collection of physical performance measures is acceptable to patients, feasible or provides comparable results to face-to-face measures.