BACKGROUND: The number of older people is increasing worldwide and public expenditure on residential aged care facilities (ACFs) is expected to at least double, and possibly triple, by 2050. Co-ordinated and timely care in residential ACFs that reduces unnecessary hospital transfers may improve residents' health outcomes and increase satisfaction with care among ACF residents, their families and staff. These benefits may outweigh the resources needed to sustain the changes in care delivery and potentially lead to cost savings. Our systematic review comprehensively and systematically presents the available evidence of the effectiveness, safety and cost-effectiveness of alternative models of providing health care to ACF residents. OBJECTIVES: Main objective To assess the effectiveness and safety of alternative models of delivering primary or secondary health care (or both) to older adults living in ACFs. Secondary objective To assess the cost-effectiveness of the alternative models. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, five other databases and two trials registers (WHO ICTRP, ClinicalTrials.gov) on 26 October 2022, together with reference checking, citation searching and contact with study authors to identify additional studies. SELECTION CRITERIA: We included individual and cluster-randomised trials, and cost/cost-effectiveness data collected alongside eligible effectiveness studies. Eligible study participants included older people who reside in an ACF as their place of permanent abode and healthcare professionals delivering or co-ordinating the delivery of healthcare at ACFs. Eligible interventions focused on either ways of delivering primary or secondary health care (or both) or ways of co-ordinating the delivery of this care. Eligible comparators included usual care or another model of care. Primary outcomes were emergency department visits, unplanned hospital admissions and adverse effects (defined as infections, falls and pressure ulcers). Secondary outcomes included adherence to clinical guideline-recommended care, health-related quality of life of residents, mortality, resource use, access to primary or specialist healthcare services, any hospital admissions, length of hospital stay, satisfaction with the health care by residents and their families, work-related satisfaction and work-related stress of ACF staff. DATA COLLECTION AND ANALYSIS: Two review authors independently selected studies for inclusion, extracted data, and assessed risk of bias and certainty of evidence using GRADE. The primary comparison was any alternative model of care versus usual care. MAIN RESULTS: We included 40 randomised trials (21,787 participants; three studies only reported number of beds) in this review. Included trials evaluated alternative models of care aimed at either all residents of the ACF (i.e. no specific health condition; 11 studies), ACF residents with mental health conditions or behavioural problems (12 studies), ACF residents with a specific condition (e.g. residents with pressure ulcers, 13 studies) or residents requiring a specific type of care (e.g. residents after hospital discharge, four studies). Most alternative models of care focused on 'co-ordination of care' (n = 31). Three alternative models of care focused on 'who provides care' and two focused on 'where care is provided' (i.e. care provided within ACF versus outside of ACF). Four models focused on the use of information and communication technology. Usual care, the comparator in all studies, was highly heterogeneous across studies and, in most cases, was poorly reported. Most of the included trials were susceptible to some form of bias; in particular, performance (89%), reporting (66%) and detection (42%) bias. Compared to usual care, alternative models of care may make little or no difference to the proportion of residents with at least one emergency department visit (risk ratio (RR) 1.01, 95% confidence interval (CI) 0.84 to 1.20; 7 trials, 1276 participants; low-certainty evidence), but may reduce the proportion of residents with at least one unplanned hospital admission (RR 0.74, 95% CI 0.56 to 0.99, I(2) = 53%; 8 trials, 1263 participants; low-certainty evidence). We are uncertain of the effect of alternative models of care on adverse events (proportion of residents with a fall: RR 1.15, 95% CI 0.83 to 1.60, I² = 74%; 3 trials, 1061 participants; very low-certainty evidence) and adherence to guideline-recommended care (proportion of residents receiving adequate antidepressant medication: RR 5.29, 95% CI 1.08 to 26.00; 1 study, 65 participants) as the certainty of the evidence is very low. Compared to usual care, alternative models of care may have little or no effect on the health-related quality of life of ACF residents (MD -0.016, 95% CI -0.036 to 0.004; I² = 23%; 12 studies, 4016 participants; low-certainty evidence) and probably make little or no difference to the number of deaths in residents of ACFs (RR 1.03, 95% CI 0.92 to 1.16, 24 trials, 3881 participants, moderate-certainty evidence). We did not pool the cost-effectiveness or cost data as the specific costs associated with the various alternative models of care were incomparable, both across models of care as well as across settings. Based on the findings of five economic evaluations (all interventions focused on co-ordination of care), we are uncertain of the cost-effectiveness of alternative models of care compared to usual care as the certainty of the evidence is very low. AUTHORS' CONCLUSIONS: Compared to usual care, alternative models of care may make little or no difference to the number of emergency department visits but may reduce unplanned hospital admissions. We are uncertain of the effect of alternative care models on adverse events (i.e. falls, pressure ulcers, infections) and adherence to guidelines compared to usual care, as the certainty of the evidence is very low. Alternative models of care may have little or no effect on health-related quality of life and probably have no effect on mortality of ACF residents compared to usual care. Importantly, we are uncertain of the cost-effectiveness of alternative models of care due to the limited, disparate data available.