Study at a glance
How was vitamin D tested for movement and falls in older adults?
Two trials compared vitamin D with a placebo, a treatment without vitamin D, in older adults. One trial, called D-Vitaal, tested daily supplements for a year in 155 people with depressive symptoms, difficulties with everyday activities and relatively low vitamin D levels. The other, called D-Health, tested monthly supplements for up to five years in a much larger group whose vitamin D levels were generally adequate.
Did supplements improve movement or reduce falls overall?
D-Vitaal raised blood vitamin D but did not improve overall walking, chair-standing, balance, grip strength or reported limits on everyday activities compared with placebo. D-Health did not show an overall reduction in falls. These findings separate a rise in blood vitamin D from a demonstrated improvement in movement or fall prevention.
Were there benefits or possible harms in smaller groups?
A small D-Vitaal subgroup reported fewer limits on everyday activities, but physical performance did not clearly improve and the finding could be due to chance. In D-Health, participants with a body mass index below 25 had more falls with vitamin D in both survey and diary analyses. This possible harm signal matters, but subgroup results need caution because many comparisons were made.
Do these findings settle how severe deficiency should be treated?
D-Vitaal included few people with very low vitamin D levels. D-Health did not measure everyone’s starting vitamin D and mainly studied people with adequate levels. Its annual surveys covered falls in the preceding month, rather than every fall over five years. Neither trial settles treatment of severe deficiency or identifies a personal dose. The results do not support assuming that more vitamin D improves physical performance or prevents falls.
Read the full article below ↓Two placebo-controlled trials separate a rise in blood vitamin D from a measurable improvement in movement or fewer falls. D-Vitaal found no overall physical-performance benefit in older adults selected for symptoms, functional difficulty and relatively low vitamin D status. D-Health found no overall fall-prevention benefit in a much larger, generally vitamin D-replete older population. For the companion appraisal, see Vitamin D, Physical Performance and Falls: Study Summary.
The question behind low vitamin D and poor performance
A low vitamin D concentration and impaired movement can occur together without proving which caused the other. Less outdoor activity, illness and other characteristics may affect both. A randomised supplementation trial tests whether changing intake improves an outcome under its particular conditions. It does not experimentally create deficiency, and a null treatment result does not establish that untreated severe deficiency is harmless.
The physiology described in Vitamin D and Skeletal Muscle Function makes a muscle effect plausible. Clinical consequences still require direct measurement. Strength, walking, balance and falling are related but distinct outcomes: a fall also depends on vision, medication, cognition and environmental hazards. Reading them as interchangeable would obscure what either trial actually established.
D-Vitaal selected participants with symptoms and relatively low status
The Dutch D-Vitaal trial randomised 155 community-dwelling adults aged sixty to eighty years. Participants had clinically relevant depressive symptoms, at least one reported functional limitation, and screening 25-hydroxyvitamin D between 15 and 50 nmol/L in winter or between 15 and 70 nmol/L in summer. The seasonal upper limits matter: not every participant was below 50 nmol/L at entry. Baseline medians were approximately 46 and 44 nmol/L in the intervention and placebo groups.
Participants received 1,200 IU/day vitamin D3 or identical placebo for twelve months. Both groups could continue up to 400 IU/day from other supplements. Calcium intake was supported through dietary advice, with calcium tablets for participants reporting low dairy intake. The study therefore tests an assigned regimen within that background care, rather than isolated vitamin D exposure without any accompanying nutritional advice. These quantities describe research methods, not an individual prescription.
Performance improved neither consistently nor selectively in low-status participants
The primary physical outcomes were reported functional limitations and a modified Short Physical Performance Battery combining walking, repeated chair stands and balance. Grip strength and timed up-and-go were secondary measurements. At six months, mean 25-hydroxyvitamin D was 85 nmol/L with supplementation and 43 nmol/L with placebo. The clear biochemical separation did not produce an overall advantage in physical performance, functional limitations, grip strength or timed up-and-go.
For the adjusted performance analysis, the estimated between-group difference was −0.20 points, with a 95% confidence interval from −0.79 to 0.38. Grip strength likewise showed no established benefit. These estimates are more informative than saying only that a significance threshold was missed: they show the direction and uncertainty of the observed contrast. Vitamin D Status vs Vitamin D Effect provides the relevant distinction between changing a blood marker and changing function.
An exploratory subgroup with starting concentrations at least 50 nmol/L reported fewer functional limitations with supplementation. The difference was small, did not establish a corresponding performance benefit and was absent in the lower-status subgroup. The authors regarded it as likely a chance finding. The trial was not powered for those subgroup comparisons. The result must remain visible, but it cannot justify selecting people for treatment by that threshold.
What D-Vitaal leaves uncertain
The physical-performance analysis included 151 of the 155 randomised participants because four lacked adequate follow-up. Dropout was relatively low and adherence was high, supporting the interpretation that the main null finding was not simply a failure to change exposure. Nevertheless, only about sixteen percent entered below 30 nmol/L. Small subgroup numbers limit conclusions about more marked deficiency, and the eligibility criteria excluded concentrations below 15 nmol/L.
Depressive symptoms and functional difficulty defined this population. Results cannot automatically be extended to every older adult, to institutional care or to a patient with a diagnosed deficiency-related myopathy. Adverse-event totals did not differ significantly between groups; one participant in the intervention group died. The few reported falls in the adverse-event table were not a sufficiently powered falls-prevention experiment. The separate D-Health trial addresses falls directly.
D-Health tested a different population and dosing schedule
D-Health randomised 21,315 Australians aged sixty to eighty-four to monthly 60,000 IU vitamin D3 or placebo for up to five years. Five withdrew consent and their data were removed. Falls were a prespecified tertiary outcome of a trial primarily designed around mortality. To preserve blinding for other analyses, the investigators used randomly selected subsets rather than presenting every participant as part of this falls analysis.
The annual-survey analysis included 15,416 participants with at least one response from a sample of 16,000. A separate three-month diary analysis included 2,200 respondents, with 2,093 in the fully adjusted comparison. The main falls endpoint asked whether a participant had fallen in the month before each annual survey. This provides repeated snapshots, not continuous observation of every fall throughout five years. The diaries offered more detailed prospective recording over a much shorter interval.
No overall reduction in falling
The odds ratio for a reported fall in the preceding month was 1.02, with a 95% confidence interval of 0.95 to 1.10. The fully adjusted diary comparison was also compatible with no effect: odds ratio 1.07, confidence interval 0.84 to 1.36. Diary fall incidence showed no clear reduction either. These are different analyses with different precision, but neither demonstrated the proposed overall prevention benefit.
An odds ratio is not the percentage of participants who fell, and it is not automatically an identical relative risk. The confidence interval also does not prove exact equivalence. Here the large survey analysis constrains the size of plausible overall effects more tightly than the smaller diary analysis. The trial concerns Vitamin D in Older Adults, but it did not directly measure a microscopic contraction mechanism or explain every contributor to an individual fall.
A safety signal in the lower-BMI subgroup
In the prespecified subgroup with body mass index below 25 kg/m², supplementation was associated with more falls in both survey and diary analyses. The survey odds ratio was 1.25, with a 95% confidence interval of 1.09 to 1.43; the diary estimate was 1.72, with an interval of 1.09 to 2.73. The corresponding interaction tests supported a difference between BMI strata. Participants with higher BMI did not show the same pattern.
This signal deserves attention alongside the overall null result. Multiple comparisons were not adjusted, and subgroup interpretation remains less secure than the principal analysis. Exploratory exclusions of underweight participants did not remove the pattern, but they do not establish its mechanism. Hypercalcaemia and kidney-stone incidence did not differ materially between groups. Vitamin D and Calcium therefore supplies physiological context without allowing the falls signal to be relabelled proven calcium toxicity.
Measured and predicted vitamin D status are different evidence
D-Health did not measure baseline vitamin D in every participant. It predicted lower starting status using a model developed from placebo-group samples. The model had only moderate discrimination and a low positive predictive value for concentrations below 50 nmol/L. Survey subgroup estimates suggested a more favourable effect at predicted lower status, but the diary analyses did not confirm that pattern. Prediction error and inconsistent outcomes prevent treating it as a demonstrated benefit in deficient people.
During follow-up, sampled mean concentrations were approximately 115 nmol/L in the vitamin D group and 78 nmol/L in placebo recipients. Only thirteen percent of placebo samples were below 50 nmol/L. The trial consequently offers strong information about the tested monthly regimen in a largely replete population, while leaving falls in populations with prevalent severe deficiency unresolved. It should not be described as a direct replication of D-Vitaal’s daily regimen and symptom-selected participants.
Measurement and follow-up limit the inference
D-Health added an explicit definition of falling from the third annual survey. Reports increased in both groups at that point, showing why a change in the questionnaire should not be mistaken for a sudden treatment effect. Excluding participants without any outcome data also makes the analysis a modified intention-to-treat comparison. The diary group was selected later in follow-up among people who remained enrolled; high adherence in that subset does not establish identical experience among everyone originally randomised.
The D-Vitaal measurements illustrate a different distinction. Reported difficulty with daily activities is meaningful to participants, but it does not measure exactly the same thing as timed walking, chair stands, balance or dynamometer grip strength. A favourable questionnaire subgroup without a corresponding performance effect cannot establish stronger muscle. Conversely, a small physical test battery cannot exclude every possible physiological response. The correct inference remains tied to the measured outcomes and their precision, rather than an unrestricted claim of either benefit or biological inactivity.
How the trials fit muscle physiology
The results do not remove the importance of Vitamin D Receptors in experimental muscle biology. Receptor deletion, low circulating precursor and additional oral supplementation are different perturbations. An intact pathway may contribute to normal tissue function without a higher intake producing better performance. Vitamin D Activation adds another distinction: an administered dose, the circulating precursor and local active-hormone signalling are not the same measurement.
For Vitamin D and Muscle Function, D-Vitaal provides a direct performance test in a relatively low-status symptomatic group, while D-Health provides a much larger falls test in a different population. Neither study traces a causal sequence from vitamin D insufficiency through a particular muscle defect to a fall. Their combined contribution is to constrain treatment claims and expose the remaining population-specific uncertainty, rather than to turn an association into a universal cause or cure.
The D-Health supplementary year-by-year analysis included one favourable second-year survey estimate, with an odds ratio of 0.84 and a 95% confidence interval of 0.71 to 0.99. Other yearly estimates and the combined analysis did not establish an overall reduction. This isolated time-point result should remain visible without replacing the overall interpretation.
Conclusion
These trials raised blood vitamin D without demonstrating overall improvements in the physical outcomes tested. D-Vitaal’s small favourable subgroup result and D-Health’s possible harm signal should be reported with their limitations. The findings argue against assuming that higher vitamin D exposure improves movement or prevents falls. They do not settle treatment of profound deficiency, identify a universal safe dose or supply a personal treatment schedule.
Sources and publication details
Vitamin D supplementation for the prevention of depression and poor physical function in older persons: the D-Vitaal study, a randomized clinical trial. de Koning EJ and colleagues. American Journal of Clinical Nutrition. 2019;110(5):1119 to 1130. DOI: 10.1093/ajcn/nqz141. PMID: 31340012.
Vitamin D supplementation and risk of falling: outcomes from the randomized, placebo-controlled D-Health Trial. Waterhouse M and colleagues. Journal of Cachexia, Sarcopenia and Muscle. 2021;12(6):1428 to 1439. DOI: 10.1002/jcsm.12759. PMID: 34337905.
Related reading
A Systematic Review of Complications Following Elective Foot and Ankle Surgery
This review aimed to synthesize complication profiles, risk factors, and mitigation strategies across elective procedures.
Original paper ↗Read more →Accidental Falls in Patients with Hyperkinetic Movement Disorders: A Systematic Review
Most of the common diseases characterized by HKMDs were investigated for both fall causes and consequences, but prevention studies were limited to spinocerebellar ataxia and Huntington’s disease.
Original paper ↗Read more →Assessment of gait and posture characteristics using a smartphone wearable system for persons with osteoporosis with and without falls
Given the close association between serum calcium, albumin, PTH, Vitamin D, and musculoskeletal health, we compared these markers in individuals with history of falls as compared to nonfallers.
Original paper ↗Read more →Association Between Physical Performance and Cognitive Function in Older Adults Across Multiple Studies: A Pooled Analysis Study
While several studies have examined the association between cognitive and physical function, the consistency of these associations across functional contexts is unclear.
Original paper ↗Read more →Association of Bone Mass, Falls, and Vertebral Fractures in Older Patients on Hemodialysis: The Role of Comprehensive Geriatric Assessment
Participants had multiple comorbidities, typical of this population, including anemia, cardiovascular disease and chronic liver disease.
Original paper ↗Read more →Beyond patient education: fall prevention knowledge, health literacy, and implementation gaps in Chinese hospitals: a patient-caregiver study
This study examined these domains and their relationships with fall outcomes.
Original paper ↗Read more →Cognitive frailty is a robust predictor of falls, injuries, and disability among community-dwelling older adults
In this study, we aimed to determine the ability of cognitive frailty in predicting the incidence of falls, injuries and disability among multi-ethnic older adults in Malaysia at 5 years follow-up.
Original paper ↗Read more →Effect of Calcifediol on Physical Performance and Muscle Strength Parameters: A Systematic Review and Meta-Analysis
The present systematic review and meta-analysis aimed to summarize the effects of calcifediol on physical performance and muscle strength.
Original paper ↗Read more →Effects of a nutritional supplement containing fish protein, vitamin D, and ω3 fatty acids, taken during high-intensity functional training, on physical performance in recreationally trained individuals: a randomized controlled trial
We examined the effects of a multi-ingredient supplement containing fish protein, vitamin D, and ω3 fatty acids, taken during HIFT, on physical performance in trained individuals.
Original paper ↗Read more →Effects of Vitamin D3 Supplementation and Resistance Training on 25-Hydroxyvitamin D Status and Functional Performance of Older Adults: A Randomized Placebo-Controlled Trial
Besides a daily dose of 400 mg of calcium, participants received either 800 IU vitamin D3 per day (VDD), 50,000 IU vitamin D3 per month (VDM) or nothing (CON).
Original paper ↗Read more →Evaluating the impact of a falls prevention community of practice in a residential aged care setting: a realist approach
Participants were 20 interdisciplinary staff nominating as CoP members between Nov 2013-Nov 2015 representing 13 facilities (approximately 780 beds) of a RAC organisation.
Original paper ↗Read more →Falls and potential therapeutic interventions among elderly and older adult patients with cancer: a systematic review
The aim of this study was to perform a systematic review for previous publications that have assessed the incidence, risk factors, and favorable procedures to prevent and manage falls among cancer survivors of elderly and older adults.
Original paper ↗Read more →“Falls in older people receiving home healthcare services at armed forces hospital Southern Region: prevalence, associated factors, and consequent injuries”
A total of 368 participants were selected using systematic random sampling.
Original paper ↗Read more →Gender-Specific Patterns of Injury in Older Adults After a Fall from a Four-Wheeled Walker (Rollator): Retrospective Study from a Swiss Level 1 Trauma Centre
Aim: As the population is aging, falls by older people, in particular falls from four-wheeled walkers (“rollators”), are a growing problem.
Original paper ↗Read more →Implementing STEADI for routine falls prevention of all older adults attending outpatient physical therapy: key partner perspectives
Participants included physical therapy providers (n = 16), physicians (n = 5), administrative staff (n = 3), patients (n = 10), and care partners (n = 10).
Original paper ↗Read more →Long-term adherence and effects on grip strength and upper leg performance of prescribed supplemental vitamin D in pregnant and recently pregnant women of Somali and Swedish birth with 25-hydroxyvitamin D deficiency: a before-and-after treatment study
Muscular weakness and severe vitamin D deficiency is prevalent in Somali (veiled) pregnant women, Sweden.
Original paper ↗Read more →Multidisciplinary care pathways for falls prevention in older adults: visualizing the needs of primary care-based health care professionals
The aim of this study was to gain insight into and visualize the needs of primary care-based health care professionals with regards to multidisciplinary falls prevention care using journey mapping.
Original paper ↗Read more →Phase angle by electrical bioimpedance is a predictive factor of hospitalisation, falls and mortality in patients with cirrhosis
The aim was to analyse the prognostic value of the phase angle to predict clinically relevant events such as hospitalisation, falls, and mortality in patients with cirrhosis.
Original paper ↗Read more →Prevalence and Associated Factors of Falls among Older Adults between Urban and Rural Areas of Shantou City, China
Methods: We used cluster random sampling to investigate the sociodemographic information, living habits, medical status, falls, home environment, and balance ability among 649 older adult participants.
Original paper ↗Read more →Relating consequential falls to individual and combinations of prescription medication use in Swedish older adults
Medication prescription and prospectively registered falls over 2.5 years were extracted from 441 participants within a Swedish National Quality Register.
Original paper ↗Read more →Serum 25-hydroxyvitamin D and health-related quality of life in patients with papillary thyroid carcinoma: a prospective cohort study
Associations between serum 25-hydroxyvitamin D and HRQoL domains were examined using rank-transformed general linear models, spearman correlation, multivariable linear regression and binary logistic regression.
Original paper ↗Read more →Original sources
Vitamin D supplementation for the prevention of depression and poor physical function in older persons: the D-Vitaal study, a randomized clinical trial
Published in de Koning EJ and colleagues. American Journal of Clinical Nutrition (2019).
Publication details
Vitamin D supplementation for the prevention of depression and poor physical function in older persons: the D-Vitaal study, a randomized clinical trial. de Koning EJ and colleagues. American Journal of Clinical Nutrition. 2019;110(5):1119 to 1130. DOI: 10.1093/ajcn/nqz141. PMID: 31340012.
Vitamin D supplementation and risk of falling: outcomes from the randomized, placebo-controlled D-Health Trial
Published in Waterhouse M and colleagues. Journal of Cachexia, Sarcopenia and Muscle (2021).
Publication details
Vitamin D supplementation and risk of falling: outcomes from the randomized, placebo-controlled D-Health Trial. Waterhouse M and colleagues. Journal of Cachexia, Sarcopenia and Muscle. 2021;12(6):1428 to 1439. DOI: 10.1002/jcsm.12759. PMID: 34337905.