MARGARIN 2002

Methods	Mediterranean alpha‐linolenic enriched Groningen dietary intervention study (MARGARIN) RCT, factorial 2 × 2 (ALA rich margarine vs LA rich margarine, also nutrition education vs no education but this is not included), 2 years Summary risk of bias: low
Participants	Hypercholestrolemic adults with 2 or more CVD risk factors N: total 282 randomised; 114 intervention (51 with nutrition education, 58 without NE) 157 control (52 with NE, 105 without NE) Level of risk for CVD: moderate (multiple cardiovascular risk factors, 10‐year IHD risk ˜20%) Men: 41.9% intervention, 45.7% control Mean age in years (SD): 54.4 (9.5) intervention, 53.9 (9.8) control Age range: 30‐70 Smokers: 49.1% intervention, 49.3% control Hypertension: 52.9% intervention, 45.3% control (on anti‐hypertensives) Medications taken by at least 50% of those in the control group: antihypertensives Medications taken by 20%‐49%: not reported Medications taken by some, but < 20%: not reported Location: the Netherlands Ethnicity: not reported
Interventions	Type: supplementary food (ALA enriched margarine) Intervention: provided with ALA rich margarine (80% fat of which 15% was ALA and 46% LA) to be eaten ad libitum. Dose: average intake 6.3 g/d ALA (was also 1 g/d ALA in the control group). Control: provided with linoleic rich margarine (80% fat of which 0.3% was ALA and 58% LA), identical in taste and packaging. Both margarines contained 0.66 mg vit E/g, 9 micro‐g vit A/g and 0.023 micro‐g vit D/g Comparison: ALA vs omega 6 Compliance: serum fatty acids used to assess, ALA rose by 0.47 mol % (SD 0.04) and 0.36 mol% (SD 0.04) intervention arms (with and without NE) and fell by 0.06 mol % (SD 0.04) and 0.11 mol % (SD 0.03) control arms (with and without NE), significantly different. Duration of intervention: 24 months
Outcomes	Main study outcome: cardiovascular risk factors and IHD risk Dropouts: unclear Available outcomes: total and CV deaths, non‐fatal MI, stroke, CABG and angioplasty, BMI, lipids, BP Response to contact: yes
Notes	Study funding: Prevent fund and Unilever Research Other intervention (2 × 2) was educational, teaching a multifactorial dietary intervention. It was excluded as multifactorial.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random allocation, allocated by an independent trial coordination centre that organised masked distribution of margarines
Allocation concealment (selection bias)	Low risk	Allocated by an independent trial coordination centre which organised masked distribution of margarines
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind; the 2 margarines are described as identical as to taste and packaging (though not reported as checked)
Blinding of outcome assessment (detection bias) All outcomes	Low risk	2 independent physicians, a cardiologist and a general practitioner validated and classified results in a blinded fashion
Incomplete outcome data (attrition bias) All outcomes	Low risk	The number randomised to each arm was unclear, but one publication clarifies (55 randomised to each arm, 51 intervention and 52 control analysed).
Selective reporting (reporting bias)	Unclear risk	No study protocol or trials registry entry was found.
Attention	High risk	There was no difference in attention between margarine types, but the dietary advice group spent more time with study staff than the control group, and some (not quite randomly allocated) were sent individual motivational letters (Siero 2000).
Compliance	Low risk	Serum fatty acids used to assess, ALA rose by 0.47 mol% (SD 0.04) and 0.36 mol % (SD 0.04) intervention arms (with and without NE) and fell by 0.06 mol % (SD 0.04) and 0.11 mol % (SD 0.03) control arms (with and without NE), Significantly different
Other bias	Low risk	No further bias noted