Skip to main content
. Author manuscript; available in PMC: 2020 Apr 1.
Published in final edited form as: Am J Epidemiol. 2019 Apr 1;188(4):709–723. doi: 10.1093/aje/kwy265

Table 5.

Empirical Example 2: Results for Binary Outcomes from Propensity Score-Adjusted Analyses using Different Combinations of Confounding Adjustment Method and Data-Sharing Approach, Non-TNFi vs TNFia

Confounding adjustment method & data-sharing approach Effectiveness outcome Safety outcome
Treatment switchingb Serious infectionsc
OR 95% CI OR 95% CI
Stratification
 Pooled individual-level 0.52 0.34, 0.76 0.97 0.46, 1.86
 Risk-set 0.52 0.34, 0.76 0.97 0.46, 1.86
 Summary-table 0.52 0.34, 0.76 0.97 0.46, 1.86
 Effect-estimate, fixed-effect 0.54 0.36, 0.79 1.06 0.56, 2.03
 Effect-estimate, random-effects 0.53 0.32, 0.86 1.03 0.56, 2.03
 Measure of heterogeneity, Qd 2.23 0.3268 0.78 0.6741
Matching
 Pooled individual-level 0.47 0.30, 0.73 1.07 0.46, 2.37
 Risk-set 0.47 0.30, 0.73 1.07 0.46, 2.37
 Summary-table 0.47 0.30, 0.73 1.07 0.46, 2.37
 Effect-estimate, fixed-effect 0.47 0.31, 0.72 1.23 0.58, 2.64
 Effect-estimate, random-effects 0.47 0.31, 0.72 1.23 0.58, 2.64
 Measure of heterogeneity, Qd 0.55 0.7589 0.03 0.9820
Inverse probability weighting
 Pooled individual-level 0.36 0.23, 0.57 2.88 1.97, 4.21
 Risk-set 0.36 0.23, 0.57 3.11 2.12, 4.55
 Effect-estimate, fixed-effect 0.39 0.25, 0.61 3.17 2.16, 4.65
 Effect-estimate, random-effects 0.45 0.18, 1.09 3.17 2.16, 4.65
 Measure of heterogeneity, Qd 4.62 0.0992 0.85 0.6531
Matching weighting
 Pooled individual-level 0.51 0.32, 0.81 0.89 0.39, 2.04
 Risk-set 0.51 0.32, 0.81 0.95 0.41, 2.19
 Effect-estimate, fixed-effect 0.51 0.32, 0.82 0.93 0.40, 2.17
 Effect-estimate, random-effects 0.51 0.32, 0.82 0.93 0.40, 2.17
 Measure of heterogeneity, Qd 1.30 0.5213 0.42 0.8105

Note: TNFi=tumor necrosis factor-alpha inhibitor; OR= odds ratio; CI= confidence interval

a

There were 407 (5.2%) patients who initiated non-TNFi, and 7,419 (94.8%) patients who initiated TNFi.

b

The incidences for treatment switching were 7.6% in new users of non-TNFi and 11.2% in new users of TNFi.

c

The incidences for serious infections were 2.9% in new users of non-TNFi and 3.1% in new users of TNFi.

d

Q is a measure of heterogeneity among the three data-contributing sites. The summary statistic and p-value from Cochran’s Q test are shown here.