Table 2.
 Methodological quality assessments of non-randomised studies. Values in brackets are maximum scores achievable
| Study (design) | Downs and Black quality score (maximum number)8 | |||||
|---|---|---|---|---|---|---|
| Reporting (11) | External validity (3) | Internal validity: bias (7) | Internal validity: confounding (6) | Power (2) | Overall (29) | |
| Boersma et al 2001w15 (prospective cohort) | 10 | 3 | 4 | 4 | 2 | 23 |
| Poldermans et al 2003w6 (case-control) | 10 | 2 | 4 | 4 | 2 | 22 |
| Landesberg et al 2003w17 (retrospective cohort) | 7 | 3 | 4 | 4 | 1 | 19 |
| Lindenauer et al 2004w7 (retrospective cohort) | 10 | 2 | 4 | 4 | 2 | 22 |
| Kertai et al 2004w13 (retrospective cohort) | 9 | 3 | 4 | 3 | 2 | 21 |
| Abbruzzese et al 2004w10 (retrospective cohort) | 9 | 3 | 4 | 3 | 2 | 22 |
| O'Neil-Callahan et al 2005w2 (retrospective cohort) | 8 | 3 | 5 | 4 | 2 | 21 |
| Schouten et al 2005w8 (retrospective cohort) | 9 | 3 | 4 | 3 | 2 | 21 |
| Kennedy et al 2005w5 (retrospective cohort) | 10 | 3 | 3 | 4 | 2 | 22 |
| McGirt et al 2005w3 (retrospective cohort) | 11 | 3 | 4 | 4 | 2 | 24 |
| Ward et al 2005w9 (retrospective cohort) | 9 | 3 | 4 | 4 | 2 | 22 |
| Amar et al 2005w18 (prospective cohort) | 6 | 2 | 3 | 3 | 1 | 15 |
| Conte et al 2005w16 (prospective cohort) | 8 | 1 | 3 | 4 | 1 | 17 |
| Dotani et al 2000w12 (retrospective cohort) | 9 | 3 | 4 | 3 | 2 | 22 |
| Pan et al 2004w11 (retrospective cohort) | 8 | 3 | 3 | 4 | 2 | 20 |
| Ali and Buth 2005w4 (retrospective cohort) | 10 | 3 | 3 | 4 | 2 | 22 |