Effect of Chitosan Supplementation and Lifestyle Changes on Health and Gut Microbiota in Icelandic Females With Obesity Authors: Thelma Rún Rúnarsdóttir, Sigurlaug Skírnisdóttir, Hildur Thors, Hjördís Harðardóttir, Karla F. Corral‐Jara, Alexandra M. Klonowski, Þórhallur Ingi Halldórsson, Marta Guðjónsdóttir Published: 2026 ABSTRACT ABSTRACTEvaluate the effect of chitosan supplementation with structured lifestyle changes on health factors and gut microbiota in females with obesity following an interdisciplinary rehabilitation program. In this randomized, double‐blind, placebo‐controlled trial (NCT04551365), 33 Icelandic females with obesity (BMI 30.0–54.1 kg/m2) received 3 g/day of LipoSan Ultra chitosan (TRG, n = 18) or placebo (PLG, n = 15) for 12 weeks, alongside structured lifestyle changes. Clinical, anthropometric (DXA), and functional data were collected at baseline (T1) and post‐intervention (T3). Gut microbiota (16S rRNA) data were collected at T1, T2 (2 weeks), T3, and T4 (2 weeks post‐intervention). Structured lifestyle interventions increased lean mass, grip strength, endurance, and self‐rated health, while reducing body fat mass and mean arterial pressure across BMI groups. Participants with BMI ≥ 40 kg/m2 showed decreased microbiota diversity over time. Chitosan supplementation was linked to a reduction in the Firmicutes/Bacteroidota ratio, although no differentially abundant taxa were identified (ANCOM‐BC2). Structured lifestyle interventions improved body composition and physical function, independently of weight loss. Chitosan supplementation modulated gut microbiota composition. Interpretation of subgroup and gut microbiota findings is constrained by the study size. Future studies should investigate long‐term effects and stratify outcomes by BMI to better evaluate chitosan's therapeutic potential. Introduction Lifestyle‐related diseases such as obesity, type 2 diabetes, and cardiovascular disease have increased significantly in recent decades [1, 2]. Since the mid‐1970s, global obesity rates have nearly tripled [3]. By 2022, a substantial proportion of the adult population worldwide, including Iceland, was classified by body mass index as having overweight or obesity [4]. Obesity has reached epidemic proportions and was associated with millions of deaths globally in 2019, according to the Global Burden of Disease [5]. Given the global rise in obesity and its associated health complications, there is an increasing demand for effective and sustainable weight management strategies. Among various approaches, dietary supplements have gained significant attention for their potential role in supporting weight loss [6]. Chitosan, an amino polysaccharide, has been marketed as a fat‐binding dietary fiber with claimed weight‐managing properties, although the European Food Safety Authority (EFSA) has found no established cause‐and‐effect relationship to support these claims [7]. Several randomized controlled trials have reported that chitosan supplementation may contribute to small to moderate weight loss in individuals with overweight or obesity [8, 9, 10]. However, other studies have found no significant effect on body weight [11, 12]. The inconsistency in findings may be attributed to variations in dosage, duration of intervention, and differences in chitosan formulations, which limit the reliability of conclusions regarding its efficacy. The evidence regarding chitosan's prebiotic effects on the human gut microbiota remains scarce, with only two clinical studies conducted to date. Consequently, understanding of its potential secondary impacts on health and obesity‐related comorbidities mediated through gut microbiota modulation is still limited. The first study, a randomized, double‐blind clinical trial conducted in Iran, investigated the effects of 12‐week chitosan supplementation in 64 adolescents with overweight or obesity [13]. In the group receiving chitosan supplementation, the intervention led to a significant reduction in the body mass index (BMI) z‐score. Correspondingly, RT‐PCR analysis of selected bacterial taxa revealed a decreased relative abundance of Firmicutes and increased levels of Bacteroidota and Akkermansia, while Bifidobacteria levels remained unchanged. No comparable changes were observed in the control group [13]. The second study was our recently published analysis from a randomized, double‐blind clinical trial conducted in Iceland, which included 60 female participants with BMI values ranging from 19 to 35 kg/m2 who did not undergo any lifestyle interventions. In the chitosan‐supplemented group, a reduction in gut microbiota diversity and evenness was observed. This shift in gut microbiota composition was characterized by an increased relative abundance of Actinobacteriota, Ruminococcus gnavus, Clostridium innocuum, Bifidobacterium, and Holdemania, along with decreased levels of Blautia, Eubacterium hallii, and CAG‐56. No significant differences were observed in the Firmicutes/Bacteroidota ratio (F/B ratio), body weight, or body composition [12]. In this analysis, the primary aim was to evaluate the effects of a 12‐week chitosan supplementation, combined with lifestyle changes, on various health factors and gut microbiota composition in Icelandic females with obesity, compared to lifestyle changes alone. A secondary objective was to determine whether the effects varied according to the degree of obesity. To our knowledge, this is the first randomized controlled trial to investigate the synergistic impact of chitosan and lifestyle intervention on gut microbiota in this population, providing novel insights into the potential mechanisms linking dietary supplementation, health factors, and gut microbiota composition. Methods Study Design and Participants This interventional study (https://www.clinicaltrials.gov NCT04551365) was a randomized, double‐blind, placebo‐controlled, clinical trial. Thirty‐six participants who were enrolled in an inpatient interdisciplinary rehabilitation program at Reykjalundur Rehabilitation Center in Iceland were randomly assigned in a 1:1 ratio to either the chitosan treatment group (TRG, n = 18) or the placebo group (PLG, n = 15). Eligible participants provided written informed consent at the baseline visit (T1) in addition to taking a pregnancy test. At the T1 visit, the subjects were randomly assigned to either the TRG or PLG, with randomization generated using an online random integer set generator [14]. Clinical measurements were collected at T1 and at the end of the intervention period (T3), corresponding to the time point at which participants discontinued supplementation. Blood and fecal samples were collected at week 2 (T2), and fecal samples were additionally collected 2 weeks after cessation of supplementation (T4). The timeline setup of the study is shown in Figure 1. The study was approved by the Icelandic National Bioethics Committee (VSNb2017090016/03.01). Flowchart of participant recruitment, randomization, intervention allocation, and analysis. Data were collected at four time points (T1–T4), including assessments of vital signs, blood markers, anthropometry, fecal samples, physical fitness, safety parameters, and questionnaires. Inclusion criteria were females, 18–80 years old, with BMI ranging from 30 to 55 kg/m2, who were ready to make healthy changes to their lifestyle in addition to halting any intake of probiotics during the trial period. Exclusion criteria included the following: pregnancy, planning pregnancy or breastfeeding during the intervention period; eating disorders (i.e., anorexia, bulimia); inflammatory bowel disease (i.e., Crohn's disease, ulcerative colitis); poorly controlled type 2 diabetes or hypertension; surgical procedures (abdominal and chest) or bariatric operation in the past 6 months; history of invasive fat reduction procedure (e.g., liposuction or abdominoplasty) within the past one year; intake of warfarin; shellfish allergy; subjects taking or having taken weight management supplements in the past 30 days and use of antibiotics in the last 3 months. If treated with semaglutide or liraglutide drugs (glucagon‐like peptide 1 receptor agonist), the treatment must have been stable for at least 3 months. Intervention Participants were enrolled in an inpatient interdisciplinary rehabilitation program involving structured intervention for 8 h daily during the first 5 weeks. The intervention was individually tailored by a team of health professionals to support the adaptation of healthy lifestyle behaviors. It included regular physical activity (at least twice daily exercise sessions), access to balanced and nutritious meal options for breakfast, lunch, and afternoon snacks, and personalized guidance on goal setting, stress reduction, sleep improvement, mental health, and behavior change techniques. Upon completion of the 5‐week program, participants were instructed to maintain these lifestyle modifications throughout the remaining 7 weeks of the study. Participants received either the investigational product, LipoSan Ultra, a proprietary chitosan formulation previously described in detail by Rúnarsdóttir et al. [12], or a placebo. The capsule composition, dosage (375 mg per capsule; 3 g/day total), and placebo formulation (187.5 mg microcrystalline cellulose and 187.5 mg native maize starch per capsule) were identical to those used in the prior analysis. Placebo and chitosan capsules were indistinguishable in appearance and taste. The capsules were dispensed at T1, and adherence was assessed at T3 by counting the number of remaining capsules returned by participants (Figure 1). Outcome Measures The primary outcome of this study was to evaluate the effects of chitosan supplementation (TRG) in combination with lifestyle changes, on various health factors and gut microbiota composition in Icelandic females with obesity, compared to lifestyle changes alone (PLG). In addition to this primary group‐based analysis, a secondary analysis was conducted in which participants were stratified according to baseline BMI (<40 vs. ≥40 kg/m2) to assess whether the response to the intervention differed by the degree of obesity. This stratification was exploratory in nature and was not part of the randomization scheme. Anthropometric Measurements Anthropometric data, vital signs, health and endurance, and strength factors were measured at T1 and again at T3. Anthropometric factors were estimated as follows: body composition (fat mass and fat‐free mass) was assessed by using dual‐energy x‐ray absorptiometry (DXA; Hologic Horizon A, Hologic Inc., USA), BMI was determined using whole body weight (kg) divided by squared height (m2) [15], waist circumference was measured following the WHO STEPS method [16] and the waist‐to‐height ratio (WtHR) was calculated by dividing the waist circumference by height [17]. Fecal Sampling and 16S rRNA Sequencing Fecal samples were self‐collected using the OMNIgene GUT kit (DNA Genotek, Canada) and stored at −80°C. DNA extraction, amplification of the 16S rRNA gene (V4 region), inclusion of controls, and sequencing on an Illumina MiSeq platform (including multiplexing and normalization) were performed following the protocol previously described by Rúnarsdóttir et al. [12]. Further methodological details on bioinformatic analysis are provided in the Supporting Methods section. Physical Function Assessment Grip strength was measured for each hand using a Saehan SH5001 hand dynamometer (Saehan Corporation, South Korea), and the mean value was calculated from three consecutive measurements [18, 19]. Endurance was evaluated using a 6‐min walk test in a 70 m‐long corridor, measuring the walking distance completed. Two tests were conducted at T1, with at least 1 hour in between tests, where the better test was used for analysis. At T3, only one test was conducted [20]. The predicted walking distance value was calculated using Gibbons’ formula [21]. Step count, activity intensity, sedentary time, and sleep duration were measured using an activity tracker (Fitbit Charge 3) that participants wore on their non‐dominant hand for the whole duration of the study. Physical activity for Period 1 was calculated as the average activity from weeks 3–5, during which participants were engaged in active inpatient rehabilitation. For Period 2, physical activity was calculated from weeks 10–12, when participants were maintaining lifestyle changes at home. Blood pressure at rest was measured at T1 and T3 using a manual arm cuff and stethoscope, with two measurements taken 2 min apart. Mean arterial pressure (MAP) was calculated using the standard clinical approximation formula [22]. Questionnaires Self‐rated health status was evaluated by the participants using the 0–100 visual analog scale (EQ VAS) from the EuroQol‐5 Dimensions 5 Levels questionnaire, with scale endpoints labeled as “The best health you can imagine” and “The worst health you can imagine” [23, 24]. Depression severity and general anxiety symptoms were assessed using two self‐reported screening questionnaires: the Patient Health Questionnaire (PHQ‐9), a nine‐item instrument used to screen for and evaluate the severity of depression [25], and the Generalized Anxiety Disorder‐7 (GAD‐7), a seven‐item scale which was used to assess anxiety symptoms [26]. The scoring for these two questionnaires is the same, where 0–4 represents non or minimal, 5–9 mild, 10–14 moderate, 15–19 moderately severe, and 20–27 severe depression/anxiety. Defecation was evaluated using the Bristol Stool Form Scale (permitted use granted by the Rome Foundation, USA), where fecal matter is categorized into constipation (types 1 and 2), ideal stools (types 3 and 4), trending toward diarrhea (type 5), and diarrhea (types 6 and 7) [27]. Physical activity levels were assessed using the Saltin–Grimby Physical Activity Level Scale (SGPALS), a self‐reported questionnaire categorizing habitual physical activity during leisure time into four graded levels, ranging from sedentary behavior to vigorous physical activity [28]. Biochemical Measurements At T1 and T3, fasting blood measurements included hemoglobin, high sensitivity C‐reactive protein (hs‐CRP), lipid profile (total cholesterol (TC), low‐density lipoprotein cholesterol (LDL), high‐density lipoprotein cholesterol (HDL), triglycerides (TG)), as well as blood glucose metabolism (HbA1c, insulin, fasting glucose) and markers of liver function (alanine transaminase (ALT) and alkaline phosphatase (ALP)). A second blood test was taken in week2 to evaluate the safety of the supplementation (T2), during which blood lipids and blood glucose metabolism were measured. Statistical Analysis Differences between groups (TRG and PLG; BMI <40 vs. ≥40 kg/m2), the effects of the intervention over time within groups (T1 vs. T3), and group‐by‐time interactions for continuous, normally distributed variables were evaluated using a mixed‐effects analysis of variance (mixed ANOVA). The results were presented as mean ± standard deviation (SD). Group differences were assessed using the paired sample t‐test, but when an indication of skewness in the variable distribution was observed, a Wilcoxon‐rank test was used. Primary analyses were adjusted for baseline age. Additional sensitivity analyses were conducted adjusting for baseline physical activity (SGPALS) and LDL cholesterol as covariates. These adjustments did not materially affect the results (Table 1S). In R, gut microbiota was evaluated using the stat_compare_means() function with the method set to “wilcox.test” from the ggpubr package (version 0.6.0) [29] to assess differences in alpha diversity between groups using the Wilcoxon test [30]. To evaluate differences in beta diversity, a PERMANOVA test was conducted with the number of permutations = 999 and a Bray–Curtis distance matrix, using the “adonis2()” function from the vegan package (version 2.6.10) [31]. P ≤ 0.05 was considered statistically significant in all analyses. Results Of the 51 females screened for eligibility, 36 met the inclusion criteria and were subsequently randomized into either the TRG or PLG. During the intervention period, three participants withdrew (8.3% drop‐out rate, two in the TRG and one in the PLG) for reasons detailed in Figure 1. Ultimately, 18 participants in the TRG and 15 in the PLG completed the 12‐week intervention. Intake adherence was 92% in the TRG and 87% in the PLG (p = 0.053). None of the participants reported the use of probiotics or weight management supplements during the study period. However, three participants, all in the TRG, were on stable treatment of glucagon‐like peptide‐1 (GLP‐1) receptor agonists throughout the study period. Use of other medications at T1 is reported in Table S2. Baseline Characteristics Baseline clinical characteristics (T1) for all participants (without grouping) are presented in Table S3. The age range of the study population was 31.8–68 y, with a BMI range from 30.0–54.1 kg/m2. At T1, no statistical differences in taxonomic composition were observed between the two intervention groups, TRG and PLG. Similarly, no differences were detected for the four alpha diversity indices (Observed, Chao1, Shannon and Simpson) between groups (Figure S1). Furthermore, both groups showed a F/B ratio dominated by Firmicutes (58%–60%). At T1, the TRG exhibited moderate depressive symptoms according to the PHQ‐9, while the PLG demonstrated mild symptoms. Both groups presented with mild anxiety levels on the GAD‐7 scale. Furthermore, both groups rated their own health similarly on the EQ‐5D 5L VAS with a score of 52.9 (Table 1). TABLE 1Effects of the intervention on anthropometric parameters, gut microbial indices, and physical and mental parameters.VariablesReference rangeTRG (n = 18)PLG (n = 15) p valueT1T3T1T3Between groupWithin groupInteractionAge (years)48.6 (10.7)54.6 (9.83) 0.02 Anthropometric parametersHeight (cm)166 (6.24)164 (8.81)1.00Weight (kg)111 (16.5)111 (16.5)105 (16.2)104 (17.0)0.770.500.72BMI (kg/m2)40.3 (4.9)40.4 (4.8)39.3 (7.1)39.4 (7.2)0.690.510.73BFM (%)48.5 (3.6)48.2 (3.4)50.1 (4.6)49.4 (4.4)0.220.120.89LBM (%)49.3 (3.5)49.7 (3.3)47.6 (4.5)48.3 (4.2)0.190.150.92WtHR≤ 0.50.72 (0.1)0.72 (0.1)0.69 (0.1)0.68 (0.1)0.640.350.67WC (cm)119 (12.1)119 (12.4)112 (9.9)111 (11.2)0.090.350.64HC (cm)131 (11.1)129 (11.3)127 (12.5)128 (13.0)0.590.300.09Gut microbial indicesObserved174.0 (38,2)155.0 (47.2)185.0 (44.0)167.0 (43.2)0.080.440.91Shannon4.3 (0.3)3.9 (0.7)4.3 (0.3)4.2 (0.4) 0.04 0.120.27Simpson1.0 (0.02)0.9 (0.1)1.0 (0.01)1.0 (0.02)0.200.210.27Physical and mental parametersMAP (mmHg)≤ 99.095.5 (8.6)94.0 (9.3)100 (5.4)95.3 (7.7)0.30 0.02 0.14Heart rate at rest (bpm)60 ‐ 10081 (13)80 (12)77 (12)78 (9)0.380.580.97EQ VAS (score)52.9 (16.1)61.4 (13.5)52.9 (16.8)64.3 (17.2)0.74 0.01 0.67PHQ‐9 (score)< 510.2 (4.5)8.1 (4.5)8.1 (4.6)7.2 (6.4)0.390.130.51GAD‐7 (score)< 57.1 (5.1)6.3 (4.0)5.5 (3.6)4.8 (5.12)0.330.370.856MWT (m)526 (96.6)549 (67.9)537 (68.3)549 (62.7)0.74 0.02 0.816MWT (% pred)80.8 (13.0)83.9 (9.2)85.4 (12.7)87.3 (12.0)0.31 0.02 0.82Grip strength D (pounds)38‐87 (65.8)69.4 (11.8)72.9 (11.5)61.9 (13.8)63.9 (13.9)0.06 0.04 0.53Grip strength ND (pounds)35‐76 (57.3)64.4 (11.0)67.3 (11.3)57.8 (11.9)59.8 (11.0)0.07 0.03 0.70Data shown as mean (standard deviation), p < 0.05 was considered statistically significant, as shown in bold. Abbreviations: TRG, chitosan group; PLG, placebo group; T1, baseline; T3, week 12; BMI, body mass index; BFM, body fat mass; LBM, lean body mass; WtHR, waist to height ratio; WC, waist circumference; HC, hip circumference; MAP, mean arterial pressure; EQ VAS, visual analog scale; PHQ‐9, depression screener patient health questionnaire; GAD‐7, generalized anxiety disorder screener; 6MWT, 6‐minute walk test; D, dominant; ND, non‐dominant. P calculated using mixed ANOVA. Model adjusted for age.John Wiley & Sons, Ltd. Effects of the intervention on anthropometric parameters, gut microbial indices, and physical and mental parameters. Data shown as mean (standard deviation), p < 0.05 was considered statistically significant, as shown in bold. Abbreviations: TRG, chitosan group; PLG, placebo group; T1, baseline; T3, week 12; BMI, body mass index; BFM, body fat mass; LBM, lean body mass; WtHR, waist to height ratio; WC, waist circumference; HC, hip circumference; MAP, mean arterial pressure; EQ VAS, visual analog scale; PHQ‐9, depression screener patient health questionnaire; GAD‐7, generalized anxiety disorder screener; 6MWT, 6‐minute walk test; D, dominant; ND, non‐dominant. P calculated using mixed ANOVA. Model adjusted for age. Despite randomization, some baseline differences were observed in age between the groups, and thus adjustments were made (Tables 1 and 3). Additionally, a greater proportion of participants in the PLG reported lower physical activity levels according to SGPALS (38% vs. 6% in the TRG, Table 2). Biochemical analysis was performed, and the results are presented in the Supporting Information (Tables S4 and S5). TABLE 2Changes in self‐reported physical activity from T1 to T3 for TRG and PLG using the Saltin–Grimby physical activity level scale (SGPALS).TRG T3T1InactiveSome light activityRegular moderate activityTotalInactive0 (0.0)0 (0.0)1 (11.1)1Some light activity0 (0.0)6 (75.0)3 (33.3)9Regular moderate activity0 (0.0)2 (25.0)5 (55.6)7Total0 (100.0)8 (100.0)9 (100.0)17 PLG T3T1InactiveSome light activityRegular moderate activityTotalInactive3 (100)2 (50.0)0 (0.0)5Some light activity0 (0.0)2 (50.0)4 (66.7)6Regular moderate activity0 (0.0)0 (0.0)2 (33.3)2Total3 (100.0)4 (100.0)6 (100.0)13Data shown as count (percentage in brackets).Inactive: Sedentary.Some light activity: Walking, cycling, or other light activities for at least 4 h/week.Regular moderate activity: More structured activity for at least 2–3 h/week.Sample size TRG n = 17, PLG n = 13 due to missing responses for this variable.John Wiley & Sons, Ltd. Changes in self‐reported physical activity from T1 to T3 for TRG and PLG using the Saltin–Grimby physical activity level scale (SGPALS). Data shown as count (percentage in brackets). Inactive: Sedentary. Some light activity: Walking, cycling, or other light activities for at least 4 h/week. Regular moderate activity: More structured activity for at least 2–3 h/week. Sample size TRG n = 17, PLG n = 13 due to missing responses for this variable. Effects of Chitosan Supplementation Anthropometrics Anthropometric, physical, and mental parameters, as well as gut microbiota indices at T1 and T3 for both groups, are presented in Table 1. No significant differences were observed for body weight or body composition. During the 12‐week intervention, no changes were seen in other biochemical analyses. Gut Microbiota A mixed ANOVA revealed no significant time x group interaction for any alpha diversity indices, indicating that temporal changes did not differ between groups. Observed richness showed a trend toward reduction over time (p = 0.08), while Shannon diversity demonstrated a significant effect of time (p = 0.036), reflecting an overall decrease in diversity from T1 to T3 across both groups. Principal coordinates analysis (PCoA) based on the Bray–Curtis distance matrix was used to visualize beta‐diversity patterns among samples. PERMANOVA revealed significant separation between T1 and T2, as well as between T2 and T4 in TRG (Figure 2A). In contrast, no distinct separation was observed in the PLG (Figure 2B). Comparisons between other time points were not significant. Principal Coordinates Analysis (PCoA) based on the Bray–Curtis distance matrix showing the gut microbiota community distance between T1, T2, T3, and T4 in the TRG (A, n = 18) and PLG (B, n = 15). The statistical difference was determined using the PERMANOVA test. Trace values of TRG and PLG were 21.47 and 14.58, respectively. Based on rarefied data, measurements were on the amplicon sequence variant (ASV) level. P values equal to or less than 0.05 were considered statistically significant. Each point represents one biological replicate. Chitosan supplementation induced a significant transient reduction in the F/B ratio between T1 and T2, characterized by increased Bacteroidota and decreased Firmicutes. Bacteroidota levels returned to baseline values by T3, while Firmicutes recovered to baseline levels by T4, and the F/B ratio was no longer significant (data not shown). An ANCOM‐BC2 differential abundance analysis between T1 and T3 showed no significant changes at any taxonomic level for either group. According to the Bristol Stool Chart (Table S6), at T1, 44% of participants in the TRG reported ideal stool consistency, while the remaining reported abnormal stool patterns, including constipation or diarrhea. In the PLG, 79% reported ideal stool consistency at T1. By T3, the proportion of participants with ideal stool consistency increased to 66% in the TRG and 86% in the PLG. Chitosan supplementation did not change stool consistency differently over time compared to placebo (p = 0.774). Physical and Mental Variables Grip strength in both hands and endurance (6MWT) improved in both groups, likely attributable to lifestyle changes (Table 1). Participants in both groups were able to maintain their physical activity levels between Period 1 and Period 2, suggesting sustained activity during the maintenance phase following completion of the rehabilitation program (Table S7). Furthermore, self‐reported physical activity indicated that participants perceived a shift toward a more physically active lifestyle at T3 compared to T1 (Table 2). Subsequently, self‐rated health (EQ VAS) improved during the intervention period. Effects of Lifestyle Changes Depending on BMI To further explore potential differential responses to lifestyle intervention, a subgroup analysis was conducted by stratifying participants into two BMI categories: <40 and ≥40 kg/m2 (Figure 3). This stratification aimed to evaluate whether the effects of the lifestyle intervention differed based on baseline BMI, particularly given the absence of clear effects of the chitosan supplementation. Flowchart of screening, enrolment, and group assignment of the study participants depending on BMI (< 40 and ≥ 40 kg/m2) used for secondary analyses. Anthropometrics As expected, significant differences in anthropometric variables were observed between the BMI subgroups. Participants in the <40 kg/m2 group had lower values across most parameters, except for lean body mass percentage, which was significantly higher (Table 3). At T3, both groups showed a reduction in body fat percentage and an increase in lean body mass percentage, despite no significant changes in total body weight (Table 3). TABLE 3Effects of lifestyle intervention on anthropometric parameters, gut microbial indices, and physical and mental parameters between those with BMI < or ≥ 40 kg/m2.VariablesReference range<40 kg/m2 (n = 19)≥40 kg/m2 (n = 14) p valueT1T3T1T3Between groupWithin groupInteractionNumber of TRG/PLG (n)10/98/6Age (years)56.1 (8.7)44.8 (9.6) 0.001 Anthropometric variablesHeight (cm)166 (0.07)163 (7.9)0.21Weight (kg)99.4 (14.1)99.6 (14.8)120 (9.8)120 (9.8) 0.04 0.900.45BMI (kg/m2)35.6 (3.1)35.8 (3.4)45.5 (3.6)45.5 (3.6) <.001 0.910.47BFM (%)47.4 (3.9)47.1 (3.9)51.4 (3.4)50.4 (3.1) 0.006 0.02 0.18LBM (%)50.3 (3.8)50.5 (3.8)46.8 (3.4)47.7 (3.1) 0.01 0.03 0.11WtHR≤ 0.50.66 (0.05)0.65 (0.06)0.80 (0.04)0.80 (0.05) <.001 0.290.51WC (cm)110 (8.91)107 (8.67)125 (8.18)125 (8.30) <.001 0.280.51HC (cm)122 (9.34)121 (7.81)138 (7.56)138 (9.23) <.001 0.210.74Microbial indicesObserved182.0 (42.6)183.0 (49.6)174.0 (38.9)133.0 (25.3) 0.05 0.01 0.03 a Shannon4.3 (0.3)4.2 (0.7)4.3 (0.3)3.7 (0.7) 0.02 0.06 0.04 a Simpson1.0 (0.01)1.0 (0.01)1.0 (0.02)0.9 (0.1)0.190.210.20Physical and mental variablesMAP (mmHg)≤ 99.095. (6.66)93.2 (6.35)99.8 (8.62)96.4 (10.6)0.10 0.03 0.98Heart rate at rest (bpm)60–10071.7 (13.4)75.7 (8.15)75.4 (6.76)82.7 (12.2)0.46 0.002 0.526MWT (m)537 (105.0)567 (76.6)524 (42.9)529 (39.6) 0.002 0.01 0.02 b 6MWT (% pred)85.5 (15.9)90.5 (11.7)79.4 (5.68)80.1 (5.85) 0.003 0.01 0.02 b Grip strength D (pounds)38–87 (65.8)62.7 (10.5)66.5 (10.2)60.3 (13.7)63.3 (15.4) 0.01 0.02 0.56Grip strength ND (pounds)35–76 (57.3)60.0 (10.7)63.0 (10.1)55.9 (14.1)55.4(19.0) 0.01 0.510.19EQ VAS (score)50.5 (13.1)63.3 (14.3)55.7 (19.4)61.8 (16.0)0.84 0.01 0.43PHQ‐9 (score)< 58.65 (5.00)6.53 (5.00)9.93 (4.14)9.21 (5.42)0.740.120.72GAD‐7 (score)< 56.00 (4.63)5.06 (4.12)6.86 (4.42)6.29 (5.09)0.580.360.85Data shown as mean (standard deviation), p < 0.05 was considered statistically significant, as shown in bold. Abbreviations: TRG, chitosan group; PLG, placebo group; T1, baseline; T3, week 12; BMI, body mass index; BFM, body fat mass; LBM, lean body mass; WtHR, waist to height ratio; WC, waist circumference; HC, hip circumference; MAP, mean arterial pressure; EQ VAS, visual analog scale; PHQ‐9, depression screener patient health questionnaire; GAD‐7, generalized anxiety disorder screener; 6MWT, 6‐min walk test; D, dominant; ND, non‐dominant. P calculated using mixed ANOVA. Model adjusted for age and supplementation use. a Change during the intervention was significant in the ≥ 40 BMI group but not in the < 40 BMI group. b Change during the intervention was significant in the < 40 BMI group but not in the ≥ 40 BMI group.John Wiley & Sons, Ltd. Effects of lifestyle intervention on anthropometric parameters, gut microbial indices, and physical and mental parameters between those with BMI < or ≥ 40 kg/m2. Data shown as mean (standard deviation), p < 0.05 was considered statistically significant, as shown in bold. Abbreviations: TRG, chitosan group; PLG, placebo group; T1, baseline; T3, week 12; BMI, body mass index; BFM, body fat mass; LBM, lean body mass; WtHR, waist to height ratio; WC, waist circumference; HC, hip circumference; MAP, mean arterial pressure; EQ VAS, visual analog scale; PHQ‐9, depression screener patient health questionnaire; GAD‐7, generalized anxiety disorder screener; 6MWT, 6‐min walk test; D, dominant; ND, non‐dominant. P calculated using mixed ANOVA. Model adjusted for age and supplementation use. Change during the intervention was significant in the ≥ 40 BMI group but not in the < 40 BMI group. Change during the intervention was significant in the < 40 BMI group but not in the ≥ 40 BMI group. Gut Microbiota Significant differences were observed in alpha diversity patterns dependent on BMI. Observed richness showed a significant main effect of BMI (p = 0.012) and a significant time x group interaction (p = 0.028), with further analysis showing that individuals with BMI ≥40 kg/m2 exhibited a marked reduction in observed richness from T1 to T3, whereas no change was detected in the <40 BMI group. Similarly, Shannon diversity showed a significant effect of time (p = 0.023) and a significant time x group interaction (p = 0.042), reflecting a decline in diversity in participants with BMI ≥40 kg/m2 but not in those with BMI<40 kg/m2. Physical and Mental Variables Consistent with the baseline anthropometric differences between BMI groups, participants in the <40 kg/m2 group exhibited higher grip strength and better endurance (Table 3). However, SGPALS scores indicated that both groups reported similar levels of physical activity at T1 (Table 4). Following the intervention, both BMI groups demonstrated improvements in physical performance, with increases in grip strength and endurance. Notably, the <40 kg/m2 group showed a significantly greater improvement in endurance compared to the ≥40 kg/m2 group. A significant decrease in MAP was observed in both groups, accompanied by an increase in resting heart rate. Furthermore, EQ VAS scores improved in both BMI groups, indicating enhanced self‐rated health status (Table 3). TABLE 4Changes in self‐reported physical activity from T1 to T3 for <40 and ≥40 BMI using the Saltin–Grimby physical activity level scale (SGPALS). < 40 BMI (n = 17)* T3 T1InactiveSome light activityRegular moderate activityTotalInactive2 (100.0)1 (11.1)0 (0.0)3Some light activity0 (0.0)6 (66.7)5 (83.3)11Regular moderate activity0 (0.0)2 (22.2)1 (16.7)3Total2 (100.0)9 (100.0)6 (100.0)17 ≥ 40 BMI (n = 13)* T3T1InactiveSome light activityRegular moderate activityTotalInactive1 (100)1 (33.3)1 (11.1)3Some light activity0 (0.0)2 (66.7)2 (22.2)4Regular moderate activity0 (0.0)0 (0.0)6 (66.7)6Total1 (100.0)3 (100.0)9 (100.0)13Data shown as count (percentage in brackets).Inactive: Sedentary.Some light activity: Walking, cycling, or other light activities for at least 4 h/week.Regular moderate activity: More structured activity for at least 2–3 h/week*Sample size < 40 BMI n = 17, ≥ 40 n = 13 due to missing responses for this variable.John Wiley & Sons, Ltd. Changes in self‐reported physical activity from T1 to T3 for <40 and ≥40 BMI using the Saltin–Grimby physical activity level scale (SGPALS). < 40 BMI (n = 17)* T3 Data shown as count (percentage in brackets). Inactive: Sedentary. Some light activity: Walking, cycling, or other light activities for at least 4 h/week. Regular moderate activity: More structured activity for at least 2–3 h/week*Sample size < 40 BMI n = 17, ≥ 40 n = 13 due to missing responses for this variable. Discussion This study investigated the effects of a structured interdisciplinary lifestyle intervention, with additional evaluation of 12 weeks of chitosan supplementation (LipoSan Ultra) in Icelandic females with obesity aged 31–68 years. The results indicate that chitosan supplementation was associated with some transient changes in gut microbiota composition, including a short‐term reduction in the F/B ratio, but no sustained changes in overall microbiota diversity. Moreover, no significant changes were observed in individual microbiota taxa using ANCOM‐BC2 analysis, nor were there specific clinical or biochemical outcomes that could be directly attributed to chitosan alone. A transient reduction in the F/B ratio was observed in the TRG during the intervention period. Similar changes have previously been reported following chitosan supplementation in adolescents with overweight or obesity [13]. However, this finding should be interpreted with caution. The change was temporary; no significantly differentially abundant taxa were identified using ANCOM‐BC2, and the study was not powered to detect subtle microbiome effects. Furthermore, the biological significance of variations in the F/B ratio remains debated, as associations with obesity and metabolic health have been inconsistent across studies [32]. Therefore, while the observed shift may suggest a modest effect of chitosan on overall gut microbiota structure, the present data do not support strong conclusions regarding its biological or clinical relevance. Our earlier study involving females with BMI ranging from 18 to 35 kg/m2 did not detect such changes following chitosan supplementation alone [12], suggesting that responsiveness may depend on population characteristics or study context. Despite these gut microbiota alterations, chitosan supplementation did not lead to significant changes in body weight or composition overall, contrasting with findings from Schiller et al. [9], who reported weight loss in participants with overweight or mild obesity using a similar 8‐week chitosan supplementation. However, the current results are more aligned with our previous analysis in which 12 weeks of LipoSan Ultra without lifestyle changes in females with BMI between 19 and 35 kg/m2 did not produce significant anthropometric or metabolic effects [12]. It is plausible that the study size limits sensitivity to detect subtle effects, particularly in subgroup and microbiome‐related analyses. This highlights the need for future studies to clarify whether chitosan efficacy varies by BMI subgroup or when used independently of lifestyle interventions. The impact of chitosan supplementation on the human gut microbiota remains poorly understood. While some shifts were observed, the absence of significant differential abundance suggests that chitosan may induce low‐magnitude, community‐wide changes not detectable at the individual taxa level and should, therefore, be interpreted with caution. The longitudinal design of the current study, with four fecal samples collected per participant, allowed us to observe these early and possibly transient effects. Moreover, a trend towards improved stool consistency reported by participants contrasts with constipation noted in our previous chitosan study without lifestyle changes [12], suggesting that the addition of lifestyle changes may help mitigate gastrointestinal side effects. However, the effects of medications on stool consistency in the present study population group cannot be ruled out as a potential confounder. The structured lifestyle intervention was the primary driver of the observed clinically meaningful health improvements. Both BMI groups experienced reductions in body fat percentage and increases in lean mass, without significant changes in total body weight. These compositional changes were accompanied by improvements in grip strength and endurance. Subsequently, participants in both groups reported better perceived health status, suggesting enhanced overall physical function and well‐being. These results support the notion that meaningful health improvements can occur in the absence of weight loss, underscoring the value of comprehensive lifestyle interventions that target body composition, physical function, metabolic and mental health as well as weight loss. This is consistent with previous findings emphasizing the importance of broader outcome measures in obesity treatment [33]. Interestingly, individuals with BMI ≥ 40 kg/m2 exhibited a reduction in gut microbiota richness and diversity over the 12‐week intervention period, whereas no comparable change was observed among participants with BMI <40 kg/m2. However, these findings should be interpreted with caution. The BMI subgroup analysis was exploratory, involving a limited number of participants, and was not powered to supply definitive gut microbiota‐related conclusions. Furthermore, no complementary functional analyses, such as metabolomic profiling, were performed to determine whether the observed changes in alpha diversity were associated with biologically meaningful alterations in microbial function. Although reduced gut microbiota diversity has previously been linked to adverse metabolic outcomes [34, 35], the clinical significance of the diversity changes observed in the present study remains uncertain. These findings should therefore be considered hypothesis‐generating and warrant confirmation in larger studies. The observed decline may reflect a gut ecosystem that is more vulnerable to perturbation in individuals with severe obesity, rather than a clearly detrimental response. However, there is a gap in the literature regarding how lifestyle‐induced changes in the gut microbiota diversity relate to clinical outcomes in individuals with severe obesity. Thus, further research is needed to elucidate both the mechanisms and potential clinical implications of these changes in this population. A key strength of this study lies in its randomized, double‐blind, placebo‐controlled design, and execution in a well‐defined human cohort. The use of DXA for body composition analysis provided accurate and sensitive detection of regional changes in fat and lean mass. The inclusion of a safety monitoring blood draw at week 2 enabled early detection of any potential adverse responses; however, no adverse effects were observed. The exclusive inclusion of female participants, while limiting generalizability, increased sample homogeneity and improved statistical power within this subgroup. Importantly, participants took part in a tailored, intensive inpatient interdisciplinary rehabilitation program daily for 5 weeks, which strengthened adherence to lifestyle modification and enhanced consistency across participants. However, further recruitment was hindered by COVID‐19‐related restrictions, ultimately resulting in a smaller sample size than originally intended. Several limitations should be acknowledged. First, the absence of objective physical activity data before the intervention limits our ability to quantify changes relative to baseline activity levels. Second, the cohort consisted exclusively of Caucasian Icelandic females aged 31–68 y with obesity, limiting generalizability to other populations. Finally, the 12‐week intervention period may have been insufficient to capture the full range of potential effects, particularly for gut microbiota or long‐term metabolic outcomes. The study size further limits the interpretability of subgroup and microbiome‐specific findings. Additionally, three participants in the intervention group were receiving stable treatment with GLP‐1 receptor agonists, which could potentially influence weight and metabolic outcomes. However, no differential response was observed in these individuals compared with the rest of the group, and no weight loss was detected at the group level. While a confounding effect cannot be entirely excluded, the limited number of participants and absence of observable differences suggest that GLP‐1 use is unlikely to have substantially influenced the overall findings. Conclusion In conclusion, the structured lifestyle intervention was the primary driver of improvements in body composition, physical function, and perceived health among women with obesity. While chitosan supplementation did not independently promote weight loss or considerable metabolic changes, it was associated with a transient decrease in the F/B ratio, suggesting potential modulation of the gut microbiota, although these effects warrant cautious interpretation due to small sample size and the biological significance of this finding remains uncertain. Future studies with larger, more diverse cohorts and longer supplement durations are needed to clarify the specific contexts in which chitosan may be most effective, and particularly whether it is best used as a standalone supplement or as an adjunct to structured lifestyle interventions. Funding This research was supported by the Technology Development Fund in Iceland (grant agreement number 198694‐0611, MicroFIBERgut). Conflicts of Interest The authors declare no conflicts of interest. Trial Registration Number This trial was registered at clinicaltrials.gov as NCT04551365. Supporting information Supporting File: mnfr70558‐sup‐0001‐SuppMat.docx.