Clinical information.

Comprehensive clinical information at the time of conception, including all aspects of the successful ART cycle, type of ART (IVF/ICSI, fresh/frozen), causes of a couples’ subfertility, as well as pregnancy outcomes were collected retrospectively from the two fertility clinics. Detailed embryological data of the index pregnancy, number and type of embryos transferred, along with details on media and duration of embryo incubation, type of ovarian stimulation used, and maximum level of maternal oestradiol were also collected.

Additional information regarding pregnancy (outcomes) was collected from the Midwives Notification System records, via data linkage through the Western Australia Department of Health. The obtained data was primarily used in case of missing information regarding pregnancy outcomes from the fertility clinics, as well as to cross check and validate variables, as fertility clinics do not always hold information regarding pregnancy outcomes.

Routine assessments of the GUHS adolescents and young adults.

Questionnaires. At all three follow-ups, the study participants answered their own questionnaires with questions appropriate for their age and with increasing complexity and detail. The questionnaire included general information regarding their physical health, self-reported pubertal development (Tanner staging), mental health, physical and social activity, relationships including family members’ support, eating, and drinking habits, education and/or occupation. At the 14-year follow-up this questionnaire included questions regarding smoking, alcohol and drug use. At the 17- and 20-year follow-up, these questions, as well as questions regarding relationships and sexual experience, were included in a separate confidential questionnaire. Information on a range of medical conditions was provided in a separate medical history questionnaire.

At the 14- and 17-year follow-up, the primary caregiver provided information on social determinants and quality of life of the family (housing, environment, primary caregiver health and wellbeing). A second primary caregiver questionnaire collected information on the participant’s physical activity, social behaviour, mental health, learning and education. Detailed medical questions were also included with emphasis on a wide range of doctor-diagnosed conditions, hospitalisation, injuries and the use of prescription and non-prescription medications, alongside a brief medical history of all family members.

At each follow-up, age-specific standardised questionnaires were completed. At all three follow-ups a standardised food-frequency questionnaire was completed measuring the dietary intake. This Dietary Questionnaire for Epidemiological Studies (DQES v.2.0) [20] was a modified version from the questionnaire developed by the Cancer Council Victoria [21]. At the 17-year follow-up participants completed the International Physical Activity Questionnaire (IPAQ) [22]. At the 14- and 17-year follow-up the Youth Self-Report [23], Cowen self-efficacy assessment [24] and Beck’s youth inventory [25] were administered. At the 20-year follow-up, the Depression, Anxiety and Stress Scale (DASS-21) [26], as well as the Autism-spectrum Quotient (AQ) [27] were administered. At the 14- and 17-year follow-up the primary caregiver completed the Child Behaviour Checklist (CBCL) [23].

All questionnaires are available as S2–S4 Files for the 14-, 17- and 20-year follow-up respectively.

Blood sample and urine analysis. Following informed consent, fasting venous blood samples were taken by a trained phlebotomist, in three different collection tubes containing the appropriate preserving agent for the specific measured analytes. Samples were delivered to the PathWest Laboratory in Perth, Western Australia, Australia, within two hours of phlebotomy for immediate analysis. The remaining aliquots of serum, plasma and urine samples have been appropriately archived at -80°C for long term storage.

The following analyses were performed: glucose homeostasis, including a calculation of the homeostatic model assessment for insulin resistance (HOMA-IR) [Fasting insulin (mU/L) X (fasting glucose (mmol/L)/22.5)] [28]; thyroid, kidney and liver function tests; iron studies; lipids and cardiac measurements, Vitamin D, fatty acids, as well as plasma and urine spot analyses for kidney function. The detailed methodology and a full list of analyses is presented in S1 Table.

Anthropometric assessments.

Quantitative anthropometric measurements were collected routinely at all three follow-ups to assess body composition.

Standing height was measured using a stadiometer (to the nearest 0.1 cm) (Wall mounted Holtain stadiometer, United Kingdom) and weight was measured seated, with participants lightly clothed, using an electronic scale (to the nearest 0.1 kg) (Wedderburn Chair Scales, Australia). Body mass index (BMI) was calculated (kg/m²).

To improve accuracy of the measurements, the following parameters: waist, hip, and wrist circumference, biacromial and anterior superior iliac spine breadth were measured twice (to the nearest 0.1cm), using a Birch plastic measuring tape, and the average of both measurements was recorded (Birch plastic measuring tape, Australia). Waist to hip ratio was calculated (average waist circumference/average hip circumference, to the nearest 0.1 cm). Similarly, skinfold thickness was measured twice using a calliper, and the average of both measurements was recorded (triceps, subscapular, biceps, mid-abdominal and suprailiac skinfolds, to the nearest 0.1 mm) (Holtain skin fold Callipers, United Kingdom).

Systolic and diastolic blood pressure measurements were taken in a supine position, on the right arm, following 5 minutes rest, using DinaMap ProCare 100 oscillometric sphygmomanometer with appropriate cuff size. Furthermore, all age groups underwent a cardiovascular endurance test to estimate physical work capacity (PWC170) using a Monark cycle ergometer and the Australian Education Fitness Award test (AFEA) encompassing four physical activities: curl ups, shoulder stretches, sit and reach and basketball throws. Weekly pedometer logs were completed. A photocopy of both hands was taken, and finger length for the second and fourth digit, as well as 2^nd to 4^th digit ratio was recorded.

At the 14 and 17-year follow-up, the psychomotor skills of the participants were assessed by administering the standardised quantitative McCarron Assessment of Neuromuscular Development tool (MAND) [33].

14-year follow-up (adolescents aged 13–15.9).

At the 14-year follow-up, the assessments of particular interest were: lung function, bronchial responsiveness and allergy profiles.

Lung function and bronchial hyperresponsiveness test. Lung function was assessed by base spirometry [34], using an electronic spirometer (KoKo Spirometer, nSpire Health Ltd. USA), to asses airway function by calculating standard spirometry variables, and to define the suitability of the participants to perform the bronchial hyperresponsiveness test. Bronchial hyperresponsiveness was assessed by administering increasing doses of Methacholine, with the forced expiratory volume (FEV1) measured within five minutes of administration, following the American Thoracic Society guidelines for Methacholine and Exercise Challenge Testing (KoKo Dosimeter, nSpire Health Ltd USA) [35]. Testing was continued until either a drop in FEV1 ≥20% from baseline assessment occurred, or maximum cumulative dose was reached (16 mg/ml). If a drop in FEV1 ≥20% occurred, 400 mcg of salbutamol was given via spacer device under medical supervision. The provocative concentration causing the 20% fall in FEV1 was calculated according to the following formula:

Where:

C1 = second last concentration (<20% FEV1 fall)

C2 = last concentration (>20% FEV1 fall)

R1 = % fall FEV1 after C1

R2 = % fall FEV1 after C2

Skin prick testing for allergies. Standard skin prick testing was completed using the following panel of allergens: egg white, 7 grass mix and grass pollen (perennial ryegrass), cat and dog hair, cow’s milk (whole), cockroach, fungus (Aspergillus fumigatus), mould (Alternaria alternata), and dust mite (Dermatophagoides farina and pteronyssinus), positive controls (histamine hydrochloride 10 mg/ml) and negative controls (glycerine/saline). In case of a reaction the size of the wheal and flare (mm) were recorded. Positive control size was recorded 10 minutes after administration, and size of allergens and negative control 15 minutes after administration. Participants were instructed to cease the use of antihistamines 24h prior to the administered test and were questioned to verify medication withholding on the day of testing. Exclusion criteria were as follows: severe reactions to respiratory irritants (smoke, fumes, perfumes etc.), diffuse dermatological conditions (the test was only performed on healthy skin) and inability to cease antihistamines.

In addition to the physical measurements, questions with emphasis on participants and familial asthma and allergy medical history, as well as a modified version of the standardised ‘International Studies of Asthma and Allergies in Childhood’ (ISAAC) questionnaire were completed by the primary caregiver.

17-year follow-up (adolescents aged 16–18.9). At the 17-year follow-up cardiometabolic health and mental health were assessed.

Abdominal ultrasonography. Abdominal ultrasonography was conducted for investigation of the presence and severity of hepatic steatosis and measurement of abdominal adipose tissue (mm) (subcutaneous, visceral and pre-peritoneal), using a Siemens Antares ultrasound machine with a CH 6–2 curved array probe (Sequoia, Siemens Medical Solutions, USA). Trained sonographers performed focused liver ultrasonography in accordance with the protocol described by Hamaguchi et al., which provides 92% sensitivity and 100% specificity for the histological diagnosis of hepatic steatosis [36].

A single specialist radiologist, who was blinded to the clinical and laboratory characteristics of the subjects, interpreted the ultrasound images. Scores of 0 to 3, 0 to 2 and 0 to 1 were determined from captured images for liver echotexture (bright liver and hepatorenal echo contrast), deep attenuation (diaphragm visibility) and vessel blurring (intrahepatic vessel visibility). The diagnosis of hepatic steatosis required a total score of ≥ 2, inclusive of echotexture score of ≥1. The severity of hepatic steatosis was classified by the total fatty liver score as 0–1 (no fatty liver), 2–3 (mild fatty liver), or 4–6 (moderate to severe fatty liver). Adolescents with sonographic fatty liver, the absence of self-reported long-term use of medication affecting the liver, and a self-reported weekly alcohol intake of < 140 g for males and < 70 g for females over the previous year were classified as having non-alcoholic fatty liver disease.

Abdominal (subcutaneous, visceral and pre-peritoneal) adipose thickness measurements were performed with previously described criteria that correlate closely with compartmental adipose areas and cardiovascular and metabolic risk factors [37–39].

Measures of blood pressure and arterial stiffness. Pulse wave velocity (m/s) and heart rate corrected augmentation index (%), as measures of arterial stiffness, as well as systolic and diastolic blood pressure (mmHg) and resting heart rate (bpm) were measured using SphygmoCor assessment (AtCor Medical Pulse Wave Analysis System SCOR-Px, Australia) [40]. Systolic and diastolic blood pressure were measured by oscillometric sphygmomanometer (DINAMAP vital signs monitor 8100, DINAMAP XL vital signs monitor or DINAMAP ProCare 100), after resting 5 minutes and using the appropriate cuff size. Six readings were taken in a supine position, every 2 minutes for 10 minutes. The average value was calculated after excluding the first reading. After 5 minutes rest, the six blood pressure measurements were recorded, and three electrocardiogram leads attached to left leg, right arm and left arm. Tonometers were applied to two sites (the carotid artery and the distal dorsalis pedis). Distance measurement was recorded in millimetres between the manubrium sternum and the two sites. The pulse wave analysis was recorded from the supported radial artery with the wrist facing up. Data were entered after the waveform was maintained for 10 seconds and the test was repeated until at least two captures were recorded with a quality index of >80. Pulse wave velocity was calculated by dividing the distance between tonometers by the transit time of the arterial pulse wave. Heart rate corrected augmentation index was defined as the difference in the second and first systolic pressure peaks as a percentage of pulse pressure and was corrected for heart rate.

In addition to the physical measurements, the participant questionnaire included a set of questions with emphasis on familial cardiometabolic and cardiovascular medical history.

Cognitive function. Measures on a range of elements of cognitive performance were collected from a sensitive and reliable computerized battery of tests, using the digital cognitive testing system via the downloaded Cogstate Research™ software (CogState Limited, USA) (www.cogstate.com) [41,42]. The following tests were administered to each participant following adequate familiarization through practice sessions: The Groton Maze Learning Test, assessing the total number of errors made in attempting to learn the same hidden pathway on five consecutive trials at a single session; The Groton Maze Learning Test Recall, assessing the total number of errors made in remembering the maze pathway after a delay; Card Detection and Card Identification Tests, both measuring speed performance; One Card Learning Test, assessing accuracy of performance; Continuous Paired Associate Learning Test, assessing accurate memory patterns; total number of errors across the five rounds.

20-year follow-up (young adults aged 20–22.9).

The 20-year follow-up investigated vision and body composition.

Vision. An accurate measurement of the refractive status of the participants’ eyes, as well as a measurement of the corneal curvature was attained using the NIDEK ARK-1 Auto ref/keratometer (NIDEK CO., LTD. Tokyo, Japan). On the day of the assessment, the participants were asked not to wear contact lenses or for them to be removed >20 minutes prior to the autorefraction test. If the participants wore glasses, they were asked to remove them right before testing. Trained research assistants operated the autorefractor according to manufacturer’s recommendation, and the following measurements were taken from both eyes: spherical refractive error, cylindrical refractive error, cylinder axis, pupillary distance, corneal curvature radius and axis angle in the steepest and flattest meridian direction.

In addition to the physical measurements, the participant questionnaire included a set of questions with emphasis on participants and familial eye-health medical history.

Body composition. The bone mineral content and body composition including body fat percentage and skeletal integrity was assessed from a dual energy X-ray absorptiometry scan (DEXA). DEXA is considered the preferred method for measuring bone mineral density and soft tissue composition [43]. Each participant completed one total body scan and standard measurements were recorded. The Norland XR-36 Quick scan machine (Norland Medical Systems, Inc., Fort Atkinson, WI, USA), that utilizes a very small dose (<10 microSieverts) ionising radiation was operated by a trained Research Assistants. Due to the use of ionising radiation, approval was obtained from the Radiation Safety Officer at the University of Western Australia.