The Effect of Vitamin D and Omega-3 Fatty Acid Supplementation on Pain Prevalence and Severity in Older Adults. A Large -Scale Ancillary Study of the VITamin D and OmegA-3 TriaL (VITAL)

Abstract

A diet supplemented with vitamin D and marine omega-3 fatty acids may prevent and treat painful disorders by promoting the resolution of inflammation. However, large, randomized placebo-controlled trials evaluating the effects of supplementation with omega-3 fatty acids and vitamin D on the presence and severity of pain are lacking.

VITamin D and OmegA-3 triaL – Pain (VITAL-Pain) is an ancillary study to the VITAL trial, a large randomized, double-blind, placebo-controlled trial of vitamin D (2000 IU/day) and omega-3 supplementation (1 g/day) over 5.3 years of median follow-up, among 25,871 older men and women. We assessed pain among those reaching the end of the VITAL intervention phase using questions from the 2012 National Health Interview Survey. We used ordinal logistic regression to test the effect of vitamin D and omega-3 fatty acids on the odds ratio (OR) and 95% confidence interval [CI] of reporting higher pain prevalence or severity.

Overall, 19,611 participants provided complete pain information at the end of the VITAL trial. The ORs for higher pain prevalence and/or severity for vitamin D or omega-3 supplementation versus placebo was 0.99 ([CI] 0.94–1.05) and 0.99 ([CI] 0.94–1.04), respectively. There was no interaction between the two active treatments.

Dietary supplementation with commonly used moderate doses of vitamin D and/or omega-3 fatty acids over a median of 5.3 years did not result in a lower prevalence or severity of pain in middle-aged and older U.S. adults.

1. Introduction

Chronic pain is a major health problem in the United States. A study by the National Institutes of Health’s National Center for Complementary and Integrative Health (NCCIH), looking at data from the 2012 National Health Interview Survey (NHIS), reported that nearly 50 million American adults have significant chronic pain.[23] The prevalence of chronic pain is particularly high in older adults with more than half reporting bothersome pain in the last month.[24] Patients with chronic pain need and use more health care services and suffer significant disability. Health economists report that yearly chronic pain costs in the United States are more than the combined costs of cancer, heart disease and diabetes.[6] Despite the high socio-economic burden, current treatments are often inadequate or have significant side effects, as evidenced by the prescription opioid crisis. There is an urgent and unmet need for novel approaches for the treatment and prevention of chronic pain.

Peripheral and central inflammation can influence excitatory and inhibitory pain signaling and play a significant role in hyperalgesia and central sensitization, which mechanistically contribute to development of chronic pain. [2,9] In observational studies, an anti-inflammatory diet, which has gained popularity in recent years, has been associated with a reduction in chronic pain.[27] It is hypothesized that a diet supplemented with marine omega-3 fatty acids, namely DHA and EPA (the principal polyunsaturated fatty acids of marine oils) and/or vitamin D may prevent and treat painful disorders by promoting the resolution of inflammation.[25]

DHA and EPA are omega-3 fatty acids which are the biosynthetic precursors to specialized pro-resolving lipid mediators (SPMs).[28] These include three separate biosynthetic pathways that lead to bioactive molecular families, the resolvins, protectins and maresins collectively termed SPMs.[28] SPMs have potent analgesic actions in preclinical inflammatory, neuropathic and postoperative pain models.[12, 17, 34] In vivo results from experimental animal models indicate that SPMs dampen abnormal pain via multiple mechanisms including reducing inflammation, glial activation and modulating spinal cord synaptic plasticity.[13] Smaller clinical trials also provide preliminary support for dietary omega-3 fatty acid supplementation to improve pain resolution, particularly at higher doses (>2.7g).[8]

Vitamin D3 supplementation may also improve pain resolution. Vitamin D3 decreases cytokine production, inhibits COX-2 expression and reduces PGE2 production.[11] PGE2 is an established lipid mediator contributing to inflammatory pain through actions on both the peripheral and central nervous system. High dose vitamin D3 supplementation (4000 IU daily) led to faster pain resolution and lower pro-inflammatory cytokine levels in a randomized, double-blind, placebo-controlled trial in 80 patients with musculoskeletal pain.[7]

Despite promising evidence to support the use of omega-3 fatty acid and vitamin D supplementation for improved pain resolution, large-scale, randomized, placebo-controlled trials evaluating the potential impact of omega-3 fatty acids and vitamin D on the incidence and severity of pain are lacking. In this paper, we report the results from the VITAL-Pain study, an ancillary study to the VITAL trial [20,21], as the first large-scale randomized trial testing the effects of supplementation with vitamin D and omega-3 fatty acids on pain prevalence and severity compared with placebo in the general U.S. population

2. Methods

2.1. Study Design

The VITAL–Pain study is an ancillary study to the VITAL trial, a large-scale, double-blind, randomized, placebo-controlled trial of vitamin D and marine omega-3 fatty acid supplements for the primary prevention of cancer and cardiovascular disease.[20,21] (https:clinicaltrial.gov/ NCT01169259). Briefly, 25,871 US men aged 50 years or older and women aged 55 years or older were recruited from throughout the United States by mail. Oversampling of African American/Black individuals was done with the goal to include at least 5000 Black participants. Inclusion criteria were no history of cancer or cardiovascular disease and willingness to limit the use of vitamin D supplementation to 800 IU per day and forego use of fish oil supplements throughout the trial. Exclusion criteria were a history of renal failure or dialysis, cirrhosis, hypercalcemia, or other serious conditions that could make study participation unsafe. After a 3-month placebo run-in phase, participants were randomized in a 2×2 factorial design to receive vitamin D3 (cholecalciferol 2000 IU/day) and/or marine omega-3 fatty acid (Omacor^® 1g/d, containing 460 mg of EPA + 380 mg of DHA) supplements or respective placebos (olive oil). Randomization took place between November 2011 to March 2014 and pill taking ended, as planned, on December 31, 2017. Randomization was computer generated within sex, race, and 5-year age groups in blocks of eight. The median follow-up was 5.3 years (range of 3.8 to 6.1 years). (Figure 1 and Flow Diagram VITAL Trial in Appendix 3) Baseline blood samples were obtained for measurement of serum 25-hydroxy vitamin D levels and plasma omega-3 index in >12,000 study participants included in the VITAL-Pain study. Among those, 1,301 provided a follow up sample at 1 year and 2,295 at year 4, that could be analyzed for the assessment of treatment compliance. Serum 25-hydroxyvitamin D assays were performed using liquid chromatography-tandem mass spectrometry by Quest Diagnostics. Measurements of the plasma omega-3 index (EPA + DHA as a percentage of total fatty acids) [10] were also performed by Quest Diagnostics using liquid chromatography- tandem mass spectrometry. These assays were performed at no cost to the trial. Quest had no role in the design of the trial, accrual of the data (other than the assays), analysis of the data (other than the assay standards), or preparation of the manuscript. . The trial was approved by the institutional review board of Partners Healthcare/Brigham and Women’s Hospital, Boston, Massachusetts and was monitored by an external data and safety monitoring board. Investigational New Drug Approval was obtained from the Food and Drug Administration. Written informed consent was obtained from all participants before trial enrollment.

Figure 1: The VITamin D and OmegA-3 TriaL (VITAL)-Pain design Flow Diagram.

The parent VITamin D and OmegA-3 TriaL (VITAL) included 25,871 US men and women, without a history of cardiovascular disease or cancer. Participants were randomized in a two-by-two factorial design to receive 2000 IU Vitamin D daily or placebo and 1g omega-3 fatty acids daily or placebo, over a median period of 5.3 years. Of the 24,817 participants that survived at the end of the trial, 19,611 completed the pain questionnaire.

2.2. Endpoints for the VITAL-Pain ancillary study

Pain was assessed via a questionnaire sent to participants reaching the end of the VITAL trial intervention period on December 31, 2017, using questions from the NHIS (National Health Interview Survey) to assess the presence, frequency and bothersomeness (intensity) of pain experienced in the previous three months.[23] The questions were developed and validated nationally and internationally by the Washington Group on Disability Statistics.[18,19,31] Specifically, participants were asked two questions for a graded assessment of pain severity: 1) In the last three months, how often did you have pain? (never, some days, most days, or every day) and 2) Thinking about the last time you had pain, how much pain did you have? (a little, between a little and a lot, or a lot). For individuals reporting pain on at least some days, we created 4 discrete categories of increasingly severe pain: 0 = pain free, 1 = a little pain some days, 2 = a little pain most or every day, or between a little and lot of pain some days, and 3 = a lot of pain some days, or between a little and a lot of pain most or every day, and 4 = a lot of pain most or every day. (Figure 2) These pain categories were used as the primary outcome for this study, for whom 19,611 (79%) VITAL participants provided responses of 24,817 known to be alive at year 5.

Figure 2: Pain Prevalence and Severity Categories.

Pain among survivors at the end of the VITAL trial using questions from the 2012 National Health Interview Survey using the last 3 months as the reference period: 1) In the last three months, how often did you have pain? (never, some days, most days, or every day). And 2) Thinking about the last time you had pain, how much pain did you have? (a little, between a little and a lot, or a lot). These questions allow a graded assessment of pain prevalence and severity, with 4 discrete categories.

2.3. Statistical Analysis

We expected the distribution of reported pain prevalence and severity categories among VITAL-Pain participants to be at least similar or higher compared to those from the 2012 National Health Interview Survey (44.3% pain free, 23.9% category 1, 14.2% category 2, 11.3% category 3, and 6.3% category 4), given that the median age in our study was higher.[23,24] Assuming that only one of the VITAL interventions would be effective at reducing pain prevalence and severity, 10,000 active-agent and 10,000 placebo participants for each intervention would provide 80% power at a two-sided alpha level of 0.05 to detect a 7% decrease in the odds of a higher pain prevalence and severity category (corresponding to an odds ratio of 0.93) at an average 5-year follow up using ordinal logistic regression. We assessed balance between randomized groups with respect to demographics and baseline characteristics via two-sample t-tests for continuous variables and chi-square tests for categorical variables. For our primary comparisons, a multivariable ordinal logistic regression model tested the effect of each intervention on pain prevalence and severity category at 5-year follow up while adjusting for age and sex. In exploratory analyses, we also tested whether the potential effect of vitamin D or omega-3 fatty acids supplementation on pain prevalence and severity category varied based on factors including baseline patient characteristics, compliance with trial supplements, or medical events. We built separate models to test each factor by expanding the primary analysis model to include the factor as a main effect and in an interaction term with randomized group. Confidence intervals for odds ratios from these subgroup analyses were not adjusted for multiple hypothesis testing, so results should be interpreted with caution. All analyses excluded patients who died by 5-year follow-up (n=1054). This exclusion was deemed to be reasonable since the incidence of death was low (4.1%) and there was lack of evidence that vitamin D or omega-3 fatty acids supplementation impacted 5-year all-cause mortality.[20,21] We calculated standardized differences to assess the magnitude and direction of differences in demographics and baseline characteristics between the participants who completed the 5-year pain questions (n=19611) vs. those who survived to 5 years but did not complete the pain questions (n= 5206) vs. those who died before 5-year follow-up (n= 1054). To account for missing pain responses from participants who survived to 5 years, sensitivity analyses for the primary comparisons of active vs. placebo vitamin D supplementation and active vs. placebo omega-3 fatty acids supplementation were performed using multiple imputation of missing data. Specifically, the method of multivariate imputation by chained equations (MICE)[26] was used to create 10 imputed datasets based on observed values of the variables included in the analysis model (pain prevalence and severity category, vitamin D randomization, omega-3 fatty acids randomization, age, and sex), as well as variables associated with missing variables and/or the probability of missingness (e.g., race, education level, and development of medical conditions such as rheumatoid arthritis and cancer over the course of the trial). Pain prevalence and severity category odds ratios and corresponding standard errors were calculated using ordinal logistic regression for each of the 10 imputed datasets, and then combined using Rubin’s rules to produce pooled odds ratios with 95% confidence intervals.[26] Three different missing data mechanism assumptions were tested in the multiple imputation analyses. The first assumed that pain prevalence and severity category was missing at random (MAR), such that any systematic differences between the reported and unreported outcome could be explained by observed data. The second and third missing data mechanisms tested assumed that the pain outcome was missing not at random (MNAR), such that even after accounting for observed data, patients not reporting the outcome had systematically higher or lower odds of being in a higher pain prevalence and severity category. To account for the possibility that the missingness was extremely nonignorable, the MNAR scenarios included testing of 8 times and 0.125 times the odds of a higher pain prevalence and severity category in patients who did versus did not report the outcome. All statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc, Cary, NC, USA).

3. Results

3.1. Patient Characteristics

In the 19,611 participants of the VITAL-Pain study, the mean (SD) age was 67.1 years and 50.1 % of study participants were female. The multi-ethnic study cohort included 75.7% non-Hispanic white and 16.3% black participants. The baseline characteristics were balanced between the different groups. (Table 1) Among VITAL-Pain participants, 33.3% were pain free, 37.1% had category 1, 15.7% category 2, 11.2% category 3 and 2.7% category 4 pain. Among survivors at the end of the trial, proportionally more black than white participants that did not complete the pain questionnaire. (Appendix 1) Women were significantly more likely than men to report a higher pain prevalence and severity category. (OR =1.77; 95 % confidence interval [CI] 1.68–1.87; p <0.0001). Current smokers (OR = 1.46; [CI] 1.30–1.64; p < 0.0001) and patients with a higher BMI (OR = 1.06; [CI] 1.06–1.07; p < 0.0001) were also more likely to report a higher pain prevalence and severity category.

Table 1.

Baseline Characteristics of Participants who answered 5-Year Pain Questionnaire, According to Randomized Assignment.

Characteristic	Total	Vitamin D		Marine n-3 fatty acids
	(N=19611)	Active (N=9,866)	Placebo (N=9,745)	Active (N=9826)	Placebo (N=9785)
Age (years), mean ± SD	67.1 ± 6.7	67.1 ± 6.7	67.1 ± 6.6	67.1 ± 6.6	67.2 ± 6.7
Female sex (%)	50.1	50.1	50.2	50.2	50.1
Race or ethnic group (%)
Non-Hispanic white	75.7	75.6	75.7	75.4	75.9
Black	16.3	16.3	16.4	16.7	15.9
Nonblack Hispanic	3.8	3.9	3.7	3.7	3.9
Asian or Pacific Islander	1.5	1.5	1.5	1.5	1.5
American Indian or Alaskan native	0.8	0.8	0.8	0.8	0.8
Other or unknown	1.9	1.9	1.9	1.8	2.0
Greater than high school education (%)	89.6	89.5	89.7	89.3	89.8
Body mass index (kg/m2), mean ± SD	27.8 ± 5.4	27.8 ± 5.4	27.8 ± 5.5	27.9 ± 5.5	27.7 ± 5.4
Current smoking (%)	5.8	5.7	5.8	5.8	5.8
Any alcohol use (%)	70.7	70.5	70.9	70.8	70.6
Current use of multivitamins (%)	46.2	46.7	45.7	45.8	45.5
Current use of supplemental vitamin D (%)	45.2	45.0	45.4	45.0	46.7
Current use of fish oil (%)	0.0	0.0	0.0	0.0	0.0
Dark-meat fish intake (servings/week), mean ± SD	1.0 ± 1.4	1.0 ± 1.3	1.0 ± 1.5	1.0 ± 1.5	1.0 ± 1.4
Other fish and seafood intake (servings/week), mean ± SD	1.0 ± 1.5	1.0 ± 1.5	1.0 ± 1.5	1.0 ± 1.4	1.1 ± 1.6
Baseline biomarkers, mean ± SD
Serum 25-hydroxyvitamin D	31.28 ± 9.83	31.38 ± 9.83	31.18 ± 9.82	31.32 ± 9.93	31.24 ± 9.73
Plasma n−3 index	2.63 ± 0.91	2.63 ± 0.91	2.62 ± 0.90	2.62 ± 0.91	2.63 ± 0.91
Aspirin use (%)	46.3	46.0	46.6	46.1	46.5
Corticosteroid use (%)	1.6	1.7	1.6	1.5	1.7
Headaches (%)	6.7	6.4	7.1	6.7	6.8
Knee pain (%)	6.3	6.3	6.4	6.2	6.5
Depression diagnosis (%)	20.0	19.6	20.4	19.8	20.2
Total annual household income (%)
Under $15,000	4.6	4.5	4.7	4.5	4.7
$15,000 to 49,999	27.8	27.9	27.7	27.6	28.0
$50,000 to 89,999	29.7	29.7	29.6	30.0	29.3
$90,000 to 120,000	17.7	17.4	18.1	17.8	17.7
over $120,000	20.2	20.5	19.9	20.0	20.3
Total MET (metabolic equivalent) hours from exercise and stairs/ week, median [Q1-Q3]	16.97 [5.50–33.07]	16.70 [5.44–32.87]	17.25 [5.71–33.31]	16.72 [5.61–32.87]	17.16 [5.44–33.37]

3.2. Vitamin D3

The mean self-reported compliance among VITAL-Pain participants, defined as taking at least two-thirds of the study pills, was 87% for the vitamin D group and 86% in the placebo group, over the 5 years of follow-up. Serum 25-hydroxyvitamin D levels were measured in 12,644 participants at baseline. The mean (SD) baseline serum total hydroxyvitamin D level was 31.3 (9.8) ng/mL and 1,426 participants (11.3%) had low baseline serum hydroxyvitamin D levels (< 20 ng/mL). Two subgroups of patients had repeat measurements at 1 year (n=1,301) and at 4 years (n=2,295) respectively. In the vitamin D₃ group, the mean (SD) serum 25-hydroxyvitamin D level increased 38.6% from 30.3 (10.4) ng/mL at baseline to 42.0 (9.8) ng/mL at 1 year and similarly, with 40% from 31.1 (9.7) ng/mL at baseline to 43.6 (10.6) ng/mL, at year 4. In the placebo group there was no significant change in serum 25- hydroxyvitamin D levels at both year 1 and year 4 from baseline.

Overall, there was no difference in reported pain prevalence or severity category at the end of the trial between participants assigned to vitamin D₃ or placebo. (OR: 0.99, 95% Confidence Interval [CI] 0.94–1.05). (Table 2) The multiple imputation sensitivity analyses including surviving patients who did not respond to the pain questions produced similar results under all missing data mechanism assumptions tested (Appendix 2). There was also no difference in use of non-steroidal anti-inflammatory drugs (NSAIDs) between the vitamin D₃ and the placebo group at the end of the trial (p=0.395). Subgroup analysis (Figure 3) indicated that age and baseline serum 25-hydroxyvitamin D-levels may modify the effect of vitamin D₃ supplementation on pain prevalence and severity, suggesting a trend towards worse pain prevalence and severity with vitamin D₃ supplementation in younger patients and patients with vitamin D deficiency (p, interactions = 0.023 and p = 0.047, respectively). In black participants, supplementation with vitamin D₃ resulted in a small increase in pain prevalence and severity (OR: 1.16 [CI] 1.02–1.32), while vitamin D₃ supplementation in non-black Hispanic participants was associated with less pain prevalence and severity (OR: 0.73 [CI] 0.56–0.96; p = 0.014 for the interaction) (Figure 3).

Table 2.

Odds Ratios and 95% Confidence Intervals for Pain Severity Category at 5-year Follow-up According to Randomized Assignment.

	Vitamin D			Marine n-3 fatty acids
Outcome	Active (N=9866)	Placebo (N=9745)	Odds ratio (95% CI)	Active (N=9826)	Placebo (N=9785)	Odds ratio (95% CI)
Pain severity category, n (%)
0	3304 (33.5)	3231 (33.2)	0.99 (0.94, 1.05)	3309 (33.7)	3226 (33.0)	0.99 (0.94, 1.04)
1	3637 (36.9)	3642 (37.4)		3606 (36.7)	3673 (37.5)
2	1555 (15.8)	1521 (15.6)		1513 (15.4)	1563 (16.0)
3	1106 (11.2)	1082 (11.1)		1117 (11.4)	1071 (11.0)
4	264 (2.7)	269 (2.8)		281 (2.9)	252 (2.6)

Figure 3:

Odds Ratios of Pain Severity Category According to Subgroup, Comparing Vitamin D with Placebo

There was no interaction between the two active treatments (vitamin D₃ and omega-3 fatty acids) in the two-by-two factorial trial design. (p = 0.57 for the interaction).

3.3. The Effect of Omega-3 Fatty Acids on Pain

Similar to the vitamin D₃ group, self-reported rates of adherence to the study medication (percentage of participants who took at least two thirds of the study pills) among VITAL -pain participants, was 87% in the omega-3 fatty acid group and 86% of the placebo group, averaged over the 5 years of follow-up. The plasma omega-3 index (EPA + DHA as a percent of total fatty acids) was analyzed by Quest Diagnostics, using liquid chromatography – tandem mass spectrometry, in 12,416 study participants at baseline and among those, 1,301 provided a follow up sample at 1 year and 2,295 at year 4 that could be analyzed. In the year 1 subgroup, the mean (SD) plasma omega-3 index increased by 56.4%, from 2.658% (1.0%) at baseline to 4.158% (1.3%) in the omega-3 fatty acid group. There was no significant change in plasma omega-3 index in the placebo group. Similarly, in the year 4 subgroup, the mean (SD) plasma omega-3 index increased by 58.8%, from 2.509% (0.8%) at baseline to 3.984% (1.1%) at year 4 in the omega-3 fatty acid group and remained unchanged in the placebo group.

There was no difference in the odds of reporting a more severe pain prevalence or severity category at the end of the trial between the omega-3 fatty acid group and placebo group. (OR: 0.99 [CI] 0.94 – 1.04). (Table 2) The multiple imputation sensitivity analyses including surviving patients who did not respond to the pain question led to identical results under all missing data mechanism assumptions tested (Appendix 2). There was also no difference in use of non-steroidal anti-inflammatory drugs (NSAIDs) between the omega-3 fatty acid (Omacor^® 1g/day) and the placebo group at the end of the trial (p=0.35).

Among other effect measure modifiers considered, there was a significant interaction between omega-3 fatty acid supplementation and BMI on pain (p=0.006 for the interaction) (Figure 4). Patients with a BMI < 25 were slightly more likely to report a lower pain prevalence and severity category with omega-3 fatty acid supplementation compared to placebo (OR: 0.9 [CI] 0.82 – 0.99). However, this effect disappeared among overweight patients (OR: 0.96 [CI] 0.89 – 1.05) and reversed in obese patients (BMI ≥ 30) (OR: 1.12 [CI] 1.02–1.24) (Figure 4).

Figure 4:

Odds Ratios of Pain Severity Category According to Subgroup, Comparing Omega-3 with Placebo

4. Discussion

In this large, blinded, randomized placebo-controlled trial in initially healthy, older Americans, daily supplementation with vitamin D₃ (2000 IU/day) and/or omega-3 fatty acids (Omacor at 1 g/day) over a median period of 5.3 years did not lead to lower pain prevalence and/or severity at trial closeout.

For vitamin D supplementation, these findings contrast with previously published findings. A meta-analysis of 19 smaller randomized, placebo-controlled trials, including 3,436 participants, reported a significantly greater reduction of pain in the vitamin D group compared to the placebo group. However, the pain reduction with vitamin D was greater in patients with pre-existing pain.[32] In our study, participants were not selected for disease risk and pain prevalence at the beginning of the trial may have been low, which could account for the different findings. In addition, the meta-analysis included studies with a wide variety of vitamin D dosing and it is unclear if higher doses of vitamin D supplementation may have a different effect on pain.[23] Similar to our study, the secondary analysis of the Vitamin D Assessment (ViDA) study, looking at the effects of 100,000 IU monthly vitamin D supplementation on pain and analgesic use did not find a difference in pain scores between the vitamin D and the placebo groups.[33] However, ViDA did observe a reduction in NSAID use in the vitamin D group versus placebo group in patients with baseline vitamin D deficiency [33], a finding not replicated in VITAL-Pain. In fact, our data suggest a trend towards worse pain prevalence and severity with vitamin D supplementation in patients with vitamin D deficiency. And recently published results from a sub cohort from the VITAL parent trial, looking at the effect of vitamin D and omega-3 fatty acids on biomarkers of systemic inflammation found that among participants receiving vitamin D, high sensitivity C-reactive protein (hsCRP), increased significantly in those with low (<20 ng/mL) but not in those with higher baseline serum 25-hydroxyvitamin D levels. A similar, but nonsignificant increase was also noted with interleukin 6 (IL-6) among patients with low baseline 25-hydroxyvitamin D levels.[4] Both hsCRP and IL-6 have been associated with more severe pain in several painful conditions.[15,30] While our findings of a potential negative effect of vitamin D supplementation on pain prevalence and severity in patients with vitamin D deficiency did not reach significance, our results highlight the need to further investigate the effect of vitamin D supplementation on pain in patients based upon serum 25-hydroxyvitamin D levels.

We also found potential racial differences in the effect of vitamin D on pain prevalence and severity. Black participants who received vitamin D supplementation were more likely to report higher pain prevalence and pain severity category compared to those who received placebo. Vitamin D deficiency is significantly more prevalent in blacks versus whites, and while vitamin D deficiency is associated with lower bone density and increased fracture risk in whites, serum 25-hydroxyvitamin D levels do not correlate with the same health outcomes in blacks.[1] In fact, a nested case-control study from the Women’s Health Initiative Observational Study (WHI-OS) found that in blacks, in contrast to whites, higher levels (≥ 20 ng/mL) of serum 25-hydroxyvitamin D were associated with a higher risk of fracture.[3] Our study highlights the importance of understanding racial differences of vitamin D supplementation on pain.

Contrary to previously published smaller trials [5, 22], our study also found no effect of 1 g/day of omega-3 fatty acids on pain prevalence and severity or NSAID use at 5 years follow-up in older American adults. Like the findings for vitamin D, it is possible that the difference between our study and earlier published positive studies, could, at least in part, be explained by a relative low prevalence of pre-existing pain in the present cohort. Another possible contributing factor is that the dose of omega-3 fatty acids (1g per day containing 840 mg of omega-3 fatty acids, including 460 mg of eicosapentaenoic acid (EPA) and 380 mg of docosahexaenoic acid (DHA)), based on recommendations by the American Heart Association for cardioprotection [14,20], may be too low for the treatment and/or prevention of pain. A meta-analysis of 17 trials suggested that 2.7 g/day omega-3 fatty acids may be required to reduce patient reported pain.[16]

In our study we found through exploratory further analyses that patients with a low BMI <25 were more likely to benefit from omega-3 fatty acids than patients with a higher BMI. If the effect of omega-3 fatty acids on pain is dose and weight dependent, it is possible that patients with a higher BMI would require higher doses, which could explain the difference in the effect of omega-3 fatty acids between participants with a low BMI compared to participants with a high BMI. Future studies evaluating different doses of omega-3 fatty acids on pain are needed to address this question.

Strengths and Limitations

Our study has several strengths, including its randomized, placebo-controlled trial design, large sample size, diverse multi-ethnic population, and high rate of adherence to the study medication. Response rate to the pain questionnaire was also high, with 76% of patients who started the VITAL trial completing all pain questions at the end of the trial. Our study also has several limitations. Pain prevalence and severity were not assessed at baseline. Given the randomized trial design and large sample size, baseline comparability between treatment groups on this measure was likely achieved. With regards to Vitamin D, stratification by region and sun exposure was not performed. For omega-3 fatty acids, a subgroup analysis for fish consumption was not performed, however dietary fish intake was generally balanced between groups at baseline. Despite a high pain questionnaire response rate, there is the possibility that the presence of missing outcome data introduced bias. However, we performed three multiple imputation sensitivity analyses to test the robustness of results to different missing data assumptions, and found consistent results across all analyses. Another limitation is that no adjustments were made for multiple comparisons and, therefore, subgroup analysis findings should be interpreted cautiously and require confirmation in future trials.

5. Conclusions

In conclusion, dietary supplementation with commonly used moderate doses of vitamin D and/or omega-3 fatty acids over a median period of 5.3 years did not result in a lower prevalence or severity of pain in a general U.S. population of older adults.

Data Sharing Statement:

We will make the data and associated documentation available to users only under a data-sharing agreement. Dataset(s) will be de-identified prior to release for sharing. Details on the availability of the study data to other investigators will be on our study website at https://vitalstudy.org/.