Background: Families can have considerable influence on the health of a patient with a chronic physical disease, and attempts have been made to optimise treatment by involving family members. This meta-analysis aimed to systematically assess the effects of family-oriented interventions on adults with chronic physical diseases. Methods: A systematic literature search was conducted to identify randomised clinical trials that have evaluated the effects of family-based interventions for adult patients compared to standard treatment. Interventions were classified as psychoeducation or addressing family relationships. The outcomes studied were the patient’s physical and mental health and the health of the family members. The effect sizes were calculated as Hedges’ g and combined using random-effects models. Results: Fifty-two relevant randomised controlled trials (including 8,896 patients) were extracted. The main disease groups were cardiovascular diseases including stroke, cancer and arthritis. Meta-analysis showed that family involvement resulted in significantly better health than standard treatment for all outcomes. The mean overall effect sizes were 0.32 for the patients’ physical health, 0.28 for the patients’ mental health and 0.35 for the family members’ health. Overall, the effects were not large, but they were broad, significant and stable over a long period of time. There was a tendency towards higher effects in favour of relationship-focused family interventions compared to educational interventions. Conclusions: Psychosocial interventions involving family members in the treatment of chronic physical diseases have positive effects on the health outcome for both the patient and his/her family. Further high-quality research is necessary to investigate differential effects.