What actually works for weight loss without drugs, and when are medications reasonable?

Why the scale fights back

Weight loss needs an energy deficit, but your body resists it. In 50 adults who lost 13.5 kg on a 10-week very-low-energy diet, leptin, peptide YY and insulin stayed lower, ghrelin stayed higher and hunger stayed greater a year later⁷. Regain reflects biology, not weak will.

Which diet: the one you keep

A network meta-analysis of 48 randomized trials (7,286 adults) found that low-carbohydrate and low-fat diets each beat no diet by about 7 kg at 12 months, with minimal differences between named diets¹. In DIETFITS, 609 adults lost 5.3 kg on a healthy low-fat diet and 6.0 kg on a healthy low-carbohydrate diet over 12 months².

Levers that help

• Food quality. In a 4-week inpatient trial, 20 adults ate about 508 kcal a day more on an ultra-processed diet than on an unprocessed one, gaining 0.9 kg versus losing 0.9 kg, even though meals were matched for presented calories, sugar, sodium and fiber⁴. See ultra-processed food.
• Protein and fiber. Diets with 1.2 to 1.6 g of protein per kg a day, at least 25 to 30 g per meal, modestly improved fullness and weight, mainly in people who stuck to them⁵.
• Sleep. Overweight adults who slept about 1.2 hours more ate 270 kcal a day less⁶. See why sleep matters.
• Meal timing. Eating only between 8 am and 4 pm on top of calorie restriction gave no significant extra loss over calorie restriction alone¹⁰.
• Support. In adults with type 2 diabetes, intensive lifestyle support kept a 4.7 percent average weight loss at 8 years, against 2.1 percent with diabetes education alone¹¹.
• Movement. 150 to 250 minutes a week prevents gain but gives only modest loss; more than 250 minutes is linked to clinically significant loss and better maintenance¹². See walking.

Where medication fits

Semaglutide 2.4 mg weekly plus lifestyle advice cut weight 14.9 percent versus 2.4 percent with placebo over 68 weeks in 1,961 adults without diabetes who had a BMI of 30 or more (or 27 or more with a related condition)⁸. A year after stopping, two-thirds of the loss had returned⁹. For food-based ways to support your own GLP-1, see natural ways to boost GLP-1.

Diet type

Johnston and colleagues pooled 59 articles on 48 trials. Compared with no diet, low-carbohydrate diets produced 8.73 kg of loss at 6 months and 7.25 kg at 12 months, and low-fat diets 7.99 and 7.27 kg. Differences between named diets were minimal (Atkins beat Zone by 1.71 kg at 6 months). Behavioral support added 3.23 kg at 6 months but 1.08 kg at 12 months (95% interval -1.82 to 3.96), while exercise added 0.64 kg at 6 months and 2.13 kg at 12 months¹. DIETFITS randomized 609 adults aged 18 to 50 (BMI 28 to 40, no diabetes) to 22 small-group sessions over 12 months. Of them, 79 percent finished; the between-diet difference was 0.7 kg (95% CI -0.2 to 1.6), and neither a genotype pattern nor insulin secretion predicted who did better on which diet².

Food, sleep and timing

Hall's trial put 20 weight-stable adults at the NIH on ultra-processed and unprocessed diets, 2 weeks each in random order. Intake was 508 kcal a day higher on the ultra-processed diet (carbohydrate and fat rose; protein did not), and weight changes tracked intake (r=0.8)⁴. In Tasali's trial of 80 adults aged 21 to 40 with overweight and habitual sleep under 6.5 hours, counseling that extended sleep by about 1.2 hours cut energy intake by 270 kcal a day (95% CI 147 to 393) with no change in expenditure, so weight fell⁶. Liu's trial randomized 139 adults with obesity to calorie restriction with or without 8 am to 4 pm eating. At 12 months weight fell 8.0 kg versus 6.3 kg (net difference -1.8 kg, 95% CI -4.0 to 0.4; P=0.11)¹⁰.

Why maintenance is hard

Sumithran's 50 participants lost 13.5 kg in 10 weeks. At 62 weeks leptin, peptide YY, cholecystokinin, insulin, ghrelin and other mediators still differed significantly from baseline, and hunger was higher⁷. In Look AHEAD, 5,145 adults with type 2 diabetes and overweight were randomized to intensive lifestyle intervention or education only. At year 8 they had lost 4.7 and 2.1 percent of body weight; 50.3 versus 35.7 percent kept off at least 5 percent, and 26.9 versus 17.2 percent at least 10 percent¹¹. The US Preventive Services Task Force found moderate benefit from intensive, multicomponent programs and from maintenance programs, with small or no harms³. The American College of Sports Medicine notes that no well-designed randomized trial shows exercise prevents regain, though observational data link more than 250 minutes a week to better maintenance¹².

Medication as the honest comparator

STEP 1 randomized 1,961 adults to semaglutide 2.4 mg or placebo, both with lifestyle advice, for 68 weeks. Weight fell 14.9 versus 2.4 percent (difference 12.4 points, 95% CI 11.5 to 13.4); 86.4 versus 31.5 percent lost at least 5 percent, and 50.5 versus 4.9 percent at least 15 percent. Nausea and diarrhea were most common, and 4.5 percent versus 0.8 percent stopped for stomach side effects. The company funded the trial⁸. In the off-treatment extension of 327 participants, those on semaglutide had lost 17.3 percent by week 68 and regained 11.6 points by week 120, leaving a net 5.6 percent loss; cardiometabolic gains drifted back toward baseline⁹.

Conflicting and null results

Time-restricted eating did not beat plain calorie restriction¹⁰. Higher-protein diets help in tightly controlled short studies, but long-term studies are limited and conflicting, with benefit mainly in adherers⁵. Behavioral support's edge faded by 12 months in the meta-analysis¹. Sleep extension was tested in adults with overweight, not obesity⁶.

Bottom line

Confidence is fairly high (4 of 5). Diet type matters little^1,2, structured support, food quality and sleep help^3,4,6, and regain is biologically driven⁷. Medication is more potent but works only while taken^8,9; the choice belongs with a clinician. For weight loss with type 2 diabetes, see type 2 diabetes and remission.