Maki 2010.

Study characteristics
Methods	Setting: USA (Bloomington, IN and St Petersburg, FL) Design: randomised parallel‐arm controlled trial Dates: trial dates not reported Intervention duration: 12 weeks Follow‐up: no postintervention follow‐up Focus: to compare the effect on LDL cholesterol and other cardiovascular disease markers of whole grain oat ready‐to‐eat cereal compared to energy‐matched low‐fibre foods as part of a dietary program for weight loss in overweight and obese adults.
Participants	N: 204 randomised, 80/101 completed in whole grain oat ready‐to‐eat cereal group; 73/103 completed in the control group. Inclusion criteria: free‐living overweight and obese adults, BMI 25 to 40 kg/m2 with baseline LDL cholesterol 3.4 to 5.2 mmol/L, aged 20 to 65 years. Exclusion criteria: participants who reported a weight change of 4.5 kg during the previous 2 months; use of weight loss medications within 2 months before screening or supplements, programs, or meal replacement products within 2 weeks before screening; use of drugs (within 4 weeks before screening), supplements, or foods (within 2 weeks before screening) known to alter lipid levels; use of fibre‐containing supplements within 2 weeks before screening; daily consumption of oat or barley products (e.g. ready‐to‐eat oat‐based cereals, oatmeal, or oat bran) or frequent consumption of foods rich in viscous fibre within 2 weeks of screening; clinically significant abnormal laboratory test results (e.g. triglycerides >/= 4.5 mmol/L, glucose >/= 7.0 mmol/L, creatinine >/= 114.4 μmol/L, and alanine aminotransferase and aspartate aminotransferase levels 1.5 times the upper limit of normal), uncontrolled hypertension (systolic/diastolic blood pressures >/= 160/100 mmHg); a history of cardiac, renal, hepatic, endocrine, pulmonary, biliary, pancreatic, gastrointestinal, or neurologic disorders, or cancer in the past 2 years; known sensitivity to any of the ingredients in the study foods; a history of weight‐reducing surgery; a history of eating disorders or alcohol abuse; or use of thyroid hormones (except stable dose replacement therapy) or systemic corticosteroids. Age (years): whole grain oat group: 50.1 (SEM); control: 47.5 (SEM 1.3) Sex (% men): whole grain oat group: 24.7%; control: 17.9% Ethnicity: US: whole grain oat group: non‐Hispanic white 69/77; African‐American 6/77; Hispanic 1/77; other 1/77 control group: non‐Hispanic white 55/67; African‐American 5/67; Hispanic 4/67; other 3/67 Baseline cardiovascular risk status: BMI (kg/m2): whole grain intervention: 32.2 (SEM 0.6); control: 32.0 (SEM 0.5) Total cholesterol (mmol/L): whole grain intervention: 6.00 (SEM 0.08); control: 5.92 (SEM 0.08) HDL cholesterol (mmol/L): whole grain intervention: 1.24 (0.03); control: 1.24 (0.04) LDL cholesterol (mmol/L): whole grain intervention: 4.02 (0.05); control: 4.00 (0.06) Triglyceride (mmol/L): whole grain intervention: 1.65 (SEM 0.0.9); control: 1.48 (SEM 0.0.8) Systolic blood pressure (mmHg): whole grain intervention: 127.2 (SEM 1.2); control: 122.6 (SEM 1.3) (P = 0.01) Diastolic blood pressure (mmHg): whole grain intervention: 79.4 (0.9); control: 78.3 (1.00) Medications used: not reported
Interventions	Whole grain diet group: whole grain oat ready‐to‐eat cereal Control: low‐fibre cereal foods Description of dietary intervention: 2 portions per day of whole grain ready‐to‐eat oat cereal or energy‐matched low‐fibre foods (control) such as corn cereals, white toast, plain bagels, English muffins, pretzels, soda crackers, or rice cakes, as part of a reduced‐energy (~500 kilocalories/day deficit) dietary program that encouraged limiting consumption of foods high in energy and fat, portion control, and regular physical activity. All participants were asked to avoid foods rich in viscous soluble fibre such as barley, oatmeal and oat bran products except for the study products provided to the whole grain oat cereal group. The whole grain oat ready‐to‐eat cereal (Cheerios, General Mills, Minneapolis, MN) was packaged in ~40 g portions by the manufacturer. A registered dietitian met biweekly with participants to monitor and reinforce dietary changes. Incentives: not reported Co‐interventions in both groups: regular physical activity (30 to 60 min/day) was encouraged in both groups as part of the intervention. Physical activity was assessed using a 7‐day recall questionnaire. At baseline, the mean activity scores above resting were 118.4 (SEM 7.6) metabolic equivalent hours in the whole grain oat group and 118.3 (SEM 10.0) in the control group. Physical activity scores increased by 9.1% (SEM 4%) in the whole grain group and 15% (SEM 5.2%) (P = 0.710) in the control group. Assessment of dietary adherence: 3‐day self reported diaries recording compliance with provided foods, participant interview, and review of unused foods. The whole grain oat group consumed 96.8% (SEM 0.6) of the expected servings of study foods, and the control group consumed 95.7% (SEM 0.7) (P = 0.202). Was the diet energy reduced? yes (~500 kilocalories/day deficit) Comparability of diet composition: (See Table 2.) As designed, both groups reduced energy intakes; there was no difference between groups at week 12, although the control group showed a larger reduction in energy intake at week 4. The percentage of total daily energy intake from carbohydrate was greater at week 12 in the whole grain oat group than in the control group (P = 0.017). Total and soluble fibre increased as expected. Change in diet over time: as above for energy. (See Table 2.)
Outcomes	Total cholesterol, LDL cholesterol, HDL cholesterol, non‐HDL cholesterol, triglycerides, BMI, waist circumference, midarm circumference, triceps skinfold thickness
Funding/conflicts of interest	General Mills Bell Institute of Health and Nutrition, Minneapolis, MN. 3 of the authors are employees of General Mills, 3 authors are employees of Provident Clinical Research and received research grant support from General Mills to conduct the study. 1 author is an employee of Meridien Research and received research grant support from General Mills.
Notes	Adverse effects: the frequencies of adverse events (of any type, whether related to the study products or not) were reported to be similar between groups (59.8% for the whole grain group and 52.4% for the control group, P = 0.321). The most common adverse events in both groups were respiratory tract infection, sinusitis, and pharyngitis. Most adverse events were mild and not related to the study product. Adverse events the authors considered related to the study products were nausea (2 people in the whole grain oat group), flatulence (2 people in the whole grain oat group), gastroenteritis (1 person in the control group), gastroesophageal reflux (1 person in the control group), and vomiting (1 person in the control group). Adverse events that led to drop out from the study were an infectious cyst (1 control) and spinal stenosis (1 control), but the study authors did not consider these to be related to the study product.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"participants were randomly assigned"; method of randomisation not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported for anthropometry or lab measurements
Incomplete outcome data (attrition bias) All outcomes	High risk	More participants completed study in whole grain oat group than in control group. 20 withdrew consent in control group, 11 in whole grain oat group; both groups lost n = 7 to follow‐up; 2 in control group reported adverse events; 1 in WG group reported product too filling. There was differential dropout between groups.
Intention to treat analysis	Unclear risk	Modified intention‐to‐treat analyses done. Of 101 participants randomised to whole grain oat group, 86 were included in MITT analysis (15 were excluded as they did not have at least 1 postrandomisation lipid value or the blood draw was invalid). Of 103 randomised to control group, 87 were included in MITT analysis (16 were excluded as they did not have at least 1 postrandomisation lipid value or the blood draw was invalid). Sensitivity analyses conducted using MITT and completers data. Results also presented as per‐protocol subset of MITT.
Selective reporting (reporting bias)	Unclear risk	Blood pressure data not fully reported; no outcome data given.
Groups comparable at baseline	Unclear risk	Yes, for all reported measures, except for systolic blood pressure, which was higher in the whole grain oat group compared to control (127.2 SEM 1.2 vs 122.6 SEM 1.3). However, outcome data not reported for systolic blood pressure. Reported only for the per‐protocol subset.
Other bias	Low risk	Power calculations: a sample size of 128 (64 per arm) was expected to provide 80% power to detect a difference of 5% between groups in the per cent change from baseline in LDL cholesterol, assuming a 10% pooled standard deviation for the LDL cholesterol response. This sample size was achieved in both the analysis on completers and the MITT analysis.