Feasibility and Preliminary Efficacy of a Theory-driven BALANCE Dietary Program Among Patients With Type 2 Diabetes

Fasting Blood Glucose (FBG) was measured after an overnight fast of at least 8 hours. Venous blood samples were collected and analyzed using a standardized enzymatic method in a certified clinical laboratory. FBG values were reported in mg/dL. Fasting was defined as no caloric intake except water during the fasting period. The measure reflects baseline glucose levels and was used to assess glycemic status according to established clinical criteria.

Serum total cholesterol was measured to assess overall lipid metabolism status. After an overnight fast of at least 8 hours, venous blood samples were collected and serum total cholesterol concentration was determined using an enzymatic colorimetric method (cholesterol oxidase-peroxidase, CHOD-PAP). Results are reported in mmol/L (conversion: 1 mmol/L = 38.67 mg/dL). According to the Chinese Guidelines for the Prevention and Treatment of Dyslipidemia in Adults (2016 revision), desirable TC is <5.18 mmol/L, borderline high is 5.18-6.19 mmol/L, and elevated TC is ≥6.22 mmol/L. Lower values indicate improved lipid control.

Serum total cholesterol was measured to assess overall lipid metabolism status. After an overnight fast of at least 8 hours, venous blood samples were collected and serum total cholesterol concentration was determined using an enzymatic colorimetric method (cholesterol oxidase-peroxidase, CHOD-PAP). Results are reported in mmol/L (conversion: 1 mmol/L = 38.67 mg/dL). According to the Chinese Guidelines for the Prevention and Treatment of Dyslipidemia in Adults (2016 revision), desirable TC is <5.18 mmol/L, borderline high is 5.18-6.19 mmol/L, and elevated TC is ≥6.22 mmol/L. Lower values indicate improved lipid control.

Diabetes knowledge assessed using the Chinese version of the Diabetes Knowledge Test (DKT) scale, originally developed by Fitzgerald et al. and culturally adapted and validated by Sun et al. The scale consists of 23 questions, each worth 1 point. Total score ranges from 0 to 23, with higher scores indicating better diabetes knowledge. The scale demonstrates acceptable internal consistency (Cronbach's α = 0.76).

Self-efficacy assessed using the Diabetes Self-Efficacy Scale (DSES) developed by Lorig et al. and validated by Sun et al. The scale consists of 8 items, each rated on a scale from 1 to 10. Total score ranges from 8 to 80, with higher scores indicating stronger self-efficacy in diabetes self-management. The scale demonstrates excellent internal consistency (Cronbach's α = 0.889).

The Summary of Diabetes Self-Care Activities (SDSCA) scale will be used to evaluate patients' diabetes self-management behaviours. Originally developed by Toobert et al., the SDSCA has been translated into Chinese and validated for the Chinese population by Sun et al. The scale assesses patients' self-care performance over the preceding 7 days across five behavioural domains: general diet, specific diet, physical activity, blood glucose self-monitoring, and foot care. It comprises 11 items, each scored on a scale of 0 to 7, where 0 indicates that the behaviour was not performed on any day and 7 indicates performance on every day of the week. Total scores are calculated as the mean of all item scores, with higher scores reflecting better self-management. The Chinese version of the SDSCA has demonstrated good internal consistency, with a Cronbach's α of 0.84.

Insulin resistance was evaluated using the Homeostasis Model As-sessment of Insulin Resistance (HOMA-IR). HOMA-IR was calculated using the formula: HOMA-IR = [FPG (mmol/L) × FINS (μU/mL)] / 22.5. Higher HOMA-IR values indicate greater insulin resistance.

Insulin sensitivity was evaluated using the Matsuda Insulin Sensitivity Index (ISI-Matsuda), a composite measure derived from a standard 75-g oral glucose tolerance test (OGTT). Higher ISI-Matsuda values indicate greater insulin sensitivity.

C-peptide levels were determined using an electrochemiluminescence immunoassay (ECLIA) method. Results are reported in ng/mL and reference range is 0.78-1.89 ng/mL

Glucose tolerance was assessed using a standard 75-g oral glucose tolerance test (OGTT) performed according to the World Health Organization (WHO) protocol. After an overnight fast of at least 8-10 hours, a fasting venous blood sample was collected for baseline plasma glucose measurement. Participants then ingested a solution containing 75 g anhydrous glucose dissolved in 250-300 mL water within 5 minutes. Subsequent venous blood samples were collected at 30, 60, 90, and 120 minutes post-ingestion for plasma glucose measurement using the glucose oxidase or hexokinase enzymatic method. The primary outcome parameter is the 2-hour postload plasma glucose (2h-PG, mmol/L). Glucose tolerance status was classified according to WHO/ADA criteria: normal glucose tolerance (NGT): FPG <6.1 mmol/L and 2h-PG <7.8 mmol/L; impaired fasting glucose (IFG): FPG 6.1-6.9 mmol/L and 2h-PG <7.8 mmol/L; impaired glucose tolerance (IGT): FPG <7.0 mmol/L and 2h-PG 7.8-11.0 mmol/L; diabetes mellitus (DM): FPG ≥

Serum high-density lipoprotein cholesterol was measured to assess reverse cholesterol transport capacity and cardiovascular protective status. After an overnight fast of at least 8 hours, venous blood samples were collected and serum HDL-C concentration was determined using a direct homogeneous enzymatic method (direct HDL assay). Results are reported in mmol/L (conversion: 1 mmol/L = 38.67 mg/dL). Unlike TC, TG, and LDL-C, higher HDL-C values are considered protective against cardiovascular disease. According to the Chinese Guidelines for the Prevention and Treatment of Dyslipidemia in Adults (2016 revision), desirable HDL-C is ≥1.04 mmol/L and reduced HDL-C is <1.04 mmol/L.

Serum low-density lipoprotein cholesterol was measured as the primary indicator of atherogenic lipid burden. After an overnight fast of at least 8 hours, venous blood samples were collected and serum LDL-C concentration was determined using a direct homogeneous enzymatic method (direct LDL assay). Where the direct assay was not available, LDL-C was calculated using the Friedewald equation: LDL-C = TC - HDL-C - TG/2.2 (all values in mmol/L), applicable only when TG <4.52 mmol/L. Results are reported in mmol/L (conversion: 1 mmol/L = 38.67 mg/dL). According to the Chinese Guidelines for the Prevention and Treatment of Dyslipidemia in Adults (2016 revision), desirable LDL-C is <3.37 mmol/L, borderline high is 3.37-4.12 mmol/L, and elevated LDL-C is ≥4.14 mmol/L. Target LDL-C levels were individualized according to each participant's cardiovascular risk stratification.

Body composition was measured using a multi-frequency bioelectrical impedance analyzer (InBody 770, InBody Co., Ltd., Seoul, South Ko-rea). Participants fasted for ≥2 hours, avoided strenuous exercise for 24 hours, removed shoes and metal accessories, emptied their blad-ders, and stood upright for 5 minutes before testing to stabilize fluid distribution. During measurement, participants grasped hand elec-trodes with arms abducted ~30° from the body and remained still. Measured parameters included weight (kg), height (cm), BMI (kg/m²), body fat percentage (%), visceral fat area (cm²), and total body water (L). All assessments were performed at the same time of day (±2 hours) across visits.

Body composition was measured using a multi-frequency bioelectrical impedance analyzer (InBody 770, InBody Co., Ltd., Seoul, South Ko-rea). Participants fasted for ≥2 hours, avoided strenuous exercise for 24 hours, removed shoes and metal accessories, emptied their blad-ders, and stood upright for 5 minutes before testing to stabilize fluid distribution. During measurement, participants grasped hand elec-trodes with arms abducted ~30° from the body and remained still. Measured parameters included weight (kg), height (cm), BMI (kg/m²), body fat percentage (%), visceral fat area (cm²), and total body water (L). All assessments were performed at the same time of day (±2 hours) across visits.

Body composition was measured using a multi-frequency bioelectrical impedance analyzer (InBody 770, InBody Co., Ltd., Seoul, South Ko-rea). Participants fasted for ≥2 hours, avoided strenuous exercise for 24 hours, removed shoes and metal accessories, emptied their blad-ders, and stood upright for 5 minutes before testing to stabilize fluid distribution. During measurement, participants grasped hand elec-trodes with arms abducted ~30° from the body and remained still. Measured parameters included weight (kg), height (cm), BMI (kg/m²), body fat percentage (%), visceral fat area (cm²), and total body water (L). All assessments were performed at the same time of day (±2 hours) across visits.

Body composition was measured using a multi-frequency bioelectrical impedance analyzer (InBody 770, InBody Co., Ltd., Seoul, South Ko-rea). Participants fasted for ≥2 hours, avoided strenuous exercise for 24 hours, removed shoes and metal accessories, emptied their blad-ders, and stood upright for 5 minutes before testing to stabilize fluid distribution. During measurement, participants grasped hand elec-trodes with arms abducted ~30° from the body and remained still. Measured parameters included weight (kg), height (cm), BMI (kg/m²), body fat percentage (%), visceral fat area (cm²), and total body water (L). All assessments were performed at the same time of day (±2 hours) across visits.

Body composition was measured using a multi-frequency bioelectrical impedance analyzer (InBody 770, InBody Co., Ltd., Seoul, South Ko-rea). Participants fasted for ≥2 hours, avoided strenuous exercise for 24 hours, removed shoes and metal accessories, emptied their blad-ders, and stood upright for 5 minutes before testing to stabilize fluid distribution. During measurement, participants grasped hand elec-trodes with arms abducted ~30° from the body and remained still. Measured parameters included weight (kg), height (cm), BMI (kg/m²), body fat percentage (%), visceral fat area (cm²), and total body water (L). All assessments were performed at the same time of day (±2 hours) across visits.

A non-elastic flexible tape at the midpoint between the lowest rib margin and the iliac crest at the end of normal expiration, with the participant standing and arms relaxed at sides. Measurements were taken twice and averaged. Results are reported in cm.

A non-elastic flexible tape at the level of the greatest posterior protru-sion of the buttocks, with the participant standing with feet together. Measurements were taken twice and averaged. Results are reported in cm.

WHR is calculated as the waist circumference divided by the hip cir-cumference.

Trained investigators assisted with app installation and provided standardized usage instructions at enrollment. Participants photo-graphed and logged all meals and beverages daily, and the app au-tomatically calculated total caloric intake (kcal/day) and macronutri-ent composition (carbohydrate, protein, and fat in g/day and % of to-tal energy) based on a built-in Chinese food composition database.

Dietary adherence assessed using the Dietary Behaviour Compliance Scale developed by Zhao et al. The scale comprises 5 dimensions with 23 total items: fruits and vegetables (5 items), oil and salt (4 items), carbohydrates and fat (5 items), cooking and dining habits (4 items), and overall dietary self-regulation (5 items). Each item is rated on a 5-point Likert scale where 1 = "never" and 5 = "always". Total score ranges from 23 to 115, with higher scores indicating better adherence to recommended dietary behaviours. The scale demonstrates excellent internal consistency (Cronbach's α = 0.886).

Physical activity assessed using the IPAQ-SF. Participants recall and report frequency (days/week) and duration (minutes/day) of physical activity, plus total daily sitting time on weekdays, over the previous 7 days. Metabolic equivalent (MET) values are: 3.3 METs for walking, 4.0 METs for moderate-intensity activity, and 8.0 METs for vigorous-intensity activity. Weekly physical activity for each intensity is calculated as: MET-min/week = MET value × frequency (days/week) × duration (min/day). The questionnaire demonstrates good validity; correlations between weekly activity time across IPAQ-SF dimensions and pedometer-recorded total weekly step counts ranged from 0.417 to 0.440 (p < 0.05).