This is a single-center, parallel-group, randomized clinical trial with three intervention conditions implemented over an 8-week period. After screening and baseline procedures, participants are assigned in a 1:1:1 ratio using a computer-generated block randomization sequence prepared by an independent researcher not otherwise involved in study conduct. Allocation concealment is maintained using sequentially numbered, opaque, sealed envelopes opened only after confirming eligibility.
Due to the dietary nature of the interventions, participant-level blinding is not feasible. However, biospecimen handling and laboratory analyses are performed on coded (anonymized) samples; laboratory personnel remain blinded to group assignment and time point to minimize measurement bias.
Personalized medical nutrition therapy (MNT) is delivered using a standardized counseling framework with individualized targets. Resting metabolic rate is estimated using the Mifflin-St Jeor equation, and daily energy prescription is derived using physical activity level (PAL). Macro-distribution targets are standardized (10-20% protein, 45-60% carbohydrate, 25-30% fat), with saturated fat \<7% of energy and trans fat \<1% to align with guidance-based cardiometabolic risk control.
To reduce confounding from background polyphenol exposure, participants are instructed to avoid high-polyphenol red/purple fruits and vegetables according to protocol-defined lists; diet plans are structured as three main meals plus 2-3 snacks, and standardized exchange lists are provided to support adherence.
Hardaliye intervention (product handling and quality assurance): The study beverage is sourced from a traditional producer in Kırklareli and manufactured via controlled spontaneous fermentation without starter culture.
Hardaliye is dispensed in 200 mL bottles and distributed weekly under cold-chain conditions; participants are instructed to consume the beverage daily within a consistent intake window (16:00-16:30) to standardize exposure timing.
To ensure batch-to-batch consistency, each production batch undergoes predefined acceptance checks (°Brix, pH, and titratable acidity), and key compositional markers are measured post-bottling (total phenolics via Folin-Ciocalteu, total monomeric anthocyanins via pH-differential method, and antioxidant capacity via ABTS-based TEAC). Analytical repeatability is monitored (replicates/parallel runs; CV target \<10% with repeat testing if unmet).
Batches outside acceptance ranges (e.g., °Brix 20.5-21.8; pH 3.50-4.00; acidity 0.68-0.75 g/100 mL; and protocol-defined total phenolic/anthocyanin targets) are excluded from participant distribution.
Where relevant, chromatographic confirmation (HPLC-DAD) is used to verify the phenolic profile of the study product and support linkage of total phenolic estimates to compound-level composition.
Participants complete an initial screening visit followed by a baseline (Day 0) visit and an end-of-intervention (Week 8) visit, with brief interim check-ins at Weeks 2, 4, and 6 for adherence review and systematic safety monitoring. Interim contacts include beverage count/log review, reinforcement of dietary counseling where applicable, and adverse event solicitation/documentation.
Adherence is operationalized as ≥80% compliance with the assigned beverage intake and/or nutrition therapy targets; intake and adherence are captured using simple participant logs reviewed at each contact.
Dietary intake is assessed using structured food records collected at multiple time points (7-day records at baseline, Week 4, and Week 8; and additional 3-day records during specified weeks) to quantify energy and nutrient intake and to support adherence monitoring. Nutrient analyses are performed using BeBIS software and interpreted against national reference intakes.
Dietary polyphenol exposure is estimated using Phenol-Explorer (v3.6) to derive total polyphenols and relevant subclasses; an antioxidant diet quality score (DAQS) is calculated based on intake of key antioxidant nutrients. Physical activity is monitored via 24-hour recall on diary days, and PAL is derived using activity-specific PAR values combined with Mifflin-St Jeor BMR estimates to support stability of lifestyle exposure during follow-up.
Fasting venous blood is collected at baseline and Week 8 and processed according to institutional standard operating procedures (timing, centrifugation, aliquoting, storage). Samples are labeled with coded identifiers to preserve analytical blinding, and assays are performed following validated methods and kit instructions with internal QC steps documented prospectively.
Data are captured in structured case report forms and reconciled against visit checklists, distribution records, and participant logs. Primary analyses follow an intention-to-treat approach. Group differences in change over time are evaluated using repeated-measures modeling (e.g., linear mixed-effects models including group, time, and group×time interaction), with prespecified sensitivity analyses based on adherence and protocol deviations; missing-data handling and multiplicity procedures are defined in the statistical analysis plan.
