Summary of findings 2.

High versus low ALA omega‐3 fats for preventing cardiovascular disease (primary outcomes)

High versus low ALA omega‐3 fats for preventing cardiovascular disease (primary outcomes)
Patient or population: adults with or without existing CVD Setting: participants were living at home for most or all of the duration of their trials. Most studies were carried out in high‐income economies (World Bank 2018), but four trials were carried out in upper‐middle income countries (Argentina, Iran, Turkey and China). No studies took place in low‐ or low‐middle income countries. Intervention: higher intake of ALA Comparison: lower intake of ALA The intervention was dietary supplementation, a provided diet or advice on diet. Supplementation may have been in oil or capsule form or as foodstuffs provided, to be consumed by mouth (excluding enteral and parenteral feeds and enemas). The foodstuffs or supplements must have been: refined ALA, linseed (flax), canola (rapeseed), perilla, purslane, mustard seed, candlenut, stillingia or walnut as a food, oil, made into a spreading fat or supplementing another food (such as bread or eggs). For ALA sources the product consumed had to have an omega‐3 fat content of at least 10% of the total fat content.
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with lower ALA	Risk with higher ALA
All‐cause mortality – deaths Assessed with: number of participants dying of any cause, whether reported as an outcome or a reason for dropout Duration: range 12 to 40 months	25 per 1000	25 per 1000 (21 to 29)	RR 1.01 (0.84 to 1.20)	19327 (5 RCTs)	⊕⊕⊕⊝ Moderatea	Meta‐analysis and sensitivity analyses suggest risk increase of less than 1%. ALA intake probably makes little or no difference to all‐cause mortality.
Cardiovascular mortality – cardiovascular deaths Assessed with: deaths from any cardiovascular cause. Where this was not available cardiac death was used instead where known. Duration: range 12 to 40 months	12 per 1000	12 per 1000 (9 to 15)	RR 0.96 (0.74 to 1.25)	18619 (4 RCTs)	⊕⊕⊕⊝ Moderateb	Meta‐analysis and sensitivity analyses suggest risk reduction of less than 5%. ALA intake probably makes little or no difference to cardiovascular mortality.
Cardiovascular events – cardiovascular events Assessed with: number of participants experiencing any cardiovascular event Duration: range 12 to 40 months	48 per 1000	47 per 1000 (39 to 57)	RR 0.95 (0.83 to 1.07)	19327 (5 RCTs)	⊕⊕⊝⊝ Lowc	Meta‐analysis and sensitivity analyses suggest risk reduction of 5% to 10%. ALA intake may reduce the risk of cardiovascular events but by a very small amount (from 4.8 to 4.7%). One thousand people would need to consume more ALA to prevent a single person experiencing a CVD event (NNT=1000).
Coronary heart mortality – CHD deaths Assessed with: Coronary deaths, or where these were not reported, IHD death, fatal MI or cardiac death (in that order) Duration: range 12 to 40 months	11 per 1000	10 per 1000 (8 to 14)	RR 0.95 (0.72 to 1.26)	18353 (3 RCTs)	⊕⊕⊕⊝ Moderated	Meta‐analysis and sensitivity analyses suggest risk reduction of 5% to 8%. ALA intake probably reduces risk of CHD mortality but by a very small amount (from 1.1 to 1.0%). One thousand people would need to consume more ALA to prevent a single person experiencing a CHD death (NNT=1000).
Coronary Heart Disease – CHD events Assessed with: number of participants experiencing the first outcome in this list reported for each trial: CHD or coronary events; total MI; acute coronary syndrome; or angina (stable and unstable) Duration: range 12 to 40 months	22 per 1000	22 per 1000 (17 to 28)	RR 1.00 (0.82 to 1.22)	19061 (4 RCTs)	⊕⊕⊝⊝ Lowe	Meta‐analysis and sensitivity analyses suggest risk reduction of 0% to 9%. ALA intake may make little or no difference to CHD events.
Stroke Assessed with: number of participants experiencing at least one fatal or non‐fatal, ischaemic or haemorrhagic stroke Duration: range 12 to 40 months	2 per 1000	3 per 1000 (2 to 5)	RR 1.15 (0.66 to 2.01)	19327 (5 RCTs)	⊕⊝⊝⊝ Very lowf	Meta‐analysis and sensitivity analyses suggest risk increase of −15% to 23%. The effect of ALA intake on stroke is unclear as the evidence is of very low quality.
Arrhythmias – AF, VT, VF Assessed with: number of participants experiencing fatal or nonfatal, new or recurrent arrhythmia, including atrial fibrillation, ventricular tachycardia and ventricular fibrillation Duration: 1 trial of 40 months	33 per 1000	26 per 1000 (19 to 36)	RR 0.79 (0.57 to 1.10)	4837 (1 RCT)	⊕⊕⊕⊝ Moderateg	Meta‐analysis and sensitivity analyses suggest risk reduction of 21%. ALA intake probably reduces the risk of arrhythmias a small amount (from 3.3 to 2.6%). 143 people would need to consume more ALA to prevent a single person experiencing an arrhythmic event (NNT=143).
Harms: bleeding Assessed with: number of participants experiencing bleeding events	The effect of ALA intake on bleeding is unclear as no studies reported this outcome.
Harms: pulmonary embolus or DVT Assessed with: number of participants experiencing pulmonary embolus or deep vein thrombosis Duration: range 24 months	3 per 1000	1 per 1000 (0 to 23)	RR 0.32 (0.01 to 7.80)	708 (1 study)	⊕⊝⊝⊝ Very lowh	The effect of ALA intake on pulmonary embolus or DVT is unclear as the evidence is of very low quality.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ALA: alpha‐linolenic acid; CHD: coronary heart disease; CI: confidence interval; DVT: deep vein thrombosis; IHD: ischaemic heart disease; MI: myocardial infarction; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aAll‐cause mortality, ALA

• Risk of bias: there was little or no effect in the main meta‐analysis or when data were limited to RCTs at low summary risk of bias, low risk of compliance problems or larger trials, though a suggestion of increased risk of death with fixed‐effect meta‐analyses. Not downgraded.
• Inconsistency: I² was < 60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies with consistent results. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^bCardiovascular mortality, ALA

• Risk of bias: there was little or no effect in the main analysis, or when data were limited to RCTs at low summary risk of bias, larger trials or fixed‐effect meta‐analysis, though a small benefit was suggested when studies were limited to trials with low risk of compliance bias. Not downgraded.
• Inconsistency: I² was < 60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies with consistent results. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^cCardiovascular events, ALA

• Risk of bias: there was a small effect in the main analysis, with larger trials and in fixed‐effect analysis, and a larger effect when data were limited to RCTs at low summary risk of bias or at low risk from compliance problems. Downgraded once.
• Inconsistency: I² was <60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies with consistent results. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^dCoronary heart disease mortality, ALA

• Risk of bias: while ALA reduced CHD mortality by 5% in the main analysis, fixed‐effect analysis and in larger trials, limiting data to RCTs at low summary risk of bias and low risk of compliance problems resulted in 7%‐8% reductions. Not downgraded.
• Inconsistency: I² was < 60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies with consistent results. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded. Downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^eCoronary heart disease events, ALA

• Risk of bias: there was little or no effect in the main analyses, in fixed‐effect meta‐analysis, or in larger studies, but some risk reduction (8 to 9%) when data were limited to RCTs at low summary risk of bias or low risk of compliance bias. Downgraded once.
• Inconsistency: I² was < 60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies with consistent results. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded. Downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^fStroke, ALA

• Risk of bias: the main analysis, fixed‐effect analysis, and larger trials suggest increased risk of stroke with more ALA, but there was little or no effect when data were limited to RCTs at low summary risk of bias, and a suggestion of benefit when limited to trials with low risk of compliance problems. Downgraded twice.
• Inconsistency: I² was < 60%; not downgraded.
• Indirectness: representative, generalisable adult population including men and women, including healthy participants and participants with CVD risk factors or previous CVD as well as non‐CVD health problems. All studies were conducted in high‐income countries. Not downgraded.
• Imprecision: large numbers of participants have taken part in RCTs in long‐term studies, but only 49 participants experienced strokes. 95% confidence intervals do not exclude important benefits or harms, downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^gArrhythmias, ALA

• Risk of bias: there was a 21% reduction in risk of arrhythmia in the main analysis, when data were limited to RCTs at low summary risk of bias, in larger trials and when data were limited to trials at low risk from compliance. Not downgraded.
• Inconsistency: only one trial, no inconsistency. Not downgraded.
• Indirectness: a single trial, which included adults with previous MI in a high‐income country and only assessed new arrhythmia. Not downgraded.
• Imprecision: large numbers of participants have taken part in this long term study. However, as 95% confidence intervals do not exclude important benefits or harms we downgraded once.
• Publication bias: funnel plot not useful as fewer than 10 trials included. Not downgraded.

^hPulmonary embolus or DVD, ALA

• Risk of bias: the single trial was not at low summary risk of bias. Downgraded once.
• Inconsistency: with one trial no inconsistency. Not downgraded.
• Indirectness: healthy men and women, no participants with CVD risk factors or previous CVD; low‐ and middle‐income countries not represented. Not downgraded.
• Imprecision: only one event included in a single trial. Downgraded twice.
• Publication bias: insufficient studies for funnel plot. Not downgraded.