Examining differences between overweight women and men in 12-month weight loss study comparing healthy low-carbohydrate vs. low-fat diets

The study addresses how sex as a biological variable and gender-related behaviors influence weight-loss responses to diet interventions. Prior work and NIH guidance emphasize incorporating sex into research design and reporting. Biological differences (e.g., higher lean mass, energy expenditure, and greater visceral fat in men) and gender-related factors (e.g., women’s higher participation in weight-loss efforts and preference for low-fat foods) may affect outcomes and adherence. Most prior trials underrepresent men and rarely report sex-specific outcomes, limiting reproducibility and insight into differential responses. This secondary analysis of the DIETFITS trial aimed to compare 12-month changes in body weight, fat mass, and lean mass by diet (healthy low-carbohydrate vs. healthy low-fat) and sex (women vs. men), and to assess whether adherence differed by diet-sex group.

Prior diet trials often included predominantly women, had small sample sizes (<100) and/or short durations (<6 months), limiting power to detect sex-specific effects. Some reported greater absolute weight loss in women than men, but differences often disappeared after adjusting for baseline weight. Few studies examined sex differences in response to differing macronutrient compositions, or effects on fat/lean mass changes, and results have been mixed. A previous analysis found adherence comparable between low-fat and low-carb diets overall, but adherence was associated with greater weight loss only in the low-carb group. These gaps motivated a sex-stratified secondary analysis in DIETFITS to clarify whether diet type interacts with sex to influence weight and body composition changes and whether adherence differs by diet-sex group.

Design: Secondary analysis of the DIETFITS randomized, single-site, parallel-group trial (NCT01826591). Participants: 609 generally healthy overweight/obese adults (BMI 28–40 kg/m²), ages 18–50 years; randomized to healthy low-carbohydrate (HLC; n=304) or healthy low-fat (HLF; n=305). Sex-diet subgroups: HLC women n=179, HLC men n=125, HLF women n=167, HLF men n=138. Key exclusions: uncontrolled metabolic disease or hypertension; pregnancy/lactation; diabetes; cancer; cardiovascular, renal, or liver disease; medications affecting weight. Intervention: 12 months, 22 small-group educational sessions. Both diets shared three components: (1) Limbo phase (first 8 weeks): progressively reduce target macronutrient to ≤20 g/day (carbohydrate for HLC; fat for HLF). (2) Titrate phase: increase target macronutrient by 5–15 g/week to an individually sustainable maintenance level. (3) Emphasis on diet quality for both arms: home-cooked whole foods; maximize fresh, seasonal vegetables; lean, grass-fed meats; minimize added sugars, refined white flour, and trans-fats. Measurements: Weight at baseline, 3, 6, 12 months; height at baseline; BMI calculated. Body composition via DEXA at baseline (not available for first 78 participants) and at 6 and 12 months thereafter; after DEXA became available, 87% of remaining participants completed scans (n=276 women, n=190 men). Dietary intake and adherence: Intake assessed at baseline and 3, 6, 12 months by three unannounced 24-h multiple-pass recalls (2 weekdays, 1 weekend). Adherence quantified using a weight-adjusted standardized adherence (WASA) score derived from deviation from the Limbo-phase macronutrient target (20 g/day). Steps: average recalls per timepoint; compute deviation from 20 g target; divide by baseline weight to form deviation score (DS); normalize DS within diet and timepoint (Z-score); average Z-scores across 3, 6, 12 months to obtain WASA (higher=better adherence relative to overall average). Food choice questionnaires: Baseline items from a validated FCQ assessed importance of eating low-fat foods and tendency to avoid refined-carbohydrate foods. Statistical analysis: Hypotheses tested were (i) 12-month changes in weight, fat mass, and lean mass differ by diet-sex group; (ii) adherence differs by diet-sex group; (iii) adherence mediates differences in outcomes. Linear mixed effects models with fixed effects for diet-sex group, time, their interaction, baseline weight, and baseline percent fat; random intercept for participants; F-tests with Satterthwaite adjustment for overall differences and t-tests with Kenward-Roger adjustment for pairwise comparisons. Adherence modeled via linear regression as a function of diet-sex group with overall F-test and specified pairwise comparisons. Mediation evaluated by adding WASA to mixed models; relationships between WASA and percent changes assessed via Spearman correlations. Two-sided alpha=0.05; no multiple testing adjustments due to exploratory nature. Analyses conducted in R 3.6.1.

• Retention at 12 months was ~79% across all groups with no significant differences by diet-sex group. Baseline age was ~39–40 years; women weighed less than men but BMI was similar (~33 kg/m²). Men had higher metabolic syndrome-related measures at baseline. - Macronutrient intake differed as intended post-randomization: HLC groups consumed lower percent carbohydrate than HLF, and HLF groups consumed lower percent fat than HLC (P<0.001 at all postrandomization time points). - Primary outcomes (adjusted for baseline weight and body fat percentage): • 12-month weight loss differed by diet-sex group: F(3, 1221.14)=5.95, P<0.001. • 12-month lean mass loss differed by diet-sex group: F(3, 816.46)=9.21, P<0.001. - Among men, HLC led to greater losses than HLF: • Weight: -2.98 kg (95% CI -4.47, -1.50); P<0.001. • Fat mass: -1.51 kg (95% CI -2.79, -0.23); P=0.02. • Lean mass: -1.33 kg (95% CI -1.97, -0.68); P<0.001. - Among women, no significant differences between HLC and HLF for weight, fat mass, or lean mass changes. - Within HLC, men lost more than women: • Weight: -2.32 kg (95% CI -3.67, -0.97); P<0.001. • Lean mass: -1.42 kg (95% CI -2.01, -0.84); P<0.001. - Within HLF, changes in weight, fat mass, and lean mass did not differ significantly between women and men. - Adherence (WASA): Overall differences by diet-sex group were not statistically significant [F(3, 553)=2.06, P=0.10], but pairwise, HLC men were more adherent than HLC women (WASA difference 0.25 [0.04, 0.46]; P=0.02). - Associations between adherence and percent changes over 12 months: • Weight: HLC women rs=-0.36 (P<0.001), HLC men rs=-0.57 (P<0.001), HLF women rs=-0.32 (P<0.001), HLF men rs=-0.36 (P<0.001). • Fat mass: HLC women rs=-0.29 (P=0.002), HLC men rs=-0.60 (P<0.001), HLF women rs=-0.27 (P=0.01), HLF men rs=-0.22 (P=0.066). • Lean mass: HLC women rs=-0.23 (P=0.014), HLC men rs=-0.45 (P<0.001), HLF women rs=-0.20 (P=0.054), HLF men rs=-0.29 (P=0.016). - Adjusting outcome models for adherence did not materially change within-group weight loss estimates, suggesting adherence was not a mediator of the observed sex-by-diet effects.

Stratifying by sex revealed a significant interaction between diet type and sex that was not apparent in the original primary analysis combining women and men. Men experienced greater weight, fat, and lean mass losses on the healthy low-carbohydrate diet compared with the healthy low-fat diet, whereas women achieved similar changes regardless of diet assignment. Within the low-carbohydrate arm, men lost more weight and lean mass than women. Adherence patterns provide a potential behavioral explanation: men were more adherent to the low-carbohydrate regimen than women, and higher adherence correlated with greater reductions in weight and body composition outcomes across groups. Women’s stronger baseline preference for low-fat foods may have made adherence to the higher-fat content of the low-carbohydrate diet more difficult. However, statistical adjustment suggested adherence did not mediate the sex-by-diet differences, indicating other biological or behavioral factors may contribute. These results underscore the importance of considering sex in the design, analysis, and reporting of diet intervention trials to better understand heterogeneous responses and to inform personalized dietary strategies.

In this secondary analysis of DIETFITS, men assigned to a healthy low-carbohydrate diet lost significantly more total weight, fat mass, and lean mass over 12 months than men assigned to a healthy low-fat diet, while women had similar outcomes on both diets. Reporting outcomes by sex uncovered differences not evident in the original combined analysis and highlights the need to incorporate sex as a key variable in diet trial design, analysis, and reporting. Future research should prospectively test sex-specific responses to dietary macronutrient composition, examine mechanisms underlying adherence and biological differences, and explore tailoring interventions to individual preferences and physiological profiles.

This was an exploratory secondary analysis with no adjustments for multiple comparisons. Body composition (DEXA) was not measured for the first 78 participants and, thereafter, was available for 87% of remaining enrollees, introducing potential selection and missing data concerns; analyses assumed data were missing at random. The study was conducted at a single site with generally healthy, nondiabetic overweight/obese adults aged 18–50 years, which may limit generalizability. Dietary intake and adherence were derived from 24-hour recalls and computed adherence metrics, and retention at 12 months was ~79%, which may affect precision and interpretation.