Background

The relationship between long-chain omega-3 (LCn3), alpha-linolenic acid (ALA), omega-6 and total polyunsaturated fatty acid (PUFA) intakes and cancer risk is unclear.

Methods

We searched Medline, Embase, CENTRAL and trials registries for RCTs comparing higher with lower LCn3, ALA, omega-6 and/or total PUFA, that assessed cancers over ≥12 months. Random-effects meta-analyses, sensitivity analyses, subgrouping, risk of bias and GRADE were used.

Results

We included 47 RCTs (108,194 participants). Increasing LCn3 has little or no effect on cancer diagnosis (RR1.02, 95% CI 0.98–1.07), cancer death (RR0.97, 95% CI 0.90–1.06) or breast cancer diagnosis (RR1.03, 95% CI 0.89–1.20); increasing ALA has little or no effect on cancer death (all high/moderate-quality evidence). Increasing LCn3 (NNTH 334, RR1.10, 95% CI 0.97–1.24) and ALA (NNTH 334, RR1.30, 95% CI 0.72–2.32) may slightly increase prostate cancer risk; increasing total PUFA may slightly increase risk of cancer diagnosis (NNTH 125, RR1.19, 95% CI 0.99–1.42) and cancer death (NNTH 500, RR1.10, 95% CI 0.48–2.49) but total PUFA doses were very high in some trials.

Conclusions

The most extensive systematic review to assess the effects of increasing PUFAs on cancer risk found increasing total PUFA may very slightly increase cancer risk, offset by small protective effects on cardiovascular diseases.

Subject terms: Risk factors, Oncology

Inclusion criteria

We included randomised controlled trials (RCTs) that compared higher versus lower LCn3, ALA, omega-6 and/or total PUFA in adults aged at least 18 years, who were not pregnant or seriously ill. Participants could be free of cancer, at increased risk of cancer or with a previous cancer diagnosis, but were excluded if they were currently undergoing cancer treatment. The minimum study duration was 1 year (≥52 weeks) reflecting metabolic studies suggesting 6 months is the minimum supplementation duration required to equilibrate LCn3 into most body compartments,⁴¹ plus a further 6 months to influence cancer development.

Interventions could consist of foods, oral supplements (oil, capsules, or enriched foods) or advice, to increase or decrease omega-3, omega-6 and/or total PUFA intake, or achieve a change of ≥10% of baseline intake, comparing higher versus lower PUFA intake. Studies were excluded if they examined lifestyle or dietary interventions in addition to PUFA unless effects of the PUFA could be separated out.

Primary outcomes included:

• New diagnosis of breast cancer

• Breast cancer mortality

• New diagnosis of any cancer

• Any cancer mortality

Secondary outcomes included prostate cancer diagnosis and mortality (added post-hoc to complement prostate-specific antigen (PSA) data), markers of cancer risk (including breast density and PSA), body weight and measures of adiposity, quality of life, and dropouts.

Methods for identification of studies

We searched Cochrane CENTRAL, Medline and Embase to 27 April 2017, ClinicalTrials.com and WHO International Clinical Trials Registry Platform to September 2016 and reassessed all ongoing trials in December 2018. We checked included trials of relevant systematic reviews, and wrote to authors of included studies for additional trial data, creating a database of trials that randomised participants to increased omega-3, omega-6 or total PUFA compared to lower omega-3, omega-6 or total PUFA.³⁹ From this database, trials with duration of at least 12 months and data collected on any primary outcome were included in this review, even if study objectives were not primarily to assess effects on cancer, or those outcomes were not published.

Study inclusion, data extraction and risk of bias assessment (onto a specially developed form) were conducted independently in duplicate. We assessed Cochrane risk of bias domains⁴² plus risk from compliance problems and attention bias.³⁹ We considered supplementation trials to be at low summary risk of bias where randomisation, allocation concealment, blinding of participants, personnel and outcome assessors were judged adequate (all other trials were considered at moderate or high risk of bias). Dietary advice trials were at low summary risk of bias where randomisation, allocation concealment and blinding of outcome assessors were assessed adequate.³⁹

Data synthesis

Primary analyses assessed effects of total PUFA, omega-6, LCn3 and ALA using random-effects Mantel–Haenszel meta-analysis (as dietary interventions are heterogeneous by their nature⁴³) in Review Manager 5.3.⁴⁴ Treatment/control differences in outcomes were combined across studies using risk ratios (RR) or mean differences (MD), the at-risk population included only men for prostate cancer and women for breast cancer. Change from baseline in each arm with standard deviations were used for continuous outcomes where available, otherwise endpoint data were used.⁴³ Pre-specified sensitivity analyses included fixed effects analysis, limiting analysis to studies at low summary risk of bias, and limiting to trials randomising ≥100 participants. At the request of our funders, we added sensitivity analyses limiting to studies at low risk for compliance issues. At the request of referees, we added sensitivity analyses using Peto fixed-effects analysis (creating odds ratios), to ensure that our findings are robust to analysis methods despite the inclusion of trials with rare events. Pre-specified subgroup analyses were conducted for outcomes with ≥8 studies by intervention type, replacement, dose, duration, age, sex and cancer risk (normal cancer risk/ increased risk/ previous cancer).³⁹ We planned to sub-group also by medications used, baseline omega-3, omega-6 or total PUFA intake, pre- or post-menopausal, BMI, ethnicity and omega-3/omega-6 ratio, however, this information was not available in most trials, so subgrouping was not attempted. Heterogeneity was assessed using I² and considered important where >50%.⁴⁵ Small study bias was assessed using funnel plots where at least 10 trials were included in a meta-analysis.⁴⁶ Data from individual participants were only counted once in any meta-analytical pooling.

Effect sizes were interpreted as agreed with the WHO Nutrition Guidance Expert Advisory Group (NUGAG) Subgroup on Diet and Health who commissioned this research.³⁹ A risk ratio less than 0.92 or greater than 1.08 was considered a potentially relevant clinical effect (RR 0.92–1.08 was considered “little or no effect”), while a mean difference between arms of at least 10% of baseline was required for a relevant clinical effect for markers. Where we found a suggested effect we quantified the effect using number needed to treat for an additional benefit (NNTB) or number needed to treat to cause an additional harm (NNTH).⁴⁷ Outcome data were interpreted using GRADE assessment, drafted by LH then discussed and agreed with WHO NUGAG.³⁹ Where sensitivity analyses using Mantel–Haenszel or Peto fixed-effects analyses were not consistent with the main random-effects analysis we downgraded (for inconsistency), and where sensitivity analyses including only trials at low summary risk of bias, or only trials with good compliance differed from the main analysis we downgraded (for risk of bias). Where GRADE suggested data of very low-quality we did not interpret effect sizes. Where data were of low-quality, we used the term “may”, moderate-quality evidence warranted “probably” in describing effects. Summary of findings’ (GRADE) tables show effects on all cancers, breast cancer and prostate cancer diagnoses and deaths (marker evidence strengthened or weakened findings for relevant cancers).

Effects of increasing long-chain omega-3

Increasing LCn3 has little or no effect on risk of diagnosis of any cancer (high-quality evidence) and probably has little or no effect on risk of cancer death (moderate-quality evidence). We meta-analysed 27 trials (113,557 participants, 7339 diagnoses, mean duration 32 months, mean dose 1.7 g/d LCn3) assessing effects of LCn3 on cancer diagnosis (RR 1.02, 95% CI 0.98 to 1.07, I² 0%, Fig. 1). This lack of effect was not altered in any sensitivity analysis. There was no suggestion of heterogeneity between trials and the funnel plot did not suggest small study bias (Supplementary Fig. 3). Subgrouping did not suggest effect differences by duration, dose, nutrients replaced, intervention type, age, sex or baseline cancer risk. Eighteen trials (99,336 participants) provided data on 2277 cancer deaths (RR 0.97, 95% CI 0.90 to 1.06, I² 0%, Fig. 2). This lack of effect did not alter in sensitivity analyses or subgrouping and there was no suggestion of small study bias (Supplementary Fig. 4) or heterogeneity.

Increasing LCn3 probably has little or no effect on risk of breast cancer diagnosis (moderate-quality evidence), but effects on breast cancer deaths are unclear as the evidence is of very low-quality (two deaths). We meta-analysed 12 trials (44,295 women, 661 diagnoses, mean duration 48 months, mean dose 1.9 g/d LCn3) assessing effects of LCn3 on breast cancer diagnosis (RR 1.03, 95% CI 0.89–1.20, I² 0%, Fig. 3). This lack of effect did not alter in sensitivity analyses, there was no suggestion of small study bias or heterogeneity. Subgrouping did not suggest differences in effect by duration, dose, replacement, intervention type, age, sex or cancer risk. Breast density data were consistent with little or no effect.

Increasing LCn3 may slightly increase prostate cancer risk (low-quality evidence), but effects on prostate cancer death were unclear (the evidence was very low-quality, five deaths). Seven trials (38,525 men, mean duration 51 months, mean dose 1.2 g/d LCn3) reported on 1021 prostate cancer diagnoses, finding higher risk of prostate cancer in men with increased LCn3 (RR 1.10, 95% CI 0.97–1.24, I² 0%, NNTH 334, Fig. 4). This slight increase in prostate cancer risk was stable to all sensitivity analyses. However, the suggestion of harm was contradicted by PSA data reported in a single large trial (25% reduction, MD −0.13 ng/ml, 95% CI −0.25 to 0.01, 1622 participants). Raised PSA was reported in 12 of 62 participants in another trial (RR 0.47, 95% CI 0.16–1.40), also contradicting the suggested LCn3 harms.

Effects of increasing ALA

Increasing ALA probably has little or no effect on risk of cancer death (moderate-quality evidence) and may slightly increase the risk of prostate cancer diagnosis (low-quality evidence). Data on any cancer diagnoses, breast cancer diagnoses, breast or prostate cancer deaths and breast density were too limited to provide useful information, so effects were unclear.

Two trials (5545 participants, durations 24 and 40 months, doses 2 and 5 g/d ALA) provided data on 123 cancer deaths and meta-analysis suggested little or no effect (RR 1.05, 95% CI 0.74–1.49, I² 0%), which did not alter in sensitivity analyses. The same two trials reported 46 prostate cancer diagnoses in 4010 male participants (RR 1.30, 95% CI 0.72–2.32, NNTH 334, I² 0%). This increased risk was consistent across all sensitivity analyses and supported by a rise in PSA in those taking more ALA in the single large trial (rise of 23% from baseline, MD 0.10 ng/ml, 95% CI −0.03 to 0.23).

Effects of increasing omega-6

Evidence for effects of omega-6 on all cancer outcomes was unclear and of very low-quality (see SupplementaryMaterials).

Effects of increasing total PUFA

Increasing total PUFA may slightly increase risk of diagnosis of any cancer and cancer death (both low-quality evidence). No trials reported breast cancer deaths or breast density, prostate cancer deaths or PSA and effects on breast and prostate cancer diagnoses were unclear (evidence of very low-quality).

Eight trials (9428 participants, 436 diagnoses, mean duration 39 months, doses ranging from 0.8% of energy to 38% of energy from PUFA) assessed effects of increasing total PUFA on cancer diagnosis (RR 1.19, 95% CI 0.99–1.42, NNTH 125, I² 0%), consistent across sensitivity analysis. While the funnel plot suggested small trials with higher risk ratios may be missing (Supplementary Fig. 10), if such trials were included the risk ratio would increase further. Subgrouping did not suggest important differences due to study duration, PUFA dose, age, sex, baseline cancer risk or replacement but data were limited for assessment of subgroup effects. Four trials (3407 participants, 73 deaths, mean duration 39 months, median dose 7%E from PUFA) reported on cancer deaths (RR 1.10, 95% CI 0.48–2.49, NNTH 500, I² 37%), consistent across all sensitivity analyses.

Secondary outcomes

Effects on body weight and measures of adiposity are reported as primary outcomes in other reviews in this series.^33–35 No trials reported on quality of life; dropouts are reported in supplementary materials.

Strengths and limitations

Strengths of this systematic review include its large size (47 long-term RCTs including more than 108,000 randomised participants worldwide). Creation of a dataset of RCTs randomising to higher or lower PUFA intakes, regardless of primary and reported outcomes, allowed the inclusion of trials and data that would otherwise have been missed or remained unpublished. This allowed us to include many large and long-term RCTs of PUFAs in populations recruited for health problems other than cancer risk, so allowing us to assess effects of increasing PUFA on diagnosis of cancers in low-risk populations. As meta-analysis of trials with rare events can produce different effect sizes when using different analytical methods we ran sensitivity analyses using Mantel–Haenszel and Peto fixed-effects meta-analyses and compared the results with the main random-effects Mantel–Haenszel analysis.^48–51 This ensures that review results are robust to analysis methods.

Review limitations include limited available data on effects of increasing ALA, omega-6 and total PUFA. It was notable that doses of total PUFA were highly variable (from 0.8% of energy to almost 38% of energy from total PUFA in trials providing cancer diagnosis data), but the small number of trials made subgrouping by dose uninformative (Supplementary Figure 11). LCn3 results resulted from meta-analyses of mainly supplementary trials, so effects of increasing oily fish consumption are unclear. As poorly concealed allocation is associated with a 40% greater effect size⁵² and lack of blinding with additional bias^53,54 caution is needed in interpreting small effects in weaker trials. As prostate cancer was not a primary outcome in this review, we did not ask trialists for additional prostate cancer data, which means that more information on prostate cancer may be available from existing trials.

What does this study add?

Our review concurs with a systematic review of observational data²³ and two including fewer trials (10 and 19 RCTs to our 34) suggesting LCn3 intake is not associated with total cancer risk.^15,16 Two previous systematic reviews of trials and observational data suggested there were inadequate data to determine whether LCn3 intake was associated with prostate cancer incidence or progression.^26,29 A systematic review of cohort studies assessing relationships between omega-3 and eleven types of cancer found mixed results, including cohorts suggesting both statistically significantly increased and decreased risk for prostate cancer.²³ This review is new in suggesting that actively increasing dietary total PUFA may slightly increase the risk of both cancer diagnosis and cancer mortality. A recent systematic review of observational studies suggested no association between total polyunsaturated fat intake and breast cancer risk,²² but as higher PUFA intake is associated with healthier lifestyles small harms may be difficult to spot in observational studies due to confounding. RCT data are insufficient to corroborate or contradict two nested case-control trials suggesting that higher PUFA intake correlates to higher prostate cancer risk.^24,25

The small harms resulting from increased LCn3, ALA and total PUFAs need to be balanced against potential gains from the other major cause of morbidity and mortality, cardiovascular disease (Table 1). For example, this review suggests that increasing LCn3 intake may increase the risk of prostate cancer in men, such that 1000 men increasing their LCn3 intake would lead to three additional men being diagnosed with prostate cancer. In a sister review, meta-analysis including 25 RCTs and over 127,000 participants suggests that if 1000 people consume more LCn3 three will avoid death from coronary heart disease. Further analyses suggest that of the 1000 six will avoid a CHD event and one will avoid arrhythmia.⁵⁵ The balance appears similar for ALA—for every 1000 people increasing their ALA intake two will avoid a CVD event, eleven will avoid arrhythmia but three will be diagnosed with prostate cancer who would not otherwise have been diagnosed (Fig. 5 represents the harms and benefits visually as number of additional diagnoses incurred or avoided per 1000 people increasing their LCn3, ALA or total PUFA intake).⁵⁵ Increasing total PUFA in 1000 people appears to prevent five people dying from CHD, but two additional people will die from cancer. Sixteen people will be protected from CVD events, nineteen from CHD events, but eight more will be diagnosed with cancer (Fig. 5).³⁴ This suggests that small benefits and small harms of increasing LCn3 intake are likely to be partially balanced out across major sources of morbidity and mortality and indeed increasing LCn3, ALA, omega-6 and total PUFA appear to have little or no effect on all-cause mortality (Table 1).^34,35,55

While increasing LCn3 has little or no effect on risk of cancer diagnosis, breast cancer diagnosis or cancer death (moderate and high-quality evidence), trial evidence suggests that increasing omega-3 may slightly increase prostate cancer risk, and increasing total PUFA may slightly increase cancer risk (low-quality evidence), although this could result from very high intakes of PUFA in some trials. Considering both cancer and cardiovascular outcomes, overall health effects of increasing LCn3, ALA, omega-6 and total PUFA appear small.

The PUFAH group

Asmaa Abdelhamid², Sarah Ajabnoor², Faye Alabdulghafoor², Lena Alkhudairy³, Priti Biswas⁴, Julii Brainard², Charlene Bridges⁵, Tracey J Brown², Katherine Deane⁴, Daisy Donaldson², Sarah Hanson⁴, Lee Hooper², Oluseyi Florence Jimoh², Nicole Martin⁵, Alex O’Brien², Karen Rees³, Lena Alkhudairy³, Fujian Song², Gabrielle Thorpe⁴, Xia Wang², Lauren Winstanley²

• 1.Cancer Research UK. Worldwide Cancer Statisticshttps://www.cancerresearchuk.org/health-professional/cancer-statistics/worldwide-cancer: Cancer Research UK. Available from:https://www.cancerresearchuk.org/health-professional/cancer-statistics/worldwide-cancer#heading-One (2019)
• 2.Ritchie, H. How many people in the world die from cancer? Our World in Data. Available from:https://ourworldindata.org/how-many-people-in-the-world-die-from-cancer (2019).
• 3.Mariotto, A. B., Yabroff, K. R., Shao, Y., Feuer, E. J. & Brown, M. L. Projections of the cost of cancer care in the United States: 2010–2020.J. Natl Cancer Inst. 103,117–128 (2011) [DOI] [PMC free article] [PubMed]
• 4.World Health Organization (WHO). Global Status Report on Noncommunicable Diseases 2014. Geneva, Switzerland.http://apps.who.int/iris/bitstream/10665/148114/1/9789241564854_eng.pdf?ua=1: (WHO, 2015).
• 5.Barnes PM, Bloom B, Nahin RL. Complementary and alternative medicine use among adults and children: United States, 2007. Hyattsville, MD: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2008. [PubMed] [Google Scholar]
• 6.Gleissman H, Johnsen JI, Kogner P. Omega-3 fatty acids in cancer, the protectors of good and the killers of evil? Exp. Cell Res. 2010;316:1365–1373. doi: 10.1016/j.yexcr.2010.02.039. [DOI] [PubMed] [Google Scholar]
• 7.Diggle CP. In vitro studies on the relationship between polyunsaturated fatty acids and cancer: tumour or tissue specific effects? Prog. Lipid Res. 2002;41:240–253. doi: 10.1016/S0163-7827(01)00025-X. [DOI] [PubMed] [Google Scholar]
• 8.Hooper, L., Abdelhamid, A., Bunn, D., Brown, T., Summerbell, C. D. & Skeaff, D. M. Effects of total fat intake on body weight.Cochrane Database Syst Rev.8, CD011834. 10.1002/14651858.CD011834, (2015). [DOI] [PMC free article] [PubMed]
• 9.Hajeb P, Sloth JJ, Shakibazadeh S, Mahyudin N, Afsah-Hejri L. Toxic elements in food: Occurrence, binding, and reduction approaches. Compr. Rev. Food Sci. Food Saf. 2014;13:457–472. doi: 10.1111/1541-4337.12068. [DOI] [PubMed] [Google Scholar]
• 10.Malisch R, Kotz A. Dioxins and PCBs in feed and food—Review from European perspective. Sci. Total Environ. 2014;491-492:2–10. doi: 10.1016/j.scitotenv.2014.03.022. [DOI] [PubMed] [Google Scholar]
• 11.Matés JM, Segura JA, Alonso FJ, Márquez J. Roles of dioxins and heavy metals in cancer and neurological diseases using ROS-mediated mechanisms. Free Radic. Biol. Med. 2010;49:1328–1341. doi: 10.1016/j.freeradbiomed.2010.07.028. [DOI] [PubMed] [Google Scholar]
• 12.Mozaffarian D, Rimm EB. Fish intake, contaminants, and human health: evaluating the risks and the benefits. JAMA. 2006;296:1885–1899. doi: 10.1001/jama.296.15.1885. [DOI] [PubMed] [Google Scholar]
• 13.de Lorgeril M, Salen P, Martin J, Monjaud I, Boucher P, Mamelle N. Mediterranean dietary pattern in a randomized trial: prolonged survival and possible reduced cancer rate. Arch. Intern. Med. 1998;158:1181–1187. doi: 10.1001/archinte.158.11.1181. [DOI] [PubMed] [Google Scholar]
• 14.MacLean CH, Newberry SJ, Mojica WA, Khanna P, Issa AM, Suttorp MJ, et al. Effects of omega-3 fatty acids on cancer risk. a systematic review. JAMA. 2006;295:403–415. doi: 10.1001/jama.295.4.403. [DOI] [PubMed] [Google Scholar]
• 15.Hooper L, Thompson RL, Harrison RA, Summerbell CD, Ness AR, Moore HJ, et al. Risks and benefits of omega 3 fats for mortality, cardiovascular disease, and cancer: Systematic review. Br. Med. J. 2006;332:752–755. doi: 10.1136/bmj.38755.366331.2F. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Zhang Y-F, Gao H-F, Hou A-J, Zhou Y-H. Effect of omega-3 fatty acid supplementation on cancer incidence, non-vascular death, and total mortality: a meta-analysis of randomized controlled trials. BMC Public Health. 2014;14:1. doi: 10.1186/1471-2458-14-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 17.Trichopoulou A, Lagiou P, Kuper H, Trichopoulos D. Cancer and Mediterranean dietary traditions. Cancer Epidemiol. Biomarkers Prev. 2000;9:869–873. [PubMed] [Google Scholar]
• 18.Toledo E, Salas-Salvadó J, Donat-Vargas C, Buil-Cosiales P, Estruch R, Ros E, et al. Mediterranean diet and invasive breast cancer risk among women at high cardiovascular risk in the PREDIMED trial: a randomized clinical trial. JAMA Intern. Med. 2015;175:1752–1760. doi: 10.1001/jamainternmed.2015.4838. [DOI] [PubMed] [Google Scholar]
• 19.Brasky TM, Lampe JW, Potter JD, Patterson RE, White E. Specialty supplements and breast cancer risk in the vitamins and lifestyle (VITAL) cohort. Cancer Epidemiol. Biomark. amp; Prev. 2010;19:1696–1708. doi: 10.1158/1055-9965.EPI-10-0318. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Fabian CJ, Kimler BF, Hursting SD. Omega-3 fatty acids for breast cancer prevention and survivorship. Breast Cancer Res. 2015;17:62. doi: 10.1186/s13058-015-0571-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Engeset D, Alsaker E, Lund E, Welch A, Khaw K-T, Clavel-Chapelon F, et al. Fish consumption and breast cancer risk. The European Prospective Investigation into Cancer and Nutrition (EPIC) Int. J. Cancer. 2006;119:175–182. doi: 10.1002/ijc.21819. [DOI] [PubMed] [Google Scholar]
• 22.Norat, T., Chan, D., Vingeliene, S., Aune, D., Polemiti, E., Vieira, A. R. et al. The Associations between Food, Nutrition and Physical Activity and the Risk of Breast Cancer: World Cancer Research Fund International Systematic Literature Review, 13 Jan 2–17 (Imperial College London,London, 2017)
• 23.MacLean CH, Newberry SJ, Mojica WA, et al. Effects of omega-3 fatty acids on cancer risk: a systematic review. JAMA. 2006;295:403–415. doi: 10.1001/jama.295.4.403. [DOI] [PubMed] [Google Scholar]
• 24.Brasky TM, Till C, White E, Neuhouser ML, Song X, Goodman P, et al. Serum phospholipid fatty acids and prostate cancer risk: results from the prostate cancer prevention trial. Am. J. Epidemiol. 2011;173:1429–1439. doi: 10.1093/aje/kwr027. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Brasky TM, Darke AK, Song X, Tangen CM, Goodman PJ, Thompson IM, et al. Plasma phospholipid fatty acids and prostate cancer risk in the SELECT trial. J. Natl Cancer Inst. 2013;105:1132–1141. doi: 10.1093/jnci/djt174. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 26.Aucoin M, Cooley K, Knee C, Fritz H, Balneaves LG, Breau R, et al. Fish-derived omega-3 fatty acids and prostate cancer: a systematic review. Integr. Cancer Therapies. 2016;16:32–62. doi: 10.1177/1534735416656052. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 27.Department of Health. Dietary Reference Values: a guide.https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/743790/Dietary_Reference_Values_-_A_Guide__1991_.pdf (HMSO, London, 1991).
• 28.Forouhi NG, Krauss RM, Taubes G, Willett W. Dietary fat and cardiometabolic health: evidence, controversies, and consensus for guidance. BMJ. 2018;361:k2139. doi: 10.1136/bmj.k2139. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Hackshaw-McGeagh LE, Perry RE, Leach VA, Qandil S, Jeffreys M, Martin RM, et al. A systematic review of dietary, nutritional, and physical activity interventions for the prevention of prostate cancer progression and mortality. Cancer Causes Control. 2015;26:1521–1550. doi: 10.1007/s10552-015-0659-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 30.Zheng J-S, Hu X-J, Zhao Y-M, Yang J, Li D. Intake of fish and marine n-3 polyunsaturated fatty acids and risk of breast cancer: meta-analysis of data from 21 independent prospective cohort studies. BMJ. 2013;346:f3706. doi: 10.1136/bmj.f3706. [DOI] [PubMed] [Google Scholar]
• 31.Flower G, Fritz H, Balneaves LG, Verma S, Skidmore B, Fernandes R, et al. Flax and breast cancer: a systematic review. Integr. Cancer Therapies. 2014;13:181–192. doi: 10.1177/1534735413502076. [DOI] [PubMed] [Google Scholar]
• 32.Ghazanfarpour M, Sadeghi R, Roudsari RL, Khadivzadeh T, Khorsand I, Afiat M, et al. Effects of flaxseed and Hypericum perforatum on hot flash, vaginal atrophy and estrogen-dependent cancers in menopausal women: a systematic review and meta-analysis. Avicenna J. Phytomedicine. 2016;6:273–283. [PMC free article] [PubMed] [Google Scholar]
• 33.Abdelhamid, A. S., Brown, T. J., Brainard, J. S., Biswas, P., Thorpe, G. C., Moore, H. J. et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease.Cochrane Database Syst. Rev.11:CD003177. 10.1002/14651858.CD003177.pub4 (2018) [DOI] [PMC free article] [PubMed]
• 34.Abdelhamid, A. S., Martin, N., Bridges, C., Brainard, J. S., Wang, X., Brown, T. J. et al. Polyunsaturated fatty acids for the primary and secondary prevention of cardiovascular disease.Cochrane Database Syst. Rev. 11:CD012345 10.1002/14651858.CD012345.pub3 (2018) [DOI] [PMC free article] [PubMed]
• 35.Hooper, L., Al-Khudairy, L., Abdelhamid, A. S., Rees, K., Brainard, J. S., Brown, T. J. et al. Omega-6 fats for the primary and secondary prevention of cardiovascular disease.Cochrane Database Syst. Rev. 11:CD011094 10.1002/14651858.CD011094.pub4 (2018) [DOI] [PMC free article] [PubMed]
• 36.Brown TJ, Brainard J, Song F, Wang X, Abdelhamid A, Hooper L. Omega-3, omega-6, and total dietary polyunsaturated fat for prevention and treatment of type 2 diabetes mellitus: systematic review and meta-analysis of randomised controlled trials. BMJ. 2019;366:l4697. doi: 10.1136/bmj.l4697. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Deane, K. H. O., Jimoh, O. F., Biswas, P., O’Brien, A., Hanson, S., Abdelhamid, A. S. et al. Omega-3 and polyunsaturated fat for prevention of depression and anxiety symptoms: systematic review and meta-analysis of randomised trials.Br. J. Psychiatry10.1192/bjp.2019.234 (2019). [DOI] [PubMed]
• 38.Abdelhamid A, Hooper L, Sivakaran R, Hayhoe RPG, Welch A, The PUFAH Group. The relationship between omega-3, omega-6 and total polyunsaturated fat and musculoskeletal health and functional status in adults: a systematic review and meta-analysis of RCTs. Calcif. Tissue Int. 2019;105:353–372. doi: 10.1007/s00223-019-00584-3. [DOI] [PubMed] [Google Scholar]
• 39.Hooper L, Abdelhamid A, Brainard J, Deane KHO, Song F. Creation of a database to assess effects of omega-3, omega-6 and total polyunsaturated fats on health: database and methodology for a set of reviews. BMJ Open. 2019;9:e029554. doi: 10.1136/bmjopen-2019-029554. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 40.Hanson, S., Thorpe, G., Winstanley, L., Abdelhamid, A. & Hooper, L. Effects of supplementary dietary polyunsaturated fat on cancer incidence.Prospero. CRD42017056109 (2017).
• 41.Browning LM, Walker CG, Mander AP, West AL, Madden J, Gambell JM, et al. Incorporation of eicosapentaenoic and docosahexaenoic acids into lipid pools when given as supplements providing doses equivalent to typical intakes of oily fish. Am. J. Clin. Nutr. 2012;96:748–758. doi: 10.3945/ajcn.112.041343. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 42.Higgins, J. P. T., Altman, D. G. & Sterne, J. A. C. Cochrane Statistical Methods Group, Cochrane Bias Methods Group. Chapter 8: Assessing risk of bias in included studies. (eds Higgins J. P. T., Green S.). Cochrane handbook for systematic reviews of interventions Version 510 [updated March 2011]. Available fromwww.handbook.cochrane.org: (The Cochrane Collaboration, 2011).
• 43.McKenzie JE, Herbison GP, Deeks JJ. Impact of analysing continuous outcomes using final values, change scores and analysis of covariance on the performance of meta-analytic methods: a simulation study. Res. Synth. Methods. 2016;7:371–386. doi: 10.1002/jrsm.1196. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 44.Review Manager 5 (RevMan 5). Copenhagen: The Nordic Cochrane Centre: The Cochrane Collaboration (2014).
• 45.Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560. doi: 10.1136/bmj.327.7414.557. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple graphical test. Br. Med J. 1997;315:629–634. doi: 10.1136/bmj.315.7109.629. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 47.Altman, D. G. Confidence intervals for the number needed to treat.BMJ317, 1309–1312 (1998) [DOI] [PMC free article] [PubMed]
• 48.Analysing data and undertaking meta-analyses. Chapter 10. Cochrane Handbook for Systematic Reviews of Interventions version 60 (eds Deeks, J. J., Higgins, J. P. T. & Altman, D. G.). (updated July 2019). Available fromwww.training.cochrane.org/handbook (Cochrane, 2019).
• 49.Efthimiou O. Practical guide to the meta-analysis of rare events. Evid. Based Ment. Health. 2018;21:72–76. doi: 10.1136/eb-2018-102911. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Sweeting MJ, Sutton AJ, Lambert PC. What to add to nothing? Use and avoidance of continuity corrections in meta-analysis of sparse data. Stat. Med. 2004;23:1351–1375. doi: 10.1002/sim.1761. [DOI] [PubMed] [Google Scholar]
• 51.Sweeting MJ, Sutton AJ, Lambert PC. Correction to sweeting 2004. Stat. Med. 2006;25:2700. doi: 10.1002/sim.2536. [DOI] [PubMed] [Google Scholar]
• 52.Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995;273:408–412. doi: 10.1001/jama.1995.03520290060030. [DOI] [PubMed] [Google Scholar]
• 53.Savovic J, Jones H, Altman D, Harris R, Jüni P, Pildal J, et al. Influence of reported study design characteristics on intervention effect estimates from randomised controlled trials: combined analysis of meta-epidemiological studies. Health Technol. Assess. 2012;16:1–82. doi: 10.3310/hta16350. [DOI] [PubMed] [Google Scholar]
• 54.Wood L, Egger M, Gluud LL, Schulz K, Jüni P, Altman DG, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ. 2008;336:601–605. doi: 10.1136/bmj.39465.451748.AD. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 55.Abdelhamid, A. S., Brown, T. J., Brainard, J. S., Biswas, P., Thorpe, G. C., Moore, H. J. et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease.Cochrane Database Syst. Rev.3, CD003177. 10.1002/14651858.CD003177.pub5 (2020) [DOI] [PMC free article] [PubMed]

Supplementary Materials

Data Availability Statement

The dataset for this review was part of our published dataset, and so is publicly available, see ref.³⁹