Instruments Measuring Integrated Care: A Systematic Review of Measurement Properties

• Published in: The Milbank quarterly Vol. 94; no. 4; pp. 862 - 917
• Main Authors: BAUTISTA, MARY ANN C., NURJONO, MILAWATY, LIM, YEE WEI, DESSERS, EZRA, VRIJHOEF, HUBERTUS JM
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01-12-2016; Milbank Memorial Fund; John Wiley and Sons Inc
• Subjects: Adequacy; Case management; Chronic diseases; Consistency; Construct validity; Content validity; Continuity of Patient Care; Coordination; Criteria; Cultural property; Delivery of Health Care, Integrated - standards; Disease models; Domains; Error analysis; Functional integration; Health care; Health care industry; Health care policy; Health outcomes; instrument validation; integrated care; Integrated services; Integration; Managed care; Management; Measurement; Measurement errors; measurement instruments; Measurement methods; Measuring instruments; Medical personnel; Patient-centered care; Patients; Primary health care; Properties (attributes); Quality; Quality assessment; Quality Indicators, Health Care - standards; quality of measurement properties; Questionnaires; Responsiveness; Review; Review Article; Search strategies; Selection criteria; Synthesis; Systematic review; Validation studies; Validity; instrument validation; quality of measurement properties; integrated care; systematic review; measurement instruments
• ISSN: 0887-378X; 1468-0009
• DOI: 10.1111/1468-0009.12233

Context: Integrated care is an important strategy for increasing health system performance. Despite its growing significance, detailed evidence on the measurement properties of integrated care instruments remains vague and limited. Our systematic review aims to provide evidence on the state of the art in measuring integrated care. Methods: Our comprehensive systematic review framework builds on the Rainbow Model for Integrated Care (RMIC). We searched MEDLINE/PubMed for published articles on the measurement properties of instruments measuring integrated care and identified eligible articles using a standard set of selection criteria. We assessed the methodological quality of every validation study reported using the COSMIN checklist and extracted data on study and instrument characteristics. We also evaluated the measurement properties of each examined instrument per validation study and provided a best evidence synthesis on the adequacy of measurement properties of the index instruments. Findings: From the 300 eligible articles, we assessed the methodological quality of 379 validation studies from which we identified 209 index instruments measuring integrated care constructs. The majority of studies reported on instruments measuring constructs related to care integration (33%) and patientcentered care (49%); fewer studies measured care continuity/comprehensive care (15%) and care coordination/case management (3%). We mapped 84% of the measured constructs to the clinical integration domain of the RMIC, with fewer constructs related to the domains of professional (3.7%), organizational (3.4%), and functional (0.5%) integration. Only 8% of the instruments were mapped to a combination of domains; none were mapped exclusively to the system or normative integration domains. The majority of instruments were administered to either patients (60%) or health care providers (20%). Of the measurement properties, responsiveness (4%), measurement error (7%), and criterion (12%) and cross-cultural validity (14%) were less commonly reported. We found <50% of the validation studies to be of good or excellent quality for any of the measurement properties. Only a minority of index instruments showed strong evidence of positive findings for internal consistency (15%), content validity (19%), and structural validity (7%); with moderate evidence of positive findings for internal consistency (14%) and construct validity (14%). Conclusions: Our results suggest that the quality of measurement properties of instruments measuring integrated care is in need of improvement with the less-studied constructs and domains to become part of newly developed instruments.