The Gothenburg H70 Birth cohort study 2014–16: design, methods and study population

Blood sampling was performed by a research nurse in 1192 individuals (555 men, 637 women; response rate 99.1%) after overnight fasting. Fasting status was confirmed by the nurse before sampling. The venipuncture was performed in the inner elbow while the participant was in an upright sitting position. Blood sampling could not be performed in 11 individuals (due to e.g. small or deep veins). A maximum of 120 ml blood was drawn, of which 15 ml was used for blood tests. A maximum of 105 ml was aliquoted into serum (after 20–30 min of coagulation, 5 ml SST tubes were centrifuged with 2000 RCF (Relative Centrifugal Force) for 10 min and then pipetted into 1.5 ml micro tubes), plasma (10 ml EDTA tubes were centrifuged with 2000 RCF (Relative Centrifugal Force) in 20 °C for 15 min and then pipetted into 1.5 ml micro tubes) and whole blood (10 ml EDTA tubes pipetted into 1.5 ml micro tubes) and were frozen at –80 °C within approximately one hour to be saved in a biobank for future analyses according to the Swedish Biobank Law (2002:297) (biobank registration date: 2005-03-17; Registration Number: 532 at the National Board of Health and Welfare). All study participants gave written consent regarding biobanking. Blood tests included thrombocytes, hemoglobin (Hb), glucose, thyroid function (TSH, T4), kidney function (creatinine), liver function (AST, ALT), lipids (total-, HDL- and LDL-cholesterol, triglycerides) and indirect measurement of vitamin B12 and folic acid status (homocysteine). Name of the assay/reagent and instrument manufacturer for each measure can be found in Supplementary 1.

Due to a limited amount of blood sampled, DNA extraction was impossible to perform for 25 of the total 1192 individuals with blood samples. DNA was thus extracted from 1165 individual samples of whole blood (548 men, 617 women; response rate 97.7%). The extraction of DNA from whole blood was performed according to standard procedures at LGC Genomics in Berlin (Germany). All the DNA samples have been genotyped at University College London (UK), using the Neuro Consortium Array (neurox2) from Illumina (www.​illumina.​com). Additional single nucleotide polymorphisms (SNPs) in genes of specific interest have been, and will be, analyzed at LGC Genomics in Hoddesdon (UK), using the KASP genotyping technology. The participants and a close informant were asked about family history of dementia, depression, stroke, diabetes mellitus and cardiovascular disorders, as described below.

A psychiatric examination was performed in 1194 individuals (555 men, 639 women; response rate 99.3%). Non-participants (n = 9) did not perform the examination due to several reasons (e.g. declining participation, aphasia or not speaking Swedish well enough to respond to the psychiatric questions). The semi-structured interview was conducted by a psychiatric research nurse, a psychiatrist, or a medical doctor, with a mean duration of 30 min. Psychiatric signs and symptoms during the past month were rated according to the Comprehensive Psychopathological Rating Scale (CPRS) [16], and the Mini-International Neuropsychiatric Interview (MINI) [17]. The Montgomery-Åsberg Depression Rating Scale (MADRS) [18] was derived from the CPRS to assess the overall burden of depressive symptoms. Both CPRS and MADRS have been found to be valid and reliable in older populations [19, 20]. The CPRS subscale, The Brief Scale for Anxiety (BSA) [21], was used to assess overall burden of anxiety symptoms. Mental and social functioning was assessed with the Global Assessment of Functioning (GAF) scale [22]. Questions about suicidal feelings were asked according to Paykel et al. [23]. The interview also comprised questions regarding symptoms of ADHD during early life and childhood [24], history of previous mental disorders; periods of hypomania according to Angst [25], the Yale-Brown Obsessive‐Compulsive Scale [26], symptoms of depression and anxiety (panic, general anxiety, agoraphobia), specific phobias, social anxiety disorder, post-traumatic stress syndrome, psychotic symptoms, thoughts about death and suicide, feelings of loneliness, periods of stress in relation to family and work, cognitive symptoms, appetite and weight loss, and sleep. The examiner also rated global severity of mental health problems and a number of observed psychiatric symptoms. All diagnoses for psychiatric disorders (e.g. depression, generalized anxiety, specific and social phobia, bipolar disorder, psychosis, obsessive–compulsive disorder, and traumatic stress disorder) followed the Diagnostic and Statistical Manual of Mental Disorders, DSM-IV [27], DSM-IV-TR [28], and DSM-5 [29] criteria as closely as possible. Questions regarding mental health were also asked in self-rating questionnaires (see Table 3: GAF; Phobias; PSS-14).

A clinical cognitive examination was performed in 1196 individuals (557 men, 639 women; response rate 99.4%). The examination was conducted by a psychiatric nurse, a psychiatrist, or a medical doctor, with a total duration of approximately 25 min. The examination comprised tests of word fluency (animals), naming 12 objects, copying five figures, finger agnosia, orientation, selective attention, understanding proverbs, ideational apraxia, memory (remembering 12 objects, word memory list, two last prime ministers), ability to follow instructions. Cognitive function was rated in accordance with a Swedish version of the Mini-Mental State Examination (MMSE) [30], the Gottfries Bråne Steen-scale (GBS) [31], as well as global ratings. Specific assessments relevant for dementia diagnosis, such as recent and remote memory, semantic memory, concentration, apraxia, spoken language, language comprehension, word finding difficulties, and other intellectual functions were performed, as well as assessments with the Clinical Dementia Rating scale (CDR) [32], and Alzheimer’s Disease Assessment Scale-Cognitive (ADAS-COG) [33]. Dementia was diagnosed following the Diagnostic and Statistical Manual of Mental Disorders DSM III-R [34] criteria as closely as possible using information from the clinical cognition examination and information from key informants (see below). We also assessed the neurocognitive domains described in DSM-5 (complex attention, executive function, learning and memory, language, perceptual-motor, and social cognition) [29].

A more detailed psychometric cognitive examination that included questions about subjective memory as well as testing of multiple cognitive domains was performed in 1173 individuals (547 men, 626 women; response rate 97.5%). The examination was conducted by research staff members, all trained by a psychologist, with a total duration of approximately 60 min. The subjective part included questions concerning participant’s everyday memory, experiences of memory decline, satisfaction with own memory, everyday memory problems, memory training, and engagement in cognitively demanding recreational activities, such as chess, bridge, crossword-puzzles, and reading habits. The cognitive test battery included several cognitive measurements intended to cover a broad range of different cognitive abilities. Memory in Reality (MIR) [35], Supra-span memory test (BUS II) [36], and Thurstone’s Picture Memory [37] measured memory abilities. Digit Span Forward and Backward test [38] measured short term memory and executive functioning. Figure Identification—Psif [39] measured mental speed, and Figure Logic (SRB 2) [39] measured inductive reasoning. Controlled Oral Word Association‐FAS (COWA-FAS) [40] measured verbal fluency and Block Design (Koh’s Block Test) [38] measured spatial ability. The Clock Test [41] was originally developed for dementia screening and measured a wide range of cognitive abilities including executive functioning [42]. Most of the tests are part of the Dureman and Sälde (1959) psychometric test battery [43] that was widely used in Sweden at the start of the H70 study 1971–72.

A general health interview was performed in 1199 individuals (558 men, 641 women; response rate 99.7%), with a total duration of approximately 30 min. The interview was conducted by a research nurse or medical doctor, and comprised questions about health care, self-rated health, pain and fatigue, weight at age 20 and 50, current and past health issues (e.g. stroke/TIA, cancer, and a number of cardiovascular, respiratory, metabolic, endocrine, neurological, musculoskeletal, gastrointestinal, kidney, genitourinary, blood, eye and ear diseases) and treatment history. It also included questions about specific symptoms, e.g. chest pain according to the Rose angina questionnaire [44], breathlessness, dizziness/vertigo, headache, joint and low-back pain, falls and fractures, head trauma, dental health, hearing and eye sight, urinary and fecal incontinence, sexual behavior, age of menarche and menopause, hormonal and contraceptives use, and alcohol, drug and tobacco use.

Questions regarding self-rated health, substance use, and factors associated with physical health were also asked in self-rating questionnaires (see Table 3: SF-36, AUDIT and DUDIT).

A total of 1200 individuals (559 men, 641 women; response rate 99.8%) reported their medications, including dosages and treatment indications. Participants were also asked to bring their medication lists. The medications were classified according to the Anatomical Therapeutic Chemical (ATC) classification system recommended by the WHO [45].

Physical examinations were carried out by research nurses in 1200 individuals (559 men, 641 women; response rate 99.8%), with a total duration of approximately 45 min. The examination was conducted by a research nurse, and comprised measures of length, weight, waist and hip circumference, ankle pressure (Minidop ES-100VX Hadeco), and pulse frequency. Systolic and diastolic blood pressure were recorded in the right arm in the sitting position after 5 min rest using a standard cuff (Umedico). Blood pressure was also recorded in the supine position after 5 min rest, and after one and 3 min in the standing position. The neurological examination included the Romberg test [46], the Grasset test, and evaluation of hypokinesia, rigidity, tremor, and a number of other motor disturbances. Electrocardiography (ECG) was performed in 1193 individuals (556 men, 637 women; response rate 99.2%) (SCHILLER AT-2 plus or SCHILLER AT-102 plus at the outpatient clinic, and SHILLER AT-101 during home visits). Non-participants (n = 10) did not perform the examination due to several reasons (e.g. declining participation or technical reasons). The ECG was coded according to the Minnesota code system [47].

Bioelectrical impedance was performed in 1147 individuals (533 men, 614 women; response rate 95.3%). Non-participants (n = 56) did not perform the examination due to several reasons (e.g. having pace-maker or wheelchair). The examination was used to measure individual body composition (Body Composition Analyzer TANITA T7063 09010007 at the outpatient clinic, and TANITA T6360 07060054 during home visits). Participants were also asked to take part in a more comprehensive body composition examination (see below).

A total of 865 individuals (412 men, 453 women; response rate 71.9%) performed a lung function examination. This comprised spirometry (Spirare SPS310 sensor and Spirare 3 software, Diagnostica AS, Oslo, Norway) without bronchodilator administration according to the American Thoracic Society/European Respiratory Society standards [48]. Maximum forced volume capacity (FVC), forced expiratory volume in 1 s (FEV1), peak expiratory flow (PEF), forced expiratory flow (FEF25, 50, 75), maximal (mid-) expiratory flow (MMEF), slow vital capacity (SVC), and forced expiratory time (FET) were measured.

As in previous H70 studies, FVC and FEV1 were also measured with a Micro Medical portable spirometer (Micro Medical Ltd, Rochester, Kent, UK) [49], and peak expiratory flow (PEF) was recorded with a Mini-Wright Peak Flow Meter (Clement Clarke International Ltd, Harlow, Essex, UK) [50], where participants were asked to exhale with a maximally forced effort from a position of maximal inspiration. A total of 1098 (515 men, 583 women; response rate 91.3%) individuals performed the Micro Medical portable spirometer test (FVC, FEV1), and 1099 (518 men, 581 women; response rate 91.4%) performed the Mini-Wright Peak Flow Meter test (PEF). Each participant performed the PEF test three times and the highest value was used as the final result. Of those who did not perform the lung function examinations, several reasons were reported (e.g. technical reasons or physically unable to take the test). Only the Micro Medical portable spirometer and Mini-Wright Peak Flow Meter could be performed during home visits.

An audiological examination was performed in 1116 individuals (523 men, 593 women; response rate 92.8%), with a total duration of approximately 30 min. Non-participants (n = 87) did not perform the examination due to several reasons (e.g. declining participation or technical reasons). The examination could not be performed during home visits. The audiological examination was conducted by a research nurse in a quiet office setting, and comprised pure tone audiometry, which measures the sensitivity of the ear to tones of differing frequency; wide band tympanometry, measuring the condition of the middle ear and the mobility of the ear drum and the construction bones; and otoscopy. Audiometry (air-conduction only) was carried out with an automated method using Sennheiser HDA200 headphones. The examination maps hearing loss caused by lesions and diseases in the middle ear and the cochlear nerve. This type of audiometry is gold standard for clinical practice and for epidemiological purposes. Questions were also asked about perceived hearing disability, hearing aid use and tinnitus. A subsample performed a more comprehensive audiological examination (see below).

Near vision acuity was tested with a Jaeger eye chart in 1181 individuals (549 men, 632 women; response rate 98.2%) with the participants’ own reading glasses if any. The general examination also included questions about ocular morbidity and vision. Questions regarding visual-related quality of life were also asked in self-rating questionnaires (see Table 3: VFQ-25). A subsample performed a more comprehensive ophthalmic examination (see below).

An interview assessing functional ability and disability was performed in 1197 individuals (558 men, 639 women; response rate 99.5%). The interview was conducted by a research nurse, with a total duration of approximately 20 min. Difficulty in self-care tasks or activities of daily living (ADL) was either self-assessed, based on observations by the research nurse, or from information from staff and carers. Several instruments were used. The Barthel Index of Activities of Daily Living [51‐53] included ten domains of function (bowels, bladder, grooming, toilet use, feeding, transfer, mobility, dressing, stairs, and bathing). Response options were either independent or dependent/unable. A summary score ranged from zero (low function, dependent) to 100 (high function, independent). The Katz Index of Independence in ADL [54] included six domains of function (bathing, dressing, toileting, transferring from bed to chair, continence, and feeding), with response options independent or dependent. A summary score ranged from zero (high function/independent) to six (low function/dependent). Independent living skills or instrumental activities of daily living (IADL) were measured with the Lawton Instrumental Activities of Daily Living Scale (IADL) [55]. The scale provides information about functional skills necessary to live in the community, and includes eight domains of function (ability to use the telephone, shopping, food preparation, housekeeping, laundry, mode of transportation, responsibility of own medications, and the ability to handle finances). Response options ranged from fully independent to completely unable/incapable. A summary score ranged from zero (low function, dependent) to eight (high function, independent). The examination also included assessment of a number of other functions used in previous H70 studies. The “ADL staircase” included a mix of questions from the Katz and Lawton scales with response options independent/dependent. Questions were also asked about use of assistive devices, ability to pick up things from the floor, ability to move around indoors and outdoors with and without assistive devices, past year travel abroad, car driving, and use of community health care, services, and personal care needs including help from family and friends. Questions regarding helping others with daily activities were also asked in self-rating questionnaires (see Table 3: Relationships).

An examination of fitness and physical activity was performed in 1173 individuals (545 men, 628 women; response rate 97.5%), with a total duration of approximately 30 min. A physiotherapist performed the examination, which included fitness tests and an interview regarding physical activity. Self-selected and maximum gait speed [56] for 30-meter indoors with standing start were measured (meter/second), as well as achieved walking distance (meters) during a six-minutes indoor walking test [57]. Pulse and oxygen uptake were measured at four time points with a pulse oximeter (NONIN Onyx II model 9550 oximeter). Grip strength was tested with a Martin Vigorimeter [58] and a JAMAR dynamometer (sub-sample) [59], with the shoulder joint in a neutral position. The test was repeated three times for each hand, and the highest value of the best hand was used as outcome. Timed chair-stand [60] tested the ability to stand up and sit down from a chair (height 43 cm) five times in a row as quickly as possible (five-repetition sit-to-stand-test) [61]. The total time (in seconds) was used as outcome. One-leg static balance [62] tested the ability to stand on one leg without shoes, for a maximum of 30 s, with eyes open. The test was interrupted if the individual moved from the standardized position and three trials for each leg were allowed. A dynamic balance test [63, 64] was also performed, in which the study participants were instructed to walk along the lines of a floor-painted “8” two times. Both time (in seconds) and number of inaccurate steps (counting left and right) were registered. A stair climbing test was conducted, as described in [65], testing the ability to climb onto a box of 50 cm (or 40, 30, 20 to 10 cm, if unable to climb the box of 50 cm) with or without support using a handle attached to the wall. Physical activity was assessed by an interview including the 6-grade physical activity scale [66] and study specific questions regarding walking habits and types of physical activities. The examination also comprised questions about walking habits, self-rated fitness, level of physical activity at certain ages (lifetime), participation in competitive sports (lifetime), and type of activities performed during the past 12 months. Questions regarding physical activity were also asked in a self-rating questionnaire (see Table 3: IPAQ).

An interview regarding social factors was performed in 1198 individuals (558 men, 640 women; response rate 99.6%), with a total duration of approximately 30 min. The interview was conducted by a research nurse, and comprised questions about past and present marital status and marital satisfaction, relationships with friends, partner’s physical and mental health, and need of assistance, housing and neighborhood, noise in surrounding, place of birth and family origin, family constellation and number of children, grandchildren, and siblings, and death of children, grandchildren, siblings or friends. Questions were also asked about serious illness and social problems in children and grandchildren, conflict with children or grandchildren and time spent helping them in everyday life, parents’ age at birth of participant, parents’ age at death, and own age at death of parent. Age at time of move to Gothenburg was recorded for those not born in the city. Study participants were also asked about childhood conditions, education, past and present working conditions, retirement and reaction to retirement transition, and partner’s work or retirement status, and current feelings of loneliness. Questions regarding social factors were also asked in self-rating questionnaires (see Table 3: Activities, Relationships, Income, and Work), which included questions about e.g. hobbies, leisure time activities, and use of internet and computers.

All participants were asked to answer 19 self-rating questionnaires. Number of participants answering each questionnaire can be seen in Table 3. The self-rating questionnaires could be filled out during the day of the general examinations or at home. In the latter case, questionnaires were sent back by mail. A continuous quality assurance comprising completion of unanswered questions was made by the research staff by telephone and letter. A total of 47 participants (23 men, 24 women) declined to answer all questionnaires (due to cognitive decline, impaired vision, poor health, or unwillingness to answer).

Following the general examination, an overall quantitative rating of medical burden was made using the Cumulative Illness Rating Scale for Geriatrics (CIRS-G) [67, 68]. Medical problems were rated by organ system (heart, vascular, hematopoietic, respiratory, eyes, ears, nose, throat and larynx, upper gastrointestinal tract, lower gastrointestinal tract, liver, renal, genitourinary, musculoskeletal and skin, neurological, endocrine, metabolic and breast, and psychiatric illness) with a scale ranging from 0 to 4, from no problems to extremely severe.

All participants were asked if they would like to supply contact information for a close informant for a collateral interview by research staff. If the participant was not able to provide this information due to dementia, a close informant was sought in other ways (e.g. through staff at residential care facilities). A close informant interview was performed by telephone for 973 individuals (470 men, 503 women; response rate 80.9%). Among those not having a close informant interview (n = 230), several reasons were reported (e.g. did not have any close informants to contact, close informants declined to participate). The interview was conducted by a research nurse or psychologist, during a mean of 45 min. The close informant was a spouse or partner in 606 cases, a daughter in 176 cases, a son in 79 cases, and others (e.g. friend, grandchild, or residential care staff) in 53 cases. The mean interval between the general examination and the close informant interview was 209 days. The close informant interview was semi-structured, and included the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) [69], which assesses changes in everyday cognitive function. The interview also comprised questions regarding cognitive symptoms and functional abilities (e.g. memory, spatial orientation, language, general intellectual symptoms, eating habits, incontinence, and a number of activities of daily living), as well as psychiatric symptoms (e.g. personality changes, irritability and aggressive behavior, motivation, emotions, depressive, anxiety, obsessive compulsive, psychotic and other psychiatric symptoms), and age at onset, course and cause of symptoms or disabilities. The interview also included questions regarding a number of background factors (e.g. education, smoking, snoring, family situation, and life events), history of diseases (e.g. cerebrovascular and cardiovascular diseases, dementia, head trauma, alcohol abuse, infectious diseases, deficiency states, hypo- and hyperthyroidism, chronic bronchitis and asthma, diabetes mellitus, Parkinson’s disease, normal pressure hydrocephalus, hip fracture), and family history of diseases (dementia, depression, stroke, cardiovascular diseases, diabetes mellitus). In cases of dementia, questions were also asked regarding course and age at onset.

All study participants without dementia were invited to take part in a dietary examination. The examination was performed in 861 individuals (385 men, 476 women; response rate 71.6%) with a total duration of approximately 60–90 min. Non-participants (n = 342) did not perform the examination due to several reasons (e.g. declining participation, poor health, lack of time or no response). A dietician performed a semi-structured interview capturing habitual dietary intake (food and beverages) during the past 3 months. The interview, which was conducted at the outpatient clinic or during a home visit, included not only open-ended questions about usual food patterns, but also structured questions in order to capture total intake as closely as possible. Dietary intake, defined as energy and nutrient intake, and food pattern were measured by the diet history (DH) method described elsewhere [70, 71]. Data from DH interviews was processed using the Swedish National Food Agency’s nutrient database (The Swedish Food Composition Database), to estimate energy and nutrient intake.

All study participants were invited to take part in a body composition examination. The examination was performed in 993 individuals (402 men, 591 women; response rate 82.5%) with total duration of approximately 20 min. Non-participants (n = 210) did not perform the examination due to several reasons (e.g. declining participation, or technical reasons or no contact). The body composition examination was conducted by a nurse at Sahlgrenska University Hospital, where dual energy X-ray absorptiometry (DXA) was measured using a Lunar Prodigy scanner (Scanex, Helsingborg, Sweden) [72]. A full body scan analyzed the amount of body fat, lean soft tissue, and bone mineral content. Fat-free mass was defined as the sum of lean soft tissue and bone mineral content. Appendicular lean soft tissue was defined as the sum of lean soft tissue in arms and legs. Body composition was measured by bioelectrical impedance spectroscopy (BIS) using an ImpediMed SFB7 device. BIS measures body composition based on the principle that tissues in the body conduct electric current. BIS calculated total body water and the amount and distribution of intracellular water and extracellular water via electrodes on hand/wrist and foot/ankle. Thus, fat free mass was estimated. Resistance and reactance measurements were made over a wide range of frequencies [73, 74]. Manufacturer’s proprietary software was used to calculate Cole Plot parameters and body composition estimates.

All study participants were invited to take part in a brain imaging examination, conducted at the Aleris Clinic in Gothenburg. Participants underwent both computed tomography (CT, approximately 5 min) and magnetic resonance imaging (MRI, approximately 40 min) during the same session. Number of participants and the acquisition parameters for CT and MRI can be observed in Table 4. The CT examination was performed in 915 individuals (response rate 76.1%), and the MRI examination in 791 individuals (response rate 65.8%). Participants were interviewed for safety procedures (e.g. incompatible implant devices such as pacemakers) and informed consent was obtained. Non-participants did not perform the examination due to several reasons (e.g. declining participation, no response or expressing fear or discomfort). For those willing to undergo MRI, five could not participate due to contraindications (i.e. pace-makers). The examination was interrupted on 22 occasions due to claustrophobia and three occasions due to noise discomfort.

The CT data was acquired on a 64-slice Philips Ingenuity CT system (Philips Medical Systems, Best, Netherlands). The MRI data was acquired on a 3.0T Philips Achieva system (Philips Medical Systems). The MRI protocol consisted of a T1 3D-isotropic acquisition (T1W_3D_TFE) for structural changes, a fluid attenuation inversion recovery (FLAIR) for detection of white matter pathology, and T2 weighted images (t2w) to exclude other types of pathologies such as tumors. Further, a Venus bold sequence was acquired for the detection of microbleeds. White matter integrity was examined using diffusion tensor imaging (DTI) sequence and functional changes were observed with resting-state fMRI sequence.

All study participants were invited to take part in CFS sampling via a lumbar puncture (LP). A total of 430 individuals consented (response rate 35.7%), but there were pharmacological contraindications for 108 persons (e.g. anticoagulant therapy, immune modulation, cancer therapy), leaving 322 participants (166 men, 156 women). Of those unwilling to participate (n = 773), several reasons were reported (e.g. expressing fear or anticipated discomfort regarding the examination or its results). The LP, with total duration of approximately 30 min, was conducted by a consultant neurologist/psychiatrist at the Neuropsychiatric outpatient department at the Sahlgrenska University Hospital. Prior to the examination, each participant had to undergo CT and/or MRI examination in order to detect possible contradictions. All CSF samples were collected by LP in the morning from the L3/L4 or L4/L5 interspace. One CSF aliquot was used for CSF analyses [75], including a CSF cell count and the CSF/serum albumin ratio, after which a total of 10 ml of CSF was collected in a polypropylene tube and immediately transported to the local laboratory for centrifugation at 1800 RCF (Relative Centrifugal Force) at 20 °C for 10 min. The supernatant was gently mixed to avoid possible gradient effects, aliquoted in polypropylene tubes and stored at − 70 °C. CSF samples in the present study underwent one freeze–thaw cycle prior to analysis. CSF total-tau, phospho-tau and Aβ42 were analyzed directly after 1 to a few days at − 20 °C. CSF total tau and tau phosphorylated at threonine 181 (p-tau) were determined using a sandwich enzyme-linked immunosorbent assay (ELISA) (INNOTEST^® htau Ag and PHOSPHO_TAU (181P), (Fujirebio, Ghent Belgium) as previously described [76, 77]. CSF amyloid-β42 (Aβ42) peptide was measured using a sandwich ELISA (INNOTEST^® β-amyloid_1-42), specifically constructed to measure amyloid-β peptides starting at amino acid 1 and ending at amino acid 42 [78]. All measurements were performed by board certified laboratory technicians blinded to clinical data. Laboratory procedures were accredited by the Swedish Board for Accreditation and Conformity Assessment. The laboratory technicians were blinded to clinical data.

A subsample of 305 participants, born on dates ending with 2 and 5 of each month, were invited for an extended audiological examination conducted at the Audiology Unit, Sahlgrenska Academy. Of those invited, 251 participants (113 men, 138 women; response rate 82.3%) underwent the examination with total duration of approximately 90 min.

A subsample of 631 participants, born on days ending with 0, 2, 5 and 8 (born January-April), and 5 and 8 (born May-December), were invited for an ophthalmic examination conducted at Sahlgrenska University Hospital, Mölndal. Of those invited, 561 participants (266 men, 295 women; response rate 88.9%) underwent the examination with a total duration of approximately 90 min.

Purposive sub-samples of study participants also took part in qualitative studies (focus groups or individual interviews).

The H70 study is permitted to be linked to a large number of databases from different state agencies in Sweden (e.g. including data regarding participants’ birth records, visits to early child care, schooling, drafting records for 18-year-old men).