BackgroundThe NHS has been challenged to adopt new integrated models of service delivery that are tailored to local populations. Evidence from the international literature is needed to support the development and implementation of these new models of care.ObjectivesThe study aimed to carry out a systematic review of international evidence to enhance understanding of the mechanisms whereby new models of service delivery have an impact on health-care outcomes.DesignThe study combined rigorous and systematic methods for identification of literature, together with innovative methods for synthesis and presentation of findings.SettingAny setting.ParticipantsPatients receiving a health-care service and/or staff delivering services.InterventionsChanges to service delivery that increase integration and co-ordination of health and health-related services.Main outcome measuresOutcomes related to the delivery of services, including the views and perceptions of patients/service users and staff.Study designEmpirical work of a quantitative or qualitative design.Data sourcesWe searched electronic databases (between October 2016 and March 2017) for research published from 2006 onwards in databases including MEDLINE, EMBASE, PsycINFO, Cumulative Index to Nursing and Allied Health Literature, Science Citation Index, Social Science Citation Index and The Cochrane Library. We also searched relevant websites, screened reference lists and citation searched on a previous review.Review methodsThe identified evidence was synthesised in three ways. First, data from included studies were used to develop an evidence-based logic model, and a narrative summary reports the elements of the pathway. Second, we examined the strength of evidence underpinning reported outcomes and impacts using a comparative four-item rating system. Third, we developed an applicability framework to further scrutinise and characterise the evidence.ResultsWe included 267 studies in the review. The findings detail the complex pathway from new models to impacts, with evidence regarding elements of new models of integrated care, targets for change, process change, influencing factors, service-level outcomes and system-wide impacts. A number of positive outcomes were reported in the literature, with stronger evidence of perceived increased patient satisfaction and improved quality of care and access to care. There was stronger UK-only evidence of reduced outpatient appointments and waiting times. Evidence was inconsistent regarding other outcomes and system-wide impacts such as levels of activity and costs. There was an indication that new models have particular potential with patients who have complex needs.LimitationsDefining new models of integrated care is challenging, and there is the potential that our study excluded potentially relevant literature. The review was extensive, with diverse study populations and interventions that precluded the statistical summary of effectiveness.ConclusionsThere is stronger evidence that new models of integrated care may enhance patient satisfaction and perceived quality and increase access; however, the evidence regarding other outcomes is unclear. The study recommends factors to be considered during the implementation of new models.Future workLinks between elements of new models and outcomes require further study, together with research in a wider variety of populations.Study registrationThis study is registered as PROSPERO CRD37725.FundingThe National Institute for Health Research Health Services and Delivery Research programme.