TABLE 2.
Quality assessment of included observational studies using the Newcastle–Ottawa Scale
| Selection | Comparability | Outcome | Total score | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Study and year1 | Representativeness of the exposed cohort2 | Selection of the non-exposed cohort3 | Ascertainment of exposure4 | Demonstration that outcome of interest was not present at start of study | Comparability of cohorts on the basis of the design or analysis | Assessment of outcome5 | Was follow-up long enough for outcomes to occur? | Adequacy of follow-up of cohorts6 | (9*) |
| Berg et al. (36), 2001 | — | * | * | * | — | * | * | — | ***** (5) |
| Chen et al. (35), 2014 | * | * | * | * | — | * | * | * | ******* (7) |
| Cuco et al. 2006 (30) | — | * | * | * | — | * | * | * | ****** (6) |
| Derbyshire et al. (26), 2009 | — | * | * | * | — | * | * | * | ****** (6) |
| George et al. (21), 2005 | — | * | * | * | — | * | * | * | ****** (6) |
| Jardí et al. (31), 2019 | * | * | * | * | — | * | * | — | ****** (6) |
| Lebrun et al. (29), 2019 | — | * | * | * | — | * | * | — | ***** (5) |
| Lundqvist et al. (28), 2016 | * | * | * | * | — | * | * | * | ******* (7) |
| Moran et al. (34), 2013 | — | * | * | * | — | * | * | * | ****** (6) |
| Murphy and Abrams (24), 1993 | * | * | * | * | — | * | * | — | ****** (6) |
| Murrin et al. (32), 2013 | * | * | * | * | — | * | * | — | ****** (6) |
| Northstone and Emmet (25), 2008 | * | * | * | * | — | * | * | — | ****** (6) |
| Olson (20), 2005 | * | * | * | * | — | * | * | * | ******* (7) |
| Sotres-Alvarez et al. (23), 2013 | — | * | * | * | — | * | * | * | ****** (6) |
| Talai Rad et al. (33), 2011 | — | * | * | * | — | * | * | * | ****** (6) |
| Tovar et al. (22), 2019 | * | * | * | * | — | * | * | — | ****** (6) |
| Wennberg et al. (27), 2016 | — | * | * | * | — | * | * | — | ***** (5) |
A study can be awarded a maximum of 1 star for each category and a maximum of 2 stars for comparability.
One star was assigned if the cohort was truly or somewhat representative of a community-/population-based study. No star was assigned if the study population was sampled from a special population—that is, a population from a company, hospital patients, data from the health insurance company or health examination organization, nurses, etc.
One star was assigned if drawn from the same community as the exposed cohort.
Star assigned where diets were assessed using validated FFQ, 24-h recalls, or multiple-day food record.
Star assigned where it was an independent blind assessment.
Star assigned where follow-up rate was >80% or the description of those lost suggested no different from those followed.