Table 4.
Quality assessment of reviewed cohort studies
| Reference | Exposed cohort represented average in community | Nonexposed cohort selected from same community | Exposure ascertained through records or structured interviews | Outcome demonstrated to be not present at study start | Exposed and nonexposed subjects matched and/or adjusted by factors | Outcome ascertained via independent blind assessment or record linkage | Follow-up long enough for outcome to occur | Loss to follow-up <20% |
|---|---|---|---|---|---|---|---|---|
| Sonestedt et al. (2008)34 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 |
| Singh & Fraser (1998)33 | 1 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Yeh et al. (2006)36 | 1 | 1 | 1 | 0 | 2 | 1 | 1 | 1 |
| Thompson et al. (2010)35 | 1 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Hedelin et al. (2011)32 | 1 | 1 | 0 | 0 | 2 | 1 | 1 | 1 |
| Bao et al. (2013)37 | 0 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Mills et al. (1989)39 | 1 | 1 | 0 | 0 | 2 | 1 | 1 | 1 |
| Pan et al. (2013)40 | 0 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Parker et al. (2003)41 | 1 | 1 | 0 | 1 | 2 | 0 | 1 | 1 |
| Villegas et al. (2008)42 | 1 | 1 | 1 | 1 | 2 | 0 | 1 | 1 |
| Kochar et al. (2010)38 | 0 | 1 | 0 | 1 | 2 | 0 | 1 | 0 |
| Thiebaut et al. (2009)46 | 0 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Jenab et al. (2004)43 | 1 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Saberi Hosnijeh et al. (2014)45 | 1 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Nettleton et al. (2008)44 | 1 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
| Farvid et al. (2014)66 | 0 | 1 | 0 | 1 | 2 | 1 | 1 | 1 |
aScoring: 2 means the subjects were fully matched and/or the analysis was fully adjusted. 1 means the study adequately fulfilled a quality criterion, 0 means it did not. Quality scale does not imply that items are of equally relevant importance.