Table 2.
Risk of bias assessment of included studies
| Studies | Representativeness of exposed cohort | Selection of non-exposed cohort | Ascertainment of exposure | Demonstration | Comparability | Assessment of outcome | Follow-up long enougha | Adequacy of follow-upb | Total score |
|---|---|---|---|---|---|---|---|---|---|
| Case Reports | |||||||||
| Franz 2006 | Somewhat representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Variable | Complete* | 5 |
| Wienecke 2006 | Somewhat representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | Complete* | 6 |
| Birca 2010 | Somewhat representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Unclear | Complete* | 5 |
| Lam 2010 | Somewhat representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | Complete* | 6 |
| Cohort Prospective | |||||||||
| Krueger 2010 | Truly representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | Complete* | 6 |
| Dabora 2011 | Truly representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | 8 out of 36 lost follow-up (22 %)* | 6 |
| Davies 2011 | Truly representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | 4 out of 16 lost follow-up (25 %)* | 6 |
| Lopez 2011 | Truly representative* | No non-exposed cohort | Secure record* | Yes* | No | Record linkage* | Yes* | 1 out of 17 lost follow-up (5 %)* | 6 |
aLength of follow-up of at least 6 months is considered sufficient
bLost of follow-up <30 % is considered acceptable; *Indicates point that contribute to Total Score