Skip to main content
. 2019 Nov 1;7:e1541–e1552. doi: 10.1016/S2214-109X(19)30412-7

Table 2.

Pooled random effects meta-analysis of the relative risk of intussusception in the periods 1–7 days and 8–21 days after doses 1 and 2 of rotavirus vaccination

w Vaccine After dose 1 After dose 2
1–7 days 8–21 days 1–7 days 8–21 days
Relative risk Lower 95% uncertainty interval limit Upper 95% uncertainty interval limit Relative risk Lower 95% uncertainty interval limit Upper 95% uncertainty interval limit Relative risk Lower 95% uncertainty interval limit Upper 95% uncertainty interval limit Relative risk Lower 95% uncertainty interval limit Upper 95% uncertainty interval limit
Relative risk per study
Yung et al25 (Singapore)   2·6 Rotarix   8·36 2·42 28·96   1·72* 0·51   5·88 3·09 0·41 23·37   1·54 0·20 11·69
Perez-Vilar et al26 (Spain)   3·5 Rotarix or RotaTeq   4·70 0·30 74·10   0·80 0·10 13·90 1·60 0·10 32·30   3·90 0·30 44·00
Carlin et al21 (Australia)   4·6 Rotarix   6·76 2·40 19·01   3·45 1·33   8·94 2·84 1·10   7·34   2·11 0·97   4·62
Carlin et al21 (Australia)   4·6 RotaTeq   9·89 3·70 26·42   6·32 2·78 14·37 2·81 1·16   6·80   1·77 0·81   3·88
Stowe et al2 (England)   4·7 Rotarix 13·81 6·44 28·32   1·59 0·34   3·75 2·20 0·50   5·02   2·77 1·36   5·32
Haber et al27 (USA)   7·0 Rotarix   7·50 2·30 24·60 ·· ·· ·· 2·40 0·80   7·50 ·· ·· ··
Yih et al3 (USA)   7·0 Rotarix ·· ·· ·· ·· ·· ·· 3·50 0·50 25·10   1·21§ 0·11 13·53
Yih et al3 (USA)   7·0 RotaTeq   9·10 2·20 38·60   2·93§ 0·46 18·55 1·80 0·40   7·20   0·76§ 0·16   3·67
Haber et al29 (USA)   7·0 RotaTeq   3·75 1·90   7·39 ·· ·· ·· 1·43 0·85   2·40 ·· ·· ··
Escolano et al30 (worldwide) >7·0 RotaTeq   3·45 1·84   6·55   0·91 0·51   1·62 1·63 0·86   3·13   1·07 0·63   1·80
Patel et al23 (Brazil) 18·4 Rotarix   1·10 0·30   3·30   0·51 0·20   1·33 2·60 1·30   5·20   1·12 0·65   1·93
Patel et al23 (Mexico) 23·2 Rotarix   5·30 3·00   9·30   0·99 0·52   1·91 1·80 0·90   3·80   2·20 1·40   3·45
Velazquez et al24 (Mexico) 23·2 Rotarix   6·49 4·17 10·09   1·19|| 0·78   1·83 1·29 0·80   2·11   1·00|| 0·66   1·52
Groome et al12 (South Africa) 47·0 Rotarix   1·86** 0·09 37·09   4·01 0·87 10·56 1·71 0·83   3·01   0·96 0·52   1·60
Tate et al11 (Africa) 67·0 Rotarix   0·25 0·00   1·16   1·01 0·26   2·24 0·76 0·16   1·87   0·74 0·39   1·20
Base-case risk scenario
Pooled RR ·· ··   6·26 4·25   9·22   1·69 1·05   2·72 1·82 1·41   2·34   1·37 1·03   1·84
l2 ·· ·· 39·0% ·· ·· 61·7% ·· ·· 0·0% ·· ·· 44·9% ·· ··
p value ·· ··   0·081 ·· ··   0·0036 ·· ·· 0·76 ·· ··   0·046 ·· ··
Scenario with overlapping data points
Pooled RR ·· ··   5·52 3·93   7·75   1·56 1·02   2·40 1·75 1·41   2·16   1·33 1·03   1·73
l2 ·· ·· 43·0% ·· ·· 61·5% ·· ·· 0% ·· ·· 41·7% ·· ··
p value ·· ··   0·044 ·· ··   0·0026 ·· ·· 0·86 ·· ··   0·057 ·· ··

RR=relative risk.

*

No intussusception cases occurred in the period 8–21 days, so the risk was calculated as: exp ([ln (RR1 – 21) – (ln (RR1 – 7) × (7/22))]/(15/22)).

RR based on method that included historical cases to adjust for age.

Indicates data points that were excluded from the main analysis because a data point was available from a more systematic study for the same population, time period, brand, dose, or risk period; two studies from Mexico reflected different populations so were both included in the main analysis; pooled RR estimates that allowed for overlapping data points indicated slightly lower risk than the pooled RRs for the base-case scenario.

§

RR in the period 8–21 days was calculated as: exp ([ln (RR1–21) – (ln (RR1–7) × (7/22))]/(15/22)).

||

RR in the period 7–30 days was used as a proxy for 8–21 days and was calculated as: exp ([ln (RR0 – 30) – (ln (RR0 – 6) × (7/31))]/(24/31)).

RR in the period 8–21 days was calculated as: exp (average [ln (RR8 – 14), ln (RR15–21)]).

**

No intussusception cases occurred in the 1–7-day period after dose 1 so the risk in this period was calculated as: exp ([ln (RR1 – 21) – ((15/22) × ln (RR8 – 21))]/(7/22)).