Skip to main content
. 2021 Jul 25;9(8):821. doi: 10.3390/vaccines9080821

Table 3.

Subgroup analysis of effect size, including parameters of strength of evidence, robustness, and prediction interval.

Subgroups Pooled ES 1 (95% CI 2) * No Serious Publication Bias (Pooled ES; 95% CI) ** 95% Prediction Interval Number of MAs 3 No Serious Studies’ Inconsistency (I2) 4 No Serious Impression (SE) 5 Power of Test (%) Robustness (Pooled ES; 95% CI) †
Autoimmune disorders
Endocrine 0.96 (0.92–1.00) Yes (0.97; 0.95–1.00) 0.72–1.29 120 Yes (45.6%) Yes (0.02) 90.0 Yes (0.93; 0.89–0.98)
Neurological 1.02 (0.95–1.10) Yes (0.91; 0.89–0.94) 0.65–1.60 95 Yes (47.0%) Yes (0.04) 73.0 No (1.10; 1.02–1.20)
Gastrointestinal 1.03 (0.93–1.13) No (1.06; 1.00–1.12) 0.63–1.66 41 No (50.9%) Yes (0.05) 40.3 Yes (1.05; 0.96–1.14)
Hematologic 1.12 (0.91–1.36) Yes (1.09; 0.97–1.24) 0.43–2.89 27 No (52.4%) No (0.10) 5.2 No (1.22; 1.02–1.47)
Dermatological 0.98 (0.85–1.13) Yes (0.98; 0.87–1.11) 0.68–1.41 10 Yes (7.9%) Yes (0.07) 17.5 Yes (1.28; 0.86–1.90)
Connective tissue 0.98 (0.93–1.04) Yes (0.98; 0.93–1.04) 0.87–1.10 47 Yes (<0.1%) Yes (0.03) 5.0 Yes (1.07; 0.97–1.19)
Vasculitis 1.09 (0.91–1.29) Yes (1.08; 0.94–1.25) 0.60–1.96 20 Yes (22.6%) Yes (0.09) 5.3 Yes (1.12; 0.96–1.31)
Type of vaccine
Killed 1.00 (0.97–1.03) Yes (1.00; 0.98–1.02) 0.73–1.36 251 Yes (38.0%) Yes (0.02) >99.9 No (1.07; 1.01–1.12)
Live 0.99 (0.92–1.07) Yes (0.94; 0.90–0.98) 0.60–1.62 72 No (54.4%) Yes (0.04) 90.5 Yes (1.00; 0.93–1.07)
Unspecific 1.03 (0.93–1.15) Yes (0.90; 0.87–0.93) 0.73–1.47 19 Yes (41.4%) Yes (0.06) 41.2 Yes (1.09; 0.95–1.25)
Lipid-adjuvants 0.94 (0.83–1.07) Yes (0.95; 0.85–1.06) 0.63–1.41 22 Yes (14.5%) Yes (0.07) 68.3 Yes (0.96; 0.86–1.06)
Study methodology
Case–control 0.96 (0.91–0.99) Yes (0.91; 0.89–0.93) 0.74–1.22 152 Yes (21.6%) Yes (0.02) >99.9 Yes (1.03; 0.98–1.09)
Cohort 1.00 (0.95–1.04) Yes (0.98; 0.95–1.00) 0.67–1.47 176 Yes (49.4%) Yes (0.02) >99.9 Yes (1.05; 0.97–1.13)
Modified 1.08 (0.99–1.16) Yes (1.03; 0.99–1.07) 0.82–1.41 36 Yes (26.9%) Yes (0.04) 11.6 No (1.11; 1.01–1.22)
Measures of association
Crude 0.97 (0.90–1.05) Yes (0.97; 0.92–1.02) 0.60–1.56 94 Yes (35.7%) Yes (0.04) 96.4 Yes (1.07; 0.98–1.18)
Adjusted 1.00 (0.97–1.03) Yes (0.96; 0.95–0.98) 0.74–1.36 270 Yes (46.7%) Yes (0.02) >99.9 Yes (1.04; 0.99–1.08)
Age of study population
<18 years 1.02 (0.98–1.06) Yes (1.02; 0.99–1.04) 0.78–1.32 184 Yes (27.3%) Yes (0.02) 99.9 Yes (1.00; 0.95–1.04)
>18 years 0.99 (0.94–1.05) Yes (0.96; 0.93–1.00) 0.65–1.50 91 No (56.3%) Yes (0.03) 98.7 Yes (1.06; 0.99–1.13)
General population 0.95 (0.90–1.02) Yes (0.90 (0.88–0.93) 0.65–1.39 89 Yes (42.3%) Yes (0.03) >99.9 Yes (1.09; 0.99–1.21)
Study locality
Asia 1.09 (0.94–1.26) Yes (1.09; 0.94–1.26) 0.80–1.48 16 Yes (<0.1%) Yes (0.08) 5.0 Yes (1.28; 0.84–1.94)
Australia 1.10 (0.86–1.41) Yes (1.10; 0.86–1.41) 0.63–1.95 6 Yes (<0.1%) No (0.13) 5.0 Yes (0.92; 0.77–1.09)
Europe 1.01 (0.98–1.04) Yes (0.97; 0.95–0.99) 0.76–1.33 229 Yes (41.2%) Yes (0.02) >99.9 Yes (1.02; 0.98–1.06)
North America 0.96 (0.90–1.02) Yes (0.92; 0.89–0.96) 0.58–1.57 113 No (53.0%) Yes (0.03) 99.9 Yes (1.07; 0.99–1.15)
Year of publication
1990 or older 1.04 (0.83–1.31) Yes (1.04; 0.83–1.31) 0.62–1.75 7 Yes (<0.1%) No (0.12) 5.0 Yes (0.88; 0.75–1.03)
1991–2000 0.98 (0.91–1.06) Yes (0.98; 0.92–1.05) 0.77–1.25 43 Yes (9.6%) Yes (0.04) 87.5 Yes (0.93; 0.86–1.01)
2001–2010 1.00 (0.93–1.08) Yes (0.99; 0.94–1.04) 0.63–1.58 73 Yes (44.5%) Yes (0.04) 82.5 Yes (1.04; 0.97–1.11)
2011 or newer 1.00 (0.96–1.03) Yes (0.96; 0.94–0.98) 0.72–1.37 241 Yes (48.5%) Yes (0.02) >99.9 No (1.09; 1.03–1.15)

1 ES: effect size, 2 CI: confidence interval; 3 MAs: measures of association; 4 I2: index of inconsistency; 5 SE: standard deviation; *: pooled effect size derived from the random-effects model including 95% confidence interval; **: pooled effect size derived from the fixed-effects model including 95% confidence interval; and †: pooled effect size derived from the random-effects model including 95% confidence interval (studies of any risk of bias).