This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
wiki:guide:qc:qc_checklist [2022/02/09 17:25] nicole_hardy_alumni.brown.edu |
wiki:guide:qc:qc_checklist [2022/02/09 17:29] nicole_hardy_alumni.brown.edu |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== QC Checklist ====== | ====== QC Checklist ====== | ||
- | 1. Do the studies meet the inclusion criteria? | + | |
- | a. Are they the primary studies and not secondary studies? | + | |
- | b. Are there any papers with biased sub-populations? | + | |
- | 2. Is the tagging complete? | + | |
- | a. Check for gap teeth \\ | + | |
- | b. Check for missing tags\\ | + | |
- | c. Use the bulk actions to apply missing tags to included studies and go back and add text. \\ | + | |
- | 3. Is extraction done correctly? | + | |
- | a. Are all the units matching? | + | |
- | i. Check this using quantitative synthesis\\ | + | |
- | b. Is the same scale used for a given outcome across papers? | + | |
- | c. Is there missing data that should have been extracted? | + | |
- | i. Can standard deviation (SD) be calculated from standard error (SE) or confidence interval (CI)?\\ | + | |
- | ii. Was there data in the results/ | + | |
- | iii. Was there data in tables that should have been extracted? | + | |
- | iv. Are there any secondary analyses with extra data we could extract? | + | |
- | 1. If so, make sure you comment where you got the data. \\ | + | |
- | v. Is there a supplement with useful data? \\ | + | |
- | 1. If so, make sure you comment where you got the data. \\ | + | |
- | vi. Is there data on clinicaltrials.gov? | + | |
- | 1. If so, make sure you comment where you got the data. \\ | + | |
- | d. Were there errors in the extraction? | + | |
- | i. Make sure means/ | + | |
- | ii. Make sure SD/SE are not confused.\\ | + | |
- | iii. Make sure IQR and range are not confused. | + | |
- | iv. Make sure LS means are treated separately from (arithmetic) means. Unless otherwise specified it will be an arithmetic mean. \\ | + | |
- | v. Make sure loss to follow up is reflected in the sample sizes. | + | |
- | e. Are outcomes extracted at compatible timepoints across papers? | + | |
- | f. Are blanks and 0s properly differentiated? | + | |
- | g. Are the definitions of data elements consistent across papers? | + | |
+ | |||
+ | If you prefer to have this in a downloadable word doc, here it is! | ||
+ |