Table II.
Validity of included studies.*
| Study | Allocation concealment | Masking of participants | Masking of outcome assessors | Industry funding or involvement | Compliance assessed and reported | Reporting of withdrawals |
|---|---|---|---|---|---|---|
| Atkinson 2004 | Yes | Yes | Yes | Partly (some funding from non-industry sources, some from Novogen Ltd) | Done, urinary isoflavones | Partial |
| Marini 2008 | Unclear | Yes | Yes | Partly (some funding from non-industry sources, some from Primus Pharmaceuticals, two authors work for Primus Pharmaceuticals) | Done, serum genistein | Done |
| Maskarinec 2002 | Unclear | Yes | Yes | Yes, funding from Pharmavite Corporation | Done, urinary isoflavones and tablet count | Done |
| Maskarinec 2004 | Unclear | No | Yes | Partly (funding from non-industry source, food donations from Aloha Tofu, Dr Soy, Solae Company) | Done, urinary isoflavones, intake logs, 24 h recalls | Partial |
| Maskarinec 2009 OPUS | Unclear | Yes | Yes | Unclear (funded by NIH but unclear if study paid for the tablets or not) | Partly done, blood isoflavones and pill counts, results not reported | Done |
| Powles 2008 | Unclear | Yes | Unclear | Partly (funding from non-industry source, supplement supplied by Novogen Ltd) | Not done | Partial |
| Tice 2009 (PREVENT) | Yes | Yes | Yes | None (US Army) | Done, packet count at 6 mo. | Done |
| Verheus 2008 (Finesse) | Yes | Yes | Yes | Partly (some funding from non-industry sources, some from DuPont Protein Technologies) | Done, serum genistein, counts of powder bags, diary | Done |
*Trial quality characteristics assessed included: (i) allocation concealment (concealment of the ability of those recruiting participants to assess which arm participants will be randomized into before recruitment is complete, coded as adequate, unclear or inadequate); (ii) participant masking (concealment of the participants to whether they are part of the intervention or control condition, coded as ‘yes’ where there was a clear and realistic attempt to mask, ‘no’ where not, or ‘unclear’); (iii) outcome assessor blinding (concealment of the outcome assessor, here the reader of the mammogram, to whether participants are part of the intervention or control condition, coded as ‘yes’ where there was a clear and realistic attempt to mask, ‘no’ where not, or ‘unclear’); (iv) industry funding or involvement [level of financial involvement of industries that may have a financial interest in the study results, coded as yes (study mainly funded by industry, or at least one author is employed by industry), partly (other impartial funding, but includes some industry funding that may include free provision of supplements) or no]; (iv) compliance assessed and reported (measurement of the degree to which participants complied with taking the intervention and control foods or supplements, and reporting of these data, coded as ‘done’ when compliance was both assessed and reported, ‘partly done’ when it was assessed but not reported or reported without any indication of the method used, and ‘not done’ when neither was addressed adequately); and (v) reporting of withdrawals [numbers of withdrawals in each group clear and reasons reported coded as done, reported as ‘done’ when numbers randomized, completed and analysed all clear, plus reasons for dropouts given (by intervention arm), ‘partially done’ when some of the above, ‘not done’ when not].