The mobile system "eSense Pulse" by Mindfield Biosystems Ltd. (Gronau, Germany) was used for RFB-HRV-BF training. The participants sat in an upright position, started the training via the eSense app on their smartphone, and were instructed to practice breathing at their individually determined resonance frequency.During training, a ball that expands (breathe in) and shrinks (breathe out) on the screen specified the breathing frequency goal to a value of six cycles per minute (five seconds inspiration, five seconds expiration). The HRV was visualized by new symbols varying in color depending on the respectively measured values, which appeared every ten seconds.

The mobile system eSense Pulse from Mindfield Biosystems is used for the digital anti-stress program. To do this, the participants put on a chest strap with a sensor and can start training via the eSense app. The anti-stress program is carried out using a procedure generated for the study. During the anti-stress program units, the participants receive application-related information and tips to better understand the topic of stress, identify stressors in their own lives and reduce stress. The active control intervention of the anti-stress program was selected to investigate the effect of a specific stress level reduction without the direct modulation of physiological mechanisms of action as with HRV-BF-RFA.

During training, a ball that expands (breathe in) and shrinks (breathe out) on the screen specified the breathing frequency goal to a value of six cycles per minute (five seconds inspiration, five seconds expiration). The HRV was visualized by new symbols varying in color depending on the respectively measured values, which appeared every ten seconds. When the participants achieved the specified breathing frequency, resulting in an increased sinus arrhythmia, the symbols appearing on the smartphone display were green. When the participants did not achieve the specified breathing frequency, the symbols appeared in yellow (non-significant deviation), orange (significant deviation), or red (very significant deviation). The participants were instructed to adjust their breathing frequency to reach emergence of as many green symbols as possible for the entire training session.

Glucose metabolism was assessed through a frequently-sampled 75 g oral glucose tolerance test (OGTT) starting at 8:00 a.m. after an overnight fast. After basal blood sampling, the participants drank a 75 g glucose solution, with further blood samples taken at 30, 60, 90, and 120 minutes after glucose ingestion. Areas under the curve were calculated based on the trapezoid method.

Glucose stimulated insulin secretion was assessed as oral Disposition Index and as ratio of the areas under the C-peptide curves and the areas under the glucose curves during the first 30 minutes of the OGTT (AUC C-peptid0-30/AUC glucose0-30). For the estimation of insulin sensitivity, HOMA-IR was used for the fasting state and Matsuda Index [29] for the post-glucose load situation. Insulin clearance was estimated by the ratio of the areas under the C-peptide and insulin curves during the OGTT (AUC C-peptide0-120/AUC insulin0-120).

The daily profiles of the interstitial fluid glucose levels are recorded using continuous glucose measurement systems (FreeStyle Libre Pro iQ). These data enable the calculation of indices to assess glucose variability (standard deviation, coefficient of variation, MAGE index, MODD index) and glucose tolerance (daily mean values, AUC). Each measurement period lasted 14 days, as per the sensor lifespan.