What to Eat on Ozempic and GLP-1: 45 Foods Scored by Insulin
By Dino Pohilj, Founder, RealFoods
Why does food still matter on Ozempic if I am barely hungry?
GLP-1 medications suppress appetite through two mechanisms: they slow gastric emptying and they reduce the hunger signal from the brain. What they do not do is alter the insulin response to food. A single glass of orange juice, even a small one, still delivers concentrated fructose and glucose to the liver, produces a rapid insulin spike, and puts the body into fat-storage mode for several hours. The medication will make you less hungry, but it cannot override the hormonal command that sugar sends.
This matters more on GLP-1 than off it, because the smaller meals you eat carry disproportionate weight. If you are eating 400 calories instead of 1,800, a single high-insulin-impact choice represents a larger share of your total daily hormonal load. A croissant and a sweetened latte eaten without hunger may be the only food-related event of your morning, and they could pause fat burning for four to six hours.
The SURMOUNT-1 trial (Jastreboff et al., NEJM 2022) found that tirzepatide at 15 mg produced a mean weight reduction of 20.9% over 72 weeks. That outcome is real, but it is anchored in participants who were given structured lifestyle guidance alongside the medication. Food quality is not optional on GLP-1. It is the mechanism that determines how much of the weight lost is fat versus lean mass, and whether results hold when the medication changes or stops.
What should I eat on GLP-1? The 45-food scored table
The scores below use the 1-10 Weight Impact (WI) Scale, which measures each food's insulin response. Scores 1-3 support fat burning. Scores 7-10 pause it. The insulin-index framework, established by Holt, Miller, and Petocz in their 1997 paper in the American Journal of Clinical Nutrition, demonstrated that food category and composition predict insulin response with measurable reliability, and that protein-rich foods and bakery products in particular prompt insulin responses disproportionate to their glucose content. The scoring below applies that framework to foods most commonly eaten on GLP-1 medication.
How to read these scores
Each score reflects the food's insulin impact at a typical serving. In the RealFoods app, your actual WI Score adjusts for the portion you logged. Low-scoring foods like salmon and broccoli stay low regardless of portion. High-scoring foods like white rice and juice scale up quickly with volume.
Proteins
| Food | WI Score | Why |
|---|---|---|
| Chicken breast | 1/10 | Pure protein, negligible insulin response. |
| Eggs | 1/10 | Protein and fat combined, keeps blood sugar perfectly stable. |
| Salmon (grilled or baked) | 1/10 | Omega-3 rich, zero carbohydrate, excellent hormone balance. |
| Steak (sirloin, ribeye) | 1/10 | Zero carbs, highly satiating, no glucose spike. |
| Canned sardines | 1/10 | High protein and omega-3, no added carbohydrate. |
| Cottage cheese (plain, full-fat) | 2/10 | Protein-dominant with fat, minimal carbs from lactose. |
| Greek yogurt (plain, full-fat) | 2/10 | Fat and protein buffer the lactose; avoid sweetened versions. |
| Turkey breast | 1/10 | Lean protein, no carbohydrate, minimal insulin response. |
Non-starchy vegetables
| Food | WI Score | Why |
|---|---|---|
| Broccoli | 2/10 | Mostly fibre, very low net carbs, outstanding fibrous armor. |
| Spinach | 1/10 | Essentially zero net carbs, nutrient dense, keeps insulin flat. |
| Courgette (zucchini) | 1/10 | Low starch, high water, almost no glucose load. |
| Bell peppers | 2/10 | Mild sweetness but high fibre keeps glucose release gentle. |
| Cauliflower | 2/10 | Fills the plate, fibre buffered, low net carb. |
| Cucumber | 1/10 | 95% water, minimal carbs, no measurable insulin spike. |
| Kale | 1/10 | High fibre, low sugar, excellent metabolic support. |
| Asparagus | 1/10 | Prebiotic fibre, very low carbohydrate content. |
Fats and dairy
| Food | WI Score | Why |
|---|---|---|
| Avocado | 1/10 | Mostly fat and fibre, near-zero net carbs, ideal GLP-1 complement. |
| Olive oil (extra virgin) | 1/10 | Pure fat, zero carbs, reduces post-meal glucose curves. |
| Almonds | 2/10 | Dense fat and fibre matrix, very low glycaemic impact. |
| Walnuts | 2/10 | Omega-3 rich, slow-digesting, supports stable blood sugar. |
| Cheese (cheddar, feta, brie) | 1/10 | High fat and protein, negligible carbs, insulin stays low. |
| Butter and ghee | 1/10 | Pure fat, zero carbohydrate, no insulin response. |
Fruit
| Food | WI Score | Why |
|---|---|---|
| Strawberries | 2/10 | High water and fibre perfectly buffers the natural sugars. |
| Blueberries (small portion) | 3/10 | Denser in sugar than strawberries, but still well buffered. |
| Raspberries | 2/10 | Very high fibre-to-sugar ratio among fruits. |
| Watermelon (small portion) | 4/10 | High glycaemic index but low glycaemic load at small portions. |
| Banana | 5/10 | Starchy fruit; ripe bananas convert starch to sugar. |
| Mango (large portion) | 6/10 | High fructose content, limit to a small portion. |
Carbohydrates and starches
| Food | WI Score | Why |
|---|---|---|
| Lentils (cooked) | 4/10 | Legume fibre slows glucose absorption significantly. |
| Chickpeas (cooked) | 4/10 | Resistant starch and fibre buffer the moderate carb load. |
| Oats (rolled, plain) | 5/10 | Better than instant but still a moderate insulin trigger. |
| White rice | 8/10 | Stripped of fibre, behaves like near-pure glucose. |
| White bread | 8/10 | Highly refined wheat breaks down to glucose almost instantly. |
| Pasta (large portion, plain) | 7/10 | Dense refined carbs, significant insulin response without protein. |
| Potatoes (boiled) | 6/10 | Moderate impact especially when cooled, but eaten alone, spikes insulin. |
Snacks and drinks
| Food | WI Score | Why |
|---|---|---|
| Sparkling water (plain) | 1/10 | No carbohydrate, insulin stays completely flat. |
| Black coffee or green tea | 1/10 | No sugar, no impact, may even modestly improve insulin sensitivity. |
| Dark chocolate (85%+, small piece) | 3/10 | High fat and fibre buffers the modest sugar at small portions. |
| Protein shake (whey, unflavoured or low sugar) | 3/10 | Protein-dominant; some whey triggers a modest insulin response. |
| Flavoured oat milk latte | 7/10 | Liquefied starch with added sugar, a disguised dessert. |
| Fruit juice (any) | 7/10 | Fibre removed, delivers concentrated sugar directly. |
| Sports drink | 9/10 | Liquid glucose and electrolytes, no fibre buffering at all. |
| Sweetened protein bar | 8/10 | High in syrups and sugar despite the protein claim on the label. |
| Soda and sweetened drinks | 9/10 | Pure added sugar, the fastest route to an insulin spike. |
The pattern is consistent with the published insulin-index data: protein sources, non-starchy vegetables, and whole fats cluster at 1-3. Fruit with intact fibre sits at 2-5. Refined starches and liquid sugars sit at 7-10. The takeaway is to build every meal around the 1-5 range, and to treat anything above 6 as an occasional choice rather than a staple.
Ready to stop guessing what to eat?
How much protein do I need to protect muscle on GLP-1?
Muscle loss is a documented side effect of rapid weight loss on GLP-1 medication. The SEMALEAN study (Alissou et al., Diabetes Obesity and Metabolism 2026) found that semaglutide produced a mean lean mass decline of approximately 3 kg at seven months, though lean mass stabilised thereafter and handgrip strength improved by 4.5 kg by 12 months. The muscle-loss risk is real in the early phase of treatment and is amplified by insufficient protein intake.
Target 1.6 to 2.0 grams of protein per kilogram of body weight per day. For a 75 kg person, that is 120 to 150 grams of protein daily. On GLP-1, where total food volume is lower, this requires deliberate choices: anchor every meal with a protein source from the 1-2 scoring group. Eggs, Greek yogurt, chicken breast, cottage cheese, salmon, and turkey breast are the highest-efficiency options because they deliver protein without raising insulin significantly.
If appetite suppression makes it difficult to reach protein targets through whole food, a plain whey or egg-white protein shake (unsweetened or very low sugar) scores approximately 3 on the WI Scale and is a practical tool. Avoid the sweetened protein bars that crowd the supplement aisle. Most score 8 due to the syrups and added sugars used to improve palatability.
What happens when I stop the medication?
The STEP 4 trial (Rubino et al., JAMA 2021) established the weight-regain risk directly. Participants who completed an initial 20 weeks on semaglutide and then switched to placebo regained roughly two-thirds of their lost weight within 48 weeks. Body weight increased by approximately 6.9% in the withdrawal group, compared to a continued 7.9% weight loss in those who stayed on the medication.
The reason is that GLP-1 medications work by reducing appetite and food intake, not by permanently altering metabolism or fat-storage pathways. When the drug is removed, appetite returns toward its previous level. If the food environment has not changed, intake increases and weight returns.
The practical implication is to treat the medication window as a training period. Use the appetite suppression to build habits around low-insulin-impact eating, longer overnight fasting windows (12-14 hours), and protein-anchored meals. These habits reduce the chronic insulin signal independently of medication. They do not replicate GLP-1's appetite effect, but they address the hormonal mechanism that drives fat storage. Talk to your prescriber before making any changes to your medication.
Do I need to count calories on GLP-1?
No. GLP-1 medications manage appetite with enough precision that intake reduction happens without manual tracking. Counting on top of that adds friction without adding useful signal. The relevant question is not how much you are eating but whether what you are eating is keeping insulin low.
Consider two meals at roughly the same volume: grilled salmon with roasted courgette and olive oil, versus a slice of white bread with fruit jam and a glass of orange juice. The first keeps insulin flat for hours. The second produces a rapid spike that pauses fat burning, triggers a glucose crash, and may paradoxically increase cravings despite the medication's appetite-suppressing effect. Same amount of food, opposite hormonal outcome.
What to track instead: score every meal by its insulin impact and target a daily WI average below 5. This takes one photo per meal with RealFoods and produces the signal that matters. Over time, the pattern tells you more than any calorie total, because it reflects what your body is doing with the food rather than simply how much you consumed.
Common questions
What should I eat on Ozempic?
Prioritise protein, non-starchy vegetables, and healthy fats. Foods scoring 1-3 on the WI Scale keep insulin low while the medication handles appetite. Eggs and smoked salmon for breakfast, a chicken and avocado salad for lunch, salmon with roasted broccoli for dinner. Avoid refined carbohydrates and sugary drinks even in small amounts, as they can still spike insulin and pause fat burning for hours.
How do I track food on semaglutide?
Track meal quality, not meal size. Score each meal by its insulin impact, target a daily WI average below 5. RealFoods generates that score from a photo of your plate. The key question for each meal is: does this food spike insulin or keep it low? That question is more useful than any calorie total when you are already eating less.
Do I need to count calories on GLP-1?
No. GLP-1 medication reduces appetite, so intake naturally decreases. Counting calories on top of pharmaceutical appetite suppression adds complexity without adding precision. Focus on the insulin quality of what you eat. A 400-calorie meal can either keep insulin flat for six hours or spike it for three, depending entirely on what you chose. That hormonal difference determines whether those hours become fat-burning time or fat-storage time.
What foods should I avoid on Ozempic?
Avoid anything scoring 7-10: sugary drinks, fruit juice, sweetened coffee, flavoured oat milk, white bread, white rice, commercial breakfast cereals, granola, protein bars with added syrups, and ultra-processed snack foods. These items trigger a full insulin spike even in small portions. On GLP-1, small portions are all you are eating, which means one high-scoring choice can dominate your entire hormonal picture for that meal window.
How do I keep the weight off after stopping?
Build the food habits while you are still on the medication. The STEP 4 trial found approximately two-thirds of lost weight returned within a year of stopping semaglutide. The people best positioned to maintain results are those who have practised low-insulin eating patterns long enough that they are habitual rather than effortful. Target a daily WI average below 5, protect lean mass with adequate protein, and extend the overnight fasting window. Then talk to your prescriber about your transition plan.
How RealFoods fits
RealFoods scores every meal you photograph on the 1-10 Weight Impact Scale, the same scale used in the food table above. It tracks your daily average, shows you which meals spiked insulin and which kept it low, and gives you a weekly pattern to work with. On GLP-1, where every meal is smaller and more deliberate, this kind of per-meal feedback is more useful than a calorie total.
The app also tracks your fasting window automatically. A 12-14 hour overnight fast, which most people on GLP-1 achieve naturally due to appetite suppression, is one of the most reliable ways to lower baseline insulin and allow fat oxidation to proceed. RealFoods detects the window, shows you where you are in the metabolic stages of fasting, and reinforces the behaviour without requiring manual logging.
Ready to stop guessing what to eat?
Continue reading: how to lower insulin naturally, what to eat with prediabetes, or the Weight Impact Score of 23 popular foods.
Sources
- Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). N Engl J Med. 2021;384(11):989-1002. PMID 33567185
- Rubino DM, Greenway FL, Khalid U, et al. Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance: The STEP 4 Randomized Clinical Trial. JAMA. 2021;325(14):1414-1425. PMID 33755728
- Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1). N Engl J Med. 2022;387(3):205-216. PMID 35658024
- Alissou M, et al. Impact of Semaglutide on fat mass, lean mass and muscle function in patients with obesity: The SEMALEAN study. Diabetes Obes Metab. 2026. PMID 41068996
- Holt SH, Miller JC, Petocz P. An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods. Am J Clin Nutr. 1997;66(5):1264-1276. PMID 9356547
