The objective of this study was to identify and appraise evidence on the direct and indirect impacts of high indoor temperatures on health; the indoor temperature threshold at which the identified health impacts are observed; and to summarise the evidence for establishing a maximum indoor temperature threshold for health. This is a systematic literature review and narrative synthesis. A review of the published literature using MEDLINE, EMBASE, Global Health, PsycINFO, Maternity and Infant Care, Cochrane Library, CINAHL and GreenFILE databases was conducted. The search criteria were kept broad to capture evidence from all countries and contexts; no date or study design limits were applied, except English language limits. We included studies that specifically measured indoor temperature and examined its effect on physical or mental health outcomes. Evidence was graded using the National Institutes of Health framework. Twenty-two articles were included in the review, including 11 observational, seven cross-sectional and three longitudinal cohort studies and one prospective case-control study. Eight main health effects were described: respiratory, blood pressure, core temperature, blood glucose, mental health and cognition, heat-health symptoms, physical functioning and influenza transmission. Five studies found respiratory symptoms worsened in warm indoor environments, with one reporting indoor temperatures higher than 26 °C, which was associated with increased respiratory distress calls being made to paramedics (odds ratio = 1.63, P = 0.056). Core symptoms of schizophrenia and dementia were found to be significantly exacerbated by indoor heat (the latter above a 26 °C cumulative exposure threshold). The absorption of insulin doses in people with type one diabetes was also significantly accelerated in hot indoor environments. Only five studies reported the temperatures at which health outcomes worsened, with thresholds ranging between 26 °C and 32 °C. However, owing to insufficient data and the heterogeneity of the included studies (design, population, setting, exposure measures, outcomes and location), meta-analysis and an upper threshold determination was not feasible. High indoor temperatures affect aspects of human health, with the strongest evidence for respiratory health, diabetes management and core schizophrenia and dementia symptoms. Exacerbation of symptoms in warm indoor environments has clinical relevance to at-risk groups and those caring for them. Care staff and facility managers need to be vigilant of high temperatures in care environments and should incorporate indoor overheating into their risk management and sustainability and/or climate change adaptation plans. The indoor temperature threshold at which adverse effects begin to occur remains unclear as studies seldom report the exposure-response relationship over a temperature continuum. Until there is extensive scientific data to support a maximum indoor temperature threshold, 26 °C may be the most suitable indoor temperature for at-risk groups in keeping with the existing guidance documents. Crown Copyright © 2019. Published by Elsevier Ltd. All rights reserved.