Data assessment

During the study center visit, trained nurses carry out detailed anthropometric measurements according to standardized procedures, using regularly calibrated instruments [8, 9]. Quality controls of the measurements are regularly conducted.

For the measurements, participants are wearing underwear only as well as no shoes. Since 1985, height or length (to 0.1 cm), weight (to 100 g), upper arm circumference (to 0.1 cm), as well as four skinfolds (biceps, triceps, subscapular and iliac skinfolds, to 0.1 mm) are measured at each examination. Among children < 2 years of age, recumbent length is measured. From the age of 2 years onwards, standing height is measured. Both are assessed using a digital stadiometer (Harpenden, Crymych, UK). Body weight is measured using an infant weighing scale (until 08/2018: Mettler PS 15; Mettler Toledo, Columbus, OH; from 08/2018 onwards: Seca 757; Seca Weighing and Measuring System, Germany) as well as an electronic scale (Seca 920; from 02/2006 onwards Seca 701; Seca Weighing and Measuring System, Germany) among older participants. To estimate body composition, [10–12], four skinfolds (biceps, triceps, iliacal, and supradings) are measured on the right side of the body using a skinfold calliper (Holtain Ltd, Croswell, Dyfed, UK), duplicate measurements are taken since 2006. Head circumference (accurate to 0.1 cm) has been assessed at every examination since 1985, from the first visit to age 5 as well as at ages 18 and 21, using diameter measuring tapes (until 09/2012: Circumference Diameter Tape, Chasmors, London, UK; from 09/2012 onwards: Diameter Measuring Tape, Richter, Germany). In addition, since 1985, sitting height (to 0.1 cm) is recorded among participants ≥ 3 years at each visit using a sitting height measuring device (custom-made for the DONALD study).

Since 2005, from the age of 3 years onwards, hip (around the widest portion of the buttocks) and waist (at the midpoint between lower rib and iliac crest) circumference (to 0.1 cm) are assessed. Abdominal circumference (to 0.1 cm) is been measured since 1985 at every visit to the study center, from study entry until 2 years of age. Measurement of upper and lower leg circumference was performed at each visit/every age of the participants between 1985 and 2006. Since 2006, these measurements are taken only at the first visit as well as at the visits at age 5, 10, 15, and 18–25.

Until 2004, anthropometric measurements were taken every 6 months during puberty (girls: 8–14 years, boys: 10–16 years) to reflect rapid anthropometric changes during this period.

At each study center visit, trained physicians conduct a medical examination. Furthermore, information about the current and past (related to the time period since the last visit) health status of the participant as well as the age-appropriate physical development are collected. A detailed anamnesis interview inquired among others chronic diseases that are documented as ICD codes. Other examinations include measurements of blood pressure, intima-media thickness of the carotid artery and thyroid volume.

The measurement of systolic and diastolic blood pressure is usually performed at regular time intervals from the age of 4 (until 2004) or 6 (since October 2004) years onwards by trained nurses, according to standard procedures (participant in seated position with legs uncrossed, feet at the floor, using the right arm, arm supported at heart level) [13, 14]. Until 1994, a random zero sphygmomanometer (Random Zero Sphygmanometer 77,075, Hawksley & Sons, UK) is used for the measurements, whereas a standard mercury sphygmomanometer (Sphygmomanometer Mercury 300, Speidel & Keller, Germany) was used between 1994 and 2020. Different cuff sizes are used according to arm circumferences. At each measurement occasion, two consecutive blood pressure readings are taken after a rest of ~ 5 min. Since 2020, a digital sphygmomanometer (Mindray VS-600, Mindray, Germany) is used.

Since 2008, intima-media thickness is measured every 2 years in schoolchildren (11–17 years) as well as during adulthood. For this, vascular condition of the left and right common carotid artery is determined by high-resolution ultrasonography (Until 2016: Mindray DP3300, Mindray Bio‐Medical Electronics Co., Ltd., Shenzhen, China; since 2016: Sonoscape S11, Sonoscape Co. Ltd., Shenzhen, China). Measurements are taken after a 10-min resting period in participants in a supine position, head slightly to the right or left according to a standard procedure [15–17].

Since 2012, the thyroid volume has also been determined by ultrasonography under the same conditions as the intima-media examination. Data are collected from children, every 2 years between 7 and 17 years, as well as at each visit during adulthood. Thyroid-lobe height, width, and depth were measured. From these data, thyroid volume is calculated by the formula of Brunn et al. [18].

Paper-based 3-day weighed dietary records are used to collect information on food and nutrient intake from all participants from infancy to adulthood. To estimate breast milk quantities by test-weighing [19] during infancy, mothers are provided a baby scale (Soehnle multina 8300, Soehnle, Germany) to weigh the child before and after each feeding to the nearest 10 g [20].

From the start of complementary feeding onwards, all foods and beverages consumed by the participant, as well as leftovers, are weighed and recorded over 3 consecutive days to the nearest 1 g by the parents or, later on, by participants themselves. If there is no digital scale in the household, families are provided with a digital scale (currently WEDO Digi 2000, WEDO, Germany) and are instructed by a dietician on how to use it for weighing. If exact weighing is not possible, e.g., when eating out-of-home meals or snacks, household measures (e.g., spoons, cups) are allowed for semiquantitative recording.

The beginning of the 3-day dietary recording is chosen by the participant themselves usually within 8 weeks after the study center visit. Requested information includes name of the food or drink, date, time and location of eating, as well as recipes (ingredients and preparation). The participants are asked to describe the food as detailed as possible, e.g., to also record the fat content of dairy products or bread with iodized salt. For commercial food items, the packages or the food labels are collected by the participants and information on the packaging (name, manufacturer, brand, labeled nutrients, list of ingredients) are added to the dietary record data. Medication and dietary supplement use are also recorded. After the 3 days of recording, a trained dietician picks up all materials at the family’s home or picture of packaging are collected via an online messenger. The dietician crosschecks the dietary record using standard questions (e.g., on brand, manufacturer, homemade recipes, preparation method, type of table salt) in a short interview/phone call with the participant. Missing information can, thus, be added. In addition, it is inquired whether unusual events during the recording days might have changed the habitual dietary behavior.

All recorded food items are coded and linked to the continuously updated in-house nutrient database „LEBensmittelTABelle“ (food table, LEBTAB [21]). LEBTAB was developed during the course of the DONALD study and contains detailed data (energy and 72 nutrients; version III) on all basic (e.g., apple or milk) and commercial composite foods (i.e., infant formula, commercial complementary food as well as ready-to-eat-meals/-products, fortified food, food for dietetic purpose, and nutrient supplements), which were recorded by DONALD participants. Data entries for basic foods are based mainly on the German standard food table “Bundeslebensmittelschlüssel“ (BLS; version 3.02; https://www.blsdb.de/). Missing food items are complemented by data from other national food tables, e.g., by the U.S. Department of Agriculture (USDA) [22]. Energy and nutrient contents of composite foods are estimated by recipe simulation using labeled ingredients and declared nutrient contents taking into account nutrient fortification. In addition, food additives and flavorings have been recorded qualitatively in the estimated recipes since 2004.

Energy and nutrient data in LEBTAB refer to the raw food item. To take the loss of vitamins during food preparation into account, corresponding subtractions are made based on average nutrient losses during preparation as published by the German Nutrition Society [23]. The 3-day weighed records have been validated against the biomarker of 24-h urinary total nitrogen excretion [24], which showed that protein intake in children and adolescents can be estimated with acceptable validity by weighed dietary records.

Due to the longitudinal character of the DONALD study, recipe modifications or reformulations of commercial food products by the manufacturers are updated continuously. Therefore, each modified product receives a new entry in the nutrient database, while the entry for the product with the old recipe is marked and retained in the database. Until the end of 2022, LEBTAB contains more than 17,000 entries, i.e., ~ 1500 basic foods, ~ 15,700 composite foods (including ~ 4900 fortified products), and ~ 470 medication (e.g., antibiotics, analgesics) and supplements (e.g., nutrients, botanicals). In addition, as part of specific research projects, LEBTAB or at least parts of the included food codes, is regularly expanded to additional nutritional information (e.g., whole grain [25], glycemic index [25], isoflavones [26] or anthocyanin [27], or free sugar [28]).

At the first visit, each child’s birth characteristics are retrieved from the German standardized pregnancy document called “Mutterpass”. This includes information on pregnancy duration, type of birth (vaginal or cesarean), birth weight, birth size, head circumference of the child as well as the APGAR score at 5 and 10 min after birth [29]. In addition, a detailed interview on pregnancy is conducted with the mother. To be able to calculate weight gain during pregnancy, maternal body weight at the beginning and the end of pregnancy is inquired about. Furthermore, information on pregnancy symptoms and illnesses is collected, e.g., occurrence and duration of nausea as well as feeling of fullness/reflux and its consequences for eating behavior, iron deficiency, high blood pressure, diabetes or kidney diseases. Questions on maternal lifestyle during pregnancy concern smoking behavior before and during pregnancy, including number of cigarettes/day; alcohol consumption and whether, how long and until which month of pregnancy the mother did exercise. In addition, mothers are asked about their intake of dietary supplements during pregnancy. A detailed questionnaire on breastfeeding behavior is completed at all visits during infancy. Thus, data on the duration of full and partial breastfeeding are collected. In addition, mothers are asked whether they had planned to breastfeed before the birth of their child and the reason for stopping breastfeeding. Formula feeding as well as the start of complementary feeding are also documented.

In DONALD, a standardized questionnaire based on the Adolescent Physical Activity Recall Questionnaires [30] and questions from the German Health Interview and Examination Survey for Children and Adolescents (KiGGS) [31] is used to assess physical activity as well as sedentary behavior since 2004. Questions on physical activity include the duration and frequency of organized (e.g., club sport, gym) and unorganized sports (e.g., playing football with friends, cycling) as well as how the participant usually gets to school (e.g., by car, bus, bike or walking). Furthermore, duration and frequency of sedentary activities, i.e., watching television or reading a book, are asked.

Since June 2004, muscle function/physical performance tests have been used every 2 years from the age of 6 and on every occasion in adulthood with the help of a Ground Reaction Force Platform (GRFP; NOVOTEC Medical GmbH). In addition, since 2004 on the same occasions, muscle strength/grip strength is determined using a dynamometer (Jamar Jackson, MI, 49203 U.S.A.) on the non-dominant hand (participant in an upright-seated position and feet at the floor).

Since 2014, chronotype of DONALD participants during adolescence has been assessed with the Munich Chronotype Questionnaire (MCTQ). The MCTQ includes information on sleep duration and bed times, separately for free and for school/working days, and therefore allows to calculate chronotypes and social jetlag according to Roenneberg [32, 33].

At each visit, further personal interviews with participants’ parents and/or older participants themselves are conducted by the study pediatricians, including inquiring about data on childcare, use of vitamin D and fluoride supplements and preventive medical services. From the age of 12 onwards, participants are also regularly asked to complete questionnaires on smoking habits, intake of alcoholic beverages, and restrictive dietary behavior in absence of their parents.

From 3 or 4 years onwards, 24-h urine is collected once annually preferably on the third day of dietary recording. Parents as well as older children receive in-person and written instructions on how to collect complete 24-h urine samples according to a standardized procedure. The collection starts after voiding of the bladder upon getting up in the morning. This micturition is completely discarded. The time of this micturition is recorded and defines the start of the 24-h collection, which ends with the first micturition on the following day. During the collection period, the participants store micturitions immediately in preservative-free, Extran-cleaned (Extran, MA03; Merck, Darmstadt, Germany) 1-L plastic containers at less than − 12 °C before transfer to the study center. The exact times of micturition as well as possible comments on the collection are recorded on a predefined protocol sheet. The dietician picks up the 24-h urine samples together with the dietary record, checks plausibility of the collection protocol, inquires parents and/or participants about the completeness of the urine samples, and adds this information to the protocol sheet if necessary. After the transfer to the study center, the containers are stored at − 22 °C until analysis and subsequent biobanking. For routine laboratory measures and aliquotation, the samples are thawed and stirred. Urine volume, creatinine, urea, sodium, potassium, osmolality, and pH value are analyzed in the DONALD study laboratory. Subsequently, five aliquots of 20 ml each from each 24-h urine are stored at − 22 °C in the DONALD biobank for further research. Since the beginning of the DONALD study in 1985, over 11,000 24-h urine samples were collected and added to the biobank (Table 2).

Number of 3-day weighed dietary records, anthropometric measurements, 24-h urine and blood samples collected between 1985 and 2022 in the DONALD study, stratified by age and sex groups

Values are absolute and relative frequencies

¹Urine samples were collected from 3–4 years onwards

²Blood samples were collected among adult participants (≥ 18) only

Until 2019, for younger participants (0 − < 3 years), a spot urine is collected at the study center visit using a urine bag (Braun AG, Germany). After creatinine concentration measurements, samples were stored preservative free at − 22 °C in the urine biobank of the DONALD study (n = 1134).

Since 2005, participants over the age of 18 have been asked to provide a venous fasting blood sample. Blood samples are centrifuged directly at 4 °C within 15 min (3100 U/ min). The blood samples are routinely analyzed to determine among others blood cell levels (e.g., leucocytes, thrombocytes, erythrocytes), hemoglobin, hematocrit, liver enzymes (GOT, GPT, GGT), cholesterol and blood lipid levels (e.g., total cholesterol, LDL, HDL), triglycerides, blood sugar, iron levels and thyroid parameters (TSH, fT3, fT4). Aliquots of serum, plasma (citrate, EDTA) and buffy coat (citrate, EDTA) samples (each 500 µl) are stored at − 80 °C in the DONALD biobank for further research.

The study nurses collect maternal and/or paternal anthropometric (body weight and height, skinfolds, hip and waist circumference) as well as medical data (blood pressure) every 4 years using the same methodology and equipment as for the participants. In addition, parents are asked about previous chronic diseases (including hypertension, hyperlipidemia, diabetes, fatty liver disease), the intake of medication as well as diagnosed allergies. Furthermore, data on parental lifestyle (e.g., smoking in household, physical activity, specific dietary practices) and socioeconomic status (e.g., parental education and employment) are collected from both parents.

During adulthood (from the age of 21 onwards), the participants themselves are asked about their parents’ data, e.g., chronic diseases as well as if necessary, year of mortality and the cause of death.

In addition to the basic protocol of the DONALD study, further modules among subgroups are established within the framework of specific projects (Table S1). Methods are described in short as follows.

Stool samples were collected between 2017 and 2018 in the HEALTHMARK (metabolic HEALTH through nutrition, microbiota and tryptophan bioMARKers) project from adult participants (≥ 18 years) to be able to explore gut microbiome-related research questions. Finally, microbiome data based on 16S ribosomal RNA sequencing of n = 128 individuals (77 double and 51 single stool samples) are available [34]. Currently, i.e., since 2023, further stool samples have been collected from the adult participants as part of the PerMiCCion (PERsonalized MIcrobiome-Based Approaches to Early Onset Colorectal Cancer PreventION, Diagnosis and Management) project (http://www.permiccion.de/).

A variety of additional modules were made possible within the framework of the nutrition research competence cluster Diet Body Brain (DietBB; www.diet-body-brain.de), including the collection of genetics, data on fluid intelligence and personality-related characteristics accelerometer. Between 2015 and 2022, saliva samples were used for DNA extraction and one time collected among children and adolescents between 6 and 17 years (n = 170). Among adult participants, blood samples (buffy coats (n = 440) and PaxGene samples (n = 287)) are used to analyze DNA and RNA. These data allow the investigation of associations between nutrition, genetics, and health.

To investigate associations between personality-related characteristics and nutrition or anthropometry as part of DietBB (2016–2020), individual personality data were collected. For this, game-based experiments were conducted. Among 5–6 year olds (n = 126; until June 2016 also 4 year olds), delay of gratification was assessed, via Stanford marshmallow experiment [35]. To assess economic preferences, experiments on time, social, and risk preferences [36] were performed among 7–9 year olds (n = 168). In addition, validated questionnaires on personality by Decker et al. [36] were used, which include personality questions on the consumption of sweets (n = 125; age 5–6 years, until June 2016 also 4 year olds) or pocket money (n = 167; age 7–9 years).

Fluid intelligence was assessed using the Culture Fair Intelligence (CFT) 1-R (n = 89) [37] or the CFT 20-R test (8.5–18 years n = 139; > 18 years n = 99) [38] among a DONALD subgroup in 2017–2018. The cognition test was carried out once per participant. CFT 1-R was used among participants between 5 and < 8.5 years, CFT 20-R was used among participants from 8.5 years onwards. Both tests were developed language-free and figure-based to minimize cultural or social disadvantages or advantages [37, 38].

Accelerometer (CentrePoint Insight Watch from the company ActiGraph) data have been collected from 2020 until 2022 to determine physical activity and sleep characteristics (quality and duration) among participants between 6–18 years (n = 127) and > 18 years (n = 36).

In a further additional module in 1989–1999, a single peripheral quantitative computed tomography (pQCT) analysis of the forearm was undertaken in 371 participants aged 6–18 years. Details of these data collection and measurements can be found in numerous publications (among others: [39–41]).

Additional analyses of biosamples (urine and blood) are carried out within the framework of specific research projects, partly in cooperating laboratories. According to the respective research question, project-specific metabolic parameters were measured in urine samples among subgroups of the study. These include among others metabolomics (e.g., described in [42]), hippuric acid as biomarkers for flavonoid intake (e.g., described in [43]), fructose/sucrose excretion as biomarkers for total sugar intake (described in [44, 45]) and nutritionally and metabolically relevant hormone metabolites, e.g., adrenal androgens, glucocorticoids (described in [46, 47]). Similar to these urinary analyses, additional blood analyses were conducted within specific research projects, e.g., untargeted metabolomics or several serum (e.g., hormones (IGF-1, leptin), inflammatory markers (IL-18, E-selectin, IL1-ra, hs-CRP) and EDTA plasma parameters (e.g., hormones (insulin, adiponectin, fT3, fT4, TSH), inflammatory markers (IL-6, IL-18, hs-CRP, sICAM-1, omentin, fetuin-A), triglycerides, liver enzymes (GOT, GGT, GPT, ALT, AST)). Data from blood sample analyses are used primarily to determine metabolic health during adulthood, e.g., type II diabetes [3, 48], non-alcoholic fatty liver [4, 49] or cardiometabolic diseases [5, 50].