Skip to main content
. 2013 Apr;4(2):53–62. doi: 10.1177/2042098613477445

Table 2.

Tenth percentile, 50th percentile (median), and 90th percentile of the distribution of the relative impact of changing a single parameter on the estimated relative risk.

Study design Parameter 10%ile 50%ile 90%ile
Self-controlled Time at risk 1.06 1.36 3.20
Self-controlled Include index date in time at risk 1.03 1.28 3.13
Self-controlled Outcomes to include 1.01 1.17 2.24
Self-controlled Variance of the prior 1.00 1.00 1.02
Cohort Time at risk 1.04 1.25 2.24
Cohort Covariate eligibility window 1.01 1.05 1.30
Cohort Analysis strategy 1.00 1.05 1.51
Cohort Covariates included in propensity score 1.00 1.01 1.08
Case control Number of matched controls 1.02 1.14 1.63
Case control Required observation time prior to outcome 1.01 1.05 1.18
Case control Required observation time after outcome 1.01 1.04 1.19