Noom

Diet Reference Archive: a commercial program

Noom

A paid smartphone program that sets a daily calorie range, sorts foods by color according to how many calories they carry for their weight, and adds short psychology-based lessons, food and weight logging, and coaching by chat. It is sold for weight loss, and the same company now also offers prescription weight-loss medication through a service called Noom Med.

Family: Commercial programs and products Promoted by: Noom, Inc., the company that sells it First appeared: research the company dates to 2008; its app was in use by 2012 GRADE: Very low for a difference in measured body weight at the end of a 16-week to 6-month program, adults with overweight, obesity or prediabetes, Noom programs as the company delivers them against usual care with printed materials, educational emails or newsletters; this page's own judgment

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. Most of the research on this program was run or paid for by the company that sells it, and section 4 says so study by study. Back to the Diet Reference Archive index.

In brief

What is it? A subscription weight-loss program run through a smartphone app by Noom, Inc. Members get a daily calorie range, log their food and their weight, read short daily lessons the company describes as based on cognitive behavioral therapy, and can message coaches. The company says no food is off limits; it labels foods green, yellow or orange by how many calories they carry for their weight (Noom Team, 2026b, 2026c). A separate service, Noom Med, adds clinicians who can prescribe weight-loss medicines, including the drugs known as GLP-1 medications (Noom, n.d.-e).

Does it work? In randomized trials, most of them small and most run or paid for by the company, people assigned to a Noom program lost somewhat more weight by the end of a 4-month to 6-month program than people assigned to usual care or to educational emails or newsletters. In one 202-person trial in adults with prediabetes, the measured difference was 2.6 kg at 6 months and 1.8 kg at 12 months, and HbA1c, a measure of average blood sugar, differed between the groups on average by less than a hundredth of a percentage point at either time (Toro-Ramos et al., 2020). This page rates the short-term weight evidence very low in certainty. The searches logged for this page did not identify a study estimating the program's effect on heart attacks, strokes, cancer or survival; recording deaths among adverse events is a different question.

Is it safe? Some studies assessed safety or reported adverse events, but their size, follow-up and comparisons do not establish the consumer program's safety for members in general. The one that posted adverse events lists no deaths and 3 serious events, without saying whether they were related to the program, in a trial where both groups used a version of the app (Icahn School of Medicine at Mount Sinai, 2025). Eating disorder questionnaire scores fell over time in both groups of the trials that measured them, and no trial found a worsening compared with its control group; that does not show the program is safe for anyone at higher risk (McCallum et al., 2026; Sysko, Michaelides, et al., 2022). The company says the program is not for anyone under 18, pregnant, underweight or living with an eating disorder (Noom, n.d.-d).

Medication needs care. A client taking insulin or a sulfonylurea keeps the prescriber informed before eating substantially less, because too little carbohydrate can bring on low blood sugar (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.); the coach never changes a medication. If a low blood sugar episode happens, the client's own treatment plan comes first, with emergency help if symptoms are severe or the person cannot manage the episode under that plan, and the prescriber is told afterward. Medication prescribed through Noom Med, including compounded versions the company says are not approved by the U.S. Food and Drug Administration, belongs entirely to the prescriber (Noom, n.d.-d).

Who should not do it without talking to a doctor or dietitian first? Anyone under 18, pregnant or breastfeeding, or underweight. Anyone with a current or past eating disorder. Anyone taking insulin, a sulfonylurea or other medicine for diabetes, or medicine for blood pressure; they keep the prescriber informed before a substantial change in eating. Anyone with kidney disease, whose needs depend on its stage, laboratory results and whether they are on dialysis. And anyone thinking about weight-loss medication, through Noom Med or anywhere else, which is a decision for a clinician who knows their full history.

Related pages: WeightWatchers, Calorie Counting and Tracking Apps, Meal Replacement Programs, Intuitive Eating, Mindful Eating and Very Low Calorie Diets. Full list on the archive index.

1. Overview

What you eat

Noom does not hand members a list of foods. A new member answers an online questionnaire about current weight, goal weight and dieting history and receives a plan, usually about 16 weeks long, with a suggested calorie range for each day; the company describes the pace it maps out as one to two pounds a week (Noom Team, 2026a, 2026b). Every food a member logs in the app is labeled green, yellow or orange by calorie density, meaning how many calories a food carries for its weight. Green foods, such as most vegetables, fruits and whole grains, carry the fewest calories for the amount; yellow foods, such as chicken, fish and low-fat dairy, carry more; orange foods, such as chocolate, salami and churros, carry the most. The app suggests a daily share for each color (Noom Team, 2026c).

The company says the colors do not mean good or bad, that no food is off limits, and that a member can eat any food as long as they log it and aim to stay within the calorie range (Noom Team, 2026b, 2026c). Its sample menu includes poached eggs on avocado toast, Greek yogurt with oats and berries, a grilled chicken pita with vegetables, a lentil and carrot salad, a shrimp rice bowl, salmon with potatoes and broccoli, and turkey burgers with chickpea pasta (Noom Team, 2026c). The company says vegan, low-carbohydrate, lactose-free and gluten-free eating can all be fitted into the range, and that its coaches have meal plans and grocery lists to help (Noom Team, 2026c).

What you avoid or limit

Nothing is forbidden by rule. What limits eating is the calorie range and the suggested balance of colors, with orange foods the ones a member is encouraged to eat in smaller amounts (Noom Team, 2026c). The program's other daily tasks are habits rather than food rules: reading lessons of 5 to 15 minutes, logging food, weighing in, which the company encourages every day, and counting steps, starting at 200 a day for people new to exercise (Noom Team, 2026a, 2026b).

The calorie range is set by the app, not by a clinician. The company's terms say the service is not medical care, that it cannot be used by anyone under 18, and that members should talk to a doctor before starting (Noom, n.d.-j). Its home page adds that the program cannot be used by anyone who is pregnant, has an eating disorder, or is underweight, which it defines as a body mass index below 18.5 (Noom, n.d.-d).

Where it came from

Noom, Inc. names Saeju Jeong and Artem Petakov as its co-founders and says their research into helping people reach a healthy weight began in 2008 (Noom, n.d.-a). The app was in use by 2012: a Korean research team analyzed the records of 35,921 people who used the Noom Coach app between October 2012 and April 2014 (Chin et al., 2016). The company built a version of the Diabetes Prevention Program, the lifestyle curriculum recognized by the Centers for Disease Control and Prevention, into the app, and says it was the first virtual program of that kind to receive the agency's recognition, in 2017 (Noom, n.d.-b).

The company now calls itself "a consumer-led digital health company" and says it has expanded from weight management to diabetes and menopause and to medication (Noom, n.d.-a). By September 2024 it was describing Noom Med, a service that adds clinician visits and prescriptions to the app (Noom Team, 2025). In its current marketing the medication programs come first: its home page opens with medication for weight loss and offers what it calls a "psychology-based" program as the alternative (Noom, n.d.-d).

What it claims to do

The company's website is titled "Noom: Lose weight and keep it off," and it describes its approach as focusing "not just on what you eat, but why" (Noom, n.d.-d). It says its products "are developed using evidence-based approaches and scientifically proven principles, such as cognitive behavioral therapy (CBT)" (Noom, n.d.-a), and that it has "more than 40 peer-reviewed scientific articles" (Noom, n.d.-i). One of its program pages says that "research has proven that Noom Weight helps people to lose weight," citing a study of almost 36,000 users in which almost 80 percent reported weight loss (Noom Team, 2026b). Its home page adds that 98 percent of members say Noom helps change their habits and behaviors for good (Noom, n.d.-d).

For Noom Med, the company gives three different figures in three places for how much more weight people lose with its program and a medication than with the medication alone: 37 percent on its home page, 48 percent at 6 months on its Noom Med page, and 39 percent in a 6-month study on an older pricing page (Noom, n.d.-d, n.d.-e; Noom Team, 2025). The home page's footnote says its figure comes from a January 2026 analysis of members who logged their weight and that "this association does not imply causation" (Noom, n.d.-d). Section 4 says what the searches logged for this page found behind these claims.

What it gets right

Several of the program's habits line up with mainstream guidance. The foods it labels green, vegetables, fruits and whole grains, are among the features of a heart-healthy eating pattern in the American Heart Association's 2026 dietary guidance, which lists eating plenty of vegetables and fruits and choosing foods made mostly with whole grains (Lichtenstein et al., 2026). Self-monitoring of diet, exercise and weight is the centerpiece of behavioral weight-loss programs, and a systematic review consistently found it associated with weight loss, although the authors judged the level of that evidence weak (Burke et al., 2011). Lowering the calorie density of meals, for example with water-rich vegetables and fruit, can increase fullness and lower how much people eat at a meal in controlled studies (Rolls, 2009).

Some of the company's framing is also careful. It says the colors are not good or bad and no food is forbidden (Noom Team, 2026c). It excludes people under 18, pregnant people, people with an eating disorder and people who are underweight (Noom, n.d.-d). Its terms state that its coaches are not licensed medical professionals and cannot advise on diagnosis or treatment (Noom, n.d.-j), and it publishes a protocol that points members in crisis to emergency services and the 988 Suicide and Crisis Lifeline (Noom, n.d.-g). These are the parts 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 Noom," "I'm a Noomer," "I'm over my calorie budget," "that's an orange food," "I need more green," "my Noom coach told me," or "I get my GLP-1 through Noom." The company uses the words Noomer, calorie budget and the color names itself (Noom Team, 2026a, 2026b; Noom, n.d.-k). A client who says "Noom" may mean the app-only program, now called Noom Weight; the diabetes prevention version; or Noom Med, which adds medication (Noom, n.d.-d, n.d.-e).

It can be confused with WeightWatchers and other commercial programs, with Meal Replacement Programs, and with general food-logging apps. Because its lessons use words such as mindfulness and hunger cues, it is sometimes confused with Mindful Eating and Intuitive Eating, each of which has its own page. And because it now offers medication, it can be confused with a clinic. Two further confusions matter for coaching. A Noom coach and a board-certified health and wellness coach are not the same role, as section 3 explains. And Noom Monitor, an app tested inside clinical treatment for binge eating disorder, is a different product from the weight program, as section 4 explains.

2. In practice

A typical day of eating

The day described here is this page's own illustration, built from the company's sample menu and program descriptions. Breakfast is low-fat Greek yogurt with oats and berries, logged in the app, which counts it against the day's range. Lunch is a grilled chicken pita with yogurt dressing, peppers and spinach. An apple fills the afternoon. Dinner is salmon with olive oil and dill, potatoes and broccoli. If a coworker's birthday brings a slice of cake, it is logged as an orange food and counted like anything else (Noom Team, 2026c).

Around the meals sit the program's habits: a morning weigh-in, a daily lesson of 5 to 15 minutes, a step count, perhaps a message to a coach or a post in a group (Noom Team, 2026a, 2026b). Nothing on that list is unusual food. What takes time is the logging, every meal and snack, and the attention to a running total.

Cost, products, and who profits

The prices on this page are a snapshot taken from the company's own pages on October 11, 2026, and they change often. Noom Weight, the app-only program, is sold as auto-renewing plans, each billed in full at the start of its period: $70 for a month, $169 for 4 months, which the company says most members start with, and $209 for a year, paid at once.

A trial of 7 or 14 days comes first, and the company lets a new member choose what to pay for it, with examples of $0.50, $3, $10 and $18, saying a trial costs it about $10 to provide (Noom Team, 2026a). Its terms state that subscriptions renew automatically at the end of each billing period unless canceled, that a trial turns into a paid plan unless canceled before it ends, and that fees are generally not refundable except where the law requires, such as a 3-business-day cancellation right for California residents (Noom, n.d.-j).

What a plan includes varies. One company page says every user is offered a personal coach (Noom, n.d.-c), while its support page says coaching requires a paid subscription, that not all programs include a one-on-one coach, and that members without one are supported first by an artificial intelligence chat called Welli, with questions passed to a team of coaches when needed (Noom, n.d.-k).

Noom Med is priced separately. On October 11, 2026, the company advertised its Microdose GLP-1Rx program at $79 plus tax for an introductory 3-week subscription that includes 4 weeks of medication, and after that $199 a month plus tax on a 12-week subscription. Its GLP-1Rx program cost $149 plus tax for the same introductory period, and after that $349 a month on a 12-week subscription on its home page but $299 a month on two program pages. All of these were described as new-account pricing from March 31, 2026. Separately, a Noom Med membership with access to brand-name medications was advertised from $39, with the medication itself paid for on top, depending on insurance (Noom, n.d.-d, n.d.-e; Noom Team, 2026b, 2026c).

The company states that the medications in its GLP-1Rx programs are made in state-regulated compounding pharmacies and are not reviewed by the U.S. Food and Drug Administration for safety, efficacy or quality (Noom, n.d.-d), and its terms say some compounded prescriptions can be filled only through a pharmacy partner on its platform (Noom, n.d.-j).

Several parties make money when a client follows the program. The company earns subscription and medication-program fees, and it also sells the program to employers and health plans: in a company overview dated October 2025 it says it covers more than 6.5 million lives and works with 5 of the 20 largest health plans in the country, figures that are the company's own (Noom, 2025). Compounding pharmacies and the makers of brand-name medications are paid when prescriptions are filled. Most of the research on the program lists company employees among its authors, as section 4 notes trial by trial. None of this makes the program or the research wrong, and it is worth knowing who has a stake in it.

Billing has been the subject of litigation. In a class action in federal court in New York, customers alleged that Noom's trial and automatic renewal practices led to payments they did not intend; the case ended in a settlement that, as the plaintiffs' lawyers described it to the court, included a $56,000,000 cash fund and $6,000,000 in subscription credits, and the court approved it in July 2022 (Nichols v. Noom, Inc., 2022, Document 495; Nichols v. Noom, Inc., 2022, Document 518). The case ended in a settlement rather than a trial, and a settlement is not a finding that the allegations were true. The case concerned billing, not the program's effects on health.

The searches logged for this page did not identify a study of what members spend on food while following the program. What follows is observation rather than measured data. The subscription is the main fixed cost; the program needs a smartphone and time for daily logging; and, because no food is required, a member can follow it with ordinary groceries. A member on a medication program faces a much larger monthly cost.

Environmental footprint

The searches logged for this page did not identify a study of the environmental footprint of following Noom. Because the program sets a calorie range rather than a food list, a member's footprint depends mostly on which foods they choose and how much they eat.

Cultural fit

Because the program forbids no food, a member can in principle keep the dishes of their own tradition, and the company names cheesecake, enchiladas, chicken curry, ice cream and samosas as foods members can still have (Noom Team, 2026b, 2026c). The colors follow calorie density, not tradition or nutrient value, so a dish at the center of a family's cooking may land in orange simply because it is dense in calories; the company's own example compares half a pound of peanut butter, at 1,334 calories, with the same weight of spinach, at 52 (Noom Team, 2026b).

Logging a home-cooked family dish means estimating its ingredients, and the company's terms say it does not verify the nutrition entries members add to its food database (Noom, n.d.-j). The company's site is offered in English, Korean, Spanish, German and French (Noom, n.d.-d). For a client whose meals are shared, unmeasured and cooked by someone else, the daily logging the program depends on asks a great deal.

What makes people stop

The trials show how many people drift away. In the prediabetes trial, 23 of the people assigned to the program never downloaded it or redeemed its code, and another 27 did not engage meaningfully; messages to coaches in the program group averaged 1.44 a week at 6 months and 0.94 a week at 12 months (Toro-Ramos et al., 2020). In a 600-person trial run with Mount Sinai, 112 of 302 people assigned to the full program and 105 of 298 assigned to a simplified version did not complete the 30-month study, most of them lost to follow-up (Icahn School of Medicine at Mount Sinai, 2025). In an early diabetes prevention study, 121 people signed up and 43 met the definition of starting the program (Michaelides et al., 2016).

Beyond those figures, what follows is observation: the reasons people leave are likely to be the ones the program creates, such as the time daily logging takes, the cost of renewing, and the moment the novelty of the lessons wears off.

What happens when you stop

The trials found for this page followed people for up to 30 months. In the prediabetes trial, the difference in measured weight between groups was 2.6 kg at 6 months and 1.8 kg at 12 months, and people who completed the program regained some weight between those points (Toro-Ramos et al., 2020). In a company-run trial of a 16-week program, in which participants weighed themselves at home and reported the result, the program group had lost 4.1 percent of body weight at 68 weeks while the control group had gained 1.5 percent (McCallum et al., 2026). In the 30-month trial, average weights posted to the trial registry were 12.9 pounds below the starting average in the full-program group and 10.1 pounds below it in the simplified-program group, among the people who provided data, with no statistical comparison posted (Icahn School of Medicine at Mount Sinai, 2025).

In a company survey of 840 former members who had lost at least 10 percent of their weight on the program, respondents reported keeping off 65 percent of the weight they had lost after 1 year and 57 percent after 2 years (May et al., 2023). That survey reached only people who had already succeeded and chose to answer, and its weights were self-reported, so it cannot say what happens to members in general. For Noom Med, what happens when a medication is stopped is a question for the prescriber; the company's own terms include a guarantee for members who taper off medication and do not maintain their weight (Noom, n.d.-j).

Common variations

Noom Weight is the app-only program; older papers and court records call it Healthy Weight (DeLuca et al., 2020; Nichols v. Noom, Inc., 2022, Document 518; Noom, n.d.-d). The diabetes prevention version runs the federal Diabetes Prevention Program curriculum through the app (Noom, n.d.-b). Noom Med adds clinician visits and prescriptions, including GLP-1 medications, and a companion curriculum for people taking them (Noom, n.d.-e). The company also lists a mental well-being program, Noom Mood, and programs for menopause (Noom, n.d.-a, n.d.-h). Employers and health plans offer a version called Noom Health under their own terms (Noom, n.d.-j).

In research, the program has been adapted in ways a client will not meet in the app store: in some Korean trials, clinical dietitians, exercise coordinators or psychologists delivered the coaching (Cho et al., 2020; Kim et al., 2020). Noom Monitor, an app used inside clinical treatment for binge eating disorder, is a separate product described in section 4. The program's food tracking overlaps with general food-logging apps.

3. For the coach

SCOPE OF PRACTICE

A health and wellness coach does not prescribe, adjust, endorse, or discourage Noom 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.

A client on Noom may already have a Noom coach, and the two roles are not the same. Noom states that all of its coaches go through its own training program, Noomiversity, with more than 75 hours of new-hire training and over 200 hours of coaching experience with users, and that all of them are given the option of its Health Coach Certification Program, which it calls "an approved transition program with the National Board for Health and Wellness Coaches," and of sitting the board's certifying examination (Noom, n.d.-f).

That is the company's description of a training program; this page could not confirm it in the board's own directory. Approval of a training program is not the same as an individual's certification: a coach becomes a National Board Certified Health and Wellness Coach only by meeting the board's requirements and passing its examination. Its home page says that many of its coaches are board-certified health professionals (Noom, n.d.-d). Its terms state that Noom Coaches are not licensed medical professionals and cannot advise on the diagnosis or treatment of any medical or mental health condition (Noom, n.d.-j). The pages opened for this page do not say how many of its coaches hold that board's credential.

Any decision about weight-loss medication, including whether to start, change or stop a GLP-1 medication prescribed through Noom Med, belongs to the prescriber. A calorie range for someone with diabetes, kidney disease or another diagnosed condition belongs with the client's clinician and a registered dietitian; with kidney disease, what fits depends on its stage, laboratory results and whether the person is on dialysis, so the kidney care team is involved. The coach never starts, changes or stops a prescription.

The program's risks fall on particular groups rather than on healthy adults in general, and the company itself excludes several of them. For a client in any of these groups, the coach's role includes making sure the client has medical oversight for the change:

• A current or past eating disorder: the company says the program cannot be used by someone with an eating disorder, so the client's clinician is involved before anything else, through the coach’s own referral process or an existing care plan.
• Pregnant or breastfeeding: the obstetric clinician and a registered dietitian; the company excludes pregnancy.
• Under 18: the pediatric clinician; the company's terms exclude anyone under 18.
• Underweight: a clinician; the company excludes anyone with a body mass index below 18.5.
• Taking insulin, a sulfonylurea or other medicine for diabetes, or medicine for blood pressure: The client keeps the prescriber informed before a substantial change in eating. A possible low blood sugar episode is treated first under the client's own plan, with emergency help if symptoms are severe or the person cannot manage the episode under that plan; the prescriber is told afterward. The coach never changes a prescription.
• Taking or considering weight-loss medication, through Noom Med or elsewhere: the prescriber, for every question about the medication.

Who does what

The physician

Decides whether calorie restriction is medically appropriate for this person, and adjusts any medication it affects, particularly insulin and sulfonylureas, whose effects change when a person eats less, and blood pressure medicine. Evaluates and prescribes, or declines to prescribe, weight-loss medication, and manages its side effects, its dose and the decision to stop it. Assesses a client whose eating, mood or weight raises concern, including a possible eating disorder.

The registered dietitian

Sets or reviews an individual calorie and nutrient plan when a diagnosed condition is involved, such as diabetes, pregnancy or kidney disease, where needs depend on the stage, laboratory results and dialysis. Checks whether the foods a client is choosing within the range meet their nutrient needs, and helps a client whose cultural or family meals are hard to log. A dietitian who specializes in eating disorders is the right person when tracking and restriction have become part of the problem.

The health and wellness coach

Helps the client say what they want from Noom and why now, and test the program against the life they live: the time logging takes, the meals they share, the budget for renewals. Helps the client notice how the program is going in their own terms, energy, mood, sleep and enjoyment of food as well as the numbers the app tracks. With the client's permission, can share general information from recognized authorities. Works alongside, never in competition with, the Noom coach in the app, and holds the referral line when a question belongs to a clinician.

What a client bringing this diet needs

Clients bring Noom for a handful of reasons: an advertisement that promised a psychological approach rather than another diet, a coworker who lost weight on it, a benefit offered through an employer or health plan, or interest in a GLP-1 medication. Some arrive already months into the program and frustrated with the logging; some arrive with a prescription from Noom Med and questions about it. The reason shapes the conversation, and a medication question is never the coach's to answer.

What the coaching conversation offers that the app cannot is room to separate the client's own goals from the program's metrics. A coach avoids moral language about food; the program's colors are not meant as moral labels, according to the company (Noom Team, 2026c), and it is worth listening for whether a client hears them that way. Knowing how a certification differs from a license helps a coach describe the difference between the client's Noom coach and themselves without disparaging either.

What to ask

• "What drew you to Noom, and what are you hoping will be different in 6 months?"
• "How does logging fit into your days, and how do you feel when you go over your range?"
• "Who else is part of this, your Noom coach, your doctor, the people you eat with, and what does each of them know about your plan?"

Referral red flags

A client taking insulin, a sulfonylurea or a meglitinide who starts eating less, skips or delays meals, or exercises more is at greater risk of low blood sugar (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.); the client keeps the prescriber informed before a substantial change. A client 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 symptoms are severe or the person cannot manage the episode under their plan, and telling the prescriber is the separate step that follows.

A client taking a GLP-1 medication, whether through Noom Med or elsewhere, who describes vomiting that does not stop, severe abdominal pain or other troubling symptoms needs the prescriber promptly; the U.S. Food and Drug Administration has received reports of adverse events with compounded versions, some requiring hospitalization (U.S. Food and Drug Administration, n.d.). The coach does not guess at the cause or comment on the dose. Weight a client is losing without meaning to, such as loss that continues after they have stopped trying, is also a reason to involve a clinician. A client who expresses thoughts of suicide or self-harm calls for the program's emergency and safety procedures at once; the company's own protocol points members to emergency services and the 988 Suicide and Crisis Lifeline (Noom, n.d.-g).

Some patterns call for a closer look at the eating itself: skipping meals to stay under the range, distress or guilt after an orange food, weighing in several times a day, refusing meals with others that cannot be logged, or a range that keeps shrinking. Logging and daily weighing are ordinary on this program, and many people track food without any problem, 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 company says the program cannot be used during pregnancy (Noom, n.d.-d). The searches logged for this page did not identify a study testing the program's health or safety effects in pregnancy or breastfeeding. A company-run analysis followed 130 women who had given birth within the previous 2 years and were offered the program free, and reported weight change over 24 weeks without a comparison group (Toro-Ramos et al., 2021). A randomized trial at Northwestern University is testing blood pressure monitoring plus a Noom subscription after high blood pressure in pregnancy, and its registry record shows no results (Northwestern University, 2026). Any calorie-restricted plan during pregnancy or breastfeeding belongs with the obstetric clinician and a registered dietitian.

Children and adolescents

The company's terms exclude anyone under 18 (Noom, n.d.-j), and the searches logged for this page did not identify a study of the program in children or adolescents. A coach works with the adult in front of them; a young person's eating belongs to the family and the pediatric clinician.

Older adults

A company-run analysis of 14,767 users aged 35 to 85 reported that older users lost more weight than younger users, with no comparison group (DeLuca et al., 2020). The searches did not identify a trial designed to test the program's benefits and harms in older adults. People aged 65 or older are among those more likely to develop low blood sugar when they take insulin or some other diabetes medicines (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.), so an older client on those medicines keeps the prescriber informed before eating substantially less.

Athletes and people with high activity

The searches logged for this page did not identify a study of the program in athletes. Whether an app-set calorie range meets the energy needs of hard training is a question for a sports dietitian.

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

Insulin, sulfonylureas and meglitinides can cause low blood sugar, and not eating enough carbohydrate, or skipping or delaying meals, makes it more likely (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.). For any of these medicines, and for blood pressure medicine and warfarin, 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. Weight-loss medication prescribed through Noom Med is managed entirely by the prescriber.

People with a history of disordered eating

The company says the program cannot be used by anyone with an eating disorder (Noom, n.d.-d). The studies reviewed here do not establish the consumer program's safety for people with a current or past eating disorder: McCallum excluded anyone ever diagnosed, while other clinical or bariatric studies did not use that same exclusion and did not provide a separate safety estimate for this population. Calorie counting, color labels and daily weighing are the features that make caution sensible for someone with this history. A pooled analysis of behavioral weight-management trials in adults found that eating disorder symptoms fell on average but that a small share of participants in some studies developed new symptoms (Jebeile et al., 2023), and that evidence concerns weight programs in general, not this one. A client with this history is referred to their clinician, or through the coach’s own referral process if they have none, before the program is discussed further.

Client misconceptions

Three come up most. The first is that Noom is not a diet. The company says it is not a diet in the usual sense (Noom Team, 2026b), but it sets a daily calorie range and asks members to log what they eat against it, which is calorie restriction by another name. The second is that research has proven the program works. The trials found for this page show a modest difference in measured weight by the end of a 4-month to 6-month program, most of them run or paid for by the company, and the size of the short-term difference is very uncertain.

The third is that a Noom coach is a health professional who can answer medical or nutrition questions. The company's own terms say its coaches are not licensed medical professionals and cannot advise on diagnosis or treatment (Noom, n.d.-j). Each is taken up, with its sources, in section 6.

🤝 COACHING CONNECTION

Danielle has used Noom for 3 months and says it is working, but she has started skipping dinner on days when lunch "went orange," and she feels guilty when she goes over her range. The coach reflects that she sounds proud and worn out. The coach asks what skipping dinner does for her, how often it happens, and how she feels the next day. Danielle says it happens most weeks and that she has stopped eating with her roommates on those nights. The coach says that because the skipped meals are leaving her guilty and eating alone, this is something her doctor should hear about, and asks how she would feel about raising it. Danielle decides to call her doctor's office and to tell her Noom coach that the guilt is the thing she wants help with. Nothing was recommended, and the program was never evaluated.

4. The evidence

(select the heading to open or close this section)

Noom has a sizable body of published research, and most of it comes from the company: its employees are authors, it sponsored or helped design the trials, or the data come from its own users. The searches logged for this page, run on October 11, 2026 in Europe PMC, PubMed and ClinicalTrials.gov, identified randomized trials of the weight program against usual care or educational materials; trials that added human coaching or psychologist-delivered therapy to the app; trials of a separate app, Noom Monitor, inside eating disorder treatment; and many company analyses of its own users. They did not identify a study estimating the program's effect on heart attacks, strokes, cancer or survival; adverse-event counts are a different question. 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 a program by chance and compares them with people assigned to something else, which is the strongest way to test whether the program itself causes a result. In these trials participants knew which program they were using, so none could be blinded, and the two largest company trials were registered only after they had ended, which makes it impossible to check that the outcomes and analyses reported were the ones planned (Noom Inc., 2024; Toro-Ramos et al., 2020).

Cardiometabolic outcomes

In a trial at Stony Brook Medicine in New York, 202 adults with prediabetes were randomized to the app's diabetes prevention program, with coaches certified for the National Diabetes Prevention Program and trained in motivational interviewing, or to usual care with a printed copy of the curriculum.

HbA1c, a laboratory measure of average blood sugar, differed between the groups by 0.004 percentage points at 6 months and 0.006 at 12 months. Approximate 95 percent intervals, which this page calculated from the standard errors the paper reports, run from about 0.09 points lower to 0.10 points higher in the program group at 6 months and from about 0.13 lower to 0.14 higher at 12 months, well inside the 0.5-point difference the trial was designed to detect. The authors note that the trial enrolled fewer people than the 224 it planned and describe it as underpowered for HbA1c (Toro-Ramos et al., 2020).

An independent trial in Seoul, funded by the Korean government, randomized 129 adults who had at least 2 features of metabolic syndrome but no diagnosis or treatment to no app, the Noom app for logging only, or the app with coaching from clinical dietitians and exercise coordinators. Its main outcome, systolic blood pressure, fell in all 3 groups over 24 weeks with no significant difference among them; the trial did not reach its planned sample size (Cho et al., 2020).

A 12-week pilot trial in 40 adults with type 2 diabetes paired the Noom Coach app with a glucose meter and medical supervision based on the data sent to a central server. Among the 35 who completed it, 47.1 percent of 17 in the app group reached an HbA1c of 6.5 percent or lower, against 11.1 percent of 18 in conventional care (Ku et al., 2020). That design tested a clinic-supervised system rather than the consumer program; it was read for this page as an abstract, and proportions among completers of a small pilot are not an estimate of the program's effect.

Cancer

The trials in people with cancer used the app for support during or after treatment and measured nutrition, quality of life and weight, not cancer outcomes. In a trial of 40 people starting chemotherapy for pancreatic cancer in Korea, the app was paired with feedback from dietitians for 12 weeks; nutritional status improved in both groups, and quality of life scores improved more in the app group (Keum et al., 2021). The searches logged for this page did not identify a trial that measured cancer incidence, recurrence or survival, and the studies reviewed do not establish an effect on cancer.

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 program worked; see the standing rule in the review record at the foot of this page. The one exception is weight a person loses without meaning to, which can signal a health problem and is a reason to involve a clinician, never a sign that a program worked.

In the prediabetes trial, weights were measured in clinic. In the intention-to-treat analysis, which keeps everyone in the group they were assigned to, the program group weighed 2.64 kg less than the usual-care group at 6 months and 1.80 kg less at 12 months. Of those assigned to the program, 23 never downloaded it or redeemed its code, and the 45 who completed it lost 5.6 percent of their weight at 6 months and 4.7 percent at 12 months, a completer result that is not a fair comparison with the whole control group (Toro-Ramos et al., 2020).

Three features of that trial matter for judging it. It was registered in March 2019, after it had begun in October 2016. Two of its authors were company employees, and the company helped design the trial and paid for participant gift cards, test supplies and study staff. And the paper reports 101 people in the program group in its abstract and 103 in its table of participant characteristics (Toro-Ramos et al., 2020).

In a company-sponsored trial published in 2026, 427 adults with a mean body mass index of 32.5 were randomized to Noom Weight for 16 weeks, 217, or to an educational control, 210, who received 16 weekly email newsletters adapted from the 2015-2020 federal Dietary Guidelines. People who had ever been diagnosed with an eating disorder, or who had diabetes, were excluded. After randomization, 15 people were dropped for missing or unreliable data, leaving 412 in the analysis, 211 and 201; weights at 68 weeks were available for 183 and 174, and the statistical model kept people with missing weights in the analysis. Participants weighed themselves on their own scales and reported the result, and neither they nor the study staff were blinded (McCallum et al., 2026; Noom Inc., 2024).

At 16 weeks the program group had lost 1.2 percent of body weight and the control group had gained 0.6 percent, a difference of 1.8 percentage points (95 percent confidence interval 0.7 to 2.8). At 68 weeks the program group had lost 4.1 percent and the control group had gained 1.5 percent, a difference of 5.5 percentage points (3.0 to 8.0). The authors note that the long-term result could partly reflect bias in self-reported weights (McCallum et al., 2026). The trial was first registered in October 2024, after it ended in December 2023, and the registry record describes cellular-connected scales as a backup source of weights, which the paper does not mention (Noom Inc., 2024). All of its authors were company employees or contractors when the work was done (McCallum et al., 2026).

Smaller trials point the same way. In a 4-month pilot, 70 adults were randomized to Noom Weight or to weekly educational newsletters, and the 47 who came to their final visit on time were included in the weight analysis: weighed on an office scale, the program group lost 7.1 pounds and the control group gained 1.0 pound, a difference at the edge of statistical significance (Michaelides et al., 2022).

In 40 adults with steatohepatitis, an inflamed form of fatty liver disease, the program group lost 5.5 kg over 16 weeks against 0.3 kg with standard clinical care; weights came from a Bluetooth scale or, for some final weights, were reported by phone or in the app, and 33 of the 40 gave a final weight (Stine et al., 2023). Both trials had company employees among their authors, and the lead author of the liver trial reported research support from the company.

The independent Seoul trial found smaller differences: over 24 weeks the group with dietitian and exercise coaching lost 0.96 kg, the logging-only group 0.35 kg and the control group 0.12 kg. In a simple comparison of change, neither app group differed significantly from control; in the trial's regression model, the coaching group lost 0.87 kg more than control by 24 weeks (P = .04), a difference whose lower limit is close to none (Cho et al., 2020).

In a trial of 70 women in Korea, psychologists delivering daily cognitive behavioral therapy through the Noom Coach app produced more weight loss at 8 weeks than the app used alone, 3.1 against 0.7 percent, in an analysis limited to people who attended follow-up; the difference in reaching a 5 percent loss did not last to 24 weeks (Kim et al., 2020). In 50 people preparing for bariatric surgery, an 8-week version of the program had small effects on body mass index that were not statistically significant (Sysko, Michaelides, et al., 2022).

Does the human coaching add anything?

Human coaching is one feature that separates Noom from a plain food-logging app, and it has been tested directly only a few times. In a 600-person trial run by the Icahn School of Medicine at Mount Sinai with the company as a collaborator, adults were randomized to the full program with coaching, 302, or to a simplified version for logging food and weight with the same advice content and no coaching, 298, for 6 months, with follow-up to 30 months and weights taken on participants' own scales during video calls (Icahn School of Medicine at Mount Sinai, 2025; Sysko, Bibeau, et al., 2022).

The results posted to the trial registry give average weights of 210.1 and 211.7 pounds at the start, 199.4 and 204.6 pounds at 6 months among the 236 and 231 people with data, and 197.2 and 201.6 pounds at 30 months among 190 and 193, for the full and simplified groups. No statistical comparison between the groups is posted, and the searches logged for this page did not identify a journal report of the results (Icahn School of Medicine at Mount Sinai, 2025). A second company trial comparing coached and uncoached versions, with 246 participants and self-reported weight at 4 months as its main outcome, was registered before it began and completed in 2021, with no results posted to its registry record and no publication found by these searches (Noom Inc., 2022b). The Seoul trial's coached and logging-only groups are described above (Cho et al., 2020).

Adverse events

Few of the trials reported harms. The Mount Sinai registry results list no deaths; 1 serious event, fainting during exercise, in the full-program group and 2, lowered breastfeeding milk supply, in the simplified group, without stating whether they were related to the program; and other events in 43 and 42 people, each labeled unrelated, most of them infections (Icahn School of Medicine at Mount Sinai, 2025). The liver disease trial reported no adverse events (Stine et al., 2023). Because both Mount Sinai groups used a version of the app, those counts cannot show whether the program causes any event, and none of the trials was designed to detect rare harms.

Mortality

The searches logged for this page did not identify a trial estimating the program's effect on survival. The Mount Sinai registry records zero deaths in both app groups among its adverse events, but those counts do not estimate a mortality benefit or harm. The trials found lasted 8 weeks to 30 months and were not designed to estimate an effect on survival.

Eating behavior and wellbeing

Three trials of the weight program reported scores on the Eating Disorder Examination Questionnaire. In the 2026 trial, whose participants had never been diagnosed with an eating disorder, scores fell from the start to 68 weeks in both groups, by 0.39 points with the program and 0.25 points with the educational control. The difference between groups, 0.14 points in favor of the program, was not statistically significant, with a 95 percent confidence interval running from 0.33 points in favor of the program to 0.04 points against it (McCallum et al., 2026). A fall in both groups is a change over time, not an effect of the program.

In the bariatric surgery trial, which did not exclude people with eating disorders, differences between groups were small and not statistically significant (Sysko, Michaelides, et al., 2022). In the 30-month Mount Sinai trial both groups used a version of the app, so it cannot show the program's own effect; average scores fell from 2.5 to 1.7 and from 2.6 to 1.8 among the 119 and 123 people with data, with no statistical comparison posted (Icahn School of Medicine at Mount Sinai, 2025). A completed company trial of the program against a wait-list control that measured disordered eating attitudes, with 160 participants, has no results posted and no publication found by these searches (Noom Inc., 2022a).

Other trials reported improvements in emotional eating at 8 weeks with psychologist-delivered therapy (Kim et al., 2020), in stress and anxiety before bariatric surgery (Sysko, Michaelides, et al., 2022), and in quality of life during chemotherapy (Keum et al., 2021). None of these trials reported results separately for people with an eating disorder history, so they cannot show how the program affects someone with that history.

A different app: Noom Monitor in eating disorder treatment

Noom Monitor is a separate app built to digitize the self-monitoring records used in guided self-help based on cognitive behavioral therapy, a recognized treatment for binge eating disorder and bulimia nervosa. In a trial of 66 adults with those diagnoses, adding the app to 8 sessions of guided self-help produced a greater fall in binge episodes over 12 weeks than guided self-help with paper records; remission rates, 17 of 27 against 11 of 27, were not statistically different, and the groups did not differ at 6-month follow-up (Hildebrandt et al., 2017).

In a second trial at Kaiser Permanente Northwest, 225 adults with binge eating disorder or bulimia nervosa were randomized to 8 telemedicine sessions of guided self-help with Noom Monitor, delivered by the health system's health coaches, or to standard care; at 52 weeks, 56.7 percent of the app group were in remission from binge eating against 30 percent of standard care, none of whom had received eating disorder treatment during the first 12 weeks (Hildebrandt et al., 2020).

That trial could be read for this page only as its abstract, so its methods, its handling of missing data and its authors' financial disclosures could not be checked; the abstract lists two authors at Noom, Inc. In the first of the two trials, read in full, two investigators held equity in the company, one of them also sat on its scientific advisory board, and two authors were company employees (Hildebrandt et al., 2017); those disclosures belong to that trial and are not assumed for the second.

These trials tested a clinical treatment for diagnosed eating disorders, delivered under a health system's supervision. They do not show that the weight program is safe or helpful for someone with an eating disorder, which the company itself excludes (Noom, n.d.-d). They also illustrate a scope point: the National Board for Health and Wellness Coaching's scope of practice says coaches do not on their own provide psychological therapeutic interventions such as cognitive behavioral therapy, although they may support a licensed professional's treatment plan when working under that professional (National Board for Health and Wellness Coaching, n.d.).

Evidence from long-term prospective cohort studies

A cohort study follows people over time and records what happens to them. The searches logged for this page did not identify a prospective cohort that followed Noom users for heart disease, diabetes, cancer or death, so they supply no cohort evidence on those outcomes.

Evidence from company user data and single-group studies

The company has published many analyses of its own users. A Korean team analyzed 35,921 users of the Noom Coach app between 2012 and 2014 and reported that 77.9 percent recorded a decrease in weight while using it, over a median of 267 days (Chin et al., 2016); this is the study behind the company's "almost 80 percent" figure (Noom Team, 2026b). Weights were entered by users, and people who stop using an app stop entering weights, so the figure describes people who kept logging. A 2016 study of the diabetes prevention version reported a mean loss of 6.58 percent at 24 weeks among 43 people who met its definition of starting, out of 121 who signed up (Michaelides et al., 2016). Analyses of 14,767 users by age (DeLuca et al., 2020) and of 130 women after childbirth (Toro-Ramos et al., 2021) also had no comparison group.

Two company-run analyses of insurance and health record data reported lower health care costs among users than among nonusers: up to $1,219.06 lower per user at 24 months in a subset of a study of 43,047 users and 14,555 nonusers (Mitchell et al., 2023), and fewer outpatient visits and less GLP-1 use at 6 months among 723 participants matched to 723 employees offered the program who did not enroll (Tak et al., 2025). The first analysis also reported fewer outpatient visits and lower costs at 12 months. The company repeats the 24-month figure in its materials for employers and investors (Noom, 2025). People who choose to use a program, and keep using it, differ in many ways from people who do not, and statistical weighting and matching can balance only the differences that were measured, so neither analysis shows that the program caused the lower use or costs.

For Noom Med, the company's figures for extra weight lost with its program alongside medication come from its own analyses of members, and its home page says the association "does not imply causation" (Noom, n.d.-d). A 2026 preprint, not yet peer reviewed, compared 16-week weight change among 29,788 members of the company's weight program and its two compounded semaglutide programs (Owen et al., 2026). This page does not rate the medication programs; medication decisions belong to the prescriber.

What independent reviews say

A 2021 review of commercial weight-loss programs by researchers at Johns Hopkins, whose disclosures list no tie to the company, stated that they were "aware of no published RCTs reporting outcomes with Noom at 12 months in the general population" and concluded that "current evidence regarding weight-loss efficacy and safety with the use of Noom is limited" (Laudenslager et al., 2021). The review described the prediabetes trial as ongoing, although its results had been published in 2020, and the 2026 trial appeared later. This page read the review as its accepted manuscript in PubMed Central; the published version could not be opened, so these quotations are checked against the manuscript only.

A 2024 Cochrane review of smartphone apps in general for adults and adolescents with overweight or obesity concluded that the available evidence "does not demonstrate a clear benefit"; against no or minimal intervention, apps may make little to no difference to body weight at 12 months, a mean difference of 2.5 kg lower with apps, from 6.8 kg lower to 1.7 kg higher, rated low certainty (Metzendorf et al., 2024). This page could open only its abstract and plain-language summary, which do not name the apps tested. The review concerns apps in general; its rating is not transferred to Noom, and it is not used on this page as evidence about Noom.

Evidence on calorie tracking and eating disorder risk

Because Noom asks members to log food and weigh in daily, evidence about tracking in general is relevant, though indirect. Observational studies show strong associations between dietary self-monitoring and eating disorder risk, but they cannot show which comes first (Hahn et al., 2021). In a randomized trial of 200 undergraduate women at low risk of an eating disorder, about a month of calorie logging with a popular app did not change eating disorder risk, anxiety, depressive symptoms or body satisfaction; the authors caution that the result may not apply to people at higher risk (Hahn et al., 2021).

A meta-analysis of 49 behavioral weight-management trials in 6,337 adults found that eating disorder scores and binge eating fell on average after the programs, while 4 studies reported new eating disorder symptoms in 0 to 6.5 percent of participants (Jebeile et al., 2023). The authors conclude that such programs do not increase symptoms for most adults but that monitoring for the emergence of symptoms is important.

A 2025 systematic review of 27 studies of fitness and diet tracking technology found reasonably consistent associations between tracker use and disordered eating in cross-sectional studies, which measure both at one point in time, but the association was not replicated in experimental studies; the authors conclude that it is not possible to say whether trackers increase disordered eating or which comes first (Moody et al., 2025). None of the studies in this subsection tested Noom, and they are not evidence about the program's own safety.

Highlighted studies

These 4 studies carry the most weight for this page: the two largest trials against usual care or education, the independent trial that separated coaching from logging, and the 30-month trial of coaching. 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 Noom

Study	Design	Sample size	Duration	Finding	The limitation that matters
Toro-Ramos et al., 2020, diabetes prevention version	Randomized controlled trial against usual care with a printed curriculum	202 adults with prediabetes randomized	12 months	Weight 2.64 kg lower at 6 months and 1.80 kg lower at 12 months in the program group, intention to treat; HbA1c differed by less than 0.01 percentage points, with intervals within about 0.14 points	Registered after it ended; company employees as authors and company support; fewer participants than planned; group counts differ between abstract and table
McCallum et al., 2026	Randomized controlled trial against educational emails	427 adults randomized, 412 analyzed	16-week program, follow-up to 68 weeks	Difference of 1.8 percentage points of body weight at 16 weeks and 5.5 at 68 weeks, favoring the program	Weights self-reported by participants who knew their group; 15 dropped after randomization; registered after it ended; company-sponsored, with company authors
Cho et al., 2020	Randomized controlled trial, 3 groups, government-funded with no declared conflicts	129 adults with features of metabolic syndrome	24 weeks	No difference in systolic blood pressure; weight fell 0.96 kg with coaching, 0.35 kg with logging only and 0.12 kg with no app; in the regression model the coaching group lost 0.87 kg more than control (P = .04)	Did not reach its planned size; coaching was by clinical dietitians and exercise coordinators, not Noom's own coaches
Icahn School of Medicine at Mount Sinai, 2025, registry results	Randomized controlled trial, full program with coaching against a simplified program without coaching	600 adults randomized	6-month program, follow-up to 30 months	Average weight fell from 210.1 to 197.2 pounds with coaching and from 211.7 to 201.6 pounds without, among those with data	Registry results only, with no statistical comparison and no journal report found; about a third did not complete the study; investigators with company ties

Head to head

The searches logged for this page did not identify a trial comparing Noom with another named diet or commercial program, such as WeightWatchers. The comparisons in the trials found were with usual care, printed materials, educational emails or newsletters, standard clinical care, or a simplified version of the same app. Comparisons of Noom Med with medication alone come only from the company's own analyses of its members (Noom, n.d.-d; Owen et al., 2026).

Where the evidence is absent

The searches logged for this page, run on October 11, 2026 in Europe PMC, PubMed and ClinicalTrials.gov, including a separate search for each special population this archive covers, did not identify a study estimating the program's effect on heart attacks, strokes, cancer or survival; a prospective cohort of users followed for health outcomes; a trial establishing the consumer program's health or safety effects in children, adolescents, pregnancy, breastfeeding, athletes or people with a current or past eating disorder; or a trial comparing it with another commercial program.

They did not identify a published journal report of the 30-month trial of coaching; results of 3 completed company trials, one of coached and uncoached versions and two of the program against a wait-list control, one of them with 370 participants and weight as its main outcome (Noom Inc., 2022a, 2022b, 2023); a study assessing micronutrient adequacy, blood nutrient levels or deficiency in members; or an independent trial in the United States of the consumer program as sold. These searches did not cover every database, regulator, court or population, so a study outside them may exist. The studies reviewed for this page do not establish that the program causes lasting changes in health, and they do not establish that it lacks them.

5. Mechanisms

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

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

Established mechanisms

The program combines several behavioral tools with long research histories. Self-monitoring, recording what one eats, how much one moves and what one weighs, is the centerpiece of behavioral weight-loss programs; a systematic review found it consistently associated with weight loss, while judging the level of that evidence weak because of the studies' methods (Burke et al., 2011). In the prediabetes trial, more frequent weigh-ins and meal logging predicted greater weight loss among people in the program group (Toro-Ramos et al., 2020), an association within one group that cannot show that logging caused the loss.

The color system rests on calorie density. In controlled studies, lowering the calorie density of foods, for example by adding water-rich vegetables and fruit, increased fullness and lowered how much people ate at a meal, because people tend to eat a fairly consistent weight of food; the reasons for that tendency are not yet well understood (Rolls, 2009). The weight changes in the trials are consistent with members eating less than before, which a daily calorie range is designed to produce.

Proposed mechanisms

The company attributes its results to psychology: lessons it describes as based on cognitive behavioral therapy and dialectical behavior therapy that help members understand why they eat as they do (Noom Team, 2026a). The trials found for this page tested the program as a whole, and none separated the lessons from the logging, the calorie range and the coaching, so the share of any effect owed to the psychological content is unknown. A company pilot reported that members kept better accuracy on two tests of executive function, the mental skills of planning and self-control, over 4 months than a newsletter control, with weaker differences in brain-wave measures (Michaelides et al., 2022); that small study is a reason to look further, not a demonstrated mechanism.

The company also proposes that its program and GLP-1 medications work together, and its overview describes the time on medication as a window for building habits (Noom, 2025). That is a hypothesis; the searches logged for this page did not identify a trial testing it. A proposed mechanism is a reason to study a program, not evidence that it works.

6. Risks, contraindications and myths

(select the heading to open or close this section)

Warnings from professional or regulatory bodies

The sources and searches logged for this page did not identify a warning about the Noom weight program issued by a professional or regulatory body; they did not cover the records of every regulator, state attorney general or court. A federal warning does apply to a product category the company sells. The U.S. Food and Drug Administration states that unapproved versions of GLP-1 drugs "do not undergo FDA's review for safety, effectiveness and quality before they are marketed," that "compounded drugs should only be used in patients whose medical needs cannot be met by an FDA-approved drug," and that it has received reports of adverse events, some requiring hospitalization, that may be related to dosing errors with compounded injectable semaglutide (U.S. Food and Drug Administration, n.d.).

The agency's list of warning signs for consumers buying through telehealth includes a company that "makes claims such as the compounded drug is the same as an FDA-approved drug" or that "does not require a screening and prescription by a licensed doctor before providing medicine" (U.S. Food and Drug Administration, n.d.). The company states that its compounded medications are not approved by the agency and require a consultation and prescription by a clinician (Noom, n.d.-d). The warning concerns compounded medication in general, not a finding about this company.

Nutrient gaps

Because the program forbids no food, its effect on nutrient intake depends on what a member chooses within the calorie range. A few studies recorded energy intake. In a single-group study of 31 breast cancer survivors, food logs from 25 of them showed about 1,300 calories a day early in the program and about the same late in it, with no significant change in carbohydrate, protein, fiber or an overall diet-quality score (Shen et al., 2024). The Mount Sinai registry posts average calorie intake from dietary recalls, falling from about 2,190 and 2,120 calories a day to about 1,680 and 1,580 at 30 months among the roughly one third of participants with data (Icahn School of Medicine at Mount Sinai, 2025).

The searches logged for this page did not identify a study assessing micronutrient adequacy, blood nutrient levels or deficiency in members. Whether a particular calorie range and the foods a person chooses meet their needs is a question for a registered dietitian, especially during pregnancy or breastfeeding, in older age, or with a diagnosed condition.

Who should not follow it without medical supervision

Following the company's own exclusions: anyone under 18, anyone pregnant, anyone with an eating disorder, and anyone underweight (Noom, n.d.-d, n.d.-j). Anyone breastfeeding, for whom the searches logged for this page did not identify a study. Anyone taking insulin, a sulfonylurea or a meglitinide, because of the risk of low blood sugar when eating less (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.), or taking medicine for blood pressure or warfarin, under the standing rule that the prescriber is kept informed before a substantial change in eating; the coach never changes a medication.

Anyone with a diagnosed condition that a calorie range affects, such as diabetes or kidney disease, whose plan belongs with their clinician and a registered dietitian; with kidney disease, needs depend on its stage, laboratory results and dialysis. And anyone considering weight-loss medication, which requires a clinician's evaluation under any program (U.S. Food and Drug Administration, n.d.). The special populations list in section 3 is built from these cautions.

Medication interactions

A coach never adjusts medication. The point of listing these is to know when the client should talk to the prescriber. Insulin, sulfonylureas and meglitinides can cause low blood sugar, and not eating enough carbohydrate, skipping or delaying meals, or increasing physical activity beyond the usual routine makes it more likely (National Institute of Diabetes and Digestive and Kidney Diseases, n.d.); a calorie range and a step goal can do all three. GLP-1 medications prescribed through Noom Med are managed by the prescriber, including their dose, their side effects and when to stop, and the agency's concerns about dosing errors with compounded versions apply (U.S. Food and Drug Administration, n.d.).

Signs it is going wrong

Physical signs that need the client's care plan, a medical assessment or emergency help include shakiness, sweating or confusion in a client taking insulin or a sulfonylurea, which may be low blood sugar and is treated first under the client's own plan, with emergency help if symptoms are severe or the person cannot manage the episode, and the prescriber told afterward; vomiting that does not stop, severe abdominal pain or other troubling symptoms in a client taking a GLP-1 medication, which go to the prescriber promptly; dizziness or unusual tiredness; missed menstrual periods; and weight loss the client did not intend, such as loss that continues after they have stopped trying. Each is a reason to involve a clinician, not a diagnosis.

Eating behaviors are a separate matter and are read with care. Logging every meal and weighing in daily are what the program asks for, and work schedules, shared kitchens, cultural practice, a medical diet or household preferences can explain many other changes. Behaviors that carry more meaning include skipping meals to stay under the range, distress or guilt after an orange food, weighing several times a day, avoiding meals with others that are hard to log, or a calorie range the client keeps lowering. 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, the coach follows their own referral process or brings it to the client's clinician under an existing care plan, and does not diagnose from a behavior alone.

Myths and where they came from

Myth: "Noom isn't a diet."

Where it came from: the company's own description of a program that does not forbid foods and "isn't a diet the way most people think of diets" (Noom Team, 2026b). What is true: the program sets a daily calorie range, asks members to log what they eat against it and suggests how much of each color to eat (Noom Team, 2026b, 2026c). It has no forbidden foods, and it is still a calorie-restricted eating plan.

Myth: "Research has proven that Noom works."

Where it came from: company pages that say research "has proven" the program helps people lose weight, citing a study in which almost 80 percent of users reported weight loss (Noom Team, 2026b). What is true: that study counted users who logged weights, with no comparison group (Chin et al., 2016). The randomized trials show a modest difference in measured weight by the end of a 4-month to 6-month program, most of them run or paid for by the company, and this page rates the short-term evidence very low in certainty; the searches logged for this page did not identify a study estimating the program's effect on heart disease, cancer or survival.

Myth: "My Noom coach is a health professional, so I can ask them anything."

Where it came from: the company's statement that many of its coaches are "board-certified health professionals" (Noom, n.d.-d). What is true: the company's terms say its coaches are not licensed medical professionals and cannot advise on diagnosis or treatment, and its support page says its one-on-one coaches cannot offer clinical or medical advice or act as licensed nutrition or mental health professionals (Noom, n.d.-j, n.d.-k). Medical, medication and nutrition-therapy questions go to the client's clinician or a registered dietitian.

Myth: "Green foods are good and orange foods are bad."

Where it came from: a traffic-light color scheme that looks like a judgment. What is true: the company says the colors describe calorie density and are not good or bad (Noom Team, 2026b, 2026c), and calorie density is a different property from nutritional value. Heart-health guidance encourages choosing sources of unsaturated fat in place of saturated fat (Lichtenstein et al., 2026), while the company's own example of a calorie-dense food is peanut butter (Noom Team, 2026b).

Myth: "The medication from Noom is the same as Ozempic or Wegovy."

Where it came from: marketing that offers GLP-1 programs alongside brand-name drugs. What is true: the company states that the medications in its GLP-1Rx programs are compounded and not reviewed by the U.S. Food and Drug Administration, and that only the makers of the approved brands have FDA-approved products containing semaglutide or tirzepatide (Noom, n.d.-d). The agency lists a claim that a compounded drug is the same as an approved one among its warning signs (U.S. Food and Drug Administration, n.d.).

7. Key terms

Noom Weight

The company's app-only weight-loss program, with a daily calorie range, food and weight logging, daily lessons and coaching by chat.

Calorie density

The number of calories a food carries for its weight; water-rich foods such as vegetables have a low calorie density.

Calorie range

The daily calorie target the app sets for a member; the company also calls it a calorie budget.

Color system

The program's labeling of foods as green, yellow or orange according to calorie density, which the company says does not mean good or bad.

Self-monitoring

Recording one's own eating, activity or weight, a standard tool of behavioral weight-loss programs.

Noom coach

A coach employed by the company who supports members by chat; the company says its coaches are not licensed medical professionals.

National Board Certified Health and Wellness Coach

A certification awarded by the National Board for Health and Wellness Coaching to a coach who meets its training and experience requirements and passes its examination.

Noom Med

The company's service that adds clinician visits and prescriptions for weight-loss medication to the app.

GLP-1 medication

A class of prescription drugs, such as semaglutide and tirzepatide, used for type 2 diabetes and weight management.

Compounded medication

A drug mixed by a pharmacy rather than made as an approved product; compounded drugs are not reviewed by the U.S. Food and Drug Administration for safety, effectiveness or quality.

Diabetes Prevention Program

A year-long lifestyle curriculum for people with prediabetes, recognized by the Centers for Disease Control and Prevention, that Noom delivers through its app.

Intention-to-treat analysis

An analysis that keeps every randomized person in the group they were assigned to, whether or not they used the program.

Completer analysis

An analysis limited to people who finished a program, which tends to make a program look more effective than it is for everyone offered it.

Retrospective registration

Registering a trial in a public registry after it has started, which makes it harder to confirm that the reported outcomes were the planned ones.

Auto-renewing subscription

A plan that charges again at the end of each billing period unless the member cancels.

Noom Monitor

A separate app built to support guided self-help for binge eating disorder and bulimia nervosa within clinical treatment.

8. Review questions

1. How does the program decide how much a member should eat, what do the green, yellow and orange labels describe, and who does the company say cannot use the program?
2. Drawing on section 4: in the prediabetes trial, how many people were randomized, what was compared, what happened to measured weight and to HbA1c at 6 and 12 months, and what features of the trial limit how much weight its result can carry?
3. A client tells you she has started Noom Med and asks whether her nausea means she should lower her dose. What does the coach do, and where is the referral line?
4. A client says his Noom coach is a health professional, so he follows the coach's advice about his blood pressure. Using the company's own statements in sections 3 and 6, how would you describe the difference between a Noom coach, a board-certified health and wellness coach and a licensed clinician, without disparaging anyone?
5. Drawing on sections 1 and 6: the company says research has proven that its program works, citing a study in which almost 80 percent of users reported weight loss. Where does that figure come from, and why can it not show that the program caused the weight loss?

9. References

Burke, L. E., Wang, J., & Sevick, M. A. (2011). Self-monitoring in weight loss: A systematic review of the literature. Journal of the American Dietetic Association, 111(1), 92-102. https://doi.org/10.1016/j.jada.2010.10.008

Chin, S. O., Keum, C., Woo, J., Park, J., Choi, H. J., Woo, J.-T., & Rhee, S. Y. (2016). Successful weight reduction and maintenance by using a smartphone application in those with overweight and obesity. Scientific Reports, 6, Article 34563. https://doi.org/10.1038/srep34563

Cho, S. M. J., Lee, J. H., Shim, J.-S., Yeom, H., Lee, S. J., Jeon, Y. W., & Kim, H. C. (2020). Effect of smartphone-based lifestyle coaching app on community-dwelling population with moderate metabolic abnormalities: Randomized controlled trial. Journal of Medical Internet Research, 22(10), Article e17435. https://doi.org/10.2196/17435

DeLuca, L., Toro-Ramos, T., Michaelides, A., Seng, E., & Swencionis, C. (2020). Relationship between age and weight loss in Noom: Quasi-experimental study. JMIR Diabetes, 5(2), Article e18363. https://doi.org/10.2196/18363

Hahn, S. L., Kaciroti, N., Eisenberg, D., Weeks, H. M., Bauer, K. W., & Sonneville, K. R. (2021). Introducing dietary self-monitoring to undergraduate women via a calorie counting app has no effect on mental health or health behaviors: Results from a randomized controlled trial. Journal of the Academy of Nutrition and Dietetics, 121(12), 2377-2388. https://doi.org/10.1016/j.jand.2021.06.311

Hildebrandt, T., Michaelides, A., Mackinnon, D., Greif, R., DeBar, L., & Sysko, R. (2017). Randomized controlled trial comparing smartphone assisted versus traditional guided self-help for adults with binge eating. International Journal of Eating Disorders, 50(11), 1313-1322. https://doi.org/10.1002/eat.22781

Hildebrandt, T., Michaeledes, A., Mayhew, M., Greif, R., Sysko, R., Toro-Ramos, T., & DeBar, L. (2020). Randomized controlled trial comparing health coach-delivered smartphone-guided self-help with standard care for adults with binge eating. American Journal of Psychiatry, 177(2), 134-142. https://doi.org/10.1176/appi.ajp.2019.19020184

Icahn School of Medicine at Mount Sinai. (2025, August 11). Noom Health for weight management (ClinicalTrials.gov Identifier NCT04797169). ClinicalTrials.gov. clinicaltrials.gov/study/NCT04797169

Jebeile, H., Libesman, S., Melville, H., Low-Wah, T., Dammery, G., Seidler, A. L., Jones, R. A., McMaster, C. M., Paxton, S. J., Hill, A. J., Ahern, A. L., Garnett, S. P., Braet, C., Wilfley, D. E., Baur, L. A., & Lister, N. B. (2023). Eating disorder risk during behavioral weight management in adults with overweight or obesity: A systematic review with meta-analysis. Obesity Reviews, 24(6), Article e13561. https://doi.org/10.1111/obr.13561

Keum, J., Chung, M. J., Kim, Y., Ko, H., Sung, M. J., Jo, J. H., Park, J. Y., Bang, S., Park, S. W., Song, S. Y., & Lee, H. S. (2021). Usefulness of smartphone apps for improving nutritional status of pancreatic cancer patients: Randomized controlled trial. JMIR mHealth and uHealth, 9(8), Article e21088. https://doi.org/10.2196/21088

Kim, M., Kim, Y., Go, Y., Lee, S., Na, M., Lee, Y., Choi, S., & Choi, H. J. (2020). Multidimensional cognitive behavioral therapy for obesity applied by psychologists using a digital platform: Open-label randomized controlled trial. JMIR mHealth and uHealth, 8(4), Article e14817. https://doi.org/10.2196/14817

Ku, E. J., Park, J. I., Jeon, H. J., Oh, T., & Choi, H. J. (2020). Clinical efficacy and plausibility of a smartphone-based integrated online real-time diabetes care system via glucose and diet data management: A pilot study. Internal Medicine Journal, 50(12), 1524-1532. https://doi.org/10.1111/imj.14738

Laudenslager, M., Chaudhry, Z. W., Rajagopal, S., Clynes, S., & Gudzune, K. A. (2021). Commercial weight loss programs in the management of obesity: An update. Current Obesity Reports, 10(2), 90-99. https://doi.org/10.1007/s13679-021-00428-y

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

May, C. N., Cox-Martin, M., Ho, A. S., McCallum, M., Chan, C., Blessing, K., Behr, H., Blanco, P., Mitchell, E. S., & Michaelides, A. (2023). Weight loss maintenance after a digital commercial behavior change program (Noom Weight): Observational cross-sectional survey study. Obesity Science & Practice, 9(5), 443-451. https://doi.org/10.1002/osp4.666

McCallum, M., Moulder, R. G., Earl, S., Blessing, K., Thompson, P., Ainsworth, M. C., Baldwin, M., & May, C. N. (2026). Longitudinal effects of a digital weight management program on weight loss: A randomized controlled trial. Obesity Science & Practice, 12(2), Article e70122. https://doi.org/10.1002/osp4.70122

Metzendorf, M.-I., Wieland, L. S., & Richter, B. (2024). Mobile health (m-health) smartphone interventions for adolescents and adults with overweight or obesity. Cochrane Database of Systematic Reviews, 2024(2), Article CD013591. https://doi.org/10.1002/14651858.CD013591.pub2

Michaelides, A., Mitchell, E. S., Behr, H., Ho, A. S., Hanada, G., Lee, J., & McPartland, S. (2022). Executive function-related improvements on a commercial CBT-based weight management intervention: Pilot randomized controlled trial. International Journal of Environmental Research and Public Health, 19(14), Article 8763. https://doi.org/10.3390/ijerph19148763

Michaelides, A., Raby, C., Wood, M., Farr, K., & Toro-Ramos, T. (2016). Weight loss efficacy of a novel mobile Diabetes Prevention Program delivery platform with human coaching. BMJ Open Diabetes Research & Care, 4(1), Article e000264. https://doi.org/10.1136/bmjdrc-2016-000264

Mitchell, E. S., Fabry, A., Ho, A. S., May, C. N., Baldwin, M., Blanco, P., Smith, K., Michaelides, A., Shokoohi, M., West, M., Gotera, K., El Massad, O., & Zhou, A. (2023). The impact of a digital weight loss intervention on health care resource utilization and costs compared between users and nonusers with overweight and obesity: Retrospective analysis study. JMIR mHealth and uHealth, 11, Article e47473. https://doi.org/10.2196/47473

Moody, S., Ross, L., Opitz, M.-C., Hemmings, A., İnce, B., Bryson, C., Kuehne, C., Douglas, D., Phillips, M., Langhorne, V., Schmidt, U., & Sharpe, H. (2025). Associations between the use of fitness and diet tracking technology and disordered eating behaviour: A systematic review. European Eating Disorders Review, 33(6), 1288-1313. https://doi.org/10.1002/erv.70006

National Board for Health and Wellness Coaching. (n.d.). Scope of practice. nbhwc.org/scope-of-practice

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

Nichols v. Noom, Inc., No. 1:20-cv-03677 (S.D.N.Y. Feb. 11, 2022) (Document 495, memorandum of law in support of plaintiffs' unopposed motion for preliminary approval of the class action settlement). storage.courtlistener.com/recap/gov.uscourts.nysd.536999/gov.uscourts.nysd.536999.495.0.pdf

Nichols v. Noom, Inc., No. 1:20-cv-03677 (S.D.N.Y. July 12, 2022) (Document 518, order and final judgment). storage.courtlistener.com/recap/gov.uscourts.nysd.536999/gov.uscourts.nysd.536999.518.0.pdf

Noom. (n.d.-a). About us. noom.com/about-us

Noom. (n.d.-b). Diabetes prevention. noom.com/research/diabetes-prevention

Noom. (n.d.-c). Lose weight easily with Noom's smart weight loss program. noom.com/lose-weight

Noom. (n.d.-d). Noom: Lose weight and keep it off. noom.com

Noom. (n.d.-e). Noom Med: GLP-1 medications for weight loss. noom.com/med

Noom. (n.d.-f). Noomiversity, Noom's health coach certification program. noom.com/nbhwc-training-program

Noom. (n.d.-g). Noom's suicide and self-harm prevention protocol. noom.com/support/faqs/coach-and-community/2026/01/nooms-suicide-self-harm-prevention-protocol

Noom. (n.d.-h). Publications. noom.com/research/publications

Noom. (n.d.-i). Research. noom.com/research

Noom. (n.d.-j). Terms and conditions of use. noom.com/terms-and-conditions-of-use

Noom. (n.d.-k). What can I ask my coach (or Welli)? noom.com/support/faqs/coach-and-community/2025/10/what-can-i-ask-my-coach-or-welli

Noom. (2025, October). Noom overview: Everyday wellbeing. noom.com/uploads/2026/09/June-Update-Noom-Overview-1.pdf

Noom Inc. (2022a, October 6). Investigation of healthy eating habits and attitudes toward food during a weight loss intervention (ClinicalTrials.gov Identifier NCT04586647). ClinicalTrials.gov. clinicaltrials.gov/study/NCT04586647

Noom Inc. (2022b, October 7). Randomized, parallel trial of overweight men and women using a coached or non-coached weight loss intervention (ClinicalTrials.gov Identifier NCT04645108). ClinicalTrials.gov. clinicaltrials.gov/study/NCT04645108

Noom Inc. (2023, November 18). Randomized, controlled trial of a weight loss intervention in improving quality of life in men and women with overweight or obesity (ClinicalTrials.gov Identifier NCT04871295). ClinicalTrials.gov. clinicaltrials.gov/study/NCT04871295

Noom Inc. (2024, October 31). Longitudinal effects of a digital WM program on weight loss: A randomized controlled trial (ClinicalTrials.gov Identifier NCT06666842). ClinicalTrials.gov. clinicaltrials.gov/study/NCT06666842

Noom Team. (2025, April 30). What is Noom Med and how much does it cost (and what do you get)? Noom. noom.com/blog/weight-management/noom-program/what-is-noom-med-and-how-much-does-it-cost-and-what-do-you-get

Noom Team. (2026a, June 3). How much does Noom cost (and what do you get)? Noom. noom.com/blog/how-much-does-noom-cost

Noom Team. (2026b, September 17). What is Noom (and how can it help you lose weight)? Noom. noom.com/blog/what-is-noom-how-does-noom-work

Noom Team. (2026c, September 17). What is the Noom "diet"? Noom. noom.com/blog/what-is-the-noom-diet

Northwestern University. (2026, June 8). Mobile health interventions to prevent heart disease after hypertensive disorders of pregnancy (ClinicalTrials.gov Identifier NCT06523569). ClinicalTrials.gov. clinicaltrials.gov/study/NCT06523569

Owen, E. C., Moulder, R. G., Morse, J. L., Ainsworth, M. C., Fabry, A., & Merner, A. R. (2026). Weight loss with a lower dose compounded semaglutide and behavioral weight management program: A real-world matched retrospective cohort study [Preprint]. medRxiv. https://doi.org/10.64898/2026.09.08.26362528

Rolls, B. J. (2009). The relationship between dietary energy density and energy intake. Physiology & Behavior, 97(5), 609-615. https://doi.org/10.1016/j.physbeh.2009.03.011

Shen, S., Salehi, E., White, C., Chen, Y., & Iyengar, N. M. (2024). Evaluation of a mobile behavior change program for weight loss in breast cancer survivors. npj Breast Cancer, 10, Article 53. https://doi.org/10.1038/s41523-024-00659-x

Stine, J. G., Rivas, G., Hummer, B., Duarte-Rojo, A., May, C. N., Geyer, N., Chinchilli, V. M., Conroy, D. E., Mitchell, E. S., McCallum, M., Michealides, A., & Schmitz, K. H. (2023). Mobile health lifestyle intervention program leads to clinically significant loss of body weight in patients with NASH. Hepatology Communications, 7(4), Article e0052. https://doi.org/10.1097/hc9.0000000000000052

Sysko, R., Bibeau, J., Boyar, A., Costello, K., Michaelides, A., Mitchell, E. S., Susanin, A., & Hildebrandt, T. (2022). A 2.5-year weight management program using Noom Health: Protocol for a randomized controlled trial. JMIR Research Protocols, 11(8), Article e37541. https://doi.org/10.2196/37541

Sysko, R., Michaelides, A., Costello, K., Herron, D. M., & Hildebrandt, T. (2022). An initial test of the efficacy of a digital health intervention for bariatric surgery candidates. Obesity Surgery, 32(11), 3641-3649. https://doi.org/10.1007/s11695-022-06258-8

Tak, C., Thompson, P., Morse, J. L., & McCallum, M. (2025). Health care resource utilization and health care costs among digital weight-loss intervention participants and nonparticipants. Obesity, 33(9), 1637-1644. https://doi.org/10.1002/oby.24337

Toro-Ramos, T., Heaner, M., Yang, Q., DeLuca, L., Behr, H., Reynolds, K., Kim, Y., & Michaelides, A. (2021). Postpartum weight retention: A retrospective data analysis measuring weight loss and program engagement with a mobile health program. Journal of Women's Health, 30(11), 1645-1652. https://doi.org/10.1089/jwh.2020.8584

Toro-Ramos, T., Michaelides, A., Anton, M., Karim, Z., Kang-Oh, L., Argyrou, C., Loukaidou, E., Charitou, M. M., Sze, W., & Miller, J. D. (2020). Mobile delivery of the Diabetes Prevention Program in people with prediabetes: Randomized controlled trial. JMIR mHealth and uHealth, 8(7), Article e17842. https://doi.org/10.2196/17842

U.S. Food and Drug Administration. (n.d.). FDA's concerns with unapproved GLP-1 drugs used for weight loss. fda.gov/drugs/drug-alerts-and-statements/fdas-concerns-unapproved-glp-1-drugs-used-weight-loss

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 program can be high for one outcome and very low for another. A row marked Not rated is an abstention, not a certainty level: where no study addresses a question, or none can be appraised, there is nothing to rate.

For Noom, every certainty judgment is this page's own, because the searches did not identify a published review that rated the program with GRADE or another named tool. The main claimed outcome is weight, so the banner reflects the first row, read as a measurement and not as a goal. Some rows report a study but abstain from a certainty level because the information needed for a complete outcome-specific appraisal was not established. These abstentions are distinguished from questions for which the searches found no adequate study.

Each rated row names its intervention, comparison, population, outcome and time point. For body weight, this page uses 1 kilogram for kilogram contrasts and 1 percentage point for percentage-weight contrasts as small measurement differences worth distinguishing from smaller changes. These are this page's analytical boundaries, not validated Noom-specific clinical minimums and not goals for a client. The steatohepatitis row instead uses the 2.5-kilogram contrast its trial was planned to detect. HbA1c uses 0.5 percentage points, also checked at 0.3. Where a symptom-score boundary or the information needed for a complete appraisal was not established, the page reports the study's results without assigning a certainty level. Company involvement is assessed through actual bias or publication concerns, not treated as a deduction by itself.

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
Measured body weight at the end of a 16-week to 6-month program (the banner rating)	Very low	GRADE, this page's own judgment. Comparison: Noom programs as the company delivers them, Noom Weight or its diabetes prevention version with coaching, against usual care with a printed curriculum, educational emails or newsletters, in adults with overweight, obesity or prediabetes. Three trials: Toro-Ramos et al. (2020), 202 randomized, 6 months, clinic weights; McCallum et al. (2026), 427 randomized and 412 analyzed, 16 weeks, self-reported weights; Michaelides et al. (2022), 70 randomized and 47 in the weight analysis, 4 months, office scale. Threshold: 1 percentage point for percentage-weight contrasts and 1 kilogram for kilogram contrasts, an analytical measurement boundary, not a client goal.	Randomized trials start high. Lowered 1 level for risk of bias: the largest trial used weights reported by participants who knew their group and excluded 15 people after randomization; the smallest omitted about a third of those randomized from its weight analysis; and the 2 largest were registered after they ended, so their analyses cannot be checked against a prospective public plan. Company employees helped design or analyze all 3, but funding or employment alone is not the deduction. The trial with clinic weights and an intention-to-treat analysis provides some protection against these concerns. Not lowered for inconsistency: all 3 favor the program. Not lowered for indirectness for the company-delivered programs and enrolled populations named here. Lowered 1 level for imprecision: the largest trial's difference of 1.8 percentage points has an interval of 0.7 to 2.8, crossing this page's 1-point boundary; the other trials do not provide a compatible pooled estimate resolving that uncertainty. Lowered 1 level for strongly suspected publication bias: NCT04871295 is a completed company trial against a wait-list control, with 370 participants and weight at 4 months as its primary outcome, for which no results are posted and the logged searches found no publication (Noom Inc., 2023). This does not establish why it remains unreported. Nothing raises the rating. High minus 3 levels is Very low; the size of the short-term difference is uncertain.
Measured body weight at 12 months, adults with prediabetes	Low	GRADE, this page's own judgment. Comparison: the diabetes prevention version with coaching against usual care with a printed curriculum. One trial, Toro-Ramos et al. (2020), 202 randomized, clinic weights. Threshold: 1 kilogram, an analytical measurement boundary, not a client goal.	Randomized trials start high. Lowered 1 level for risk of bias: the trial was registered in March 2019, after it ended in June 2018; the company helped design it and paid for parts of it; and about 18 percent of participants missed a follow-up visit, handled with a statistical model. Inconsistency cannot be judged from one trial, which is not a reason to lower it. Not lowered for indirectness. Lowered 1 level for imprecision: the 1.80-kilogram difference has an approximate interval of 0.21 to 3.39 kilograms, calculated from the reported standard error, spanning differences below and above this page's 1-kilogram boundary. Publication bias not detected for this outcome. No factor raises it.
Measured body weight at 16 weeks, adults with steatohepatitis	Low	GRADE, this page's own judgment. Comparison: Noom Weight against standard clinical care in adults with steatohepatitis, an inflamed form of fatty liver disease. One trial, Stine et al. (2023), 40 randomized, 33 with a final weight. Threshold: 2.5 kilograms, the difference this pilot trial was planned to detect, used here as an analytical boundary rather than a client goal.	Randomized trials start high. Lowered 1 level for risk of bias: allocation was not concealed, participants knew their group, final weights were reported by phone or in the app, and 5 of 20 program participants and 2 of 20 controls had no final weight. Company employees were among the authors; employment alone is not the deduction. Inconsistency cannot be assessed from one trial. Not lowered for indirectness for this program in the adults with steatohepatitis named here. Lowered 1 level for imprecision: the reported mean changes differ by 5.2 kilograms. A conservative unadjusted approximation using the reported change standard deviations, 5.8 and 4.6 kilograms, and 15 and 18 observed final weights gives an interval of roughly 1.4 to 9.0 kilograms, crossing the 2.5-kilogram boundary. This is the page's observed-data calculation, not a published intention-to-treat confidence interval. Publication bias not detected; the trial was registered before enrollment. Nothing raises the rating. High minus 2 levels is Low.
Measured body weight at 24 weeks, the Noom app with clinic coaching, Korean adults with metabolic risk factors	Moderate	GRADE, this page's own judgment. Comparison: the Noom app with personalized coaching from clinical dietitians and exercise coordinators against no app, after the same baseline education for everyone. One independent, government-funded trial, Cho et al. (2020), 129 randomized across 3 groups. The logging-only group is described in section 4 and not rated, because the paper gives only a P value for its weight comparison. Threshold: 1 kilogram, an analytical measurement boundary, not a client goal.	Randomized trials start high. Risk of bias not judged serious for this measured outcome: assessors were masked, the trial was registered, it was independently funded, and the reported analysis included participants with at least one follow-up; the limitations of attendance remain relevant. Inconsistency cannot be assessed from one trial. Indirectness not serious for the app plus clinical dietitian and exercise-coordinator package expressly named in this row; this judgment does not transfer to ordinary company coaching. Lowered 1 level for imprecision: the estimated advantage is 0.87 kilograms, with an approximate interval from 0.05 to 1.69 kilograms, crossing this page's 1-kilogram boundary. Publication bias not detected. Nothing raises the rating. High minus 1 level is Moderate.
Measured body weight at 68 weeks, 52 weeks after a 16-week program	Low	GRADE, this page's own judgment. Comparison: Noom Weight for 16 weeks against 16 weekly educational email newsletters, in adults with overweight or obesity who had never been diagnosed with an eating disorder and did not have diabetes. One trial, McCallum et al. (2026), read in full: 427 randomized, 412 analyzed, weights at 68 weeks for 183 and 174. Threshold: 1 percentage point, an analytical measurement boundary, not a client goal.	Randomized trials start high. Lowered 2 levels for very serious risk of bias: weights were taken at home and reported by participants who knew their group, and the authors say the long-term result could partly reflect that; 15 people were dropped after randomization for missing or unreliable data; the trial was registered in October 2024, after it ended, so its 68-week analysis cannot be checked against a plan, and the registry's description of weighing differs from the paper's; and the company ran and analyzed it. Inconsistency cannot be judged from one trial, which is not a reason to lower it. Not lowered for indirectness: this is the program as sold, in the people it is sold to. Not lowered for imprecision: the 5.5-percentage-point difference has an interval of 3.0 to 8.0, entirely above the 1-point boundary; one trial does not itself require a precision deduction. Publication bias not detected for this outcome; the unreported company trials ended at 4 months. No factor raises it. Reported as a measurement, not as a goal.
HbA1c at 6 months, adults with prediabetes	Moderate, for little or no difference	GRADE, this page's own judgment. Comparison: the diabetes prevention version with coaching against usual care with a printed curriculum. One trial, Toro-Ramos et al. (2020), 202 randomized, laboratory HbA1c. Threshold: 0.5 percentage points, the difference the trial was designed to detect; the judgment is the same at 0.3.	Randomized trials start high. Lowered 1 level for risk of bias: registration occurred after the trial ended, the company helped design and fund it, and about 18 percent missed a follow-up, with missing data handled by maximum likelihood. HbA1c was a laboratory outcome. Inconsistency cannot be assessed from one trial. Not lowered for indirectness for this prevention package in adults with prediabetes. Not lowered for imprecision at the stated boundaries: the estimated difference is 0.004 percentage points, with an approximate interval of −0.094 to 0.102 calculated from the reported standard error of 0.05. The whole interval lies within both 0.5 and 0.3 points in either direction. The authors describe the trial as underpowered for HbA1c, but that observation alone does not show that this interval permits an important difference at these boundaries. Publication bias not detected. Nothing raises the rating. High minus 1 level is Moderate for little or no difference at these boundaries, not proof of an exactly zero effect.
HbA1c at 12 months, adults with prediabetes	Moderate, for little or no difference	GRADE, this page's own judgment. Comparison and trial as in the row above. Threshold: 0.5 percentage points; the judgment is the same at 0.3.	Randomized trials start high. Lowered 1 level for risk of bias for the same reasons as the 6-month row. Inconsistency cannot be assessed from one trial. Not lowered for indirectness for this prevention package in adults with prediabetes. Not lowered for imprecision at the stated boundaries: the estimated difference is 0.006 percentage points, with an approximate interval of −0.131 to 0.143 calculated from the reported standard error of 0.07, wholly within both 0.5 and 0.3 points in either direction. The authors' underpowering statement alone is not a further deduction for this threshold-specific conclusion. Publication bias not detected. Nothing raises the rating. High minus 1 level is Moderate for little or no difference at these boundaries, not proof of an exactly zero effect.
What added human support adds to the app, measured body weight	Very low	GRADE, this page's own judgment. Target comparison: a Noom app with added company coaching against the same app with less or no company coaching, in adults with overweight, obesity or metabolic risk factors. Supporting evidence covers 8 weeks to 30 months: the Mount Sinai registry comparison, Cho et al. (2020) and Kim et al. (2020). The latter 2 used clinical staff or psychologists and are supplementary indirect evidence for company coaching. NCT04645108 is the completed, unreported company trial of coached versus uncoached versions.	Randomized trials start high. Lowered 1 level for risk of bias: one trial analyzed only people who attended follow-up; in the Mount Sinai trial, weights were read from home scales during video calls, about a third of participants had no 30-month weight, and two investigators reported equity in the company in their other work (Sysko, Michaelides, et al., 2022). Not lowered for inconsistency: the point estimates lean the same way. Lowered 1 level for indirectness: in two of the three trials the added support came from clinic staff or psychologists rather than the company's coaches. Lowered 1 level for imprecision: the evidence available for this target does not supply a compatible set of between-group estimates and intervals that resolves the size of the added company coaching effect; descriptive changes within registry groups are not that comparison. Publication bias strongly suspected: the company's own completed trial of exactly this comparison, coached against uncoached, with 246 people, has no results posted or publication found (Noom Inc., 2022b). That reaches the floor. Very low means the size of any added effect is very uncertain; it does not mean coaching adds nothing.
Eating disorder questionnaire score at 68 weeks, adults never diagnosed with an eating disorder	Not rated, appraisal incomplete	GRADE, this page's own judgment. Comparison: Noom Weight for 16 weeks against educational email newsletters; outcome the Eating Disorder Examination Questionnaire global score. One trial, McCallum et al. (2026), read in full. No defensible smallest important change was established by this page for this score; descriptive results are reported without a certainty level.	The trial reported an EDE-Q global-score difference of −0.14 points, with a confidence interval of −0.33 to 0.04. Scores fell over time in both groups; that is not a program-specific treatment effect. This page did not establish a defensible smallest important difference on this scale, so the interval's crossing zero cannot by itself establish either meaningful-effect imprecision or little or no important difference. A complete outcome-specific certainty appraisal is therefore not assigned. Relevant bias concerns include unblinded questionnaire responses, late registration and postrandomization exclusions. Inconsistency cannot be assessed from this trial; its population is direct for adults never diagnosed with an eating disorder, not for excluded high-risk patients. Publication bias is not separately assigned for this questionnaire at 68 weeks; the different 4-month questionnaire trial remains a concern for the wider question. No upgrade is assigned. Not rated here means appraisal incomplete, not that no study exists or that the program is safe or harmful.
Eating disorder questionnaire scores at 8 weeks, bariatric surgery candidates	Not rated, appraisal incomplete	GRADE, this page's own judgment. Comparison: an 8-week Noom program for bariatric surgery against standard bariatric care; outcome the Eating Disorder Examination Questionnaire. One trial, Sysko, Michaelides, et al. (2022), 50 randomized, 36 at follow-up, read in full; eating disorders were not an exclusion.	This 50-person trial had 36 participants at follow-up and was powered only for comparatively large effects. The reported EDE-Q comparisons were not statistically significant, but that does not establish little or no important effect. This page did not establish a smallest important score difference or a usable interval for the relevant comparison, so it does not assign a completed outcome-specific certainty rating. Unblinded self-report and missing follow-up are bias concerns; investigator equity and company participation are disclosed but are not deductions by themselves. Inconsistency cannot be assessed from one trial; the intervention and bariatric population are direct for this named question. Precision remains unappraised against an important-score boundary; publication bias was not detected for this outcome; no upgrade is assigned. Not rated means appraisal incomplete, not an absence of a trial or evidence of safety for people with an eating disorder history.
Remission from binge eating at 52 weeks, adults with binge eating disorder or bulimia nervosa	Not rated, full-report appraisal incomplete	GRADE, this page's own judgment. Comparison: 8 telemedicine sessions of guided self-help based on cognitive behavioral therapy, delivered by a health system's health coaches with the separate Noom Monitor app, against standard care in that health system, which included no eating disorder treatment in the first 12 weeks. One trial, Hildebrandt et al. (2020), 225 randomized, read as its abstract only. This row describes that clinical package, not Noom Weight.	The abstract reports remission from binge eating at 52 weeks in 56.7 percent of the clinical-package group and 30 percent of standard care. It does not give the remission assessment denominators or an interval for this remission contrast. Its interval of −1.06 to −0.25 concerns binge-eating days, a different outcome, and cannot establish the precision of remission. The full published report and its financial declaration were not read, leaving allocation, assessment masking and missing-data handling unresolved. The clinical package is direct for the question named here; the earlier trial used a different active comparator, so inconsistency cannot be judged from it. Publication bias was not detected, and no upgrade is assigned. This page reports the abstract's finding without completing a 5-domain certainty rating. Not rated means appraisal incomplete; it does not mean the treatment has no benefit, and it is not evidence that Noom Weight treats an eating disorder.
Health care use, 6 and 12 months after starting	Very low	GRADE, this page's own judgment. Comparison: people who used Noom Weight against people who did not, in insurance and health record data. Two company-run analyses: Mitchell et al. (2023), 43,047 users and 14,555 nonusers, 12 months; Tak et al. (2025), 723 participants matched to 723 employees offered the program who did not enroll, 6 months.	Observational studies start low. Lowered 1 level for risk of bias: users were defined by their own engagement, and people who choose a program and keep using it differ from those who do not in ways that weighting and matching on measured traits cannot remove. Not lowered for inconsistency: both found fewer visits among users. Not lowered for indirectness for users and comparators in these datasets. Precision was not fully appraised against an outcome-specific important difference; large samples alone do not establish it. No additional precision deduction is assigned, and the serious selection/confounding concern already brings the observational evidence to Very low. Publication bias not counted, though both analyses come from the company. No factor raises it. These are associations, not effects of the program.
Health care costs, 6 to 24 months after starting	Very low	GRADE, this page's own judgment. Comparison and analyses as in the row above: Mitchell et al. (2023) at 12 months, and at 24 months in the subset of users followed that long; Tak et al. (2025) at 6 months.	Observational studies start low. Lowered 1 level for risk of bias, for the same reasons as health care use; the 24-month figure also comes from a subset whose members may differ from the rest. Not lowered for inconsistency: both found lower costs among users. Not lowered for indirectness for the users and comparators named here. Precision was not fully appraised against an important-cost boundary; no additional precision deduction is assigned, and the selection/confounding concern already brings the observational evidence to Very low. Publication bias not counted. No factor raises it. These are associations, not savings caused by the program.
Heart attacks, strokes, cancer and death	Not rated	This page's own judgment. Noom programs against any comparison.	The searches logged for this page did not identify a trial or cohort estimate of the program's effect on heart attacks, strokes, cancer or survival; the Mount Sinai registry's zero deaths are adverse-event counts in 2 app groups, not an estimate of a mortality effect. Not rated means there is no body of evidence to rate; it does not mean the program has no effect on them and does not show that it is safe.
Weight with Noom Med against medication alone	Not rated	This page's own judgment. Comparison behind the company's figures of 37, 39 and 48 percent more weight lost.	The company's figures come from its own analyses of members, described in footnotes, and its home page says the association does not imply causation (Noom, n.d.-d). The 2026 preprint compares the company's own programs, not medication alone (Owen et al., 2026). This page does not rate medication outcomes, which belong to the prescriber.

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 11, 2026, after a combined final review of the fourth batch of 5 archive pages, the commercial programs, the repairs it asked for and a targeted recheck. Written October 11, 2026. The judgments in the table are this page's own, assigned from the evidence. Some sources could be read only as abstracts or author manuscripts, and the page says so where each is used. Company prices and terms are a snapshot from the company's own pages on October 11, 2026, and can change.

What changed: After the combined final review, the 2026 company trial was read in full and its methods, numbers and limits replaced the earlier account from its abstract. The weight row was narrowed to the end of the program, with separate rows for 12 months, fatty liver disease and the independent Seoul trial, and the 68-week weight row moved from very low, provisional, to low. HbA1c was split into 6-month and 12-month rows with intervals and a stated threshold. The eating disorder row was split by trial, the binge eating treatment row moved from low to moderate, still provisional, and health care use and costs were split.

Adverse events, energy intake records, 2 unreported company trials and a 2025 review of tracking technology were added; the Seoul trial's regression result and the number randomized in the 4-month pilot were corrected; and absence statements, kidney and medication cautions and the wording on coach training were tightened. First version October 11, 2026.

What changed after the targeted recheck: The weight rows now compare each interval with a stated analytical boundary, 1 kilogram or 1 percentage point, and 2.5 kilograms for the fatty liver trial, rather than with no difference; these boundaries are not goals. The end-of-program weight row, and with it the banner, moved from low to very low; the Seoul trial row moved from low to moderate; both HbA1c rows moved from low to moderate, for little or no difference. The 2 eating disorder questionnaire rows and the binge eating remission row are now not rated, appraisal incomplete, with their trial results still reported. The emergency step for low blood sugar, the eating disorder history caution, and the statements about deaths and safety were made more exact.

Next scheduled review: Each year, because the company changes its programs, prices and medication offerings often. 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.