Background: Loneliness is a subjective and distressing experience and a global demand for both mental and physical health, yet its prevalence and features across medical diseases remain insufficiently characterized. This systematic review aimed to synthesize evidence on the prevalence and associated features of loneliness in medically ill adults. Methods: A systematic review of studies assessing loneliness in medically ill adults was conducted via PubMed, Web of Science, EMBASE, Cochrane from inception to June 2026. Prevalence rates were pooled using random-effects generalized linear mixed models with logit transformation. Heterogeneity was assessed using Q, τ², and I² statistics. Risk of bias was assessed using the JBI critical appraisal tools. Data was also meta-analyzed in two subgroups of studies homogenous for diagnosis and for which data was sufficient (i.e., breast cancer, IHV) Findings: Up to 50% of patients with cancer, with pooled rates of 29.0% (95% CI 16.1–46.4) in patients with breast cancer and up to 60% of patients with HIV with pooled rates of 41.3% (95% CI 25.4–59.3), suffer from loneliness. Moderate prevalence rates were found among patients with rheumatological, endocrine, cardiovascular, dermatological diseases, with an overall prevalence calculated at 34.7% (95% CI 25.0–45.9). Older age, female sex, unemployment, not being in a relationship, poor social support was associated with loneliness. Clinical correlates included anxiety and depressive symptoms, distress, poorer quality of life and functioning, and greater disease severity. Interpretation: Loneliness is common across chronic medical conditions and is associated with increased morbidity. However, standardization of loneliness assessment instruments is lacking. Associated psychosocial and clinical characteristics represent potentially modifiable risk factors and may therefore provide targets for clinical care.