INDEPENDENT THINKING. WHOLE-FOOD LIVING.NO ANIMALS / NO OIL
THENANODIET
← Nutrition & Adequacy / Health Desk

Dietary calcium and fractures: a nonlinear cohort result

The Swedish findings support neither unlimited calcium nor ignoring low intake.

Whole-food ingredients prepared without added oil

Illustrative food image, not a photograph of the study intervention.

Food intake observed over time

The Swedish Mammography Cohort analysis followed 61,433 women for 19 years. Investigators estimated dietary calcium using repeated food-frequency questionnaires and identified fractures through registry data. A smaller clinical subcohort supplied bone-density information for an osteoporosis analysis. This is prospective observational evidence, not a trial assigning calcium doses. Its exposure is reported dietary intake, so results should not be presented as the effect of starting a calcium supplement. The population and design also matter when considering whether the findings transfer to men, younger adults, or other dietary settings.

Warensjö E, Byberg L, Melhus H, Gedeborg R et al.. BMJ (Clinical research ed.), 2011. PMID 21610048 · DOI

The lower-intake comparison has bounds

Compared with the third intake quintile, women in the lowest calcium quintile had an adjusted hazard ratio of 1.18 for a first fracture of any type, with a 95% confidence interval of 1.12 to 1.25. For first hip fracture, the corresponding hazard ratio was 1.29, with bounds of 1.17 to 1.43. These are relative hazard estimates, not percentage-point increases in an individual's probability. The reference category is essential: neither estimate compares calcium consumption with consuming none, and neither by itself establishes that low intake caused the difference.

Warensjö E, Byberg L, Melhus H, Gedeborg R et al.. BMJ (Clinical research ed.), 2011. PMID 21610048 · DOI

More was not uniformly better

The highest intake quintile did not show further reductions in overall fracture or osteoporosis risk. It was associated with a higher hip-fracture hazard, 1.19 relative to the reference, with a 95% confidence interval from 1.06 to 1.32. The overall pattern was nonlinear, so the data do not support extrapolating a simple straight-line benefit as intake rises. Equally, the high-intake association does not demonstrate that calcium itself damages bone. Observed intake groups can differ in other ways even after statistical adjustment for measured characteristics.

Warensjö E, Byberg L, Melhus H, Gedeborg R et al.. BMJ (Clinical research ed.), 2011. PMID 21610048 · DOI

Do not turn quintiles into a prescription

The useful takeaway is a bounded interpretation: low dietary intake was associated with higher fracture risk, while progressively higher intake was not associated with progressively lower risk. Questionnaire measurement error and residual confounding remain possible in this cohort. The study does not identify a universal optimum for every person or justify stopping clinically indicated supplementation. Its quintile comparisons are research categories, not individualized treatment thresholds. Bring questions about dietary adequacy and fracture risk to a clinician rather than treating either the lowest or highest category as a universal target to copy.

Warensjö E, Byberg L, Melhus H, Gedeborg R et al.. BMJ (Clinical research ed.), 2011. PMID 21610048 · DOI