Skip to main content

Table 1 Main characteristics of the reviewed studies

From: Do health economic evaluations using observational data provide reliable assessment of treatment effects?

Type of study
Cost-effectiveness analysis (70%) Cost-Utility Analysis (30%)    
Type of journal
Statistics/Econometrics (7%) Health Economics (19%) Health Services (30%) Medical (39%) Working Paper (5%)
Year of publication
1990 – 2000 (14%) 2001 – 2010 (86%)    
Type of intervention
Surgical (37%) Medical (33%) Rehabilitation (5%) Public Health Policy (14%)  
Diagnostic (2%) Preventative (9%)    
Number of interventions
Two (74%) Three or more (26%)    
Sample size
100 – 1000 (33%) 1001 – 5000 (29%) 5001+ (33%) Not reported (5%)  
Summary outcome
None (35%) Incremental Cost-Effectiveness Ratio (51%) Net Benefit (14%)   
Evaluation of uncertainty on summary outcome
Yes (43%) No (6%) Partial/Unclear (51%)   
Method addressing selection bias
Regression Analysis (28%) Covariate Matching (7%) Propensity Scores (49%) Instrumental Variables (7%) Difference-in-Differences (9%)
Assessment of methods’ assumptions
Detailed (9%) Partial or None (91%)    
Comparison of analytical methods
Yes (23%) No (77%)    
Effort to contrast findings with other studies
Yes (47%) No (53%)