Next Article in Journal
Unraveling the Role of Foods on Chronic Anti- and Pro-Inflammatory Cytokines: A Systematic Review of Chronic Dietary Intervention Trials in Humans
Previous Article in Journal
Associations Between Diet, Oral Health, and General Development in Romanian School-Age Children
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Nutrients
Volume 17
Issue 17
10.3390/nu17172833
nutrients-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Comparing the Effects of Dairy and Soybean on Bone Health in Women: A Food- and Component-Level Network Meta-Analysis

by
Li You
1,2,†,
Langrun Wang
1,†,
Shiwen Zhou
1,
Yiran Guan
1,
Yan Liu
3,4,5,
Ruixin Zhu
1,
Huiyu Chen
1,
Jie Guo
1,
Keji Li
6,
Xingyu Bao
3,4,5,
Haotian Feng
3,4,5,
Ignatius M. Y. Szeto
3,4,5,
Jian He
3,4,5,
Ran Wang
7,* and
Jingjing He
1,*
1
Key Laboratory of Precision Nutrition and Food Quality, Department of Nutrition and Health, China Agricultural University, Beijing 100193, China
2
College of Physical Education and Health, Chongqing College of International Business and Economics, Chongqing 401520, China
3
National Center of Technology Innovation for Dairy, Hohhot 010100, China
4
Inner Mongolia Dairy Technology Research Institute Co., Ltd., Hohhot 010100, China
5
Inner Mongolia Yili Industrial Group Co., Ltd., Hohhot 010100, China
6
Department of Nutrition and Food Hygiene, School of Public Health, Peking University, Beijing 100191, China
7
Research Center for Probiotics, China Agricultural University, Beijing 100193, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Nutrients 2025, 17(17), 2833; https://doi.org/10.3390/nu17172833
Submission received: 31 July 2025 / Revised: 22 August 2025 / Accepted: 29 August 2025 / Published: 30 August 2025
(This article belongs to the Special Issue Dietary Interventions for the Prevention and Treatment of Osteoporosis)
Browse Figures
Versions Notes

Abstract

Background/Objectives: Dairy and soybean are important potential dietary sources of bone health. However, their comparative effectiveness and the role of specific components remain unclear. In this network meta-analysis (NMA), we aimed to compare the effects of various dairy and soy products (food level) and their key bioactive components (component level) on bone health in healthy women. Methods: We systematically searched PubMed, Embase, Cochrane Library, and Web of Science (up to 28 February 2025) for randomized controlled trials. A frequentist random-effects NMA was used to compare interventions for lumbar spine (LS) and total body (TB) bone mineral density (BMD) and bone turnover markers [osteocalcin (OC), deoxypyridinoline (DPD)]. Mean differences (MDs) and 95% confidence intervals were pooled. Interventions were ranked using the surface under the cumulative ranking curve (SUCRA). Results: Sixty RCTs involving 6284 participants (mean age: 54.2 years) were included. At the food level, no dairy or soy interventions significantly improved outcomes versus control, although milk + yogurt ranked numerically highest based on SUCRA values. At the bioactive-component level, the combination of casein + whey protein (MD 0.04 g/cm2, 95% CI 0.01–0.06) and soybean protein (MD: 0.03 g/cm2, 95% CI: 0.01–0.05) significantly increased TB BMD. Whey protein alone (SUCRA 74.4% for LS BMD) and casein + whey protein (SUCRA 86.3% for TB BMD and 75.9% for DPD) were among the highest-ranked interventions for bone health. Conclusions: The combination of milk and yogurt may be relatively promising among dairy products for bone health. Whey protein appears to be a key bioactive component beneficial for women’s bone health.

1. Introduction

Bone is a dynamic tissue that undergoes continuous remodeling throughout life, involving formation and resorption [1,2]. While bone formation predominates during growth, bone resorption increases after age 40, particularly in women experiencing postmenopausal estrogen decline, elevating the risk of osteoporosis and fractures due to reduced bone mineral density (BMD) and strength [3,4]. Osteoporosis, characterized by low BMD and deteriorated bone microarchitecture, disproportionately affects women, highlighting the critical need for effective strategies to maintain bone health [5,6]. Beyond estrogen deficiency, obesity and related chronic low-grade inflammation are recognized as critical modifiable risk factors for osteoporosis [7]. These conditions disrupt the bone marrow microenvironment—altering stem cell differentiation, impairing osteogenesis while promoting adipogenesis—thereby compromising bone remodeling balance [8]. Furthermore, specific dietary patterns (e.g., “milk/dairy” patterns) have been associated with a reduced risk of low BMD and fracture [9,10]. Consequently, effective nutritional strategies for bone health should not only deliver key nutrients but also account for their potential to modulate systemic inflammatory status and be embedded within beneficial dietary patterns.
Dairy products (e.g., milk, yogurt, powder) and soybeans are widely consumed and recognized as potential dietary supports for bone health [11,12,13,14]. Dairy is rich in protein, calcium, and vitamin D [15], while soy provides high-quality plant protein and isoflavones with estrogen-like properties that may influence bone cell activity [16,17,18]. Critically, processing milk or soybeans into different forms (e.g., liquid milk vs. milk powder, whole soybeans vs. soymilk) alters the content and bioavailability of key components like whey proteins [19] and isoflavones [20]. Consequently, the effects of various dairy and soy products on bone health may differ. Furthermore, consumer preferences vary; individuals following vegan/vegetarian diets or with lactose intolerance/milk allergies may avoid dairy [21,22], creating a need to understand the relative efficacy of soy alternatives. However, the current evidence has dual limitations: it neither establishes the comparative superiority between dairy and soy products overall nor provides a systematic ranking of the efficacy of different product forms within each category (e.g., milk vs. yogurt vs. milk powder; soybeans vs. soymilk vs. tofu).
Beyond variations in product forms, elucidating the mechanistic roles of their core bioactive components is critical for developing targeted nutritional interventions. At the bioactive-component level, while guidelines recommend the consumption of protein, calcium, and vitamin D to support bone health [23], evidence regarding the efficacy of calcium or vitamin D supplementation alone remains inconsistent. Several systematic reviews and meta-analyses have demonstrated that vitamin D and calcium supplementation do not have beneficial effects on bone health [24,25], including in postmenopausal women [26]. In contrast, proteins have exhibited benefits for bone health, although the research results are heterogeneous [27,28,29]. Notably, milk protein and soy protein differ in their amino acid composition [30], and recent comparative studies of animal versus plant-based proteins have produced conflicting results, often without differentiating between specific protein types or focusing on women specifically [31]. The distinct effects of soy isoflavones also remain to be fully elucidated. Furthermore, although both protein and isoflavones are recognized as key bioactive components, their relative contributions—particularly when comparing specific milk proteins (such as whey and casein) with soy protein and isoflavones—have not been adequately evaluated.
Network meta-analysis (NMA) is an advanced statistical technique uniquely suited to address these gaps. Unlike traditional pairwise meta-analyses, NMA allows for the simultaneous comparison and ranking of multiple interventions by synthesizing both direct (head-to-head trials) and indirect evidence (trials connected through common comparators) within a unified analysis [32,33,34]. This capability is crucial for comprehensively evaluating the diverse range of dairy and soy products and their bioactive components relevant to bone health. This study thus aims to achieve the following outcomes: (1) at the food-class level, compare the effects of various dairy and soy products on bone health; and (2) at the bioactive-component level, evaluate the relative contributions of specific components (milk proteins: casein, whey, and milk basic protein; soy protein; soy isoflavones).

2. Materials and Methods

The present study adhered rigorously to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Statement [35]. It was registered in the Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42024560199 (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42024560199), as of 5 July 2024.

2.1. Search Strategy

We systematically searched PubMed, Embase, Cochrane Library, and Web of Science (up to 28 February 2025). The keywords and MeSH terms used in the search strategy included dairy products, milk, cultured milk products, butter, buttermilk, cheese, Kefir, Koumiss, yogurt, milk proteins, whey proteins, caseins, lactalbumin, lactoglobulins, lactoferrin, osteopontin, glycine max, soybean proteins, isoflavones, equol, bone density, bone remodeling, osteogenesis, bone resorption, and osteoporosis. The full search strategies used in the above databases are provided in Tables S1–S4.

2.2. Inclusion and Exclusion Criteria

Studies were included if they met the following criteria:
(1)
Participants were healthy female adults aged ≥18 years, which included those with overweight or obesity but excluded individuals with other diseases known to affect bone metabolism (see exclusion criteria 2 below).
(2)
Study design was a randomized controlled trial (RCT).
(3)
Intervention groups received single or combined treatments of dairy products (milk, yogurt, milk powder, cheese), milk-derived proteins (milk protein, casein, whey protein, milk basic protein), soybean, soy protein, or soy isoflavones.
(4)
Control groups received no treatment, placebo, or any intervention listed in (3).
(5)
Data for ≥1 bone health outcome (BMD or bone turnover markers) were reported or calculable.
Primary outcome: BMD of the whole body or any specific site.
Secondary outcomes: Bone turnover markers (osteocalcin [OC], bone-specific alkaline phosphatase [BAP], procollagen type I N-terminal propeptide [PINP], deoxypyridinoline [DPD], C-terminal telopeptide of type I collagen [CTx], N-terminal telopeptide of type I collagen [NTx], etc.) and bone metabolism hormones (parathyroid hormone [PTH], insulin-like growth factor 1 [IGF-1], 25-hydroxyvitamin D [(25(OH)D]).
Exclusion criteria:
(1)
Non-English publications, abstracts, letters, conference reports, or duplicate publications.
(2)
Studies involving women with any disease that affects bone metabolism, including renal failure, liver disorders, hyperparathyroidism, hyperthyroidism, diabetes mellitus, or cancer.
(3)
Interventions confounded by nontarget components (e.g., vitamin D).
(4)
Cointerventions (e.g., exercise programs).
(5)
No complete data regarding the effect sizes could be extracted from this study, nor could such information be derived through reliable methods.

2.3. Data Extraction

EndNote X9.2 was employed for the screening and management of the literature. Following predefined inclusion and exclusion criteria, two researchers (L.Y. and S.W.Z.) independently performed the initial screening using the titles and abstracts, and subsequently reviewed the full texts to decide on the inclusion of the studies. A third researcher (J.J.H.) was consulted in case of disagreement in the selection of studies.
Two researchers (S.W.Z. and L.Y.) independently extracted data as outlined in the Cochrane handbook. The extracted details included study title, authors, year of publication, country, participant description (race/ethnicity, age, and BMI), intervention period, sample size, treatment method for the intervention or the control group, and outcome indicators (mean and standard deviation [SD] of the change from baseline to endpoint in BMD and bone turnover markers). When mean was not reported, we estimate it using the median (quartile) or the median (minimum, maximum) as appropriate [36]. When SD was not directly reported, it was calculated from the standard error or confidence interval (CI) following the Cochrane Handbook for Systematic Reviews of Interventions [37]. If the mean or SD could not be obtained through a reliable method, the data were not included in the meta-analysis. Disagreements were resolved through consensus with the third researcher (J.J.H.).

2.4. Risk-of-Bias Assessment

The quality of the included studies was assessed using the Risk of Bias tool (RoB 2) and the Cochrane Risk of Bias Assessment Tool [38]. The evaluation criteria encompassed five components: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Each component was assessed as ‘low risk’, ‘high risk’, or ‘uncertain’ [34,39]. The process of evaluating the quality of the literature was carried out independently by two researchers (L.Y. and L.R.W.), and in case of disagreement, a third researcher (J.J.H.) was added to discuss the matter together until there was unanimity of opinion.

2.5. Statistical Analysis

We conducted two independent frequentist random-effects network meta-analyses (NMAs) using Stata 17.0 and R 4.3.1: one at the food-class level (comparing dairy products and soy products) and another at the bioactive-component level (comparing milk-derived protein, soybean protein, and soy isoflavones). Empirical and simulation studies have demonstrated that, in most scenarios, the frequentist and Bayesian NMA methodologies typically yield consistent results [40,41]. We chose the frequentist approach due to its ease of implementation and the availability of a comprehensive set of tools in the Stata software platform. Furthermore, this approach is consistent with the practices adopted in recent high-quality network meta-analyses [42].
Statistical significance was set at p < 0.05. Based on the overall characteristics of the included literature, this study’s NMAs focused on the BMD values of LS and TB, as well as the bone turnover markers OC and DPD. For the BMD of other body parts and other bone turnover markers such as CTx and P1NP, NMA was not performed due to a limited number of original studies or intervention comparisons.
All outcome measures in this study were continuous variables; therefore, the mean differences (MDs) and SDs between the intervention group and the control group were utilized as effect sizes for each trial comparison. We employed a design-by-treatment interaction model as the global approach and a loop-specific method alongside a side-splitting model as the local approach to assess inconsistency within the network [43,44,45]. If the p-value from the chi-square test in the global inconsistency assessment was <0.05, a consistency model was adopted for the NMA. Conversely, an inconsistency model was utilized, necessitating further investigation into potential sources of inconsistency. For each outcome measure with sufficient data, the surface under the cumulative ranking curve (SUCRA) was used to rank the efficacy of different interventions, with higher values (ranging from 0% to 100%) indicating superior efficacy.
Sources of heterogeneity were explored using subgroup analysis and sensitivity analysis. Subgroup analysis was performed for each outcome that included at least three studies according to BMI values (<24.0 vs. ≥24.0 kg/m2). When significant local inconsistencies arise during loop analysis, sensitivity analyses were performed by excluding studies contributing to inconsistent loops. Publication bias was assessed using funnel plots combined with Egger’s tests for outcome measures involving more than 10 comparisons.

3. Results

3.1. Literature Search and Screening

Initial database searches identified 5784 records. After removing duplicates (EndNote X9.2), 3937 records remained. Screening of titles and abstracts excluded 3632 records as irrelevant. Full-text assessment of the remaining 305 records excluded 245 records not meeting the inclusion criteria, resulting in 60 records for final inclusion (Figure 1).

3.2. Study Characteristics

As shown in Table 1, the included RCTs (1995–2022) involved 6284 healthy women. The mean age of the participants was 54.2 years, with a range from 19.6 to 74.2 years. The average body mass index (BMI) was 25.1 kg/m2, ranging from 20.4 to 32.2 kg/m2. The duration of the interventions averaged 11.3 months.
Thirteen intervention groups were analyzed: control (placebo/no treatment), soybean protein, isoflavone, soybean protein + isoflavone, milk basic protein (MBP), casein, whey protein, casein + whey protein, milk, yogurt, milk + yogurt, milk powder, soymilk.
Table S5 summarizes the distribution of included study outcomes. Among the sixty included RCTs, BMD was assessed for the lumbar spine (LS; 32 RCTs); total body (TB; 20 RCTs); femoral neck (FN; 24 RCTs); total hip (TH; 16 RCTs); trochanter (13 RCTs); intertrochanter and ward triangle (WT; 10 RCTs each); total spine (TS; 4 RCTs); whole femurs (WF; 3 RCTs); and arms, femoral trochanter, legs, and pelvis (2 RCTs each). Single RCTs assessed BMD at additional forearm sites (distal radius and ulna 1/10, forearm 33% radius, radius 1/3 of styloid process, total radius, ultradistal radius, whole forearms).
Bone turnover markers were reported as follows: OC (23 RCTs), DPD (18 RCTs), 25(OH)D (14 RCTs), CTx (16 RCTs), NTx (12 RCTs), BAP (19 RCTs), PTH (12 RCTs), PINP (10 RCTs), IGF-1 (8 RCTs), and pyridinoline (6 RCTs).

3.3. Network Meta-Analysis of the Impact of Interventions on Bone Health

As shown in Table 1, the included RCTs (1995–2022) involved 6284 healthy women. The mean age of the participants was 54.2 years, with a range from 19.6 to 74.2 years. The average BMI was 25.1 kg/m2, ranging from 20.4 to 32.2 kg/m2. The duration of the interventions averaged 11.3 months.

3.3.1. Food-Class Level: Effects of Dairy and Soybean Products on Bone Health Outcomes

The NMA on LS BMD included 5 RCTs (n = 458) evaluating five intervention pairs, forming one closed loop (Figure 2a and Table S6). The outcomes of both direct and indirect assessments within the NMA are presented in Figure 3a and Figure S15. No dairy interventions significantly differed from control, though milk + yogurt ranked numerically highest (SUCRA: 73.1%) (Table 2). The LS BMD analysis showed good global consistency (p = 0.325), supporting the use of the consistency model (Figure S4). Closed-loop analysis revealed no local inconsistencies (all p > 0.05; IF = 0.078; Table S12). Table S6 presents the local inconsistencies from “direct versus indirect” comparisons in five intervention pairs, using the side-splitting model.
For TB BMD, the NMA of six RCTs (n = 834) assessed three intervention pairs without closed loops (Figure 2b and Table S7). No dairy interventions significantly differed from control, though milk + yogurt ranked numerically highest (SUCRA: 81.9%) (Table 2). No closed loops were detected.
Analyses of bone turnover markers revealed no significant effects for dairy interventions on OC (7 RCTs, n = 944) or DPD (3 RCTs, n = 295), yet milk consistently exhibited the highest SUCRA values (OC: 62.3%; DPD: 69.1%) (Figure 2c,d and Figure 3a,b, Table 2).

3.3.2. Bioactive-Component Level: Effects of Milk-Derived Proteins, Soybean Proteins, and Soybean Isoflavones on Bone Health Outcomes

At the bioactive-component level, LS BMD analysis (22 RCTs, n = 2159) evaluated twelve intervention pairs with six closed loops (Figure 2e and Table S6). No interventions significantly outperformed control, but whey protein (SUCRA: 74.4%), soybean protein (SUCRA: 43.9%), and casein + whey protein (SUCRA: 60.8%) ranked the highest (Figure 3c, Table 2). LS BMD exhibited good global consistency (p = 0.255), supporting the consistency model (Figure S5). Loop inconsistency was detected in three loops (p < 0.05; Table S12). Table S6 presents the local inconsistency from “direct versus indirect” comparisons in twelve intervention pairs using the side-splitting model.
For TB BMD (12 RCTs, n = 1220), the NMA evaluated nine intervention pairs with five closed loops (Figure 2f and Table S7). Casein + whey protein significantly increased BMD versus control (MD: 0.04 g/cm2, 95% CI: 0.01–0.06) and soy isoflavones (MD: 0.04 g/cm2, 95% CI: 0.01–0.07). Soybean protein also exceeded control (MD: 0.03 g/cm2, 95% CI: 0.01–0.05) and isoflavones alone (MD: 0.03 g/cm2, 95% CI: 0.02–0.05). Casein + whey protein ranked highest (SUCRA: 86.3%), followed by soybean protein (SUCRA: 79.7%) and MBP (SUCRA: 63.2%). (Figure 3c, Table 2). TB BMD showed good global consistency (p = 0.169), supporting the consistency model (Figure S7). No loop inconsistency was found (all p > 0.05; IF range: 0.005–0.053; Table S13). Table S7 presents the local inconsistencies from the “direct versus indirect” comparisons in nine intervention pairs using the side-splitting model.
Among bone turnover markers, casein + whey protein showed the highest efficacy for OC (SUCRA: 75.9%; 15 RCTs, n = 918), while soybean protein + isoflavone ranked highest for DPD (SUCRA: 61.0%; 13 RCTs, n = 1168), though no interventions significantly differed from control (Figure 2g,h and Figure 3d; Table 2, Tables S8 and S9). OC analysis indicated good global consistency (p = 0.322), supporting the consistency model (Figure S9). Loop inconsistency was detected in one loop (p < 0.05; Table S14). Table S8 presents the local inconsistency from “direct versus indirect” comparisons in seven intervention pairs using the side-splitting model. No closed loops were detected in the DPD analysis.

3.4. Subgroup Analysis and Sensitivity Analysis

Subgroup analyses were conducted based on baseline BMI (<24 vs. ≥24 kg/m2) to explore its role as a potential effect modifier, as suggested in the Methods. The results are detailed in Tables S16–S18. Notably, in participants with a BMI ≥ 24 kg/m2, the ranking of bioactive components was highly consistent with the primary findings from the overall analysis. The combination of casein + whey protein (SUCRA: 90.3%) and soybean protein (SUCRA: 80.4%) remained the most effective interventions for improving TB BMD, and whey protein (SUCRA: 72.5%) remained the most effective for improving LS BMD, underscoring the robustness of these findings in women with higher BMI. In contrast, a different pattern emerged in the lower BMI subgroup (<24 kg/m2), where MBP demonstrated the highest efficacy for both LS and TB BMD (SUCRA: 90.3% and 85.3%, respectively). For bone turnover markers, the rankings also varied between subgroups. Isoflavone ranked highest for OC in the lower BMI group, while casein + whey protein and isoflavone were top-ranked for OC and DPD, respectively, in the higher BMI group.
To assess the robustness of the primary network against local inconsistencies, sensitivity analyses were performed by excluding studies contributing to inconsistent loops. As shown in Tables S19–S21, the exclusion of these data did not materially alter the effect estimates or the relative SUCRA rankings, confirming the stability and reliability of our main conclusions.

3.5. Risk-of-Bias Assessment

The results of the RCT risk-of-bias evaluation are shown in Figures S1 and S2. Of the sixty RCTs, the majority of studies were classified as “low risk” in the categories “Missing outcome data” (49 studies, 81.7%), “Selection of the reported result” (48 studies, 80%), “Measurement of the outcome” (47 studies, 78.3%), and “Deviations from intended interventions” (42 studies, 70.0%). Conversely, a significant number of studies raised “some concerns” in the “Randomization process” category (33 studies, 55%).

3.6. Publication Bias

Publication bias was assessed for outcomes with sufficient data (≥10 studies) using funnel plots and formal statistical testing via Egger’s test. As shown in Figure S3, the funnel plots for LS, TB, OC, and DPD all demonstrated approximate symmetry. This visual interpretation was corroborated by Egger’s test, which yielded non-significant p-values (all p > 0.05), indicating no substantial evidence of publication bias.

4. Discussion

This NMA study represents the first comprehensive evaluation of the effects of both food-level (dairy vs. soy products) and component-level (specific proteins and isoflavones) interventions on bone health in healthy women. By synthesizing direct and indirect evidence from 60 RCTs involving 6284 participants, we established two principal findings. First, at the food level, no dairy or soy product was found to have a significant effect in improving BMD or bone turnover markers compared to control, although milk + yogurt or milk ranked numerically highest. Second, at the bioactive-component level, whey protein—either alone or combined with casein—and soy protein emerged as the highest-ranked interventions. Furthermore, subgroup analysis suggests that BMI may influence the effect of certain components on bone health, with MBP demonstrating more pronounced benefits in individuals with a lower BMI. These findings provide nuanced insights for developing targeted dietary strategies to maintain bone health.
The lack of statistically significant differences at the food level, contrary to some previous hypotheses, may be attributed to several factors. The inherent heterogeneity in the forms of dairy and soy products (e.g., liquid milk vs. powder, fortified vs. non-fortified, varying fat content) likely diluted the specific effects of any single food category. Furthermore, the control groups in many studies may have had relatively adequate baseline nutrient intakes, diminishing the observable benefit of additional dietary interventions. Despite the absence of statistical significance, the consistently highest ranking of the milk + yogurt combination is noteworthy. This may be due to a synergistic effect: the components in milk (calcium, protein, vitamin D) are complemented by the probiotics and fermentation products in yogurt, which may enhance calcium bioavailability, reduce systemic inflammation, and positively modulate the gut–bone axis, potentially leading to more favorable bone remodeling outcomes [12,15,105]
As essential bioactive ingredients that benefit bone health, calcium and vitamin D have long been regarded as crucial [15]. However, recent studies have indicated that these two nutrients may not confer the anticipated substantial benefits [24,25,26]. Conversely, the role of high-quality protein in maintaining bone health has increasingly garnered attention [27,28,29]. Previous systematic reviews have generally compared animal and plant proteins without differentiating between specific protein types [31]. In our component-level NMA, whey protein appeared to be central to dairy’s osteoprotective effects. Its highest ranking for LS BMD and superior performance for TB BMD when combined with casein underscore its biological potency. Mechanistically, whey’s rapid digestibility and rich essential amino acid composition enhance anabolic signaling pathways, while its bioactive peptides may stimulate osteoblast activity through insulin-like growth factor-1 upregulation and activation of the Runx2/GSK-3β/Nrf2 pathway [106,107,108]. Interestingly, soy protein ranked second in efficacy for multiple bone health outcomes, outperforming isolated casein or milk basic protein. Animal models corroborate this hierarchy, showing soy-fed subjects exhibit better bone metrics than casein-fed counterparts, potentially due to isoflavone-enhanced osteogenic activity [109]. However, our analysis revealed that isoflavones alone showed minimal efficacy, suggesting that their action requires the synergistic presence of soy protein rather than functioning as isolated agents. Furthermore, the biological effects of isoflavones are known to vary significantly due to metabolic differences, particularly between individuals who can produce equol and those who cannot [110]. This variability in bioavailability likely contributes to the lack of observed significant effect of isoflavones in our overall analysis. Baseline dietary calcium/protein intake may also influence the intervention outcomes. Previous studies have indicated that baseline levels of dietary calcium and protein can modulate each other’s beneficial effects on bone health. Specifically, in postmenopausal women, the dietary calcium-to-protein ratio has potential to be a critical factor in bone health interventions [111,112]. Additionally, there are notable regional disparities worldwide in the intake of key nutrients, including calcium [113]. However, as none of the original studies included in this network meta-analysis reported data on equol producer status, and comprehensive data on nutritional intake were unavailable, we were unable to conduct further subgroup analyses to evaluate these hypotheses, restricting the generalizability of our conclusions.
An intriguing observation pertains to the discordance between BMD outcomes and bone turnover markers. While multiple interventions significantly improved BMD, their effects on OC and DPD were inconsistent. This discrepancy may reflect several factors. Studies measuring turnover markers typically employed shorter intervention periods (often <6 months) compared to BMD trials, potentially capturing only acute-phase remodeling rather than structural adaptation. Additionally, biochemical markers exhibit greater analytical variability and diurnal fluctuations than dual-energy X-ray absorptiometry measurements. Furthermore, effective interventions may normalize elevated bone resorption without suppressing formation markers, maintaining metabolic equilibrium while progressively improving bone density.
Our subgroup analysis based on BMI revealed important differences in the efficacy of bioactive components. In women with a lower BMI (<24 kg/m2), MBP demonstrated the highest efficacy for improving BMD. This suggests that leaner individuals, who may have a more efficient anabolic response and lower inflammatory burden, might benefit more from specific bioactive peptides like those found in MBP. Conversely, in women with higher BMI (≥24 kg/m2), the whey protein and the combination of whey protein and casein were most effective for LS and TB BMD. This could be attributed to whey protein’s well-documented anti-inflammatory and antioxidant properties, which may help mitigate the chronic low-grade inflammation associated with obesity—a known contributor to bone resorption and impaired bone remodeling [7,108]. These findings suggest that body composition may be an important effect modifier when recommending protein supplements for bone health.
The methodological approach of this study presents notable advantages compared to previous systematic reviews. By employing a dual-level NMA framework, we simultaneously addressed food-level and component-level questions. The SUCRA values offer clinically meaningful insights into the relative efficacy of various interventions. At the practical level, findings at the food level can inform dietary recommendations (e.g., recommending powdered dairy products for lactose-tolerant women), whereas component-level results contribute valuable guidance for the development of functional foods (e.g., whey-fortified products tailored for bone health).
This study also has important limitations warranting consideration when interpreting these findings. First, many intervention comparisons relied solely on indirect evidence due to the scarcity of head-to-head trials. This, unfortunately, led to several local inconsistencies in certain intervention pairs, potentially undermining confidence in those specific estimates. To address this, we conducted sensitivity analyses by excluding the inconsistent loops, which confirmed that the overall SUCRA rankings and effect estimates remained stable. Furthermore, no global inconsistency was detected across the networks, which together support the robustness of our primary conclusions. Second, more than half (55%) of the trials raised some concerns regarding the randomization process according to the ROB2 assessment, potentially compromising the robustness of the evidence. Third, considerable variability was observed in intervention characteristics—including differences in dosage, formulation matrices, and baseline nutrient status—which may contribute to clinical heterogeneity. Fourth, our analysis was limited to surrogate markers of bone health, specifically BMD and biochemical turnover markers (e.g., OC, DPD), as data on microarchitectural properties and fracture outcomes were unavailable in the included trials. Although improvements in these markers are biologically plausible and widely used, their ability to predict fracture risk is limited and non-linear. For example, BMD changes only partially explain anti-fracture effects in interventions. Thus, while certain interventions improved bone density and metabolism, their actual clinical benefit for fracture prevention remains uncertain. Fifth, the analysis could not stratify effects based on key physiological variables such as menopausal status or genetic polymorphisms influencing nutrient metabolism, limiting its applicability for personalized interventions. Finally, generalizability to populations with pre-existing bone pathologies remains uncertain, as the analysis included only healthy women.
Future research should address these gaps using several approaches. Several intervention pairs that currently have limited evidence for direct comparisons are worth further investigation. Clinical trials should prioritize long-term intervention periods with fracture outcomes or incorporate three-dimensional bone imaging endpoints to more directly assess bone quality and strength. Direct comparator trials evaluating whey protein against soy protein, or milk powder versus fortified soymilk, would strengthen the evidence base. Mechanistic human studies examining whey’s effects on bone-related signaling pathways would further elucidate its osteoprotective actions.

5. Conclusions

In conclusion, our findings suggest that at the food level, while no specific dairy or soy product significantly outperformed control, the combination of milk and yogurt showed the highest ranking potential for benefiting bone health. At the bioactive-component level, whey protein (alone or combined with casein) and soybean protein are the most effective interventions for enhancing BMD outcomes. The efficacy of these components may be influenced by BMI, highlighting the potential for personalized nutrition strategies. These results offer evidence-based guidance for dietary choices supporting women’s bone health, suggesting the possible benefits of milk + yogurt consumption and identifying whey protein as a key functional component. Soybean protein presents a viable alternative for non-dairy consumers. Future research should focus on direct comparisons of these high-ranking interventions and investigate their effects on bone microarchitecture and fracture risk.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/nu17172833/s1, Table S1: Search strategy in PubMed; Table S2: Search strategy in Cochrane Library; Table S3: Search strategy in Embase; Table S4: Search strategy in Web of Science; Table S5: Summary of included research outcomes; Table S6: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for LS BMD; Table S7: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for TB BMD; Table S8: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for OC; Table S9: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for DPD; Table S10: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for CTx; Table S11: Inconsistency test between direct and indirect treatment comparisons in mixed treatment comparison for P1NP; Table S12: Loop-specific heterogeneity of LS BMD; Table S13: Loop-specific heterogeneity of TB BMD; Table S14: Loop-specific heterogeneity of OC; Table S15: Comparative results of network meta-analysis of bone turnover markers CTx and P1NP in the included studies; Table S16: Comparative results of subgroup analysis in LS and TB BMD; Table S17: Comparative results of subgroup analysis in bone turnover markers OC and DPD; Table S18: Subgroup analysis of intervention rankings using surface under the cumulative ranking (SUCRA) values; Table S19: Comparative results of sensitivity analysis in LS and TB BMD (excluding MBP); Table S20: Comparative results of sensitivity analysis in bone turnover markers OC and DPD (excluding MBP); Table S21: Sensitivity analysis of intervention rankings using surface under the cumulative ranking (SUCRA) values (excluding MBP); Figure S1: Graph of Cochrane risk-of-bias assessment; Figure S2: Bias risk of the included studies; Figure S3: Publication bias assessed via funnel plots; Figure S4: Global inconsistency test for LS BMD of dairy products vs. soybean; Figure S5: Global inconsistency test for LS BMD of milk-derived protein vs. soy protein and isoflavone; Figure S6: Global inconsistency test for TB BMD of dairy products vs. soybean; Figure S7: Global inconsistency test for TB BMD of milk-derived protein vs. soy protein and isoflavone; Figure S8: Global inconsistency test for OC of dairy products vs. soybean; Figure S9: Global inconsistency test for OC of milk-derived protein vs. soy protein; Figure S10: Global inconsistency test for DPD of dairy products vs. soybean; Figure S11: Global inconsistency test for DPD of milk-derived protein vs. soy; Figure S12: Global inconsistency test for CTx of dairy products vs. soybean; Figure S13: Global inconsistency test for CTx of milk-derived protein vs. soy; Figure S14: Global inconsistency test for PINP of dairy products vs. soybean; Figure S15: Effect size for LS BMD of dairy vs. soybean using forest plots; Figure S16: Effect size for LS BMD of milk-derived protein vs. soy protein and isoflavone using forest plots; Figure S17: Effect size for TB BMD of dairy products using forest plots; Figure S18: Effect size for TB BMD of milk-derived protein vs. soy protein and isoflavone using forest plots; Figure S19: Effect size for OC of dairy products using forest plots; Figure S20: Effect size for OC of milk-derived protein vs. soy protein and isoflavone using forest plots; Figure S21: Effect size for DPD of dairy products using forest plots; Figure S22: Effect size for DPD of soy protein vs. isoflavone using forest plots; Figure S23: Effect size for CTx of dairy products using forest plots; Figure S24: Effect size for CTx of milk-derived protein vs. soy isoflavone using forest plots; Figure S25: Effect size for P1NP of dairy products using forest plots; Figure S26: Ranking diagram of each intervention for lumbar spine (LS) BMD and total body (TB) BMD; Figure S27: Ranking diagram of each intervention for osteocalcin (OC) and deoxypyridinoline and (DPD); Figure S28: Ranking diagram of each intervention for C-terminal telopeptide of type I collagen (CTx) and procollagen type I N-terminal propeptide (P1NP); Figure S29: Network plots for C-terminal telopeptide of type I collagen (CTx) and procollagen type I N-terminal propeptide (P1NP).

Author Contributions

Conceptualization, J.H. (Jian He), R.W. and I.M.Y.S.; methodology, J.H. (Jian He) and L.Y.; software, L.Y. and L.W.; validation, S.Z., Y.G. and R.Z.; formal analysis, L.Y., L.W., R.Z. and H.C.; investigation, L.Y., L.W., S.Z. and Y.G.; resources, J.H. (Jingjing He) and R.W.; data curation, L.Y., Y.L. and J.G.; writing—original draft preparation, L.Y., L.W. and S.Z.; writing—review and editing, Y.L., K.L., X.B., H.F., J.H. (Jian He) and J.H. (Jingjing He); visualization, L.Y. and L.W.; supervision, J.H. (Jian He) and R.W.; project administration, L.Y., J.H. (Jian He) and J.H. (Jingjing He); funding acquisition, J.H. (Jian He) and J.H. (Jingjing He). All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Key Research and Development Program: “Precision Nutrition and Health Food Development and Industrial Demonstration for Typical Chronic Disease Populations” (No. 2024YFF1106000), the National Center of Technology Innovation for Dairy (No. 2022-KYGG-5), and the 9th China Association for Science and Technology Youth Talent Promotion Project 2023-2026 (No. 202404623140567).

Data Availability Statement

The data presented in the manuscript will be provided upon request by contacting the corresponding authors. The data are not publicly available due to ethical and participant privacy.

Conflicts of Interest

Ran Wang reports grants from the National Center of Technology Innovation for Dairy during the study period. Yan Liu, Xingyu Bao, Haotian Feng, Ignatius M.Y. Szeto, and Jian He were employed by the National Center of Technology Innovation for Dairy, Inner Mongolia Dairy Technology Research Institute Co., Ltd., and Inner Mongolia Yili Industrial Group Co., Ltd. They contributed to the conceptualization, methodology, validation, formal analysis, resources, data curation, visualization, and project administration of this study in their professional capacity as research scientists, based on their expertise in nutrition and programming. This research aims to “compare the effects of various dairy and soy products (food level) and their key bioactive components level) on bone health in healthy women”, which is unrelated to the commercial activities of the National Center of Technology Innovation for Dairy. The funder did not impose any restrictions or requirements related to commercial interests during the research process. All other authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
BAPbone-specific alkaline phosphatase
BMDbone mineral density
CTxC-terminal telopeptide of type I collagen
DPDdeoxypyridinoline
IGF-1insulin-like growth factor 1
LSlumbar spine
MBPmilk basic protein
MDsmean differences
NMAnetwork meta-analysis
NTxN-terminal telopeptide of type I collagen
OCosteocalcin
PINPprocollagen type I N-terminal propeptide
PROSPEROProspective Register of Systematic Reviews
PTHparathyroid hormone
Pyrpyridinoline
RCTsrandomized controlled trials
SDstandard deviations
SEstandard errors
SUCRAsurface under the cumulative ranking curve
TBtotal body
25(OH)D25-hydroxyvitamin D

References

  1. Shetty, S.; Kapoor, N.; Bondu, J.D.; Thomas, N.; Paul, T.V. Bone turnover markers: Emerging tool in the management of osteoporosis. Indian J. Endocrinol. Metab. 2016, 20, 846–852. [Google Scholar] [CrossRef]
  2. Eastell, R.; Szulc, P. Use of bone turnover markers in postmenopausal osteoporosis. Lancet Diabetes Endocrinol. 2017, 5, 908–923. [Google Scholar] [CrossRef]
  3. Akhlaghi, M.; Ghasemi Nasab, M.; Riasatian, M.; Sadeghi, F. Soy isoflavones prevent bone resorption and loss, a systematic review and meta-analysis of randomized controlled trials. Crit. Rev. Food Sci. Nutr. 2019, 60, 2327–2341. [Google Scholar] [CrossRef]
  4. Crandall, C.J.; Larson, J.; Wright, N.C.; Laddu, D.; Stefanick, M.L.; Kaunitz, A.M.; Watts, N.B.; Wactawski-Wende, J.; Womack, C.R.; Johnson, K.C.; et al. Serial bone density measurement and incident fracture risk discrimination in postmenopausal women. JAMA Intern. Med. 2020, 180, 1232–1240. [Google Scholar] [CrossRef]
  5. Cheng, C.H.; Chen, L.R.; Chen, K.H. Osteoporosis due to hormone imbalance: An overview of the effects of estrogen deficiency and glucocorticoid overuse on bone turnover. Int. J. Mol. Sci. 2022, 23, 1376. [Google Scholar] [CrossRef]
  6. Kanis, J.A.; McCloskey, E.V.; Johansson, H.; Cooper, C.; Rizzoli, R.; Reginster, J.Y. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos. Int. 2013, 24, 23–57. [Google Scholar] [CrossRef]
  7. Benova, A.; Tencerova, M. Obesity-Induced Changes in Bone Marrow Homeostasis. Front. Endocrinol. 2020, 11, 294. [Google Scholar] [CrossRef] [PubMed]
  8. Moscatelli, F.; Monda, A.; Messina, G.; Picciocchi, E.; Monda, M.; Di Padova, M.; Monda, V.; Mezzogiorno, A.; Dipace, A.; Limone, P.; et al. Exploring the Interplay between Bone Marrow Stem Cells and Obesity. Int. J. Mol. Sci. 2024, 25, 2715. [Google Scholar] [CrossRef] [PubMed]
  9. Denova-Gutiérrez, E.; Méndez-Sánchez, L.; Muñoz-Aguirre, P.; Tucker, K.L.; Clark, P. Dietary Patterns, Bone Mineral Density, and Risk of Fractures: A Systematic Review and Meta-Analysis. Nutrients 2018, 10, 1922. [Google Scholar] [CrossRef]
  10. Fabiani, R.; Naldini, G.; Chiavarini, M. Dietary Patterns in Relation to Low Bone Mineral Density and Fracture Risk: A Systematic Review and Meta-Analysis. Adv. Nutr. 2019, 10, 219–236. [Google Scholar] [CrossRef] [PubMed]
  11. Matía-Martín, P.; Torrego-Ellacuría, M.; Larrad-Sainz, A.; Fernández-Pérez, C.; Cuesta-Triana, F.; Rubio-Herrera, M. Effects of Milk and Dairy Products on the Prevention of Osteoporosis and Osteoporotic Fractures in Europeans and Non-Hispanic Whites from North America: A Systematic Review and Updated Meta-Analysis. Adv. Nutr. 2019, 10, S120–S143. [Google Scholar] [CrossRef]
  12. Wallace, T.C.; Bailey, R.L.; Lappe, J.; O’Brien, K.O.; Wang, D.D.; Sahni, S.; Weaver, C.M. Dairy intake and bone health across the lifespan: A systematic review and expert narrative. Crit. Rev. Food Sci. Nutr. 2021, 61, 3661–3707. [Google Scholar] [CrossRef]
  13. Hidayat, K.; Chen, J.S.; Wang, T.C.; Liu, Y.J.; Shi, Y.J.; Su, H.W.; Liu, B.; Qin, L.Q. The Effects of Milk Supplementation on Bone Health Indices in Adults: A Meta-Analysis of Randomized Controlled Trials. Adv. Nutr. 2022, 13, 1186–1199. [Google Scholar] [CrossRef]
  14. Lydeking-Olsen, E.; Beck-Jensen, J.E.; Setchell, K.D.; Holm-Jensen, T. Soymilk or progesterone for prevention of bone loss—A 2 year randomized, placebo-controlled trial. Eur. J. Nutr. 2004, 43, 246–257. [Google Scholar] [CrossRef]
  15. Rizzoli, R. Dairy products, yogurts, and bone health. Am. J. Clin. Nutr. 2014, 99, 1256s–1262s. [Google Scholar] [CrossRef]
  16. Kostelac, D.; Rechkemmer, G.; Briviba, K. Phytoestrogens modulate binding response of estrogen receptors alpha and beta to the estrogen response element. J. Agric. Food Chem. 2003, 51, 7632–7635. [Google Scholar] [CrossRef]
  17. Margeat, E.; Bourdoncle, A.; Margueron, R.; Poujol, N.; Cavaillès, V.; Royer, C. Ligands differentially modulate the protein interactions of the human estrogen receptors alpha and beta. J. Mol. Biol. 2003, 326, 77–92. [Google Scholar] [CrossRef] [PubMed]
  18. Zheng, X.; Lee, S.K.; Chun, O.K. Soy Isoflavones and Osteoporotic Bone Loss: A Review with an Emphasis on Modulation of Bone Remodeling. J. Med. Food 2016, 19, 1–14. [Google Scholar] [CrossRef] [PubMed]
  19. Haas, J.; Kim, B.J.; Atamer, Z.; Wu, C.; Dallas, D.C. Effects of spray drying and freeze drying on the protein profile of whey protein concentrate. J. Food Sci. 2024, 89, 7477–7493. [Google Scholar] [CrossRef] [PubMed]
  20. Zhang, Y.; Chang, S.K.; Liu, Z. Isoflavone profile in soymilk as affected by soybean variety, grinding, and heat-processing methods. J. Food Sci. 2015, 80, C983–C988. [Google Scholar] [CrossRef]
  21. Kent, G.; Kehoe, L.; Flynn, A.; Walton, J. Plant-based diets: A review of the definitions and nutritional role in the adult diet. Proc. Nutr. Soc. 2022, 81, 62–74. [Google Scholar] [CrossRef]
  22. Lee, J.T.; Moore, C.E.; Radcliffe, J.D. Consumption of calcium-fortified cereal bars to improve dietary calcium intake of healthy women: Randomized controlled feasibility study. PLoS ONE 2015, 10, e0125207. [Google Scholar] [CrossRef]
  23. Kanis, J.A.; Cooper, C.; Rizzoli, R.; Reginster, J.Y. Executive summary of the European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Calcif. Tissue Int. 2019, 104, 235–238. [Google Scholar] [CrossRef]
  24. Zhao, J.G.; Zeng, X.T.; Wang, J.; Liu, L. Association Between Calcium or Vitamin D Supplementation and Fracture Incidence in Community-Dwelling Older Adults: A Systematic Review and Meta-analysis. JAMA 2017, 318, 2466–2482. [Google Scholar] [CrossRef] [PubMed]
  25. Reid, I.R.; Bolland, M.J.; Grey, A. Effects of vitamin D supplements on bone mineral density: A systematic review and meta-analysis. Lancet 2014, 383, 146–155. [Google Scholar] [CrossRef] [PubMed]
  26. Méndez-Sánchez, L.; Clark, P.; Winzenberg, T.M.; Tugwell, P.; Correa-Burrows, P.; Costello, R. Calcium and vitamin D for increasing bone mineral density in premenopausal women. Cochrane Database Syst. Rev. 2023, 1, Cd012664. [Google Scholar] [CrossRef] [PubMed]
  27. Shams-White, M.M.; Chung, M.; Du, M.; Fu, Z.; Insogna, K.L.; Karlsen, M.C.; LeBoff, M.S.; Shapses, S.A.; Sackey, J.; Wallace, T.C.; et al. Dietary protein and bone health: A systematic review and meta-analysis from the National Osteoporosis Foundation. Am. J. Clin. Nutr. 2017, 105, 1528–1543. [Google Scholar] [CrossRef]
  28. Darling, A.L.; Manders, R.J.F.; Sahni, S.; Zhu, K.; Hewitt, C.E.; Prince, R.L.; Millward, D.J.; Lanham-New, S.A. Dietary protein and bone health across the life-course: An updated systematic review and meta-analysis over 40 years. Osteoporos. Int. 2019, 30, 741–761. [Google Scholar] [CrossRef]
  29. Hidayat, K.; Tong, X.; Rizzoli, R.; Fan, J.B.; Shi, Y.J.; Su, H.W.; Liu, B.; Qin, L.Q. The skeletal safety of milk-derived proteins: A meta-analysis of randomized controlled trials. Osteoporos. Int. 2023, 34, 1937–1949. [Google Scholar] [CrossRef]
  30. Gorissen, S.H.M.; Crombag, J.J.R.; Senden, J.M.G.; Waterval, W.A.H.; Bierau, J.; Verdijk, L.B.; van Loon, L.J.C. Protein content and amino acid composition of commercially available plant-based protein isolates. Amino Acids 2018, 50, 1685–1695. [Google Scholar] [CrossRef]
  31. Shams-White, M.M.; Chung, M.; Fu, Z.; Insogna, K.L.; Karlsen, M.C.; LeBoff, M.S.; Shapses, S.A.; Sackey, J.; Shi, J.; Wallace, T.C.; et al. Animal versus plant protein and adult bone health: A systematic review and meta-analysis from the National Osteoporosis Foundation. PLoS ONE 2018, 13, e0192459. [Google Scholar] [CrossRef]
  32. Mills, E.J.; Ioannidis, J.P.; Thorlund, K.; Schünemann, H.J.; Puhan, M.A.; Guyatt, G.H. How to use an article reporting a multiple treatment comparison meta-analysis. JAMA 2012, 308, 1246–1253. [Google Scholar] [CrossRef]
  33. Shu, W.-B.; Zhang, X.-b.; Lu, H.-y.; Wang, H.-H.; Lan, G.-H. Comparison of effects of four treatment methods for unstable intertrochanteric fractures: A network meta-analysis. Int. J. Surg. 2018, 60, 173–181. [Google Scholar] [CrossRef]
  34. Xu, Z.; Wang, H.; Shi, Y.; Shen, Q.; Tsamlag, L.; Wang, Z.; Yu, S.; Shen, T.; Wang, Y.; Cai, Y. Impact of calcium, vitamin D, vitamin K, oestrogen, isoflavone and exercise on bone mineral density for osteoporosis prevention in postmenopausal women: A network meta-analysis. Br. J. Nutr. 2019, 123, 84–103. [Google Scholar] [CrossRef]
  35. Hutton, B.; Salanti, G.; Caldwell, D.M.; Chaimani, A.; Schmid, C.H.; Cameron, C.; Ioannidis, J.P.; Straus, S.; Thorlund, K.; Jansen, J.P.; et al. The PRISMA Extension Statement for Reporting of Systematic Reviews Incorporating Network Meta-analyses of Health Care Interventions: Checklist and Explanations. Ann. Intern. Med. 2015, 162, 777–784. [Google Scholar] [CrossRef] [PubMed]
  36. Wan, X.; Wang, W.; Liu, J.; Tong, T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med. Res. Methodol. 2014, 14, 135. [Google Scholar] [CrossRef]
  37. Inpan, R.; Na Takuathung, M.; Sakuludomkan, W.; Dukaew, N.; Teekachunhatean, S.; Koonrungsesomboon, N. Isoflavone intervention and its impact on bone mineral density in postmenopausal women: A systematic review and meta-analysis of randomized controlled trials. Osteoporos. Int. 2023, 35, 413–430. [Google Scholar] [CrossRef]
  38. Higgins, J.P.T.; Altman, D.G.; Gotzsche, P.C.; Juni, P.; Moher, D.; Oxman, A.D.; Savovic, J.; Schulz, K.F.; Weeks, L.; Sterne, J.A.C. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011, 343, d5928. [Google Scholar] [CrossRef] [PubMed]
  39. Zhang, S.; Huang, X.; Zhao, X.; Li, B.; Cai, Y.; Liang, X.; Wan, Q. Effect of exercise on bone mineral density among patients with osteoporosis and osteopenia: A systematic review and network meta-analysis. J. Clin. Nurs. 2021, 31, 2100–2111. [Google Scholar] [CrossRef]
  40. Sadeghirad, B.; Foroutan, F.; Zoratti, M.J.; Busse, J.W.; Brignardello-Petersen, R.; Guyatt, G.; Thabane, L. Theory and practice of Bayesian and frequentist frameworks for network meta-analysis. BMJ Evid.-Based Med. 2023, 28, 204–209. [Google Scholar] [CrossRef]
  41. Seide, S.E.; Jensen, K.; Kieser, M. A comparison of Bayesian and frequentist methods in random-effects network meta-analysis of binary data. Res. Synth. Methods 2020, 11, 363–378. [Google Scholar] [CrossRef]
  42. Marcucci, M.; Etxeandia-Ikobaltzeta, I.; Yang, S.; Germini, F.; Gupta, S.; Agarwal, A.; Ventresca, M.; Tang, S.; Morgano, G.P.; Wang, M.; et al. Benefits and harms of direct oral anticoagulation and low molecular weight heparin for thromboprophylaxis in patients undergoing non-cardiac surgery: Systematic review and network meta-analysis of randomised trials. BMJ 2022, 376, e066785. [Google Scholar] [CrossRef]
  43. Fujii, T.; Salanti, G.; Belletti, A.; Bellomo, R.; Carr, A.; Furukawa, T.A.; Luethi, N.; Luo, Y.; Putzu, A.; Sartini, C.; et al. Effect of adjunctive vitamin C, glucocorticoids, and vitamin B1 on longer-term mortality in adults with sepsis or septic shock: A systematic review and a component network meta-analysis. Intensive Care Med. 2022, 48, 16–24. [Google Scholar] [CrossRef]
  44. Ho, E.K.; Chen, L.; Simic, M.; Ashton-James, C.E.; Comachio, J.; Wang, D.X.M.; Hayden, J.A.; Ferreira, M.L.; Ferreira, P.H. Psychological interventions for chronic, non-specific low back pain: Systematic review with network meta-analysis. BMJ 2022, 376, e067718. [Google Scholar] [CrossRef]
  45. Yeh, W.H.; Ju, Y.J.; Shaw, F.Z.; Liu, Y.T. Comparative effectiveness of electroencephalogram-neurofeedback training of 3-45 frequency band on memory in healthy population: A network meta-analysis with systematic literature search. J. Neuroeng. Rehabil. 2025, 22, 94. [Google Scholar] [CrossRef]
  46. Albertazzi, P.; Steel, S.A.; Bottazzi, M. Effect of pure genistein on bone markers and hot flushes. Climacteric 2005, 8, 371–379. [Google Scholar] [CrossRef]
  47. Alekel, D.L.; Germain, A.S.; Peterson, C.T.; Hanson, K.B.; Stewart, J.W.; Toda, T. Isoflavone-rich soy protein isolate attenuates bone loss in the lumbar spine of perimenopausal women. Am. J. Clin. Nutr. 2000, 72, 844–852. [Google Scholar] [CrossRef] [PubMed]
  48. Anderson, J.J.B.; Chen, X.; Boass, A.; Symons, M.; Kohlmeier, M.; Renner, J.B.; Garner, S.C. Soy isoflavones: No effects on bone mineral content and bone mineral density in healthy, menstruating young adult women after one year. J. Am. Coll. Nutr. 2002, 21, 388–393. [Google Scholar] [CrossRef] [PubMed]
  49. Aoe, S.; Koyama, T.; Toba, Y.; Itabashi, A.; Takada, Y. A controlled trial of the effect of milk basic protein (MBP) supplementation on bone metabolism in healthy menopausal women. Osteoporos. Int. 2005, 16, 2123–2128. [Google Scholar] [CrossRef] [PubMed]
  50. Arjmandi, B.H.; Khalil, D.A.; Smith, B.J.; Lucas, E.A.; Juma, S.; Payton, M.E.; Wild, R.A. Soy protein has a greater effect on bone in postmenopausal women not on hormone replacement therapy, as evidenced by reducing bone resorption and urinary calcium excretion. J. Clin. Endocrinol. Metab. 2003, 88, 1048–1054. [Google Scholar] [CrossRef] [PubMed]
  51. Arjmandi, B.H.; Lucas, E.A.; Khalil, D.A.; Devareddy, L.; Smith, B.J.; McDonald, J.; Arquitt, A.B.; Payton, M.E.; Mason, C. One year soy protein supplementation has positive effects on bone formation markers but not bone density in postmenopausal women. Nutr. J. 2005, 4, 8. [Google Scholar] [CrossRef]
  52. Atteritano, M.; Mazzaferro, S.; Frisina, A.; Cannata, M.L.; Bitto, A.; D’Anna, R.; Squadrito, F.; Macrì, I.; Frisina, N.; Buemi, M. Genistein effects on quantitative ultrasound parameters and bone mineral density in osteopenic postmenopausal women. Osteoporos. Int. 2009, 20, 1947–1954. [Google Scholar] [CrossRef]
  53. Bonjour, J.-P.; Brandolini-Bunlon, M.; Boirie, Y.; Morel-Laporte, F.; Braesco, V.; Bertière, M.-C.; Souberbielle, J.-C. Inhibition of bone turnover by milk intake in postmenopausal women. Br. J. Nutr. 2008, 100, 866–874. [Google Scholar] [CrossRef]
  54. Bonjour, J.-P.; Dontot-Payen, F.; Rouy, E.; Walrand, S.; Rousseau, B. Evolution of serum 25OHD in response to vitamin D3–fortified yogurts consumed by healthy menopausal women: A 6-month randomized controlled trial assessing the interactions between doses, baseline vitamin D status, and seasonality. J. Am. Coll. Nutr. 2018, 37, 34–43. [Google Scholar] [CrossRef]
  55. Brink, E.; Coxam, V.; Robins, S.; Wahala, K.; Cassidy, A.; Branca, F. Long-term consumption of isoflavone-enriched foods does not affect bone mineral density, bone metabolism, or hormonal status in early postmenopausal women: A randomized, double-blind, placebo controlled study. Am. J. Clin. Nutr. 2008, 87, 761–770. [Google Scholar] [CrossRef] [PubMed]
  56. Brooks, J.D.; Ward, W.E.; Lewis, J.E.; Hilditch, J.; Nickell, L.; Wong, E.; Thompson, L.U. Supplementation with flaxseed alters estrogen metabolism in postmenopausal women to a greater extent than does supplementation with an equal amount of soy. Am. J. Clin. Nutr. 2004, 79, 318–325. [Google Scholar] [CrossRef] [PubMed]
  57. Chee, W.S.S.; Suriah, A.R.; Chan, S.P.; Zaitun, Y.; Chan, Y.M. The effect of milk supplementation on bone mineral density in postmenopausal Chinese women in Malaysia. Osteoporos. Int. 2003, 14, 828–834. [Google Scholar] [CrossRef] [PubMed]
  58. Chen, Y.-M.; Ho, S.C.; Lam, S.S.H.; Ho, S.S.S.; Woo, J.L.F. Soy isoflavones have a favorable effect on bone loss in chinese postmenopausal women with lower bone mass: A double-blind, randomized, controlled trial. J. Clin. Endocrinol. Metab. 2003, 88, 4740–4747. [Google Scholar] [CrossRef]
  59. Chilibeck, P.D.; Vatanparast, H.; Pierson, R.; Case, A.; Olatunbosun, O.; Whiting, S.J.; Beck, T.J.; Pahwa, P.; Biem, H.J. Effect of exercise training combined with isoflavone supplementation on bone and lipids in postmenopausal women: A randomized clinical trial. J. Bone Miner. Res. 2013, 28, 780–793. [Google Scholar] [CrossRef]
  60. Cleghom, D.B.; O’Loughlin, P.D.; Schroeder, B.J.; Nordin, B.E.C. An open, crossover trial of calcium-fortified milk in prevention of early postmenopausal bone los. Med. J. Aust. 2001, 175, 242–245. [Google Scholar] [CrossRef]
  61. Choquette, S.; Riesco, É.; Cormier, É.; Dion, T.; Aubertin-Leheudre, M.; Dionne, I.J. Effects of soya isoflavones and exercise on body composition and clinical risk factors of cardiovascular diseases in overweight postmenopausal women: A 6-month double-blind controlled trial. Br. J. Nutr. 2011, 105, 1199–1209. [Google Scholar] [CrossRef]
  62. Dalais, F.S.; Ebeling, P.R.; Kotsopoulos, D.; McGrath, B.P.; Teede, H.J. The effects of soy protein containing isoflavones on lipids and indices of bone resorption in postmenopausal women. Clin. Endocrinol. 2003, 58, 704–709. [Google Scholar] [CrossRef] [PubMed]
  63. Gallagher, J.C.; Satpathy, R.; Rafferty, K.; Haynatzka, V. The effect of soy protein isolate on bone metabolism. Menopause 2004, 11, 290–298. [Google Scholar] [CrossRef] [PubMed]
  64. Green, J.H.; Booth, C.; Bunning, R. Impact of supplementary high calcium milk with additional magnesium on parathyroid hormone and biochemical markers of bone turnover in postmenopausal women. Asia Pac. J. Clin. Nutr. 2002, 11, 268–273. [Google Scholar] [CrossRef] [PubMed]
  65. Gui, J.C.; Brašić, J.R.; Liu, X.D.; Gong, G.Y.; Zhang, G.M.; Liu, C.J.; Gao, G.Q. Bone mineral density in postmenopausal Chinese women treated with calcium fortification in soymilk and cow’s milk. Osteoporos. Int. 2012, 23, 1563–1570. [Google Scholar] [CrossRef] [PubMed]
  66. Harkness, L.S.; Fiedler, K.; Sehgal, A.R.; Oravec, D.; Lerner, E. Decreased bone resorption with soy isoflavone supplementation in postmenopausal women. J. Women’s Health 2004, 13, 1000–1007. [Google Scholar] [CrossRef] [PubMed]
  67. Huang, H.-Y.; Yang, H.-P.; Yang, H.-T.; Yang, T.-C.; Shieh, M.-J.; Huang, S.-Y. One-year soy isoflavone supplementation prevents early postmenopausal bone loss but without a dose-dependent effect. J. Nutr. Biochem. 2006, 17, 509–517. [Google Scholar] [CrossRef]
  68. Ilich, J.Z.; Kelly, O.J.; Liu, P.-Y.; Shin, H.; Kim, Y.; Chi, Y.; Wickrama, K.K.A.S.; Colic-Baric, I. Role of calcium and low-fat dairy foods in weight-loss outcomes revisited: Results from the randomized trial of effects on bone and body composition in overweight/obese postmenopausal women. Nutrients 2019, 11, 1157. [Google Scholar] [CrossRef]
  69. Kenny, A.M.; Mangano, K.M.; Abourizk, R.H.; Bruno, R.S.; Anamani, D.E.; Kleppinger, A.; Walsh, S.J.; Prestwood, K.M.; Kerstetter, J.E. Soy proteins and isoflavones affect bone mineral density in older women: A randomized controlled trial. Am. J. Clin. Nutr. 2009, 90, 234–242. [Google Scholar] [CrossRef]
  70. Kreijkamp-Kaspers, S.; Kok, L.; Grobbee, D.E.; de Haan, E.H.F.; Aleman, A.; Lampe, J.W.; van der Schouw, Y.T. Effect of soy protein containing isoflavones on cognitive function, bone mineral density, and plasma lipids in postmenopausal women. JAMA 2004, 292, 65–74. [Google Scholar] [CrossRef]
  71. Kruger, M.C.; Booth, C.L.; Coad, J.; Schollum, L.M.; Kuhn-Sherlock, B.; Shearer, M.J. Effect of calcium fortified milk supplementation with or without vitamin K on biochemical markers of bone turnover in premenopausal women. Nutrition 2006, 22, 1120–1128. [Google Scholar] [CrossRef]
  72. Kruger, M.C.; Schollum, L.M.; Kuhn-Sherlock, B.; Hestiantoro, A.; Wijanto, P.; Li-Yu, J.; Agdeppa, I.; Todd, J.M.; Eastell, R. The effect of a fortified milk drink on vitamin D status and bone turnover in post-menopausal women from South East Asia. Bone 2010, 46, 759–767. [Google Scholar] [CrossRef] [PubMed]
  73. Kruger, M.C.; Ha, P.C.; Todd, J.M.; Kuhn-Sherlock, B.; Schollum, L.M.; Ma, J.; Qin, G.; Lau, E. High-calcium, vitamin D fortified milk is effective in improving bone turnover markers and vitamin D status in healthy postmenopausal Chinese women. Eur. J. Clin. Nutr. 2012, 66, 856–861. [Google Scholar] [CrossRef]
  74. Lau, E.M.C.; Woo, J.; Lam, V.; Hong, A. Milk supplementation of the diet of postmenopausal Chinese women on a low calcium intake retards bone loss. J. Bone Miner. Res. 2001, 16, 1704–1709. [Google Scholar] [CrossRef]
  75. Lau, E.M.C.; Lynn, H.; Chan, Y.H.; Woo, J. Milk supplementation prevents bone loss in postmenopausal chinese women over 3 years. Bone 2002, 31, 536–540. [Google Scholar] [CrossRef]
  76. Lee, H.; Choue, R.; Lim, H. Effect of soy isoflavones supplement on climacteric symptoms, bone biomarkers, and quality of life in Korean postmenopausal women: A randomized clinical trial. Nutr. Res. Pract. 2017, 11, 223–231. [Google Scholar] [CrossRef]
  77. Levis, S. Soy isoflavones in the prevention of menopausal bone loss and menopausal symptoms. Arch. Intern. Med. 2011, 171, 1363–1369. [Google Scholar] [CrossRef]
  78. Liu, Z.-M.; Chen, B.; Li, S.; Li, G.; Zhang, D.; Ho, S.C.; Chen, Y.-m.; Ma, J.; Qi, H.; Ling, W.-h. Effect of whole soy and isoflavones daidzein on bone turnover and inflammatory markers: A 6-month double-blind, randomized controlled trial in Chinese postmenopausal women who are equol producers. Ther. Adv. Endocrinol. Metab. 2020, 11, 14. [Google Scholar] [CrossRef]
  79. Manios, Y.; Moschonis, G.; Trovas, G.; Lyritis, G.P. Changes in biochemical indexes of bone metabolism and bone mineral density after a 12-mo dietary intervention program: The Postmenopausal Health Study. Am. J. Clin. Nutr. 2007, 86, 781–789. [Google Scholar] [CrossRef] [PubMed]
  80. Marini, H.; Minutoli, L.; Polito, F.; Bitto, A.; Altavilla, D.; Atteritano, M.; Gaudio, A.; Mazzaferro, S.; Frisina, A.; Frisina, N.; et al. Effects of the Phytoestrogen Genistein on Bone Metabolism in Osteopenic Postmenopausal Women. Ann. Intern. Med. 2007, 146, 839–847. [Google Scholar] [CrossRef] [PubMed]
  81. Morabito, N.; Crisafulli, A.; Vergara, C.; Gaudio, A.; Lasco, A.; Frisina, N.; D’Anna, R.; Corrado, F.; Pizzoleo, M.A.; Cincotta, M.; et al. Effects of genistein and hormone-replacement therapy on bone loss in early postmenopausal women: A randomized double-blind placebo-controlled study. J. Bone Miner. Res. 2002, 17, 1904–1912. [Google Scholar] [CrossRef]
  82. Mori, M.; Sagara, M.; Ikeda, K.; Miki, T.; Yamori, Y. Soy isoflavones improve bone metabolism in postmenopausal Japanese women. Clin. Exp. Pharmacol. Physiol. 2004, 31, S44–S46. [Google Scholar] [CrossRef]
  83. Mori, M.; Aizawa, T.; Tokoro, M.; Miki, T.; Yamori, Y. Soy isoflavone tablets reduce osteoporosis risk factors and obesity in middle-aged Japanese women. Clin. Exp. Pharmacol. Physiol. 2004, 31, S39–S41. [Google Scholar] [CrossRef]
  84. Moschonis, G.; Katsaroli, I.; Lyritis, G.P.; Manios, Y. The effects of a 30-month dietary intervention on bone mineral density: The Postmenopausal Health Study. Br. J. Nutr. 2010, 104, 100–107. [Google Scholar] [CrossRef]
  85. Norton, C.; Hettiarachchi, M.; Cooke, R.; Kozior, M.; Kontro, H.; Daniel, R.; Jakeman, P. Effect of 24-week, late-evening ingestion of a calcium-fortified, milk-based protein matrix on biomarkers of bone metabolism and site-specific bone mineral density in postmenopausal women with osteopenia. Nutrients 2022, 14, 3486. [Google Scholar] [CrossRef] [PubMed]
  86. Prince, R.; Devine, A.; Dick, I.; Criddle, A.; Kerr, D.; Kent, N.; Randell, A.; Price, R. The effects of calcium supplementation (milk powder or tablets) and exercise on bone density in postmenopausal women. J. Bone Miner. Res. 1995, 10, 1068–1075. [Google Scholar] [CrossRef]
  87. Radhakrishnan, G.; Rashmi; Agarwal, N.; Vaid, N.B. Evaluation of isoflavone rich soy protein supplementation for postmenopausal therapy. Pak. J. Nutr. 2009, 8, 1009–1017. [Google Scholar] [CrossRef]
  88. Sathyapalan, T.; Aye, M.; Rigby, A.S.; Fraser, W.D.; Thatcher, N.J.; Kilpatrick, E.S.; Atkin, S.L. Soy reduces bone turnover markers in women during early menopause: A randomized controlled trial. J. Bone Miner. Res. 2017, 32, 157–164. [Google Scholar] [CrossRef] [PubMed]
  89. Spence, L.A.; Lipscomb, E.R.; Cadogan, J.; Martin, B.; Wastney, M.E.; Peacock, M.; Weaver, C.M. The effect of soy protein and soy isoflavones on calcium metabolism in postmenopausal women: A randomized crossover study. Am. J. Clin. Nutr. 2005, 81, 916–922. [Google Scholar] [CrossRef]
  90. Tai, T.Y.; Tsai, K.S.; Tu, S.T.; Wu, J.S.; Chang, C.I.; Chen, C.L.; Shaw, N.S.; Peng, H.Y.; Wang, S.Y.; Wu, C.H. The effect of soy isoflavone on bone mineral density in postmenopausal Taiwanese women with bone loss: A 2-year randomized double-blind placebo-controlled study. Osteoporos. Int. 2012, 23, 1571–1580. [Google Scholar] [CrossRef]
  91. Tousen, Y.; Ezaki, J.; Fujii, Y.; Ueno, T.; Nishimuta, M.; Ishimi, Y. Natural S-equol decreases bone resorption in postmenopausal, non-equol-producing Japanese women. Menopause 2011, 18, 563–574. [Google Scholar] [CrossRef]
  92. Turhan, N.O.; Bolkan, F.; Duvan, C.I.; Ardiçoglu, Y. The effect of isoflavones on bone mass and bone remodelling markers in postmenopausal women. Turk. J. Med. Sci. 2008, 38, 145–152. [Google Scholar]
  93. Uenishi, K.; Ishida, H.; Toba, Y.; Aoe, S.; Itabashi, A.; Takada, Y. Milk basic protein increases bone mineral density and improves bone metabolism in healthy young women. Osteoporos. Int. 2007, 18, 385–390. [Google Scholar] [CrossRef]
  94. Uesugi, T.; Fukui, Y.; Yamori, Y. Beneficial effects of soybean isoflavone supplementation on bone metabolism and serum lipids in postmenopausal Japanese women: A four-week study. J. Am. Coll. Nutr. 2002, 21, 97–102. [Google Scholar] [CrossRef]
  95. Uesugi, T.; Toda, T.; Okuhira, T.; Chen, J.-T. Evidence of estrogenic effect by the three-month-intervention of isoflavone on vaginal maturation and bone metabolism in early postmenopausal women. Endocr. J. 2003, 50, 613–619. [Google Scholar] [CrossRef] [PubMed][Green Version]
  96. Vupadhyayula, P.M.; Gallagher, J.C.; Templin, T.; Logsdon, S.M.; Smith, L.M. Effects of soy protein isolate on bone mineral density and physical performance indices in postmenopausal women—A 2-year randomized, double-blind, placebo-controlled trial. Menopause 2009, 16, 320–328. [Google Scholar] [CrossRef] [PubMed]
  97. Wangen, K.E.; Duncan, A.M.; Merz-Demlow, B.E.; Xu, X.; Marcus, R.; Phipps, W.R.; Kurzer, M.S. Effects of soy isoflavones on markers of bone turnover in premenopausal and postmenopausal women. J. Clin. Endocrinol. Metab. 2000, 85, 3043–3048. [Google Scholar] [CrossRef][Green Version]
  98. Woo, J.; Lau, W.; Xu, L.; Lam, C.W.K.; Zhao, X.; Yu, W.; Xing, X.; Lau, E.; Kuhn-Sherlock, B.; Pocock, N.; et al. Milk supplementation and bone health in young adult Chinese women. J. Womens Health 2007, 16, 692–702. [Google Scholar] [CrossRef] [PubMed]
  99. Wu, J.; Oka, J.; Tabata, I.; Higuchi, M.; Toda, T.; Fuku, N.; Ezaki, J.; Sugiyama, F.; Uchiyama, S.; Yamada, K.; et al. Effects of Isoflavone and Exercise on BMD and Fat Mass in Postmenopausal Japanese Women: A 1-Year Randomized Placebo-Controlled Trial. J. Bone Miner. Res. 2006, 21, 780–789. [Google Scholar] [CrossRef]
  100. Wu, J.; Oka, J.; Ezaki, J.; Ohtomo, T.; Ueno, T.; Uchiyama, S.; Toda, T.; Uehara, M.; Ishimi, Y. Possible role of equol status in the effects of isoflavone on bone and fat mass in postmenopausal Japanese women. Menopause 2007, 14, 866–874. [Google Scholar] [CrossRef] [PubMed]
  101. Yamori, Y.; Moriguchi, E.H.; Teramoto, T.; Miura, A.; Fukui, Y.; Honda, K.-i.; Fukui, M.; Nara, Y.; Taira, K.; Moriguchi, Y. Soybean isoflavones reduce postmenopausal bone resorption in female Japanese immigrants in Brazil: A ten-week study. J. Am. Coll. Nutr. 2002, 21, 560–563. [Google Scholar] [CrossRef]
  102. Ye, Y.-B.; Tang, X.-Y.; Verbruggen, M.A.; Su, Y.-X. Soy isoflavones attenuate bone loss in early postmenopausal Chinese women. Eur. J. Nutr. 2006, 45, 327–334. [Google Scholar] [CrossRef] [PubMed]
  103. Zhu, K.; Meng, X.; Kerr, D.A.; Devine, A.; Solah, V.; Binns, C.W.; Prince, R.L. The effects of a two-year randomized, controlled trial of whey protein supplementation on bone structure, IGF-1, and urinary calcium excretion in older postmenopausal women. J. Bone Miner. Res. 2011, 26, 2298–2306. [Google Scholar] [CrossRef]
  104. Zou, Z.-Y.; Lin, X.-M.; Xu, X.-R.; Xu, R.; Ma, L.; Li, Y.; Wang, M.-F. Evaluation of milk basic protein supplementation on bone density and bone metabolism in Chinese young women. Eur. J. Nutr. 2009, 48, 301–306. [Google Scholar] [CrossRef]
  105. Ong, A.M.; Kang, K.; Weiler, H.A.; Morin, S.N. Fermented Milk Products and Bone Health in Postmenopausal Women: A Systematic Review of Randomized Controlled Trials, Prospective Cohorts, and Case-Control Studies. Adv. Nutr. 2020, 11, 251–265. [Google Scholar] [CrossRef]
  106. Hulmi, J.J.; Lockwood, C.M.; Stout, J.R. Effect of protein/essential amino acids and resistance training on skeletal muscle hypertrophy: A case for whey protein. Nutr. Metab. 2010, 7, 51. [Google Scholar] [CrossRef] [PubMed]
  107. Bar-Maisels, M.; Menahem, C.; Gabet, Y.; Hiram-Bab, S.; Phillip, M.; Gat-Yablonski, G. Different Effects of Soy and Whey on Linear Bone Growth and Growth Pattern in Young Male Sprague-Dawley Rats. Front. Nutr. 2021, 8, 739607. [Google Scholar] [CrossRef]
  108. Bu, T.; Huang, J.; Yu, Y.; Sun, P.; Yang, K. Whey Protein Hydrolysate Ameliorated High-Fat-Diet Induced Bone Loss via Suppressing Oxidative Stress and Regulating GSK-3β/Nrf2 Signaling Pathway. Nutrients 2023, 15, 2863. [Google Scholar] [CrossRef]
  109. Duque, G.; Al Saedi, A.; Rivas, D.; Miard, S.; Ferland, G.; Picard, F.; Gaudreau, P. Differential Effects of Long-Term Caloric Restriction and Dietary Protein Source on Bone and Marrow Fat of the Aging Rat. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 2020, 75, 2031–2036. [Google Scholar] [CrossRef] [PubMed]
  110. Miura, A.; Sugiyama, C.; Sakakibara, H.; Simoi, K.; Goda, T. Bioavailability of isoflavones from soy products in equol producers and non-producers in Japanese women. J. Nutr. Intermed. Metab. 2016, 6, 41–47. [Google Scholar] [CrossRef]
  111. Liu, Z.M.; Huang, Q.; Li, S.Y.; Liu, Y.P.; Wu, Y.; Zhang, S.J.; Li, B.L.; Chen, Y.M. A 1:1 matched case-control study on dietary protein intakes and hip fracture risk in Chinese elderly men and women. Osteoporos. Int. 2021, 32, 2205–2216. [Google Scholar] [CrossRef] [PubMed]
  112. Zhong, Y.; Okoro, C.A.; Balluz, L.S. Association of total calcium and dietary protein intakes with fracture risk in postmenopausal women: The 1999–2002 National Health and Nutrition Examination Survey (NHANES). Nutrition 2009, 25, 647–654. [Google Scholar] [CrossRef] [PubMed]
  113. GBD 2017 Diet Collaborators. Health effects of dietary risks in 195 countries, 1990–2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet 2019, 393, 1958–1972. [Google Scholar] [CrossRef] [PubMed]
Nutrients 17 02833 g001
Figure 1. Flow diagram of literature search and screen.
Figure 1. Flow diagram of literature search and screen.
Nutrients 17 02833 g001
Nutrients 17 02833 g002
Figure 2. Network plots for lumbar spine (LS) BMD, total body (TB) BMD, osteocalcin (OC), and deoxypyridinoline (DPD). Each node symbolizes an intervention, and each connecting line represents a direct comparison between two interventions. The node size and the line thickness reflect the number of participants evaluating each intervention and comparison, respectively. CON: control (placebo/no treatment); SP: soy protein; ISO: isoflavone; SP + ISO: soy protein + isoflavone; MBP: milk basic protein; CAS: casein; WP: whey protein; CAS + WP: casein + whey protein; MK: milk; MK + YOG: milk + yogurt; MP: milk powder; SMK: soymilk.
Figure 2. Network plots for lumbar spine (LS) BMD, total body (TB) BMD, osteocalcin (OC), and deoxypyridinoline (DPD). Each node symbolizes an intervention, and each connecting line represents a direct comparison between two interventions. The node size and the line thickness reflect the number of participants evaluating each intervention and comparison, respectively. CON: control (placebo/no treatment); SP: soy protein; ISO: isoflavone; SP + ISO: soy protein + isoflavone; MBP: milk basic protein; CAS: casein; WP: whey protein; CAS + WP: casein + whey protein; MK: milk; MK + YOG: milk + yogurt; MP: milk powder; SMK: soymilk.
Nutrients 17 02833 g002
Nutrients 17 02833 g003
Figure 3. Comparative results of the network meta-analysis of the included studies with regard to LS BMD, TB BMD, and the bone turnover markers OC and DPD. The green area on the left represents the comparison outcomes of the food-class level (a,b), while the blue area on the right illustrates the comparison results of the bioactive-component level (c,d). CON: control (placebo/no treatment); SP: soy protein; ISO: isoflavone; SP + ISO: soy protein + isoflavone; MBP: milk basic protein; CAS: casein; WP: whey protein; CAS + WP: casein + whey protein; MK: milk; MK + YOG: milk + yogurt; MP: milk powder; SMK: soymilk.
Figure 3. Comparative results of the network meta-analysis of the included studies with regard to LS BMD, TB BMD, and the bone turnover markers OC and DPD. The green area on the left represents the comparison outcomes of the food-class level (a,b), while the blue area on the right illustrates the comparison results of the bioactive-component level (c,d). CON: control (placebo/no treatment); SP: soy protein; ISO: isoflavone; SP + ISO: soy protein + isoflavone; MBP: milk basic protein; CAS: casein; WP: whey protein; CAS + WP: casein + whey protein; MK: milk; MK + YOG: milk + yogurt; MP: milk powder; SMK: soymilk.
Nutrients 17 02833 g003
Table 1. The characteristics of the included RCTs.
Table 1. The characteristics of the included RCTs.
Study (Year)CountryIntervention
Period		nAge
(Years)		BMI (kg/m2)ComparisonOutcomes
(M ± SD)
1 Albertazzi (2005) [46]UK6 weeks10053.5 ± 3.027.0 ± 6.1Isoflavone (genistein, 90 mg) vs. placeboOC, CTx
2 Alekel (2000) [47]USA24 weeks6950.2 ± 3.924.0 ± 3.2Soy protein (40 g/d) vs. whey protein (40 g/d) vs. soy protein (40 g/d) with isoflavone (80.4 mg/d)BMD (LS)
3 Anderson (2002) [48]USA12 months2823.9 ± 1.021.4 ± 3.8Isoflavone (90 mg/d) vs. soy proteinBMD (TB, FN, LS, trochanter, intertrochanter, WT)
4 Aoe (2005) [49]Japan6 months2750.5 ± 3.021.6 ± 2.9MBP (40 mg/d) vs. placeboOC, NTx
5 Arjmandi (2003) [50]USA3 months4262.1 ± 4.932.2 ± 7.9Soy protein (40 g/d) vs. milk protein (casein-whey protein, 40 g/d)BAP, DPD, IGF-1
6 Arjmandi (2005) [51]USA12 months6254.3 ± 5.728.0 ± 5.3Soy protein (25 g/d) with isoflavone (60 mg/d) vs. placeboBMD (TB, LS, TH), OC, BAP, DPD, IGF-1
7 Atteritano (2009) [52]Italy2 years13852.5 ± 2.124.5 ± 3.5Isoflavone (genistein, 54 mg/d) vs. placeboBMD (FN, LS)
8 Bonjour (2008) [53]France6 weeks6059.5 ± 3.323.7 ± 3.4Semi-skimmed milk (500 mL/d) vs. no treatmentOC, BAP, PINP, CTx, PTH, IGF-1, 25(OH)D
9 Bonjour (2018) [54]France16 weeks13361.5 ± 5.024.6 ± 2.9Yogurts (125 g/d or 250 g/d) vs. no treatment25(OH)D
10 Brink (2008) [55]Netherlands, Italy, France53 weeks23753.0 ± 3.024.5 ± 2.1Isoflavone vs. placeboBMD (TB, LS), BAP, PINP, DPD, Pyr, PTH, 25(OH)D
11 Brooks (2004) [56]Canada16 weeks2853.4 ± 3.227.4 ± 5.3Soy (25 g/d) with isoflavones (41.9 mg/d) vs. placeboBAP, DPD
12 Chee (2003) [57]China24 months17358.8 ± 3.523.8 ± 3.5High-calcium skimmed milk powder (50 g/d) vs. no treatmentBMD (TB, FN, LS, TH), OC, DPD, PTH, 25(OH)D
13 Chen (2003) [58]China1 year17554.2 ± 3.124.0 ± 3.5Isoflavones (40 mg/d or 80 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, intertrochanter)
14 Chilibeck (2013) [59]Canada2 years14955.7 ± 6.927.1 ± 3.9Isoflavone (165 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, WT)
15 Cleghorn (2001) [60]Australia1 year7252.0 ± 3.026.1 ± 5.4Calcium-fortified milk (3 L/wk) vs. no treatmentDPD
16 Choquette (2011) [61]Canada6 months4559.0 ± 5.030.1 ± 2.8Isoflavones (70 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, WT)
17 Dalais (2003) [62]Australia3 months7860.0 ± 6.225.4 ± 4.7Soy protein (40 g/d) with isoflavones (118 mg/d) vs. casein (40 g/d)DPD, Pyr
18 Gallagher (2004) [63]USA15 months5055.4 ± 4.626.4 ± 5.3soy protein (40 g/d) vs. soy protein (40 g/d) with isoflavones (52 mg/d or 96 mg/d)BMD (FN, LS, trochanter), OC, NTx
19 Green (2002) [64]New Zealand4 weeks5067.6 ± 6.727.5 ± 4.5High-calcium skim milk powder (50 g/d) vs. placeboPINP, DPD, NTx, CTx, PTH
20 Gui (2012) [65]China18 months9856.3 ± 4.224.5 ± 3.0Calcium-fortified milk (250 mL/d) vs. soy milk (250 mL/d) vs. no treatmentBMD (FN, LS, TH)
21 Harkness (2004) [66]USA6 months3870.6 ± 6.325.9 ± 3.5Soy isoflavone (110 mg/d) vs. placeboBMD (TS, trochanter, intertrochanter), OC, BAP
22 Huang (2006) [67]China12 months4252.4 ± 2.823.5 ± 2.6Soy isoflavone (100 mg/d or 200 mg/d) vs. no treatmentBMD (FN, femoral trochanter, LS, WT), BAP, DPD, NTx
23 Ilich (2019) [68]USA6 months6055.8 ± 4.331.5 ± 5.1Low-fat dairy foods (cheeses, milk, yogurt, pudding, low-fat ice cream, 4–5 servings/d) vs. placeboBMD (TB, FN, LS, WF, radius 1/3 of styloid process, whole forearms), OC, NTx, CTx, PTH, 25(OH)D
24 Kenny (2009) [69]USA1 year9773.1 ± 5.928.3 ± 5.4Soy protein (20 g/d) vs. soy isoflavones (105 mg/d) vs. placeboBMD (TB, FN, LS, WF, trochanter, WT, forearm 33% radius, ultradistal radius, total radius), BAP, NTx
25 Kreijkamp-Kaspers (2004) [70]Netherlands12 months17566.6 ± 4.726.2 ± 3.8Soy protein (25.6 g/d) with isoflavones (99 mg/d) vs. 25.6 g of total milk protein (casein-whey protein)BMD (LS, TH, trochanter, intertrochanter, WT), BAP
26 Kruger (2006) [71]New Zealand16 weeks5527.1 ± 4.824.1 ± 2.9High calcium skim milk powder (50 g/d) vs. no treatmentOC, PINP, CTx, PTH, IGF-1, 25(OH)D
27 Kruger (2010) [72]South East Asia16 weeks11357.5 ± 4.424.8 ± 3.8High-calcium vitamin D fortified milk powder (60 g/d) vs. placeboOC, PINP, CTx, PTH
28 Kruger (2012) [73]China12 weeks5862.1 ± 4.225.7 ± 2.3High-calcium vitamin D fortified milk (two servings/d) vs. placeboPINP, CTx, PTH, 25(OH)D
29 Lau (2001) [74]China24 months18556.7 ± 1.726.2 ± 3.8High calcium, low-fat, low-lactose milk powder (50 g/d) vs. no treatmentBMD (TB, FN, LS, TH, intertrochanter)
30 Lau (2002) [75]China3 years19756.9 ± 1.724.1 ± 4.1High calcium, low-fat, low-lactose milk powder (50 g/d) vs. no treatmentBMD (TB, FN, LS, TH, intertrochanter)
31 Lee (2017) [76]Korea12 weeks8453.6 ± 3.4——Isoflavones (70 mg/d) vs. placeboOC, BAP, DPD, NTx, CTx
32 Levis (2011) [77]USA24 months17752.5 ± 3.326.3 ± 3.3Isoflavones (200 mg/d) vs. placeboBMD (FN, LS, TH), NTx, 25(OH)D
33 Liu (2020) [78]China6 months27057.9 ± 5.0——Soy flour (40 g/d) vs. low-fat milk powder (40 g/d)OC, BAP, PINP, CTx, 25(OH)D
34 Lydeking-Olsen (2004) [14]Denmark2 years4557.1 ± 7.623.9 ± 3.8Soy milk (500 mL/d) with isoflavone (76.0 mg/d) vs. soy milk (500 mL/d)BMD (LS)
35 Manios (2007) [79]Greece12 months5560.9 ± 4.829.4 ± 4.8Dairy products fortified with calcium and vitamin D3 (3 portions/d, one portion equals 250 mL milk and 200 g yogurt) vs. no treatmentBMD (TB, LS, TS, pelvis, arms, legs), OC, CTx, PTH, IGF-1, 25(OH)D
36 Marini (2007) [80]Italy2 years30454.5 ± 3.125.0 ± 3.8Isoflavone (genistein, 54 mg/d) vs. placeboBMD (FN, LS), BAP, DPD, Pyr, IGF-1, 25(OH)D
37 Morabito (2002) [81]Italy12 months6051.5 ± 3.524.0 ± 2.5Isoflavone (genistein, 54 mg/day) vs. placeboBMD (FN, LS, WT), OC, BAP, DPD, PTH, 25(OH)D
38 Mori (2004) [82]Japan4 weeks4340–6322.3 ± 2.0Isoflavones (40 mg/d) vs. placeboOC, DPD
39 Mori (2004) [83]Japan24 weeks7049.7 ± 4.822.0 ± 2.7Isoflavones (100 mg/d) vs. placeboBMD (TB)
40 Moschonis (2010) [84]Greece30 months6659.8 ± 4.728.9 ± 5.1Low-fat dairy products fortified with Ca and vitamin D3 (3 portions/d, milk and yogurt) vs. no treatmentBMD (TB, TS, pelvis, arms, legs)
41 Norton (2022) [85]Ireland24 weeks6762.8 ± 6.026.2 ± 4.3Milk protein (casein-whey protein) vs. placeboPINP, CTx
42 Prince (1995) [86]Australia2 years8463.0 ± 4.0——Milk powder (208 mL/d) vs. placeboBMD (FN, trochanter, intertrochanter)
43 Radhakrishnan (2009) [87]India6 months8548.9 ± 6.425.5 ± 4.4Soy protein (25 g/d) with isoflavone (75 mg/d) vs. casein protein (25 g/d)BMD (FN, LS)
44 Sathyapalan (2017) [88]UK6 months20052.0 ± 4.525.5 ± 4.5Soy protein (15 g/d) with isoflavones (66 mg/d) vs. soy protein (15 g/d)PINP, CTx
45 Spence (2005) [89]New Zealand12 weeks4557.0 ± 6.029.0 ± 7.0Soy protein (40 g/d) with isoflavones vs. soy protein (40 g/d) vs. milk protein (casein-whey protein, 40 g/d)OC, BAP, NTx, PTH, 25(OH)D
46 Tai (2012) [90]China96 weeks39655.8 ± 3.822.9 ± 2.6Isoflavone (300 mg/d) vs. placeboBMD (LS), BAP, NTx
47 Tousen (2011) [91]Japan12 months9353.8 ± 3.722.0 ± 3.0Isoflavone (S-equol, 2 mg/d or 6 mg/d or 10 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, intertrochanter, WT), OC, BAP, DPD, NTx
48 Turhan (2008) [92]Turkey6 months9053.9 ± 7.127.0 ± 3.1Isoflavone (genistein, 60 mg/d) vs. placeboOC, CTx
49 Uenishi (2007) [93]Japan6 months3521.0 ± 1.020.8 ± 2.3MBP (40 mg/d) vs. placeboBMD (LS), OC, NTx, CTx
50 Uesugi (2002) [94]Japan4 weeks2351.4 ± 5.922.6 ± 2.8Isoflavones (61.8 mg/d) vs. placeboOC, DPD, Pyr
51 Uesugi (2003) [95]Japan3 months2151.5 ± 5.322.5 ± 2.7Isoflavones (61.8 mg/d) vs. placeboBMD (LS), Pyr
52 Vupadhyayula (2009) [96]USA24 months15763.6 ± 4.526.2 ± 4.0Soy protein (25 g/d) vs. casein and whey protein (25 g/d) vs. soy protein (25 g/d) with isoflavones (90 mg/d)BMD (TB, FN, femoral trochanter, LS, WF)
53 Wangen (2000) [97]USA12 weeks5157.1 ± 5.925.2 ± 3.7Isoflavones (65 mg/d or 130 mg/d) vs. placeboOC, BAP, DPD, CTx, IGF-1
54 Woo (2007) [98]China24 months40828.0 ± 8.020.4 ± 3.6Milk powder (two sachets/d) vs. no treatmentBMD (TB, TH, TS), OC, BAP, PINP, NTx, CTx, PTH, 25(OH)D
55 Wu (2006) [99]Japan12 months6654.4 ± 2.921.1 ± 2.3Isoflavone (75 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, WT), OC, BAP, DPD
56 Wu (2007) [100]Japan12 months5454.4 ± 2.921.1 ± 2.5Isoflavone (75 mg/d) vs. placeboBMD (TB, FN, LS, TH, trochanter, intertrochanter, WT)
57 Yamori (2002) [101]Japan10 weeks4053.2 ± 3.525.8 ± 3.7Isoflavones (37.3 mg/d) vs. placeboDPD, Pyr
58 Ye (2006) [102]China6 months8452.3 ± 3.322.7 ± 2.4Soy isoflavones (84 mg/d or 126 mg/d) vs. placeboBMD (FN, LS, TH, trochanter, intertrochanter), OC, BAP, DPD
59 Zhu (2011) [103]Australia2 years19674.2 ± 2.726.7 ± 3.9Whey protein (30 g/d) vs. placeboBMD (FN, TH), IGF-1
60 Zou (2009) [104]China8 months8119.6 ± 0.620.4 ± 1.8MBP (40 mg/d) vs. whole milk (250 mL/d) vs. no treatmentBMD (TB, LS, Dist R + U 1/10)
BAP: bone-specific alkaline phosphatase; BMD: bone mineral density; BMI: body mass index; CTx: C-terminal telopeptide of type I collagen; DPD: deoxypyridinoline; FN: femoral neck; IGF-1: Insulin-like growth factor 1; LS: lumbar spine; MBP: milk basic protein; NTx: N-terminal telopeptide of type I collagen; OC: osteocalcin; PINP: procollagen type I N-terminal propeptide; PTH: parathyroid hormone; Pyr: pyridinoline; SD: standard deviation; TB: total body; TH: total hip; TS: total spine; WF: whole femurs; WT: ward triangle; [25(OH)D]: 25-hydroxyvitamin D.
Table 2. Intervention rankings using surface under the cumulative ranking (SUCRA) values.
Table 2. Intervention rankings using surface under the cumulative ranking (SUCRA) values.
RankLS BMDTB BMDOCDPD
InterventionSUCRA (%)InterventionSUCRA (%)InterventionSUCRA (%)InterventionSUCRA (%)
Comparison of dairy and soy products
1MK + YOG73.1MK + YOG81.9MK62.3MK69.1
2CON55.8CON60.7MP61.2MP60.2
3MK53.0MK46.5MK + YOG49.5CON20.7
4SMK38.7MP1.1CON27.0
5MP29.4
Comparison of milk protein, soy protein, and soy isoflavones
1WP74.4CAS + WP86.3CAS + WP75.9SP + ISO61.0
2SP63.9SP79.7SP54.9ISO53.4
3CAS + WP60.8MBP63.2ISO49.2CON35.6
4CAS56.4SP + ISO43.4SP + ISO48.2
5ISO48.7CON18.1CON47.6
6MBP40.5ISO9.1MBP24.3
7SP + ISO39.4
8CON15.9
DPD: deoxypyridinoline; LS: lumbar spine; OC: osteocalcin; SUCRA: surface under the cumulative ranking curve; TB: total body; CON: control (placebo/no treatment); SP: soy protein; ISO: isoflavone; SP + ISO: soy protein + isoflavone; MBP: milk basic protein; CAS: casein; WP: whey protein; CAS + WP: casein + whey protein; MK: milk; MK + YOG: milk + yogurt; MP: milk powder; SMK: soymilk.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

You, L.; Wang, L.; Zhou, S.; Guan, Y.; Liu, Y.; Zhu, R.; Chen, H.; Guo, J.; Li, K.; Bao, X.; et al. Comparing the Effects of Dairy and Soybean on Bone Health in Women: A Food- and Component-Level Network Meta-Analysis. Nutrients 2025, 17, 2833. https://doi.org/10.3390/nu17172833

AMA Style

You L, Wang L, Zhou S, Guan Y, Liu Y, Zhu R, Chen H, Guo J, Li K, Bao X, et al. Comparing the Effects of Dairy and Soybean on Bone Health in Women: A Food- and Component-Level Network Meta-Analysis. Nutrients. 2025; 17(17):2833. https://doi.org/10.3390/nu17172833

Chicago/Turabian Style

You, Li, Langrun Wang, Shiwen Zhou, Yiran Guan, Yan Liu, Ruixin Zhu, Huiyu Chen, Jie Guo, Keji Li, Xingyu Bao, and et al. 2025. "Comparing the Effects of Dairy and Soybean on Bone Health in Women: A Food- and Component-Level Network Meta-Analysis" Nutrients 17, no. 17: 2833. https://doi.org/10.3390/nu17172833

APA Style

You, L., Wang, L., Zhou, S., Guan, Y., Liu, Y., Zhu, R., Chen, H., Guo, J., Li, K., Bao, X., Feng, H., Szeto, I. M. Y., He, J., Wang, R., & He, J. (2025). Comparing the Effects of Dairy and Soybean on Bone Health in Women: A Food- and Component-Level Network Meta-Analysis. Nutrients, 17(17), 2833. https://doi.org/10.3390/nu17172833

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop