Unpaid caregivers often experience sleep impairments as an unintended consequence of providing care. This systematic review and meta-analysis investigated the efficacy of interventions to improve sleep in unpaid caregivers. Six databases were searched from journal inception to 7-Sep-2021 to identify randomised controlled trials. Random-effects meta-analyses estimated mean differences (MD) at end-of-intervention. Twenty-one studies were identified (15 eligible for meta-analysis). Compared to control, interventions improved sleep quality (Pittsburgh Sleep Quality Index; 12 studies, 1153 participants, MD = −1.66, 95% CI [−2.91, −0.41], p = 0.009, I² = 90.51%, GRADE = low), and total sleep time (hours; two studies, 122 participants, MD = 0.45, 95% CI [0.42, 0.48], p = 0.003, I² = 0.00%, GRADE = low), but not sleeping problems (sleep item on Symptom Distress Scale of the Omega Screening Questionnaire; two studies, 529 participants, MD = −0.06, 95% CI [−0.69, 0.58], p = 0.458, I² = 0.01%, GRADE = low). There is low quality evidence that interventions improve sleep quality in unpaid caregivers compared to control. Limitations include the lack of evidence for specific intervention modes and limited use of objective sleep measures. Future research should explore potential effect modifiers such as care recipient condition (CRD42021278670).