Study participants were recruited at Hannover Medical School, Dept. for Clinical Psychiatry, Social Psychiatry and Psychotherapy, Division of Clinical Psychology and Sexual Medicine, and at the Geneva Musical Minds Lab, Geneva School of Health Sciences, University of Applied Sciences and Arts Western Switzerland (HES-SO), Geneva, Switzerland, via newspaper advertisement and placards at public places. Demographics are given in the results part in Table 2. Prerequisites for participation in the study were overall good health, absence of psychiatric diseases, being right-handed, retired, and non-reliant on hearing aids.

The first time point for data assessment was before the weekly practical piano lessons were started (or theoretical music lessons without any practical exercises in the control group, respectively), the second time point was after half a year, and the third time point was 1 year after the first assessment (for the long-term observation of potential beneficial effects of piano-lessons) for every individual participant. The time points each cover a period of several months, as all the participants could not be examined on the same day, since MRI scans were also performed as part of the original study (T0: March to June 2019, T1: August 2019 to February 2020, T2: June 2020 to November 2020). A small subgroup of participants were recruited as early as in January 2019 (T0), and had their consecutive examinations in August 2019 (T1) and in the end of February/beginning of March 2020 (T2). In other words, their T2 was directly before the lockdown started. Therefore, this small subgroup was excluded from the main analyses. However, 55 patients were finally included in the present manuscript.

In both countries, Germany and Switzerland, lockdown measures were similar. It was forbidden to meet more than one other person except for the nuclear family. Major events were canceled as well as private marriages, and church services. It was not allowed to travel for vacation. Restaurants, cafes, bars, and discos were closed as well as barber shops, nail salons, and physiotherapy practices. Concerning retail, only supermarkets were kept open to ensure supplying of the population. Many companies stopped operating. Schools, most kindergartens, universities, libraries, swimming baths, and gyms closed down as well as all cultural facilities. Even playgrounds had been locked. Children were not allowed to meet other children and had to learn at home as school teaching was canceled for months. Most adults had to work in home office and for families, the situation at home became very dense. Furthermore, it was forbidden to visit hospitalized relatives or friends. Also, there was no possibility to visit old people in rest homes. Therefore, many old or sick people had to die lonely without any support from the ones they loved.

Genomic DNA (gDNA) for telomere measurements was isolated in the Institute of Molecular and Translational Therapeutic Strategies of Hannover Medical School according to standard procedures from 50 µL blood using the DNeasy Blood & Tissue Kit (Qiagen #69,506) and stored at − 20 °C. DNA samples were diluted in 96-well plates to a fixed concentration of 10 ng/µL. The blood samples were received anonymized and prior to gDNA isolation, the order of the samples was further randomized to minimize potential batch effects.

A quantitative polymerase chain reaction (qPCR)-based assay was performed to measure the relative telomere length as described before [7]. The ratio of the mean telomere (T) repeat sequence to a reference single copy (S) gene is calculated for each sample as previously described [8].

DNA samples were amplified in a total reaction volume of 10 µL containing 1 × iQ SYBR Green Supermix (Biorad #172-5006CUST), 2 µM ROX reference dye (Thermofischer #AB-4136/B) with either HPLC grade primer pairs for T or S amplification (see Table 1 below), and 20 ng of gDNA. For telomere repeat amplification, 300 nM of Human T FW primer and 900 nM of Human T RV primer were used. Whereas for S gene—36B4 primers were used for amplification for which the concentration of primers were as follows: 300 nM of 36B4 FW and 500 nM of 36B4 RV.

All DNA samples were run as triplicates on the qPCR (ABI Viia 7 Real-Time PCR system) for both the telomere (T) primer and 36B4 (S) primer, in 384-well format. The distribution of samples on the plates was randomized. The thermal cycling profile was the same for both primers: 95 °C incubation for 10 min followed by 35 cycles of 95 °C for 15 s, 54 °C for 2 min, and 72 °C for 15 s. Inter-run calibrators were included in each qPCR run comprising one gDNA sample with long telomeres (human-induced pluripotent stem cells) and another with short telomeres (human umbilical vein endothelial cells, passage 5). A standard curve was generated from a serially diluted reference DNA sample of human gDNA (Roche #11,691,112,001) for each of the qPCR runs to ensure primer and qPCR efficiency.

The final relative telomere length (T/S ratio) for each sample was the ratio of Telomere (T) and 36B4 (S) amplification which was calculated based on the qPCR efficiency and relative to the inter-run calibrators, as previously described [9].

The entire telomere length analysis was performed in a blinded, anonymized, and randomized experimental setup.

IBM SPSS Statistics (IBM, New York, NY), version 29.0, was used for the statistical analyses. For the analysis, a mixed linear model was calculated using a restricted estimated maximum likelihood approach to compare telomere length between different time points. The mixed linear model in SPSS can in general be used to develop different variance analysis models. It can deal with a lot of variables and covariates. The level of significance was set at p ≤ 0.05 throughout the analysis while correcting for multiple comparisons for the post hoc t-tests (Bonferroni). The telomere lengths were defined as the dependent variable and fixed factors were the time points of measurement. Age, sex, smoking behavior, BMI, nicotine consumption, and ethanol consumption were used as covariates. The covariates were used to exclude whether they have a significant influence on alterations in telomere length.

For correlation analyses of telomere lengths and BMI, nicotine consumption, alcohol consumption, and the WHOQOL-BREF Social Score, we used a bivariate correlation in SPSS.