Statistical problems with study on milk intake and mortality and fracturesBMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g6991 (Published 26 November 2014) Cite this as: BMJ 2014;349:g6991
- Christopher Labos, research fellow1,
- James Brophy, professor of medicine and epidemiology2
- 1Institute for Clinical and Evaluate Sciences, University of Toronto, Toronto, Canada
- 2McGill University, Montreal, Canada
Michaëlsson and colleagues’ article attempted to link milk consumption with fracture risk and mortality, especially cardiovascular mortality.1 The authors correctly note that residual confounding may be at play here, especially because their multivariate model did not adjust for osteoporosis or bone mineral density.
Adjusting only for calcium and vitamin D supplementation probably …