Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Nov 15.
Published in final edited form as: Arthritis Rheumatol. 2024 Feb 20;76(6):973–983. doi: 10.1002/art.42811

Vitamin D and Marine n-3 Fatty Acids for Autoimmune Disease Prevention: Outcomes Two Years After Completion of a Double-Blind, Placebo-Controlled Trial

Karen H Costenbader 1, Nancy R Cook 1, I-Min Lee 1, Jill Hahn 2, Joseph Walter 3, Vadim Bubes 3, Gregory Kotler 3, Nicole Yang 1, Sonia Friedman 1, Erik K Alexander 1, JoAnn E Manson 1
PMCID: PMC11565399  NIHMSID: NIHMS2032846  PMID: 38272846

Abstract

Objective.

In the 5.3-year randomized, 2 × 2 factorial, double-blind, placebo-controlled Vitamin D and Omega-3 Trial (VITAL), vitamin D supplementation reduced autoimmune disease (AD) incidence (hazard ratio [HR] 0.78, 95% confidence interval [CI] 0.61–0.99). Omega-3 (n-3) fatty acid supplementation showed a statistically nonsignificant reduction (HR 0.85, 95% CI 0.67–1.08). We aimed to confirm further AD cases arising during and after randomization and assess sustained effects with two years of postintervention observation.

Methods.

Of the 12,786 men aged ≥50 and 13,085 women aged ≥55 initially randomized, we observed surviving and willing participants for two more years. We continued to confirm annual participant-reported new AD by medical record review. Cox models calculated HRs for all confirmed incident AD, (and secondary endpoints, including probable cases, and individual ADs), during the observational and randomized periods.

Results.

A total of 21,592 participants (83.5%) were observed for two more years; 514 participants developed incident confirmed AD (236 since prior report), of whom 255 had been randomized to vitamin D versus 259 to vitamin D placebo (HR 0.98 [95% CI 0.83–1.17] at 7 years). AD was confirmed in 234 participants initially randomized to n-3 fatty acids versus 280 randomized to its placebo (HR 0.83 [95% CI 0.70–0.99] at 7 years). Of newly confirmed cases, 65 had onset during randomization; their inclusion changed randomized results as follows: HR 0.85 (95% CI 0.70–1.04) for vitamin D and HR 0.87 (95% CI 0.71–1.06) for n-3 fatty acids.

Conclusion.

Two years after trial termination, the protective effects of 2000 IU/day of vitamin D dissipated, but 1,000 mg/day of n-3 fatty acids had a sustained effect in reducing AD incidence.

INTRODUCTION

Autoimmune diseases (ADs), including rheumatoid arthritis (RA), polymyalgia rheumatica (PMR), psoriasis, autoimmune thyroid disease (AITD), inflammatory bowel disease (IBD), systemic vasculitides, and others, are a leading cause of morbidity and mortality among older adults, appear to have rising incidence, and share many underlying pathogenic mechanisms.1–4 Most ADs are chronic conditions, and many have increasing prevalence with age.5 ADs incur high burdens of disease and costs due to use of corticosteroids and toxic and expensive immunosuppressive therapies, disability, and lost productivity.5,6 Thus, there is keen interest in the identification of effective strategies for the prevention of the development of ADs.7

Vitamin D and marine-derived long-chain omega-3 (n-3) fatty acids are widely available nutritional supplements with many biologic mechanisms that may act to forestall the development of ADs in humans. These include effects on T and B lymphocyte development and activity, autoantibody production, and systemic inflammation in in vitro, animal, and human studies.8–15 In vitro, the lipid-soluble active form of vitamin D [25-hydroxyvitamin D: 1,25(OH)D2] regulates a wide array of genes involved in inflammation and acquired and innate immune responses.16 The n-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are essential lipids the body cannot synthesize, found in high quantities in fish oils.8 In the 1980s and 1990s, randomized controlled trials proved that fish oil supplements were effective in decreasing pain and inflammation in RA, improving patient and physician assessments of disease activity, and decreasing use of analgesics.17–19

We previously reported the preventive effects of vitamin D and n-3 fatty acid supplementation on the reduction of AD incidence within the large-scale Vitamin D and Omega-3 Trial (VITAL) over a median of 5.3 years of randomized follow-up.20–22 Randomization to vitamin D reduced AD incidence20 (hazard ratio [HR] 0.78, 95% confidence interval [CI] 0.61–0.99). Receiving n-3 fatty acids led to suggestive protection, with nonsignificantly lower incidence of confirmed AD (HR 0.85, 95% CI 0.67–1.08) but significantly lower incidence when probable incident AD cases were included20 (HR 0.82, 95% CI 0.68–0.99).

The effects on AD risk when these supplements, which have been taken continuously for a period of years, are discontinued are still unknown. When VITAL ended on December 31, 2017, trial participants were invited to be observed with similar trial questionnaires in a postrandomized noninterventional follow-up. We herein report the results of follow-up for incident ADs using the same case ascertainment and confirmation methods over the next two years of observational follow-up to examine whether the effects of these two supplements on reducing AD incidence were sustained in this population of older American participants. We have re-examined prior VITAL randomized trial results with longer follow-up (two more years of AD case finding and confirmation). We also examined other prespecified endpoints, including those with lagging of the onset for inclusion of incident AD by two years given the known long latency period before diagnosis of many ADs, and testing for interactions of the vitamin D effect with participant body mass index (BMI).

PATIENTS AND METHODS

VITAL was a randomized, double-blind, placebo-controlled 2 × 2 factorial design trial testing the benefits and risks of vitamin D (cholecalciferol; 2000 IU/day) and marine n-3 fatty acids (1 g/day as a fish oil capsule containing 460 mg of EPA and 380 mg of DHA) in the prevention of cancer and cardiovascular disease (CVD) among 25,871 men (aged ≥50 years) and women (≥55 years)21,22 (ClinicalTrials.gov identifier: NCT01169259). Aggregate incident AD and confirmed and validated cases, as well as incidence of the most common ADs, were prespecified end points of a funded ancillary study initiated before trial recruitment20 (ClinicalTrials.gov identifier: NCT01351805).

As previously described, participants were recruited throughout the United States, and during the trial, participants were requested to limit vitamin D use from outside sources to no more than 800 IU/day and forego the use of n-3 fatty acid supplements.20–22 Individuals with a history of renal failure or dialysis, cirrhosis, hypercalcemia, cancer (except nonmelanoma skin cancer), CVD, or other serious illness were ineligible for the trial. A total of 25,871 individuals, including 5,106 Black individuals and 2,152 individuals of not Black or White race and ethnicity, consented and were randomized to treatment (for vitamin D, n = 12,927; for n-3 fatty acids, n = 12,933) or placebo arms (for vitamin D placebo, n = 12,944; for n-3 fatty acid placebo, n = 12,938) within sex, race, and five-year age groups. Calendar packs containing similar-appearing trial capsules were mailed to participants with questionnaires. The vitamin D pill and its matching placebo, which contained soybean oil, were prepared by Pharmavite, LLC. The n-3 fatty acid pill and its matching placebo, which contained olive oil, were prepared by Pronova BioPharma. Baseline questionnaires collected clinical and lifestyle risk factors, and queried vitamin D supplement use, as well as fish and dairy intake.23 Questionnaires at six months and one year post randomization, and then annually, asked about trial supplement adherence, new doctor-diagnosed diseases, potential side effects of trial agents, and new cancer or CVD risk factors.

Randomization occurred between November 2011 and March 2014, and the intervention was completed as prespecified after a median of 5.3 years of randomized assignment22 on December 31, 2017. After unblinding, participants were no longer provided with any supplements but were invited to continue being observed while off assigned supplements, with annual questionnaires and reporting of end points occurring through December 31, 2019. The trial and observational extension were approved by the institutional review board (IRB) of Partners HealthCare and monitored by an external data and safety monitoring board.

AD end points.

Annual questionnaires assessing incident AD during the randomized follow-up were continued in the observational extension. Questionnaires inquired about new-onset doctor-diagnosed RA, PMR, AITD, psoriasis, and IBD and had spaces for participants to write in all other new-onset doctor-diagnosed ADs. As in the randomized portion of the trial, participants who reported a new incident AD were asked to sign a release for medical records. A trained physician (including a board-certified rheumatologist, endocrinologist, and gastroenterologist) blinded to treatment assignment reviewed each record and confirmed or disconfirmed the AD according to classification criteria, where available. We classified cases with evidence of incident AD but insufficient documentation for certainty as “probable” cases and added them to confirmed cases as a secondary end point. Dates of doctors’ diagnoses and first symptoms were recorded from the medical records. Deaths were confirmed by review of medical records and death certificates, as previously reported.23 In the published analyses of VITAL AD, we continued case confirmation efforts through December 31, 2019, and included cases diagnosed through trial completion,20 December 31, 2017. Here, we have closed the follow-up case file on April 27, 2022, for this analysis and have included cases confirmed as diagnosed through December 31, 2019, two years after randomized therapy ended. Thus, during the ongoing effort to identify AD cases in follow-up, we confirmed some AD cases that, because of delays in diagnosis and medical record documentation, had onset during the randomized portion of the trial.

Statistical analyses.

We conducted t-tests and chi square tests to compare baseline characteristics of participants in the observational extension study according to their initial randomization group and compared them to the entire group of VITAL randomized participants. All analyses were prespecified in the protocol. For our primary analyses, we compared the main effects of randomization to vitamin D and to n-3 fatty acid supplements on AD incidence using cumulative incidence curves and Cox regression models, both after the two years of observational extension, and then reanalyzed the effects of the supplements during the randomized portion of the trial with the most recently updated end points file. To account for randomization stratification and study design, Cox models were adjusted for age, sex, self-reported race, and randomization to the other supplement.24 Person-time was counted until diagnosis of an incident confirmed AD, death, withdrawal, or the end of the trial or the observational period for AD, December 31, 2019.

In addition to our strict definition of confirmed AD, we ran models including both the confirmed and the probable incident ADs as end points. We assessed effects of treatment on prespecified individual disease end points (RA, PMR, AITD, psoriasis, and IBD) and grouped all other ADs as an additional end point of “other AD.” We continued to observe those with AD predating the trial for a new AD given that underlying shared genetic factors across multiple ADs predisposes individuals with one AD to developing another new AD. For analyses of individual ADs, we excluded those with the same AD previously reported at baseline. To assess for effects of supplementation with both n-3 fatty acids and vitamin D, specified a priori, we examined four group cumulative incidence curves and repeated Cox models with the placebo/placebo group as the reference group compared to the three intervention arms.

We repeated prior analyses using a two-year lagged onset for the inclusion of incident AD given the known long latency period, which had shown stronger effects of vitamin D, and again tested for interactions with participant BMI and baseline reported family history of AD.20 We also repeated main analyses excluding those who self-reported taking >800 IU/day of vitamin D or n-3 fatty acid supplementation in the observational period. Data analyses were performed using SAS 9.4 (SAS Institute Inc). The study protocol was approved by the IRB at Partners HealthCare (now Mass General Brigham). Data from this study may be considered for release if the appropriate IRB approval and publication clearances have been obtained, and a project is approved by the VITAL oversight committee. Patients and the public were not involved in the design, conduct, reporting, or dissemination plans of our research.

RESULTS

Overall, 21,592 of VITAL participants agreed and were observed after the randomized trial ended (83.5% of those initially randomized and 87.9% of those alive with a good address at trial end). Characteristics of these participants at the start of the two-year observational follow-up are shown in Table 1 (more details of the initially randomized VITAL participants are published20,21). Overall, 50.6% were women, and the mean age was 67 (SD ±6.7) years. The self-identified trial composition was 74.3% non-Hispanic White participants, 17.5% Black participants, and 8.2% other racial and ethnic groups. These and other characteristics, such as BMI (mean ± SD 28.1 ± 5.7 vs 27.9 ± 5.5) and family history of AD (34.3% vs 34.9%), did not differ in those initially randomized versus those who continued in the observational extension. Other demographic and lifestyle factors did not differ across the randomization arms. During the observational period, 6,125 participants (28.3%) reported personal use of >800 IU/day of vitamin D and 4,944 (22.9%) reported outside n-3 fatty acid supplement intake.

Table 1.

Characteristics of the VITAL participants at baseline and those entering the observational extension at a median of 5.3 y according to randomized assignment to active supplementation (vitamin D or n-3 fatty acids) or placebo*

Characteristic Randomized
(n = 25,871)		 Observational extension
(n = 21,592)		 Vitamin D
(observational extension)		 n-3 fatty acid arm
(observational extension)
Active
(n = 10,866)		 Placebo
(n = 10,726)		 Active
(n = 10,812)		 Placebo
(n = 10,780)
Female, n (%) 13,085 (50.6) 10,933 (50.6) 5,512 (50.7) 5,421 (50.5) 5,477 (50.7) 5,456 (50.6)
Age, mean ± SD, y 67.1 ± 7.1 67.03 ± 6.7 67.1 ± 6.7 67.0 ± 6.7 67.0 ± 6.6 67.0 ± 6.7
Race and ethnicity, n (%)a
 Non-Hispanic White 18,046 (71.3) 15,699 (74.3) 7,877 (74.1) 7,822 (74.5) 7,868 (74.4) 7,831 (74.2)
 Black 5,106 (20.2) 3,704 (17.5) 1,879 (17.7) 1,825 (17.4) 1,858 (17.6) 1,846 (17.5)
 Not self-reporting Black or White 2,152 (8.5) 1,723 (8.2) 876 (8.2) 847 (8.1) 852 (8.1) 871 (8.3)
Geographic region of residence, n (%)
 Northeast 7,161 (27.7) 6,077 (28.1) 3,086 (28.4) 2,991 (27.9) 3,002 (27.8) 3,075 (28.5)
 Midwest 5,541 (21.4) 4,621 (21.4) 2,348 (21.6) 2,273 (21.2) 2,318 (21.4) 2,303 (21.4)
 West 5,926 (22.9) 5,023 (23.3) 2,494 (23.0) 2,529 (23.6) 2,547 (23.6) 2,476 (23.0)
 Southeast 7,242 (28.0) 5,870 (27.2) 2,938 (27.0) 2,932 (27.3) 2,944 (27.2) 2,926 (27.1)
High school education or less, n (%) 3,304 (12.8) 2,373 (11.0) 1,203 (11.1) 1,170 (10.9) 1,224 (11.3) 1,149 (10.7)
Self-reported annual income <$50,000, n (%)b 8,523 (36.6) 6,481 (33.3) 3,286 (33.6) 3,195 (33.1) 3,220 (33.0) 3,261 (33.6)
Body mass index, mean ± SDc 28.1 ± 5.7 27.9 ± 5.5 27.9 ± 5.5 27.9 ± 5.5 28.0 ± 5.6 27.8 ± 5.5
Physical activity total MET hr/wk, median (IQR) 15.4 (4.6–31.6) 16.6 (5.4–32.7) 16.4 (5.2–32.5) 16.8 (5.5–33.0) 16.6 (5.4–32.6) 16.6 (5.3–32.8)
Current smoking, n (%) 1,836 (7.2) 1,253 (5.9) 643 (6.0) 610 (5.8) 618 (5.8) 635 (6.0)
Self-reported family history of any autoimmune disease, n (%) 8,168 (34.3) 6,965 (34.9) 3,517 (34.9) 3,448 (34.9) 3,508 (34.9) 3,457 (34.8)
Open in a new tab
*

Percentages may not sum to 100% because of rounding. There were no significant differences between groups regarding baseline characteristics. IQR, interquartile range; MET, metabolic equivalent; VITAL, Vitamin D and Omega-3 Trial.

a

Race and ethnicity reported by participants.

b

Income data missing for 10% of participants.

c

Body mass index is weight in kilograms divided by the square of height in meters.

Primary and secondary end points: Confirmed and confirmed plus probable incident AD.

A total of 236 new AD cases were confirmed since the prior publication.20 This included 98 new confirmed cases of AD with onset during the randomized portion of the trial and 138 new confirmed cases in the observational extension. There were also 65 additional probable AD cases identified during the median 5.3 years of randomized time and 42 probable AD cases diagnosed after trial completion during the 2-year observational phase (Tables 2 and 3). Deaths and adverse events during the randomized trial period were low, as previously reported22; in the observational follow-up, there were 785 additional deaths, equally distributed across the initially randomized trial arms.

Table 2.

HRs and 95% CIs for the primary and secondary end points after a median 5.3 y of randomization and 2 y of observational follow-up: Vitamin D or placebo*

Autoimmune disease end points Published VITAL resultsa Updated VITAL resultsb VITAL results with 2-y observational extensionc
Vitamin D, n Placebo, n HR (95% Cl) P Vitamin D, n Placebo, n HR (95% Cl) P Vitamin D, n Placebo, n HR (95% Cl) P
Primary endpoint
 Confirmed autoimmune diseases 123 155 0.78 (0.61–0.99) 0.045 173 203 0.85 (0.70–1.04) 0.12 255 259 0.98 (0.83–1.17) 0.84
Secondary end point
 Confirmed and probable autoimmune diseases 210 247 0.85 (0.70–1.02) 0.09 285 335 0.85 (0.73–1.00) 0.045 391 409 0.95 (0.83–1.10) 0.50
Individual autoimmune diseases
 Confirmed rheumatoid arthritis 15 24 0.58 (0.30–1.13) 0.11 20 25 0.80 (0.44–1.44) 0.46 29 35 0.83 (0.51–1.35) 0.45
 Confirmed and probable rheumatoid arthritis 18 27 0.63 (0.34–1.15) 0.13 23 30 0.77 (0.45–1.32) 0.34 35 44 0.79 (0.51–1.24) 0.31
 Confirmed polymyalgia rheumatica 31 43 0.70 (0.44–1.12) 0.14 36 54 0.67 (0.44–1.02) 0.06 56 63 0.89 (0.62–1.27) 0.52
 Confirmed and probable polymyalgia rheumatica 32 43 0.72 (0.46–1.15) 0.17 38 57 0.67 (0.44–1.01) 0.054 61 66 0.92 (0.65–1.31) 0.66
 Confirmed autoimmune thyroid disease 21 11 1.63 (0.77–3.45) 0.20 27 18 1.49 (0.82–2.71) 0.19 36 23 1.56 (0.92–2.63) 0.10
 Confirmed and probable autoimmune thyroid disease 99 94 1.05 (0.78–1.41) 0.74 120 126 0.95 (0.74–1.22) 0.67 149 142 1.04 (0.83–1.30) 0.72
 Confirmed psoriasis 15 23 0.72 (0.37–1.39) 0.32 21 32 0.66 (0.38–1.14) 0.14 28 46 0.61 (0.38–0.98) 0.04
 Confirmed and probable psoriasis 17 25 0.70 (0.37–1.32) 0.27 24 35 0.69 (0.41–1.16) 0.16 31 50 0.62 (0.40–0.97) 0.04
 Confirmed inflammatory bowel diseased 3 9 0.33 (0.09–1.23) 0.10 4 9 0.45 (0.14–1.45) 0.18 10 11 0.91 (0.39–2.14) 0.83
 Confirmed and probable inflammatory bowel diseased 4 10 0.40 (0.13–1.28) 0.12 5 10 0.50 (0.17–1.46) 0.21 11 12 0.92 (0.40–2.08) 0.84
 Confirmed other autoimmune diseased 37 47 0.82 (0.53–1.27) 0.38 51 51 1.00 (0.68–1.47) 0.99 78 66 1.18 (0.85–1.64) 0.33
 Confirmed and probable other autoimmune diseased 41 53 0.80 (0.53–1.21) 0.30 58 59 0.98 (0.68–1.41) 0.92 85 78 1.09 (0.80–1.48) 0.60
Open in a new tab
*

N = 25,871: 12,927 randomized to vitamin D and 12,944 randomized to placebo. Analyses from Cox regression models were controlled for age, sex, race, and n-3 fatty acid randomization group. Boldface indicates statistical significance. CI, confidence interval; HR, hazard ratio; VITAL, Vitamin D and Omega-3 Trial.

a

Cases with onset during the randomized time that were confirmed through December 2019.20

b

Cases with onset during the randomized time that were confirmed through April 2022.

c

Cases with onset during the randomized and observational times that were confirmed through April 2022.

d

Inflammatory bowel disease was grouped with other autoimmune diseases in the published trial results.20

Table 3.

HRs and 95% CIs for the primary and secondary end points after 5.3 y of randomization and 2 y of observational follow-up: n-3 fatty acids or placebo*

End points: numbers of cases Published VITAL resultsa Updated VITAL resultsb VITAL results with 2-y observational extensionc
n-3 fatty acids, n Placebo, n HR (95% Cl) P n-3 fatty acids, n Placebo, n HR (95% Cl) P n-3 fatty acids, n Placebo, n HR (95% Cl) P
Primary end point
 Confirmed autoimmune diseases 130 148 0.85 (0.67–1.08) 0.19 175 201 0.87 (0.71–1.06) 0.17 234 280 0.83 (0.70–0.99) 0.04
Secondary end point
 Confirmed and probable autoimmune diseases 208 249 0.82 (0.68–0.99) 0.04 287 333 0.86 (0.73–1.00) 0.06 370 430 0.86 (0.74–0.98) 0.03
Individual autoimmune diseases
 Confirmed rheumatoid arthritis 15 24 0.58 (0.30–1.13) 0.11 18 27 0.66 (0.37–1.21) 0.18 25 39 0.64 (0.39–1.06) 0.08
 Confirmed and probable rheumatoid arthritis 17 28 0.57 (0.31–1.05) 0.07 21 32 0.65 (0.38–1.13) 0.13 34 45 0.75 (0.48–1.18) 0.21
 Confirmed polymyalgia rheumatica 34 40 0.87 (0.55–1.38) 0.55 42 48 0.87 (0.58–1.32) 0.51 54 65 0.83 (0.58–1.19) 0.30
 Confirmed and probable polymyalgia rheumatica 34 41 0.85 (0.54–1.34) 0.48 43 52 0.82 (0.55–1.23) 0.35 57 70 0.81 (0.57–1.15) 0.24
 Confirmed autoimmune thyroid disease 12 20 0.53 (0.25–1.14) 0.10 18 27 0.67 (0.37–1.22) 0.19 23 36 0.64 (0.38–1.08) 0.10
 Confirmed and probable autoimmune thyroid disease 85 108 0.80 (0.59–1.07) 0.13 112 134 0.84 (0.65–1.08) 0.17 135 156 0.87 (0.69–1.09) 0.23
 Confirmed psoriasis 23 15 1.57 (0.80–3.07) 0.19 33 20 1.65 (0.95–2.88) 0.08 47 27 1.74 (1.09–2.80) 0.02
 Confirmed and probable psoriasis 25 17 1.44 (0.76–2.72) 0.26 36 23 1.57 (0.93–2.65) 0.09 51 30 1.70 (1.08–2.67) 0.02
 Confirmed inflammatory bowel diseased 3 9 0.33 (0.09–1.23) 00.1 4 9 0.44 (0.14–1.44) 0.18 8 13 0.61 (0.25–1.48) 0.28
 Confirmed and probable inflammatory bowel diseased 3 11 0.27 (0.08–0.98) 0.046 4 11 0.36 (0.12–1.14) 0.08 8 15 0.53 (0.23–1.26) 0.15
 Confirmed other autoimmune diseased 42 42 0.95 (0.61–1.47) 0.82 52 50 1.03 (0.70–1.52) 0.87 66 78 0.84 (0.61–1.17) 0.30
 Confirmed and probable other autoimmune diseased 45 49 0.87 (0.58–1.32) 0.52 57 60 0.94 (0.66–1.36) 0.76 71 92 0.77 (0.56–1.05) 0.09
Open in a new tab
*

N = 25,871: 12,933 randomized to n-3 fatty acids and 12,938 randomized to placebo. Analyses fromCox regression models were controlled for age, sex, race, and n-3 fatty acid randomization group. Boldface indicates statistical significance. CI, confidence interval; HR, hazard ratio; VITAL, Vitamin D and Omega-3 Trial.

a

Cases with onset during the randomized time that were confirmed through December 2019.20

b

Cases with onset during the randomized time that were confirmed through April 2022.

c

Cases with onset during the randomized and observational times that were confirmed through April 2022.

d

Inflammatory bowel disease was grouped with other autoimmune diseases in the published trial results.20

Among those initially randomized to receive vitamin D, after two more years of observational nonrandomized follow-up, 255 participants had newly developed AD, compared to 259 among those initially randomized to vitamin D placebo (HR 0.98, 95% CI 0.83–1.17; Table 2, Figure 1). When we included the probable AD cases as well, the adjusted HR through the observational period was 0.95 (95% CI 0.83–1.10) for randomization to vitamin D. The numbers of individual confirmed and confirmed plus probable AD cases (RA, PMR, AITD, psoriasis, IBD, and all others) through the observational period with their adjusted HRs are also shown in Table 2. The only AD for which active vitamin D resulted in prolonged and significantly reduced HRs was psoriasis (adjusted HRs of 0.61 [95% CI 0.38–0.98] for confirmed cases and 0.62 [95% CI 0.40–0.97] when probable cases were included).

Figure 1.

Figure 1.

Open in a new tab

Adjusted cumulative incidence curves for autoimmune disease incidence among those randomized to vitamin D or placebo in VITAL (N = 25,871) over a median of 5.3 years of randomized follow-up and 2 years of observational follow-up after trial termination. CI, confidence interval; VITAL, Vitamin D and Omega-3 Trial.

Among those initially randomized to active n-3 fatty acids, after two more years of observational follow-up, confirmed incident AD had occurred in 234 participants, compared to 280 among those initially randomized to n-3 fatty acid placebo (adjusted HRs 0.83 [95% CI 0.70–0.99] and 0.86 [95% CI 0.74–0.98] when the probable cases were included; Table 3, Figure 2). The breakdown into the most common individual ADs is shown in Table 3 as well; of these, only psoriasis had a significant adjusted HR, this time for increased risk associated with initial randomization to n-3 fatty acids versus placebo (adjusted HRs 1.74 [95% CI 1.09–2.80] for confirmed cases and 1.70 [95% CI 1.08–2.67] for confirmed and probable cases together).

Figure 2.

Figure 2.

Open in a new tab

Cumulative incidence curves for autoimmune disease incidence among those randomized to n-3 fatty acids or placebo in VITAL (N = 25,871) over a median of 5.3 years of randomized follow-up and 2 years of observational follow-up after trial termination. CI, confidence interval; VITAL, Vitamin D and Omega-3 Trial.

In reanalyses of the AD cases diagnosed during the randomized trial period, now including the 65 new cases confirmed through April 2022, there were 173 confirmed cases in the vitamin D arm and 203 in the vitamin D placebo arm (HR 0.85, 95% CI 0.70–1.04), and there were 175 confirmed cases in the n-3 fatty acid arm and 201 in the n-3 fatty acid placebo arm (HR 0.87, 95% CI 0.71–1.06). When including the probable cases, these updated HRs were 0.85 (95% CI 0.73–1.00) for vitamin D and 0.86 (95% CI 0.73–1.00) for n-3 fatty acids.

Secondary analyses.

When analyzed in the four group factorial arms (those randomized to two active supplements, two placebo pills, vitamin D alone, and n-3 fatty acids alone), all of the intervention arms continued to have lower cumulative incidence of ADs over follow-up than did those initially on placebo (Figure 3). The reductions associated with active vitamin D only were no longer statistically significant, but they were significant for placebo vitamin D and active n-3 fatty acids versus two placebos (HR 0.71, 95% CI 0.55–0.91). When the cases identified during the observational extension (but with onset during the randomized time) were included, the HRs were similarly reduced for all groups but no longer statistically significant (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Cumulative incidence curves for autoimmune disease incidence in the four arms of VITAL (N = 25,871) over a median of 5.3 years of randomized follow-up and 2 years of observational follow-up after trial termination. CI, confidence interval; VITAL, Vitamin D and Omega-3 Trial.

In analyses lagging the period for incident AD case confirmation such that it started two years after randomization, as prespecified in the main trial analyses, to examine the delay in onset of supplement effects and delay in AD diagnoses, there were 274 confirmed incident cases: 131 in the active vitamin D arm versus 143 in the placebo vitamin D arm (HR 0.91, 95% CI 0.71–1.15). There were also 124 confirmed incident cases in the active n-3 fatty acid arm versus 150 in the placebo n-3 fatty acid arm (HR 0.83, 95% CI 0.65–1.05). Additionally, in analyses stratified by BMI World Health Organization categories, as in the initial trial analyses, we continued to find a stronger protective effect of vitamin D among those in the lowest BMI category (BMI < 25; HR 0.80 [95% CI 0.60–1.08], P = 0.006 for interaction with BMI for confirmed AD; HR 0.75 [95% CI 0.59–0.95] for confirmed and probable AD, P = 0.01 for interaction with BMI). We again found that the beneficial effect of n-3 fatty acids for AD prevention was greater among those with a family history of AD (HR 0.70, 95% CI 0.55–0.90) compared to those with no family history (HR 0.90, 95% CI 0.73–1.13, P interaction < 0.01 for confirmed AD).

When analyses including the observational period were repeated, censoring those who reported they were taking >800 IU/day of supplemental vitamin D in the observational period (n = 6,115 [28.3%]), the adjusted HR for overall confirmed AD associated with initial randomization to vitamin D was 0.95 (95% CI 0.78–1.15). After similarly censoring 4,944 individuals (22.9%) who reported that they were taking n-3 fatty acids during the observational period, the HR for confirmed AD with initial randomization to n-3 fatty acids was 0.89 (95% CI 0.73–1.08).

DISCUSSION

In this observational extension after the five-year VITAL randomized trial, we observed 21,592 participants (83.5%) who had been initially randomized to 2000 IU/day of vitamin D, 1 g/day of marine n-3 fatty acids, both, or neither. After randomized trial completion, participants were observed for the next few years, completing the same questionnaires, including reporting of incident diseases of particular ADs. Thus, using the same methods as during the trial, we identified an additional 98 confirmed incident AD cases with diagnosis before December 31, 2017, the randomized portion of the trial. Although not all participants were observed in the observational period, when the two years after randomization were included, the protective effect of vitamin D had dissipated, and the HR returned to 0.98 (95% CI 0.83–1.17). For n-3 fatty acid supplementation, on the other hand, a similar magnitude effect was sustained, with an adjusted HR for AD risk of 0.83 (95% CI 0.70–0.99). Reanalysis of the results of the randomized portion of the trial with additional more recently confirmed cases revealed slightly higher HRs (less protective effects) during the trial20: an HR of 0.78 (95% CI 0.61–0.99) for vitamin D became 0.85 (95% CI 0.70–1.04), and an initially reported HR of 0.85 (95% CI 0.67–1.08) for n-3 fatty acids became 0.87 (95% CI 0.71–1.06).

For both supplements, similar results were found when probable incident ADs were included as end points along with the confirmed cases. Moreover, prior observations that during the randomized trial, patients in all intervention arms had lower risks of developing AD compared to patients in the placebo/placebo arm remained true, although with randomized supplement discontinuation, the effect of vitamin D diminished the most (HR for vitamin D alone 0.85 (95% CI 0.67–1.07) versus the placebo/placebo arm as a referent). We observed a sustained statistically significant reduction in AD risk for the active n-3 fatty acid/placebo vitamin D versus placebo/placebo arms, with an HR of 0.71 (95% CI 0.55–0.91). There was some heterogeneity of the effect of both supplements observed as well; in particular, vitamin D was associated with reduced risk of psoriasis, whereas n-3 fatty acids led to increased risk of psoriasis. Relatively small numbers of cases made it difficult to interpret differential effects in other individual ADs, although it also appeared that the effects on autoimmune thyroid disease incidence were disparate (elevated with vitamin D and reduced with n-3 fatty acids, but neither result was statistically significant). Although use of supplements and their formulations were potentially more difficult to assess during the observation period, in sensitivity analyses excluding the <9% who reported use of >800 IU/day of vitamin D or of n-3 fatty acids at years one or two of the observational period, the results were similar. We also continued to detect an interaction of vitamin D’s effect with BMI, with stronger protective effect among those with normal BMI, consistent with its known biologic mechanisms of action and results for other disease states as well and a stronger protective effect of n-3 fatty acids among those who reported a family history of AD at baseline.25–27

The contrasting sustained effects of vitamin D and n-3 fatty acids at two years after termination of the randomized trial may reflect differences in the underlying biologic mechanisms of action of these two supplements. Active 1,25 vitamin D binds to vitamin D receptor to regulate an array of genes involved in inflammation and immune responses.28,29 The bioavailability of active vitamin D is known to fluctuate, depending on intake, UV light exposure, and renal and hepatic function.30 Moreover, vitamin D levels decrease during systemic inflammation, with levels of circulating vitamin D inversely proportional to systemic inflammation from a wide variety of sources.31 High BMI and smoking are important factors associated with lower circulating vitamin D levels and responses to vitamin D supplementation.20,25,26 On the population level, vitamin D levels have been shown to be lower in the individuals with obesity, smokers, and people with darker skin tones.32,33 Within VITAL, we found that supplemental vitamin D led to a decrease in C-reactive protein levels, a biomarker of inflammation, after two years of supplementation (although this effect was not consistent over time).34,35 Thus, it could be that 2000 IU/day of oral supplementation acted to increase vitamin D reserves temporarily, with beneficial effects on AD risk, but with discontinuation, these reserves were depleted. Conversely, during the extension period participants were allowed to take outside supplements, unlike during the trial, and outside use of vitamin D in all arms may have diluted the effect of the initial randomized therapy. For this reason, we conducted secondary analyses excluding those participants who reported taking >800 IU/day of vitamin D in the extension period, but the results were similar.

There is also abundant literature concerning the beneficial effects of n-3 fatty acids on inflammation and immunity. In in vitro and animal studies, n-3 fatty acids block nuclear translocation of NF-κB.36 NF-κB is a recognized target for control of inflammation and autoimmunity, activating inflammatory cytokines, adhesion molecules, and cyclooxygenase genes.37–41 Omega-3 fatty acids inhibit production of inflammatory mediators prostaglandin E2 (PGE2) and leukotriene B4 (LTB4).42,43 Increased n-3 fatty acid intake also upregulates production of an array of bioactive lipid mediators, the specialized proresolving mediators (SPMs).44 Reports of the potential proresolving actions of SPMs continue to accumulate, including dampening production of inflammatory mediators such as PGE2, LTB4, interleukin-1β (IL-1β), and IL-6; promoting phagocytosis and efferocytosis; and increasing polarization of proresolving type 2 macrophages (M2).44,45 Several past trials have proven the beneficial effects of n-3 fatty acid supplementation on systemic inflammation and tender and swollen joints for patients with RA, although most of these trials tested higher doses of these supplements.46–48 It is thus interesting that marine n-3 fatty acid supplementation was here found to have an ongoing protective effect on AD incidence, an approximately 15% reduction, throughout the trial, which was sustained through the observational period, suggesting a more sustained effect of this supplement.

Limitations to this study include those of the initial trial, including that only these doses and preparations of the supplements were selected for testing (for the prevention of CVD and cancer in the originally funded trial) because they were thought to balance safety and potential efficacy, but higher doses may have had more potent effects.47 Although olive oil may have beneficial effects on inflammation as well, the small amount (less than one teaspoon) of olive oil in the n-3 fatty acid placebo is unlikely to have had a measurable biologic effect.17 During the observational extension, there was some participant dropout (87.9% of those alive with a good address were observed), and after completion of the randomized trial, those who were not initially randomized to a supplement may have started taking it, complicating interpretations. We did conduct secondary analyses excluding those who reported taking outside supplements during the observational period, and the results were similar. We have attempted to address the long latency period before diagnosis of many ADs by including in this report cases that were subsequently reported and verified to have been diagnosed during the randomized trial, as well as by extending case follow-up well beyond the end of the two-year observational extension. Nevertheless, the confirmation of incident AD and the exact timing of its onset and diagnosis is challenging via medical record review, as it can be in clinical practice. Additionally, the smaller number of cases during the observational period alone precluded detailed analyses in the observational period and of the individual ADs during this period.

Strong rationale supports biologic effects of both vitamin D and marine n-3 fatty acids for AD prevention, and the results of the first five years of VITAL supported the beneficial effects of both supplements for AD prevention. VITAL observational extension results suggest that vitamin D supplementation should be given on a continuous basis for long-term prevention of AD. The beneficial effects of n-3 fatty acids, however, may be prolonged for at least two years after discontinuation.

Acknowledgments

Supported by the NIH (grants R01-AR-059086, K24-AR-066109, U01-CA-138962, R01-CA-138962, and R01-AT-011729).

Footnotes

The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/art.42811.

REFERENCES

  • 1.Wang L, Wang FS, Gershwin ME. Human autoimmune diseases: a comprehensive update. J Intern Med 2015;278(4):369–395. [DOI] [PubMed] [Google Scholar]
  • 2.Jacobson DL, Gange SJ, Rose NR, et al. Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clin Immunol Immunopathol 1997;84(3):223–243. [DOI] [PubMed] [Google Scholar]
  • 3.Cotsapas C, Voight BF, Rossin E, et al. Pervasive sharing of genetic effects in autoimmune disease. PLoS Genet 2011;7(8):e1002254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Conrad N, Misra S, Verbakel JY, et al. Incidence, prevalence, and co-occurrence of autoimmune disorders over time and by age, sex, and socioeconomic status: a population-based cohort study of 22 million individuals in the UK. Lancet 2023;401(10391):1878–1890. [DOI] [PubMed] [Google Scholar]
  • 5.Roberts MH, Erdei E. Comparative United States autoimmune disease rates for 2010–2016 by sex, geographic region, and race. Autoimmun Rev 2020;19(1):102423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Al Maini M, Adelowo F, Al Saleh J, et al. The global challenges and opportunities in the practice of rheumatology: white paper by the World Forum on Rheumatic and Musculoskeletal Diseases. Clin Rheumatol 2015;34(5):819–829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sparks JA, Costenbader KH. Genetics, environment, and gene-environment interactions in the development of systemic rheumatic diseases. Rheum Dis Clin North Am 2014;40(4):637–657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Serhan CN, Clish CB, Brannon J, et al. Novel functional sets of lipid-derived mediators with antiinflammatory actions generated from omega-3 fatty acids via cyclooxygenase 2-nonsteroidal antiinflammatory drugs and transcellular processing. J Exp Med 2000;192(8):1197–1204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Calder PC. Marine omega-3 fatty acids and inflammatory processes: effects, mechanisms and clinical relevance. Biochim Biophys Acta 2015;1851(4):469–484. [DOI] [PubMed] [Google Scholar]
  • 10.Yan Y, Jiang W, Spinetti T, et al. Omega-3 fatty acids prevent inflammation and metabolic disorder through inhibition of NLRP3 inflammasome activation. Immunity 2013;38(6):1154–1163. [DOI] [PubMed] [Google Scholar]
  • 11.Yamamoto E, Jørgensen TN. Immunological effects of vitamin D and their relations to autoimmunity. J Autoimmun 2019;100:7–16. [DOI] [PubMed] [Google Scholar]
  • 12.Murdaca G, Tonacci A, Negrini S, et al. Emerging role of vitamin D in autoimmune diseases: an update on evidence and therapeutic implications. Autoimmun Rev 2019;18(9):102350. [DOI] [PubMed] [Google Scholar]
  • 13.Fisher SA, Rahimzadeh M, Brierley C, et al. The role of vitamin D in increasing circulating T regulatory cell numbers and modulating T regulatory cell phenotypes in patients with inflammatory disease or in healthy volunteers: a systematic review. PLoS One 2019;14(9):e0222313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Ding Y, Liao W, He XJ, et al. Effects of 1,25(OH)2D3 and vitamin D receptor on peripheral CD 4+/CD 8+ double-positive T lymphocytes in a mouse model of systemic lupus erythematosus. J Cell Mol Med 2017;21(5):975–985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Dankers W, Colin EM, van Hamburg JP, et al. Vitamin D in autoimmunity: molecular mechanisms and therapeutic potential. Front Immunol 2017;7:697. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kriegel MA, Manson JE, Costenbader KH. Does vitamin D affect risk of developing autoimmune disease?: a systematic review. Semin Arthritis Rheum 2011;40(6):512–531.e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kremer JM, Lawrence DA, Jubiz W, et al. Dietary fish oil and olive oil supplementation in patients with rheumatoid arthritis. Clinical and immunologic effects. Arthritis Rheum 1990;33(6):810–820. [DOI] [PubMed] [Google Scholar]
  • 18.Geusens P, Wouters C, Nijs J, et al. Long-term effect of omega-3 fatty acid supplementation in active rheumatoid arthritis. A 12-month, double-blind, controlled study. Arthritis Rheum 1994;37(6):824–829. [DOI] [PubMed] [Google Scholar]
  • 19.Sperling RI, Weinblatt M, Robin JL, et al. Effects of dietary supplementation with marine fish oil on leukocyte lipid mediator generation and function in rheumatoid arthritis. Arthritis Rheum 1987;30(9):988–997. [DOI] [PubMed] [Google Scholar]
  • 20.Hahn J, Cook NR, Alexander EK, et al. Vitamin D and marine omega 3 fatty acid supplementation and incident autoimmune disease: VITAL randomized controlled trial. BMJ 2022;376:e066452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Manson JE, Cook NR, Lee IM, et al. Marine n-3 fatty acids and prevention of cardiovascular disease and cancer. N Engl J Med 2019;380(1):23–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Manson JE, Cook NR, Lee IM, et al. Vitamin D supplements and prevention of cancer and cardiovascular disease. N Engl J Med 2019;380(1):33–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Bassuk SS, Manson JE, Lee IM, et al. Baseline characteristics of participants in the VITamin D and OmegA-3 TriaL (VITAL). Contemp Clin Trials 2016;47:235–243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Althouse AD, Below JE, Claggett BL, et al. Recommendations for statistical reporting in cardiovascular medicine: a special report from the American Heart Association. Circulation 2021;144(4):e70–e91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ceglia L, Nelson J, Ware J, et al. Association between body weight and composition and plasma 25-hydroxyvitamin D level in the Diabetes Prevention Program. Eur J Nutr 2017;56(1):161–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Young KA, Engelman CD, Langefeld CD, et al. Association of plasma vitamin D levels with adiposity in Hispanic and African Americans. J Clin Endocrinol Metab 2009;94(9):3306–3313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gangloff A, Bergeron J, Pelletier-Beaumont E, et al. Effect of adipose tissue volume loss on circulating 25-hydroxyvitamin D levels: results from a 1-year lifestyle intervention in viscerally obese men. Int J Obes (Lond) 2015;39(11):1638–1643. [DOI] [PubMed] [Google Scholar]
  • 28.Arnson Y, Amital H, Shoenfeld Y. Vitamin D and autoimmunity: new aetiological and therapeutic considerations. Ann Rheum Dis 2007;66(9):1137–1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Cantorna MT, Mahon BD. Mounting evidence for vitamin D as an environmental factor affecting autoimmune disease prevalence. Exp Biol Med (Maywood) 2004;229(11):1136–1142. [DOI] [PubMed] [Google Scholar]
  • 30.Bikle DD. Vitamin D metabolism, mechanism of action, and clinical applications. Chem Biol 2014;21(3):319–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Quraishi SA, Camargo CA Jr. Vitamin D in acute stress and critical illness. Curr Opin Clin Nutr Metab Care 2012;15(6):625–634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Brot C, Jørgensen NR, Sorensen OH. The influence of smoking on vitamin D status and calcium metabolism. Eur J Clin Nutr 1999;53(12):920–926. [DOI] [PubMed] [Google Scholar]
  • 33.Freedman BI, Register TC. Effect of race and genetics on vitamin D metabolism, bone and vascular health. Nat Rev Nephrol 2012;8(8):459–466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Costenbader KH, MacFarlane LA, Lee IM, et al. Effects of one year of vitamin D and marine omega-3 fatty acid supplementation on biomarkers of systemic inflammation in older US adults. Clin Chem 2019;65(12):1508–1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Dong Y, Zhu H, Chen L, et al. Effects of vitamin D3 and marine omega-3 fatty acids supplementation on biomarkers of systemic inflammation: 4-year findings from the VITAL randomized trial. Nutrients 2022;14(24):5307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Chitranjali T, Anoop Chandran P, Muraleedhara Kurup G. Omega-3 fatty acid concentrate from Dunaliella salina possesses antiinflammatory properties including blockade of NF-κB nuclear translocation. Immunopharmacol Immunotoxicol 2015;37(1):81–89. [DOI] [PubMed] [Google Scholar]
  • 37.Kliewer SA, Sundseth SS, Jones SA, et al. Fatty acids and eicosanoids regulate gene expression through direct interactions with peroxisome proliferator-activated receptors α and γ. Proc Natl Acad Sci U S A 1997;94(9):4318–4323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Forman BM, Chen J, Evans RM. Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are ligands for peroxisome proliferator-activated receptors α and δ. Proc Natl Acad Sci U S A 1997;94(9):4312–4317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Novak TE, Babcock TA, Jho DH, et al. NF-κB inhibition by ω−3 fatty acids modulates LPS-stimulated macrophage TNF-α transcription. Am J Physiol Lung Cell Mol Physiol 2003;284(1):L84–L89. [DOI] [PubMed] [Google Scholar]
  • 40.Zhao Y, Joshi-Barve S, Barve S, et al. Eicosapentaenoic acid prevents LPS-induced TNF-α expression by preventing NF-κB activation. J Am Coll Nutr 2004;23(1):71–78. [DOI] [PubMed] [Google Scholar]
  • 41.Housley WJ, Fernandez SD, Vera K, et al. Genetic variants associated with autoimmunity drive NFκB signaling and responses to inflammatory stimuli. Sci Transl Med 2015;7(291):291ra93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Simopoulos AP. Evolutionary aspects of diet, the omega-6/omega-3 ratio and genetic variation: nutritional implications for chronic diseases. Biomed Pharmacother 2006;60(9):502–507. [DOI] [PubMed] [Google Scholar]
  • 43.Larsson SC, Kumlin M, Ingelman-Sundberg M, et al. Dietary long-chain n-3 fatty acids for the prevention of cancer: a review of potential mechanisms. Am J Clin Nutr 2004;79(6):935–945. [DOI] [PubMed] [Google Scholar]
  • 44.Serhan CN, Chiang N, Dalli J. The resolution code of acute inflammation: novel pro-resolving lipid mediators in resolution. Semin Immunol 2015;27(3):200–215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Chiang N, Serhan CN. Specialized pro-resolving mediator network: an update on production and actions. Essays Biochem 2020;64(3):443–462. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kremer JM, Lawrence DA, Petrillo GF, et al. Effects of high-dose fish oil on rheumatoid arthritis after stopping nonsteroidal antiinflammatory drugs. Clinical and immune correlates. Arthritis Rheum 1995;38(8):1107–1114. [DOI] [PubMed] [Google Scholar]
  • 47.Kremer JM, Jubiz W, Michalek A, et al. Fish-oil fatty acid supplementation in active rheumatoid arthritis. A double-blinded, controlled, crossover study. Ann Intern Med 1987;106(4):497–503. [DOI] [PubMed] [Google Scholar]
  • 48.Rajaei E, Mowla K, Ghorbani A, et al. The effect of omega-3 fatty acids in patients with active rheumatoid arthritis receiving DMARDs therapy: double-blind randomized controlled trial. Glob J Health Sci 2015;8(7):18–25. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES