ROB scale and accepted criteria | Akiskal et al. 1995 | Angst et al. 2003 | Egeland et al. 2012 | Gan et al. 2011 | Hafeman et al. 2017 | Ratheesh et al. 2015 | Salvatore et al. 2013 | Tohen et al. 2012 | DeGeorge et al. 2014 | Kochman et al. 2005 | Sperry et al. 2020 |
---|---|---|---|---|---|---|---|---|---|---|---|
Selection | |||||||||||
 Exposed cohort is representative of the average person without BD | – | – | – | – | – | – | – | – | – | – | – |
 Non-exposed cohort is drawn from the same community as the exposed cohort | * | * | * | * | * | * | * | * | * | * | * |
Exposure ascertained through secure record or structured interview | – | – | * | * | – | – | * | * | – | – | – |
 Demonstration that outcome of interest was not present at start of study | * | * | * | * | * | * | * | * | * | * | * |
Comparability | |||||||||||
 Study controls for family history of BD or additional factor | ** | * | – | – | ** | – | * | – | – | – | – |
Outcome | |||||||||||
 Assessment of outcome uses structured clinical assessment or record linkage | * | * | * | * | * | * | * | * | * | * | * |
 Follow-up long enough for outcome to occur (5 + years)? | * | * | * | – | * | – | – | – | – | – | – |
 Adequacy of follow up of cohorts (> 80% follow up or description provided of those lost) | – | – | – | * | * | – | – | * | – | – | – |
Total (max = 9) | 6 | 5 | 5 | 5 | 7 | 3 | 5 | 5 | 3 | 3 | 3 |