It might be assumed though that when the people at risk start taking extra dairy, this will be a substitution—either full or partly—for other food products. Hence, in this situation, the total cost of dairy foods might only be slightly higher. If a strict health care perspective is adopted, Thiazovivin ic50 the costs of purchasing dairy foods as part of a normal diet do not need to be taken into account. The scope of the analysis can be limited to the health

care costs made for hip fractures. Some remarks should be made on the data used as input in the calculations, especially regarding the relative risk for hip Selleckchem Pinometostat fracture associated with low calcium intake. First, reviews with pooled study results do not take into account different starting levels of calcium intake. This might hamper the interpretation of the effect size of low calcium intake on the occurrence of hip fractures. The data existing in the literature did not allow us to correct for a different start point in calcium intake of these elderly in our model. This probably resulted in an underestimation of the effect size of the main outcomes in this study. Second,

the relative risk for hip fracture was derived from the meta-analysis of Cumming et al. [37]. Although more recent studies are available on the relationship between calcium intake and osteoporotic fracture, MLN2238 this study mentioned a dose–response relationship. In another meta-analysis, it was found that a supplement of 500 to 1,200 mg calcium would reduce the risk of hip fracture with 12 % (RR 0.88; 95 % CI 0.83–0.95) [50]. This study only took into account randomized controlled trials, with calcium supplementation as Terminal deoxynucleotidyl transferase intervention. However, both studies are concordant. Recently, a meta-analysis by Bischoff-Ferrari et al. did not find a significant reduction in hip fracture by drinking milk for men and women [51]. However, by deleting a Swedish study (considered to be an outlier) from their

analyses, the authors found a statistically significant risk reduction of 5 %. Also in a meta-analysis by Kanis et al. [44], it was found that a low intake of milk was not associated with a marked increase in hip fracture risk. However, low intake was defined as drinking less than one glass of milk daily. Dairy products such as cheese and yogurt were not taken into account. We defined low calcium intake to be under 600 mg, we took a risk reduction of 8 % based on the data of Cumming et al [37], thereby following a conservative approach. Finally, our approach was supported by the results of a recent population-based cohort study by Warensjö et al. In this study, it was found that a dietary calcium intake below approximately 700 mg per day in women was associated with an increased risk of hip fracture [52]. This risk estimate was somewhat higher than in our study. However, this comprehensive study was not specifically directed at dairy calcium intake.