Nineteen studies (n=2,814, range 31 to 623) were included in the review. Seventeen studies were included in the meta-analysis. Six studies were at low risk of bias, five were at high risk of bias and risk of bias in remaining eight studies was unclear.
Injection of different formulations of the same vaccine (five studies): Priorix injection caused less pain than the M-M-R II vaccine according to children’s self-report (SMD -0.66, 95% CI -0.81 to -0.50, NNT=3.7; two studies) and resulted in less crying (RR 0.66, 95% CI 0.59 to 0.74, I2=48%, NNT=3.2; three studies). Similar findings were found for parental report studies (three studies) and physician-reported scores, but there was some evidence of statistical heterogeneity.
Position of the child during injection (four studies): Significant heterogeneity was found (I2=65%). The authors reported that three studies showed greater pain scores in children lying supine during vaccination compared with children who were sitting or being held by a parent. A forest plot showed that the difference was only statistically significant in two studies.
Stroking the skin close to the injection site before and during injection (one study): Children who received stroking reported less pain than those who did not receive stroking.
Order of vaccine injection when two vaccines were administered sequentially (one study): Overall, pain was reported to be lower in children who received diphtheria-polio-tetanus-acellular pertussis-Haemophilus influenzae type b (DPTaP-Hib) vaccine first and pneumococcal conjugate vaccines (PCV) second compared to children who received the vaccines in the opposite order.
Rapid injection without aspiration for intramuscular injection (one study): Rapid injection without aspiration was found to be less painful as assessed by observers, cry duration and parent and physician assessment.
Insufficient evidence was found to support greater effectiveness for intramuscular versus subcutaneous injection (three studies), cooling the skin at the injection site with ice before injection (two studies), simultaneous versus sequential injection of two vaccines (one study) and vaccine temperature (one study).
Sensitivity analysis was not performed. The authors reported potential for publication bias.