This systematic review was conducted following the methodological standards of the Cochrane Handbook Systematic Reviews of Interventions (35) and adhered to the PRISMA 2020 statement (36) to ensure transparent and comprehensive reporting. The planning, coordination, iterative updates, and evidence synthesis followed the structured approach proposed by Muka et al. (37) and EFSA guidance (38, 39), with adaptations based on the PIMENTO Study Protocol. The EFSA guidance was followed specially to incorporate food characterization, mechanism of action, safety and mechanistic substantiation and evidence grading, which was included in our review as a part of the non-systematic analysis described in Section “2.2 Non-systematic components of the review.”

In accordance with EFSA’s scientific guidance (38, 39), we conducted non-systematic reviews on three key aspects to support the systematic review findings:

Observational and interventional studies, including randomized controlled trials (RCTs) and non-randomized interventions, were among our eligibility criteria and both types of studies were found in the 1302 records analyzed (Figure 1). However, after screening titles, abstracts, and full texts in CADIMA according to predefined inclusion and exclusion criteria, only 12 articles reporting observational studies, encompassing cohort, case-control and cross-sectional studies, met the established eligibility criteria and were finally selected (Figure 1). The selected studies were conducted in various adult populations including postmenopausal women (Table 2). They assessed endpoints such as BMD, bone turnover markers, and risk of fracture. We did not find any suitable randomized or non-randomized clinical trials that met the eligibility criteria.

Although 12 articles were selected, they corresponded to 14 studies, because in one article two different cohorts, one from the UK (The UK Biobank Cohort) and one from Spain (The Seniors-ENRICA Study), were reported and compared (23), and in another article two different end-points, hip fractures and BMD, were assessed independently (21). Notably, only a single study (19) specifically addressed the effect of yogurt consumption on bone health outcomes. In the remaining studies, yogurt intake was considered as part of a broader category of milk or total dairy food consumption, making it challenging to isolate yogurt’s unique impact. Five out of the 14 studies were conducted in Europe (19, 23, 26, 48), two in Asia (20, 49), and seven in the USA (21, 22, 27, 50–52). Although a majority studies were from the USA, these included populations from only two independent cohorts, The Framingham Original Cohort and its derivatives (The Framingham Offspring Study, The Framingham Heart Study, and The Framingham Osteoporosis Study) (21, 50, 52) and the Nurses’ Health Study (22, 27, 51). Of the Asian cohorts, one was from Korea (20) and the other one from Japan (49).

*Reasons:

Population-related exclusions (n = 28) arose from studies involving adults with pre-existing fractures at enrolment, those specifically recruiting pregnant women or infants, or animal models, which did not meet our target demographic of healthy adults or individuals with osteoporosis/osteopenia.

Intervention/exposure exclusions (n = 117) were primarily due to non-eligible fermented products, including alcoholic beverages exceeding 1.25% alcohol content, non-nutritional applications (e.g., topical or nasal use), or interventions containing confounders such as added prebiotics or bioactive compounds. Probiotics were excluded unless the microbial strains carried out the milk fermentation. Studies employing prebiotic-enriched foods, postbiotics, or pill-based supplements (e.g., algal extracts) were also omitted.

Outcome-related exclusions (n = 42) involved studies failing to report predefined bone health metrics, such as changes in BMD (e.g., osteoporosis-to-osteopenia transitions), incidence of fractures (hip, wrist, spine), serum calcium/alkaline phosphatase levels, or adverse effects per CTCAE criteria. Additionally, studies lacking valid effect measures (e.g., hazard ratios, weighted mean differences) were excluded.

Comparator-related exclusions (n = 32) pertained to studies without a defined control group (e.g., placebo, non-fermented food comparator).

Other exclusions (n = 6) included unavailable full texts (n = 5) and duplicate records (n = 1).

†Reasons:

Ineligible intervention (n = 25). Independent exposure to yogurt not recorded (n = 18): studies analyzed mixed dairy intake without disaggregating yogurt-specific effects, rendering data non-extractable for our review.

Yogurt with non-permitted additives (n = 7): included trials testing yogurt enriched with probiotics (n = 2), vitamin D (n = 1), or other bioactive compounds (n = 4), which confounded the assessment of yogurt’s intrinsic properties.

Inappropriate study design (n = 6): (a) non-yogurt-specific RCTs (n = 3): although randomized, these trials evaluated composite interventions (e.g., yogurt combined with supplements or other fermented foods) without isolated yogurt arms; (b) uncontrolled or non-comparative designs (n = 3): lacked placebo or fermented food comparators such as milk or milk plus vitamin D.

Irrelevant Outcomes (n = 5): insufficient follow-up (n = 4): (a) study durations < 6 months, inadequate to detect bone density or fracture risk changes; (b) on-skeletal endpoints (n = 1): focused on general well-being or inflammatory markers without assessing BMD, fractures, or bone-related biomarkers (e.g., serum calcium, alkaline phosphatase).

Two studies reported a statistically significant association between yogurt consumption and improvement of the endpoint under consideration, BMD and bone-associated biomarkers (19) and risk of osteoporosis (20). In the latter work, a protective effect on radius osteoporosis risk was found in high frequency yogurt consumers as compared with non-consumers [HR = 0.51, 95% CI: 0.30–0.85 for >5–6 servings/week vs. non-consumers (P for trend = 0.0167)] (20). Similarly, total hip and femoral neck BMD in females were 3.1%–3.9% higher among those with the highest yogurt intakes compared to the lowest (P < 0.05) (19). These authors further detected a positive effect in some physical function measures, such as the Timed Up and Go response (−6.7% reduction; P = 0.013) in high yogurt consumers. In one additional study (21), a weak association between yogurt intake and a protective effect on hip fractures was identified, although the reported Hazard Ratios was not significant [HR(95% CI): ≤4 serv/wk: 0.46 (0.21–1.03) vs. >4 serv/wk: 0.43 (0.06–3.27)]. However, a statistically significant protective effect was found for BMD at the trochanter (P-value for > 4 serv/wk intake vs. no intake = 0.03).

The studies assessing hip fracture demonstrated high methodological quality in the Newcastle Ottawa Scale (NOS), supported by well-defined cohorts, extended follow-up, and robust outcome validation (Tables 3, 4). Overall, the methodological quality of the included evidence was deemed satisfactory to support the conclusions of the meta-analysis. Studies differ widely in number of individuals followed (from 764 up to 103,003), male/female sex ratio (with five studies considering only women), mean or range age, endpoints (general fractures, hip fracture, falls, osteoporosis, BMD, and bone-associated biomarkers), duration of the follow up period (from one-point analysis up to 20.08 years), and yogurt consumption (both in serving size and servings per time unit). These large differences make difficult a comparison of the results from the evaluated studies.

The certainty of the evidence was assessed using the GRADE approach, following Cochrane guidance (35, 44). The two outcomes under evaluation, hip fracture and BMD, were graded separately, and the results summarized in Table 5. As all studies were observational in design, the initial certainty rating for each outcome began at “low.” We evaluated upgrading criteria, including large magnitude of effect, presence of a dose-response gradient, and potential residual confounding. None of these criteria were met: the size effect was small (e.g., SMD = 0.009 for BMD; HRs close to 1.0 for hip fracture), no dose-response relationships were observed, and confounding factors were well controlled in most studies. Therefore, no upgrading was applied. We also assessed the five downgrading domains: risk of bias: overall low, with average NOS scores between 7.0 and 8.3; inconsistency: none; results were consistent across studies (e.g., I² = 0%); indirectness: no serious concerns; populations, exposures, and outcomes were relevant; imprecision: confidence intervals were relatively narrow; publication bias: not suspected.

Our systematic review and meta-analysis found no significant association between yogurt consumption and hip fracture risk, based on data from large, long-term cohort studies. The pooled estimates were consistent across populations and showed hazard ratios close to unity. For BMD, the evidence indicated a small but statistically significant positive effect. However, the magnitude of this effect was clinically negligible. Overall, the aggregated findings suggest that yogurt consumption is not convincingly linked to fracture prevention and, at best, has a marginal impact on BMD. Therefore, based on these assessments, the certainty of the evidence for the two outcomes, hip fracture and BMD, remained low.

According to international standards, to ensure the functional properties of yogurt, viable counts of the two lactic acid bacteria (LAB) species in yogurt, S. thermophilus and L. bulgaricus must reach at least 10⁷ colony-forming units (cfu) per gram at the end of shelf life (≈28 days). However, none of the studies included in this review reported bacterial counts or verified the microbial viability in the yogurts consumed, limiting the ability to assess the potential contribution of the live cultures to the observed outcomes.

The mechanisms through which yogurt could support bone health may involve both nutritional, biochemical, and microbiological pathways. During fermentation, S. thermophilus and L. bulgaricus metabolize lactose into lactic acid, decreasing pH and enhancing calcium solubility, which improves intestinal absorption of this mineral (54). Acidification not only facilitates passive diffusion but may also stimulate active transport through the upregulation of expression of calcium-binding proteins.

Yogurt is a natural source of essential micronutrients, including vitamins B₂, B₆, B₁₂, and K₂, which serve as cofactors in osteoblast differentiation, collagen synthesis, and the γ-carboxylation of osteocalcin –crucial for bone matrix mineralization (55). Of particular interest is vitamin K₂ (menaquinone), which plays a key role in calcium metabolism and bone strength (17). The specific isoform of vitamin K₂ (e.g., MK-4 to MK-13) present in yogurt varies depending on the starter cultures and fat content of the milk used for production (56), which may partially explain the heterogeneity in outcomes across studies. Further, recent findings suggest a synergistic effect between vitamins K₂ and D₃ in the regulation of calcium homeostasis and bone formation (57).

S. thermophilus and L. bulgaricus possess proteolytic systems capable of breaking down milk proteins into peptides and amino acids with potential biological activity, including anti-inflammatory and antioxidant effects (58, 59). Additionally, these bacteria can synthesize extracellular polysaccharides (EPS) and other fermentation-derived compounds that have been linked to improved gut barrier function and systemic immunomodulation (3, 4).

Finally, strains of LAB species, including those of S. thermophilus and L. delbrueckii, can modulate the gut microbiota, promoting the production of short-chain fatty acids (SCFAs), such as acetate, butyrate, and propionate. SCFAs have demonstrated anti-inflammatory effects and the capacity to inhibit osteoclast differentiation, thereby reducing bone resorption (60). The gut–bone axis, mediated by immune signaling and cytokine modulation, further supports the potential of yogurt-derived microbiota interactions to influence systemic bone metabolism (61).

Fermented dairy products such as yogurt may enhance the bioavailability of specific nutrients and bioactive compounds through modifications induced by fermentation (62). Compared to non-fermented milk, yogurt exhibits improved nutrient solubility, enzymatic liberation of active compounds, and interaction with gut microbiota–all of which may facilitate systemic absorption and delivery to bone tissue.

1. Calcium

Yogurt’s acidic pH, resulting from lactic acid production during fermentation, increases calcium ionization and its passive diffusion across intestinal membranes (21, 63). Some studies indicate that yogurt consumption leads to higher calcium uptake when compared to milk, suggesting that fermentation enhances mineral bioavailability (64). Increased calcium uptake from fermented milk is further supported by in vitro assays using Caco-2 cells (65). In addition, the favorable calcium-to-sodium ratio in yogurt may further support its efficient utilization, minimizing the risk of phosphate imbalance often associated with high-sodium dairy products.

2. Protein and amino acids

Proteins and peptides in yogurt are partially hydrolyzed during fermentation by the proteolytic systems of S. thermophilus and L. bulgaricus, resulting in the release of amino acids and short peptides (2, 58, 66). This hydrolysis may improve intestinal absorption, thereby facilitating the availability of essential amino acids to peripheral tissues such as bone. In particular, branched-chain amino acids (BCAAs) may contribute to musculoskeletal integrity. However, direct evidence on the transport and utilization of these peptides by bone cells remain limited in the reviewed studies.

3. Vitamins (B₂, B₁₂, K₂, and D)

Yogurt can serve as a dietary source of vitamin B₂ (riboflavin) and B₁₂ (cobalamin), both of which may be synthesized or preserved during fermentation depending on the starter strains (67, 68). Although the specific mechanisms by which these vitamins contribute to bone metabolism are not fully elucidated, their increased content and stability in yogurt matrices may enhance systemic availability, particularly in populations at risk of deficiency (69).

Vitamin K₂ (menaquinone) bioavailability also appears to benefit from the dairy matrix. Although variability in content across different yogurt types due to differences in starter cultures and fat content, some forms, such as MK-7, demonstrate more efficient intestinal absorption when delivered via yogurt (56). The interplay between vitamin D and K₂, noted in the mechanistic studies, may be partly mediated by their co-delivery in fermented dairy products, supporting their concurrent availability for the regulation of calcium uptake (57).

4. Fermentation and gut interactions

Beyond the liberation of nutrients, the yogurt fermentation process itself may influence bioavailability via gut-level interactions. Some studies report that yogurt consumption modulates gut microbiota composition, promoting the production of short-chain fatty acids (SCFAs) (70), which may enhance nutrient absorption and barrier function. Yogurt intake has also been associated with lower serum levels of tartrate-resistant acid phosphatase 5b (Trap 5b), a marker of bone resorption (19). While anti-inflammatory effects are often discussed, their role in facilitating systemic transport of bone-relevant compounds remains an emerging area of research and was not fully explored in the reviewed articles.

The characterization of the yogurts in the clinical and observational studies included in this review was generally very limited. In most cases, yogurt consumption was assessed through self-reported FFQs or dietary recalls, with no detailed specification regarding the physicochemical or microbiological properties of the products consumed. Only a few studies explicitly distinguished between yogurt types, such as plain versus flavored (22, 23, 48). However, even in these cases, further compositional details –such as fat content (whole, semi-skimmed, or skimmed), added sugars, or presence of fortifying agents (e.g., calcium or vitamin D)– were rarely reported (Supplementary Table 2). Similarly, none of the studies provided information on whether the yogurt was pasteurized after fermentation (a process allowed in certain countries), whether it contained probiotic strains of Bifidobacterium or other LAB species (also allowed in certain regions), or on the viability and concentration of the manufacturing cultures (Supplementary Table 2).

This insufficient characterization constrains the interpretability of the findings and the reproducibility of the observed associations. Given that the bioavailability and functionality of yogurt-derived nutrients and bioactive compounds –such as vitamins K₂, B₁₂, peptides, or short-chain fatty acids– are strongly influenced by fermentation processes, strain selection, and matrix composition, the lack of details limits the ability to attribute specific effects to yogurt consumption per se. Moreover, a limited product definition introduces exposure heterogeneity, which may partially account for the variability on the observed outcomes across cohorts.

Although yogurt is generally regarded as safe, the evidence base remains limited for specific populations (71). Most studies did not stratify results by lactose intolerance status, and no study systematically evaluated safety outcomes in individuals with dairy protein allergies or gastrointestinal disorders such as irritable bowel syndrome (IBS). Among the reviewed studies, only two reported to have taken into consideration possible adverse effects associated with yogurt consumption (48, 49). These effects were minor –mainly abdominal bloating and self-reported lactose intolerance– and no study attributed adverse events to the milk fermentation processes, the bacterial strains used as starters, or to the pH changes occurring during yogurt manufacture. In addition, yogurt is generally considered microbiologically stable due to its production from heat-treated milk, refrigerated storage, and because it contains a competitive microbiota that acidify the product through the synthesis of organic acids (71, 72).

The results of this systematic review partially agree with findings of previous reviews but also offer important nuances that enrich the current understanding of the role of yogurt in bone health. Similar to what Ong et al. (32) reported in their systematic review and meta-analysis, positive associations were identified in this work between frequent yogurt consumption and higher BMD or a lower incidence of hip fracture in postmenopausal women. However, while Ong and co-workers reported a significant 24% reduction in the risk of hip fracture (RR = 0.76; 95% CI: 0.63–0.92), the meta-analysis performed in this review, based on three large cohorts with long follow-ups and no heterogeneity between studies, did not find a statistically significant association (HR = 1.01; 95% CI: 0.96–1.07; p = 0.85). This discrepancy may be explained, at least in part, by differences in the definition of exposure (e.g., >0 vs. ≥5 times/week vs. ≥2 times/day), cohort inclusion criteria, and adjustment for relevant covariates such as physical activity, nutritional status, or intake of supplements, as well as by possible differences in the compositional characteristics of the yogurts included in the studies.

A key methodological difference between the two reviews lies in their focus on dietary exposure. While Ong et al. (32) exclusively included studies that distinguished yogurt and cheese as independent exposures, in this review only one study assessed specifically the effect of yogurt alone (19). In all others, yogurt was considered within the broad category of “dairy.” This lack of exposure specificity likely contributes to the attenuation of the effect observed in the meta-analysis. Furthermore, the review presented here focused exclusively on hip fracture as a clinical outcome, while Ong et al. (32) also considered BMD, T-score, and bone turnover markers, which might have allowed them to capture potential effects at earlier stages of bone deterioration.

Both this review and previous ones agree in pointing out the critical lack of high-quality randomized controlled trials (RCTs) evaluating the effect of a nutritionally and microbiologically well-characterized yogurt as specific dietary interventions on fracture outcomes. The few available RCTs, such as those reviewed by Ong et al. (32) are of short duration (<2 months), with intermediate outcomes (biochemical markers), and lacking statistical power to assess fractures. This methodological limitation prevents establishing robust causal relationships and underscores the urgent need for well-designed intervention studies with clear definitions of exposure and follow-up of relevance to clinical events. Studies with a duration shorter than 1 year are not considered reliable (38, 39) and were not included in our systematic review.

The systematic review and meta-analysis by Malmir et al. (73) included 34 studies (15 on osteoporosis and 21 on hip fracture) and performed separate analyses by study design. In cohort studies, no significant association was found between dairy product consumption and the risk of osteoporosis (RR = 0.82; 95% CI: 0.56–1.18) or hip fracture (RR = 0.90; 95% CI: 0.73–1.11), which is consistent with the results of our review. However, in case-control studies, yogurt consumption was associated with a 25% reduction in the risk of hip fracture (RR = 0.75; 95% CI: 0.57–0.99), suggesting that the observed effects could be influenced by selection or recall biases inherent to the study design. These authors also highlighted the heterogeneity in the definition of exposure and the lack of standardization in the portions and types of dairy products evaluated, a limitation that also affects the present review. However, the current systematic review incorporates dedicated sections aligned with EFSA guidance –such as characterization of the yogurt products, discussion of bioavailability of bioactive compounds, and mechanistic pathways– which may help to contextualize the observed associations and facilitate future standardization efforts in this research field.

Another aspect shared by all reviews is the lack of systematic assessment of adverse effects associated with yogurt consumption. This omission is particularly relevant considering the variability in the content of added sugars, saturated fats, and additives in commercial products, as well as potential intolerance to lactose or specific proteins such as A1 β-casein (55). The digestion of this casein isoform can generate bioactive peptides causing gastrointestinal pain or discomfort.

Finally, all reviews agree in pointing out the ambiguity in the definition of “yogurt” as a cross-sectional limitation. There is no systematic distinction between plain, Greek, probiotic-containing, sweetened, or fortified yogurt, which makes comparison of the results across studies and interpretation difficult. This lack of standardization, coupled with the conceptual breadth of the “dairy” category, dilutes the specific effect of yogurt and limits the applicability of the findings to produce dietary recommendations. In this sense, this review provides a more conservative and specific analysis of yogurt as a dairy product and highlights the need to move toward a precise characterization of the foods evaluated in nutritional studies.

This review has several limitations, primarily due to the observational nature of the included studies, which inherently carry risks of residual confounding and selection bias. The hip fracture meta-analysis was limited to only three studies, reducing statistical power and precluding meaningful subgroup analysis.

The variability in yogurt characterization across studies (e.g., dry matter, fat content, etc.) limits the capability to perform subgroup or dose-response analyses, which may be clinically relevant to identify more effective yogurt types, probiotic formulations, or consumption patterns for bone health outcomes. Addressing these gaps in future studies would enhance both the clinical interpretability and mechanistic understanding of the observed associations. These limitations further impede the mechanistic interpretation of the findings and the ability to identify potentially more effective yogurt formulations for bone health. While a statistically significant association was observed for BMD, the magnitude of the effect (SMD = 0.009) falls below clinical thresholds relevant for osteoporosis management or fracture prevention.

A critical gap identified in this review is the complete absence of RCT studies meeting the eligibility criteria. While the observational studies included provide suggestive evidence of a neutral to marginally positive association between yogurt consumption and bone health, they are insufficient to establish causality, which is pivotal to inform clinical or policy guidelines (74). This lack of trial-based evidence is especially relevant in the context of emerging fields such as nutrieconomics and nutrigenomics. Nutritional interventions –such as consumption of fermented dairy products– are increasingly being evaluated not only for their biological efficacy but also for their cost-effectiveness and potential for personalized implementation. As highlighted by Vélez-Cuellar et al. (75), nutrieconomic approaches are essential for informing public health strategies in resource-limited settings, especially regarding the prevention of chronic disease in older populations. Similarly, the work of Kassem et al. (76) underscores the potential of nutrigenomics and microbiome modulation in shaping personalized nutrition strategies. Fermented foods like yogurt can play a key role in this interface by influencing gut microbiota and interacting with host genetic pathways involved in bone and metabolic health. However, these mechanistic insights remain largely unexplored in clinical trials focusing on skeletal outcomes.

To advance the field, future research should prioritize the design and conduct of high-quality RCTs with the following features:

Addressing these gaps will provide more robust causal evidence and help move from observational associations toward evidence-based dietary recommendations that can inform both clinical practice and public health strategies.