What Is Food Noise? What the Research Shows on GLP-1 Drugs

Food noise is a term that spread from patient anecdotes into media and clinical conversation ^[2], but the formal definition — persistent, intrusive thoughts about food that are disruptive to daily life and make healthy behaviors difficult — was written by the research team behind the Food Noise Questionnaire ^[1]. A second expert panel defines it as persistent food thoughts that feel unwanted or distressing and may cause social, mental, or physical harm ^[2]. It is a research construct rather than an established clinical diagnosis, and it is not the same as ordinary hunger or an occasional craving ^[2][8]. What the research actually shows is this: two questionnaires now measure it, and in a 417-person observational cohort, people who started a GLP-1 receptor agonist alongside a behavioral weight program saw a larger one-month drop in food noise scores than people in the program alone, with a baseline-adjusted between-group difference of -3.0 points on a 0-20 scale ^[5]. A separate survey of 550 semaglutide users reported a recalled drop from a median of 13 to 6 ^[7]. Neither study proves GLP-1 drugs cause the reduction, and the long-term picture is still largely unmapped ^[5][7][8].

What is food noise, exactly?

The team behind the Food Noise Questionnaire (FNQ), based at Pennington Biomedical Research Center, defines food noise as persistent, intrusive thoughts about food that are disruptive to daily life and make healthy behaviors difficult ^[1]. An expert panel writing in Nutrition & Diabetes arrived at a similar formal definition: persistent thoughts about food that the person perceives as unwanted and/or distressing, which may cause harm including social, mental, or physical problems ^[2].

That panel drew a line between food noise and routine food thoughts. Everyone thinks about food. Food noise is set apart by its intensity and intrusiveness, and at its more dysfunctional levels it resembles rumination, the repetitive, obsessional thinking that interferes with other mental activity ^[2].

A third framing comes from a 2023 narrative review in Nutrients by Hayashi and colleagues, which treats food noise as heightened and/or persistent food cue reactivity, often leading to intrusive food thoughts and maladaptive eating behaviors ^[3]. The FNQ team noted a limitation of that framing: patients report food noise even when no food cues are present, even though cues can trigger or worsen it ^[1]. The INFORM survey authors make the same point, noting that these thoughts can occur independently of hunger or food-related environmental cues ^[7].

What does food noise feel like day to day?

Patient accounts describe a constant inner dialogue about food: what to eat, how much to eat, and when to eat again ^[1]. One patient quoted in the Nutrition & Diabetes paper put it this way: “I feel like it is always on my mind… am I doing the right thing? Am I eating enough, or not enough? Was that the right amount of calories? Should that have been more protein?… I’d say 80% of the time, it’s on my mind” ^[2].

The Hayashi review lists concrete examples from anecdotal reports: thinking about food all the time, particularly highly palatable and energy-dense foods; feeling tempted to check food delivery apps multiple times a day; and thinking about the next meal while still eating the current one ^[3].

Notably, food noise is not only about cravings. The expert panel found that the mental chatter also includes worry over calories, macronutrient balance, meal timing, and whether a person is eating the “right” way in the “right” amounts ^[2]. Related research uses the term food-related intrusive thoughts (FRITs), which are believed to occur in people with and without clinically diagnosed eating disorders, particularly when struggling with body weight or body image ^[3].

Where did the term food noise come from?

Nobody knows who first said it. The Nutrition & Diabetes panel reports that the phrase appears to have originated from patient anecdotes, which then spread to the media and clinicians, and that the original source is difficult to isolate ^[2]. What made the term take off, the panel suggests, is that food noise became noticeable mainly through its absence: the sudden quieting people reported after starting a GLP-1 receptor agonist for weight loss ^[2].

Google search data tracks the rise. Queries for “food noise” noticeably increased in the fall of 2022 and peaked in spring 2024, remaining fairly high since ^[2]. The FNQ team noted that, to date, one publication had reviewed food noise from a scientific viewpoint ^[1]: the Hayashi et al. narrative review, published in Nutrients in November 2023, which proposed the Cue-Influencer-Reactivity-Outcome (CIRO) model of food cue reactivity as a framework for studying it ^[3].

How is food noise measured?

There are two published instruments, and they differ in both design and availability.

The Food Noise Questionnaire (FNQ) was developed and validated by Diktas and colleagues at Pennington Biomedical, with coauthors affiliated with WW International, and published in the journal Obesity in January 2025 ^[1][8]. The study enrolled 400 participants, with data from 396 analyzed ^[1]. The FNQ showed excellent internal consistency (Cronbach’s alpha of 0.93) and high test-retest reliability (r = 0.79 at a mean of about 7 days between administrations). Factor analysis found the five items loaded onto a single factor, and the questionnaire showed good convergent validity (all r > 0.78) and discriminant validity (all r < 0.39) ^[1]. The authors state that further research is needed to evaluate its clinical utility ^[1].

The second instrument, the Ro Allison Indiana Dhurandhar Food Noise (RAID-FN) Inventory, came out of an expert panel convened at the American Society of Nutrition meeting in Chicago on June 28, 2024 ^[2]. The panel built the draft items around four features of food noise: cognitive burden, persistence, dysphoria, and self-stigma ^[2]. It circulated 34 draft items to 10 experts; 27 drew a high level of agreement as relevant and representative, and the panel added 2 items for a final total of 29 draft items for future sample testing ^[2]. At the time of publication, the validity analysis was still under way, so the RAID-FN’s final form is not yet established ^[2].

The two tools also differ in access. The FNQ is available for non-commercial purposes under a Creative Commons license and was developed in partnership with and deployed by WW International ^[8]. The RAID-FN is proprietary, and any use must be requested through the US telehealth company Ro, which began marketing it to consumers in late 2025 as a way to assess whether they could benefit from pharmacological quieting of food noise ^[8].

How is the Food Noise Questionnaire scored?

The FNQ asks five questions, each answered on a 5-point scale: strongly disagree (0), disagree (1), neither agree nor disagree (2), agree (3), and strongly agree (4). The responses are summed for a total score from 0 to 20, with higher scores indicating more food noise. The instructions ask respondents to reflect on their thoughts over the last 2 weeks only, a recall period chosen so that brief situational experiences would not drive the rating ^[1][5].

The five statements, as listed in the ECO 2026 presentation of the questionnaire, are ^[5]:

1. I find myself constantly thinking about food throughout the day.
2. My thoughts about food feel uncontrollable.
3. I spend too much time thinking about food.
4. My thoughts about food have negative effects on me and/or my life.
5. My thoughts about food distract me from what I need to do.

For context, in the FNQ validation sample, participants with a BMI of 25.0-39.9 had a median score of 7.0, and those with a BMI of 40 or higher had a median of 9.5 ^[7]. No cutoff separates “food noise” from “no food noise”: the score is a continuum, the minimal important difference has not been defined, and the FNQ is a research and measurement tool, not a diagnostic test ^[1][7].

What did the 2026 European Congress on Obesity study find?

At the 33rd European Congress on Obesity in Istanbul in May 2026, Dr. Hanim Diktas and colleagues presented an observational cohort study of 417 adults in a digital behavioral weight management program (WeightWatchers) ^[5]. Of these, 92 started a GLP-1 receptor agonist alongside the program (the Med+ group) and 325 did not (the Core+ group). People already using GLP-1s or any weight-loss medication at baseline were excluded ^[5].

The two groups started at different levels. Unadjusted mean FNQ scores fell from 13.1 to 8.7 in the Med+ group and from 10.7 to 9.7 in the Core+ group, and the baseline difference between groups was statistically significant ^[5]. After statistical adjustment for those baseline scores, the one-month change was -4.05 points for the Med+ group versus -1.15 for the Core+ group, a between-group difference of -3.0 ^[5].

The study’s demographics are worth knowing when weighing how far the findings generalize: 94% of participants were white, 93% were female, mean age was 59, and mean BMI was 34 ^[5]. The one-month mark was used because it was the first follow-up time point available for most participants; analyses of weight change and its relationship to food noise were still pending the full manuscript ^[5].

Because this was an observational poster, not a randomized trial, it shows an association between starting a GLP-1 and a larger drop in food noise scores, not proof that the drugs caused it ^[5].

What did the INFORM survey of semaglutide users find?

The INFORM survey, published in 2026, included 550 US adults using injectable semaglutide for weight management, recruited from the Numerator opinion panel, with responses collected between May 13 and May 21, 2025 ^[7]. Respondents were 86% women and 79% white, with a mean age of 53, and 81% had been treated for at least 4 months ^[7].

Based on recollection of food noise before starting semaglutide, the median FNQ score was 13 (interquartile range 10-16). After starting semaglutide, the median was 6 (IQR 3-10), for a median change of -5 (IQR -10 to -2) ^[7]. The share of participants agreeing or strongly agreeing with statements indicating food noise fell from 47-63% before to 15-20% after, and 83% reported satisfaction with treatment ^[7].

The survey has real limits. It was cross-sectional with a single measurement taken after treatment had already started, so the “before” scores are recalled after the fact, not measured at baseline. The authors flag recall bias that could inflate baseline estimates, along with a convenience sample and the potential for selection bias. There was no control group, and the survey was designed collaboratively by Novo Nordisk and Numerator ^[7].

How might GLP-1 drugs quiet food thoughts?

The most direct controlled evidence comes from appetite and eating-behavior trials rather than food noise studies per se. In a 20-week randomized, placebo-controlled trial of 72 adults with obesity, once-weekly semaglutide 2.4 mg cut ad libitum energy intake at a test lunch by 35% versus placebo ^[2]. Participants on semaglutide also reported less hunger and prospective food consumption, more fullness and satiety, better control of eating, and fewer and weaker food cravings ^[2]. A smaller 12-week trial of 30 people at a lower semaglutide dose (1 mg) found similar effects on appetite and cravings ^[2].

For tirzepatide, a 6-week phase 1 randomized controlled trial found reductions in appetite, hunger, cravings, tendency to overeat, and reactivity to environmental food cues versus placebo ^[7].

On mechanism, the Nutrition & Diabetes panel notes that GLP-1 receptor agonists may promote weight loss by influencing brain pathways that regulate appetite, including in the hypothalamus and hindbrain, and that bariatric surgery, which also raises post-meal GLP-1 secretion, similarly reduces appetite and cravings ^[2].

Is food noise the same as food tasting different on GLP-1s?

Not quite, and the distinction matters. A 2026 review in Frontiers in Nutrition separates food-related processing into three levels: sensory (detecting taste and texture), liking (how pleasurable food feels), and wanting (the motivational drive toward food) ^[9]. The review’s wanting endpoints include craving, cue reactivity, and food preoccupation ^[9] — the territory food noise describes.

That review found no consistent evidence of a primary impairment of basic taste function during GLP-1 receptor agonist treatment. Instead, altered eating experience on these drugs fits better with state-dependent food revaluation and reduced reward-driven motivation ^[9]. The review also notes responses vary between people, and can include limited response, persistent appetite, or weight regain in some individuals ^[9].

What do scientists criticize about the food noise concept?

A 2026 commentary in Appetite by Brewis, Hayashi, and colleagues raised several concerns ^[8]. One is measurement: the existing instruments may overlap conceptually with established scales for food preoccupation and cravings, evidence of clinical utility is limited, and it is unclear whether food noise is a distinct construct or an overlapping dimension of known appetite-related processes ^[8].

The commentary also raises ethical questions. It asks whether it is clinically desirable, safe, and ethical to pharmacologically suppress food-related cognition in contexts of food insecurity or among people with restrictive eating disorders, and whether the popularization of food noise as a treatable pathology is redefining what counts as “normal” appetite and thoughts ^[8]. The authors call for stronger theory, mixed-methods instrument development grounded in lived experience, cross-cultural validation, and very cautious clinical use until those questions are resolved ^[8]. The FNQ authors themselves state that clinical utility still needs research ^[1].

Does food noise go away completely, or come back?

The evidence so far suggests food noise is a matter of degree, not an on/off switch. The FNQ is a 0-20 continuum, and in the INFORM survey the median score after starting semaglutide was 6, not 0 ^[7]. The expert panel likewise describes food noise as existing to varying degrees, and not always at levels that cause distress ^[2].

On return, the picture is thin. The Nutrition & Diabetes panel notes that after weight loss, ghrelin rises and GLP-1 and PYY fall, and it calls food noise’s possible contribution to weight regain a plausible but untested hypothesis ^[2]. The ECO study measured only one month, and INFORM measured a single retrospective time point, so neither study tells us how food noise behaves over the long run ^[5][7].

Why does the “food noise” label matter?

The expert panel points out that if food noise is seen as a weakness or a lack of willpower, people may feel shame and self-blame for experiencing it ^[2]. A biological framing could shift the focus away from willpower, but the panel notes a possible downside: it could also reduce how much control people feel they have without biologically active interventions such as GLP-1 medications ^[2].

The ECO authors suggest one practical angle: a change in FNQ score may serve as an early indicator of treatment response, though that idea awaits the fuller analyses ^[5].

If you are taking a GLP-1 and notice food thoughts changing, returning, or persisting, that is information worth bringing to your prescriber or clinician rather than acting on alone.

Sources

1. Obesity. Development and validation of the Food Noise Questionnaire. 2025.
2. Nutrition & Diabetes. Food noise: definition, measurement, and future research directions.
3. PubMed Central. What Is Food Noise? A Conceptual Model of Food Cue Reactivity.
4. ClinicalTrials.gov. Development and Validation of a Brief Food Noise Questionnaire.
5. News-Medical. Adding GLP-1 drugs to behavioral interventions silences food noise. May 13, 2026.
6. PubMed Central. Retrospective Assessment of Food Noise Changes After Initiation of Injectable Semaglutide for Weight Management in the USA: The INFORM Survey.
7. Appetite. Food noise: Conceptual, methodological, and ethical considerations. November 1, 2026.
8. Frontiers in Nutrition. Altered eating experience during GLP-1 receptor agonist therapy: a sensory–liking–wanting framework for food preference and nutritional behaviour.
9. Diabetes, Obesity and Metabolism. The effect of semaglutide 2.4 mg once weekly on energy intake, appetite, control of eating, and gastric emptying in adults with obesity. 2021.