Prediabetic participants (n = 200), both males and females (non-pregnant/non-lactating), between 30 - 65 years of age were recruited in the study. Prediabetes was diagnosed on HbA1C ≥ 5.7% to ≤6.4% and/or fasting blood glucose between 100 - 125 mg/dl [5] [27] . The participants who were not receiving any anti-diabetic medication and willing to consent were recruited in the study. Participants who had BMI < 18.0 and >32.5 kg/m², or had any acute infections, eating disorders, or were taking any medication that could have profoundly affected blood glucose, were habitual substance abuse (alcohol, smoking, tobacco) consumers or had any known history of kidney or liver diseases in past 3 months were excluded at the screening.

The sample size was calculated considering an average 10% change in iAUC as outcome, 80% power and an alpha of 0.05 and with a dropout rate of 10%, making a total sample size to be 126 (n = 63 in each arm of the study). In total, 1286 participants were screened, out of which 992 did not meet the inclusion/exclusion criteria. Considering dropouts, two hundred participants were finally recruited after receiving consent for the study. They were randomized into test and control groups according to a schedule generated using SAS® statistical software (Version: 9.4; SAS Institute Inc., USA). One participant from the control group dropped out after randomization. Thus, a total of 199 participants [test (n = 100) and control (n = 99)] participated in this parallel phase study ( Figure 1 ). The study was conducted in two parallel phases for 6 weeks: Acute Blood Glucose Response (ABGR) and Intervention phase including Continuous Glucose Monitoring (CGM). The same participants were involved in both phases ( Figure 1 ).

The test DSNS is a low GI [68] nutrition supplement (Horlicks Diabetes Plus provided by Hindustan Unilever Limited). It is a dual fiber blend (Nutriose® and Fibersol®) formulation containing 22% soluble fiber and with a high MUFA composition (70% of total fatty acids delivering ~18% of energy). It delivers 13 vitamins and 7 minerals (≥15% RDA) along with the three macronutrients and fiber (refer Table S1 for nutrient Composition of DSNS). The carbohydrate delivers 40%, fiber 12%, protein 23% and fat 25% of total energy. This formula has been designed in line with the ICMR-INDIAB 2021 recommendations for Type 2 Diabetes (T2D) remission and prevention of progression to T2D in prediabetic and Normal Glucose Tolerance (NGT) individuals [38] .

ABGR phase was conducted on four visit days during the study. During each visit, participants reported at the clinical site on the previous day evening and were provided with a standard dinner. It was ensured that the time of food administration was kept nearly the same for all the participants (window period ± 1 hour). Participants were to consume their meal at the latest by 10 pm and were required to stay overnight on-site to maintain fasting for nearly 10 hours. On visit days 1, 15, 29, and 43, FBG was obtained by withdrawing fingertip capillary blood through finger prick method and using validated Hemocue 201+. The FBG sample was collected prior to product administration [15 mins (±2 mins)]. Pre-weighed 60 g DSNS in 250 ml of double toned milk was provided to the test group (n = 100). Isocaloric snack of 56 g of cornflakes in 250 ml of double toned milk was provided to the control group (n = 99). The participants were instructed to consume the same within 15-minutes. Participants did not consume anything for the next 180 min (till the last 180 min. blood sample was drawn). After administration of test product/isocaloric snack, estimation of Postprandial Glucose (PPG) was performed at 30, 60, 90, 120, 150 and 180 min (±2 minutes) and it was recorded. The details of the test (DSNS) and control (isocaloric snack) products are presented in Table 1 .

Note: ^aAssessed on days 1, 15, 29 and 43. *In 250 mL double toned milk. **In 200 ml warm water. DSNS is a micronutrient fortified supplement rich in vitamins and minerals.

Initially, for the ABGR phase, the DSNS (60 g) was administered in water and isocaloric snack of 28 g of cornflakes in 200 ml of toned milk. During the first day (visit 2) for the first set of participants (n = 57; not included in the analysis for ABGR phase), the study team received feedback that participants felt hungry even after consuming the test and control products, likely due to the overnight fast. Also, the participants could not consume anything for the next 180 mins post the product administration. Hence, to support study compliance, smooth ethical conduct, and to improve the wholesomeness of both test and control snacks, the DSNS (60 g) was agreed to be reconstituted in 250 ml double toned milk. To match the energy delivery in control group, 56 g of cornflakes were administered in 250 ml double toned milk. This was well accepted by the following study participants.

This phase was conducted to assess the impact of the DSNS along with lifestyle counselling (DSNS + LC) versus lifestyle counselling alone (LC alone) in prediabetic adults ( Figure 2 ) post 4 weeks of intervention. At the baseline, data on FBG, Hb1Ac, body weight, height, BMI, body composition, blood pressure, along with demography, medical history, 24 hours dietary recall and physical activity questionnaire were collected. The actual intervention phase for DSNS + LC versus LC alone was for 4 weeks (28 days) from day 15 to day 42 with three visits in between visit 3 (day 15), visit 4 (day 29) and visit 5 (day 43, endline). The test group participants received DSNS + LC while the control group received LC alone during the intervention phase. All subjects were reassessed on the above parameters on visits days which coincided with their ABGR assessments visits.

The participants in DSNS + LC group were provided with a 14 day supply of DSNS sachets and counselled on method of reconstitution for consumption at home ( Table 1 ) on day 15 and reconfirmed on Day 29. The participants were instructed to consume the DSNS twice a day (30 g each time with 200 ml warm water)—before breakfast and during early evening (between lunch and dinner). Shakers were provided to participants to ensure accurate dilution and compliance with consumption. The compliance was ensured by regular telephonic follow-ups and collecting empty sachets by the study staff before the next batch of sachets was handed over to the participants on day 28 (visit 4). The second batch of empty sachets was collected on day 42 (visit 5). The LC sessions were conducted twice (on days 15 and 29), for all participants during the intervention phase. The product compliance was 100% among study completers. The LC sessions focused on the importance of having a balanced diet, carbohydrate sources, reducing simple sugars and fats, increasing protein and fiber intake along with increasing intakes of fruits and vegetables. The sessions also emphasized daily walk at least for 30 minutes.

Body weight was measured to the nearest 0.1 kg using digital weighing scale (BodySCAN™ HBF 214 AP, Omron). Height was measured using a stadiometer (26SM, Biocon) to the nearest 0·1 cm. Body Mass Index (BMI) was calculated as

weight (kg)/height (m²). Waist Circumference (WC) was measured midway between the iliac crest and the lowermost margin of the ribs, with the subject breathing normally using a non-stretchable tape. Body fat percentage (%) and skeletal mass (kg) were estimated using BodySCAN™ HBF-214 Body Composition Monitor (Omron). Systolic and diastolic blood pressure was measured using calibrated digital BP apparatus (BP 02XL, Dr. Morepen). Fasting blood glucose and HBA1c were assessed using venous blood samples on visit 1 and visit 5 (day 43). Twenty-four-hour dietary recalls were collected using a standard questionnaire. Information on physical activity was collected using modified Madras Diabetes Research Foundation physical activity questionnaire (MPAQ) [69] .

Impact of DSNS on the ambulatory glucose profile was studied using a CGM device between two 14-day phases: CGM1 phase (baseline, days 1 - 14) and CGM2 phase (endline, days 28 - 42). The CGM system used was the Free Style Libre Pro (manufactured by Abbott Healthcare Pvt Ltd) and assessed the interstitial fluid glucose concentration. The CGM 1 phase was designed as the baseline for the 4-week intervention to elicit any differences in the CGM outcomes before and during/after the 4-week intervention ( Figure 2 ). Randomized participants reported at the study site on day 0 and the CGM device was inserted in the back of the upper arm post dinner for all study participants. They were asked to continue wearing the device for the next 14 days. Participants in both the groups were instructed to continue with their habitual diet and routine physical activity for this CGM1 baseline. The CGM device was removed for all the study participants on Day 14 at bedtime (visit 3) providing a 14-day CGM1 phase data (baseline). All participants received lifestyle counselling on Day 15. In addition to LC, the test group was also provided the DSNS for daily consumption for the next 14 days. During this period, participants were not wearing any CGM device. All study participants reported back on day 29 (visit 4) for the CGM2 baseline.

On day 28, the participants reported back to the study site and a new CGM device was inserted post dinner (visit 4), for all participants. Again, the participants in both groups received lifestyle counselling and the test group was given the DSNS for daily consumption for the next 14 days. All the participants reported back on day 42 (visit 5) for the endline visit and CGM removal. CGM 1 phase (days 1 - 14) and CGM 2 phase (days 29 - 42) records were compared to assess the efficacy of intervention on mean blood glucose, estimated HbA1c, and Time in Range (TIR), among DSNS + LC as well as LC alone groups.

The statistical analysis was done by using SAS^® statistical software (Version: 9.4 or higher; SAS Institute Inc., USA). Intention to treat analysis was conducted for 199 participants [test (n = 100) and control (n = 99)] who participated in this parallel phase study. During the study period, 35 participants in the test and 37 participants in the control group did not complete the study (reasons for dropout as listed in Figure 2 ). Consequently, a per protocol analysis was conducted, encompassing 65 participants in the test, and 63 in control group ( Figure 1 ).

Continuous variables were summarized using tables of descriptive statistics: number of participants with recorded observations, mean, standard deviation, median, minimum, and maximum. Categorical variables were summarized using counts and percentages. Least square means with standard error were computed, wherever necessary. For continuous variables, the within-treatment analyses were conducted to compare baseline to post-intervention data using paired t-test. The data between the groups for various parameters was analyzed using an independent t-test. Mean glucose concentrations and TIR values during the CGM1 and 2 were obtained from the CGM device report. Using 54 - 140 mg/dl as normal range for CGM interstitial values, percent TIR was calculated. Estimated HbA1c% was calculated using mean glucose (mg/dl) values into a formula [eHbA1c = 3.38 + 0.023453 (mean glucose) [70] . The nutrient intake of foods was calculated using IFCT 2017 [71] and food atlas (MDRF) [72] . The mean intakes and nutrient adequacy of major nutrients was compared with Recommended Dietary Allowances (RDAs) given by Indian Council of Medical Research (ICMR), RDA 2024 [73] . Percent energy contribution from macronutrients was also calculated. ANCOVA was used for assessing the differences between the two groups in various nutrient intakes and adequacies changes from baseline to endline. The physical activities are classified into sedentary, moderate, and vigorous categories in the MPAQ [69] . We have used modified MAPQ, thus changes in physical activities under each category were calculated as mean duration (minutes) per day. Safety data was analyzed for the safety population defined as all enrolled participants who received at least one of the test products or isocaloric meals. Adverse Events were summarized with a number and the percentage. All statistical tests of hypothesis employed a level of significance of 0.05.

DSNS group showed significantly lower PPG versus isocaloric snack on all visit days at 30 (p < 0.0001), 60 (p < 0.0001), 90 (p < 0.0001), 120 (p < 0.05) and 150 (p < 0.05) minutes. As per Intent to Treat (ITT) analysis, PPG response from

Note: ^aData from visit 2. ^#Wilcoxan rank test of significance was used for comparing gender differences in the groups. All other parameters were compared using unpaired t-test, at p < 0.05 level of significance.

0 to 180 minutes after DSNS and isocaloric snack consumption by the participants in DSNS and control groups, respectively, on days 1, 15, 29, 43 is shown in Figure 3 .

The maximum glucose peak was achieved at 60 minutes for both the groups on all the days. However, it was significantly lower in the DSNS group as compared to control group at all the days [day 1 (<0.0001), day 15 (<0.0001), day 29 (<0.0001) and day 43 (<0.0001)]. The differences in PPG levels reduced between DSNS and control group at 180 minutes for all the four days. The Per Protocol (PP) data showed similar results. Detailed data is given in Table S2 .

The iAUC was significantly lower in DSNS group versus the control group at day 1 (<0.0001), day 15 (<0.0001), day 29 (<0.0001), and day 43 (<0.0001) as per ITT analysis ( Figure 4 ). Similar results were obtained from PP data. Detailed data provided in Table S3 .

At baseline, no significant differences were observed in any of the parameters between the two groups as per ITT analysis. Post intervention, FBG, waist circumference, systolic and diastolic blood pressure reduced from baseline significantly

after 4 weeks of intervention for both the groups with the mean (SD) change being more pronounced in DSNS + LC group versus the LC alone group [FBG: DSNS + LC: −25.3 (26.51) mg/dl, p < 0.0001; LC alone: −24.4 (21.66) mg/dl, p < 0.0001; waist circumference: DSNS + LC: −1.13 (1.86) cm, p < 0.0001; LC alone: −0.88 (1.86) cm, p = 0.0004; Systolic BP: DSNS + LC: −3.3 (4.63) mm Hg, p < 0.0001; LC alone: −2.5 (4.51) mm Hg, p < 0.0001; Diastolic BP: DSNS + LC: −3.1 (3.63) mm Hg, p < 0.0001; LC alone: −2.6 (3.72) mm Hg, p < 0.0001]. A reduction in weight and percent body fat from baseline was reported by both the DSNS + LC and LC alone groups. While the change in weight was significant for DSNS + LC group [DSNS + LC: −0.4 (1.27) p = 0.0104, LC alone: −1.33 (6.27), kg/m², p > 0.05], the change in percent body fat was significant for LC alone group [DSNS + LC: −1.02 (4.34)%, p = 0.058; LC alone: −0.98 (3.5)%, p < 0.028]. Mean (SD) reductions in BMI were significant for both the groups [DSNS + LC: −0.29 (0.74) p = −0.0022, LC alone: −0.33 (0.74), kg/m², p = 0.0007]. Percent skeletal mass and Hb1AC did not show any significant changes after 4 weeks intervention in either of the groups. There were no significant differences observed between the test and control groups for any of the measures. The results were observed to be similar in PP analysis. Details are given in Table S4 .

The dietary intake of participants was assessed at baseline (day 1) and at the end of the study (day 43) using 24-hour recall method.

Macronutrients: Changes in percent energy contribution and nutrient adequacy

From the dietary data, average nutrient intake among the participants was computed. The percentage energy contribution from carbohydrate, protein and fat was similar between DSNS + LC and LC alone group at baseline. Post intervention, DSNS + LC group reported a significant reduction in the contribution of energy from carbohydrates, (DSNS + LC: −5.3 (9.22)%, LC alone: 1.6 (24.44)%, p = 0.0349), and a significant increase in protein energy percent (DSNS + LC: 1.5 (1.51)%, LC alone: 0.03 (1.37)%, p < 0.0001) compared to LC alone group, shown in Figure 5 . There were no significant changes reported in fat energy percent.

Mean protein intake increased significantly (p < 0.05) in DSNS + LC group post intervention from baseline [50.4 (20.24) g] to endline [55.3 (15.96) g]; while there were no significant changes in the energy, protein, carbohydrate, and fat intakes in LC alone group as compared to baseline. Post intervention, percentage adequacy for protein intake increased in DSNS + LC group as compared to LC alone group (DSNS + LC: 10.3 (41.83)%, LC alone: −7.2 (31.64)%, p < 0.001). On adjusting the change from baseline, the percent adequacy for fibre intake also improved in DSNS + LC, compared to LC alone group (DSNS + LC:30.2 (13.10)%, LC alone: −7 (13.30)%, p = 0.0513).

Micronutrients: Changes in mean intake and percent adequacy

Micronutrient intake data was collected and analyzed for calcium, iron, vitamin B12 and chromium. Post intervention, the mean intakes of calcium, iron, vitamin B12, and chromium were significantly (p < 0.0001 for all) higher in DSNS + LC as compared to LC alone group ( Table 3 ).

Percent adequacies of calcium, iron, vitamin B12, and chromium were found to be inadequate at baseline for both the groups. The percent adequacies for intakes of calcium, iron, vitamin B12, and chromium, not only increased substantially (p < 0.0001 for all) in DSNS + LC group post intervention as compared to baseline but were also significantly (p < 0.0001 for all) higher than LC alone group at endline ( Table 3 ). The participants in the DSNS + LC group displayed significantly augmented nutrient intake at the endline compared to baseline levels. Similar results were obtained from per-protocol analysis.

Changes in physical activity

At baseline, majority of the participants were sedentary (DSNS + LC: 97%, LC alone: 97%) and engaged in light physical activities (DSNS + LC: 44%, LC alone: 47%). Only 13% of participants in DSNS + LC and 19% in LC alone groups were engaged in moderate activities. Following the lifestyle counseling sessions, a significant increase in the proportion of participants engaged in moderate activities was observed in both groups (DSNS + LC: 98%, p < 0.0001; LC alone: 100%, p < 0.0001). There was also a significant rise in the time (mean) spent on moderate activities, with participants in the DSNS + LC group increasing by 33.6 (15.47) minutes (p < 0.005) and those in the LC alone group increasing by 30 (18.71) minutes (p = 0.0007) as compared to their respective baseline. However, no significant differences in physical activity patterns were noted between the two groups. Data from ITT and PP analysis of physical activity pattern of prediabetic participants is shown in Table S5 .

The comparison between baseline and endline CGM data suggested that 4 weeks intervention (both DSNS + LC and LC alone) did not show any substantial changes in glycemic parameters in prediabetic individuals using a CGM device in prediabetics. Data not shown in the paper.

Safety data was analyzed for all participants who received either the DSNS or isocaloric snack. Only two minor adverse events (1%), one in DSNS + LC and one in LC alone group were reported. Both AEs were of mild intensity. Both subjects experienced discomfort and mild diarrhoea after consumption of test products. Their vital signs were within normal limits and both the participants

Note: Values are Mean (SD) for baseline and endline. % nutrient adequacies were computed by comparing the nutrient intake values with RDA 2024 (ICMR), p-value for within group was calculated using paired t-test. Least square mean (SE) was used to compare the mean change from baseline between groups by ANCOVA, where p-value is significant at * < 0.05 and ** < 0.0001.

felt comfortable after taking a rest. No other participant reported any other adverse events either after consuming the DSNS or isocaloric snack.