# paper.pdf ## Metadata - PDFFormatVersion=1.4 - Language=fr - IsLinearized=false - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - Title=Supplements_final - Producer=Skia/PDF m149 Google Docs Renderer ## Contents ### Page 1 Supplemental content Supplement to : Calcium and/or vitamin D supplementation to prevent fractures and falls: systematic review and meta-analysis Olivier Massé, Claudia Mei Mercurio, Sébastien Dupuis, Maya Al Sahwi, Alexandra Arruda, Gabriel Dallaire, Katherine Desforges, Nicolas Dugré, David Williamson Table of contents Figure S1. PRISMA flow diagram showing identification, selection and inclusion ofsystematic reviews..................................................................................................................2 Figure S2. Risk of bias............................................................................................................6 Figure S3. Funnel plots for calcium monotherapy............................................................... 7 Figure S4. Funnel plots for vitamin D monotherapy............................................................8 Figure S5. Funnel plots for vitamin D combined with calcium..........................................11 Figure S6. Pooled effects of calcium versus placebo or no treatment for : (A) risk of hipfracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture, (D) risk offalling), and (E) total number of falls.................................................................................. 12 Figure S7. Pooled effects of vitamin D versus placebo or no treatment for : (A) risk of hipfracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture, (D) risk offalling), and (E) total number of falls.................................................................................. 14 Figure S8. Meta-regression analyses...................................................................................17 Figure S9. Pooled effects of calcium and vitamin D versus placebo or no treatment for :(A) risk of hip fracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture,(D) risk of falling), and (E) total number of falls................................................................ 23 Table S1. PRISMA checklist................................................................................................ 26 Table S2. Table of protocol modifications and key dates.................................................... 30 Table S3. Search strategies.................................................................................................. 31 Table S4. Additional fracture or fall data............................................................................36 Table S5. Characteristics of included trials........................................................................ 37 Table S6. Risk of bias assessment by individual studies....................................................50 Table S7. Funding and conflict of interest statements of included studies.......................55 Table S8. GRADE summary of findings for calcium monotherapy....................................57 Table S9. GRADE summary of findings for vitamin D monotherapy................................60 Table S10. GRADE summary of findings for vitamin D combined with calcium..............63 Table S11. Subgroup analyses for calcium monotherapy................................................... 66 Table S12. Sensitivity analyses for calcium monotherapy................................................. 70 Table S13. Subgroup analyses for vitamin D monotherapy............................................... 72 Table S14. Sensitivity analyses for vitamin D monotherapy............................................. 80 Table S15. Subgroup analyses for vitamin D combined with calcium...............................82 Table S16. Sensitivity analyses for vitamin D combined with calcium............................. 86 Table S17. Post-hoc sensitivity analyses............................................................................. 89 ### Page 2 # Figure S1. PRISMA flow diagram showing identification, selection and inclusion of systematic reviews * We added six systematic reviews in the February 2025 update. Included systematic reviews 1. Yao P, Bennett D, Mafham M, et al. Vitamin D and Calcium for the Prevention of Fracture: A Systematic Review and Meta-analysis. JAMA Netw Open. 2019;2(12):e1917789. 2. Zhao JG, Zeng XT, 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 Dec 26;318(24):2466-2482. 3. Avenell A, Mak JC, O'Connell D. Vitamin D and vitamin D analogues for preventing fractures in post-menopausal women and older men. Cochrane Database Syst Rev. 2014 Apr 14;2014(4):CD000227. 4. Bolland MJ, Grey A, Avenell A. Effects of vitamin D supplementation on musculoskeletal health: a systematic review, meta-analysis, and trial sequential analysis. Lancet Diabetes Endocrinol. 2018 Nov;6(11):847-858. 5. Myung SK, Cho H. Effects of intermittent or single high-dose vitamin D supplementation on risk of falls and fractures: a systematic review and meta-analysis. Osteoporos Int. 2023 Apr 29. ### Page 3 6. Eleni A, Panagiotis P. A systematic review and meta-analysis of vitamin D and calcium in preventing osteoporotic fractures. Clin Rheumatol. 2020 Dec;39(12):3571-3579. 7. Kahwati LC, Weber RP, Pan H, Gourlay M, LeBlanc E, Coker-Schwimmer M, Viswanathan M. Vitamin D, Calcium, or Combined Supplementation for the Primary Prevention of Fractures in Community-Dwelling Adults: Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2018 Apr 17;319(15):1600-1612. 8. Yeung WG, Palmer SC, Strippoli GFM, Talbot B, Shah N, Hawley CM, Toussaint ND, Badve SV. Vitamin D Therapy in Adults With CKD: A Systematic Review and Meta-analysis. Am J Kidney Dis. 2023 Jun 23:S0272-6386. 9. Zittermann A, Trummer C, Theiler-Schwetz V, Pilz S. Long-term supplementation with 3200 to 4000 IU of vitamin D daily and adverse events: a systematic review and meta-analysis of randomized controlled trials. Eur J Nutr. 2023 Jun;62(4):1833-1844. 10. Manoj P, Derwin R, George S. What is the impact of daily oral supplementation of vitamin D3 (cholecalciferol) plus calcium on the incidence of hip fracture in older people? A systematic review and meta-analysis. Int J Older People Nurs. 2023 Jan;18(1):e12492. 11. Barrionuevo P, Kapoor E, Asi N, et al.. Efficacy of pharmacological therapies for the prevention of fractures in postmenopausal women: a network meta-analysis. J Clin Endocrinol Metab. 2019;104(5):1623-1630. 12. Hu ZC, Tang Q, Sang CM, et al.. Comparison of fracture risk using different supplemental doses of vitamin D, calcium or their combination: a network meta-analysis of randomised controlled trials. BMJ Open. 2019;9(10):e024595. 13. Bolland MJ, Leung W, Tai V, et al.. Calcium intake and risk of fracture: systematic review. BMJ. 2015;351:h4580 14. Li S, Xi C, Li L, et al.. Comparisons of different vitamin D supplementation for prevention of osteoporotic fractures: a Bayesian network meta-analysis and meta-regression of randomised controlled trials. Int J Food Sci Nutr. 2021;72(4):518-528 15. Weaver CM, Alexander DD, Boushey CJ, et al.. Calcium plus vitamin D supplementation and risk of fractures: an updated meta-analysis from the National Osteoporosis Foundation. Osteoporos Int. 2016;27(1):367-376. 16. Thanapluetiwong S, Chewcharat A, Takkavatakarn K, Praditpornsilpa K, Eiam-Ong S, Susantitaphong P. Vitamin D supplement on prevention of fall and fracture: a meta-analysis of randomized controlled trials. Medicine. 2020;99(34):e21506. 17. Zheng YT, Cui QQ, Hong YM, Yao WG. A meta-analysis of high dose, intermittent vitamin D supplementation among older adults. PLoS One. 2015;10(1):e0115850 18. Cameron ID, Dyer SM, Panagoda CE, et al.. Interventions for preventing falls in older people in care facilities and hospitals. Cochrane Database Syst Rev. 2018;9:CD005465. 19. Kong SH, Jang HN, Kim JH, Kim SW, Shin CS. Effect of Vitamin D Supplementation on Risk of Fractures and Falls According to Dosage and Interval: A Meta-Analysis. Endocrinol Metab (Seoul). 2022 Apr;37(2):344-358. 20. Wei FL, Li T, Gao QY, Huang Y, Zhou CP, Wang W, Qian JX. Association Between Vitamin D Supplementation and Fall Prevention. Front Endocrinol (Lausanne). 2022 Aug 10;13:919839 21. Ling Y, Xu F, Xia X, Dai D, Xiong A, Sun R, Qiu L, Xie Z. Vitamin D supplementation reduces the risk of fall in the vitamin D deficient elderly: An updated meta-analysis. Clin Nutr. 2021 Nov;40(11):5531-5537. 22. Bolland MJ, Grey A, Gamble GD, Reid IR. The effect of vitamin D supplementation on skeletal, vascular, or cancer outcomes: a trial sequential meta-analysis. Lancet Diabetes Endocrinol. 2014 Apr;2(4):307-320. 23. Bolland MJ, Grey A, Gamble GD, Reid IR. Vitamin D supplementation and falls: a trial sequential meta-analysis. Lancet Diabetes Endocrinol. 2014 Jul;2(7):573-80. 24. Scheel. [Is vitamin D supplementation effective for the prevention of falls in elderly people?]. ### Page 4 Octary T, Gautama MSN, Duong H. Effectiveness of Vitamin D Supplements in Reducing the Risk of Falls among Older Adults: A Meta-Analysis of Randomized Controlled Trials. Ann Geriatr Med Res. 2023 Sep;27(3):192-203. 25. Khatri K, Kaur M, Dhir T, Kankaria A, Arora H. Role of calcium &/or vitamin D supplementation in preventing osteoporotic fracture in the elderly: A systematic review & meta-analysis. Indian J Med Res. 2023 Jan;158(1):5-16. 26. Przedlacki J, Ołdakowska-Jedynak U. The antifracture efficacy of vitamin D in adults - are we assessing it reliably? A systematic review. Endokrynol Pol. 2023;74(5):499-510. 27. Reis AR, Santos RKF, Dos Santos CB, Santos BDC, de Carvalho GB, Brandão-Lima PN, de Oliveira E Silva AM, Pires LV. Supplementation of vitamin D isolated or calcium-associated with bone remodeling and fracture risk in postmenopausal women without osteoporosis: A systematic review of randomized clinical trials. Nutrition. 2023 Dec;116:112151. 28. Tang O, Juraschek SP, Appel LJ. Design Features of Randomized Clinical Trials of Vitamin D and Falls: A Systematic Review. Nutrients. 2018 Jul 26;10(8):964. 29. Tricco AC, Thomas SM, Veroniki AA, Hamid JS, Cogo E, Strifler L, Khan PA, Robson R, Sibley KM, MacDonald H, Riva JJ, Thavorn K, Wilson C, Holroyd-Leduc J, Kerr GD, Feldman F, Majumdar SR, Jaglal SB, Hui W, Straus SE. Comparisons of Interventions for Preventing Falls in Older Adults: A Systematic Review and Meta-analysis. JAMA. 2017 Nov 7;318(17):1687-1699. 30. Guirguis-Blake JM, Michael YL, Perdue LA, Coppola EL, Beil TL. Interventions to Prevent Falls in Older Adults: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2018 Apr 24;319(16):1705-1716. 31. Kahwati LC, LeBlanc E, Weber RP, Giger K, Clark R, Suvada K, Guisinger A, Viswanathan M. Screening for Vitamin D Deficiency in Adults: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2021 Apr 13;325(14):1443-1463. 32. Wu H, Pang Q. The effect of vitamin D and calcium supplementation on falls in older adults : A systematic review and meta-analysis. Orthopade. 2017 Sep;46(9):729-736. 33. Guo JL, Tsai YY, Liao JY, Tu HM, Huang CM. Interventions to reduce the number of falls among older adults with/without cognitive impairment: an exploratory meta-analysis. Int J Geriatr Psychiatry. 2014 Jul;29(7):661-9. 34. Liu C , Kuang X , Li K , Guo X , Deng Q , Li D . Effects of combined calcium and vitamin D supplementation on osteoporosis in postmenopausal women: a systematic review and meta-analysis of randomized controlled trials. Food Funct. 2020 Dec 1;11(12):10817-10827 35. Sivakumar G, Koziarz A, Farrokhyar F. Vitamin D Supplementation in Military Personnel: A Systematic Review of Randomized Controlled Trials. Sports Health. 2019 Sep/Oct;11(5):425-431. 36. Hill KD, Suttanon P, Lin SI, Tsang WWN, Ashari A, Hamid TAA, Farrier K, Burton E. What works in falls prevention in Asia: a systematic review and meta-analysis of randomized controlled trials. BMC Geriatr. 2018 Jan 5;18(1):3 37. Ajwah IM, Albalawi EM, Alhawiti WM, Alahmari NF, Albalawi HZ, Alanazil TK, Alwabsi WH, Alenezil MF, Alotaibi OS, Alenezi BA, Rayyan HR. Vitamin D supplementation as a fall prevention method: A systematic review. AMJ 2020;13(4):148-155] 38. Sivakumar G, Koziarz A, Farrokhyar F. Vitamin D Supplementation in Military Personnel: A Systematic Review of Randomized Controlled Trials. Sports Health. 2019 Sep/Oct;11(5):425-431. 39. Pillay J, Gaudet LA, Saba S, Vandermeer B, Ashiq AR, Wingert A, Hartling L. Falls prevention interventions for community-dwelling older adults: systematic review and meta-analysis of benefits, harms, and patient values and preferences. Syst Rev. 2024 Nov 26;13(1):289. 40. de Souza MM, Moraes Dantas RL, Leão Durães V, Defante MLR, Mendes TB. Vitamin D Supplementation and the Incidence of Fractures in the Elderly Healthy Population: A Meta-analysis of Randomized Controlled Trials. J Gen Intern Med. 2024 Nov;39(14):2829-2836. ### Page 5 41. Tan L, He R, Zheng X. Effect of vitamin D, calcium, or combined supplementation on fall prevention: a systematic review and updated network meta-analysis. BMC Geriatr. 2024 May 2;24(1):390. 42. Lewis SR, McGarrigle L, Pritchard MW, Bosco A, Yang Y, Gluchowski A, Sremanakova J, Boulton ER, Gittins M, Spinks A, Rapp K, MacIntyre DE, McClure RJ, Todd C. Population-based interventions for preventing falls and fall-related injuries in older people. Cochrane Database Syst Rev. 2024 Jan 5;1(1):CD013789. 43. Tao X, Yang W, Zhang Q, Wang Y, Gao F, Wang Y, Zhang T, Liu H, Chen J. Effects of intermittent overload doses of oral vitamin D3 on serum 25(OH)D concentrations and the incidence rates of fractures, falls, and mortality in elderly individuals: A systematic review and meta-analysis. Biomol Biomed. 2024 Sep 6;24(5):1068-1076. 44. Shah VP, Nayfeh T, Alsawaf Y, Saadi S, Farah M, Zhu Y, Firwana M, Seisa M, Wang Z, Scragg R, Kiely ME, Lips P, Mitchell DM, Demay MB, Pittas AG, Murad MH. A Systematic Review Supporting the Endocrine Society Clinical Practice Guidelines on Vitamin D. J Clin Endocrinol Metab. 2024 Jul 12;109(8):1961-1974. # # # # # ### Page 6 # # # Figure S2. Risk of bias A. Fracture assessment B. Fall assessment # ### Page 7 # Figure S3. Funnel plots for calcium monotherapy Funnel plots could be generated for the risk of any fracture and vertebral fracture, but not for any other outcome (less than 10 trials included). A. Risk of any fracture Begg’s test : Kendall's tau = -0.4182, p = 0.0866 Egger’s test : z = -1.8474, p = 0.0647 B. Risk of vertebral fracture Begg’s test : Kendall's tau = -0.1556, p = 0.6007 Egger’s test : z = -0.5068, p = 0.6123 ### Page 8 # Figure S4. Funnel plots for vitamin D monotherapy Funnel plots could be generated for all outcomes, except for vertebral fracture (less than 10 trials included). A. Risk of any fracture Begg’s test : Kendall's tau = -0.1746, p = 0.1384 Egger’s test : z = -1.3612, p = 0.1735 B. Risk of hip fracture Begg’s test : Kendall's tau = -0.1619, p = 0.4351 ### Page 9 Egger’s test : z = -0.6209, p = 0.5347 C. Risk of non-vertebral fracture Begg’s test : Kendall's tau = -0.0316, p = 0.8728 Egger’s test : z = 0.2863, p = 0.7746 D. Risk of falling Begg’s test : Kendall's tau = -0.1653, p = 0.1906 Egger’s test : z = -2.0813, p = 0.0374 * Although Egger’s test was statistically significant, we considered the risk of publication bias to be low, as the funnel plot appeared symmetrical. ### Page 10 E. Total number of falls Begg’s test : Kendall's tau = -0.1515, p = 0.5452 Egger’s test : z = -0.4206, p = 0.6740 # # # # # ### Page 11 # Figure S5. Funnel plots for vitamin D combined with calcium Funnel plots could be generated for the risk of any fracture and risk of falling, but not for any other outcome (less than 10 trials included). A. Risk of any fracture Begg’s test : Kendall's tau = -0.0667, p = 0.7705 Egger’s test : -1.6032, p = 0.1089 B. Risk of falling Begg’s test : Kendall's tau = 0.0222, p = 1.0000 Egger’s test : z = -1.2280, p = 0.2194 ### Page 12 # Figure S6. Pooled effects of calcium versus placebo or no treatment for : (A) risk of hip fracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture, (D) risk of falling), and (E) total number of falls. A. B. ### Page 13 C. D. E. ### Page 14 # Figure S7. Pooled effects of vitamin D versus placebo or no treatment for : (A) risk of hip fracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture, (D) risk of falling), and (E) total number of falls. A. B. ### Page 15 C. D. ### Page 16 E. # # # # # # ### Page 17 # Figure S8. Meta-regression analyses A. Vitamin D : Risk of any fracture according to baseline vitamin D levels * Three trials with outlier results do not appear on the figure. ### Page 18 B. Vitamin D : Risk of non-vertebral fracture according to baseline vitamin D levels ### Page 19 C. Vitamin D : Risk of falling according to baseline vitamin D levels * Two trials with outlier results do not appear on the figure. ** The test for residual heterogeneity was significant (QE = 36.64, p = 0.048), indicating unexplained heterogeneity. ### Page 20 D. Vitamin D : Risk of any fracture according to achieved vitamin D levels * One trial with outlier results does not appear on the figure. ### Page 21 E. Vitamin D : Risk of non-vertebral fracture according to achieved vitamin D levels ### Page 22 F. Vitamin D : Risk of falling according to achieved vitamin D levels # # ### Page 23 # Figure S9. Pooled effects of calcium and vitamin D versus placebo or no treatment for : (A) risk of hip fracture, (B) risk of non-vertebral fracture, (C) risk of vertebral fracture, (D) risk of falling), and (E) total number of falls. A. B. ### Page 24 C. D. ### Page 25 E. # # # # # # # # # # # ### Page 26 # Table S1. PRISMA checklist | Sectionand Topic | Item# | Checklist item | Location whereitem is reported | | - | - | - | - | | TITLE | | Title | 1 | Identify the report as a systematic review. | Title | | ABSTRACT | | Abstract | 2 | See the PRISMA 2020 for Abstracts checklist. | Abstract | | INTRODUCTION | | Rationale | 3 | Describe the rationale for the review in the context ofexisting knowledge. | Introduction | | Objectives | 4 | Provide an explicit statement of the objective(s) orquestion(s) the review addresses. | Introduction | | METHODS | | Eligibilitycriteria | 5 | Specify the inclusion and exclusion criteria for the reviewand how studies were grouped for the syntheses. | Methods | | Informationsources | 6 | Specify all databases, registers, websites, organisations,reference lists and other sources searched or consulted toidentify studies. Specify the date when each source waslast searched or consulted. | Methods | | Searchstrategy | 7 | Present the full search strategies for all databases,registers and websites, including any filters and limitsused. | Table S3 | | Selectionprocess | 8 | Specify the methods used to decide whether a study metthe inclusion criteria of the review, including how manyreviewers screened each record and each reportretrieved, whether they worked independently, and ifapplicable, details of automation tools used in theprocess. | Methods | | Datacollectionprocess | 9 | Specify the methods used to collect data from reports,including how many reviewers collected data from eachreport, whether they worked independently, anyprocesses for obtaining or confirming data from studyinvestigators, and if applicable, details of automationtools used in the process. | Methods | | Data items | 10a | List and define all outcomes for which data were sought.Specify whether all results that were compatible witheach outcome domain in each study were sought (e.g. forall measures, time points, analyses), and if not, themethods used to decide which results to collect. | Methods | | 10b | List and define all other variables for which data weresought (e.g. participant and intervention characteristics,funding sources). Describe any assumptions made about | MethodsFull protocol | ### Page 27 | any missing or unclear information. | | - | | Study riskof biasassessment | 11 | Specify the methods used to assess risk of bias in theincluded studies, including details of the tool(s) used,how many reviewers assessed each study and whetherthey worked independently, and if applicable, details ofautomation tools used in the process. | Methods | | Effectmeasures | 12 | Specify for each outcome the effect measure(s) (e.g. riskratio, mean difference) used in the synthesis orpresentation of results. | Methods | | Synthesismethods | 13a | Describe the processes used to decide which studies wereeligible for each synthesis (e.g. tabulating the studyintervention characteristics and comparing against theplanned groups for each synthesis (item #5)). | Methods | | 13b | Describe any methods required to prepare the data forpresentation or synthesis, such as handling of missingsummary statistics, or data conversions. | Methods | | 13c | Describe any methods used to tabulate or visuallydisplay results of individual studies and syntheses. | Methods | | 13d | Describe any methods used to synthesize results andprovide a rationale for the choice(s). If meta-analysis wasperformed, describe the model(s), method(s) to identifythe presence and extent of statistical heterogeneity, andsoftware package(s) used. | Methods | | 13e | Describe any methods used to explore possible causes ofheterogeneity among study results (e.g. subgroupanalysis, meta-regression). | Methods | | 13f | Describe any sensitivity analyses conducted to assessrobustness of the synthesized results. | Methods | | Reportingbiasassessment | 14 | Describe any methods used to assess risk of bias due tomissing results in a synthesis (arising from reportingbiases). | Methods | | Certaintyassessment | 15 | Describe any methods used to assess certainty (orconfidence) in the body of evidence for an outcome. | Methods | | RESULTS | | Studyselection | 16a | Describe the results of the search and selection process,from the number of records identified in the search to thenumber of studies included in the review, ideally using aflow diagram. | ResultsFigure 1Figure S1 | | 16b | Cite studies that might appear to meet the inclusioncriteria, but which were excluded, and explain why theywere excluded. | Not reported | | Studycharacteristics | 17 | Cite each included study and present its characteristics. | Table 1Tables S5 | ### Page 28 | Risk of biasin studies | 18 | Present assessments of risk of bias for each includedstudy. | Figure 2Table S6 | | - | - | - | - | | Results ofindividualstudies | 19 | For all outcomes, present, for each study: (a) summarystatistics for each group (where appropriate) and (b) aneffect estimate and its precision (e.g. confidence/credibleinterval), ideally using structured tables or plots. | Figures 2-4 | | Results ofsyntheses | 20a | For each synthesis, briefly summarise the characteristicsand risk of bias among contributing studies. | ResultsTables S5Tables S6 | | 20b | Present results of all statistical syntheses conducted. Ifmeta-analysis was done, present for each the summaryestimate and its precision (e.g. confidence/credibleinterval) and measures of statistical heterogeneity. Ifcomparing groups, describe the direction of the effect. | Results | | 20c | Present results of all investigations of possible causes ofheterogeneity among study results. | ResultsTables S10,S12,S14 | | 20d | Present results of all sensitivity analyses conducted toassess the robustness of the synthesized results. | ResultsTables S11,S13,S15 | | Reportingbiases | 21 | Present assessments of risk of bias due to missing results(arising from reporting biases) for each synthesisassessed. | ResultsFigures S2-4 | | Certainty ofevidence | 22 | Present assessments of certainty (or confidence) in thebody of evidence for each outcome assessed. | Table 2Tables S7-9 | | DISCUSSION | | Discussion | 23a | Provide a general interpretation of the results in thecontext of other evidence. | Discussion | | 23b | Discuss any limitations of the evidence included in thereview. | Discussion | | 23c | Discuss any limitations of the review processes used. | Discussion | | 23d | Discuss implications of the results for practice, policy,and future research. | Discussion | | OTHER INFORMATION | | Registrationand protocol | 24a | Provide registration information for the review, includingregister name and registration number, or state that thereview was not registered. | MethodsCRD42022307009 | | 24b | Indicate where the review protocol can be accessed, orstate that a protocol was not prepared. | MethodsFull protocolCRD42022307009 | | 24c | Describe and explain any amendments to informationprovided at registration or in the protocol. | MethodsCRD42022307009Table S2 | | Support | 25 | Describe sources of financial or non-financial support forthe review, and the role of the funders or sponsors in the | Funding | ### Page 29 | review. | | - | | Competinginterests | 26 | Declare any competing interests of review authors. | Declaration ofcompeting interests | | Availabilityof data, codeand othermaterials | 27 | Report which of the following are publicly available andwhere they can be found: template data collection forms;data extracted from included studies; data used for allanalyses; analytic code; any other materials used in thereview. | None are publiclyavailable. | # # # # # # # # # # # # # ### Page 30 # Table S2. Table of protocol modifications and key dates | November 30, 2023 | First protocol registration on PROSPERO. | | - | - | | January 15, 2024 | First update of the protocol after critical appraisal by all co-authors :- Major changes to the search strategy after consultation with a research librarian- Minor changes to statistical plan and subgroup analysisFull modifications available on the PROSPERO registration | | February 17, 2024 | Second update of the protocol on PROSPERO after group discussions andconsultation with the research librarian :- We will sought all relevant RCTs in adults and not only in older adults- Subgroup analysis added for the age criteriaFull modifications available on PROSPERO on February 27, 2024. | | February 17, 2024 | Start of abstracts and titles screening. | | February 29, 2024 | Submission of full protocol to the BMJ Open. | | April 07, 2024 | End of abstracts and titles screening. Start of full text assessment. | | July 02, 2024 | End of full text assessment. | | July 26, 2024 | Start of data extraction. | | August 20, 2024 | We noticed that some problems occurred with the PROSPERO registration onFebruary 27, 2024. Some subgroups analyses were not added to the registration(although they were planned in the BMJ Open submission of February 2024, beforefull text assessment and data extraction) : baseline dietary calcium intake (< 800mg vs ≥ 800mg); frequency of administration of vitamin D (daily to monthlyadministration vs intermittent or high doses); mean chronic daily dose of vitamin Dexcluding intermittent or high doses (< 1000 vs 1000-2000 vs > 2000 units/day).PROSPERO registration was updated accordingly. | | November 2024 | Following peer review by the BMJ Open, minor modifications were made to the fullprotocol. We added details on how to extract data for the fractures (i.e. removingvertebral fractures as a possible surrogate for total fractures, adding osteoporoticfractures as a possible surrogate for total fractures). We slightly changed thesubgroup analysis plan, specifying when to conduct some subgroup analysis (i.e.when at least ten trials are included for a meta-analysis, to limit type 1 errors andfalse positives). These changes were made before data analysis. | | December 2024 | Publication of full protocol on BMJ Open. | | February 19, 2025 | Update of abstracts and titles screening. | | March 10, 2025 | End of abstracts and titles screening. Start of full text assessment. | | March 15, 2025 | End of full text assessment. Start of data extraction. | | April 26, 2025 | End of data extraction. | | May 3, 2025 | Start of meta-analysis, GRADE assessment and writing process. | | July 19, 2025 | Conducting the GRADE assessment, we concluded that vitamin D and/or calciumhas little to no effect, except for only one comparison and outcome. To furtherstrengthen our confidence in these negative findings, we decided to conductpost-hoc meta-regression analyses to explore whether baseline and achievedvitamin D levels, treated as continuous covariates, modified the treatment effectsof vitamin D monotherapy. | | September 2025 | End of data synthesis, GRADE assessment and writing process. Submission to theBMJ. | ### Page 31 # Table S3. Search strategies Step 1. Identifying systematic reviews | Ovid - Medline1. exp Vitamin D/2. ((d? adj1 vitamin*) or cholecalciferol* or calciol or calcifediol or hydroxycholecalciferol* ordihydroxycholecalciferol* or calciferol* or hydroxyvitamin d or hydroxyvitamin d? ordihydroxyvitamin d or dihydroxyvitamin d? or dihydrotachysterol*).tw.3. calcium.hw.4. calcium.tw.5. exp Fractures, Bone/6. (fractur* or microfractur*).tw.7. Accidental Falls/8. (falls or faller$1).tw.9. 1 or 2 or 3 or 410. 5 or 6 or 7 or 811. 9 and 1012. (systematic review or meta-analysis).pt.13. meta-analysis/ or systematic review/ or systematic reviews as topic/ or meta-analysis astopic/ or "meta analysis (topic)"/ or "systematic review (topic)"/ or exp technology assessment,biomedical/ or network meta-analysis/14. ((systematic* adj3 (review* or overview*)) or (methodologic* adj3 (review* oroverview*))).ti,ab,kf.15. ((quantitative adj3 (review* or overview* or synthes*)) or (research adj3 (integrati* oroverview*))).ti,ab,kf.16. ((integrative adj3 (review* or overview*)) or (collaborative adj3 (review* or overview*)) or(pool* adj3 analy*)).ti,ab,kf.17. (data synthes* or data extraction* or data abstraction*).ti,ab,kf.18. (handsearch* or hand search*).ti,ab,kf.19. (mantel haenszel or peto or der simonian or dersimonian or fixed effect* or latinsquare*).ti,ab,kf.20. (met analy* or metanaly* or technology assessment* or HTA or HTAs or technologyoverview* or technology appraisal*).ti,ab,kf.21. (meta regression* or metaregression*).ti,ab,kf.22. (meta-analy* or metaanaly* or systematic review* or biomedical technology assessment* orbio-medical technology assessment*).mp,hw.23. (medline or cochrane or pubmed or medlars or embase or cinahl).ti,ab,hw.24. (cochrane or (health adj2 technology assessment) or evidence report).jw.25. (comparative adj3 (efficacy or effectiveness)).ti,ab,kf.26. (outcomes research or relative effectiveness).ti,ab,kf.27. ((indirect or indirect treatment or mixed-treatment or bayesian) adj3 comparison*).ti,ab,kf.28. (meta-analysis or systematic review).md.29. (multi* adj3 treatment adj3 comparison*).ti,ab,kf.30. (mixed adj3 treatment adj3 (meta-analy* or metaanaly*)).ti,ab,kf.31. umbrella review*.ti,ab,kf.32. (multi* adj2 paramet* adj2 evidence adj2 synthesis).ti,ab,kf.33. (multiparamet* adj2 evidence adj2 synthesis).ti,ab,kf. | | - | ### Page 32 | 34. (multi-paramet* adj2 evidence adj2 synthesis).ti,ab,kf.35. or/12-3436. 11 and 3537. limit 36 to yr="2014 -Current" | | - | | | | Ovid - Embase1. exp Vitamin D/2. ((d? adj1 vitamin*) or cholecalciferol* or calciol or calcifediol or hydroxycholecalciferol* ordihydroxycholecalciferol* or calciferol* or hydroxyvitamin d or hydroxyvitamin d? ordihydroxyvitamin d or dihydroxyvitamin d? or dihydrotachysterol*).tw.3. Calcium/ or calcium intake/ or calcium carbonate/ or Calcium Gluconate/4. calcium.tw.5. 1 or 2 or 3 or 46. exp bone injury/7. (fractur* or microfractur*).tw.8. Falling/9. (falls or fallers).tw.10. 6 or 7 or 8 or 911. 5 and 1012. (systematic review or meta-analysis).pt.13. meta-analysis/ or systematic review/ or systematic reviews as topic/ or meta-analysis astopic/ or "meta analysis (topic)"/ or "systematic review (topic)"/ or exp technology assessment,biomedical/ or network meta-analysis/14. ((systematic* adj3 (review* or overview*)) or (methodologic* adj3 (review* oroverview*))).ti,ab,kf.15. ((quantitative adj3 (review* or overview* or synthes*)) or (research adj3 (integrati* oroverview*))).ti,ab,kf.16. ((integrative adj3 (review* or overview*)) or (collaborative adj3 (review* or overview*)) or(pool* adj3 analy*)).ti,ab,kf.17. (data synthes* or data extraction* or data abstraction*).ti,ab,kf.18. (handsearch* or hand search*).ti,ab,kf.19. (mantel haenszel or peto or der simonian or dersimonian or fixed effect* or latinsquare*).ti,ab,kf.20. (met analy* or metanaly* or technology assessment* or HTA or HTAs or technologyoverview* or technology appraisal*).ti,ab,kf.21. (meta regression* or metaregression*).ti,ab,kf.22. (meta-analy* or metaanaly* or systematic review* or biomedical technology assessment* orbio-medical technology assessment*).mp,hw.23. (medline or cochrane or pubmed or medlars or embase or cinahl).ti,ab,hw.24. (cochrane or (health adj2 technology assessment) or evidence report).jw.25. (comparative adj3 (efficacy or effectiveness)).ti,ab,kf.26. (outcomes research or relative effectiveness).ti,ab,kf.27. ((indirect or indirect treatment or mixed-treatment or bayesian) adj3 comparison*).ti,ab,kf.28. (meta-analysis or systematic review).md.29. (multi* adj3 treatment adj3 comparison*).ti,ab,kf.30. (mixed adj3 treatment adj3 (meta-analy* or metaanaly*)).ti,ab,kf.31. umbrella review*.ti,ab,kf.32. (multi* adj2 paramet* adj2 evidence adj2 synthesis).ti,ab,kf. | ### Page 33 | 33. (multiparamet* adj2 evidence adj2 synthesis).ti,ab,kf.34. (multi-paramet* adj2 evidence adj2 synthesis).ti,ab,kf.35. or/12-3436. 11 and 3537. limit 36 to yr="2014 -Current" | | - | Step 2. Identifying RCTs | Ovid - Medline1. exp Vitamin D/2. ((d? adj1 vitamin*) or cholecalciferol* or calciol or calcifediol or hydroxycholecalciferol* ordihydroxycholecalciferol* or calciferol* or hydroxyvitamin d or hydroxyvitamin d? ordihydroxyvitamin d or dihydroxyvitamin d? or dihydrotachysterol*).tw.3. calcium.hw.4. calcium.tw.5. exp Fractures, Bone/6. (fractur* or microfractur*).tw.7. Accidental Falls/8. (falls or faller$1).tw.9. 1 or 2 or 3 or 410. 5 or 6 or 7 or 811. 9 and 1012. (Randomized Controlled Trial or Controlled Clinical Trial or Pragmatic Clinical Trial orClinical Study or Adaptive Clinical Trial or Equivalence Trial).pt.13. (Clinical Trial or Clinical Trial, Phase I or Clinical Trial, Phase II or Clinical Trial, PhaseIII or Clinical Trial, Phase IV or Clinical Trial Protocol).pt.14. Multicenter Study.pt.15. Clinical Studies as Topic/16. exp Clinical Trial/ or exp Clinical Trials as Topic/ or Clinical Trial Protocol/ or Clinical TrialProtocols as Topic/ or exp "Clinical Trial (topic)"/17. Multicenter Study/ or Multicenter Studies as Topic/ or "Multicenter Study (topic)"/18. Randomization/19. Random Allocation/20. Double-Blind Method/21. Double Blind Procedure/22. Double-Blind Studies/23. Single-Blind Method/24. Single Blind Procedure/25. Single-Blind Studies/26. Placebos/27. Placebo/28. Control Groups/29. Control Group/30. Cross-Over Studies/ or Crossover Procedure/31. (random* or sham or placebo*).ti,ab,hw,kf.32. ((singl* or doubl*) adj (blind* or dumm* or mask*)).ti,ab,hw,kf.33. ((tripl* or trebl*) adj (blind* or dumm* or mask*)).ti,ab,hw,kf. | | - | ### Page 34 | 34. (control* adj3 (study or studies or trial* or group*)).ti,ab,hw,kf.35. (clinical adj3 (study or studies or trial*)).ti,ab,hw,kf.36. (Nonrandom* or non random* or non-random* or quasi-random* orquasirandom*).ti,ab,hw,kf.37. (phase adj3 (study or studies or trial*)).ti,ab,hw,kf.38. ((crossover or cross-over) adj3 (study or studies or trial*)).ti,ab,hw,kf.39. ((multicent* or multi-cent*) adj3 (study or studies or trial*)).ti,ab,hw,kf.40. allocated.ti,ab,hw.41. ((open label or open-label) adj5 (study or studies or trial*)).ti,ab,hw,kf.42. ((equivalence or superiority or non-inferiority or noninferiority) adj3 (study or studies ortrial*)).ti,ab,hw,kf.43. (pragmatic study or pragmatic studies).ti,ab,hw,kf.44. ((pragmatic or practical) adj3 trial*).ti,ab,hw,kf.45. ((quasiexperimental or quasi-experimental) adj3 (study or studies or trial*)).ti,ab,hw,kf.46. trial.ti,kf.47. or/12-4648. 11 and 4749. limit 48 to yr="2017 -Current" | | - | | | | Ovid - Embase1. exp Vitamin D/2. ((d? adj1 vitamin*) or cholecalciferol* or calciol or calcifediol or hydroxycholecalciferol* ordihydroxycholecalciferol* or calciferol* or hydroxyvitamin d or hydroxyvitamin d? ordihydroxyvitamin d or dihydroxyvitamin d? or dihydrotachysterol*).tw.3. Calcium/ or calcium intake/ or calcium carbonate/ or Calcium Gluconate/4. calcium.tw.5. 1 or 2 or 3 or 46. exp bone injury/7. (fractur* or microfractur*).tw.8. Falling/9. (falls or fallers).tw.10. 6 or 7 or 8 or 911. 5 and 1012. (Randomized Controlled Trial or Controlled Clinical Trial or Pragmatic Clinical Trial orClinical Study or Adaptive Clinical Trial or Equivalence Trial).pt.13. (Clinical Trial or Clinical Trial, Phase I or Clinical Trial, Phase II or Clinical Trial, PhaseIII or Clinical Trial, Phase IV or Clinical Trial Protocol).pt.14. Multicenter Study.pt.15. Clinical Studies as Topic/16. exp Clinical Trial/ or exp Clinical Trials as Topic/ or Clinical Trial Protocol/ or Clinical TrialProtocols as Topic/ or exp "Clinical Trial (topic)"/17. Multicenter Study/ or Multicenter Studies as Topic/ or "Multicenter Study (topic)"/18. Randomization/19. Random Allocation/20. Double-Blind Method/21. Double Blind Procedure/22. Double-Blind Studies/23. Single-Blind Method/ | ### Page 35 | 24. Single Blind Procedure/25. Single-Blind Studies/26. Placebos/27. Placebo/28. Control Groups/29. Control Group/30. Cross-Over Studies/ or Crossover Procedure/31. (random* or sham or placebo*).ti,ab,hw,kf.32. ((singl* or doubl*) adj (blind* or dumm* or mask*)).ti,ab,hw,kf.33. ((tripl* or trebl*) adj (blind* or dumm* or mask*)).ti,ab,hw,kf.34. (control* adj3 (study or studies or trial* or group*)).ti,ab,hw,kf.35. (clinical adj3 (study or studies or trial*)).ti,ab,hw,kf.36. (Nonrandom* or non random* or non-random* or quasi-random* orquasirandom*).ti,ab,hw,kf.37. (phase adj3 (study or studies or trial*)).ti,ab,hw,kf.38. ((crossover or cross-over) adj3 (study or studies or trial*)).ti,ab,hw,kf.39. ((multicent* or multi-cent*) adj3 (study or studies or trial*)).ti,ab,hw,kf.40. allocated.ti,ab,hw.41. ((open label or open-label) adj5 (study or studies or trial*)).ti,ab,hw,kf.42. ((equivalence or superiority or non-inferiority or noninferiority) adj3 (study or studies ortrial*)).ti,ab,hw,kf.43. (pragmatic study or pragmatic studies).ti,ab,hw,kf.44. ((pragmatic or practical) adj3 trial*).ti,ab,hw,kf.45. ((quasiexperimental or quasi-experimental) adj3 (study or studies or trial*)).ti,ab,hw,kf.46. trial.ti,kf.47. or/12-4648. 11 and 4749. limit 48 to yr="2017 -Current" | | - | | | | Cochrane Central Register of Controlled Trials1. exp Vitamin D/2. ((d? adj1 vitamin*) or cholecalciferol* or calciol or calcifediol or hydroxycholecalciferol* ordihydroxycholecalciferol* or calciferol* or hydroxyvitamin d or hydroxyvitamin d? ordihydroxyvitamin d or dihydroxyvitamin d? or dihydrotachysterol*).ti,ab.3. calcium.hw.4. calcium.ti,ab.5. exp Fractures, Bone/6. (fractur* or microfractur*).ti,ab.7. Accidental Falls/8. (falls or faller*).ti,ab.9. or/1-410. or/5-811. 9 and 1012. limit 11 to yr="2017 -Current" | * We consulted the clinical trials database on January 1st, 2024 and screened abstracts from scientific meetings published from 2021 to 2023. ### Page 36 # Table S4. Additional fracture or fall data | Source | Additional fracture or fall data | | - | - | | Leboff (2022) | IRR for falls : 0.97 [0.92-1.02]Extracted from USPTSK Draft Evidence Review [1] | | Dawson-Hughes (1997) | IRR for falls : 1.12 [0.94-1.33]Extracted from Gillepsie et al. [2] (listed as Bischoff-Ferrari 2006) | | Witham (2010) | Fall/fracture data unclear in the original articleExtracted from Avenell et al. [3] and Bolland et al. [4] | | Witham (2013) | Fall/fracture data unclear in the original articleExtracted from Avenell et al. [3] and Bolland et al. [4] | | Harwood (2004) | Hip fractures extracted from Avenell et al. [3] | | Uusi-Rasi (2017) | Fall/fracture obtained through personal communication in April 2024 | | Reid (1993) | Hip fractures extracted from Bolland et al. [5] | | Dhesi (2004) | IRR for falls : 0.61 [0.32-1.17]Extracted from Gillepsie et al. [2] | | Porthouse (2005) | IRR for falls : 0.91 [0.83-1.01]Extracted from Gillepsie et al. [2] | | Chapuy (2002) | Some fracture data extracted from Avenell et al. [3] | | Baron (1999) | Fracture data extracted from Bolland et al. [5] | | Avenell (2004) | Fracture data extracted from Avenell et al. [3] | | Bolton-Smith (2007) | Fracture data extracted from Avenell et al. [3] | | Breslavsky (2013) | Fracture data extracted from Bolland et al. [4] | References 1. Agency for Healthcare Research and Quality (Publication No. 24-05319-EF-1). Vitamin D, Calcium, or Combined Supplementation for the Primary Prevention of Falls and Fractures in Community-Dwelling Adults: A Draft Evidence Update for the U.S. Preventive Services Task Force. December 2024. 2. Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson LM, Lamb SE. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev. 2012 Sep 12;2012(9):CD007146. 3. Avenell A, Mak JC, O'Connell D. Vitamin D and vitamin D analogues for preventing fractures in post-menopausal women and older men. Cochrane Database Syst Rev. 2014 Apr 14;2014(4):CD000227. 4. Bolland MJ, Grey A, Avenell A. Effects of vitamin D supplementation on musculoskeletal health: a systematic review, meta-analysis, and trial sequential analysis. Lancet Diabetes Endocrinol. 2018 Nov;6(11):847-858. 5. Bolland MJ, Leung W, Tai V, Bastin S, Gamble GD, Grey A, Reid IR. Calcium intake and risk of fracture: systematic review. BMJ. 2015 Sep 29;351:h4580. ### Page 37 # Table S5. Characteristics of included trials A. | Source | Country | Treatments and controls | Other interventions | Duration(y) | | - | - | - | - | - | | Inkovaara 1983 | Finland | Calcium 1.2 g/d (n = 78)D3 1000 IU/d (n = 88)Calcium 1.2 g/d + D3 1000 IU/d (n = 93)Placebo (n = 68) | Factorial design : a)calcium, b) vitamin D, c)methandienone. | 1,0 | | Hansson 1987 | Sweden | Calcium 1 g/d (n = 25)Placebo (n = 25) | Groups with sodium fluoridewere excluded from theanalysis. | 3,0 | | Chapuy 1992 | France | Calcium 1.2 g + D3 800 IU/d (n = 1634)Placebo (n = 1636) | None | 3,0 | | Schwellnus 1992 | South Africa | Calcium 0.5 g/d (n = 247)No treatment (n = 1151) | Military training program | 0,2 | | Reid 1993 | NewZealand | Calcium 1 g/d (n = 68)Placebo (n = 67) | None | 4,0 | | Graafmans 1996 | Netherlands | D3 400 IU/d (n = 177)Placebo (n = 177) | None | 0,5 | | Lips 1996 | Netherlands | D3 400 IU/d (n = 1291)Placebo (n = 1287) | All participants wereadvised to consume at leastthree servings of dairyproducts per day (targetedcalcium intake of at least800 to 1000 mg/d). | 3,5 | | Recker 1996 | USA | Calcium 1.2 g/d (n = 95)Placebo (n = 102) | None | 4,3 | | Dawson-H. 1997 | USA | Calcium 0.5 g + D3 700 IU/d (n = 219)Placebo (n = 226) | None | 3,0 | | Riggs 1998 | USA | Calcium 1.6 g/d (n = 119)Placebo (n = 117) | None | 4,0 | | Baron 1999 | USA | Calcium 1.2 g/d (n = 464)Placebo (n = 466) | None | 3,2 | | Ruml 1999 | USA | Calcium 0.8 g/d (n = 29)Placebo (n = 34) | None | 2,0 | | Peacock 2000 | USA | Calcium 0.75g/d (n = 126)D3 600 IU/d (n = 132)Placebo (n = 135) | None | 4,0 | | Chapuy 2002 | France | Calcium 1.2 g + D3 800 IU/d (n = 389)Placebo (n = 194) | None | 2,0 | | Meyer 2002 | Norway | D3 400 IU/d (n = 569) | None | 2,0 | ### Page 38 | Placebo (n = 575) | | - | | Latham 2003 | Australiaand NewZealand | VitD (D2 or D3?) 300 000 once (n = 121)Placebo (n = 122) | None | 0,5 | | Trivedi 2003 | UK | D3 100 000 IU every 4 mths (n = 1345)Placebo (n = 1341) | None | 5,0 | | Avenell 2004 | UK | Calcium 1 g/d (n = 29)D3 800 IU/d (n = 35)Calcium 1 g + D3 800 IU/d (n = 35)Placebo (n = 35) | None | 3,8 | | Dhesi 2004 | UK | D2 600000 IU once (IM) (n = 70)Placebo (IM) once (n = 69) | None | 0,5 | | Fujita 2004 | Japan | Calcium 0.9 g/d (n = 38)Placebo (n = 20) | None | 2,0 | | Harwood 2004 | UK | D2 300000 IU inj once (n = 38)Calcium 1g/d + D2 inj once (n = 36)Calcium 1g/d + D3 800 IU/d (n = 39)No treatment (n = 37) | None | 1,0 | | Grant 2005 | UK | Calcium 1 g/d (n = 1311)D3 800 IU/d (n = 1343)Calcium 1 g + D3 800 IU/d (n = 1306)Placebo (n = 1332) | None | 3,8 | | Porthouse 2005 | UK | Calcium 1 g + D3 800 IU/d (n = 1321)No treatment (n = 1993) | All participants receivedgeneral lifestyle advice onhow to reduce their risk offracture and fall. | 2,0 | | Jackson 2006 | USA | Calcium 1 g + D3 400 IU/d (n = 18176)Placebo (n = 18106) | Factorial design : a) calciumand vitamin D, b) hormonetherapy, c) dietarymodification. | 7,0 | | Law 2006 | UK | D2 100 000 IU every 3 mths (n = 1762)No treatment (n = 1955) | None | 0,8 | | Prince 2006 | Australia | Calcium 1.2 g/d (n = 730)Placebo (n = 730) | None | 5,0 | | Reid 2006 | NewZealand | Calcium 1 g/d (n = 732)Placebo (n = 739) | None | 4,5 | | Bolton-S. 2007 | UK | Calcium 1 g/d + D3 400 IU/d (n = 61)Placebo (n = 61) | Groups with vitamin K wereexcluded from the analysis,as data on fractures wasunavailable for thesegroups. | 2,0 | | Broe 2007 | USA | D2 200-800 IU/d (n = 99)Placebo (n = 25) | None | 0,4 | ### Page 39 | Lyons 2007 | UK | D2 100000 every 4 mths (n = 1725)Placebo every 4 mths (n = 1715) | None | 3,0 | | - | - | - | - | - | | Smith 2007 | UK | IM D2 300 000 IU every yr (n = 4727)IM Placebo every yr (n = 4713) | None | 3,0 | | Reid 2008 | NewZealand | Calcium 0.6-1.2 g/d (mean 0.9) (n = 216)Placebo (n = 107) | None | 2,0 | | Salovaara 2010 | Finland | Calcium 1 g/d + D3 800 IU/d (n = 1586)No treatment (n = 1609) | None | 3,0 | | Sanders 2010 | Australia | D3 500000 IU every yr (n = 1131)Placebo (n = 1127) | None | 3,0 | | Witham 2010 | UK | D3 100 000 IU every 10 wks (n = 53)Placebo (n = 52) | None | 0,4 | | Mitri 2011 | USA | Calcium 0.8 g/d (n = 22)D3 2000 IU/d (n = 23)Calcium 0.8 g + D3 2000 IU/d (n = 23)Placebo (n = 24) | None | 0,3 | | Glendenning2012 | Australia | D3 150 000 IU every 3 mths (n = 353)Placebo (n = 333) | None | 0,75 | | Punthakee 2012 | Multiple | D3 1000 IU/d (n = 607)Placebo (n = 614) | Factorial design : a) vitaminD, b) rosiglitazone, c)pioglitazone. | 0,4 | | Breslavsky 2013 | Israel | D3 1000 IU/d (n = 24)Placebo (n = 23) | None | 1,0 | | Hewitt 2013 | Australiaand NewZealand | D3 50 000 IU/month with LD (n = 30)Placebo (n = 30) | Multiple phosphate binders(including calcium) andcalcitriol | 0,5 | | Macdonald 2013 | UK | D3 400-1000 IU/d (mean 700) (n = 203)Placebo (n = 102) | None | 1,0 | | Witham 2013 | UK | D3 100 000 IU every 3 mths (n = 80)Placebo (n = 79) | None | 1,0 | | Hidalgo 2014 | Spain | Calcium 1 g/d + D3 800 UI/d (n = 188)Placebo (n = 210) | None | 2,0 | | Massart 2014 | Belgium | D3 25 000 IU/week (n = 26)Placebo (n = 29) | Active vitamin D andmineral bonedisorder–related drugs wereaccepted. | 0,3 | | Hansen 2015 | USA | D3 800 IU/d (n = 75)D3 50000 IU bimonthly (n = 79) with LDPlacebo (n = 76) | Patients consuming lessthan 600 mg/d or more than1400 mg/d of calcium werecounseled to consume600-1400 mg/d. | 1,0 | | Liu 2015 | China | Calcium 1.5 g/d + D3 600 IU/d (n = 100) | Factorial design : a) calcium | 1,0 | ### Page 40 | Placebo (n = 98) | and vitamin D, b) specificset of exercises (modifiedeighth section ofeight-section Brocadetherapy). | | - | - | | Cangussu 2016 | Brazil | D3 1000 IU/d (n = 80)Placebo (n = 80) | None | 0,75 | | Imaoka 2016 | Japan | Calcium 0.2 g + D3 900 IU/d (n = 23)Usual care (n = 23) | Usual care : environmentaladaptations,falls-prevention education, aconference with caregivers,a selection of walking aids,and sessions of exercises. | 0,5 | | Jin 2016 | Australia | D3 50000 IU/month (n = 209)Placebo (n = 204) | None | 2,0 | | Hin 2017 | UK | D3 2000-4000 IU/d (n = 204)Placebo (n = 101) | None | 1,0 | | Khaw 2017 | NewZealand | D3 100000 IU/month (n = 2558) with LDPlacebo (n = 2552) | None | 3,4 | | Levis 2017 | USA | D3 4000 IU/d (n = 66)Placebo (n = 64) | None | 0,75 | | Schwetz 2017 | Austria | D3 90000 IU/month (n = 249) with LDPlacebo (n = 243) | None | 0,5 | | Uusi-Rasi 2017 | Finland | D3 800 IU/d (n = 204)Placebo (n = 205) | Factorial design : a) vitaminD, b) exercise program | 2,0 | | Xue 2017 | China | Calcium 0.6 g + D3 800 IU/d (n = 139)No treatment (n = 170) | Maintain original lifestyleand receive regular healtheducation | 2,0 | | Larsen 2018 | Norway | D3 20000 IU/week (n = 256)Placebo (n = 255) | None | 5,0 | | Aloia 2019 | USA | D3 2400-4800 IU/d (n = 130)Placebo (n = 130) | Calcium supplements toeach group to achieve a totalintake of 1200 mg per day | 3,0 | | Shea 2019 | USA | D3 800 IU/d (n = 49)Placebo (n = 51) | After 4 mths, the vitamin Dgroup with serum 25(OH)Dwas < 70 nmol/L was given1600 IU/d instead of 800. | 1,0 | | Urashima 2019 | Japan | D3 2000 IU/d (n = 251)Placebo (n = 166) | None | 3,5 | | Bischoff-F. 2020 | Europe | D3 2000 IU/d (n = 1076)Placebo (n = 1081) | Factorial design : a) vitaminD, b) omega-3s, c) strengthtraining exercise program. | 3,0 | | Prithiani 2021 | Pakistan | D3 100 000 IU monthly (n = 170) | None | 2,0 | ### Page 41 | Placebo (n = 173) | | - | | Leboff 2022 | USA | D3 2000 IU/d (n = 12927)Placebo (n = 12944) | Factorial design : a) vitaminD, b) n-3 fatty acids. | 5,3 | | Brimble 2022 | Canada | D3 10000 IU/wk (n = 34)Placebo (n = 31) | Factorial RCT with BIAguided volume management | 2,5 | | Garcia-G. 2022 | Spain | Calcium 0.6 g + D3 400 IU/d (n = 16)No treatment (n = 9) | Both groups receivedhigh-impact training. | 2,0 | | Houston 2023 | USA | D3 2000 IU/d (n = 66) with LDPlacebo (n = 70) | None | 1,0 | | Joseph 2023 | Multiple | D3 60 000 IU monthly (n = 2835)Placebo (n = 2835) | Factorial RCT with polypill,aspirin and vitD vs placebo | 4,6 | | Waterhouse2023 | Australia | D3 60000 IU monthly (n = 10662)Placebo (n = 10653) | None | 5,1 | | Bischoff-F. 2024 | Switzerland | D3 24000 IU monthly (n = 46)Placebo (n = 45) | Factorial design : a)testosterone, b) vitamin D. | 1,0 | | Bowles 2024 | UK | D3 50000 IU once (n = 11)D3 150000 IU once (n = 11)D3 500000 IU once (n = 11)No treatment (n = 27) | None | 0,25 | Legend. LD : Loading dose. B. | Source | Nontrial use of calcium and vitamin D | Diet calciumintake, mg/d | Baseline 25-OHD, nmol/L | Achieved 25-OHD, nmol/L | | - | - | - | - | - | | Inkovaara 1983 | Accepted, not recorded. | - | Missing | Missing | | Hansson 1987 | Not stated, not recorded. | Missing | Missing | Missing | | Chapuy 1992 | Not stated, not recorded. | 513 | Cal+VitD : 40,0C : 32,5 | Cal+VitD : 105,0C : 27,5 | | Schwellnus 1992 | Not stated, not reported | Missing | Missing | Missing | | Reid 1993 | Not stated, not reported | 750 | Cal : 95,0C : 93,0 | Missing | | Graafmans 1996 | Not stated, not recorded. | Missing | Missing | Missing | | Lips 1996 | Calcium : not stated, not recorded.VitD : discouraged, but accepted. 2.8%in the treatment group and 2.9% in thecontrol group took nontrial vitD. | 867 | Missing | Missing | | Recker 1996 | Not stated, not recorded. | 434 | Cal : 62,5C : 65,0 | Missing | | Dawson-H. 1997 | Refused, not recorded. | 722 | Cal+VitD : 75,4 | Cal+VitD : 106,7 | ### Page 42 | C : 71,9 | C : 71,7 | | - | - | | Riggs 1998 | Calcium : accepted up to 500 mg/d, notrecorded. VitD : accepted up to 800IU/d, not recorded. | 714 | Cal : 76,0C : 74,3 | Missing | | Baron 1999 | Calcium : refused, but nearly half tooknontrial calcium supplements (nodifferences between groups). VitD : notstated, not recorded. | 877 | Cal : 73,0C : 72,8 | Missing | | Ruml 1999 | Not stated, not recorded. | 613 | Missing | Missing | | Peacock 2000 | Not stated, not recorded. | 604 | VitD : 59,6Cal : 64,0C : 61,4 | VitD : 118,6Missing forother groups | | Chapuy 2002 | Not stated, not recorded. | 557 | Cal+VitD : 21,9C : 22,8 | Missing | | Meyer 2002 | Accepted. 3.4% in the control groupand 2.4% in the treatment group tooknontrial calcium. VitD accepted up to400 IU/d : 43% in the control group and40% in the treatment group tooknontrial vitD. | 451 | VitD : 47,0C : 51,0 | VitD : 64,0C : 46,0 | | Latham 2003 | Not stated, not recorded | Missing | VitD : 37,5C : 47,5 | VitD : 60C : 47,5 | | Trivedi 2003 | Calcium : not stated, not recorded.VitD : accepted, not recorded. | 742 | Missing | Missing | | Avenell 2004 | Calcium accepted up to 1000 mg/d andvitD up to 200 IU/d. Less than 3% weretaking nontrial supplements. | Missing | Missing | Missing | | Dhesi 2004 | Not stated, not recorded. | Missing | VitD : 26,0C : 25,0 | VitD : 43,8C : 31,5 | | Fujita 2004 | Not stated, not recorded. | 600 | Missing | Missing | | Harwood 2004 | Calcium : not stated, not recorded.VitD : refused, not recorded. | Missing | VitD : 28,0Cal+VitD : 30,0and 29,0C : 29,0 | VitD : 40,0Cal+VitD : 44,0and 50,0C : - | | Grant 2005 | Calcium accepted up to 1000 mg/d andvitD up to 200 IU/d. Less than 3% weretaking nontrial supplements. | Missing | All : 38,0 | Cal : 41,5VitD : 62,3Cal+VitD : 62,0C : 45,8 | | Porthouse 2005 | Calcium accepted up to 500 mg/d, notrecorded. VitD : not stated, notrecorded. | 1081 | Missing | Missing | | Jackson 2006 | Calcium : accepted up to 1000 mg/d,not recorded. VitD : accepted up to 600 | Missing | Missing | Missing | ### Page 43 | IU/d, not recorded. | | - | | Law 2006 | Refused. Less than 1% took nontrialcalcium or vitD supplements. | Missing | Missing | Missing | | Prince 2006 | Not stated, not recorded | 906 | Total :Winter : 67,0Summer : 87,0 | Missing | | Reid 2006 | Not stated, not recorded | 857 | Cal : 51,5C : 52,0 | Missing | | Bolton-S. 2007 | Calcium : accepted up to 500 mg/d, notrecorded. VitD : accepted up to 400IU/d, not recorded. | 1073 | Cal+VitD : 62,5C : 57,0 | Cal+VitD : 74,5C : 48,8 | | Broe 2007 | Calcium : accepted, not recorded. VitD :accepted in a multivitamin (up to 400IU/d). 68,0f% in the control group and61,6% in the treatment group tooknontrial vitD supplements. | Missing | VitD : 47,6C : 53,0 | VitD : 62,7C : 60,3 | | Lyons 2007 | Calcium : not stated, not recorded.VitD : accepted up to 400 IU/d, use notrecorded. | Missing | Missing | VitD : 80,1C : 54,0 | | Smith 2007 | Calcium : not stated, not recorded.VitD : accepted up to 400 IU/d, notrecorded. | Missing | Missing | Missing | | Reid 2008 | Not stated, not recorded. | 867 | Cal : 91,3C : 95,0 | Missing | | Salovaara 2010 | Both accepted. Calcium : 16.0% in thetreatment group and 18.0% in thecontrol group took nontrial calciumsupplements. VitD : unavailable. | 895 | Cal+VitD : 50,0C : 49,1 | Cal+VitD : 74,6C : 55,9 | | Sanders 2010 | Calcium : not stated, not recorded.VitD : accepted up to 400 IU/d, usepartially recorded (6% in the controlgroup and 3% in the treatment grouptook nontrial vitD supplements beyond400 UI/d). | 976 | VitD : 53,0C : 45,0 | Missing | | Witham 2010 | Not stated, not recorded. | Missing | VitD : 20,5C : 23,7 | VitD : 40,0C : 25,0 | | Mitri 2011 | Refused, but not recorded. | 859 | Cal : 62,5VitD : 66,3Cal+VitD : 56,0C : 60,5 | - | | Glendenning2012 | Calcium : accepted, not recorded.VitD : not stated, not recorded. | 863 | VitD : 65,0C : 66,5 | VitD : 74,6C : 60,2 | | Punthakee 2012 | Not stated. VitD : 13% in the treatment | Missing | Missing | Missing | ### Page 44 | group and 12% in the placebo grouptook nontrial vitD supplements. | | - | | Breslavsky 2013 | Not stated, not recorded. | Missing | VitD : 32,3C : 27,0 | VitD : 44,0C : 35,0 | | Hewitt 2013 | Calcium : accepted, mean use from 4.6to 5.2 g/d. VitD : accepted, notrecorded. | Missing | VitD : 45,0C : 40,0 | VitD : 87,5C : 40,0 | | Macdonald 2013 | Calcium : not stated, not recorded.VitD : refused, but not recorded. | 1281 | VitD : 33,3C : 35,8 | VitD : 70,6C : 32,0 | | Witham 2013 | Not stated, not recorded. | 1125 | VitD : 45,0C : 45,0 | VitD : 70,0C : 52,5 | | Hidalgo 2014 | Not stated, not recorded | Missing | VitD : 87,3C : 79,8 | Missing | | Massart 2014 | Calcium : accepted, mean calciumsupplement of 0.85 g/d (treatment) and0.91 g/d (placebo). VitD : refused. | 881 | VitD : 42,7C : 45,9 | VitD : 88,0C : 41,0 | | Hansen 2015 | Calcium : accepted, not recorded.VitD : accepted up to 400 IU/d (meanuse 196 IU/d). | 905 | VitD : 52,5C : 52,5 | VitD low : 67,5VitD high : 105,0C : 45,0 | | Liu 2015 | Not stated, not recorded. | 1500 | Missing | Missing | | Cangussu 2016 | Calcium : not stated, not reported.VitD : accepted up to 100 UI/d, notrecorded. | 1145 | VitD : 37,5C : 42,3 | VitD : 68,8C : 34,5 | | Imaoka 2016 | Not stated, not recorded. | Missing | Cal+VitD : 35,3C : 28,3 | Cal+VitD : 94,5C : 32,5 | | Jin 2016 | Not stated, not recorded | Missing | VitD : 43,7C : 43,8 | VitD : 84,3C : 50,5 | | Hin 2017 | Calcium : not stated. Approximately3% took nontrial calcium supplements.VitD : accepted up to 400 IU/d. 12%took nontrial vitD supplements | 711 | VitD : 49-55,0C : 47,0 | VitD : 102-137,0C : 53,0 | | Khaw 2017 | Calcium : not stated. 5% in both groupstook nontrial calcium supplements.VitD : accepted up to 600 or 800 IU/d(according to age). 8% in both groupstook nontrial vitD supplements. | Missing | VitD : 64,0C : 63,0 | VitD : 135,0C : 66,0 | | Levis 2017 | Calcium : accepted, not recorded. VitD :accepted up to 1000 IU/D, not recorded. | Missing | VitD : 57,8C : 56,5 | VitD : 114,9C : 59,9 | | Schwetz 2017 | Calcium : not stated, not recorded.VitD : accepted at the discretion oftreating physicians, not recorded. | Missing | VitD : 32,8C : 34,3 | VitD : 115,0C : 65,5 | | Uusi-Rasi 2017 | Not stated, not recorded | 1098 | VitD : 65,6 | VitD : 92,5 | ### Page 45 | C : 68,5 | C : 68,8 | | - | - | | Xue 2017 | Not stated, not recorded. | Missing | All : 30,1 | - | | Larsen 2018 | Calcium : accepted. 13.1% in thecontrol group and 8.0% in thetreatment group took nontrial calciumsupplements. VitD : accepted up to 400IU/d. 34.7% in the control group and25.9% in the treatment group tooknontrial vitD supplements. | Missing | D3 : 59,7C : 61,6 | D3 : 114,7C : 61,0 | | Aloia 2019 | Refused, not recorded. | 828 | VitD : 53,8C : 56,3 | VitD : 117,5C : 52,5 | | Shea 2019 | Calcium : not stated, not recorded.VitD : accepted. 9,8% in the controlgroup and 4,1 % in the treatmentgroup took nontrial vitD supplements. | 871 | VitD : 49,0C : 52,0 | VitD : 81,3C : 49,5 | | Urashima 2019 | Calcium : not stated, not recorded.VitD : accepted if medically necessary(but resulted in discontinuation of trialsupplement), use not recorded. | Missing | Missing | Missing | | Bischoff-F. 2020 | Calcium : accepted up to 500 mg/d, notrecorded. VitD : accepted up to 800IU/d, not recorded. | Missing | D3 : 56,0C : 56,0 | D3 : 94,0C : 61,0 | | Prithiani 2021 | Not stated, not recorded. | Missing | VitD : 56,1C : 58,1 | Missing | | Leboff 2022 | Calcium : accepted up to 1200mg/d.20.0% took nontrial calciumsupplements. VitD : accepted up to 800UI/d. 42,7% in the control group and42,5% in the treatment group tooknontrial vitD supplements. | Missing | VitD : 76,8C : 76,8 | VitD : 103,0C : 73,5 | | Brimble 2022 | Calcium : not stated, not recorded.ViD : refused, use recorded, butunavailable. | Missing | VitD : 51,6C : 48,5 | VitD : 68,9C : 40,3 | | Garcia-G. 2022 | Accepted, not recorded | Missing | Missing | Missing | | Houston 2023 | Accepted if already taken. 28.8% in thetreatment group and 40% in thecontrol group took nontrial calciumsupplements. VitD : accepted up to1000 IU/d. 45.5% in the treatmentgroup and 52.9% in the control grouptook nontrial vitD supplements. | 474 | VitD : 48,5C : 71,5 | VitD : 49,8C : 50,5 | | Joseph 2023 | Calcium : not stated, not recorded.VitD : accepted up to 400 IU/d, notrecorded. | Missing | Missing | Missing | ### Page 46 | Waterhouse 2023 | Calcium : not stated, not recorded.VitD : accepted up to 2000 IU/d. 20% inthe treatment group and 27% in thecontrol group took nontrial vitDsupplements. | Missing | Missing | VitD : 115,0C : 77,0 | | - | - | - | - | - | | Bischoff-F. 2024 | Not stated, not recorded | Missing | VitD : 65,3C : 68,5 | Missing | | Bowles 2024 | Not stated, not recorded | 790 | VitD : 20,7C: 55,3 | VitD : 54,2C: 50,5 | C. | Source | Population | Highrisk | Age(y) | Women,% | Previousfracture, % | Previousfall, % | OP | Residentialstatus | | - | - | - | - | - | - | - | - | - | | Inkovaara1983 | Institutionalisedolder adults | Yes | 79,5 | 82,6 | - | - | No | Institution | | Hansson 1987 | Postmenopausalwomen withvertebral fracture | Yes | 66,0 | 100 | 100 | - | Yes | Community | | Chapuy 1992 | Institutionalisedwomen | Yes | 84,0 | 100 | - | 13,0 | No | Institution | | Schwellnus1992 | Male military | No | - | - | - | - | No | Community | | Reid 1993 | Postmenopausalwomen | No | 58,0 | 100 | - | - | No | Community | | Graafmans1996 | ≥70 yrsinstitutionalised | Yes | 83,0 | 84,0 | - | 33,0 | No | Institution | | Lips 1996 | ≥70 yrs | Yes | 80,0 | 74,3 | - | - | No | Both | | Recker 1996 | Postmenopausalwomen | No | 74,2 | 100 | - | - | No | Community | | Dawson-H.1997 | ≥65 yrs | No | 71,0 | 55,3 | - | - | No | Community | | Riggs 1998 | Postmenopausalwomen | No | 66,3 | 100 | - | - | No | Community | | Baron 1999 | <80 yrs with ahistory of boweladenoma | No | 61,0 | 27,7 | - | - | No | Community | | Ruml 1999 | Postmenopausalwomen | No | 51,9 | 100 | - | - | No | Community | | Peacock 2000 | ≥60 yrs | No | 74,0 | 71,8 | - | - | No | Community | | Chapuy 2002 | Institutionalised | Yes | 85,1 | 100 | - | 16,2 | No | Institution | ### Page 47 | women | | - | | Meyer 2002 | Institutionalised | Yes | 84,7 | 75,9 | - | 33,1 | No | Institution | | Latham 2003 | Frail ≥ 65 years | No | 79,5 | 53,0 | 43,5 | 56,0 | No | Community | | Trivedi 2003 | 65-85 yrs | No | 74,8 | 24,2 | - | - | No | Community | | Avenell 2004 | ≥70 yrs with alow-traumafracture | Yes | 76,7 | 82,3 | 100 | - | Yes | Community | | Dhesi 2004 | ≥65 yrs fallers | Yes | 76,8 | 77,7 | - | 100 | No | Community | | Fujita 2004 | Hospitalised olderwomen | Yes | 80,7 | 100 | - | - | No | ? | | Harwood 2004 | Older adults afterhip fracture | Yes | 81,0 | 100 | 100 | - | No | Community | | Grant 2005 | ≥70 yrs with afracture | Yes | 77,0 | 85,0 | 100 | - | Yes | Community | | Porthouse2005 | ≥70 yrs | No | 76,9 | 100 | 58,5 | 34,0 | No | Community | | Jackson 2006 | 50-79 yrs women | No | 62,4 | 100 | 34,6 | 38,3 | No | Community | | Law 2006 | ≥60 yrsinstitutionalised | Yes | 85,0 | 76,0 | - | - | No | Institution | | Prince 2006 | ≥ 70 years women | No | 74,6 | 100 | 27,4 | - | No | Community | | Reid 2006 | Postmenopausalwomen | No | 74,3 | 100 | 28,6 | - | No | Community | | Bolton-S. 2007 | ≥60 yrs women | No | 68,6 | 100 | - | - | No | Community | | Broe 2007 | Institutionalised | Yes | 89,0 | 73,0 | - | 36,0 | No | Institution | | Lyons 2007 | Institutionalised | Yes | 84,0 | 76,0 | - | - | No | Institution | | Smith 2007 | ≥75 yrs | No | 79,1 | 53,9 | 37,9 | - | No | Community | | Reid 2008 | ≥40 yrs men | No | 56,3 | 0 | 52,3 | - | No | Community | | Salovaara2010 | ≥65 yrs women | No | 67,4 | 100 | 35,4 | - | No | Community | | Sanders 2010 | ≥70 yrs women | No | 76,0 | 100 | 34,6 | - | No | Community | | Witham 2010 | ≥70 yrs with heartfailure | No | 79,7 | 34,3 | - | - | No | Community | | Mitri 2011 | Adults at risk oftype 2 diabetes | No | 57,0 | 51,0 | - | - | No | Community | | Glendenning | ≥70 yrs women | No | 76,9 | 100 | - | 33,4 | No | Community | ### Page 48 | 2012 | | - | | Punthakee2012 | ≥50 years with type2 diabetes | No | 66,4 | 40,9 | 14,3 | - | No | Community | | Breslavsky2013 | Adults with type 2diabetes | No | 66,3 | 53,2 | - | - | No | Community | | Hewitt 2013 | Adults on chronichaemodialysis | No | 63,5 | 51,7 | 33,4 | 10,0 | No | Community | | Macdonald2013 | 60-70 yrs women | No | 64,6 | 100 | - | - | No | Community | | Witham 2013 | ≥70 yrs withhigh-blood pressure | No | 76,8 | 48,4 | - | - | No | Community | | Hidalgo 2014 | ≥65 yrs | No | 72,5 | 52,3 | - | 26,4 | No | Community | | Massart 2014 | Adults onhemodialysis withserum vitD < 75nmol/L | No | 64,1 | 43,6 | - | - | No | Community | | Hansen 2015 | ≥60 yrs women | No | 61,0 | 100 | - | - | No | Community | | Liu 2015 | Postmenopausalwomen withosteoporosis | Yes | 62,2 | 100 | - | - | Yes | Community | | Cangussu 2016 | 50-65 yrs womenwith a fall history | Yes | 59,1 | 100 | - | 100 | No | Community | | Imaoka 2016 | Instutionnalised | Yes | 83,6 | 76,1 | - | - | No | Institution | | Jin 2016 | 50-79 yrs with kneeosteoarthritis | No | 63,2 | 50,0 | - | - | No | Community | | Hin 2017 | ≥65 yrs | No | 71,3 | 50,0 | 30,0 | 13,0 | No | Community | | Khaw 2017 | 50-84 yrs | No | 65,9 | 42,0 | 46,5 | 6,0 | No | Community | | Levis 2017 | Male veterans | No | 72,4 | 0 | - | 6,2 | No | Community | | Schwetz 2017 | ICU adults withserum vitD ≤50nmol/L | No | 64,6 | 35,0 | - | - | No | Community | | Uusi-Rasi 2017 | 70-79 yrs womenwith a fall history | Yes | 74,2 | 100 | - | 100 | No | Community | | Xue 2017 | ≥45 yrs women | No | 63,6 | 100 | 14,1 | - | No | Community | | Larsen 2018 | 25-80 yrspre-diabetes | No | 61,8 | 38,2 | - | - | No | Community | | Aloia 2019 | ≥60 yrs women | No | 68,2 | 100 | 14,2 | - | No | Community | ### Page 49 | Shea 2019 | ≥60 yrs | No | 69,6 | 36,0 | - | - | No | Community | | - | - | - | - | - | - | - | - | - | | Urashima2019 | 30-90 yrs adultswith a digestivetract cancer | No | 66,0 | 33,8 | - | - | No | Community | | Bischoff-F.2020 | ≥60 yrs | No | 74,9 | 61,7 | - | 42,3 | No | Community | | Prithiani 2021 | 50-75 yrs | No | 61,5 | 52,8 | - | - | No | Community | | Leboff 2022 | ≥50 yrs men, ≥55yrs women | No | 67,1 | 50,6 | 10,3 | 26,9 | No | Community | | Brimble 2022 | Adults onperitoneal dialysis | No | 61,6 | 48,0 | - | - | No | Community | | Garcia-G. 2022 | Post-menopausal ≥55 yrs | No | 63,3 | 100 | 8,0 | 24,0 | No | Community | | Houston 2023 | 65-89 yrs | No | 73,4 | 49,3 | - | 54,4 | No | Community | | Joseph 2023 | ≥50 or 55 yrs | No | 63,9 | 53,0 | - | - | No | Community | | Waterhouse2023 | 60-85 yrs | No | 69,3 | 45,7 | - | - | No | Community | | Bischoff-F.2024 | Pre-frail,hypogonadal ≥ 65years men | No | 72,2 | 0 | - | 53,8 | No | Community | Legend. OP : Osteoporosis. # # # ### Page 50 # Table S6. Risk of bias assessment by individual studies A. Fracture assessment | Source | D1 | D2 | D3 | D4 | D5 | Overall | | - | - | - | - | - | - | - | | Inkovaara 1983 | High | Low | High | High | Low | High | | Hansson 1987 | High | Someconcerns | High | Low | High | High | | Chapuy 1992 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Schwellnus 1992 | High | High | High | Low | Low | High | | Reid 1993 | Someconcerns | Low | Someconcerns | High | Low | High | | Lips 1996 | Low | Low | Someconcerns | High | Low | High | | Recker 1996 | Low | Low | High | Low | Low | High | | Dawson-H. 1997 | Low | Low | Someconcerns | High | Low | High | | Riggs 1998 | Someconcerns | Low | Someconcerns | Low | Low | Someconcerns | | Baron 1999 | Low | Low | Someconcerns | High | Low | High | | Ruml 1999 | Someconcerns | Someconcerns | High | High | Low | High | | Peacock 2000 | Someconcerns | Someconcerns | High | High | Low | High | | Chapuy 2002 | Someconcerns | Low | Someconcerns | Low | Low | Someconcerns | | Meyer 2002 | Someconcerns | Low | Someconcerns | Low | Low | Someconcerns | | Trivedi 2003 | Low | Low | Low | High | Low | High | | Avenell 2004 | Low | Someconcerns | Someconcerns | High | Low | High | | Fujita 2004 | High | Someconcerns | High | Low | Someconcerns | High | ### Page 51 | Harwood 2004 | Low | High | High | High | Low | High | | - | - | - | - | - | - | - | | Grant 2005 | Low | Low | Low | Low | Low | Low | | Porthouse 2005 | Low | Someconcerns | Low | High | Low | High | | Jackson 2006 | Low | Low | Low | Low | Low | Low | | Law 2006 | Low | Low | Someconcerns | Someconcerns | Low | Someconcerns | | Prince 2006 | Low | Low | Low | High | Low | High | | Reid 2006 | Low | Low | Low | High | Low | High | | Bolton-S. 2007 | Low | Low | Low | High | Low | High | | Lyons 2007 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Smith 2007 | Low | Low | Low | High | Low | High | | Reid 2008 | Low | Low | Low | High | Low | High | | Salovaara 2010 | Low | Low | Low | High | Low | High | | Sanders 2010 | Low | Low | Low | Low | Low | Low | | Witham 2010 | Low | Low | Low | High | Low | High | | Mitri 2011 | Low | Low | Low | High | Low | High | | Glendenning 2012 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Punthakee 2012 | Low | Low | Low | Low | Low | Low | | Breslavsky 2013 | Someconcerns | Someconcerns | High | High | Low | High | | Hewitt 2013 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Macdonald 2013 | Low | Low | High | Low | Low | High | | Witham 2013 | Low | Low | Low | Low | Low | Low | | Hidalgo 2014 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Massart 2014 | Low | Low | Low | Low | Low | Low | | Hansen 2015 | Low | Low | Low | High | Low | High | | Liu 2015 | Someconcerns | Someconcerns | High | Someconcerns | High | High | ### Page 52 | Hin 2017 | Low | Low | Low | High | Low | High | | - | - | - | - | - | - | - | | Khaw 2017 | Low | Low | Low | Low | Low | Low | | Schwetz 2017 | Low | Someconcerns | Someconcerns | High | Low | High | | Uusi-Rasi 2017 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Xue 2017 | Someconcerns | High | High | Someconcerns | Someconcerns | High | | Larsen 2018 | Low | Low | Someconcerns | High | Low | High | | Aloia 2019 | Low | Low | Low | Low | Low | Low | | Urashima 2019 | Low | Low | Low | Low | Low | Low | | Bischoff-F. 2020 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Prithiani 2021 | High | High | Someconcerns | Low | Low | High | | Leboff 2022 | Low | Low | Low | Low | Low | Low | | Brimble 2022 | Low | Low | Low | Low | Low | Low | | Garcia-G. 2022 | High | High | Low | High | Low | High | | Houston 2023 | Low | Low | Low | Low | Low | Low | | Joseph 2023 | Low | Low | Low | Low | Low | Low | | Waterhouse 2023 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Bowles 2024 | High | Someconcerns | Low | Low | Low | High | B. Fall assessment | Source | D1 | D2 | D3 | D4 | D5 | Overall | | - | - | - | - | - | - | - | | Graafmans 1996 | Someconcerns | Low | Low | Someconcerns | Low | Someconcerns | | Dawson-H. 1997 | Low | Low | Someconcerns | High | Low | High | | Riggs 1998 | Someconcerns | Low | Someconcerns | Low | Low | Someconcerns | | Peacock 2000 | Some | Some | High | High | Low | High | ### Page 53 | concerns | concerns | | - | - | | Chapuy 2002 | Someconcerns | Low | Someconcerns | Low | Low | Someconcerns | | Latham 2003 | Low | High | Low | Low | Low | High | | Trivedi 2003 | Low | Low | Someconcerns | High | Someconcerns | High | | Dhesi 2004 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Harwood 2004 | Low | High | High | High | Low | High | | Grant 2005 | Low | Low | Low | Low | Low | Low | | Porthouse 2005 | Low | Someconcerns | Low | High | Low | High | | Jackson 2006 | Low | Low | Low | Low | Low | Low | | Law 2006 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Reid 2006 | Low | Low | Low | High | Low | High | | Broe 2007 | Low | Low | Low | Low | Low | Low | | Smith 2007 | Low | Low | High | High | Low | High | | Reid 2008 | Low | Low | Low | High | Low | High | | Salovaara 2010 | Low | Low | Low | High | Low | High | | Sanders 2010 | Low | Low | Low | Low | Low | Low | | Witham 2010 | Low | Low | Low | High | Low | High | | Glendenning 2012 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Breslavsky 2013 | Someconcerns | Someconcerns | High | High | Low | High | | Hewitt 2013 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Macdonald 2013 | Low | Low | High | Low | Low | High | | Witham 2013 | Low | Low | Low | Low | Low | Low | | Hidalgo 2014 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Massart 2014 | Low | Low | Low | Low | Low | Low | | Hansen 2015 | Low | Low | Low | High | Low | High | ### Page 54 | Cangussu 2016 | Low | Low | Low | High | Low | High | | - | - | - | - | - | - | - | | Imaoka 2016 | Low | High | Someconcerns | High | Low | High | | Jin 2016 | Low | Low | Someconcerns | High | Low | High | | Hin 2017 | Low | Low | Low | High | Low | High | | Khaw 2017 | Low | Low | Low | Low | Low | Low | | Levis 2017 | Low | Low | Low | Low | Low | Low | | Schwetz 2017 | Low | Someconcerns | Someconcerns | High | Low | High | | Uusi-Rasi 2017 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Xue 2017 | Someconcerns | High | High | Someconcerns | Someconcerns | High | | Aloia 2019 | Low | Low | Low | Low | Low | Low | | Shea 2019 | Low | Low | Low | Low | Low | Low | | Bischoff-F. 2020 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Prithiani 2021 | High | High | Someconcerns | Low | Low | High | | Leboff 2022 | Low | Low | Low | Low | Low | Low | | Garcia-G. 2022 | High | High | Low | High | Low | High | | Houston 2023 | Low | Low | Low | Low | Low | Low | | Joseph 2023 | Low | Low | Low | Low | Low | Low | | Waterhouse 2023 | Low | Low | Someconcerns | Low | Low | Someconcerns | | Bischoff-F. 2024 | Low | Low | Low | Low | Low | Low | | Bowles 2024 | High | Someconcerns | Low | Low | Low | High | # # ### Page 55 # Table S7. Funding and conflict of interest statements of included studies | Source | Potential conflict of interest | Funding | | - | - | - | | Statement | Exists | | Inkovaara 1983 | No | ? | Not stated, drugs from industry | | Hansson 1987 | No | ? | Non-industry | | Chapuy 1992 | No | ? | Non-industry, drugs from industry | | Schwellnus 1992 | No | ? | Unavailable, drugs from industry | | Reid 1993 | No | ? | Non-industry, drugs from industry | | Graafmans 1996 | No | ? | Unavailable | | Lips 1996 | No | ? | Non-industry, drugs from industry | | Recker 1996 | No | ? | Non-industry, drugs from industry | | Dawson-H. 1997 | No | ? | Non-industry | | Riggs 1998 | No | ? | Non-industry, drugs from industry | | Baron 1999 | Yes | No | Non-industry, drugs from industry | | Ruml 1999 | No | ? | Unavailable, drugs from industry | | Peacock 2000 | No | ? | Non-industry, drugs from industry | | Chapuy 2002 | No | Yes | Industry | | Meyer 2002 | Yes | Yes | Industry | | Latham 2003 | No | ? | Non-industry | | Trivedi 2003 | Yes | No | Non-industry | | Avenell 2004 | Yes | Yes | Non-industry, drugs from industry | | Dhesi 2004 | Yes | No | Non-industry | | Fujita 2004 | No | ? | Unavailable | | Harwood 2004 | No | ? | Industry | | Grant 2005 | Yes | Yes | Non-industry, drugs from industry | | Porthouse 2005 | Yes | Yes | Mixed industry/non-industry, drugs fromindustry | | Jackson 2006 | Yes | Yes | Non-industry, drugs from industry | | Law 2006 | No | ? | Non-industry, drugs from industry | | Prince 2006 | Yes | No | Non-industry, drugs from industry | | Reid 2006 | No | ? | Non-industry, drugs from industry | | Bolton-S. 2007 | Yes | No | Non-industry, drugs from industry | | Broe 2007 | No | ? | Non-industry | | Lyons 2007 | No | ? | Non-industry | | Smith 2007 | Yes | Yes | Non-industry | | Reid 2008 | Yes | Yes | Non-industry, drugs from industry | | Salovaara 2010 | Yes | No | Non-industry, drugs from industry | | Sanders 2010 | No | ? | Non-industry, drugs from industry | ### Page 56 | Witham 2010 | Yes | No | Non-industry | | - | - | - | - | | Mitri 2011 | Yes | No | Non-industry, drugs from industry | | Glendenning 2012 | Yes | No | Non-industry | | Punthakee 2012 | Yes | Yes | Industry | | Breslavsky 2013 | Yes | No | No funding | | Hewitt 2013 | Yes | Yes | Non-industry | | Macdonald 2013 | Yes | No | Non-industry | | Witham 2013 | Yes | No | Non-industry, drugs from industry | | Hidalgo 2014 | Yes | No | Non-industry, drugs from industry | | Massart 2014 | Yes | Yes | Industry | | Hansen 2015 | Yes | Yes | Non-industry | | Liu 2015 | Yes | No | Non-industry, drugs from industry | | Cangussu 2016 | Yes | No | Non-industry | | Imaoka 2016 | Yes | Yes | Industry | | Jin 2016 | Yes | No | Non-industry | | Hin 2017 | Yes | No | Non-industry, drugs from industry | | Khaw 2017 | Yes | No | Non-industry | | Levis 2017 | Yes | No | Non-industry, drugs from industry | | Schwetz 2017 | Yes | No | Non-industry | | Uusi-Rasi 2017 | Yes | No | Non-industry | | Xue 2017 | Yes | No | Non-industry | | Larsen 2018 | Yes | No | Mixed industry/non-industry, drugs fromindustry | | Aloia 2019 | Yes | No | Non-industry | | Shea 2019 | Yes | Yes | Mixed industry/non-industry | | Urashima 2019 | No | ? | Non-industry | | Bischoff-F. 2020 | Yes | Yes | Mixed industry/non-industry, drugs fromindustry | | Prithiani 2021 | Yes | No | Unavailable | | Leboff 2022 | Yes | Yes | Non-industry, drugs from industry | | Brimble 2022 | Yes | Yes | Mixed industry/non-industry | | Garcia-G. 2022 | Yes | No | Unfunded | | Houston 2023 | Yes | No | Non-industry | | Joseph 2023 | Yes | Yes | Mixed industry/non-industry | | Waterhouse 2023 | Yes | Yes | Non-industry | | Bischoff-F. 2024 | Yes | No | Non-industry, drugs from industry | | Bowles 2024 | Yes | No | Non-industry, drugs from industry | ### Page 57 # Table S8. GRADE summary of findings for calcium monotherapy | Outcomes | Risk of anyfracture | Risk of hipfracture | Risk ofnon-vertebralfracture | Risk ofvertebralfracture | Risk of falling | Total numberof falls | | - | - | - | - | - | - | - | | Absoluteeffect | Control | 10.9% | 1.6% | 12.5% | 4.2% | 17.2% | 59 per 100patient-years | | Treatment** | 9.9%(8.8 to 11.0) | 2.6%(1.4 to 4.8) | 11.9%(9.9 to 14.3) | 3.4%(2.7 to 4.4) | 15.7%(13.4 to 18.2) | 60 per 100(57 to 64) | | Relative effect (95%CI) | RR 0.91(0.81 to 1.01) | RR 1.63(0.86 to 3.03) | RR 0.95(0.79 to 1.14) | RR 0.82(0.65 to 1.04) | RR 0.91(0.78 to 1.06) | IRR 1.02(0.96 to 1.09) | | No of patients (Trials) | 9067 (11) | 6703 (6) | 3204 (4) | 6576 (10) | 2966 (2) | 1794 (2) | | CoE (GRADE) | ⊕⊕⊕⊝Moderate | ⊕⊝⊝⊝Very low | ⊕⊕⊕⊝Moderate | ⊕⊕⊕⊝Moderate | ⊕⊕⊕⊝Moderate | ⊕⊕⊕⊝Moderate | | Summary | Likely little tono effect | Very uncertaineffect | Likely little tono effect | Likely little tono effect | Likely little tono effect | Likely little tono effect | | Anticipated absolutedifference (95% CI) | -1%(-2.1 to +0.1) | +1%(-0.2 to +3.2) | -0.6%(-2.4 to +1.8) | -0.8%(-1.5 to +0.2) | -1.5%(-3.8 to + 1.0) | +1(-2 to +5) | | MCID | 2.0 | 0.7 | 2.0 | 2.0 | 3.0 | 5 | | Target of certaintyrating | Little to nodifference | Increase theoutcome | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | | Risk of bias | Rate down one.Almost alltrials (10/11)are at highROB. The onlytrial atlow/moderateROB (Grant2005,accounting for38% of thepooling weight: RR 0.98; 95%CI, 0.81-1.18)was notconsideredsufficient tonot rate downone level. | Rate down one.Many trials(5/6) are athigh ROB, andresults weredifferent in theonly trial atlow/moderateROB (Grant2005,accounting for49.3% of thepooling weight: RR 1.21; 95%CI, 0.81-1.83). | Not serious.Many trials(3/4) are athigh ROB, butresultsremainedrobust in theonly trial atlow/moderateROB (Grant2005,accounting for87.5% of thepooling weight: RR 0.93; 95%CI, 0.76-1.13). | Rate down one.Many trials(9/10) are athigh ROB, andthe only trialatlow/moderateROB (Grant2005)accounted foronly 1.1% ofthe poolingweight (with 4vertebralfracturesreported for2644participants). | Not serious.One trial(Grant 2005) isat low ROB,and anotherone (Reid 2008)is at high ROB.Results wereconsistent inthe two trials. | Rate down one.The twoincluded trialsare at highROB. | | Imprecision | Not ratingdown even ifthe CI includesimportantbenefits. As wealready rateddown one forROB, the panelagreed thatrating downagain would be | Rate down one,as the CI isvery wide. | Rate down one,as the CIincludesimportantbenefits. | Not serious.The CI doesnot includeimportantbenefits orharms. | Rate down one,as the CIincludesimportantbenefits. | Not seriouseven if the CIincludesimportantbenefits. As wealready rateddown one forROB, the panelagreed thatrating downtwo levels | ### Page 58 | too severe. | would be toosevere. | | - | - | | Inconsistency | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Rate down one,even if lowstatisticalheterogeneity(Tau² = 0.10; I²= 24%). Thepanel found itvery unlikelythat calciumalone wouldincrease hipfractures, andvitamin Dcombined withcalcium woulddecrease it. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | | Indirectess | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious,but the lownumber ofstudies limitedthisassessment. | | Publication bias | Not detected | Not detected | Not detected | Not detected | Not detected.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | Not detected.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | ### Page 59 *The risk in the control group was estimated using the sample-size–weighted mean of the control-group event rates, calculated by dividing the total number of control-group events by the total number of participants in the control arms across all studies in each meta-analysis. **The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect Legend CoE : Certainty of Evidence. MCID : Minimal clinically important difference. # ### Page 60 # Table S9. GRADE summary of findings for vitamin D monotherapy | Outcomes | Risk of anyfracture | Risk of hipfracture | Risk ofnon-vertebralfracture | Risk ofvertebralfracture | Risk of falling | Total numberof falls | | - | - | - | - | - | - | - | | Absoluteeffect | Control | 6.5% | 1.3% | 6.4% | 1.4% | 25.3% | 73 per 100patient-years | | Treatment** | 6.5%(6.2 to 6.9) | 1.4%(1.3 to 1.8) | 6.4%(6.1 to 6.8) | 1.4%(0.9 to 2.3) | 25.3%(24.8 to 26.3) | 73 per 100(69 to 77) | | Relative effect (95%CI) | RR 1.00(0.95 to 1.06) | RR 1.13(1.00 to 1.27) | RR 1.01(0.95 to 1.06) | RR 1.04(0.66 to 1.65) | RR 1.01(0.98 to 1.04) | IRR 1.00(0.95 to 1.06) | | No of patients (Trials) | 92415 (36) | 74344 (15) | 77329 (20) | 11803 (7) | 65234 (32) | 68022 (12) | | CoE (GRADE) | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | | Summary | Little to noeffect | Little to noeffect | Little to noeffect | Little to noeffect | Little to noeffect | Little to noeffect | | Anticipated absolutedifference (95% CI) | 0%(-0.3 to +0.4) | +0.1%(0 to +0.5) | 0%(-0.3 to + 0.4) | 0%(-0.5 to +0.9) | 0%(-0.5 to +1.0) | 0%(-4 to +4) | | MCID | 2.0 | 0.7 | 2.0 | 2.0 | 3.0 | 5 | | Target of certaintyrating | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | | Risk of bias | Not serious.After excludingtrials at highROB (17/36),resultsremainedrobust in the19 remainingtrials (RR 1.00;95% CI,0.95-1.06). | Not serious.After excludingtrials at highROB (6/15),resultsremainedrobust in the 9remainingtrials (RR 1.10;95% CI,0.96-1.26). | Not serious.After excludingtrials at highROB (7/20),resultsremainedrobust in the13 remainingtrials (RR 0.99;95% CI,0.94-1.05). | Not serious.After excludingtrials at highROB (3/7),resultsremainedrobust in the 4remainingtrials (RR 1.18;95% CI,0.77-1.82). | Not serious.After excludingtrials at highROB (12/32),resultsremainedrobust in the20 remainingtrials (RR 1.02;95% CI,0.99-1.06). | Not serious.After excludingtrials at highROB (3/12),resultsremainedrobust in the 9remainingtrials (IRR1.01; 95% CI,0.93-1.09). | | Imprecision | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | | Inconsistency | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 4%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Tau² = 0.07; I²= 25%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 15%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 25%. | ### Page 61 | Indirectess | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious,but the lownumber ofstudies limitedthisassessment.The majority oftrials assessedpeople > 65 yrsliving in thecommunity,using lowvitamin D dose(< 1000 IU/d). | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | | - | - | - | - | - | - | - | | Publication bias | Not detected | Not detected | Not detected | Not detected | Detected, butonly withEgger’s test (p= 0.0301). Notdowngraded,due to thefunnel plot'svisualinspection.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | Not detected.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | *The risk in the control group was estimated using the sample-size–weighted mean of the control-group event rates, calculated by dividing the total number of control-group events by the total number of participants in the control arms across all studies in each meta-analysis. **The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect ### Page 62 Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect Legend CoE : Certainty of Evidence MCID : Minimal clinically important difference. # ### Page 63 # Table S10. GRADE summary of findings for vitamin D combined with calcium | Outcomes | Risk of anyfracture | Risk of hipfracture | Risk ofnon-vertebralfracture | Risk ofvertebralfracture | Risk of falling | Total numberof falls | | - | - | - | - | - | - | - | | Absoluteeffect | Control* | 11.3% | 1.8% | 12.4% | 1.0% | 33.7% | 40 per 100patient-years | | Treatment** | 10.3%(9.5 to 11.2) | 1.5%(1.3 to 1.7) | 10.8%(9.7 to 11.9) | 0.9%(0.7 to 1.1) | 31.0%(28.3 to 33.7) | 38 per 100(35 to 41) | | Relative effect (95%CI) | RR 0.91(0.84 to 0.99) | RR 0.84(0.74 to 0.96) | RR 0.87(0.78 to 0.96) | RR 0.90(0.74 to 1.09) | RR 0.92(0.84 to 1.00) | IRR 0.95(0.88 to 1.03) | | No of patients (Trials) | 51126 (15) | 49909 (9) | 10324 (7) | 42185 (4) | 11068 (10) | 7532 (4) | | CoE (GRADE) | ⊕⊕⊕⊕High | ⊕⊕⊕⊕High | ⊕⊕⊕⊝Moderate | ⊕⊕⊕⊕High | ⊕⊕⊕⊝Moderate | ⊕⊕⊝⊝Low | | Summary | Little to noeffect | Little to noeffect | Likely little tono effect | Little to noeffect | Likely little tono effect | May have littleto no effect | | Anticipated absolutedifference (95% CI) | -1.0%(-1.8 to -0.1) | -0.3%(-0.5 to -0.1) | -1.6%(-2.7 to -0.5) | -0.1%(-0.3 to +0.1) | -2.7%(-5.4 to 0) | -2(-5 to +1) | | MCID | 2.0 | 0.7 | 2 | 2 | 3 | 5 | | Target of certaintyrating | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | Little to nodifference | | Risk of bias | Not serious.After excludingtrials at highROB (10/15),resultsremainedrobust in the 5remainingtrials (RR 0.94;95% CI,0.86-1.02). | Not serious.After excludingtrials at highROB (5/9),resultsremainedrobust in the 4remainingtrials (RR 0.84;95% CI,0.73-0.96). | Not serious.After excludingtrials at highROB (4/7),resultsremainedrobust in the 3remainingtrials (RR 0.88;95% CI,0.79-0.99). | Not serious.After excludingtrials at highROB (2/4),resultsremainedrobust in the 2remainingtrials (RR 0.91;95% CI,0.75-1.11). | Not serious.After excludingtrials at highROB (7/10),resultsremainedrobust in the 3remainingtrials (RR 0.94;95% CI,0.70-1.26). | Rate down.Many trials(3/4) are at highROB(accounting for92.3% of thepooled weight),and the onlytrial atlow/moderateROB is showinga trend towardimportantreduction intotal falls(Hidalgo 2014 :IRR 0.79; 95%CI, 0.61-1.03). | | Imprecision | Not serious.The CI doesnot includeimportantbenefits orharms. | Not serious.The CI doesnot includeimportantbenefits orharms. | Rate down one,as the CIincludesimportantbenefits. | Not serious.The CI doesnot includeimportantbenefits orharms. | Rate down one,as the CIincludesimportantbenefits. | Rate down one,as the CIincludesimportantbenefits. | ### Page 64 | Inconsistency | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 13%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Not serious.Lowheterogeneity.Tau² = 0.00; I²= 0%. | Although therewas someinconsistency(Tau² = 0.01; I²= 40%), wejudged thatthis wascapturedlargely by theimprecision,which wasrated down. | Although therewas someinconsistency(Tau² = 0.00; I²= 40%), wejudged that thiswas capturedlargely by theimprecision,which was rateddown. | | - | - | - | - | - | - | - | | Indirectess | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious,but the lownumber ofstudies limitedthisassessment. | Not serious. Awide range ofpopulationsandinterventionswere included.The resultswere consistentin subgroupanalysesassessingdifferences inpopulationsandinterventions. | Not serious, butthe low numberof studieslimited thisassessment. | | Publication bias | Not detected | Not detected | Not detected | Not detected | Not detected.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | Possible.Althoughreporting biasis a potentialconcern forfalls (due toinconsistentreporting offalls versusfallers), wedecided not todowngradebecause thereis insufficientevidence thatunreporteddata wouldmeaningfullychange theconclusions. | *The risk in the control group was estimated using the sample-size–weighted mean of the control-group event rates, calculated by dividing the total number of control-group events by the total number of participants in the control arms across all studies in each meta-analysis. **The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ### Page 65 GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect Legend CoE : Certainty of Evidence MCID : Minimal clinically important difference. ### Page 66 # Table S11. Subgroup analyses for calcium monotherapy A. Risk of any fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 8 | 6214 | 0.87 (0.75-1.00) | 0.33 | | High risk | 3 | 2853 | 0.97 (0.81-1.17) | | Age (years) | | ≥65 | 7 | 6281 | 0.93 (0.83-1.05) | 0.01** | | <65 | 4 | 2786 | 0.46 (0.27-0.80) | | Age (years) | | Not conducted. All trials reported a mean age of <80 yrs. | | Sex | | Women | 3 | 1842 | 0.90 (0.74-1.10) | 0.96 | | Both or men | 8 | 7225 | 0.91 (0.79-1.04) | | Type of residence | | Community | 10 | 8921 | 0.91 (0.81-1.02) | 0.39 | | Institution | 1 | 146 | 0.44 (0.08-2.31) | | Previous fracture | | Yes | 2 | 2707 | 0.98 (0.82-1.18) | 0.27 | | No | 9 | 6360 | 0.86 (0.74-1.00) | | Previous fall | | Not conducted. No trial studied only people with a previous fall. | | Osteoporosis diagnosis | | Yes | 2 | 2707 | 0.98 (0.82-1.18) | 0.27 | | No | 9 | 6360 | 0.86 (0.74-1.00) | | Mean baseline vitamin D levels (nmol/L) | | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Mean baseline vitamin D levels (nmol/L) | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported mean dietary calcium intake. | | Trial size (participants) | | ≥ 1000 | 4 | 6972 | 0.93 (0.82-1.05) | 0.19 | | < 1000 | 7 | 2095 | 0.72 (0.49-1.04) | ### Page 67 | Mean follow-up (years) | | - | | ≤ 1 | 2 | 1544 | 0.39 (0.11-1.41) | 0.20 | | > 1 | 9 | 7523 | 0.91 (0.81-1.02) | * Test for subgroup differences. P < 0.05 : statistically significant. ** We found that the risk of any fracture was significantly lower with calcium in the subgroup of participants younger than 65 years (RR 0.46, 95% CI 0.27-0.80) compared to those 65 years and older (RR 0.93, 95% CI 0.83-1.05) (p for interaction = 0.01). However, we considered this finding biologically and clinically very implausible. B. Risk of hip fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 4 | 3996 | 2.23 (0.70-7.06) | 0.12 | | High risk | 2 | 2707 | 1.24 (0.28-5.39) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. C. Risk of non-vertebral fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 2 | 494 | 1.02 (0.58-1.80) | 0.78 | | High risk | 2 | 2707 | 0.94 (0.77-1.14) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. D. Risk of vertebral fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 6 | 3760 | 0.82 (0.65-1.05) | 0.87 | | High risk | 4 | 2815 | 0.75 (0.23-2.39) | | Age (years) | | ≥ 65 | 9 | 6440 | 0.82 (0.65-1.04) | 0.57 | ### Page 68 | < 65 | 1 | 135 | 0.33 (0.01-7.92) | | - | - | - | - | | Age (years) | | ≥ 80 | 1 | 58 | 0.35 (0.06-1.93) | 0.32 | | < 80 | 9 | 6517 | 0.83 (0.66-1.05) | | Sex | | Women | 6 | 2147 | 0.78 (0.58-1.04) | 0.54 | | Both or men | 4 | 4428 | 0.90 (0.62-1.32) | | Type of residence | | No subgroup analysis. All trials included only community-dwelling adults. | | Previous fracture | | Yes | 3 | 2757 | 1.34 (0.29-6.15) | 0.52 | | No | 7 | 3818 | 0.81 (0.64-1.03) | | Previous fall | | Not conducted. No trial studied only people with a previous fall. | | Osteoporosis diagnosis | | Yes | 3 | 2757 | 1.34 (0.29-6.15) | 0.52 | | No | 7 | 3818 | 0.81 (0.64-1.03) | | Mean baseline vitamin D levels (nmol/L) | | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Mean baseline vitamin D levels (nmol/L) | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported mean dietary calcium intake. | | Trial size (participants) | | ≥ 1000 | 3 | 5574 | 0.87 (0.63-1.20) | 0.59 | | < 1000 | 7 | 1001 | 0.77 (0.55-1.08) | | Mean follow-up (years) | | Not conducted. All trials had a follow-up of > 1 year. | * Test for subgroup differences. P < 0.05 : statistically significant. E. Risk of falling | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 1 | 323 | 0.83 (0.64-1.07) | 0.36 | | High risk | 1 | 2643 | 0.96 (0.80-1.15) | ### Page 69 | Other subgroup analyses | | - | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. F. Total number of falls | Subgroup | No of studies | No of participants | IRR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not conducted, as the two included trials studied not at high-risk populations. | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. ### Page 70 # Table S12. Sensitivity analyses for calcium monotherapy | Analysis | No of participants(studies) | RR or IRR(95% CI) | Comparison to primary analysis | | - | - | - | - | | Fixed-effect model | | Risk of any fracture | 9067 (11) | RR 0.91 (0.81-1.01) | RR 0.91 (0.81-1.01) | | Risk of hip fracture | 6703 (6) | RR 1.45 (1.03-2.04) | RR 1.63 (0.86-3.08) | | Risk of non-vertebral fracture | 3204 (4) | RR 0.95 (0.79-1.14) | RR 0.95 (0.79-1.14) | | Risk of vertebral fracture | 6575 (10) | RR 0.82 (0.65-1.04) | RR 0.82 (0.65-1.04) | | Risk of falling | 2966 (2) | RR 0.91 (0.78-1.06) | RR 0.91 (0.78-1.06) | | Total number of falls | 1794 (2) | IRR 1.02 (0.96-1.09) | IRR 1.02 (0.96-1.09) | | Excluding trials at high risk of bias | | Risk of any fracture | 2879 (2) | RR 0.97 (0.81-1.17) | RR 0.91 (0.81-1.01) | | Risk of hip fracture | 2643 (1) | RR 1.21 (0.81-1.83) | RR 1.63 (0.86-3.08) | | Risk of non-vertebral fracture | 2879 (2) | RR 0.93 (0.77-1.13) | RR 0.95 (0.79-1.14) | | Risk of vertebral fracture | 2879 (2) | RR 1.05 (0.44- 2.47) | RR 0.82 (0.65-1.04) | | Risk of falling | 2643 (1) | RR 0.96 (0.80-1.15) | RR 0.91 (0.78-1.06) | | Total number of falls | Not conducted. All trials were at high risk of bias. | | Excluding trials not reporting at least 1000 patient-years of follow-up | | Risk of any fracture | 8163 (6) | RR 0.92 (0.82-1.04) | RR 0.91 (0.81-1.01) | | Risk of hip fracture | 6504 (4) | RR 1.75 (0.80-3.84) | RR 1.63 (0.86-3.08) | | Risk of non-vertebral fracture | 2904 (2) | RR 0.94 (0.78-1.14) | RR 0.95 (0.79-1.14) | | Risk of vertebral fracture | 5835 (4) | RR 0.83 (0.62-1.12) | RR 0.82 (0.65-1.04) | | Risk of falling | 2643 (1) | RR 0.96 (0.80-1.15) | RR 0.91 (0.78-1.06) | | Total number of falls | 1471 (1) | IRR 1.02 (0.96-1.09) | IRR 1.02 (0.96-1.09) | | Excluding trials that allowed nontrial supplements of calcium and/or vitamin D | | Risk of any fracture | 1460 (1) | RR 0.87 (0.69-1.10) | RR 0.91 (0.81-1.01) | | Risk of hip fracture | 1460 (1) | RR 1.83 (0.68-4.93) | RR 1.63 (0.86-3.08) | | Risk of non-vertebral fracture | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Risk of vertebral fracture | 1460 (1) | RR 0.97 (0.63-1.51) | RR 0.82 (0.65-1.04) | | Risk of falling | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Total number of falls | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Leave-one-out method | | Risk of any fracture | Results remained similar. | | Risk of hip fracture | Results remained similar. | ### Page 71 | Risk of non-vertebral fracture | Results remained similar. | | - | - | | Risk of vertebral fracture | Results remained similar. | | Risk of falling | Results remained similar. | | Total number of falls | Results remained similar. | # ### Page 72 # Table S13. Subgroup analyses for vitamin D monotherapy A. Risk of any fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 25 | 78705 | 1.01 (0.93-1.08) | 0.87 | | High risk | 10 | 13293 | 1.02 (0.92-1.12) | | Age (years) | | ≥ 65 | 25 | 84637 | 1.01 (0.95-1.06) | 0.58 | | < 65 | 10 | 7778 | 0.91 (0.63-1.31) | | Age (years) | | ≥ 80 | 5 | 9878 | 1.01 (0.89-1.14) | 1.00 | | < 80 | 30 | 82537 | 1.01 (0.94-1.07) | | Sex | | Women | 9 | 14004 | 1.10 (0.97-1.24) | 0.11 | | Both or men | 26 | 78411 | 0.98 (0.93-1.04) | | Type of residence | | Community | 31 | 89837 | 1.01 (0.94-1.07) | 0.62 | | Institution | 4 | 7381 | 0.97 (0.84-1.12) | | Previous fracture | | Yes | 2 | 2745 | 1.07 (0.89-1.28) | 0.49 | | No | 33 | 89670 | 1.00 (0.94-1.06) | | Previous fall | | Yes | 1 | 409 | 0.75 (0.32-1.75) | 0.50 | | No | 34 | 92006 | 1.01 (0.95-1.06) | | Osteoporosis diagnosis | | Yes | 2 | 2745 | 1.07 (0.89-1.28) | 0.49 | | No | 33 | 89670 | 1.00 (0.94-1.06) | | Mean baseline vitamin D levels (nmol/L) | | < 25 | 1 | 105 | 1.96 (0.18-20.99) | 0.59 | | ≥ 25 | 24 | 43665 | 1.03 (0.96-1.10) | | Mean baseline vitamin D levels (nmol/L) | | < 50 | 12 | 7509 | 1.06 (0.93-1.22) | 0.51 | | ≥ 50 | 13 | 36361 | 1.01 (0.9-1.09) | | Type of vitamin D | ### Page 73 | D2 | 4 | 15626 | 1.06 (0.90-1.23) | 0.48 | | - | - | - | - | - | | D3 | 31 | 76714 | 1.00 (0.94-1.05) | | Frequency of administration of vitamin D | | Daily to monthly | 23 | 65260 | 0.98 (0.93-1.04) | 0.45 | | Intermittent or highdoses | 12 | 27155 | 1.04 (0.91-1.19) | | Mean chronic daily dose of vitamin D excluding intermittent or high doses (IU/d) | | < 1000 | 7 | 7448 | 1.05 (0.92-1.19) | 0.56 | | 1000-2000 | 11 | 56108 | 0.97 (0.90-1.04) | | > 2000 | 5 | 1474 | 0.96 (0.57-1.64) | | Mean baseline dietary calcium intake (mg) | | < 800 | 5 | 4538 | 0.86 (0.72-1.03) | 0.03** | | ≥ 800 | 10 | 6985 | 1.11 (0.95-1.29) | | Trial size (participants) | | ≥ 1000 | 14 | 87215 | 1.01 (0.96-1.07) | 0.16 | | < 1000 | 21 | 5200 | 0.82 (0.62-1.09) | | Mean follow-up (years) | | ≤ 1 | 16 | 6720 | 0.89 (0.60-1.32) | 0.54 | | > 1 | 19 | 85695 | 1.00 (0.95-1.06) | * Test for subgroup differences. P < 0.05 : statistically significant. ** We found that the risk of any fracture was significantly lower with vitamin D in the subgroup of trials with mean dietary calcium intake of less than 800 mg/d (RR 0.86, 95% CI 0.72-1.03) compared to trials with higher intake (RR 1.11, 95% CI, 0.95-1.29) (p for interaction = 0.03). As the two treatment effects remained statistically non-significant and included low absolute differences (about 1.5%), we considered this finding of limited relevance. B. Risk of hip fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 7 | 61312 | 1.14 (0.96-1.35) | 0.90 | | High risk | 8 | 13032 | 1.12 (0.95-1.32) | | Age (years) | | Not conducted. All trials included only people with a mean age of ≥ 65 years. | | Age (years) | | ≥ 80 | 5 | 9878 | 1.11 (0.93-1.33) | 0.85 | | < 80 | 10 | 64466 | 1.14 (0.97-1.33) | ### Page 74 | Sex | | - | | Women | 5 | 12866 | 1.42 (1.02-1.96) | 0.14 | | Both or men | 10 | 61478 | 1.09 (0.96-1.24) | | Type of residence | | Community | 11 | 64541 | 1.14 (0.97-1.33) | 0.77 | | Institution | 3 | 7225 | 1.09 (0.89-1.34) | | Previous fracture | | Yes | 2 | 2745 | 1.11 (0.74-1.68) | 0.95 | | No | 13 | 71599 | 1.13 (1.00-1.28) | | Previous fall | | Yes | 1 | 409 | 7.03 (0.37-135.32) | 0.22 | | No | 14 | 73935 | 1.12 (1.00-1.27) | | Osteoporosis diagnosis | | Yes | 2 | 2745 | 1.11 (0.74-1.68) | 0.95 | | No | 13 | 71599 | 1.13 (1.00-1.28) | | Mean baseline vitamin D levels (nmol/L) | | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Mean baseline vitamin D levels (nmol/L) | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Type of vitamin D | | D2 | 3 | 15521 | 1.23 (0.95-1.60) | 0.44 | | D3 | 11 | 58748 | 1.10 (0.95-1.27) | | Frequency of administration of vitamin D | | Daily to monthly | 8 | 53120 | 1.11 (0.95-1.29) | 0.71 | | Intermittent or highdoses | 7 | 21224 | 1.16 (0.96-1.40) | | Mean chronic daily dose of vitamin D excluding intermittent or high doses (IU/d) | | Not conducted. Less than 10 trials reported daily to monthly administration ofvitamin D. | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported baseline calcium intake. | | Trial size (participants) | | ≥ 1000 | 10 | 73057 | 1.13 (1.01-1.28) | 0.43 | | < 1000 | 5 | 1287 | 0.64 (0.16-2.60) | | Mean follow-up (years) | | ≤ 1 | 4 | 3449 | 0.94 (0.37-2.42) | 0.71 | | > 1 | 11 | 70895 | 1.13 (1.00-1.27) | ### Page 75 * Test for subgroup differences. P < 0.05 : statistically significant. C. Risk of non-vertebral fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 13 | 66845 | 1.02 (0.95-1.09) | 0.58 | | High risk | 7 | 10484 | 0.98 (0.88-1.10) | | Age (years) | | ≥ 65 | 18 | 76939 | 1.01 (0.95-1.06) | 0.80 | | < 65 | 2 | 390 | 0.90 (0.38-2.15) | | Age (years) | | ≥ 80 | 3 | 7225 | 0.96 (0.83-1.12) | 0.55 | | < 80 | 17 | 70104 | 1.01 (0.96-1.07) | | Sex | | Women | 5 | 13134 | 1.11 (0.97-1.26) | 0.11 | | Both or men | 15 | 64195 | 0.99 (0.93-1.05) | | Type of residence | | Community | 17 | 70104 | 1.01 (0.96-1.07) | 0.55 | | Institution | 3 | 7225 | 0.96 (0.83-1.12) | | Previous fracture | | Yes | 2 | 2745 | 1.02 (0.85-1.24) | 0.84 | | No | 18 | 74584 | 1.00 (0.95-1.06) | | Previous fall | | Yes | 1 | 409 | 0.73 (0.30-1.78) | 0.48 | | No | 19 | 76920 | 1.01 (0.95-1.06) | | Osteoporosis diagnosis | | Yes | 2 | 2745 | 1.02 (0.85-1.24) | 0.84 | | No | 18 | 74584 | 1.00 (0.95-1.06) | | Mean baseline vitamin D levels (nmol/L) | | < 25 | 1 | 105 | 1.96 (0.18-20.99) | 0.58 | | ≥ 25 | 14 | 41307 | 1.01 (0.94-1.09) | | Mean baseline vitamin D levels (nmol/L) | | < 50 | 7 | 6522 | 1.05 (0.92-1.20) | 0.57 | | ≥ 50 | 8 | 34890 | 1.00 (0.92-1.09) | | Type of vitamin D | | D2 | 4 | 15626 | 1.05 (0.88-1.25) | 0.56 | ### Page 76 | D3 | 16 | 61703 | 1.00 (0.94-1.06) | | - | - | - | - | | Frequency of administration of vitamin D | | Daily to monthly | 12 | 53492 | 0.97 (0.91-1.04) | 0.10 | | Intermittent or highdoses | 8 | 23837 | 1.09 (0.97-1.22) | | Mean chronic daily dose of vitamin D excluding intermittent or high doses (IU/d) | | < 1000 | 5 | 4565 | 1.00 (0.86-1.17) | 0.87 | | 1000-2000 | 6 | 48584 | 0.97 (0.90-1.04) | | > 2000 | 1 | 343 | 0.81 (0.33-2.01) | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported baseline calcium intake. | | Trial size (participants) | | ≥ 1000 | 10 | 75060 | 1.01 (0.95-1.06) | 0.72 | | < 1000 | 10 | 2269 | 0.94 (0.64-1.37) | | Mean follow-up (years) | | ≤ 1 | 13 | 73508 | 1.00 (0.95-1.06) | 0.35 | | > 1 | 7 | 3821 | 1.19 (0.83-1.70) | * Test for subgroup differences. P < 0.05 : statistically significant. D. Risk of vertebral fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 3 | 5209 | 1.06 (0.36-3.12) | 0.91 | | High risk | 4 | 6594 | 0.99 (0.24-4.11) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. E. Risk of falling | Subgroup | No of studies | No of participants | RR (95% CI) | P value | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 23 | 58552 | 1.01 (0.98-1.05) | 0.36 | | High risk | 9 | 6682 | 0.96 (0.87-1.07) | | Age (years) | | ≥ 65 | 23 | 57506 | 1.02 (0.99-1.05) | 0.06 | ### Page 77 | < 65 | 9 | 7728 | 0.84 (0.70-1.02) | | - | - | - | - | | Age (years) | | ≥ 80 | 4 | 3194 | 1.02 (0.94-1.10) | 0.89 | | < 80 | 28 | 62040 | 1.01 (0.98-1.04) | | Sex | | Women | 10 | 14164 | 1.00 (0.93-1.07) | 0.72 | | Both or men | 22 | 51070 | 1.01 (0.98-1.04) | | Type of residence | | Community | 29 | 62115 | 1.01 (0.98-1.04) | 0.76 | | Institution | 3 | 3119 | 1.02 (0.94-1.11) | | Previous fracture | | Yes | 1 | 2675 | 1.11 (0.93-1.32) | 0.30 | | No | 31 | 62559 | 1.01 (0.98-1.04) | | Previous fall | | Yes | 3 | 708 | 0.75 (0.50-1.12) | 0.14 | | No | 29 | 64526 | 1.02 (0.99-1.05) | | Osteoporosis diagnosis | | Yes | 1 | 2675 | 1.11 (0.93-1.32) | 0.30 | | No | 31 | 62559 | 1.01 (0.98-1.04) | | Mean baseline vitamin D levels (nmol/L) | | < 25 | 1 | 105 | 0.39 (0.08-1.93) | 0.24 | | ≥ 25 | 25 | 42893 | 1.02 (0.98-1.06) | | Mean baseline vitamin D levels (nmol/L) | | < 50 | 15 | 7397 | 0.95 (0.82-1.10) | 0.45 | | ≥ 50 | 11 | 35601 | 1.01 (0.97-1.04) | | Type of vitamin D | | D2 | 5 | 12449 | 0.99 (0.96-1.02) | 0.36 | | D3 | 25 | 52467 | 1.01 (0.98-1.05) | | Frequency of administration of vitamin D | | Daily to monthly | 20 | 41850 | 1.01 (0.97-1.06) | 0.96 | | Intermittent or highdoses | 12 | 23384 | 1.01 (0.97-1.05) | | Mean chronic daily dose of vitamin D excluding intermittent or high doses (IU/d) | | < 1000 | 6 | 3967 | 1.00 (0.91-1.10) | 0.77 | | 1000-2000 | 8 | 36560 | 0.93 (0.77-1.12) | | > 2000 | 5 | 1093 | 1.00 (0.81-1.23) | ### Page 78 | Mean baseline dietary calcium intake (mg) | | - | | < 800 | 3 | 2479 | 0.96 (0.83-1.11) | 0.24 | | ≥ 800 | 9 | 4460 | Not estimable | | Trial size (participants) | | ≥ 1000 | 10 | 59951 | 1.02 (0.99-1.05) | 0.15 | | < 1000 | 22 | 5283 | 0.96 (0.89-1.04) | | Mean follow-up (years) | | ≤ 1 | 19 | 6499 | 1.00 (0.93-1.07) | 0.66 | | > 1 | 13 | 58735 | 1.01 (0.98-1.05) | F. Total number of falls | Subgroup | No of studies | No of participants | IRR (95% CI) | P value | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 9 | 67350 | 1.02 (0.96-1.07) | 0.11 | | High risk | 3 | 672 | 0.82 (0.63-1.06) | | Age (years) | | ≥ 65 | 10 | 62047 | 1.01 (0.96-1.07) | 0.36 | | < 65 | 2 | 5975 | 0.88 (0.65-1.19) | | Age (years) | | ≥ 80 | 1 | 124 | 0.98 (0.65-1.49) | 0.90 | | < 80 | 11 | 67898 | 1.01 (0.95-1.06) | | Sex | | Women | 4 | 12412 | 1.02 (0.90-1.16) | 0.61 | | Both or men | 8 | 55610 | 0.98 (0.94-1.03) | | Type of residence | | Community | 11 | 67898 | 1.01 (0.95-1.06) | 0.90 | | Institution | 1 | 124 | 1.02 (0.90-1.16) | | Previous fracture | | Not conducted. No trial studied only persons with a previous fracture. | | Previous fall | | Yes | 2 | 548 | 0.73 (0.53-1.02) | 0.06 | | No | 10 | 67474 | 1.01 (0.96-1.07) | | Osteoporosis diagnosis | | Not conducted. No trial studied only persons with osteoporosis. | | Mean baseline vitamin D levels (nmol/L) | | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | ### Page 79 | Mean baseline vitamin D levels (nmol/L) | | - | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Type of vitamin D | | D2 | 3 | 9703 | 0.98 (0.92-1.04) | 0.43 | | D3 | 8 | 58076 | 1.02 (0.94-1.10) | | Frequency of administration of vitamin D | | Daily to monthly | 8 | 55942 | 0.98 (0.93-1.02) | 0.29 | | Intermittent or highdoses | 4 | 12080 | 1.04 (0.93-1.17) | | Mean chronic daily dose of vitamin D excluding Intermittent or high doses (IU/d) | | Not conducted. Less than 10 trials used daily to monthly administration of vitaminD. | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported baseline calcium intake. | | Trial size (participants) | | ≥ 1000 | 6 | 66711 | 1.01 (0.95-1.07) | 0.91 | | < 1000 | 6 | 1311 | 1.00 (0.84-1.18) | | Mean follow-up (years) | | ≤ 1 | 5 | 902 | 1.06 (0.90-1.24) | 0.53 | | > 1 | 7 | 67120 | 1.00 (0.95-1.06) | ### Page 80 # Table S14. Sensitivity analyses for vitamin D monotherapy | Analysis | No of participants(studies) | RR or IRR(95% CI) | Comparison to primary analysis | | - | - | - | - | | Fixed-effect model | | Risk of any fracture | 92415 (36) | RR 1.00 (0.95-1.05) | RR 1.00 (0.95-1.06) | | Risk of hip fracture | 74344 (15) | RR 1.13 (1.00-1.27) | RR 1.13 (1.00-1.27) | | Risk of non-vertebral fracture | 77329 (20) | RR 1.01 (0.95-1.06) | RR 1.01 (0.95-1.06) | | Risk of vertebral fracture | 11803 (7) | RR 1.04 (0.77-1.41) | RR 1.04 (0.66-1.65) | | Risk of falling | 65234 (32) | RR 1.01 (0.99-1.03) | RR 1.01 (0.98-1.04) | | Total number of falls | 48798 (12) | IRR 0.99 (0.96-1.03) | IRR 1.00 (0.95-1.06) | | Excluding trials at high risk of bias | | Risk of any fracture | 74693 (19) | RR 1.00 (0.95-1.06) | RR 1.00 (0.95-1.06) | | Risk of hip fracture | 59448 (9) | RR 1.10 (0.96-1.26) | RR 1.13 (1.00-1.27) | | Risk of non-vertebral fracture | 67010 (13) | RR 0.99 (0.94-1.05) | RR 1.01 (0.95-1.06) | | Risk of vertebral fracture | 8780 (4) | RR 1.18 (0.62-2.26) | RR 1.04 (0.66-1.65) | | Risk of falling | 51085 (20) | RR 1.02 (0.99-1.06) | RR 1.01 (0.98-1.04) | | Total number of falls | 38810 (9) | IRR 1.01 (0.93-1.09) | IRR 1.00 (0.95-1.06) | | Excluding trials not reporting at least 1000 patient-years of follow-up | | Risk of any fracture | 87189 (16) | RR 1.01 (0.96-1.07) | RR 1.00 (0.95-1.06) | | Risk of hip fracture | 73057 (10) | RR 1.13 (1.01-1.28) | RR 1.13 (1.00-1.27) | | Risk of non-vertebral fracture | 75327 (11) | RR 1.01 (0.96-1.07) | RR 1.01 (0.95-1.06) | | Risk of vertebral fracture | 11324(5) | RR 1.04 (0.57-1.91) | RR 1.04 (0.66-1.65) | | Risk of falling | 49951 (10) | RR 1.02 (0.99-1.05) | RR 1.01 (0.98-1.04) | | Total number of falls | 66711 (6) | IRR 1.01 (0.95-1.07) | IRR 1.00 (0.95-1.06) | | Excluding trials that allowed nontrial supplements of calcium and/or vitamin D | | Risk of any fracture | 2948 (3) | RR 1.37 (0.90-2.10) | RR 1.00 (0.95-1.06) | | Risk of hip fracture | 2641 (1) | RR 1.35 (0.67-2.72) | RR 1.13 (1.00-1.27) | | Risk of non-vertebral fracture | 2688 (2) | RR 1.41 (0.92-2.16) | RR 1.01 (0.95-1.06) | | Risk of vertebral fracture | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Risk of falling | 2901 (1) | RR 1.02 (0.94-1.11) | RR 1.01 (0.98-1.04) | | Total number of falls | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Leave-one-out method | | Risk of any fracture | Results remained similar. | | Risk of hip fracture | Results remained similar. | ### Page 81 | Risk of non-vertebral fracture | Results remained similar. | | - | - | | Risk of vertebral fracture | Results remained similar. | | Risk of falling | Results remained similar. | | Total number of falls | Results remained similar. | ### Page 82 # Table S15. Subgroup analyses for vitamin D combined with calcium A. Risk of any fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 8 | 44094 | 0.96 (0.89-1.03) | 0.20 | | High risk | 7 | 7032 | 0.87 (0.78-0.98) | | Age (years) | | ≥ 65 | 11 | 14309 | 0.87 (0.79-0.96) | 0.06 | | < 65 | 4 | 36817 | 0.97 (0.92-1.03) | | Age (years) | | ≥ 80 | 3 | 3965 | 0.85 (0.74-0.97) | 0.10 | | < 80 | 12 | 47161 | 0.96 (0.91-1.01) | | Sex | | Women | 9 | 47216 | 0.92 (0.84-1.01) | 0.33 | | Both or men | 6 | 3910 | 0.71 (0.42-1.20) | | Type of residence | | Community | 12 | 47112 | 0.96 (0.91-1.01) | 0.10 | | Institution | 3 | 4014 | 0.85 (0.74-0.97) | | Previous fracture | | Yes | 2 | 2708 | 0.94 (0.78-1.14) | 0.68 | | No | 13 | 48418 | 0.90 (0.81-1.00) | | Previous fall | | Not conducted. No trial studied only persons with a previous fall. | | Osteoporosis diagnosis | | Yes | 3 | 2906 | 0.93 (0.76-1.15) | 0.79 | | No | 12 | 48220 | 0.90 (0.82-1.00) | | Mean baseline vitamin D levels (nmol/L) | | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Baseline vitamin D levels (nmol/L) | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Type of vitamin D | | Not conducted. No trial studied vitamin D2. | | Frequency of administration of vitamin D | | Daily to monthly | 14 | 41014 | 0.92 (0.84-1.00) | 0.44 | ### Page 83 | Intermittent orhigh doses | 1 | 112 | 0.59 (0.19-1.81) | | - | - | - | - | | Mean chronic daily dose of vitamin D excluding Intermittent or high doses (IU/d) | | < 1000 | 13 | 50853 | 0.92 (0.85-1.00) | 0.28 | | 1000-2000 | 1 | 161 | 0.37 (0.07-1.97) | | > 2000 | 0 | 0 | - | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported baseline calcium intake. | | Trial size (participants) | | ≥ 1000 | 5 | 48699 | 0.93 (0.85-1.01) | 0.33 | | < 1000 | 10 | 2427 | 0.75 (0.49-1.14) | | Mean follow-up (years) | | ≤ 1 | 3 | 471 | 0.48 (0.20-1.13) | 0.14 | | > 1 | 12 | 50655 | 0.92 (0.85-1.00) | * Test for subgroup differences. P < 0.05 : statistically significant. B. Risk of hip fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 4 | 43236 | 0.87 (0.72-1.06) | 0.75 | | High risk | 5 | 6673 | 0.83 (0.63-1.08) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. C. Risk of non-vertebral fracture | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 3 | 3763 | 0.70 (0.40-1.22) | 0.42 | | High risk | 4 | 6561 | 0.88 (0.79-0.99) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. D. Risk of new vertebral fracture ### Page 84 | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not athigh-risk | 2 | 39477 | 0.70 (0.40-1.22) | 0.39 | | High risk | 2 | 2708 | 0.34 (0.04-3.20) | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. E. Risk of falling | Subgroup | No of studies | No of participants | RR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not at high-risk | 6 | 7689 | 0.92 (0.84-1.00) | 0.56 | | High risk | 4 | 3379 | 0.84 (0.55-1.31) | | Age (years) | | ≥ 65 | 8 | 10731 | 0.91 (0.82-1.02) | 0.53 | | < 65 | 2 | 337 | 1.02 (0.12-8.83) | | Age (years) | | ≥ 80 | 3 | 741 | 0.77 (0.29-2.06) | 0.48 | | < 80 | 7 | 10327 | 0.91 (0.84-0.98) | | Sex | | Women | 6 | 7541 | 0.93 (0.79-1.10) | 0.31 | | Both or men | 4 | 3527 | 0.87 (0.79-0.95) | | Type of residence | | Community | 8 | 10439 | 0.89 (0.81-0.99) | 0.09 | | Institution | 2 | 629 | 1.02 (0.44-2.39) | | Previous fracture | | Yes | 1 | 2638 | 0.84 (0.69-1.02) | 0.35 | | No | 9 | 8430 | 0.93 (0.84-1.02) | | Previous fall | | Not conducted. No trial studied only persons with a previous fall. | | Osteoporosis diagnosis | | Yes | 1 | 2638 | 0.84 (0.69-1.02) | 0.35 | | No | 9 | 8430 | 0.93 (0.84-1.02) | | Mean baseline vitamin D levels (nmol/L) | ### Page 85 | < 25 vs ≥ 25. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | - | | Mean baseline vitamin D levels (nmol/L) | | < 50 vs ≥ 50. Not conducted. Less than 10 trials reported baseline vitamin D levels. | | Type of vitamin D | | Not conducted. No trial studied vitamin D2. | | Frequency of administration of vitamin D | | Daily to monthly | 9 | 10956 | 0.93 (0.86-1.00) | 0.06 | | Intermittent orhigh doses | 1 | 112 | 0.49 (0.25-0.95) | | Mean chronic daily dose of vitamin D excluding intermittent or high doses (IU/d) | | Not conducted. All trials studied vitamin D < 1000 IU/d. | | Mean baseline dietary calcium intake (mg) | | Not conducted. Less than 10 trials reported baseline calcium intake. | | Trial size (participants) | | ≥ 1000 | 3 | 9147 | 0.90 (0.70-1.15) | 0.73 | | < 1000 | 7 | 1921 | 0.93 (0.79-1.09) | | Mean follow-up (years) | | ≤ 1 | 2 | 158 | 0.55 (0.09-3.51) | 0.0005** | | > 1 | 8 | 10910 | 0.93 (0.86-1.00) | * Test for subgroup differences. P < 0.05 : statistically significant. ** We found that the risk of falling was significantly lower with vitamin D combined with calcium in the subgroup of trials with a mean follow-up of less than 1 year (RR 0.55, 95% CI 0.09-3.51) compared to those with longer follow-up (RR 0.93, 95% CI 0.86-1.00) (p for interaction = 0.0005). However, this finding is likely to be spurious, given the wide confidence interval in the shorter follow-up subgroup, indicating high uncertainty. F. Total number of falls | Subgroup | No of studies | No of participants | IRR (95% CI) | P value* | | - | - | - | - | - | | Risk of fractures and falls | | Not conducted, as all trials studied not at high-risk populations. | | Other subgroup analyses | | Not conducted, as this analysis included less than 10 trials. | * Test for subgroup differences. P < 0.05 : statistically significant. ### Page 86 # Table S16. Sensitivity analyses for vitamin D combined with calcium | Analysis | No of participants(studies) | RR or IRR(95% CI) | Comparison to primary analysis | | - | - | - | - | | Fixed-effect model | | Risk of any fracture | 51126 (15) | RR 0.94 (0.90-0.99) | RR 0.91 (0.84-0.99) | | Risk of hip fracture | 49909 (9) | RR 0.84 (0.74-0.96) | RR 0.84 (0.74-0.96) | | Risk of non-vertebral fracture | 10324 (7) | RR 0.87 (0.78-0.96) | RR 0.87 (0.78-0.96) | | Risk of vertebral fracture | 42185 (4) | RR 0.90 (0.74-1.09) | RR 0.90 (0.74-1.09) | | Risk of falling | 11068 (10) | RR 0.92 (0.84-0.99) | RR 0.92 (0.84-1.00) | | Total number of falls | 7532 (4) | IRR 0.95 (0.91-1.00) | IRR 0.95 (0.88-1.03) | | Excluding trials at high risk of bias | | Risk of any fracture | 43171 (5) | RR 0.94 (0.86-1.02) | RR 0.91 (0.84-0.99) | | Risk of hip fracture | 42773 (4) | RR 0.84 (0.73-0.96) | RR 0.84 (0.74-0.96) | | Risk of non-vertebral fracture | 6491 (3) | RR 0.88 (0.79-0.99) | RR 0.87 (0.78-0.96) | | Risk of vertebral fracture | 38920 (2) | RR 0.91 (0.75-1.11) | RR 0.90 (0.74-1.09) | | Risk of falling | 3619 (3) | RR 0.94 (0.70-1.26) | RR 0.92 (0.84-1.00) | | Total number of falls | 398 (1) | IRR 0.79 (0.61-1.03) | IRR 0.95 (0.88-1.03) | | Excluding trials not reporting at least 1000 patient-years of follow-up | | Risk of any fracture | 49727 (7) | RR 0.92 (0.85-1.00) | RR 0.91 (0.84-0.99) | | Risk of hip fracture | 49727 (7) | RR 0.84 (0.74-0.96) | RR 0.84 (0.74-0.96) | | Risk of non-vertebral fracture | 10131 (5) | RR 0.87 (0.78-0.96) | RR 0.87 (0.78-0.96) | | Risk of vertebral fracture | 42115 (3) | RR 0.90 (0.74-1.09) | RR 0.90 (0.74-1.09) | | Risk of falling | 10175 (5) | RR 0.93 (0.82-1.05) | RR 0.92 (0.84-1.00) | | Total number of falls | 7134 (3) | IRR 0.97 (0.89-1.05) | IRR 0.95 (0.88-1.03) | | Excluding trials that allowed nontrial supplements of calcium and/or vitamin D | | Risk of any fracture | 445 (1) | RR 0.44 (0.22-0.86) | RR 0.91 (0.84-0.99) | | Risk of hip fracture | 445 (1) | RR 0.34 (0.01-8.40) | RR 0.84 (0.74-0.96) | | Risk of non-vertebral fracture | 445 (1) | RR 0.44 (0.22-0.86) | RR 0.87 (0.78-0.96) | | Risk of vertebral fracture | Not conducted. All trials allowed (or did not provide clear instructions toavoid) nontrial supplements of calcium and/or vitamin D. | | Risk of falling | 445 (1) | RR 0.89 (0.74-1.07) | RR 0.92 (0.84-1.00) | | Total number of falls | 445 (1) | IRR 1.12 (0.94-1.33) | IRR 0.95 (0.88-1.03) | | Leave-one-out method | | Risk of any fracture | One study had a disproportionate influence on this finding. Using a | ### Page 87 | leave-one-out approach, exclusion of this trial substantially widened theconfidence intervals for total fractures : RR 0.91 (95% CI 0.84 to 0.99)changed to 0.96 (95% CI 0.91 to 1.01). | | - | | Risk of hip fracture | One study had a disproportionate influence on this finding. Using aleave-one-out approach, exclusion of this trial substantially widened theconfidence intervals for total fractures : RR 0.84 (95% CI 0.74 to 0.96)changed to 0.88 (95% CI 0.75 to 1.05). | | Risk of non-vertebral fracture | One study had a disproportionate influence on this finding. Using aleave-one-out approach, exclusion of this trial substantially widened theconfidence intervals for total fractures : RR 0.87 (95% CI 0.78 to 0.96)changed to 0.90 (95% CI 0.78 to 1.05). | | Risk of vertebral fracture | Results remained similar. | | Risk of falling | Results remained similar. | | Total number of falls | Results remained similar. | ### Page 88 # # # # # # # # # # # # # # # # # # # ### Page 89 # Table S17. Post-hoc sensitivity analyses | Analysis | No of participants(studies) | RR or IRR(95% CI) | Comparison to primary analysis | | - | - | - | - | | Calcium monotherapy, excluding studies in which osteoporotic or non-vertebral fractures were counted astotal fractures, or where a specific subtype of falls (e.g., injurious) was used as the falls outcome. | | Risk of any fracture | 7666 (8) | RR 0.90 (0.80-1.02) | RR 0.91 (0.81-1.01) | | Risk of falling | Not conducted. No trial used a specific subtype of falls as the falls outcome. | | Total number of falls | | Vitamin D monotherapy, excluding studies in which osteoporotic or non-vertebral fractures were counted astotal fractures, or where a specific subtype of falls (e.g., injurious) was used as the falls outcome. | | Risk of any fracture | 72337 (25) | RR 0.98 (0.93-1.04) | RR 1.00 (0.95-1.06) | | Risk of falling | 39363 (31) | RR 1.00 (0.97-1.04) | RR 1.01 (0.98-1.04) | | Total number of falls | 22927 (11) | IRR 1.02 (0.95-1.09) | IRR 1.00 (0.95-1.06) | | Vitamin D combined with calcium, excluding studies in which osteoporotic or non-vertebral fractures werecounted as total fractures, or where a specific subtype of falls (e.g., injurious) was used as the falls outcome. | | Risk of any fracture | 47363 (11) | RR 0.93 (0.86-1.01) | RR 0.91 (0.84-0.99) | | Risk of falling | Not conducted. No trial used a specific subtype of falls as the falls outcome. | | Total number of falls |