A Randomized Controlled Trial of the Effects of Vitamin D on Muscle Strength and Mobility in Older Women with Vitamin D Insufficiency

Kun Zhu, PhD; Nicole Austin, PhD; Amanda Devine, PhD; David Bruce, MD; Richard L. Prince, MD


J Am Geriatr Soc. 2010;58(11):2063-2068. 

In This Article

Materials and Methods


Three hundred two women aged 70 to 90 were recruited between April 2003 and October 2004 in Perth, Australia (latitude 32°S). The recruitment procedure has been reported elsewhere;[5] in brief, participants were recruited from the following three sources: the emergency departments of teaching hospitals, the local community home nursing service (Silver Chain), and the Electoral Roll. The inclusion criteria were a plasma 25(OH)D concentration less than 24 ng/mL and a history of at least one fall in the previous 12 months. Exclusion criteria were current consumption of vitamin D or bone or mineral active agents apart form calcium, a bone mineral density (BMD) Z-score at the total hip site of less than −2.0, medical conditions or disorders that influence bone mineral metabolism; a fracture in the past 6 months, a Mini-Mental State Examination score less than 24 or the presence of significant neurological conditions likely to substantially impair balance or physical activity such as stroke, and Parkinson's disease.

This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and the Human Research Ethics Committee of the Sir Charles Gairdner Hospital approved all procedures involving human subjects. Written informed consent was obtained from each participant. The study was registered with the Australian Clinical Trials Registry (registration number ACTRN012606000331538).


Participants received 1,000 mg/d of calcium as calcium citrate (Citracal, Mission Pharmacal, Key Pharmaceutical Pty Ltd, Rhodes, Australia) for 1 year as two calcium tablets in the morning with breakfast and two calcium tablets with the evening meal. They were randomized to receive 1,000 IU ergocalciferol (vitamin D2) per day or identical placebo (Ostelin, Boots Healthcare, North Ryde, Australia) consumed with the evening meal for 1 year.

An independent research scientist who labelled the bottles and dispensed the study medications to subjects generated the randomization schedule to vitamin D or placebo, which was kept in the Pharmacy Department of the Sir Charles Gairdner Hospital. The randomization procedure used a random number generator with a block size of 10 to assign participants to vitamin D or placebo in a ratio of 1:1. The study subjects and study staff remained blinded to the treatment code until all data had been entered and evaluated for accuracy and the a priori hypotheses reviewed. Adherence to the study medications was established by counting tablets returned at the clinic visits at 6 and 12 months.

Muscle Strength and Mobility

At baseline and 12 months, ankle dorsiflexion, knee flexor, knee extensor, hip abductor, hip flexor, hip extensor, and hip adductor strength were assessed using a strain gauge. The subjects were requested to exert a maximal muscle contraction against the strain gauge after one practice. The best of three attempts was recorded for each muscle group. The coefficient of variation (CV) error was between 14% and 20% for the different muscle groups.

Mobility functioning was measured using the Timed Up and Go Test (TUAG), which timed subjects while getting up, walking 3 m, turning, returning to chair and sitting down again.[21] The CV error was 7%.

Biochemistry Analysis

At baseline and 12 months, venous blood was collected after an overnight fast, and serum 25(OH)D concentrations were assessed using radioimmunoassay (DiaSorin, Stillwater, MN).

Other Assessments

At screening, demographic information including smoking history, use of community services, medications, patient recall of prevalent morbidity, and socioeconomic status was obtained. Calcium intake was assessed using a food frequency questionnaire developed in a previous study.[22] This questionnaire includes 39 food items and uses the Australian Tables of Food Composition—NUTTAB 90 database, a nutritional database that uses chemical analysis of Australian foods. Activity levels were calculated in kcal/d using a validated method using body weight, number of hours and type of physical activity, and energy costs of such activities.[23,24] Weight and height were measured at baseline and 12 months with light cloths and without shoes.

Sample Size Calculation and Statistical Analysis

Power calculations were performed before commencement of the study. As the primary purpose of this trial was to study the effects of vitamin D supplementation on risk of falling, the sample size was calculated in that 113 subjects were needed in each group to detect a relative risk reduction of 0.37 in a population with a 1-year fall risk of 0.6. Allowing for 30% dropout, the sample size was determined to be 150 per group. At this sample size, a 10% difference in muscle strength or TUAG could be detected at 80% power and 5% level of significance.

Descriptive statistics are reported as means±standard deviations and differences as means±standard errors of the mean for all variables, unless otherwise stated. The normality of continuous variables was checked through the construction of histograms. One variable that was not normally distributed (TUAG) was log-transformed. Baseline values between the two groups were compared using the Student t-test or Mann-Whitney test when appropriate. A linear regression model was used to test whether baseline values (baseline 25(OH)D or baseline muscle strength and mobility measurements) were effect modifiers for the effects of ergocalciferol supplementation on muscle strength and mobility, with changes from baseline to 12 months in outcome measures as the dependent variable and treatment group, baseline values, and the interaction terms of baseline value and treatment group as the independent variables. If there was a significant interaction, subgroup analyses were conducted according to tertile of baseline value. The normality and independence of the residuals and the homogeneity of variance of each model were checked using residual plots (normal probability plot and plot of residuals vs predicted values). All tests were two tailed, and significance was set at P<.05. The data analyses were performed with SPSSPC for Windows (SPSS, version 15, Chicago, IL).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.