Effects of Supplemental Vitamin D₃, Omega-3 Fatty Acids on Physical Performance Measures in the VITamin D and OmegA-3 TriaL

Abstract

Context

Declining muscle strength and performance in older adults are associated with falls, fractures, and premature death.

Objective

This work aimed to determine whether supplementation with vitamin D₃ or omega-3 fatty acids vs placebo for 2 years improves physical performance measures.

Methods

VITamin D and OmegA-3 TriaL (VITAL) was a double-blinded, placebo-controlled randomized trial of supplemental vitamin D₃ and/or omega-3 fatty acids vs placebo in the prevention of cancer and cardiovascular disease in 25 871 US adults. This ancillary study was completed in a New England subcohort that had in-person evaluations at baseline and 2-year follow-up. This study was conducted with 1054 participants (age: men ≥50 and women ≥55 years) at the Center for Clinical Investigations in Boston. Interventions included a 2 × 2 factorial design of supplemental vitamin D₃ (cholecalciferol, 2000 IU/day) and/or marine omega-3 fatty acids (1 g/day). Main outcome measures included 2-year changes in physical performance measures of grip strength, walking speed, standing balance, repeated chair stands, and Timed-up and Go (TUG).

Results

At 2 years, all randomized groups showed worsening walking speeds and TUG. There were no differences in changes in grip strength, walking speeds, Short Physical Performance Battery (composite of walking speed, balance, and chair stands), and TUG between the vitamin D₃-treated and the placebo-treated groups and between the omega-3-treated and the placebo-treated groups. Effects overall did not vary by sex, age, body mass index, or baseline measures of total or free 25-hydroxyvitamin D (25[OH]D) or plasma omega-3 index; TUG slightly worsened with vitamin D supplementation, compared to placebo, in participants with baseline total 25(OH)D levels above the median (P = .01; P for interaction = .04).

Conclusion

Neither supplemental vitamin D₃ nor marine omega-3 fatty acids for 2 years improved physical performance in this generally healthy adult population.

vitamin D, omega-3 fatty acids, physical performance

Muscle mass, strength, and physical function decline with advancing age and are strongly associated with falls, fractures, disability, hospitalization, and premature death (1, 2). While lean mass contributes about 50% of body weight in young adults, it makes up only 25% in adults aged 75 to 80 years (3). Additionally, fast type II muscle fibers are converted to slow type I muscle fibers with age, causing loss of muscle power, impaired mobility, and disability (4).

Skeletal muscle expression of receptors for 1,25-dihydroxyvitamin D (VDR) decrease with age (5), and vitamin D deficiency has been associated with loss of type II muscle fibers in older adults (6). In a randomized study, vitamin D supplementation has been shown to increase intramyonuclear VDR concentration and muscle fiber size in older, mobility-limited women with low vitamin D status (6). Cross-sectional and longitudinal observational studies suggest an association between low serum 25(OH)D levels and reduced muscle mass, strength, and performance, including walking speed (7, 8). These observational studies can be confounded by low sun exposure, reduced dietary intake, and decreased physical activity among older adults. Several randomized controlled trials (RCTs) have evaluated the effect of vitamin D supplementation on physical performance measures, but results have been inconsistent (9-16).

Omega-3 fatty acids, both marine (eicosapentaenoic acid [EPA] and docosahexaenoic acid [DHA]) and plant-based (alpha-linolenic acid [ALA]), may also potentially improve muscle function and strength via anti-inflammatory effects, increased muscle protein synthesis, and reduced muscle protein breakdown (17, 18). Data on the effects of supplemental omega-3 fatty acids on muscle mass and physical performance are sparse. A cross-sectional study from the National Health and Nutrition Examination Survey (NHANES) 1999 to 2002 in 2141 older adults aged 50 to 85 years found that total omega-3 intake was positively associated with knee extensor strength in older men but not in older women (19). Another cohort study in Iceland found that although higher polyunsaturated fatty acid intake was positively associated with increased thigh muscle size and greater knee strength, total omega-3 intake was not associated with muscle size or strength (20). In addition, no associations between intakes of polyunsaturated fatty acids or omega-3 fatty acids and changes in muscle size or strength were found over 5 years of follow-up. A meta-analysis of 10 RCTs (7 with marine omega-3 supplements, 1 with ALA from flax oil, and 2 with dietary pattern of omega-6 to omega-3 ratio <2) found that omega-3 supplementation did have a small-to-moderate positive effect on muscle mass in 202 adults aged 60 years or older and slightly improved performance on the Timed-Up and Go (TUG) test in a total of 136 adults; no differences were seen in walking speed or grip strength (21).

Given the conflicting and sparse literature, additional research from large RCTs is needed to clarify the effects of vitamin D and omega-3 supplementation on physical performance measures. The VITamin D and OmegA-3 TriaL (VITAL) is a 2 × 2 factorial, double-blinded, randomized, placebo-controlled trial that found that supplemental vitamin D₃ supplementation (2000 IU/day) or marine omega-3 fatty acid (1 g/day), compared to placebo, did not prevent cancer or cardiovascular disease overall, though supplemental omega-3 fatty acids did reduce myocardial infarction risk, in 25 871 participants (22, 23). In this ancillary study to VITAL, we tested the randomized effects of 2 years of supplementation with vitamin D₃ and/or omega-3 fatty acids, compared with placebo, on physical performance measures in a subcohort of participants at the Center for Clinical Investigations.

Materials and Methods

Trial Design

The parent VITAL investigated effects of supplemental vitamin D₃ supplementation (2000 IU/day) and/or omega-3 fatty acid (1 g/day of Omacor containing 840 mg of marine omega-3 fatty acids, including 460 mg of EPA and 380 mg of DHA) on the primary prevention of cancer and cardiovascular disease in women aged 55 years or older and men aged 50 years or older from all 50 US states. Exclusion criteria included self-reported prior history of cancer (except nonmelanoma skin cancer) or cardiovascular disease, hypercalcemia, hypoparathyroidism or hyperparathyroidism, renal failure, severe liver disease, sarcoidosis or other granulomatous disease, or other serious illness. In addition, participants completed a 3-month run-in period, during which they needed to take at least two-thirds of placebo study pills to qualify for randomization. During the run-in period and the interventional trial, personal supplementation of vitamin D was limited to less than or equal to 800 IU daily and calcium was limited to less than or equal to 1200 mg daily; no supplementation of fish oil was permitted. A total of 25 871 participants, including 5106 Black individuals, were treated for a median of 5.3 years (range, 3.8-6.1 years). The parent trial protocol has been described in detail (24, 25), and the results of the parent trial have been published (22, 23).

Participants within driving distance of Boston, Massachusetts, and able to provide informed consent were recruited for in-person clinical visits, including physical performance measures, anthropometric assessments, and blood draws, at the Center for Clinical Investigations under the Harvard Catalyst Clinical and Translational Science Center (CTSC) in Boston. A total of 1054 VITAL participants completed baseline visits, and retention rate at the 2-year follow-up was 92%. The baseline visits were conducted between January 2012 and March 2014, and the 2-year follow-up visits were completed between January 2014 and July 2016. A total of 711 participants (67%) returned for 4-year follow-up visits.

Information on this study is available on the Clinical Trials.gov website (NCT01747447). Pharmavite LLC donated vitamin D and Pronova BioPharma of Norway and BASF donated fish oil study pills and matching placebos in the form of calendar packs. These ancillary studies were approved by the Partners Human Research Committees and the institutional review board of Brigham and Women's Hospital. Patients provided written informed consent for this ancillary study.

Physical Performance Measures

Physical performance was assessed at baseline and 2-year follow-up and included grip strength, TUG, walking speed, standing balance, and repeated chair stands. Grip strength, which has been found to be inversely related to risk of falls and hip fractures (26, 27), was tested by JAMAR Plus+ Digital Hand Dynamometer (Sammons Preston Rolyan). In the TUG test, which also correlates with fracture risk (28, 29), participants were timed as they stood up from a chair, walked 3 meters, turned around, and returned to sit in the chair. Walking speed was assessed over 6 meters at normal and fast paces. In addition to walking speed, standing balance and repeated chair stands were also performed as part of the Short Physical Performance Battery (SPPB, scored 0-12), which is a validated measure of lower body function and predictive of recurrent falls (26, 30). For standing balance, participants were asked to hold 3 basic standing positions (side-by-side, semi-tandem, and full tandem [or heel-to-toe]) for 10 seconds each. For the chair stands, participants stood and sat down 5 times with arms folded, using a standard straight-backed chair (31). Of note, TUG and components of the SBBP were added to the study protocol later; thus, not every CTSC participant completed all physical performance measures at baseline. A total of 731 participants had assessments of all physical performance measures, including TUG, 766 completed all components of the SBBP, and all participants performed grip strength at baseline. Measures of grip strength, TUG, walking speeds, and chair stands were performed twice at each visit, and the means were used for analyses.

Measurement of Blood Samples

Blood samples were collected at baseline in all participants. Serum total 25-hydroxyvitamin D (25[OH]D) levels were assayed using liquid chromatography–tandem mass spectrometry by Quest Diagnostics, and assays were calibrated to standards of the Centers for Disease Control and Prevention (CDC) (32). Free 25(OH)D levels were measured by a competitive enzyme-linked immunosorbent assay 2-step immunoassay by Future Diagnostics Solutions B.V., developed in collaboration with DIAsource ImmunoAssays S.A. (RRID:AB_2890998). The coefficients of variation for the VITAL samples were 4.9% and 6.8% for total and free 25(OH)D levels, respectively. Plasma omega-3 index (EPA plus DHA as a percentage of total fatty acids) was performed by liquid chromatography–tandem mass spectrometry by Quest Diagnostics. Quest Diagnostics performed the serum total 25(OH)D levels and plasma omega-3 indices at no cost to the trial. Aside from the blinded assay analyses, Quest Diagnostics and Future Diagnostics had no role in the study design, data analysis, or manuscript preparation.

Statistical Analysis

We examined changes over time in physical performance measures from baseline to year 2 post randomization in the Harvard Catalyst CTSC cohort in an intention-to-treat fashion.

To assess for balance among the randomized groups in this subcohort, baseline characteristics were compared by treatment assignment. T tests and analysis of variance (or the Wilcoxon rank sum and Kruskal-Wallis tests if nonnormal) were used to compare continuous variables, and chi-square tests were used to compare proportions across randomized groups.

The primary analysis was the effect of supplemental vitamin D₃ or omega-3 fatty acid vs placebo on 2-year changes in the measures of physical performance, adjusted for age and sex. Proc mixed model was used with all measures as outcomes, including baseline. We explored interactions between the interventions, as well as with the use of personal vitamin D supplements and baseline biomarkers of serum total and free 25(OH)D levels for vitamin D and dark-meat fish intake and plasma omega-3 index for fish oil. In addition, we evaluated effect modification by age, sex, race, and body mass index (BMI).

Statistical analyses were performed using SAS 9.4 (SAS Institute). P values less than .05 were considered statistically significant. Statistical analyses in this ancillary study were not adjusted for multiple hypotheses testing; secondary and subgroup analyses should be interpreted cautiously.

Results

Ancillary Study Participants

A total of 1054 participants were recruited for the Harvard Catalyst CTSC subcohort for physical performance measures. Two-year follow-up was completed by 968 participants (92% retention).

Table 1 shows the baseline characteristics of the CTSC subcohort. Mean (±SD) age was 64.9 ± 6.5 years and 48.9% were women. Mean 25(OH)D level was 28.1 ± 9 ng/mL, and mean plasma omega-3 index was 2.9%. At baseline, few participants had muscle weakness or low muscle performance, as defined by the European Working Group on Sarcopenia in Older People 2 (EWGSOP2) (33). A total of 5.9% had low grip strength (<27 kg in men, <16 kg in women), 2.1% had slow completion of repeated chair stands of more than 15 seconds, 1.9% had slow gait speed of 0.8 m/s or less, 1.3% had low SPPB score of 8 or less, and no participants had a TUG time longer than 20 seconds.

Table 1.

Baseline characteristics of the clinic subcohort at baseline, according to randomized assignment

	Vitamin D (n = 520)	No vitamin D (n = 534)	P	Omega-3 (n = 527)	No omega-3 (n = 527)	P
Age, mean (SD), y	64.7 (6.3)	65.1 (6.6)	.33	64.8 (6.5)	64.9 (6.4)	.88
Female sex, No. (%)	256 (49.2%)	259 (48.5%)	.81	260 (49.3%)	255 (48.4%)	.76
Race, No. (%)			.62			.39
White	429 (84.0%)	442 (84.8%)		429 (83.5%)	442 (85.3%)
Black	42 (8.2%)	46 (8.8%)		43 (8.4%)	45 (8.7%)
Other	40 (7.8%)	33 (6.3%)		42 (8.2%)	31 (6.0%)
BMI, mean (SD)	28.1 (5.3)	28.3 (5.4)	.64	28.6 (5.4)	27.7 (5.1)	.005
Current smoking, No. (%)	30 (5.8%)	27 (5.1%)	.59	26 (4.9%)	31 (6.0%)	.47
Alcohol use, No. (%)			>.99			.61
Never	107 (22.0%)	110 (21.8%)		108 (21.9%)	109 (21.8%)
Rarely to < weekly	33 (6.8%)	33 (6.5%)		37 (7.5%)	29 (5.8%)
1-6 drinks/wk	196 (40.2%)	204 (40.4%)		191 (38.7%)	209 (41.9%)
Daily	151 (31.0%)	158 (31.3%)		157 (31.8%)	152 (30.5%)
MET, h/wk, median (interquartile range)	21.6 (7.2-38.4)	20.9 (7.2-36.3)	.60	21.6 (7.8-38.9)	21.0 (6.9-36.9)	.40
History of fragility fracture (%)	49 (9.4%)	46 (8.6%)	.65	44 (8.3%)	51 (9.7%)	.45
History of fall in last year, No. (%)	105 (27.7%)	112 (29%)	.67	109 (28.2%)	108 (28.4%)	>.99
Diabetes, No. (%)	57 (11%)	51 (9.6%)	.44	51 (9.7%)	57 (10.8%)	.54
Rheumatoid arthritis, No. (%)	13 (2.5%)	14 (2.6%)	.70	12 (2.3%)	15 (2.8%)	.16
General health, No. (%)			.20			.14
Excellent	208 (42.6%)	184 (36.2%)		208 (41.7%)	184 (37.0%)
Very good	196 (40.2%)	231 (45.5%)		196 (39.3%)	231 (46.5%)
Good	76 (15.6%)	82 (16.1%)		84 (16.8%)	74 (14.9%)
Fair	8 (1.6%)	11 (2.2%)		11 (2.2%)	8 (1.6%)
Personal use of supplemental vitamin D, No. (%)	230 (44.2%)	247 (46.3%)	.51	238 (45.2%)	239 (45.4%)	.95
Baseline total 25(OH)D, ng/mL, mean (SD)	27.6 (8.8)	28.7 (9.3)	.06	28.3 (9.5)	28.0 (8.6)	.56
Baseline free 25(OH)D, pg/mL, mean (SD)	5.90 (1.85)	6.01 (1.94)	.35	6.02 (2.02)	5.88 (1.75)	.23
Dark-meat fish intake, servings/wk, median (interquartile range)	0.93 (0.47-1.00)	0.93 (0.47-1.47)	.36	0.93 (0.47-1.47)	0.93 (0.47-1.47)	.34
Plasma omega-3 index, %, mean (SD)	2.93 (0.99)	2.96 (1.00)	.60	2.97 (1.00)	2.92 (0.98)	.34

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; MET, metabolic equivalent of task.

As the CTSC subcohort had to travel and complete extensive clinical assessments, the subcohort was younger and healthier than the overall cohort of 25 871 participants, which had a mean age of 67.1 ± 7.1 years (25). The CTSC subcohort was also less racially/ethnically diverse, reflecting the geographic region, compared to the overall cohort, which was 20.2% Black.

Baseline characteristics were generally balanced between the vitamin D and non–vitamin D-treated groups and between the omega-3 and non–omega-3-treated groups (see Table 1). Although there was no difference in mean BMI between vitamin D and non–vitamin D groups, the omega-3–treated group had slightly higher BMI than the non–omega-3-treated group but the difference was small and of little clinical significance. There was no difference in number of participants taking personal vitamin D supplements at baseline (limited to ≤800 IU/day), and no difference in serum total and free 25(OH)D levels among randomized groups. There was also no difference in mean plasma omega-3 index or dark-meat fish intake at baseline.

Over 2 years, total and free 25(OH)D levels increased from mean (±SE) of 27.6 ± 0.40 ng/mL to 40.0 ± 0.41 ng/mL and 5.90 ± 0.09 pg/mL to 9.12 ± 0.10 pg/mL, respectively, in the vitamin D–treated groups, while there were no changes in total and free 25(OH)D levels in the non–vitamin D-treated groups (treatment effect P < .0001 for both). The plasma omega-3 index increased from 2.97% to 4.32% at 2-year follow-up in the omega-3-treated groups, with no differences observed in the non–omega-3-treated groups (treatment effect P < .0001).

Effects of Supplemental Vitamin D on Physical Performance Measures

At 2 years, both vitamin D–treated and placebo-treated groups had slightly slower normal walking speed (change of −0.04 m/s; P < .001, and −0.05 m/s; P < .001, respectively) and fast walking speed (change of −0.06 m/s; P < .001, and −.07 m/s; P < .001, respectively) (Table 2). Participants also took longer to complete the TUG test at 2-year follow-up (0.40 seconds longer; P < .001, in the vitamin D group and 0.30 seconds longer; P < .001, in the placebo group) (see Table 2).

Table 2.

Effect of supplemental vitamin D and omega-3 fatty acids on physical performance measures, adjusted for age and sex

		Vitamin D				Omega-3
		n	Active group	Placebo group	P, treatment effect	N	Active group	Placebo group	P, treatment effect
Grip strength, men, mean ± SE, kg	Baseline	533	40.00 ± 0.50	39.26 ± 0.49		533	40.44 ± 0.50	38.82 ± 0.49
	2-y change	498	−0.24 ± 0.34	0.19 ± 0.33	.36	498	−0.21 ± 0.34	0.17 ± 0.33	.42
	P		.48	.57			.53	.62
Grip strength, women, mean ± SE, kg	Baseline	507	24.32 ± 0.30	23.91 ± 0.30		507	23.90 ± 0.30	24.33 ± 0.30
	2-y change	456	0.12 ± 0.23	−0.19 ± 0.23	.35	456	−0.09 ± 0.23	0.02 ± 0.23	.72
	P		.61	.42			.69	.92
Normal walking speed, mean ± SE, m/s	Baseline	989	1.23 ± 0.01	1.24 ± 0.01		989	1.23 ± 0.01	1.24 ± 0.01
	2-y change	909	−0.04 ± 0.01	−0.05 ± 0.01	.53	909	−0.05 ± 0.01	−0.04 ± 0.01	.55
	P		<.001	<.001			<.001	<.001
Fast walking speed, mean ± SE, m/s	Baseline	818	1.79 ± 0.01	1.79 ± 0.01		818	1.78 ± 0.01	1.80 ± 0.01
	2-y change	747	−0.06 ± 0.01	−0.07 ± 0.01	.74	747	−0.06 ± 0.01	−0.07 ± 0.01	.24
	P		<.001	<.001			<.001	<.001
Standing balance, median ± SE, s	Baseline	823	29.31 ± 0.11	29.44 ± 0.11		823	29.55 ± 0.11	29.21 ± 0.11
	2-y change	752	0.07 ± 0.15	−0.17 ± 0.14	.24	752	−0.19 ± 0.14	0.10 ± 0.14	.16
	P		.61	.25			.18	.51
Repeated chair stands, mean ± SE, s	Baseline	794	3.59 ± 0.03	3.64 ± 0.03		794	3.60 ± 0.03	3.63 ± 0.03
	2-y change	719	0.02 ± 0.03	−0.00 ± 0.03	.46	719	0.02 ± 0.03	−0.00 ± 0.03	.55
	P		.48	.74			.54	.82
TUG, mean ± SE, s	Baseline	705	8.08 ± 0.06	8.08 ± 0.06		705	8.11 ± 0.06	8.05 ± 0.06
	2-y change	648	0.40 ± 0.06	0.30 ± 0.06	.23	648	0.33 ± 0.06	0.38 ± 0.06	.56
	P		<.001	<.001			<.001	<.001
SPPB score, median ± SE	Baseline	793	11.42 ± 0.04	11.50 ± 0.04		793	11.46 ± 0.04	11.45 ± 0.04
	2-y change	717	0.05 ± 0.04	−0.04 ± 0.04	.15	717	−0.01 ± 0.04	0.01 ± 0.04	.82
	P		.29	.33			.90	.83

Abbreviations: SPPB, Short Physical Performance Battery; TUG, Timed-Up and Go.

There were no significant effects of supplementation with vitamin D vs placebo on 2-year changes in grip strength (P = .36 for men; P = .35 for women), normal walking speed (P = .53), fast walking speed (P = .74), standing balance (P = .24), repeated chair stands (P = .46), TUG times (P = .23), or SPPB scores (P = .15) (see Table 2).

Effects also did not vary by sex, age (divided at median), and BMI (divided at the median) (data not shown). Interventional vitamin D, compared to placebo, improved standing balance in participants who also took their own personal vitamin D supplements (change of 0.15 and −0.41 seconds, respectively; P = .04), but there were no between-group differences in changes in standing balance in those who did not take personal vitamin D supplements (P = .94) and the interaction was not significant (P for interaction = .18). Compared to placebo, vitamin D supplementation worsened TUG times in participants with baseline total 25(OH)D levels greater than the median (P = .01), while there was no difference between vitamin D–treated and placebo-treated groups in participants with lower baseline total 25(OH)D levels (P for interaction = .04; Table 3). In exploratory analyses, we found no correlation between TUG times and 25(OH)D levels at baseline (data not shown). There was also a significant interaction (P = .003) between vitamin D intervention and baseline free 25(OH)D levels (divided at the median) for grip strength in women (see Table 3). Vitamin D supplementation, compared to placebo, improved grip strength in female participants with baseline free 25(OH)D above the median (P = .004), and vitamin D supplementation seemed to worsen grip strength in women with baseline free 25(OH)D levels less than the median (P = .15). Due to the number of comparisons and the unexpected direction of the association (greater improvement with supplementation among those above vs below the median), this could be a false-positive finding.

Table 3.

Changes in physical performance measures by vitamin D intervention vs placebo, stratified by baseline total and free 25(OH)D levels

		Baseline total 25(OH)D level (n = 1053)							Baseline free 25(OH)D level (n = 1053)
		<median (28 ng/mL)			≥median			P for inter-action	<median (5.9 pg/mL)			≥median			P for inter-action
		Vitamin D group	Placebo group	P, treatment effect	Vitamin D group	Placebo group	P, treatment effect		Vitamin D group	Placebo group	P, treatment effect	Vitamin D group	Placebo group	P, treatment effect
Grip strength, men, mean ± SE, kg	Baseline	40.13 ± 0.67	38.58 ± 0.70		39.89 ± 0.77	39.93 ± 0.69			40.13 ± 0.68	38.49 ± 0.70		39.93 ± 0.74	40.06 ± 0.68
	2-y change	−0.20 ± 0.48	0.76 ± 0.50	.17	−0.30 ± 0.48	−0.38 ± 0.43	.91	.27	0.01 ± 0.47	0.98 ± 0.48	.15	−0.57 ± 0.49	−0.64 ± 0.45	.92	.27
Grip strength, women, mean ± SE, kg	Baseline	24.29 ± 0.40	23.89 ± 0.45		24.40 ± 0.44	23.88 ± 0.40			24.70 ± 0.44	23.93 ± 0.45		24.01 ± 0.42	23.89 ± 0.41
	2-y change	−0.07 ± 0.33	0.22 ± 0.37	.56	0.28 ± 0.32	−0.41 ± 0.29	.11	.14	−0.71 ± 0.37	0.05 ± 0.38	.15	0.79 ± 0.28	−0.37 ± 0.28	.004	.003
Normal walking speed, mean ± SE, m/s	Baseline	1.23 ± 0.01	1.23 ± 0.01		1.24 ± 0.01	1.25 ± 0.01			1.23 ± 0.01	1.23 ± 0.01		1.24 ± 0.01	1.25 ± 0.01
	2-y change	−0.05 ± 0.01	−0.05 ± 0.01	.61	−0.04 ± 0.01	−0.04 ± 0.01	.65	.95	−0.05 ± 0.01	−0.05 ± 0.01	.65	−0.03 ± 0.01	−0.05 ± 0.01	.16	.19
Fast walking speed, mean ± SE, m/s	Baseline	1.78 ± 0.02	1.77 ± 0.02		1.82 ± 0.02	1.80 ± 0.02			1.78 ± 0.02	1.76 ± 0.02		1.81 ± 0.02	1.81 ± 0.02
	2-y change	−0.06 ± 0.01	−0.06 ± 0.02	.87	−0.07 ± 0.01	−0.08 ± 0.01	.66	.71	−0.06 ± 0.01	−0.05 ± 0.01	.55	−0.07 ± 0.02	−0.09 ± 0.01	.35	.32
Standing balance, median ± SE, s	Baseline	29.27 ± 0.16	29.32 ± 0.18		29.40 ± 0.16	29.55 ± 0.14			29.23 ± 0.16	29.33 ± 0.17		29.40 ± 0.16	29.52 ± 0.15
	2-y change	0.21 ± 0.17	0.09 ± 0.19	.65	−0.10 ± 0.23	−0.37 ± 0.21	.40	.80	0.16 ± 0.19	−0.16 ± 0.20	.23	−0.03 ± 0.22	−0.16 ± 0.20	.64	.61
Repeated chair stands, mean ± SE, s	Baseline	3.58 ± 0.04	3.59 ± 0.05		3.61 ± 0.04	3.68 ± 0.04			3.61 ± 0.04	3.59 ± 0.04		3.58 ± 0.04	3.69 ± 0.04
	2-y change	−0.02 ± 0.04	−0.02 ± 0.05	.93	0.07 ± 0.04	−0.01 ± 0.04	.19	.34	−0.02 ± 0.04	−0.00 ± 0.04	.80	0.06 ± 0.04	−0.02 ± 0.04	.16	.24
TUG, mean ± SE, s	Baseline	8.22 ± 0.09	8.10 ± 0.10		7.93 ± 0.09	8.07 ± 0.08			8.24 ± 0.10	8.13 ± 0.10		7.92 ± 0.08	8.04 ± 0.08
	2-y change	0.30 ± 0.09	0.38 ± 0.10	.56	0.50 ± 0.07	0.25 ± 0.07	.01	.04	0.44 ± 0.08	0.36 ± 0.09	.49	0.36 ± 0.08	0.25 ± 0.07	.29	.90
SPPB score, median ± SE	Baseline	11.39 ± 0.06	11.43 ± 0.07		11.46 ± 0.06	11.56 ± 0.05			11.40 ± 0.06	11.43 ± 0.06		11.44 ± 0.06	11.57 ± 0.06
	2-y change	0.03 ± 0.06	−0.03 ± 0.07	.49	0.06 ± 0.06	−0.06 ± 0.06	.19	.72	0.04 ± 0.06	−0.03 ± 0.06	.40	0.05 ± 0.06	−0.06 ± 0.06	.22	.81

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; SPPB, Short Physical Performance Battery; TUG, Timed-Up and Go.

Among the 711 participants who had 4-year follow-up visits at the Center for Clinical Investigations, vitamin D supplementation did not affect changes in physical performance measures over 4 years, compared to placebo (data not shown). Furthermore, there were no differences in 4-year changes in TUG times between the vitamin D–treated and placebo groups stratified by baseline 25(OH)D levels above and below the median (P = .22 and .91, respectively; P for interaction = .29).

Effects of Supplemental Omega-3 Fatty Acids on Physical Performance Measures

Similarly, both omega-3-treated and placebo groups had slightly slower normal walking speed (change of −0.05 m/s; P < .001, and −.04 m/s; P < .001, respectively) and fast walking speed (change of −0.06 m/s; P < .001, and −.07 m/s; P < .001, respectively) and longer TUG times (0.33 seconds longer; P < .001, in the omega-3 group and 0.38 seconds longer; P < .001, in the placebo group) after 2 years (see Table 2).

There were no differences in effects of supplementation with omega-3 fatty acids vs placebo on 2-year changes in grip strength (P = .42 in men, P = .72 in women), normal walking speed (P = .55), fast walking speed (P = .24), standing balance (P = .16), repeated chair stands (P = .63), TUG tests (P = .56), or SPPB scores (P = .82) (see Table 2).

Effects did not vary by sex (data not shown), age (data not shown), BMI (data not shown), dark-meat fish intake (divided at the median; Table 4), or baseline plasma omega-3 index (divided at the median; see Table 4).

Table 4.

Changes in physical performance measures by omega-3 intervention vs placebo, stratified by baseline dark-meat fish intake and plasma omega-3 index

		Baseline dark-meat fish intake							Baseline plasma omega-3 index
		<median (0.9 servings/wk)			≥median			P for inter-action	<median (2.7%)			≥median			P for inter-action
		Omega-3 group	Placebo group	P, treatment effect	Omega-3 group	Placebo group	P, treatment effect		Omega-3 group	Placebo group	P, treatment effect	Omega-3 group	Placebo group	P, treatment effect
Grip strength, men, mean ± SE, kg	Baseline	39.49 ± 0.94	38.68 ± 0.83		40.44 ± 0.62	38.99 ± 0.65			40.72 ± 0.74	38.74 ± 0.68		40.27 ± 0.68	38.97 ± 0.72
	2-y change	0.28 ± 0.58	0.45 ± 0.51	.83	−0.19 ± 0.44	−0.03 ± 0.46	.80	.995	−0.59 ± 0.52	−0.04 ± 0.48	.44	0.06 ± 0.45	0.35 ± 0.46	.65	.78
Grip strength, women, mean ± SE, kg	Baseline	23.88 ± 0.44	24.52 ± 0.43		24.11 ± 0.43	24.27 ± 0.45			24.07 ± 0.42	24.19 ± 0.45		23.77 ± 0.43	24.38 ± 0.41
	2-y change	0.08 ± 0.31	0.14 ± 0.30	.89	−0.21 ± 0.33	−0.06 ± 0.35	.76	.87	−0.13 ± 0.28	0.04 ± 0.31	.69	−0.11 ± 0.35	0.00 ± 0.33	.81	.98
Normal walking speed, mean ± SE, m/s	Baseline	1.22 ± 0.01	1.24 ± 0.01		1.24 ± 0.01	1.24 ± 0.01			1.21 ± 0.01	1.24 ± 0.01		1.24 ± 0.01	1.25 ± 0.01
	2-y change	−0.05 ± 0.01	−0.04 ± 0.01	.65	−0.05 ± 0.01	−0.04 ± 0.01	.55	.95	−0.03 ± 0.01	−0.04 ± 0.01	.83	−0.06 ± 0.01	−0.05 ± 0.01	.33	.42
Fast walking speed, mean ± SE, m/s	Baseline	1.76 ± 0.02	1.77 ± 0.02		1.79 ± 0.02	1.83 ± 0.02			1.76 ± 0.02	1.76 ± 0.02		1.80 ± 0.02	1.84 ± 0.2
	2-y change	−0.05 ± 0.02	−0.07 ± 0.02	.42	−0.06 ± 0.01	−0.07 ± 0.01	.40	.84	−0.06 ± 0.01	−0.05 ± 0.01	.81	−0.06 ± 0.01	−0.10 ± 0.01	.05	.12
Standing balance, median ± SE, s	Baseline	29.57 ± 0.19	29.16 ± 0.18		29.55 ± 0.15	29.23 ± 0.16			29.46 ± 0.18	29.13 ± 0.18		29.60 ± 0.14	29.28 ± 0.15
	2-y change	−0.12 ± 0.22	0.15 ± 0.21	.39	−0.24 ± 0.19	0.20 ± 0.20	.11	.73	−0.08 ± 0.23	−0.09 ± 0.22	.98	−0.29 ± 0.19	0.24 ± 0.19	.047	.19
Repeated chair stands, mean ± SE, s	Baseline	3.56 ± 0.05	3.63 ± 0.05		3.62 ± 0.04	3.64 ± 0.04			3.64 ± 0.05	3.59 ± 0.04		3.56 ± 0.04	3.67 ± 0.04
	2-y change	0.05 ± 0.05	−0.04 ± 0.05	.20	−0.00 ± 0.04	0.01 ± 0.04	.78	.24	−0.02 ± 0.04	−0.01 ± 0.04	.83	0.05 ± 0.04	−0.01 ± 0.04	.31	.40
TUG, mean ± SE, s	Baseline	8.11 ± 0.10	7.96 ± 0.10		8.09 ± 0.08	8.16 ± 0.09			8.07 ± 0.10	8.05 ± 0.09		8.15 ± 0.08	8.04 ± 0.09
	2-y change	0.30 ± 0.08	0.46 ± 0.08	.16	0.34 ± 0.08	0.26 ± 0.08	.48	.13	0.39 ± 0.08	0.37 ± 0.08	.83	0.27 ± 0.08	0.39 ± 0.08	.31	.39
SPPB score, median ± SE	Baseline	11.44 ± 0.07	11.45 ± 0.06		11.49 ± 0.06	11.47 ± 0.06			11.47 ± 0.07	11.38 ± 0.06		11.46 ± 0.06	11.53 ± 0.06
	2-y change	0.03 ± 0.07	−0.03 ± 0.06	.57	−0.03 ± 0.06	0.06 ± 0.06	.30	.27	0.01 ± 0.06	−0.00 ± 0.06	.92	−0.01 ± 0.06	0.01 ± 0.06	.81	.81

Abbreviations SPPB, Short Physical Performance Battery; TUG, Timed-Up and Go.

Omega-3 fatty acids supplementation did not affect changes in the majority of the physical performance measures at 4-year follow-up, compared to placebo (data not shown). The placebo group did have slightly better standing balance results than the omega-3 group at 4 years (P for trend over time = .01).

Finally, in exploratory analyses, the combination of vitamin D and omega-3 supplementation for 2 years, compared to other treatment groups or double placebo, did not improve physical performance measures (data not shown), except for fast gait speed. The combination of supplemental vitamin D and omega-3 slowed the decline in fast gait speed (−0.04 ± 0.01 m/s; P = .01) compared to the other treatment groups (−0.08 ± 0.01 m/s; P < .001; P for treatment effect = .02).

Discussion

In this generally healthy adult population not selected for low vitamin D levels or low fish intake, supplementation with vitamin D₃ and/or omega-3 fatty acids for 2 years did not improve physical performance measures, compared to respective placebos. Due to multiple comparisons, subgroup analyses and secondary end points should be interpreted with caution.

This VITAL ancillary study included participants who were generally healthy midlife to older adults, and there were few participants with muscle weakness or poor muscle performance, as delineated by EWGSOP2 (33). Furthermore, the mean baseline 25(OH)D level of 28.1 ng/mL was likely sufficient for most health outcomes. Other RCTs have targeted participants with low vitamin D status and/or functional limitations. Two small RCTs suggested that older, frail women with low vitamin D status may derive the most benefit (15, 16). In an RCT in community-dwelling older women with mean (±SD) baseline 25(OH)D of 17.7 ± 4.2 ng/mL, only the weakest and slowest participants had significant improvements in hip muscle strength and TUG times with vitamin D supplementation (15). In another RCT in older women in long-term geriatric care, vitamin D supplementation for 3 months improved musculoskeletal function (16). More recent studies have not found a benefit for vitamin D supplementation on physical performance measures. An RCT in Black women with 25(OH)D levels less than 30 ng/mL did not find a benefit for vitamin D supplementation for 3 years on SPPB, 6-minute walk test, or grip strength (34). Two recent RCTs targeted older adults with low vitamin D status and functional limitations but also did not find an effect of vitamin D supplementation on physical performance measures (9, 35). These trials were smaller and shorter in duration (≤12 months) compared to this ancillary VITAL study. The large DO-HEALTH RCT in 2157 European adults aged 70 years or older also did not find a benefit for 2000 IU/day of vitamin D₃ for 3 years on SPPB (36). Finally, a recent meta-analysis published in 2021 suggested that vitamin D supplementation may have adverse effects on muscle health, including decreased knee flexion strength and slower TUG time (37). In VITAL, we observed longer TUG times in participants treated with supplemental vitamin D with higher baseline total 25(OH)D levels, compared to placebo (P for interaction = .04).

Our findings of no improvement in physical performance with supplementation of omega-3 fatty acids were also consistent with the findings from recent larger RCTs, including the DO-HEALTH RCT (36). The Multidomain Alzheimer Preventive Trial (MAPT) randomly assigned 1679 participants with mean age of 75 years to omega-3 fatty acids supplementation (800 mg DHA and 225 mg EPA) alone, omega-3 fatty acids and multidomain intervention (including physical activity, nutrition, and cognitive training), the multidomain intervention plus placebo, or placebo for 3 years. In this trial, omega-3 fatty acids supplementation also did not improve physical performance, as assessed by repeated chair stands, grip strength, walking speed, and SPPB, including in participants with low DHA + EPA erythrocyte level at baseline (38).

Our results were consistent with our recent findings that vitamin D₃ supplementation vs placebo did not improve measures of lean mass (lean mass index, appendicular lean mass [ALM], and ALM/BMI) in a VITAL subcohort of 771 participants who had body composition assessments by dual-energy x-ray absorptiometry (39). Furthermore, we had also found that vitamin D₃ supplementation vs placebo did not decrease risk of incident falls over 5.3 years in the overall VITAL cohort of 25 871 participants (40). Finally, compared to placebo, supplemental vitamin D or omega-3 fatty acids also did not affect frailty score over 5 years, as assessed by annual questionnaires using the Rockwood frailty index, which includes measures of function, cognition, mood, and comorbidities (41). Similarly, the DO-HEALTH study also did not find a benefit for vitamin D supplementation alone on falls or frailty, according to Fried physical frailty phenotype (42, 43). However, omega-3 fatty acid supplementation modestly reduced total falls by 10%, and the combination of vitamin D₃, omega-3 fatty acids, and exercise decreased odds of becoming prefrail (42, 43).

A major strength of this ancillary VITAL study is the size. We had 1054 participants, allowing for subgroup analyses, including by baseline 25(OH)D levels and plasma omega-3 index. Total 25(OH)D levels were calibrated to CDC standards, and free 25(OH)D levels were measured by Future Diagnostics Solutions B.V. This study had high retention (92%). We also had excellent adherence, as indicated by the increase in total 25(OH)D levels meeting the prespecified target of 36 ng/mL in the active intervention group. A limitation was that some of the SBBP components and the TUG test were initiated later in the interventional trial, so not every participant in the CTSC subcohort had all physical performance measurements assessed at baseline and 2 years. As by trial design, VITAL was performed in a generally healthy population with low rates of muscle weakness and poor performance and few participants with low baseline 25(OH)D levels. As allowed by trial design, 46.3% of the participants in the placebo group took their own personal vitamin D supplements (limited to ≤800 IU/day) at baseline, along with 44.2% in the active vitamin D group. Finally, a daily dose of 2000 units of cholecalciferol may be considered moderate to high, which was selected to achieve maximal efficacy for the primary outcomes of the parent trial while maintaining safety (44).

In a US population of midlife to older adults, not selected for muscle weakness or function, low vitamin D levels, or low fish intake, supplementation with vitamin D₃ and/or omega-3 fatty acids for 2 years did not improve physical performance measures compared to placebo. These findings do not support supplementation with vitamin D or omega-3 acids for muscular health in the general population. Along with other findings from VITAL, these data suggest that public health guidelines should recommend against vitamin D supplementation for the prevention of adverse musculoskeletal health outcomes in healthy midlife to older adults.

Acknowledgments

We thank the trial participants, staff, and investigators. We also thank the CDC (Drs Hubert Vesper and Julianne Cook Botelho) for their collaboration on the standardization and calibration of the total 25(OH)D measurements.

Funding

This work was supported by the National Institute of Arthritis Musculoskeletal and Skin Diseases/National Institutes of Health (NIAMS/NIH; grant Nos. R01 AR059775, R01 AR070854, and R01 AR060574; PI, M.S..L). This research was also supported by the VITAL parent grants and U01 CA138962 and R01 CA138962 (PIs, J.E.M. and J.E.B.) from the National Cancer Institute, the National Heart, Lung, and Blood Institute, the Office of Dietary Supplements, the National Institute of Neurological Disorders and Stroke, and the National Center for Complementary and Integrative Health. This work was conducted with support from grants 1 UL1 RR025758 and 8 UL1 TR000170, the Harvard Clinical and Translational Science Center, from the National Center for Research Resources and by 1UL1TR001102. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAMS, National Center for Research Resources, the National Center for Advancing Translational Science, or the NIH.

Disclosure

J.E.B. reports Pharmavite: other financial or material support. The other authors have nothing to disclose.

Data Availability

Some or all data sets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Clinical Trial Information

Clinical trial registration numbers NCT01747447 and NCT01704859 (registered December 11, 2012).

Abbreviations

 
• 25(OH)D

  25-hydroxyvitamin D

 
• ALA

  alpha-linolenic acid

 
• ALM

  appendicular lean mass

 
• BMI

  body mass index

 
• CDC

  Centers for Disease Control and Prevention

 
• CTSC

  Clinical and Translational Science Center

 
• DHA

  docosahexaenoic acid

 
• EPA

  eicosapentaenoic acid

 
• EWGSOP2

  European Working Group on Sarcopenia in Older People 2

 
• RCT

  randomized controlled trial

 
• SPPB

  Short Physical Performance Battery

 
• TUG

  Timed-Up and Go

 
• VDR

  1,25-dihydroxyvitamin D receptor

 
• VITAL

  VITamin D and OmegA-3 TriaL