BackgroundIncreasingly, cancer is recognised as a chronic condition with a growing population of informal caregivers providing care for cancer patients. Informal caregiving can negatively affect the health and well‐being of caregivers. We need a synthesised account of best evidence to aid decision‐making about effective ways to support caregivers for individuals 'living with cancer'.ObjectivesTo assess the effectiveness of psychosocial interventions designed to improve the quality of life (QoL), physical health and well‐being of informal caregivers of people living with cancer compared with usual care.Search methodsWe searched CENTRAL, MEDLINE, Embase, PsycINFO, ProQuest, Open SIGLE, Web of Science from inception up to January 2018, trial registries and citation lists of included studies.Selection criteriaWe included randomised and quasi‐randomised controlled trials comparing psychosocial interventions delivered to adult informal caregivers of adults affected by cancer on a group or individual basis with usual care. Psychosocial interventions included non‐pharmacological interventions that involved an interpersonal relationship between caregivers and healthcare professionals. We included interventions delivered also to caregiver‐patient dyads. Interventions delivered to caregivers of individuals receiving palliative or inpatient care were excluded. Our primary outcome was caregiver QoL. Secondary outcomes included patient QoL, caregiver and patient depression, anxiety, psychological distress, physical health status and intervention satisfaction and adverse effects.Data collection and analysisPairs of review authors independently screened studies for eligibility, extracted data and conducted 'Risk of bias' assessments. We synthesised findings using meta‐analysis, where possible, and reported remaining results in a narrative synthesis.Main resultsNineteen trials (n = 3, 725) were included in the review. All trials were reported in English and were undertaken in high‐income countries. Trials targeted caregivers of patients affected by a number of cancers spanning newly diagnosed patients, patients awaiting treatment, patients who were being treated currently and individuals post‐treatment. Most trials delivered interventions to caregiver‐patient dyads (predominantly spousal dyads) and there was variation in intervention delivery to groups or individual participants. There was much heterogeneity across interventions though the majority were defined as psycho‐educational. All trials were rated as being at 'high risk of bias'.Compared to usual care, psychosocial interventions may improve slightly caregiver QoL immediately post intervention (standardised mean difference (SMD) 0.29, 95% confidence interval (CI) 0.04 to 0.53; studies = 2, 265 participants) and may have little to no effect on caregiver QoL at 12 months (SMD 0.14, 95% CI ‐ 0.11 to 0.40; studies = 2, 239 participants) post‐intervention (both low‐quality evidence).Psychosocial interventions probably have little to no effect on caregiver depression immediately to one‐month post‐intervention (SMD 0.01, 95% CI ‐0.14 to 0.15; studies = 9, 702 participants) (moderate‐quality evidence). Psychosocial interventions may have little to no effect on caregiver anxiety immediately post‐intervention (SMD ‐0.12, 95 % CI ‐0.33 to 0.10; studies = 5, 329 participants), depression three‐to‐six months (SMD 0.03, 95% CI ‐0.33 to 0.38; studies = 5. 379 participants) post‐intervention and patient QoL six to 12 months (SMD ‐0.05, 95% CI ‐0.37 to 0.26; studies = 3, 294 participants) post‐intervention (all low‐quality evidence). There was uncertainty whether psychosocial interventions improve patient QoL immediately (SMD ‐0.03, 95 %CI ‐0.50 to 0.44; studies = 2, 292 participants) or caregiver anxiety three‐to‐six months (SMD‐0.25, 95% CI ‐0.64 to 0.13; studies = 4, 272 participants) post‐intervention (both very low‐quality evidence). Two studies which could not be pooled in a meta‐analysis for caregiver physical health status found little to no effect immediately post‐intervention and a small intervention effect 12 months post‐intervention. Caregiver or patient satisfaction or cost‐effectiveness of interventions were not assessed in any studies. Interventions demonstrated good feasibility and acceptability.Psychosocial interventions probably have little to no effect on patient physical health status immediately post‐intervention (SMD 0.17, 95 % CI ‐0.07 to 0.41; studies = 4, 461 participants) and patient depression three to six months post‐intervention (SMD‐0.11, 95% CI ‐0.33 to 0.12; studies = 6, 534 participants) (both moderate‐quality evidence).Psychosocial interventions may have little to no effect on caregiver psychological distress immediately to one‐month (SMD ‐0.08, 95% CI ‐0.42 to 0.26; studies = 3, 134 participants), and seven to 12 months (SMD 0.08, 95% CI ‐0.42 to 0.58; studies = 2, 62 participants) post‐intervention; patient depression immediately (SMD ‐0.12, 95% CI ‐0.31 to 0.07; studies = 9, 852 participants); anxiety immediately (SMD ‐0.13, 95% CI ‐0.41 to 0.15;studies = 4, 422 participants), and three to six months (SMD ‐0.22, 95% CI ‐0.45 to 0.02; studies = 4, 370 participants); psychological distress immediately (SMD ‐0.02, 95% CI ‐0.47 to 0.44; studies = 2, 74 participants) and seven to 12 months (SMD ‐0.27, 95% CI ‐0.78 to 0.24; studies = 2, 61 participants); and physical health status six to 12 months (SMD 0.06, 95% CI ‐0.18 to 0.30; studies = 2, 275 participants) post‐intervention (all low‐quality evidence).Three trials reported adverse effects associated with the interventions, compared with usual care, including higher distress, sexual function‐related distress and lower relationship satisfaction levels for caregivers, higher distress levels for patients, and that some content was perceived as insensitive to some participants.Trials not able to be pooled in a meta‐analysis did not tend to report effect size and it was difficult to discern intervention effectiveness. Variable intervention effects were reported for patient and caregiver outcomes.Authors' conclusionsHeterogeneity across studies makes it difficult to draw firm conclusions regarding the effectiveness of psychosocial interventions for this population. There is an immediate need for rigorous trials with process evaluations and clearer, detailed intervention descriptions. Cost‐effectiveness studies should be conducted alongside future trials.