Ok... let’s see if I can. So this graph is looking at relative risk of leukaemia if BF. 1 equals no effect of being BF. Less than one is a reduction in relative risk and more than one is an increase in relative risk.
-
-
Replying to @RuthAnnHarpur @_breastfeeding and
But also note these are associations and causality is uncertain.
2 replies 0 retweets 2 likes -
Replying to @RuthAnnHarpur @_breastfeeding and
So on the graph the dots are the relative risk (RR) found in the study. The long lines are 95% confidence intervals so we are 95% that the true RR is between these 2 numbers. A statistically significant association needs to have both CIs on the same side of 1.
1 reply 0 retweets 2 likes -
Replying to @RuthAnnHarpur @_breastfeeding and
5 studies out of 17 meet this criteria First figure in each set of brackets is the RR, next ones are 95% CIs Bener (0.34, 0.18-0.64) Altinkaynak (0.36, 0.18-0.73) Perrillat (0.49, 0.24-0.97) Smulevich (0.64, 0.44-0.93) Shu (0.75, 0.65-0.86)
2 replies 0 retweets 3 likes -
Replying to @RuthAnnHarpur @MariaTeresaEs and
The OR is less than 1 in 13. The whole point of doing a metanalysis is that CI's don't tell you much when the sample size is too small. So either you deal with each study alone or the entire analysis. In the latter case, there is an impact, as I pointed out in an earlier tweet.
1 reply 0 retweets 0 likes -
Replying to @_breastfeeding @MariaTeresaEs and
You claimed nearly all studies found an effect... (the majority did not). The OR in the meta-analysis was .80 (95% CIs 0.72-0.90) which is an association between breastfeeding and lower leukaemia but this is insufficient to determine causality.
7 replies 0 retweets 1 like -
Replying to @RuthAnnHarpur @_breastfeeding and
Meta-analysis gives the cumulative impact - and a 20% reduction with 95% CI 0.72-0.90 is important.
1 reply 0 retweets 1 like -
Replying to @PhyllBuc @_breastfeeding and
I disagree... this is a small effect on something rare is not demonstrated to be causal. How important that is is down to individuals to decide based on what is most important to them.
2 replies 0 retweets 2 likes -
Replying to @RuthAnnHarpur @PhyllBuc and
Would also be helpful to see absolute risks here
@justsaysrisks1 reply 0 retweets 3 likes -
Replying to @RuthAnnHarpur @PhyllBuc and
Challenging to calculate absolute risks directly from the study, because it's a meta analysis of case-control studies. That being said, pediatric leukemia is very rare, a relative risk reduction of 20% would equate to an absolute reduction of about 0.001%
2 replies 1 retweet 6 likes
Also worth noting that this study doesn't really speak to causation, especially considering the wide variance in rigor of the included studies
-
-
Replying to @GidMK @RuthAnnHarpur and
Thank you for your input Health Nerd. For some reason
@_breastfeeding has blocked me. I can only think it's because I called him out over a tweet alluding to not breastfeeding causing higher leukaemia rates. Parents deserve to be given accurate information to make feeding choices1 reply 0 retweets 4 likes -
Replying to @AlieshiaCross @RuthAnnHarpur and
Mmmm, based on the evidence presented I'm not sure how you could argue that breastfeeding prevents leukemia
1 reply 0 retweets 2 likes - Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.