Study

Population and setting

Summary of results

Calcium and vitamin D:

Nursing home:

Chapuy et al, 1992

2790 women (mean age 84) in French nursing homes

Significant reduction in all fractures (odds ratio 0.72, 95% confidence interval 0.57 to 0.90)

Chapuy et al, 2002

583 women (mean age 85) in French nursing homes

No reduction in all fractures (odds ratio 0.99, 95% confidence interval 0.62 to 1.62)

Community:

Dawson-Hughes, 1997

389 community dwelling men and women (>65) in USA

Significant reduction in osteoporotic fractures (odds ratio 0.42, 95% confidence interval 0.18 to 0.92)

Larsen et al, 2004

9605 community dwelling men and women in Denmark

16% reduction in fracture incidence

Calcium with or without vitamin D:

Medical Research Council RECORD trial, 2005

5292 men and women (aged over 70), recruited from fracture clinics in UK to a factorial trial comparing calcium with or without cholecalciferol

No difference in fracture incidence (hazard ratio 0.94, 95% confidence interval 0.81 to 1.09 for calcium v no calcium, and 1.02, 95% confidence interval 0.88 to 1.19 for vitamin D v no vitamin D)

Vitamin D alone:

Lips et al, 1996

2578 men and women (aged over 70) randomised to 400 IU of vitamin D daily

Slight increase in risk of hip fracture among participants allocated to vitamin D (hazard ratio 1.18, 95% confidence interval 0.81 to 1.71)

Trivedi et al, 2003

2686 mainly retired male doctors allocated to 100 000 IU of vitamin D every four months

Slight decrease in relative risk of fracture (0.78, 95% confidence interval 0.61 to 0.99)

Smith et al, 2004

9440 men and women (aged 75 and over) allocated to annual injection of 300 000 IU of vitamin D

Slight increase in risk of all fractures and large increase in hip fractures (hazard ratio 1.10, 95% confidence interval 0.94 to 1.29 for all fractures and 1.48, 95% confidence interval 1.01 to 2.17 for hip fractures)