Portfolio Diet

Diet Reference Archive: a research-developed eating pattern for lowering LDL cholesterol

Portfolio Diet

A plant-based eating pattern that adds a set daily amount of 5 kinds of food, each studied for lowering LDL cholesterol, to a diet low in saturated fat and cholesterol: nuts and seeds, soy or pulse protein, foods rich in viscous fiber, plant sterols, and plant oils rich in monounsaturated fat. It was built and tested by a research group in Toronto to lower LDL cholesterol, the main blood fat linked to heart disease.

Family: Guideline-endorsed and evidence-based dietary patterns Promoted by: David Jenkins and colleagues at the University of Toronto and St. Michael's Hospital, Toronto First appeared: 2002, in the first published trial of the combination GRADE: High for lowering LDL cholesterol, the Portfolio diet added to a low-saturated-fat diet compared with that diet alone, in adults with high cholesterol, over a median of 4 weeks (range 4 to 24 weeks); published review (Chiavaroli et al., 2018)

ON THIS PAGE

1. In brief
2. Overview
3. In practice
4. For the coach
5. The evidence (opens on request)
6. Mechanisms (opens on request)
7. Risks, contraindications and myths (opens on request)
8. Key terms
9. Review questions
10. References
11. Evidence rating and review record

The summary and sections 1 to 3 are the short version and are open when the page loads. Sections 4 to 6 hold the research detail, the biology, and the cautions. Each of those opens when you select its heading, and you can read the page without opening them. Back to the Diet Reference Archive index.

In brief

What is it? A way of eating built around 5 kinds of food that each lower LDL cholesterol a little: nuts and seeds; protein from soy foods or beans, peas and lentils; foods with sticky, gel-forming fiber such as oats, barley, psyllium, okra and eggplant; foods or supplements with added plant sterols; and plant oils such as olive and canola oil. They are added, in set daily amounts, to an eating plan low in saturated fat and cholesterol. A research group in Toronto designed it in the early 2000s, and its developers still describe the amounts and publish most of the research (Chiavaroli et al., 2018; Kavanagh, Chiavaroli, et al., 2025).

Does it work? For LDL cholesterol, yes, and the evidence is strong. Pooling 7 controlled comparisons in 439 adults with high cholesterol, mostly 4-week studies in which every meal was supplied, a published review found LDL cholesterol about 17 percent lower than on a low-saturated-fat diet alone, and rated that result high in certainty (Chiavaroli et al., 2018). When people were given advice rather than food, the effect was smaller: in the largest trial, LDL fell about 13 to 14 percent over 6 months, against 3 percent on low-saturated-fat advice (Jenkins et al., 2011). Small drops in blood pressure and in a marker of inflammation were rated moderate in certainty.

What has not been shown is whether eating this way prevents heart attacks, strokes or deaths: no trial has measured those, and a trial of artery wall changes finished in 2023 without results the searches for this page could find (Unity Health Toronto, 2024). Large cohort studies link usual diets that resemble the pattern with somewhat lower rates of heart disease and death (Glenn, Guasch-Ferré, et al., 2023; Kavanagh, Zurbau, et al., 2025), which this page rates low in certainty. Most of the trials and cohort analyses come from the developers' own research network.

Is it safe? In the 6-month advice trial, there were no serious adverse events or hospitalizations; 36 minor events were recorded, with no clear difference in their number between groups, and 2 participants had reactions associated with soy or soy and almonds (Jenkins et al., 2011). Some trials also measured blood levels of plant sterols and fat-soluble vitamins: plant sterol levels rose and beta-carotene fell a little, with no difference once cholesterol levels were allowed for (Ramprasath et al., 2014). These short trials do not establish long-term safety, and the searches did not identify a study designed to test it.

The cautions are specific. Plant sterol products are labeled in the European Union as not for people who do not need to lower cholesterol, as needing medical supervision for people on cholesterol medicine, and as possibly unsuitable in pregnancy, breastfeeding and for children under 5 (Regulation (EU) No 1169/2011, 2011). Nuts, peanuts and soy are major food allergens (Food and Drug Administration, 2026). Soy, walnuts and fiber can reduce how well the thyroid medicine levothyroxine is absorbed (AbbVie, 2024).

Who should not do it without talking to a doctor or dietitian first? Anyone taking a statin or another cholesterol medicine, because any change to that medicine is the prescriber's decision, and no trial has tested the diet as a substitute for one in preventing heart disease. Anyone taking levothyroxine, warfarin, insulin or a sulfonylurea, who keeps the prescriber informed before a substantial change in eating; a possible low blood sugar episode is treated first under the person's own plan, with emergency help if symptoms are severe or the person cannot manage it, and the prescriber is told afterward. Anyone with kidney disease, whose needs depend on the stage of disease, lab results and dialysis. Anyone with a nut, peanut or soy allergy, or with the rare inherited condition sitosterolemia. Anyone pregnant or breastfeeding, and children. Anyone with a history of disordered eating.

Related pages: DASH Diet, Mediterranean Diet, Whole-Food Plant-Based Diet and Vegan Diet. Full list on the archive index.

1. Overview

What you eat

The Portfolio Diet asks for 5 kinds of food every day, each in a set amount. The amounts its developers give for a 2,000-calorie day are about 45 grams of nuts and seeds, such as almonds, walnuts, pumpkin seeds or peanut butter; about 50 grams of plant protein, from tofu, soy milk, soy-based meat alternatives, beans, chickpeas or lentils; about 20 grams of viscous fiber, the gel-forming kind found in oats, oat bran, barley, psyllium, eggplant, okra, apples, oranges and berries; about 2 grams of plant sterols, from fortified foods or supplements; and about 45 grams of monounsaturated fat from plant sources, such as extra virgin olive oil, canola or soybean oil, and avocado (Kavanagh et al., 2024; Kavanagh, Chiavaroli, et al., 2025).

In everyday terms, 45 grams of nuts is a little over 1.5 ounces, and reaching 50 grams of plant protein and 20 grams of viscous fiber usually means soy, pulse, oat or barley foods at more than one meal. For the plant sterols, the developers point to fortified foods and supplements (Kavanagh et al., 2024). The earliest trials used 4 kinds of food and added plant sterols mostly as an enriched margarine; plant oils became the fifth part in a 2010 trial (Chiavaroli et al., 2018; Jenkins et al., 2010).

What you avoid or limit

The 5 foods are added to a background diet low in saturated fat and cholesterol, the National Cholesterol Education Program Step II diet that the trials used: no more than 7 percent of calories from saturated fat and less than 200 milligrams of cholesterol a day (Chiavaroli et al., 2018; Kavanagh et al., 2024). In practice that means limiting the foods richest in saturated fat and cholesterol, such as fatty meats, butter, full-fat dairy and many baked goods.

The limit is a guideline rather than a list of forbidden foods. The sources opened for this page define the general pattern by the plant foods it adds and by its saturated fat and cholesterol limits, and none of them describes it as vegetarian or vegan by rule, although the study diets in the 2003 feeding trial and the 2011 advice trial, including the comparison diets, were vegetarian (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003; Jenkins et al., 2011). The score researchers use to measure it gives points for the 5 plant foods and takes points away for foods high in saturated fat and cholesterol (Glenn, Guasch-Ferré, et al., 2023).

Where it came from

The diet was developed by David Jenkins and colleagues at the University of Toronto and St. Michael's Hospital in Toronto (Chiavaroli et al., 2018). Jenkins was already known for an earlier idea: he was first author of a 1981 paper, "Glycemic index of foods," that compared how much 62 common foods raised blood sugar (Jenkins et al., 1981). In the early 2000s, the Adult Treatment Panel III recommendations of the National Cholesterol Education Program emphasized diets low in saturated fat together with plant sterols and viscous fiber, and the American Heart Association supported soy protein and nuts; the Toronto group set out to test what happens when all of these are eaten together, in one diet (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003; Kendall & Jenkins, 2004).

The first published test, in 2002, already called the combination a dietary portfolio. It gave 13 adults with high cholesterol a month of a diet combining plant sterols, soy protein and viscous fiber from foods "available in supermarkets and health food stores"; LDL cholesterol fell by 29 percent, and the diet was rated acceptable and "very filling" (Chiavaroli et al., 2018; Jenkins et al., 2002). Almonds were added in the 2003 trials (Jenkins, Kendall, Marchie, Faulkner, Vidgen, et al., 2003). The group then tested it against a cholesterol-lowering medicine, as dietary advice for a year, across 4 Canadian centers, and with added plant oils, the trials described in section 4.

In recent years the same group has developed a score for measuring how closely people's ordinary diets match the pattern, a free web app, and handouts for patients and health professionals. The developers report that the Canadian Cardiovascular Society endorses their infographic and app and hosts their toolkit for health professionals (Kavanagh et al., 2022; Viscardi, Eskandari, et al., 2025). The authors of the 2018 review list Canadian, European and British guidelines that recognize the diet (Chiavaroli et al., 2018); those guidelines themselves were not opened for this page, except the 2023 European diabetes nutrition recommendations, which describe the diet and its evidence but make no specific recommendation about it (Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, 2023).

What it claims to do

Its developers present it as a diet to lower LDL cholesterol and, through that, the risk of heart disease. Their 2018 review concludes that the pattern "leads to clinically meaningful improvements in LDL-C as well as other established cardiometabolic risk factors" (Chiavaroli et al., 2018). Early papers went further, suggesting that such a diet "may be as effective as the starting dose of older first-line drugs" (Jenkins et al., 2002). The developers' later reviews suggest it could serve as an addition to cholesterol medicine for people who need more lowering (Chiavaroli et al., 2018).

Books sold under the name make broader claims. An academic book on the diet was published by Elsevier in 2019, and several independently published cookbooks use the name, one subtitled as a proven program to prevent heart disease (Open Library, n.d.). Those are their authors' claims. Section 4 says what the research does and does not show, including the absence of trials of heart attacks, strokes or deaths.

What it gets right

Every part of the pattern points the same way as mainstream heart-health guidance. The American Heart Association's 2026 dietary guidance lists, among the features of a heart-healthy pattern, choosing healthy sources of protein, choosing unsaturated fats in place of saturated fats, eating plenty of vegetables and fruits, and choosing mostly whole grains (Lichtenstein et al., 2026). The National Heart, Lung, and Blood Institute tells people with high LDL cholesterol that eating plans which limit saturated fats and include whole grains, fruits, vegetables and a variety of nuts can help lower it (National Heart, Lung, and Blood Institute, 2024).

What the Portfolio Diet adds is specificity: named foods in measured amounts, chosen because each one has been studied for cholesterol lowering, and a trial base that measured what happens when they are eaten together. It also adds rather than subtracts. Most of the change it asks for is eating more of something, a feature a coach can honor in a client's reasons without endorsing the whole package.

How clients describe it, and what it gets confused with

Clients say "I'm doing the Portfolio Diet," "I'm eating the cholesterol foods," "my cardiologist gave me a handout with nuts, soy, oats and sterols," "I'm trying to lower my cholesterol without a statin," or "I use the Portfolio app." Some mean the full 5-part pattern; some mean one part, such as adding oats or a sterol-enriched spread; and some are following a version from another country, such as a Mediterranean Portfolio diet.

It is confused with the DASH Diet, which also lowers LDL cholesterol a little but was designed for blood pressure; with the Mediterranean Diet, which shares nuts and olive oil; with the Whole-Food Plant-Based Diet, which emphasizes whole or minimally processed plant foods; with the Vegan Diet, which excludes animal foods; with the Vegetarian Diet, which excludes meat, poultry and fish; and with the TLC Diet, the older Therapeutic Lifestyle Changes plan from the National Heart, Lung, and Blood Institute. Because Jenkins also introduced the glycemic index, clients sometimes mix it up with the Low-Glycemic Index Diet. The DASH Diet, Mediterranean Diet, Whole-Food Plant-Based Diet, Vegan Diet and Vegetarian Diet pages carry their own evidence, which this page does not repeat.

2. In practice

A typical day of eating

A day built from the foods the developers name might start with oatmeal or oat bran cooked in soy milk, topped with berries and a spoonful of ground seeds, with a plant sterol-enriched drink or supplement if the person uses one with their clinician's or dietitian's agreement. Lunch might be a lentil or bean soup with barley, a slice of whole-grain bread with a little avocado, and an orange. A small handful of almonds or walnuts serves as a snack.

Dinner might be tofu or a soy-based burger with eggplant and okra roasted in olive or canola oil, over brown rice, with a salad dressed with oil and vinegar, and an apple later in the evening. Psyllium, if used, is stirred into water or another drink. Nothing in that day is unusual food; what changes for most people is how often soy, pulses, oats and nuts appear, and how much saturated fat from meat, butter and full-fat dairy is left out. The example illustrates the food groups the sources name; it is not a menu a coach hands to a client.

Cost, products, and who profits

The searches for this page did not identify a study of what the Portfolio Diet costs at the grocery store, so what follows is observation rather than measured data. Oats, barley, dried or canned beans, lentils and soy milk are among the less expensive protein and fiber sources in most stores, while nuts, plant sterol-enriched products and sterol supplements cost more. Availability varies: a participant in the developers' 2025 app study described finding a sterol powder in a health food store to replace a sterol "margarine that's no longer produced" (Kavanagh, Chiavaroli, et al., 2025). Eggplant and okra are easy to find in some neighborhoods and hard in others.

The searches for this page found no company that owns the Portfolio Diet or sells a membership in it. The developers' web app, PortfolioDiet.app, is described by them as freely available, and their toolkit for health professionals is hosted on the Canadian Cardiovascular Society's website (Kavanagh et al., 2022; Viscardi, Eskandari, et al., 2025). An Elsevier academic book and several independently published cookbooks carry the name (Open Library, n.d.); their authors and publishers profit from sales.

The companies that make the component foods have an interest too. The 2018 review reports that every trial it included was funded by a combination of public agencies and industry (Chiavaroli et al., 2018). The trial registration for the largest trial lists Loblaw Companies, Solae and Unilever R&D as collaborators (University of Toronto, 2018), and in a 2023 paper David Jenkins declared research grants from many food and agricultural organizations, among them pulse growers' associations, Loblaw, Unilever, the Almond Board of California, Quaker Oats, soy and nut trade groups, and Solae (Glenn, Guasch-Ferré, et al., 2023). Funding does not make a result wrong; it is something a reader should know before weighing it, and section 4 explains how it bears on the evidence.

Environmental footprint

The searches for this page did not identify a study of the environmental footprint of the Portfolio Diet itself. Because the pattern adds plant protein and plant fats on top of a diet that may or may not still include meat and dairy, its footprint depends mostly on what a person eats less of to make room; the Vegan Diet and Whole-Food Plant-Based Diet pages describe the research on plant-based diets and the environment.

Cultural fit

Soy foods, beans, lentils, nuts, oats, barley, okra and eggplant belong to many food traditions, and the developers' app offers recipes it describes as culturally adapted, including African, Mediterranean and South Asian dishes, with filters for other dietary needs (Kavanagh, Chiavaroli, et al., 2025). Research teams elsewhere have built local versions: a Mexican portfolio using nopal, soy protein, chia seeds, inulin and oats (Vázquez-Manjarrez et al., 2021), an Italian Portfolio-Mediterranean diet (Ferro et al., 2020), and a Mediterranean Portfolio diet in England (Stradling et al., 2021).

The evidence comes from some populations more than others. The trials were run in Canada, in mostly middle-aged adults (Chiavaroli et al., 2018). Cohort studies have included older adults in Hong Kong (Lo et al., 2021), a racially diverse United States sample, where the authors found that the association differed by race and ethnicity (Kavanagh, Zurbau, et al., 2025), and adults in Tehran, where no clear association with type 2 diabetes was found and the authors suggest that local eating patterns and the way the score weights its parts may matter (Malmir et al., 2025). Two features fit some lives less easily: a nut, peanut or soy allergy rules out a core part of the pattern, and sterol-enriched products are hard to find in some places.

What makes people stop

The trials report how many people stayed and how closely they followed the plan, though not why people left. In the 6-month multicenter trial, 18 percent of people given intensive advice, 23 percent given routine advice and 26 percent on the control diet dropped out (Jenkins et al., 2011). In a year-long study of people following the diet on their own, 55 of 66 completed it, and LDL lowering tracked how closely they followed it (Jenkins et al., 2006). The 2018 review describes adherence as over 90 percent in feeding trials, where food was supplied, and under 50 percent in trials where people received advice (Chiavaroli et al., 2018).

An independent pilot trial in England reported 12 percent attrition and adherence of 59 percent to the portfolio foods over 6 months (Stradling et al., 2021). Beyond those figures, what follows is observation: the likely reasons are the work of eating 5 specific foods every day, the cost and availability of nuts and sterol products, meals with family or friends who eat differently, the fiber load for people not used to it, and the slow, invisible nature of cholesterol change, which a person sees only on a lab report.

What happens when you stop

The searches for this page did not identify a study that followed people after they stopped the full Portfolio Diet. Two studies come close. When the developers removed plant sterols for 10 weeks in the middle of an 80-week study, LDL lowering fell from 15.4 percent to 9.0 percent below the starting level, suggesting that each part contributes only while it is eaten (Jenkins, Kendall, Nguyen, et al., 2008).

In the independent English pilot trial, a Mediterranean Portfolio diet supported for 6 months lowered LDL cholesterol by 0.4 mmol/L compared with low-saturated-fat advice at 6 months, but the difference was not sustained at 12 months, after the 6-month intervention period (Stradling et al., 2021). That trial was small and in people living with HIV, so it cannot say how others would fare, but it is the only randomized evidence the searches found beyond 6 months, and it suggests that the effect lasts only as long as the eating does.

Common variations

The original 4-part diet, used in the trials from 2002 to 2011, combined plant sterols, soy protein, viscous fiber and almonds (Chiavaroli et al., 2018). The enhanced or 5-part version adds plant oils rich in monounsaturated fat, tested in 2010 (Jenkins et al., 2010), and is the version the developers now teach. A version without plant sterols was tested in adults with type 2 diabetes after heart bypass surgery (Keith et al., 2015), and the developers' finished artery-imaging trial used an enhanced version that also favored low glycemic index foods (Unity Health Toronto, 2024).

Other research teams have tested looser variations under the name: a Portfolio-Mediterranean diet that left out soy and nuts and added oats or barley, plant sterols, chitosan and green tea (Ferro et al., 2020); a Mediterranean Portfolio diet (Stradling et al., 2021); a Mexican functional-food portfolio, including a reduced-calorie version for type 2 diabetes (Medina-Vera et al., 2019; Vázquez-Manjarrez et al., 2021); and a "portfolio moderate-carbohydrate" diet in Iran, compared with a ketogenic diet in women with polycystic ovary syndrome and with 5:2 fasting and an anti-inflammatory diet in people with prediabetes (Pasdar et al., 2025; Sharifi et al., 2024). The Ketogenic Diet, Intermittent Fasting and Anti-Inflammatory Diet pages describe those comparison diets. A study of one variation does not describe the others.

3. For the coach

SCOPE OF PRACTICE

A health and wellness coach does not prescribe, adjust, endorse, or discourage the Portfolio Diet or any other way of eating, and does not diagnose or treat. The coach helps the client understand what they have chosen, notice how it is going, and decide what to do next. Cholesterol-lowering medicine belongs to the prescriber: a coach does not compare this diet with a statin or any other medicine, and does not suggest it in place of one. Whether to use plant sterol-enriched foods or supplements, and how much, is a question for the client's clinician and a registered dietitian.

The diet's documented cautions fall on particular groups rather than on healthy adults in general. For a client in any of these groups, the coach's role includes making sure the client has medical oversight for the change:

Taking a statin or another cholesterol medicine: any change to the medicine is the prescriber's decision, and European labeling says people on cholesterol medicine should use sterol-enriched foods only under medical supervision. Taking levothyroxine: the client keeps the prescriber informed before a substantial change in eating, because soy, walnuts and fiber can reduce its absorption. Taking warfarin: the client keeps the prescriber informed before a substantial change in eating, because soybeans and soybean oil contain vitamin K and steady intake matters. Taking insulin or a sulfonylurea: the client keeps the prescriber informed before a substantial change in eating.

Kidney disease: the kidney care team, because needs depend on the stage of disease, lab results and dialysis, and beans, lentils, nuts and some fruits are higher in potassium. A nut, peanut or soy allergy: the client's clinician or allergist. Sitosterolemia, a rare inherited condition in which plant sterols build up: the client's physician. Pregnancy or breastfeeding, and children: the obstetric or pediatric clinician and a registered dietitian. Difficulty swallowing, for anyone using psyllium: the client's clinician. A current or past eating disorder or disordered eating: the client's clinician, through the coach's own referral process or an existing care plan, before the plan is discussed further.

Who does what

The physician

Measures and interprets the lipid panel, including LDL cholesterol, non-HDL cholesterol and apolipoprotein B where used, and works out with the person what their own numbers should be. Decides whether a cholesterol-lowering medicine is needed, and whether it should start, stay, change or stop when the person changes how they eat. Reviews other medicines the diet can affect, such as levothyroxine and warfarin, and checks kidney function and potassium where they matter. Evaluates weight a person is losing without meaning to.

The registered dietitian

Designs and monitors the individual eating plan, which a handout or app does not do. For this diet that usually means working out realistic amounts of each of the 5 foods for the person's calorie needs, budget and food traditions; deciding with the clinician whether plant sterol-enriched products or supplements fit; adjusting the plan for kidney disease, diabetes, allergies or digestive conditions; and checking nutrients if the person also cuts back on meat and dairy.

The health and wellness coach

Helps the client get clear about why their cholesterol matters to them and what they want from changing how they eat. Helps them put their own choices, and any plan agreed with their clinician or dietitian, into practice: the shopping, cooking and meals out that make the foods a habit, and notice how it is going in their own terms: energy, digestion, enjoyment of food, cost and the habits themselves, not only the next lab result. With the client's permission, can share general information about what the research does and does not show. Holds the referral line when a question belongs to the prescriber or the dietitian.

What a client bringing this diet needs

Clients usually bring the Portfolio Diet because a number on a lab report worried them. Some were told by a physician to try diet first; some were handed the developers' infographic, which the Canadian Cardiovascular Society endorses, or their app; some read that the diet works "like a statin" and hope to avoid a medicine or stop one they take. Some have a family history of heart disease, and a few have familial hypercholesterolemia, an inherited condition for which the National Heart, Lung, and Blood Institute lists specific medicines (National Heart, Lung, and Blood Institute, 2024).

What the coaching conversation can offer that a handout cannot is room for the person's own reasons and a realistic plan. A coach can help the client separate two questions that are easy to blur: what to eat, which the client can work on with the coach and a dietitian, and what to do about medicine, which belongs with the prescriber. The coach can also help a client put their own choices, and any plan agreed with their clinician or dietitian, into practice in their own kitchen and budget, since the research suggests that eating more of the foods, more consistently, is what moves the result (Jenkins et al., 2006; Jenkins et al., 2011).

What to ask

• "What made the Portfolio Diet the one you wanted to look at?"
• "Of the 5 kinds of food, which ones are already part of your week, and which feel furthest away?"
• "What does your doctor know about this plan, and what would you like to ask them?"
• "How will you know, a few months from now, whether this way of eating is working for you, beyond the next cholesterol test?"

Referral red flags

A client who says they plan to stop, skip or cut down a statin or another cholesterol medicine because they have started this diet is describing a decision that belongs with the prescriber. The coach does not argue for or against it; the coach asks what the prescriber knows and supports the client in raising it. The same goes for a client who wants to add sterol supplements to a medicine, or who asks how much plant sterol to take.

A client on insulin or a sulfonylurea who reports shakiness, sweating or confusion may be having a low blood sugar episode right now. The first step is the one in their own diabetes plan for treating it, with emergency help if the symptoms are severe or the person cannot manage the episode under their plan; telling the prescriber is the separate step that follows (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.). Eating fewer carbohydrates, or skipping or delaying meals, makes low blood sugar more likely for people taking these medicines (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.).

A client taking levothyroxine who plans to eat much more soy, walnuts or fiber keeps the prescriber informed before a substantial change in eating, because the label says those foods can bind the medicine and reduce its absorption (AbbVie, 2024). A client taking warfarin keeps the prescriber informed before a substantial change in eating, because people taking warfarin need a consistent intake of vitamin K, which soybeans and soybean oil supply (National Institutes of Health Office of Dietary Supplements, 2021). A client with kidney disease who plans to add beans, lentils, nuts and oranges every day takes that to the kidney care team, because those foods are higher in potassium and potassium needs depend on lab results (National Kidney Foundation, 2023).

Physical signs also call for a clinician: signs of an allergic reaction after eating nuts, peanuts or soy, such as hives, swelling of the lips or throat, or trouble breathing, where a severe reaction needs emergency help (Food and Drug Administration, 2026); chest pain, vomiting, or trouble swallowing or breathing after taking psyllium, for which its required label warning says to seek immediate medical attention (Water-Soluble Gums, Hydrophilic Gums, and Hydrophilic Mucilloids as Active Ingredients, 2026); and weight a person is losing without meaning to.

Some patterns call for a closer look at the eating itself, such as distress over missing a day's targets, cutting out whole food groups well beyond what the plan asks, or turning down meals with others. Many of these have ordinary reasons, from a dietitian's plan to a work schedule to a household's preferences, so on its own a behavior tells a coach little, and none of this is a diagnosis.

The coach asks what the behavior is for, whether it causes distress, whether it is narrowing what the person eats, and whether it is getting in the way of daily life. Where the answers raise a real concern about disordered eating, the coach follows their own referral process, or brings the concern to the client's clinician under an existing care plan.

Special populations

Pregnancy and breastfeeding

The searches for this page did not identify a study of the Portfolio Diet in pregnancy or breastfeeding. European Union labeling rules require sterol-enriched foods to carry a statement that they may not be nutritionally appropriate for pregnant or breastfeeding women (Regulation (EU) No 1169/2011, 2011). Any plan to change eating during pregnancy or breastfeeding belongs with the obstetric clinician and a registered dietitian.

Children and adolescents

The searches did not identify a study of the Portfolio Diet in children or teenagers; the youngest people studied were young adults, average age 23, in a cross-sectional study (Chen et al., 2025). The same European labeling rules require a statement that sterol-enriched foods may not be nutritionally appropriate for children under 5 (Regulation (EU) No 1169/2011, 2011). A child's or teenager's high cholesterol, including inherited high cholesterol, is managed by the pediatric clinician and a dietitian.

Older adults

The trials enrolled adults with a median age of 57 (Chiavaroli et al., 2018), and cohort studies included postmenopausal women aged 50 to 79 and Hong Kong adults aged 65 and over (Glenn, Lo, et al., 2021; Lo et al., 2021). The searches did not identify a trial designed for older adults. An older adult who takes several medicines, eats little, or has trouble chewing nuts or swallowing fiber supplements is a client for the physician and a dietitian before the plan changes.

Athletes and people with high activity

The searches did not identify a study of the Portfolio Diet in athletes. Whether the plan meets the energy and protein needs of training is a question for a sports dietitian.

People taking medication for diabetes, blood pressure, or blood thinning

The trials excluded people with diabetes (Chiavaroli et al., 2018). Insulin and sulfonylureas can cause low blood sugar, and eating fewer carbohydrates or skipping or delaying meals makes it more likely (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.). People taking warfarin need a consistent intake of vitamin K (National Institutes of Health Office of Dietary Supplements, 2021). The standing rule is that the client keeps the prescriber informed before a substantial change in eating, rather than asking permission for every food choice. Small blood pressure reductions were seen in the trials (Chiavaroli et al., 2018), so a client on blood pressure medicine also keeps the prescriber informed.

People with a history of disordered eating

The searches did not identify a study measuring disordered eating in people following the Portfolio Diet. The pattern mostly adds foods rather than removing them, but it does set daily gram targets and ask people to track 5 food groups, and the developers' app asks users to enter their Portfolio foods every day (Kavanagh, Chiavaroli, et al., 2025). A client with this history is referred to their clinician, or through the coach's own referral process if they have none, before the plan is discussed further.

Client misconceptions

Three come up most. The first is that the diet works as well as a statin, so a person can stop theirs. Two short trials in the developers' clinic compared LDL lowering over 4 weeks with a low dose of an older statin, lovastatin (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003; Jenkins et al., 2005), but neither measured heart attacks, strokes or deaths, and when people followed the diet as advice rather than supplied food, LDL fell by roughly half as much (Chiavaroli et al., 2018). Whether to take, change or stop a medicine is the prescriber's decision.

The second is that it is a weight-loss diet. The trials were designed to keep weight steady, and body weight changed little, by an average of 0.1 kg (Chiavaroli et al., 2018). The third is that adding one cholesterol-lowering food will do what the whole pattern does. Each food lowers LDL cholesterol only a little, and the trials tested them together; when plant sterols were taken out for 10 weeks, the LDL reduction fell by about a third (Chiavaroli et al., 2018; Jenkins, Kendall, Nguyen, et al., 2008). Each myth is taken up in section 6.

🤝 COACHING CONNECTION

Marcus tells his coach that his LDL cholesterol came back high, that he does not want to "be on pills forever," and that he read the Portfolio Diet works as well as a statin, so he plans to stop the one his doctor started last month. The coach reflects how much he wants to feel in charge of his health and asks what his doctor knows about the plan. Marcus has not said anything yet. With his permission, the coach shares that decisions about the medicine belong with the prescriber, and asks what he would want to raise at that conversation. Marcus decides to call his doctor's office about the diet and the medicine before changing anything, and meanwhile to start with oatmeal at breakfast and a handful of nuts in the afternoon. The coach asks how he will notice whether those 2 changes are fitting into his week.

4. The evidence

(select the heading to open or close this section)

The searches for this page, run on October 19, 2026, in Europe PMC, which indexes PubMed, and in ClinicalTrials.gov, found 3 kinds of evidence, and this section keeps them apart. The first is trials of the whole diet, which show with high certainty that it lowers LDL cholesterol in adults with high cholesterol over weeks to months, by more when food is supplied than when people follow advice. The second is cohort studies that score how closely people's ordinary diets resemble the pattern; they link higher scores with somewhat lower rates of heart disease and death, but they cannot show that the diet caused the difference. The third is research on the single foods, which the developers used to build the pattern; this page names it briefly and does not rate it as if it were the diet.

No trial has measured heart attacks, strokes or deaths, and almost all the trials of the full diet, and most of the cohort analyses, come from one research network, the developers'. The certainty ratings are in the Evidence rating and review record at the end of the page.

Evidence from randomized controlled trials

A trial assigns people to the diet by chance and compares them with people assigned to something else, which is the strongest way to test whether the diet itself causes a result. For the Portfolio Diet, the trials come in 2 kinds. In efficacy trials, the research team supplied all the food, so people ate the diet as designed; these lasted 4 weeks. In effectiveness trials, people were taught the diet and bought and cooked their own food; the longest randomized one lasted 6 months.

The developers' 2018 systematic review, commissioned to update European diabetes nutrition guidelines, found 5 reports with 7 trial comparisons in 439 adults with high cholesterol who did not have diabetes or heart disease (Chiavaroli et al., 2018). All were from the Toronto group, all were in Canada, and 3 of the 5 reports were randomized trials; the other 2 compared the diet with a run-in diet in the same people. In every comparison the Portfolio foods were added to a low-saturated-fat National Cholesterol Education Program Step II diet, and the comparison was that diet alone, with calories matched so that weight stayed steady. Five comparisons were 4-week feeding trials and 2 were the 24-week arms of one advice trial.

Cardiometabolic outcomes

LDL cholesterol and other blood fats. Pooled across the 7 comparisons, LDL cholesterol was 0.73 mmol/L lower on the Portfolio diet, with a 95 percent confidence interval of 0.56 to 0.89 lower, which the authors describe as about 17 percent (Chiavaroli et al., 2018). In the feeding trials alone the reduction was 21 percent, and in the advice arms 11 percent; restricting to either kind removed the differences between trials. Non-HDL cholesterol was 0.83 mmol/L lower (0.64 to 1.03), apolipoprotein B 0.19 g/L lower (0.15 to 0.23), total cholesterol 0.81 mmol/L lower and triglycerides 0.28 mmol/L lower (0.14 to 0.42). HDL cholesterol did not change, 0.01 mmol/L (0.03 lower to 0.05 higher). The review rated the certainty high for LDL cholesterol, non-HDL cholesterol, apolipoprotein B, total cholesterol and triglycerides, and moderate for HDL cholesterol.

The single trials behind that figure. In the best-known feeding trial, 46 adults with high cholesterol were randomly assigned for 1 month to a very low-saturated-fat diet, the same diet plus 20 mg of lovastatin, or the Portfolio diet; LDL cholesterol fell 8.0 percent, 30.9 percent and 28.6 percent, and the statin and Portfolio groups each differed from the control group but were not shown to differ from each other (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003). In a crossover trial in 34 adults who had all 3 treatments for a month each, LDL fell 8.5, 33.3 and 29.6 percent; here the statin lowered LDL significantly more than the diet, though 9 of the 34 reached their lowest LDL on the diet (Jenkins et al., 2005).

The advice trial. The largest trial randomized 351 men and postmenopausal women with high cholesterol, at low or intermediate estimated heart risk and not taking cholesterol medicine, at 4 Canadian centers to 6 months of advice on a low-saturated-fat diet, or on the Portfolio diet given in 2 clinic visits or in 7 (Jenkins et al., 2011). Dietitians counseled all 3 groups to follow vegetarian diets that kept weight steady; the control advice stressed low-fat dairy, whole grains, fruit and vegetables and avoided the Portfolio foods. The trial registration describes the control as a low-fat vegetarian diet (University of Toronto, 2018), and a later paper from the same trial calls it a DASH-type diet (Jenkins et al., 2015).

In the 345 analyzed, LDL cholesterol fell 13.8 percent with intensive advice, 13.1 percent with routine advice and 3.0 percent on the control diet, from an average starting level of 171 mg/dL. Compared with the control group, LDL was 18 mg/dL lower with intensive advice (10 to 25 lower) and 16 mg/dL lower with routine advice (9 to 23 lower), in an analysis that allowed for starting level and sex. The 2 Portfolio groups were not shown to differ, and LDL fell more in those who followed the diet more closely. About 23 percent of participants dropped out, and missing values were filled in statistically (Jenkins et al., 2011).

Blood pressure. Pooled, systolic blood pressure was 1.75 mm Hg lower (0.26 to 3.23) and diastolic 1.36 mm Hg lower (0.38 to 2.33), both rated moderate in certainty because the intervals did not rule out unimportant effects (Chiavaroli et al., 2018). These reductions depended on the advice trial: removing its intensive-advice arm made both no longer statistically clear, removing its routine-advice arm did so for diastolic pressure, and restricting the analysis to the feeding trials did so for both. An exploratory analysis of that trial found diastolic pressure 1.8 mm Hg lower than on the control diet at 24 weeks, with systolic pressure 2.1 mm Hg lower, an interval that included no change (Jenkins et al., 2015).

In the year-long study without a control group, blood pressure was 4.2/2.3 mm Hg lower than at the start (Jenkins, Kendall, Faulkner, et al., 2008), a change that cannot be separated from time or other changes.

Inflammation and estimated heart risk. C-reactive protein, a marker of inflammation, was 0.58 mg/L lower on the Portfolio diet (0.15 to 1.01), in 7 comparisons with 435 adults, and the estimated 10-year risk of coronary heart disease, calculated from risk factors, was 1.34 percentage points lower (0.49 to 2.19), in 5 comparisons with 415 adults; both were rated moderate (Chiavaroli et al., 2018).

The review is not consistent with itself on these 2 outcomes: its summary figure, forest plot and rating table give 0.58 mg/L for C-reactive protein, its results text gives 0.53 and its discussion 0.74, and its results text counts 7 comparisons for the risk estimate where its participant summary, figure and tables count 5; this page uses the figure and table values without resolving the difference. The 2023 European diabetes recommendations, citing the same review, list the 10-year risk estimate as very low certainty (Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, 2023), a different judgment from the review's own moderate.

Blood sugar. The 2018 review found no trial that measured fasting glucose, insulin or HbA1c (Chiavaroli et al., 2018). Later, the developers pooled 267 adults with type 2 diabetes from 2 trials of low glycemic index diets and found that those whose eating moved further toward the Portfolio pattern over 6 months had lower HbA1c, 0.04 percentage points per point on a 25-point score (Kavanagh et al., 2024). That is an observational analysis inside trials of a different diet, not a test of the Portfolio Diet. Small trials of portfolio-style diets in other countries, including a reduced-calorie Mexican version in type 2 diabetes and an Iranian version in prediabetes, reported lower HbA1c than their comparison diets (Medina-Vera et al., 2019; Pasdar et al., 2025).

Heart attacks, strokes and deaths. The searches for this page did not identify a trial of the Portfolio Diet that measured cardiovascular events. The developers' artery-imaging trial, which assigned 232 adults at high cardiovascular risk to Portfolio or DASH-like diets, each with structured or routine exercise advice, for 3 years and measured carotid artery wall volume by MRI, was completed in June 2023; no results were posted in its registration when this page was written, and the searches did not find a published report (Unity Health Toronto, 2024).

Independent trials. Trials by other teams tested variations rather than the developers' exact diet. In a randomized trial in Mexico in 62 adults, LDL cholesterol fell 18 percent over about 10 weeks when a functional-food portfolio was added to a low-saturated-fat diet, more than with the same diet and a placebo (Vázquez-Manjarrez et al., 2021). A randomized pilot trial in England in 60 adults living with HIV found LDL 0.4 mmol/L lower at 6 months with a Mediterranean Portfolio diet than with low-saturated-fat advice, a difference not sustained at 12 months (Stradling et al., 2021). These broadly agree on the short-term direction, with smaller and less certain effects outside the developers' feeding trials.

Cancer

The searches for this page did not identify a randomized trial of the Portfolio Diet with any cancer outcome. The trials reviewed do not establish whether the diet affects cancer risk.

Body weight, as measured

This archive reports body weight as something a study measured, in that study's own numbers, and not as a goal or as a measure of whether a diet worked; see the standing rule in the review record at the foot of this page. The Portfolio trials were designed to keep weight steady, with calories matched between diets, so weight change was not their question. Pooled, body weight was 0.10 kg lower on the Portfolio diet, with an interval from 0.48 kg lower to 0.27 kg higher, which the review rated high certainty for no meaningful difference (Chiavaroli et al., 2018). In one feeding trial, weight fell by about 1 kg on both diets (Jenkins, Kendall, Marchie, Faulkner, Vidgen, et al., 2003).

Mortality

The searches for this page did not identify a trial of the Portfolio Diet that measured deaths. The trials reviewed were 4 to 24 weeks long and do not establish whether the diet affects how long people live.

Eating behavior and wellbeing

The first combination diet was rated acceptable and very filling by participants (Jenkins et al., 2002). In a small randomized study within the artery-imaging trial, 14 adults given the developers' app rated it usable, with an average usability score of 80.9 out of 100, but adherence to the diet did not differ clearly from the comparison group over 12 weeks (Kavanagh, Chiavaroli, et al., 2025). The searches did not identify a study that measured mood, quality of life or disordered eating in people following the diet.

Evidence from long-term prospective cohort studies

A cohort study follows people who already eat a certain way and records what happens to them. It can show that a pattern goes with an outcome; it cannot by itself show that the pattern caused it. For the Portfolio Diet, the cohorts measure something different from the trials: a Portfolio Diet Score, calculated from food questionnaires, that gives points for eating more of the 5 plant foods and fewer foods high in saturated fat and cholesterol (Glenn, Boucher, et al., 2021). Almost nobody in these cohorts was following the Portfolio Diet; the score ranks how much their usual diet happened to resemble it, and the authors of one analysis note that intake of some Portfolio foods stayed low even in the top fifth (Glenn, Guasch-Ferré, et al., 2023).

When the score was tested in a 6-month trial of overweight adults, a rise in the score went with a small fall in LDL cholesterol, and the score from the food questionnaire agreed reasonably with a 7-day diet record, a correlation of 0.69 (Glenn, Boucher, et al., 2021). A 2026 systematic review by the same network found 10 observational studies of the diet with 513,284 participants and pooled 4 prospective cohorts; it was read for this page only as its abstract (Chan et al., 2026).

Cardiometabolic outcomes

Cardiovascular disease. In the Women's Health Initiative, 123,330 postmenopausal women followed for an average of 15.3 years, those in the highest quarter of the score had 11 percent lower rates of cardiovascular disease than those in the lowest, a hazard ratio of 0.89 (0.83 to 0.94), and lower rates of coronary heart disease, 0.86, and heart failure, 0.83, but no clear difference in stroke, 0.97 (0.87 to 1.08) (Glenn, Lo, et al., 2021). In 3 Harvard cohorts of about 210,000 nurses and health professionals followed for up to 30 years, the highest fifth of the score had 14 percent lower rates of cardiovascular disease, coronary heart disease and stroke, each about 0.86, after adjusting for lifestyle and overall diet quality (Glenn, Guasch-Ferré, et al., 2023).

The 2026 review's pooled estimate was 6 percent lower cardiovascular disease risk for each standard deviation higher score, a relative risk of 0.94 (0.92 to 0.96) (Chan et al., 2026). In a subset of the Harvard cohorts, higher scores also went with lower blood fats and markers of inflammation (Glenn, Guasch-Ferré, et al., 2023), and in the PREDIMED-Plus trial in Spain, older adults with metabolic syndrome whose Portfolio score rose over a year had small falls in HbA1c, triglycerides and waist size (Glenn, Hernández-Alonso, et al., 2021). In young adults in Toronto, a cross-sectional study found higher scores with slightly lower LDL cholesterol (Chen et al., 2025).

Type 2 diabetes. Among 145,299 women in the Women's Health Initiative followed for an average of 16 years, the highest fifth of the score had lower rates of type 2 diabetes than the lowest, a hazard ratio of 0.77 (0.72 to 0.82) in the main model, which allowed for lifestyle and medical factors, similar to the figures for DASH and Mediterranean scores in the same women. The rate was 23 percent lower before body mass index was added to the model and 13 percent lower after it was added, a hazard ratio of 0.87 (0.82 to 0.93). Whether body size is a confounder, a mediator or both affects how these models are interpreted; the authors note it could be both (Glenn, Li, et al., 2023).

In 2,188 adults in the Tehran Lipid and Glucose Study followed for 8.9 years, the score was not clearly associated with type 2 diabetes, a hazard ratio of 0.91 (0.64 to 1.31) (Malmir et al., 2025).

Cancer

In older adults in Hong Kong, the highest quarter of the score had lower cancer death rates, a hazard ratio of 0.72 (0.54 to 0.96) (Lo et al., 2021). The other studies of cancer the searches found were case-control studies, which compare people who already have cancer with people who do not and ask about past diet: an Italian study of pancreatic cancer using a related cholesterol-lowering diet score (Di Maso et al., 2024) and an Iranian study of breast cancer, which found an association only in premenopausal women (Yarmand et al., 2025). The 2026 review describes the cancer findings as preliminary (Chan et al., 2026). They do not establish whether the diet affects cancer risk.

Body weight, as measured

In the PREDIMED-Plus analysis, a rise in the Portfolio score over a year went with a small fall in body mass index, 0.17 kg/m² per standard deviation (Glenn, Hernández-Alonso, et al., 2021), and in Toronto young adults, higher scores went with slightly lower body weight in a single measurement (Chen et al., 2025). These are associations; the trials, which held weight steady, are the better guide to whether the diet itself changes weight.

Mortality

In 3,991 Hong Kong Chinese adults aged 65 and over followed for a median of 11.1 years, the highest quarter of the score had 28 percent lower death rates from any cause, a hazard ratio of 0.72 (0.62 to 0.86), while the association with cardiovascular death was not statistically clear, 0.90 (0.64 to 1.26); the associations were stronger in men than in women (Lo et al., 2021). In 14,835 United States adults from a national survey followed for 22 years, an 8-point higher score went with 12 percent lower cardiovascular death, a hazard ratio of 0.88 (0.78 to 0.99), and 12 percent lower death from any cause, 0.88 (0.82 to 0.95) (Kavanagh, Zurbau, et al., 2025). The 2026 review pooled a 10 percent lower all-cause death rate per standard deviation, 0.90 (0.86 to 0.95) (Chan et al., 2026).

Highlighted studies

These 5 studies carry the most weight for this diet: the feeding trial that made it famous, the largest advice trial, the published review that rates the evidence, the largest cohort analysis, and the one independent randomized trial that followed people to a year. Each is described so that a reader can judge it without opening the paper.

On a narrow screen, scroll the table sideways to see every column.

Highlighted studies for the Portfolio Diet

Study	Design	Sample size	Duration	Finding	The limitation that matters
Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003	Randomized controlled trial, 3 parallel groups, all food supplied	46 adults with high cholesterol	1 month	LDL cholesterol fell 28.6 percent on the Portfolio diet, 30.9 percent on a low-saturated-fat diet with 20 mg lovastatin, and 8.0 percent on the low-saturated-fat diet alone; the Portfolio and statin groups were not shown to differ	Small and only 4 weeks long, run by the developers, with food supplied; it measured LDL cholesterol, not heart attacks or deaths, and it cannot show how people do on their own
Jenkins et al., 2011	Randomized controlled trial at 4 Canadian centers, dietary advice	351 randomized, 345 analyzed	6 months	LDL cholesterol fell 13.8 percent with intensive advice, 13.1 percent with routine advice and 3.0 percent with low-saturated-fat advice; compared with control advice, LDL was 18 mg/dL lower with intensive advice (10 to 25) and 16 mg/dL lower with routine advice (9 to 23)	18 to 26 percent dropped out, some because LDL rose or they started cholesterol medicine, and missing values were filled in statistically; participants knew which advice they received, though the laboratory staff and statistician did not; run by the developers with food-industry collaborators; participants were mostly white and at low to intermediate heart risk
Chiavaroli et al., 2018	Systematic review and meta-analysis with GRADE	7 trial comparisons from 5 reports, 439 adults	4 to 24 weeks, median 4	LDL cholesterol 0.73 mmol/L lower, about 17 percent; also lower non-HDL cholesterol, apolipoprotein B, triglycerides, blood pressure and C-reactive protein; body weight unchanged	All trials came from the review authors' own group, 2 of the 5 reports were not randomized, and most comparisons lasted 4 weeks
Glenn, Guasch-Ferré, et al., 2023	Prospective cohort studies (Nurses' Health Study, Nurses' Health Study II, Health Professionals Follow-up Study)	About 210,000 adults	Up to 30 years	Highest fifth of the Portfolio Diet Score had about 14 percent lower rates of cardiovascular disease, coronary heart disease and stroke than the lowest fifth	Observational: measures resemblance of usual diets, self-reported by questionnaire, with possible residual confounding; it cannot show that the diet caused the difference
Stradling et al., 2021	Randomized pilot trial, independent of the developers	60 adults living with HIV and high LDL cholesterol	6 months of support, follow-up to 12 months	LDL cholesterol 0.4 mmol/L lower than with low-saturated-fat advice at 6 months; not sustained at 12 months	A small pilot in one population, testing a Mediterranean variation; read for this page as its abstract only

Head to head

Against a statin. Two 4-week trials in the developers' clinic compared the diet with 20 mg of lovastatin, an older statin, on a background low-saturated-fat diet, with all food supplied. In one, LDL lowering was not shown to differ (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003); in the other, the statin lowered LDL significantly more (Jenkins et al., 2005). In a year of following the diet at home, LDL fell 12.8 percent on average, and about a third of the 66 participants reached reductions above 20 percent, which in that group were not shown to differ from their own earlier responses to the statin in a feeding trial (Jenkins et al., 2006).

These trials compared LDL cholesterol over weeks under research conditions. They did not compare heart attacks, strokes or deaths, they are not a basis for choosing between a diet and a medicine, and that choice belongs with the prescriber.

Against other diets. Compared with a DASH-type control diet in the 6-month advice trial, the Portfolio diet lowered diastolic blood pressure slightly more (Jenkins et al., 2015); the DASH Diet page carries DASH's own evidence, including its moderate-certainty effect on LDL cholesterol. In the Women's Health Initiative, Portfolio, DASH and Mediterranean scores had similar associations with lower type 2 diabetes (Glenn, Li, et al., 2023), and in PREDIMED-Plus, rising Portfolio and DASH scores both went with small improvements in risk factors (Glenn, Hernández-Alonso, et al., 2021). An Italian study that was not randomized found larger LDL reductions with a Portfolio-Mediterranean diet than with a Mediterranean Diet alone over about 7 weeks (Ferro et al., 2020).

In Iran, a portfolio moderate-carbohydrate diet lowered LDL cholesterol more than a ketogenic diet over 8 weeks in women with polycystic ovary syndrome, while the ketogenic diet lowered body mass index more (Sharifi et al., 2024). The searches did not identify a trial comparing the Portfolio Diet directly with the Vegan Diet, the Whole-Food Plant-Based Diet or a commercial program. The Vegan Diet and Whole-Food Plant-Based Diet pages describe their own evidence.

Where the evidence is absent

The searches for this page, run on October 19, 2026, in Europe PMC and ClinicalTrials.gov, including a separate search for each special population this archive covers, did not identify a trial of the Portfolio Diet that measured heart attacks, strokes, cancer or death; a published result from the completed artery-imaging trial; a randomized trial of the full diet longer than 6 months or after people stop; a trial of the full diet in people with diabetes, kidney disease or heart disease; or a study of its cost, its environmental footprint, or disordered eating in people following it.

They did not identify a study testing the diet in pregnancy, breastfeeding, children, adolescents, athletes, or older adults as a group, and almost all trials of the full diet come from one research group in Canada. Of the main trial reports, the 2003 feeding trial and the 2011 advice trial were read in full; the 2005 statin comparison and the year-long study were read only as abstracts. The studies reviewed do not establish that the diet prevents heart disease, and they do not establish that it does not. These searches were not an exhaustive check of every database, regulator or special population.

5. Mechanisms

(optional reading; select the heading to open or close this section)

This section explains how the diet is thought to work in the body. You can skip it. Nothing in the coaching sections depends on it.

Established mechanisms

The idea behind the diet is that several foods, each of which lowers LDL cholesterol a little, can be combined so that their effects add up. The developers' review summarizes earlier reviews of the single foods as each lowering LDL cholesterol by roughly 5 to 10 percent, and treats the larger reduction seen in the feeding trials as the expected sum (Chiavaroli et al., 2018). One test of that idea was made in people: when plant sterols were removed from the diet for 10 weeks of an 80-week study, LDL lowering fell from 15.4 to 9.0 percent below the starting level, a difference of about 6 percentage points that the authors attribute to the sterols (Jenkins, Kendall, Nguyen, et al., 2008).

Plant sterols, sometimes called phytosterols, are cholesterol-like compounds from plants that reduce how much cholesterol the intestine absorbs (Ramprasath et al., 2014). People on the Portfolio diet absorb some of them: blood levels of the plant sterols campesterol and beta-sitosterol rose in the diet's trials (Jones et al., 2005; Ramprasath et al., 2014). The diet also replaces some foods high in saturated fat with plant foods, and the low-saturated-fat background diet by itself lowered LDL cholesterol in the comparison groups, by about 8 percent in the feeding trials (Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003; Jenkins et al., 2005).

Proposed mechanisms

Viscous fibers, such as the beta-glucan in oats and barley and the fiber in psyllium, form a gel in the gut. The main proposed mechanism for their effect is that the gel traps bile acids and cholesterol, so that less is absorbed or reabsorbed; the liver then makes more LDL receptors, which pull LDL particles out of the blood (Guarneiri et al., 2024). Why soy protein, other plant proteins and nuts lower LDL cholesterol is less settled, and the sources opened for this page do not establish a single mechanism.

Adding plant oils rich in monounsaturated fat raised HDL cholesterol and lowered the ratio of total to HDL cholesterol in a 1-month feeding trial, without further lowering LDL (Jenkins et al., 2010). For blood pressure, an exploratory analysis found that more nuts, soy and viscous fiber, and more potassium in the urine, went with larger falls in mean arterial pressure (Jenkins et al., 2015); those are associations within one trial, not a tested mechanism. A proposed mechanism never supports a claim in section 4.

6. Risks, contraindications and myths

(select the heading to open or close this section)

Warnings from professional or regulatory bodies

The searches for this page did not identify a warning about the Portfolio Diet as a whole. They covered Europe PMC, ClinicalTrials.gov and the Federal Register, searched for "portfolio diet" and "dietary portfolio" on October 19, 2026, and the United States and European Union labeling rules for its components; they did not cover every regulator, court or country.

Two of its parts carry official labeling rules. In the European Union, foods with added plant sterols or stanols must carry statements that the product "is not intended for people who do not need to control their blood cholesterol level," that "patients on cholesterol lowering medication should only consume the product under medical supervision," that it "may not be nutritionally appropriate for pregnant or breastfeeding women and children under the age of 5 years," and that more than 3 grams a day of added plant sterols or stanols should be avoided (Regulation (EU) No 1169/2011, 2011).

In the United States, over-the-counter products with psyllium and similar gums must carry a choking warning: taking them "without adequate fluid may cause it to swell and block your throat or esophagus and may cause choking," and they should not be taken by people who have difficulty swallowing (Water-Soluble Gums, Hydrophilic Gums, and Hydrophilic Mucilloids as Active Ingredients, 2026). Those are warnings about products, quoted as written, not findings about the diet.

The food labeling rules also bear on what clients read on packages. The Food and Drug Administration authorizes heart-disease health claims for soluble fiber from oats, barley and psyllium, for soy protein, and for plant sterol and stanol esters, each "as part of a diet low in saturated fat and cholesterol" (Health Claims: Plant Sterol/Stanol Esters and Risk of Coronary Heart Disease (CHD), 2026; Health Claims: Soluble Fiber From Certain Foods and Risk of Coronary Heart Disease (CHD), 2026; Health Claims: Soy Protein and Risk of Coronary Heart Disease (CHD), 2026). In 2017 the agency proposed revoking the soy protein claim, tentatively concluding that the evidence no longer supported "significant scientific agreement" (Food Labeling: Health Claims; Soy Protein and Coronary Heart Disease, 2017); as of October 1, 2026, the soy regulation remained in the Code of Federal Regulations (Health Claims: Soy Protein and Risk of Coronary Heart Disease (CHD), 2026).

Nutrient gaps

The pattern adds foods rather than removing food groups, so it does not by itself create a known nutrient gap. Plant sterols do lower blood levels of some carotenoids: a meta-analysis of 41 trials found beta-carotene about 16 percent lower, or 10 percent lower after allowing for the fall in cholesterol, with levels staying within normal ranges and no effect on vitamins A or D (Baumgartner et al., 2017). In the 6-month Portfolio trial, beta-carotene fell slightly but did not differ from the control group once cholesterol was allowed for (Ramprasath et al., 2014). European labeling asks sterol products to advise regular fruit and vegetables "to help maintain carotenoid levels" (Regulation (EU) No 1169/2011, 2011).

A person who also cuts back sharply on meat, fish, eggs and dairy to make room for soy and pulses takes on the planning questions those changes raise, such as vitamin B12, calcium and iodine; the Vegan Diet page discusses them. The searches for this page did not identify a study measuring vitamin or mineral status in people following the full Portfolio Diet over the long term, and planning for a medical condition belongs with a registered dietitian.

Who should not follow it without medical supervision

Anyone taking a statin or another cholesterol-lowering medicine, because any change to that medicine is the prescriber's decision, and because European labeling says people on cholesterol medicine should use plant sterol products only under medical supervision (Regulation (EU) No 1169/2011, 2011). Anyone with sitosterolemia, a rare inherited condition in which plant sterols from foods such as vegetable oils and nuts build up in the blood and tissues because the body cannot remove them normally (MedlinePlus, n.d.). Anyone with a nut, peanut or soy allergy, since all 3 are among the 9 major food allergens in the United States (Food and Drug Administration, 2026).

The 6-month advice trial reported no serious adverse events or hospitalizations. It recorded 36 minor events, with no clear difference in their number between groups; 2 participants had reactions associated with soy or soy and almonds. These observations do not establish long-term safety (Jenkins et al., 2011).

Anyone with kidney disease, whose needs depend on the stage of disease, lab results and dialysis; beans, lentils, legumes, okra, oranges, nuts and peanut butter are among the foods the National Kidney Foundation lists as higher in potassium, and a person with high potassium may be advised to limit them (National Kidney Foundation, 2023). Anyone pregnant or breastfeeding, and children, for whom sterol-enriched foods may not be appropriate and no Portfolio study was found. Anyone who has difficulty swallowing, if psyllium is part of the plan (Water-Soluble Gums, Hydrophilic Gums, and Hydrophilic Mucilloids as Active Ingredients, 2026). Anyone taking the medicines in the next section. The special populations list in section 3 is built from this.

Medication interactions

Cholesterol-lowering medicines. The diet's developers suggest it could be added to statin treatment for people who need more LDL lowering (Chiavaroli et al., 2018), and the National Heart, Lung, and Blood Institute tells people who are prescribed cholesterol medicine to "be sure to continue your healthy lifestyle changes" (National Heart, Lung, and Blood Institute, 2024). Neither is a reason for a client to change a medicine on their own. Bile acid sequestrants, one kind of cholesterol medicine, can make some other medicines less effective (National Heart, Lung, and Blood Institute, 2024); how a person times medicines and fiber supplements is a question for the prescriber or pharmacist.

Levothyroxine. The prescribing label for one brand says that "soybean flour, cottonseed meal, walnuts, and dietary fiber may bind and decrease the absorption" of the medicine, and that dosing may need adjusting when certain foods are eaten regularly within an hour of it (AbbVie, 2024). A client who takes levothyroxine and plans to eat much more soy, walnuts or fiber keeps the prescriber informed before a substantial change in eating.

Warfarin. People taking warfarin need a consistent intake of vitamin K (National Institutes of Health Office of Dietary Supplements, 2021). Roasted soybeans, edamame, soybean oil and canola oil supply some vitamin K, less than leafy greens; a substantial change in how much of them a person eats is a change to tell the prescriber about before it happens.

Insulin and sulfonylureas. The Portfolio diets tested in the trials were not low in carbohydrate; in most of them about half of calories came from carbohydrate (Chiavaroli et al., 2018), but a substantial change in eating, especially one that replaces some carbohydrate foods or changes meal timing, can change blood sugar for someone taking these medicines; eating fewer carbohydrates or skipping or delaying meals makes low blood sugar more likely (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.). The client keeps the prescriber informed before a substantial change. A coach never adjusts medication; the point of listing these medicines is to know when to send the client back to the prescriber.

Signs it is going wrong

Some physical signs need the client's care plan, a medical assessment or emergency help. Hives, swelling of the face, lips or throat, wheezing or trouble breathing after eating nuts, peanuts or soy can be an allergic reaction, and a severe one is an emergency (Food and Drug Administration, 2026). Chest pain, vomiting, or trouble swallowing or breathing after taking psyllium calls for immediate medical attention (Water-Soluble Gums, Hydrophilic Gums, and Hydrophilic Mucilloids as Active Ingredients, 2026).

In someone taking insulin or a sulfonylurea, shakiness, sweating or confusion may be low blood sugar: the person treats it first under their own diabetes plan, gets emergency help if the symptoms are severe or the episode cannot be managed under that plan, and tells the prescriber afterward (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.). Persistent digestive symptoms, unusual tiredness, or weight lost without meaning to are reasons to involve a clinician.

Eating behaviors are a separate question. A client who becomes distressed about missing a day's gram targets, checks every food against the 5 categories long after it stopped being useful, or turns down meals with family or friends because the food does not fit may be describing ordinary effort to learn a new pattern, or something more. Ordinary explanations are common: a work schedule, a tight food budget, a household that eats differently, a cultural or religious practice, or a dietitian's plan for another condition. The coach asks what the behavior is for, whether it causes distress, whether it is narrowing what the person eats, and whether it gets in the way of daily life, and routes a real concern through the coach's own referral process or the client's existing clinical plan. A behavior alone is not a diagnosis.

Myths and where they came from

Myth: "The Portfolio Diet works as well as a statin, so I can stop mine."

Where it came from: the developers' early papers, which compared LDL lowering over 4 weeks with 20 mg of lovastatin (Jenkins et al., 2002; Jenkins, Kendall, Marchie, Faulkner, Wong, et al., 2003). What the evidence shows: those were short trials with all food supplied, one of the 2 found the statin lowered LDL more (Jenkins et al., 2005), and with advice instead of supplied food, LDL fell about half as much (Chiavaroli et al., 2018). No trial has compared heart attacks, strokes or deaths. The decision about any medicine stays with the prescriber.

Myth: "It is a weight-loss diet."

Where it came from: observational studies that link higher Portfolio scores with slightly lower body weight or body mass index (Chen et al., 2025; Glenn, Hernández-Alonso, et al., 2021). What the evidence shows: the trials matched calories to keep weight steady, and body weight changed by an average of 0.1 kg, which the published review rated as high-certainty evidence of no meaningful difference (Chiavaroli et al., 2018). Its purpose, as tested, is lowering LDL cholesterol.

Myth: "Just eat oatmeal, or use a sterol spread, and you are doing the Portfolio Diet."

Where it came from: package health claims for single foods (Health Claims: Plant Sterol/Stanol Esters and Risk of Coronary Heart Disease (CHD), 2026; Health Claims: Soluble Fiber From Certain Foods and Risk of Coronary Heart Disease (CHD), 2026). What the evidence shows: each food lowers LDL cholesterol only a little, and the trials tested them together on a low-saturated-fat background; when one part, plant sterols, was taken away, the LDL reduction fell by about a third (Chiavaroli et al., 2018; Jenkins, Kendall, Nguyen, et al., 2008). Single foods are components, not the diet.

Myth: "Soy acts like estrogen in the body, so it is risky."

Where it came from: long-running controversy about soy isoflavones, plant compounds whose effects researchers compare with those of the hormone estrogen (Viscardi, Back, et al., 2025). What the evidence shows: a 2025 meta-analysis of 40 randomized trials in postmenopausal women found no statistically significant effect of soy isoflavones on 4 measures of estrogen activity, rated high to moderate certainty; it was written by the same research network (Viscardi, Back, et al., 2025). Questions about soy for a person with a hormone-sensitive condition belong with that person's clinician.

Myth: "It is proven to prevent heart attacks."

Where it came from: the strong LDL evidence, and cohort studies linking higher Portfolio scores with less heart disease (Glenn, Guasch-Ferré, et al., 2023). What the evidence shows: lower LDL cholesterol is a recognized target, but no trial of the Portfolio Diet has measured heart attacks, and the cohort studies measure resemblance of usual diets, so they cannot show cause (Chan et al., 2026; Chiavaroli et al., 2018).

7. Key terms

Portfolio Diet

An eating pattern that adds set daily amounts of nuts and seeds, soy or pulse protein, viscous fiber, plant sterols and plant oils rich in monounsaturated fat to a diet low in saturated fat and cholesterol, to lower LDL cholesterol.

LDL cholesterol

Cholesterol carried in low-density lipoprotein particles in the blood; high levels are linked with plaque buildup in the arteries.

Viscous fiber

Soluble fiber that forms a gel in water, found in oats, barley, psyllium, beans, okra, eggplant and some fruits.

Beta-glucan

The viscous fiber in oats and barley.

Psyllium

A fiber from the husk of Plantago seeds, sold as a powder or capsule and used in some fiber supplements.

Pulses

The dried seeds of legumes, such as beans, lentils, chickpeas and dried peas.

Plant sterols

Cholesterol-like compounds from plants, also called phytosterols, that reduce how much cholesterol the gut absorbs; the diet uses fortified foods or supplements to reach about 2 grams a day.

Monounsaturated fat

A kind of unsaturated fat found in olive oil, canola oil, avocado and many nuts.

NCEP Step II diet

The National Cholesterol Education Program's low-saturated-fat, low-cholesterol diet, the background diet and comparison diet in the Portfolio trials.

Statin

A prescription medicine that lowers LDL cholesterol by reducing how much cholesterol the liver makes.

Efficacy trial

A trial that tests a diet under close control, here with all food supplied by the researchers.

Effectiveness trial

A trial that tests a diet as people would follow it in ordinary life, here with advice rather than supplied food.

Non-HDL cholesterol

Total cholesterol minus HDL cholesterol, a measure of the cholesterol carried in all the particles that can build up in arteries.

Apolipoprotein B

A protein found on each LDL and similar particle, so its level reflects the number of those particles.

C-reactive protein

A protein in the blood that rises with inflammation.

Portfolio Diet Score

A score, calculated from food questionnaires, that ranks how closely a person's usual diet resembles the Portfolio pattern.

Hazard ratio

A comparison of how often an event happens over time in 2 groups; a value below 1 means it happened less often in the first group.

Confidence interval

The range of values a study's result is compatible with; a wide interval means the result is uncertain.

Health claim

A statement on a food label, allowed by regulators, that links a food or nutrient to a lower risk of a disease.

Sitosterolemia

A rare inherited condition in which plant sterols build up in the blood and tissues because the body cannot remove them normally.

8. Review questions

1. What are the 5 kinds of food the Portfolio Diet adds, and what background diet are they added to? Give an example of a food from each group.
2. Drawing on section 4: the published review rated the LDL cholesterol result high in certainty. In whom were the trials done, for how long, and who supplied the food? Why was the LDL reduction smaller in the 6-month advice trial than in the 4-week feeding trials?
3. A client tells you she plans to stop her statin because she read that the Portfolio Diet works as well. What does the coach do, what does the coach not do, and who makes the decision about the medicine?
4. What is the difference between a trial of the Portfolio Diet and a cohort study of the Portfolio Diet Score? What can each kind of study show about heart disease, and what can it not show?
5. Drawing on section 6: where did the myth that the diet is "just oatmeal and a sterol spread" come from, and what does the evidence on combining the foods show?

9. References

AbbVie. (2024). Synthroid (levothyroxine sodium) tablets: Prescribing information DailyMed, National Library of Medicine. dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1e11ad30-1041-4520-10b0-8f9d30d30fcc

Baumgartner, S., Ras, R. T., Trautwein, E. A., Mensink, R. P., & Plat, J. (2017). Plasma fat-soluble vitamin and carotenoid concentrations after plant sterol and plant stanol consumption: A meta-analysis of randomized controlled trials. European Journal of Nutrition, 56(3), 909-923. https://doi.org/10.1007/s00394-016-1289-7

Chan, V. W. K., Gebretsadik, G. G., Tse, M. T. H., Lai, G. Z., Leung, L. Y., Glenn, A. J., Kendall, C. W. C., Kavanagh, M. E., Chiavaroli, L., Jenkins, D. J. A., Sievenpiper, J. L., & Lo, K. K. H. (2026). The Portfolio diet and disease outcomes: A systematic review and meta-analysis of observational studies. Critical Reviews in Food Science and Nutrition. Advance online publication. https://doi.org/10.1080/10408398.2026.2712865

Chatzi, C. A., Basios, A., Markozannes, G., Ntzani, E. E., Tsilidis, K. K., Kazakos, K., Agouridis, A. P., Barkas, F., Pappa, M., Katsiki, N., & Rizos, E. C. (2024). Effect of different dietary patterns on cardiometabolic risk factors: An umbrella review of systematic reviews and meta-analyses. Nutrients, 16(22), Article 3873. https://doi.org/10.3390/nu16223873

Chen, V., Chiavaroli, L., Glenn, A. J., Kavanagh, M. E., Zeitoun, T., Mahdavi, S., Kendall, C. W. C., Jenkins, D. J. A., El-Sohemy, A., & Sievenpiper, J. L. (2025). Portfolio diet and LDL-C in a young, multiethnic cohort: Cross-sectional analyses with cumulative exposure modeling. BMC Public Health, 25(1), Article 1761. https://doi.org/10.1186/s12889-025-22479-9

Chiavaroli, L., Nishi, S. K., Khan, T. A., Braunstein, C. R., Glenn, A. J., Blanco Mejia, S., Rahelić, D., Kahleová, H., Salas-Salvadó, J., Jenkins, D. J. A., Kendall, C. W. C., & Sievenpiper, J. L. (2018). Portfolio dietary pattern and cardiovascular disease: A systematic review and meta-analysis of controlled trials. Progress in Cardiovascular Diseases, 61(1), 43-53. https://doi.org/10.1016/j.pcad.2018.05.004

Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes. (2023). Evidence-based European recommendations for the dietary management of diabetes. Diabetologia, 66(6), 965-985. https://doi.org/10.1007/s00125-023-05894-8

Di Maso, M., Augustin, L. S. A., Jenkins, D. J. A., Crispo, A., Toffolutti, F., Negri, E., La Vecchia, C., Ferraroni, M., & Polesel, J. (2024). Adherence to a cholesterol-lowering diet and the risk of pancreatic cancer: A case-control study. Nutrients, 16(15), Article 2508. https://doi.org/10.3390/nu16152508

Dinu, M., Pagliai, G., Angelino, D., Rosi, A., Dall'Asta, M., Bresciani, L., Ferraris, C., Guglielmetti, M., Godos, J., Del Bo', C., Nucci, D., Meroni, E., Landini, L., Martini, D., & Sofi, F. (2020). Effects of popular diets on anthropometric and cardiometabolic parameters: An umbrella review of meta-analyses of randomized controlled trials. Advances in Nutrition, 11(4), 815-833. https://doi.org/10.1093/advances/nmaa006

Ferro, Y., Mazza, E., Salvati, M., Santariga, E., Giampà, S., Spagnuolo, R., Doldo, P., Pujia, R., Coppola, A., Gazzaruso, C., Pujia, A., & Montalcini, T. (2020). Effects of a Portfolio-Mediterranean diet and a Mediterranean diet with or without a sterol-enriched yogurt in individuals with hypercholesterolemia. Endocrinology and Metabolism, 35(2), 298-307. https://doi.org/10.3803/EnM.2020.35.2.298

Food and Drug Administration. (2026, March 11). Food allergies. fda.gov/food/nutrition-food-labeling-and-critical-foods/food-allergies

Food Labeling: Health Claims; Soy Protein and Coronary Heart Disease, 82 Fed. Reg. 50324 (proposed October 31, 2017). federalregister.gov/documents/2017/10/31/2017-23629

Glenn, A. J., Boucher, B. A., Kavcic, C. C., Khan, T. A., Paquette, M., Kendall, C. W. C., Hanley, A. J., Jenkins, D. J. A., & Sievenpiper, J. L. (2021). Development of a Portfolio Diet Score and its concurrent and predictive validity assessed by a food frequency questionnaire. Nutrients, 13(8), Article 2850. https://doi.org/10.3390/nu13082850

Glenn, A. J., Guasch-Ferré, M., Malik, V. S., Kendall, C. W. C., Manson, J. E., Rimm, E. B., Willett, W. C., Sun, Q., Jenkins, D. J. A., Hu, F. B., & Sievenpiper, J. L. (2023). Portfolio Diet Score and risk of cardiovascular disease: Findings from 3 prospective cohort studies. Circulation, 148(22), 1750-1763. https://doi.org/10.1161/CIRCULATIONAHA.123.065551

Glenn, A. J., Hernández-Alonso, P., Kendall, C. W. C., Martínez-González, M. Á., Corella, D., Fitó, M., Martínez, J. A., Alonso-Gómez, Á. M., Wärnberg, J., Vioque, J., Romaguera, D., López-Miranda, J., Estruch, R., Tinahones, F. J., Lapetra, J., Serra-Majem, J. L., Bueno-Cavanillas, A., Tur, J. A., Celada, S. R., . . . Salas-Salvadó, J. (2021). Longitudinal changes in adherence to the portfolio and DASH dietary patterns and cardiometabolic risk factors in the PREDIMED-Plus study. Clinical Nutrition, 40(5), 2825-2836. https://doi.org/10.1016/j.clnu.2021.03.016

Glenn, A. J., Li, J., Lo, K., Jenkins, D. J. A., Boucher, B. A., Hanley, A. J., Kendall, C. W. C., Shadyab, A. H., Tinker, L. F., Chessler, S. D., Howard, B. V., Liu, S., & Sievenpiper, J. L. (2023). The Portfolio Diet and incident type 2 diabetes: Findings from the Women's Health Initiative prospective cohort study. Diabetes Care, 46(1), 28-37. https://doi.org/10.2337/dc22-1029

Glenn, A. J., Lo, K., Jenkins, D. J. A., Boucher, B. A., Hanley, A. J., Kendall, C. W. C., Manson, J. E., Vitolins, M. Z., Snetselaar, L. G., Liu, S., & Sievenpiper, J. L. (2021). Relationship between a plant-based dietary portfolio and risk of cardiovascular disease: Findings from the Women's Health Initiative prospective cohort study. Journal of the American Heart Association, 10(16), Article e021515. https://doi.org/10.1161/JAHA.121.021515

Guarneiri, L. L., Kirkpatrick, C. F., & Maki, K. C. (2024). Viscous fiber from oats and barley: Keep them in your cholesterol-lowering portfolio. Annals of Translational Medicine, 12(1), Article 18. https://doi.org/10.21037/atm-23-1816

Health Claims: Plant Sterol/Stanol Esters and Risk of Coronary Heart Disease (CHD), 21 C.F.R. § 101.83 (2026). ecfr.gov/current/title-21/section-101.83

Health Claims: Soluble Fiber From Certain Foods and Risk of Coronary Heart Disease (CHD), 21 C.F.R. § 101.81 (2026). ecfr.gov/current/title-21/section-101.81

Health Claims: Soy Protein and Risk of Coronary Heart Disease (CHD), 21 C.F.R. § 101.82 (2026). ecfr.gov/current/title-21/section-101.82

Jenkins, D. J., Chiavaroli, L., Wong, J. M., Kendall, C., Lewis, G. F., Vidgen, E., Connelly, P. W., Leiter, L. A., Josse, R. G., & Lamarche, B. (2010). Adding monounsaturated fatty acids to a dietary portfolio of cholesterol-lowering foods in hypercholesterolemia. CMAJ, 182(18), 1961-1967. https://doi.org/10.1503/cmaj.092128

Jenkins, D. J., Jones, P. J., Frohlich, J., Lamarche, B., Ireland, C., Nishi, S. K., Srichaikul, K., Galange, P., Pellini, C., Faulkner, D., de Souza, R. J., Sievenpiper, J. L., Mirrahimi, A., Jayalath, V. H., Augustin, L. S., Bashyam, B., Leiter, L. A., Josse, R., Couture, P., . . . Kendall, C. W. (2015). The effect of a dietary portfolio compared to a DASH-type diet on blood pressure. Nutrition, Metabolism and Cardiovascular Diseases, 25(12), 1132-1139. https://doi.org/10.1016/j.numecd.2015.08.006

Jenkins, D. J., Jones, P. J., Lamarche, B., Kendall, C. W., Faulkner, D., Cermakova, L., Gigleux, I., Ramprasath, V., de Souza, R., Ireland, C., Patel, D., Srichaikul, K., Abdulnour, S., Bashyam, B., Collier, C., Hoshizaki, S., Josse, R. G., Leiter, L. A., Connelly, P. W., & Frohlich, J. (2011). Effect of a dietary portfolio of cholesterol-lowering foods given at 2 levels of intensity of dietary advice on serum lipids in hyperlipidemia: A randomized controlled trial. JAMA, 306(8), 831-839. https://doi.org/10.1001/jama.2011.1202

Jenkins, D. J., Kendall, C. W., Faulkner, D., Vidgen, E., Trautwein, E. A., Parker, T. L., Marchie, A., Koumbridis, G., Lapsley, K. G., Josse, R. G., Leiter, L. A., & Connelly, P. W. (2002). A dietary portfolio approach to cholesterol reduction: Combined effects of plant sterols, vegetable proteins, and viscous fibers in hypercholesterolemia. Metabolism, 51(12), 1596-1604. https://doi.org/10.1053/meta.2002.35578

Jenkins, D. J., Kendall, C. W., Faulkner, D. A., Kemp, T., Marchie, A., Nguyen, T. H., Wong, J. M., de Souza, R., Emam, A., Vidgen, E., Trautwein, E. A., Lapsley, K. G., Josse, R. G., Leiter, L. A., & Singer, W. (2008). Long-term effects of a plant-based dietary portfolio of cholesterol-lowering foods on blood pressure. European Journal of Clinical Nutrition, 62(6), 781-788. https://doi.org/10.1038/sj.ejcn.1602768

Jenkins, D. J., Kendall, C. W., Faulkner, D. A., Nguyen, T., Kemp, T., Marchie, A., Wong, J. M., de Souza, R., Emam, A., Vidgen, E., Trautwein, E. A., Lapsley, K. G., Holmes, C., Josse, R. G., Leiter, L. A., Connelly, P. W., & Singer, W. (2006). Assessment of the longer-term effects of a dietary portfolio of cholesterol-lowering foods in hypercholesterolemia. The American Journal of Clinical Nutrition, 83(3), 582-591. https://doi.org/10.1093/ajcn.83.3.582

Jenkins, D. J., Kendall, C. W., Marchie, A., Faulkner, D., Vidgen, E., Lapsley, K. G., Trautwein, E. A., Parker, T. L., Josse, R. G., Leiter, L. A., & Connelly, P. W. (2003). The effect of combining plant sterols, soy protein, viscous fibers, and almonds in treating hypercholesterolemia. Metabolism, 52(11), 1478-1483. https://doi.org/10.1016/S0026-0495(03)00260-9

Jenkins, D. J., Kendall, C. W., Marchie, A., Faulkner, D. A., Wong, J. M., de Souza, R., Emam, A., Parker, T. L., Vidgen, E., Lapsley, K. G., Trautwein, E. A., Josse, R. G., Leiter, L. A., & Connelly, P. W. (2003). Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA, 290(4), 502-510. https://doi.org/10.1001/jama.290.4.502

Jenkins, D. J., Kendall, C. W., Marchie, A., Faulkner, D. A., Wong, J. M., de Souza, R., Emam, A., Parker, T. L., Vidgen, E., Trautwein, E. A., Lapsley, K. G., Josse, R. G., Leiter, L. A., Singer, W., & Connelly, P. W. (2005). Direct comparison of a dietary portfolio of cholesterol-lowering foods with a statin in hypercholesterolemic participants. The American Journal of Clinical Nutrition, 81(2), 380-387. https://doi.org/10.1093/ajcn.81.2.380

Jenkins, D. J., Kendall, C. W., Nguyen, T. H., Marchie, A., Faulkner, D. A., Ireland, C., Josse, A. R., Vidgen, E., Trautwein, E. A., Lapsley, K. G., Holmes, C., Josse, R. G., Leiter, L. A., Connelly, P. W., & Singer, W. (2008). Effect of plant sterols in combination with other cholesterol-lowering foods. Metabolism, 57(1), 130-139. https://doi.org/10.1016/j.metabol.2007.08.016

Jenkins, D. J., Wolever, T. M., Taylor, R. H., Barker, H., Fielden, H., Baldwin, J. M., Bowling, A. C., Newman, H. C., Jenkins, A. L., & Goff, D. V. (1981). Glycemic index of foods: A physiological basis for carbohydrate exchange. The American Journal of Clinical Nutrition, 34(3), 362-366. https://doi.org/10.1093/ajcn/34.3.362

Jones, P. J., Raeini-Sarjaz, M., Jenkins, D. J., Kendall, C. W., Vidgen, E., Trautwein, E. A., Lapsley, K. G., Marchie, A., Cunnane, S. C., & Connelly, P. W. (2005). Effects of a diet high in plant sterols, vegetable proteins, and viscous fibers (dietary portfolio) on circulating sterol levels and red cell fragility in hypercholesterolemic subjects. Lipids, 40(2), 169-174. https://doi.org/10.1007/s11745-005-1372-6

Kavanagh, M. E., Back, S., Chen, V., Glenn, A. J., Viscardi, G., Houshialsadat, Z., Sievenpiper, J. L., Kendall, C. W. C., Jenkins, D. J. A., & Chiavaroli, L. (2024). The Portfolio Diet and HbA1c in adults living with type 2 diabetes mellitus: A patient-level pooled analysis of two randomized dietary trials. Nutrients, 16(17), Article 2817. https://doi.org/10.3390/nu16172817

Kavanagh, M. E., Chiavaroli, L., Glenn, A. J., Heijmans, G., Grant, S. M., Chow, C. M., Josse, R. G., Malik, V. S., Watson, W., Lofters, A., Holmes, C., Rackal, J., Srichaikul, K., Sherifali, D., Snelgrove-Clarke, E., Udell, J. A., Juni, P., Booth, G. L., Farkouh, M. E., . . . Sievenpiper, J. L. (2022). A web-based health application to translate nutrition therapy for cardiovascular risk reduction in primary care (PortfolioDiet.app): Quality improvement and usability testing study. JMIR Human Factors, 9(2), Article e34704. https://doi.org/10.2196/34704

Kavanagh, M. E., Chiavaroli, L., Quibrantar, S. M., Viscardi, G., Ramboanga, K., Amlin, N., Paquette, M., Sahye-Pudaruth, S., Patel, D., Grant, S. M., Glenn, A. J., Ayoub-Charette, S., Zurbau, A., Josse, R. G., Malik, V. S., Kendall, C. W. C., Jenkins, D. J. A., & Sievenpiper, J. L. (2025). Acceptability of a web-based health app (PortfolioDiet.app) to translate a nutrition therapy for cardiovascular disease in high-risk adults: Mixed methods randomized ancillary pilot study. JMIR Cardio, 9, Article e58124. https://doi.org/10.2196/58124

Kavanagh, M. E., Zurbau, A., Glenn, A. J., Oguntala, J. O., Josse, R. G., Malik, V. S., Chiavaroli, L., Liu, S., Kendall, C. W. C., Jenkins, D. J. A., & Sievenpiper, J. L. (2025). The portfolio dietary pattern and risk of cardiovascular disease mortality during 1988-2019 in US adults: A prospective cohort study. BMC Medicine, 23(1), Article 287. https://doi.org/10.1186/s12916-025-04067-1

Keith, M., Kuliszewski, M. A., Liao, C., Peeva, V., Ahmed, M., Tran, S., Sorokin, K., Jenkins, D. J., Errett, L., & Leong-Poi, H. (2015). A modified portfolio diet complements medical management to reduce cardiovascular risk factors in diabetic patients with coronary artery disease. Clinical Nutrition, 34(3), 541-548. https://doi.org/10.1016/j.clnu.2014.06.010

Kendall, C. W., & Jenkins, D. J. (2004). A dietary portfolio: Maximal reduction of low-density lipoprotein cholesterol with diet. Current Atherosclerosis Reports, 6(6), 492-498. https://doi.org/10.1007/s11883-004-0091-9

Lichtenstein, A. H., Khera, A., Anderson, C. A. M., Appel, L. J., DeSilva, D. M., Gardner, C., Hu, F. B., Jones, D. W., Petersen, K. S., & American Heart Association. (2026). 2026 dietary guidance to improve cardiovascular health: A scientific statement from the American Heart Association. Circulation, 153(18), e1285-e1295. https://doi.org/10.1161/CIR.0000000000001435

Lo, K., Glenn, A. J., Yeung, S., Kendall, C. W. C., Sievenpiper, J. L., Jenkins, D. J. A., & Woo, J. (2021). Prospective association of the Portfolio Diet with all-cause and cause-specific mortality risk in the Mr. OS and Ms. OS Study. Nutrients, 13(12), Article 4360. https://doi.org/10.3390/nu13124360

Malmir, H., Hosseinpour-Niazi, S., Gaeini, Z., Mirmiran, P., & Azizi, F. (2025). Adherence to Portfolio diet and risk of type 2 diabetes mellitus in a population-based prospective cohort study. Journal of Diabetes and Metabolic Disorders, 24(2), Article 178. https://doi.org/10.1007/s40200-025-01695-5

Medina-Vera, I., Sanchez-Tapia, M., Noriega-López, L., Granados-Portillo, O., Guevara-Cruz, M., Flores-López, A., Avila-Nava, A., Fernández, M. L., Tovar, A. R., & Torres, N. (2019). A dietary intervention with functional foods reduces metabolic endotoxaemia and attenuates biochemical abnormalities by modifying faecal microbiota in people with type 2 diabetes. Diabetes & Metabolism, 45(2), 122-131. https://doi.org/10.1016/j.diabet.2018.09.004

MedlinePlus. (n.d.). Sitosterolemia. National Library of Medicine. medlineplus.gov/genetics/condition/sitosterolemia

National Heart, Lung, and Blood Institute. (2024, April 19). Blood cholesterol: Treatment. National Institutes of Health. nhlbi.nih.gov/health/blood-cholesterol/treatment

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.). Low blood glucose (hypoglycemia). National Institutes of Health. niddk.nih.gov/health-information/diabetes/overview/​preventing-problems/low-blood-glucose-hypoglycemia

National Institutes of Health Office of Dietary Supplements. (2021, March 29). Vitamin K: Fact sheet for health professionals. ods.od.nih.gov/factsheets/VitaminK-HealthProfessional

National Kidney Foundation. (2023, May 5). Potassium in your CKD diet. kidney.org/kidney-topics/potassium-your-ckd-diet

Open Library. (n.d.). Search results for "portfolio diet". Internet Archive. Retrieved October 19, 2026, from openlibrary.org/search?title=portfolio+diet

Pasdar, Y., Saber, A., Rezvani, N., Bozorgi, S., Moradi, S., Cheshmeh, S., Niazi, H., & Najafi, F. (2025). Effect of intermittent fasting, portfolio-moderate-carbohydrate, and anti-inflammatory diets on cardio-metabolic status in pre-diabetic patients; an open-label randomized clinical trial. Diabetology & Metabolic Syndrome, 17(1), Article 319. https://doi.org/10.1186/s13098-025-01901-7

Ramprasath, V. R., Jenkins, D. J., Lamarche, B., Kendall, C. W., Faulkner, D., Cermakova, L., Couture, P., Ireland, C., Abdulnour, S., Patel, D., Bashyam, B., Srichaikul, K., de Souza, R. J., Vidgen, E., Josse, R. G., Leiter, L. A., Connelly, P. W., Frohlich, J., & Jones, P. J. (2014). Consumption of a dietary portfolio of cholesterol lowering foods improves blood lipids without affecting concentrations of fat soluble compounds. Nutrition Journal, 13, Article 101. https://doi.org/10.1186/1475-2891-13-101

Regulation (EU) No 1169/2011 of the European Parliament and of the Council of 25 October 2011 on the provision of food information to consumers. (2011). Official Journal of the European Union, L 304, 18. Annex III, consolidated text of January 1, 2018. eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:02011R1169-20180101

Sharifi, M., Saber, A., Moludi, J., Salimi, Y., & Jahan-Mihan, A. (2024). The effects of portfolio moderate-carbohydrate and ketogenic diets on anthropometric indices, metabolic status, and hormonal levels in overweight or obese women with polycystic ovary syndrome: A randomized controlled trial. Nutrition Journal, 23(1), Article 152. https://doi.org/10.1186/s12937-024-01056-7

Stradling, C., Thomas, G. N., Hemming, K., Taylor, S., & Taheri, S. (2021). Randomized parallel-group pilot trial (Best foods for your heart) comparing the effects of a Mediterranean Portfolio diet with a low saturated fat diet on HIV dyslipidemia. Clinical Nutrition, 40(3), 860-869. https://doi.org/10.1016/j.clnu.2020.08.038

Unity Health Toronto. (2024). The Combined Portfolio Diet and Exercise Study (ClinicalTrials.gov Identifier NCT02481466) [Clinical trial registration]. ClinicalTrials.gov. clinicaltrials.gov/study/NCT02481466

University of Toronto. (2018). Portfolio 5: Multicentre dietary advice on serum lipids in hyperlipidemia (ClinicalTrials.gov Identifier NCT00438425) [Clinical trial registration]. ClinicalTrials.gov. clinicaltrials.gov/study/NCT00438425

Vázquez-Manjarrez, N., Guevara-Cruz, M., Flores-López, A., Pichardo-Ontiveros, E., Tovar, A. R., & Torres, N. (2021). Effect of a dietary intervention with functional foods on LDL-C concentrations and lipoprotein subclasses in overweight subjects with hypercholesterolemia: Results of a controlled trial. Clinical Nutrition, 40(5), 2527-2534. https://doi.org/10.1016/j.clnu.2021.02.048

Viscardi, G., Back, S., Ahmed, A., Yang, S., Blanco Mejia, S., Zurbau, A., Khan, T. A., Selk, A., Messina, M., Kendall, C. W., Jenkins, D. J., Sievenpiper, J. L., & Chiavaroli, L. (2025). Effect of soy isoflavones on measures of estrogenicity: A systematic review and meta-analysis of randomized controlled trials. Advances in Nutrition, 16(1), Article 100327. https://doi.org/10.1016/j.advnut.2024.100327

Viscardi, G., Eskandari, S., Chen, A., Chiang, N., Kavanagh, M. E., Back, S., Vallis, M., Scourboutakos, M. J., Malik, V. S., Kendall, C. W. C., Jenkins, D. J. A., Sievenpiper, J. L., & Chiavaroli, L. (2025). Translating cardiovascular clinical practice guidelines on nutrition therapy: Validation of the Portfolio Diet toolkit for healthcare providers. CJC Open, 7(12), 1662-1671. https://doi.org/10.1016/j.cjco.2025.08.015

Water-Soluble Gums, Hydrophilic Gums, and Hydrophilic Mucilloids as Active Ingredients; Required Warnings and Directions, 21 C.F.R. § 201.319 (2026). ecfr.gov/current/title-21/section-201.319

Yarmand, S., Atashi, N., Kazemi, I., Mahmoudi-Zadeh, M., Yazdani, S., Behzadi Nejad, H., Shateri, Z., Amiri, S. M., Rashidkhani, B., & Nouri, M. (2025). Association of portfolio diet score with breast cancer risk: Insights from a case-control analysis. BMC Cancer, 25(1), Article 274. https://doi.org/10.1186/s12885-025-13634-x

Evidence rating and review record

Each row identifies the method, source, outcome or recommendation, and comparison being rated. Ratings from different tools are not interchangeable. The source is either a named published review or this page's own judgment. The rating in the banner is for this diet's main claimed outcome or recommendation, and the banner names the tool that produced it. A high rating means the research is solid; it does not mean the diet is right for everyone, and it does not mean a coach should recommend it.

The rated outcomes on this page use GRADE, the same 4-level scale as the clinical guidelines you will meet in practice. High means further research is very unlikely to change the conclusion. Moderate means further research may change it. Low means further research is likely to change it. Very low means the conclusion is uncertain. GRADE rates each outcome separately, so a diet can be high for one outcome and very low for another. A row marked Not rated is an abstention, not a certainty level.

The first 9 rows are carried exactly as the 2018 review published them, for its own comparison of the Portfolio diet on a low-saturated-fat background against that background alone, mostly over 4 weeks; they are not transferred to longer periods, to advice without supplied food, or to heart attacks. The remaining rows are this page's own judgments. Two umbrella reviews, which grade whole bodies of meta-analyses with their own credibility criteria rather than GRADE, reached more cautious labels: one graded the Portfolio evidence weak (Dinu et al., 2020), and one classed the LDL cholesterol, non-HDL cholesterol and apolipoprotein B results as highly suggestive rather than strong, rating the 2018 review's methodological quality critically low (Chatzi et al., 2024). Those labels come from different tools and are not converted into GRADE.

On a narrow screen, scroll the table sideways to see every column.

Evidence ratings, their sources and reasons

Outcome	Certainty	Who rated it, and for which comparison	Why it is rated this way
LDL cholesterol (the banner rating)	High	GRADE, published review (Chiavaroli et al., 2018). Outcome: LDL cholesterol measured in blood; pooled difference 0.73 mmol/L lower, 95 percent confidence interval 0.56 to 0.89 lower, about 17 percent. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published: the review started the evidence from controlled trials at high and found no serious risk of bias, did not lower it for inconsistency because the differences between trials were explained by feeding versus advice, did not lower it for indirectness although it noted that one investigator group ran every trial and none included people with diabetes, pointing to the large multicenter advice trial, and found no serious imprecision against its minimally important difference of 0.1 mmol/L. Its rating table, in the supplement read for this page, shows no domain lowered and says publication bias could not be assessed with fewer than 10 trials, so it was not lowered. This page carries the label as published and adds what a reader should weigh: the review authors include the diet's developers, 2 of the 5 reports were not randomized, and most comparisons lasted 4 weeks.
Non-HDL cholesterol	High	GRADE, published review (Chiavaroli et al., 2018). Outcome: non-HDL cholesterol; 0.83 mmol/L lower, 0.64 to 1.03. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published, for the same reasons as the first row: started high, with no domain lowered. The same added cautions apply.
Apolipoprotein B	High	GRADE, published review (Chiavaroli et al., 2018). Outcome: apolipoprotein B; 0.19 g/L lower, 0.15 to 0.23. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published, for the same reasons as the first row: started high, with no domain lowered. The same added cautions apply.
Triglycerides	High	GRADE, published review (Chiavaroli et al., 2018). Outcome: triglycerides; 0.28 mmol/L lower, 0.14 to 0.42. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published, for the same reasons as the first row: started high, with no domain lowered; the differences between trials were explained when the routine-advice arm was removed. The same added cautions apply.
Systolic blood pressure	Moderate	GRADE, published review (Chiavaroli et al., 2018). Outcome: systolic blood pressure; 1.75 mm Hg lower, 0.26 to 3.23. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published: started high and lowered 1 level for serious imprecision, because the 95 percent confidence interval included the review's prespecified minimally important difference of 2 mm Hg, so unimportant effects could not be ruled out. Risk of bias, inconsistency and indirectness were not lowered, for the reasons given in the first row, and publication bias could not be assessed with fewer than 10 trials and was not lowered. This page carries the label as published. The reduction was no longer statistically clear when the intensive-advice arm was removed.
Diastolic blood pressure	Moderate	GRADE, published review (Chiavaroli et al., 2018). Outcome: diastolic blood pressure; 1.36 mm Hg lower, 0.38 to 2.33. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published: started high and lowered 1 level for serious imprecision, because the 95 percent confidence interval included the review's prespecified minimally important difference of 2 mm Hg, so unimportant effects could not be ruled out. Risk of bias, inconsistency and indirectness were not lowered, for the reasons given in the first row, and publication bias could not be assessed with fewer than 10 trials and was not lowered. This page carries the label as published. The reduction was no longer statistically clear when either advice arm was removed.
C-reactive protein	Moderate	GRADE, published review (Chiavaroli et al., 2018). Outcome: C-reactive protein; 0.58 mg/L lower, 0.15 to 1.01, about 32 percent, as given in the review's summary figure, forest plot and rating table. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 435 adults.	As published: started high and lowered 1 level for serious imprecision, because the 95 percent confidence interval included the review's prespecified minimally important difference of 0.5 mg/L, so unimportant effects could not be ruled out. Risk of bias, inconsistency and indirectness were not lowered, for the reasons given in the first row, and publication bias could not be assessed with fewer than 10 trials and was not lowered. This page carries the label as published. The review reports 0.53 mg/L lower in its results paragraph, 0.58 lower in its summary figure, forest plot and rating table, and 0.74 lower in its discussion; this page reports 0.58, the value the review's rating table uses, without resolving that inconsistency.
Estimated 10-year risk of coronary heart disease	Moderate	GRADE, published review (Chiavaroli et al., 2018). Outcome: Framingham-estimated 10-year risk, calculated from risk factors, not measured events; 1.34 percentage points lower, 0.49 to 2.19. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 3 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 5 comparisons, 415 participants in the participant summary, Figure 2, forest plot and rating table; the results paragraph says 7 comparisons.	As published: started high and lowered 1 level for serious imprecision, because the 95 percent confidence interval included the review's prespecified minimally important difference of 1 percentage point, so unimportant effects could not be ruled out. Risk of bias, inconsistency and indirectness were not lowered, for the reasons given in the first row, and publication bias could not be assessed with fewer than 10 trials and was not lowered. This page carries the label as published. The 2023 European diabetes recommendations list this outcome as very low certainty while citing the same review (Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, 2023); this page carries the review's own label and reports the difference.
Body weight	High, for little or no difference	GRADE, published review (Chiavaroli et al., 2018). Outcome: body weight in trials designed to keep weight steady; 0.10 kg lower, from 0.48 lower to 0.27 higher. Population: adults with high cholesterol, without diabetes or heart disease, median age 57, all in Canada. Intervention: the Portfolio foods added to a National Cholesterol Education Program Step II diet, with all food supplied in 5 comparisons and as advice in 2. Comparison: the Step II diet alone, calories matched. Follow-up: 4 to 24 weeks, median 4. 7 comparisons from 5 reports, 439 adults.	As published: started high, no domain lowered, and the interval was judged to exclude an important difference, using the review's boundary of 0.5 kg. Weight is reported as a measurement; the trials did not set out to change it.
LDL cholesterol at 6 months with dietary advice	Moderate	GRADE, this page's own judgment, from one randomized trial (Jenkins et al., 2011), read in full. Population: 351 men and postmenopausal women with high cholesterol at low or intermediate estimated heart risk, not taking cholesterol medicine, at 4 Canadian centers; average starting LDL 171 mg/dL. Intervention: advice on the Portfolio diet, in 2 or 7 clinic visits. Comparison: advice on a low-saturated-fat diet. Outcome: LDL cholesterol measured in blood, difference between groups in mg/dL at 6 months, allowing for starting level and sex. Follow-up: 6 months; 345 analyzed.	Randomized evidence starts high. Risk of bias is serious, lowered 1 level, for missing data: 22.6 percent dropped out, 18 to 26 percent by group, some for reasons tied to the outcome, such as a rise in LDL or starting cholesterol medicine, and missing values were filled in by multiple imputation, which assumes that, given age, sex and the values measured, the people who left resembled those who stayed. Allocation used numbered envelopes from a statistician's random blocks, and the statistician, investigators and laboratory staff did not know the groups; participants and dietitians did, which a diet trial cannot avoid and which is not a deduction by itself. Inconsistency cannot be assessed from one trial and is not a deduction. Indirectness is not serious for the population, advice and comparison named. Imprecision is not serious: compared with control advice, LDL was 18 mg/dL lower with intensive advice (10 to 25) and 16 mg/dL lower with routine advice (9 to 23). The threshold is 5 percent of the 171 mg/dL starting level, about 8.6 mg/dL, the difference the trial was designed to detect and the low end of the 5 to 10 percent the developers attribute to each single food; both intervals exclude smaller effects, the routine-advice interval only just. Publication bias cannot be assessed; not lowered. No factor raises it. High minus 1 is Moderate.
LDL cholesterol 12 months after starting, 6 months after supported eating ended	Low, provisional	GRADE, this page's own judgment, from one randomized pilot trial independent of the developers (Stradling et al., 2021), read as its abstract. Population: 60 adults in England living with HIV, on antiretroviral treatment, with LDL above 3 mmol/L. Intervention: 6 months of support for a Mediterranean Portfolio diet with nuts, stanols, soy, oats and beans. Comparison: advice to cut saturated fat below 10 percent of calories. Outcome: LDL cholesterol. Follow-up: 12 months.	Randomized evidence starts high. Risk of bias is serious, lowered 1 level, provisionally: the trial's main aim was feasibility, the LDL comparisons were a secondary analysis, and how allocation was concealed and missing values handled could not be checked from the abstract. Inconsistency cannot be assessed from one trial. Indirectness is not serious for the question as stated, which names people living with HIV and the Mediterranean Portfolio variation; this row does not answer the question for other populations or for the developers' diet. Imprecision is serious, lowered 1 level: the difference was 0.05 mmol/L lower, 0.33 lower to 0.23 higher, an interval that includes both no effect and differences larger than the 0.2 mmol/L threshold used here, about 5 percent of a typical starting LDL. Publication bias cannot be assessed; not lowered. High minus 2 is Low. Provisional until the full report's methods are read.
Developing cardiovascular disease, higher compared with lower Portfolio Diet Score	Low	GRADE, this page's own judgment, from 2 cohort analyses covering 4 cohorts (Glenn, Guasch-Ferré, et al., 2023; Glenn, Lo, et al., 2021), with a pooled estimate read as an abstract (Chan et al., 2026). Population: about 333,000 United States adults, mostly women, initially free of cardiovascular disease. Exposure: the highest compared with the lowest quarter or fifth of a score of how closely usual diets resembled the pattern. Outcome: total cardiovascular disease events. Follow-up: about 15 to 30 years.	Observational evidence starts low. Risk of bias is not lowered further: diet was self-reported and residual confounding is possible, the usual reasons for starting low, and both analyses adjusted for major risk factors and one for overall diet quality. Inconsistency is not serious: hazard ratios of 0.89 (0.83 to 0.94) and 0.86 (0.81 to 0.92) agree. Indirectness is not serious for this question, which concerns the score, not the prescribed diet. Imprecision is not serious: both upper limits, 0.94 and 0.92, lie below a hazard ratio of 0.95. The 0.95 hazard-ratio boundary is this page's analytical convention, not a published patient-important threshold; it concerns the highest-versus-lowest score categories specified here. It asks whether an interval rules out associations so small, under a 5 percent relative difference between the top and bottom of the score, that residual confounding could readily account for them; the Women's Health Initiative upper limit, 0.94, is close to it. Publication bias cannot be assessed; all analyses come from one research network. No factor raises it; a dose-response pattern was reported but not used to raise the rating. Low.
Death from any cause, highest compared with lowest quarter of the Portfolio Diet Score (Hong Kong cohort)	Low	GRADE, this page's own judgment, from one cohort analysis (Lo et al., 2021). Population: 3,991 Hong Kong Chinese adults aged 65 and over. Exposure: highest compared with lowest quarter of a score of how closely usual diets resembled the pattern. Outcome: death from any cause. Follow-up: a median of 11.1 years.	Observational evidence starts low. Risk of bias is not lowered further, for the same reasons as the cardiovascular row above. Inconsistency cannot be assessed within one cohort; the association was clear in men and mostly not in women. Indirectness is not serious for this question. Imprecision is not serious: the hazard ratio was 0.72 (0.62 to 0.86), and the upper limit lies below the 0.95 convention for highest-versus-lowest categories explained in the cardiovascular row. Publication bias cannot be assessed; not lowered. No factor raises it. Low.
Death from any cause, per 8 points higher on the Portfolio Diet Score (United States cohort)	Low	GRADE, this page's own judgment, from one cohort analysis (Kavanagh, Zurbau, et al., 2025). Population: 14,835 United States adults from a national survey. Exposure: each 8 points higher on a score that runs from 6 to 30, a step the authors describe as about 33 percentile points. Outcome: death from any cause. Follow-up: 22 years.	Observational evidence starts low. Risk of bias is not lowered further, for the same reasons as the cardiovascular row above. Inconsistency cannot be assessed within one cohort; the authors reported a difference by race and ethnicity. Indirectness is not serious for this question. Imprecision is not serious, a close call: the hazard ratio was 0.88 (0.82 to 0.95) per 8 points in the abstract, 0.81 to 0.95 in the paper's table. This page applies its 0.95 convention to this 8-point step as an analytical choice, not a published threshold; the upper limit reaches 0.95 but does not cross it, and the same paper's highest-versus-lowest third comparison, 0.86 (0.78 to 0.96), crosses it slightly. The 2026 review's pooled estimate per standard deviation, read as an abstract, is a different contrast and is not used for either row (Chan et al., 2026). Publication bias cannot be assessed; not lowered. No factor raises it. Low.
Developing type 2 diabetes, higher compared with lower Portfolio Diet Score	Low	GRADE, this page's own judgment, from 2 cohorts (Glenn, Li, et al., 2023, read in full; Malmir et al., 2025, read as its abstract). Population: 145,299 postmenopausal women in the United States and 2,188 adults in Tehran, free of diabetes at the start. Exposure: highest compared with lowest fifth (United States) or third (Tehran) of the score. Outcome: new type 2 diabetes. Follow-up: an average of 16 years and 8.9 years.	Observational evidence starts low. Risk of bias is not lowered further, for the reasons in the cardiovascular row. The Women's Health Initiative estimates come from 2 models: 0.77 (0.72 to 0.82) in the main model, which allowed for lifestyle and medical factors, and 0.87 (0.82 to 0.93) after body mass index was added; the authors note body size may be a confounder, a mediator or both. Inconsistency is not serious: the Tehran estimate from its fully adjusted model, 0.91 (0.64 to 1.31), overlaps both, and a smaller nonsignificant study is not by itself inconsistency. Indirectness is not serious for this question, though most of the evidence is from older women. Imprecision is not serious: the upper limits in both Women's Health Initiative models, 0.82 and 0.93, lie below 0.95, this page's analytical convention for highest-versus-lowest score categories, not a published patient-important threshold, as explained in the cardiovascular row. Publication bias cannot be assessed. No factor raises it. Low.
HbA1c in adults with type 2 diabetes	Very low	GRADE, this page's own judgment, from one pooled observational analysis inside 2 trials of other diets (Kavanagh et al., 2024). Population: 267 adults with type 2 diabetes taking oral medicines. Exposure: change in a 25-point clinical Portfolio Diet Score. Outcome: change in HbA1c. Follow-up: 6 months.	An analysis of self-selected diet change starts low. Risk of bias is serious, lowered 1 level: people chose how far their eating moved toward the pattern, within trials of low glycemic index diets, so confounding by other changes may explain the association. Inconsistency cannot be assessed from one analysis. Indirectness is not serious for the question as stated, which concerns change in the score rather than the prescribed diet; the analysis remains observational. Imprecision was not resolved for a meaningful score change: the estimate is 0.04 percentage points lower HbA1c per score point (0.02 to 0.07), and this page has not set an HbA1c threshold for a defined change in the score; the rating is already Very low because of confounding. Publication bias cannot be assessed. Low minus 1 is Very low.
Heart attacks, strokes and deaths in trials of the diet	Not rated	This page's own judgment. Population: any. Intervention: the Portfolio Diet. Comparison: any. Outcome: cardiovascular events or death. Follow-up: any.	The searches did not identify a trial with these outcomes. Not rated is an abstention: there is no trial evidence to rate, which does not mean the diet has no effect.
Progression of artery wall disease	Not rated	This page's own judgment. Population: adults at high cardiovascular risk. Intervention: an enhanced Portfolio diet with exercise. Comparison: a DASH-like diet with exercise. Outcome: carotid vessel wall volume by MRI. Follow-up: 3 years.	The trial that measured this was completed in June 2023, with no results posted or published report found (Unity Health Toronto, 2024). This row should be revisited when it reports.
Cancer	Not rated	This page's own judgment. Population: adults. Exposure: the diet or the score. Outcome: new cancer or cancer death. Follow-up: any.	No trial was found. The observational evidence is one cohort analysis of cancer deaths (Lo et al., 2021) and 2 case-control studies using different scores (Di Maso et al., 2024; Yarmand et al., 2025), which the 2026 review calls preliminary (Chan et al., 2026). That is not an adequate body to rate.
Side effects and long-term safety	Not rated	This page's own judgment. Population: adults with high cholesterol following the diet. Intervention: the Portfolio Diet. Comparison: a low-saturated-fat diet. Outcome: adverse events in the short term, and long-term safety. Follow-up: 6 months in the one trial report read for adverse events; no long-term study found.	Short term, reported as observations: the 6-month advice trial reported no serious adverse events or hospitalizations and 36 minor events, with no clear difference in their number between groups; 2 participants had reactions associated with soy or soy and almonds (Jenkins et al., 2011). Short trials also measured plant sterol and fat-soluble vitamin levels (Ramprasath et al., 2014). One trial's event counts, not designed as a safety comparison, are too few to rate. Long term: no study designed to test long-term safety was found, so there is no body of evidence to rate. Not rated does not mean the diet is safe or unsafe.
Disordered eating	Not rated	This page's own judgment. Population: people with and without a history of disordered eating. Intervention: the Portfolio Diet. Comparison: any. Outcome: onset or worsening of disordered eating. Follow-up: any.	The searches did not identify a study. Not rated is an abstention, not a finding of safety.

How this archive talks about weight. This page reports what studies measured, including body weight, and does not treat weight change as a goal or as a measure of a diet's success. Weight a person loses without meaning to can be a sign of a health problem, and where it is, the page names it as a reason to see a clinician. People are described the way the study described them.

Last reviewed: Reviewed and approved by James Peth, MS, MPH, PhD, on October 21, 2026, after a combined final review of the fifth batch of 5 archive pages, the repairs it asked for and a targeted recheck. Written October 19, 2026. The first 9 ratings are carried from a published review; the others are this page's own judgments, assigned from the evidence. Several sources could be read only as abstracts, including the 2005 statin comparison, the year-long study and the independent HIV pilot trial, and the page says so where each is used.

What changed: After the combined final review, the 2011 advice trial, the 2003 feeding trial, the Women's Health Initiative diabetes report and the 2018 review's supplement were read in full. The C-reactive protein and heart risk rows now give the review's own counts and its conflicting figures. The advice trial is described from its full report, with its differences between groups, safety observations and vegetarian study diets. The diabetes cohort is reported with and without body mass index in the model. Those 2 ratings are no longer provisional. The HIV pilot trial row (now Low, provisional) and the HbA1c row are no longer lowered for the population or exposure their questions name, and death from any cause is split by cohort. The coach's part in food choices was narrowed, and the American Heart Association is named as an author of its 2026 statement. First version October 19, 2026.

Next scheduled review: Each year; the developers' artery-imaging trial (NCT02481466) was completed in 2023 and should be checked for published results, and the United States soy protein health claim, proposed for revocation in 2017, should be checked for a final decision. Pages built on annual guidance are rechecked each year.

---

Diet Reference Archive | One James

Prepared by James Peth, MS, MPH, PhD

© 2026–2027 One James. All rights reserved.

This page is for education and coaching practice. It is not medical or dietary advice.