Table 1.

Description of the characteristics of the included studies

Source	Study type	Data source (study period)	Study location	Patients, n	Disease of admission	Study outcome	Socioeconomic variables	Follow-up period	Analytic model	Quality score
Krumholz et al 200024	Retrospective cohort study	Medicare provider analysis and review; medical records; Hospital administrative data (1994–1995)	USA	2,176	Heart failure	All-cause readmission; heart failure readmission	Race/ethnicity	6 months	Cox proportional hazards regression	82%
Rathore et al 200325	Retrospective cohort study	Medical records; Medicare administrative data; American Hospital Association’s Annual Survey (1998–1999)	USA	29,732	Heart failure	All-cause readmission	Race/ethnicity	1 year	Multivariate logistic regression	80%
Schwarz et al 200328	Prospective cohort study	Medical records; self report (not given)	USA	156	Heart failure	All-cause readmission	Education	90 days	Cox proportional hazards regression	84%
Rathore et al 200629	Retrospective cohort study	Medicare administrative data; medical records (1998–1999)	USA	25,086	Heart failure	All-cause readmission	Socioeconomic status (residence- based socioeconomic information)	1 year	Multilevel variables hierarchical model	95%
Rodríguez-Artalejo et al 200630	Prospective cohort study	Medical records; self report (not given)	Spain	371	Heart failure	All-cause readmission	Social network (married, living alone, contact with family daily, stay alone >2 hours)	6 months (on average)	Cox proportional hazards regression	75%
Roe-Prior et al 200727	Retrospective cohort study	Self report (interview); randomized controlled trial cohort (1994–2004)	USA	103	Heart failure	All-cause readmission; heart failure readmission	Marital status; income; education	90 days	Multivariate logistic regression	61%
McHugh et al 201022	Retrospective cohort study	Medicare provider analysis and review; American Hospital Association’s Annual Survey (2008)	USA	590,114	Heart failure + acute myocardial infarction	All-cause readmission	Race/ethnicity	30 days	Generalized estimating equation models	50%
Joynt et al 201114	Retrospective cohort study	Medicare provider analysis and review; American Hospital Association’s Annual Survey (2006–2008)	USA	3,163,011	Acute myocardial infarction + CHF	All-cause readmission	Race/ethnicity	30 days	Multivariate logistic regression	84%
Rodriguez et al 201121	Retrospective cohort study	Medicare provider analysis review; American Hospital Association; Hospital Quality Alliance (2006–2008)	USA	1,734,101	Heart failure + acute myocardial infarction	All-cause readmission	Race/ethnicity	30 days	Multivariate logistic regression	84%
Howie-Esquivel et al 201223	Retrospective cohort study	Medical records (2009)	USA	809	Heart failure	All-cause readmission	Race/ethnicity; marital status	90 days	Multivariate logistic regression	80%
Lindenauer et al 201326	Retrospective cohort study	Medicare administrative data; US Census current population survey and other claims files (2006–2008)	USA	558,200	Acute myocardial infarction + heart failure	All-cause readmission	Income	30 days	Multivariate logistic regression	89%

Abbreviation: CHF, congestive heart failure.