Component | Questions | Assessment | Ratings |
---|---|---|---|
Selection bias | Are the individuals selected to participate likely to be representative of the national population? | • Very likely | Studies that are very likely to be representative and have greater than 80% participation will be rated as strong. |
• Somewhat likely | |||
• Not likely | |||
• Cannot tell | |||
What percentage of selected individuals agreed to participate? | • 80 to 100% | ||
• 60 to 79% | |||
• Less than 60% | |||
• Can’t tell | |||
Study design | Were study methods comparable over time? | • Yes | Studies with comparable methods over time will be rated as strong. |
• No | |||
• Cannot tell | |||
Confounders | Were confounders (age, sex, race/ethnicity) controlled for in study design or analysis? | • Yes | Studies that control for confounding will be rated as strong. |
• No | |||
• Cannot tell | |||
Data collection methods | Was anthropometry measured (as opposed to self reported)? | • Yes | Studies where anthropometry was measured will be rated as strong. |
• No | |||
• Cannot tell | |||
Analyses | Are the statistical analyses appropriate to detect differences by SEP? | • Yes | Studies that have performed analyses to detect differences by SEP will be rated as strong. |
• No | |||
• Cannot tell | |||
SEP, socioeconomic position. |