Updating the study protocol: Insight 46 – a longitudinal neuroscience sub-study of the MRC National Survey of Health and Development – phases 2 and 3

Figure 4 provides a list of the tests that are administered as part of the main neuropsychological assessment block during phases 2 and 3. With few exceptions, this assessment block is administered during the morning of day 1 to reduce participant fatigue. All efforts have been made by the research team to ensure test administration continuity between the phases of data collection. The order of the test administration presented in Fig. 4 was only altered if an issue arose with the equipment or the participant requested a break.

The 15-item word list learning task and the visual search speed task (letter cancellation) were added to phase 3 to measure verbal memory and processing speed [65]. These assessments were performed as part of NSHD data collections at ages 43, 53, 60–64, and 68–69 [66, 67]. The word list task consists of 15 words that are presented serially to the participant for approximately two seconds each. Participants are given one minute to write down as many words as they can recall. This procedure is repeated for a total of three trials. After a delay of approximately 90 s (during which the visual search task is completed), the participants are given another minute to write down as many words as they can recall. The visual search task requires participants to cross out the letters P and W, which are randomly embedded within a page of other letters, as quickly and accurately as they can within one minute.

The Adult Memory and Information Processing Battery (AMIPB) complex figure drawing is a test of visual memory recall [68]. This was administered for the first time in phase 2. In phase 3, a different version of the complex figure is administered to mitigate practice effects for the participants in the TAU group who had performed this task previously. Participants are asked to copy a complex figure as accurately as possible. The figure and copy are then removed and participants are immediately asked to draw the figure again from memory. After a 30-min delay in which participants complete other cognitive assessments, participants are asked to draw the figure again from memory.

The choice reaction time test has been described previously [10, 19]. In phase 3, the response inhibition condition was removed.

The Graded Naming Test (GNT) is administered in phases 2 and 3, and it measures naming ability for objects graded for their difficulty [24]. Participants are shown 30 black and white line images, starting with common objects, before moving to less common objects. Participants are asked to name each item aloud. The task is audio recorded to measure response latency. An automated software-based solution was developed using the Speech Filing System (SFS) [69] to extract temporal information about the onset of voicing.

The instructionless eye tracking test was introduced during phase 1 (n = 179) and continued to be administered in phase 2. This is a free-viewing test designed to assess multiple cognitive domains, including recognition memory, semantic processing, and social interaction. Participants are shown 48 images for three seconds each, and their eye movements are recorded using a desk-mounted video-based eye tracker (Eyelink 1000 plus) at 100 Hz [25]. Participants are not given any explicit instructions other than to look at the screen. The stimuli are divided into four groups: i) scene exploration, which contains 28 images of social scenes, social interactions, and missing items; ii) a spatial anticipation task (based on the Brixton test), which assesses participants’ ability to detect and follow rules in sequences of stimuli [70]; iii) ten sentences, where half the sentences are semantically congruent and half semantically incongruent; and iv) a recognition memory test of ten pairs of images, one of which is a new image and the other an image shown previously in the social interaction task.

The visuomotor integration circle-tracing task, with concurrent (dual-task) serial subtraction, has been described in detail [10, 27]. The protocol remained unchanged from phase 1 to phase 2; however, in phase 3, additional trials were added to measure individual aspects of the tasks. Participants are asked to perform the circle-tracing and serial subtraction tasks independently (as single tasks). This will allow performance to be compared between single- and dual-task conditions and to investigate dual-task cost. All subtraction trials are audio-recorded.

The digit span forward and backward task from the Wechsler Memory Scale–Revised (WMS-R) was added to phase 3 to capture verbal working memory. Participants are asked to repeat increasingly longer strings of numbers exactly as presented (forward condition) and in reverse order (backward condition) [28].

Accelerated long-term forgetting [71] is assessed during phases 2 and 3. Participants are given a sealed envelope at the end of the research visit. They are asked to have this envelope to hand and to open as instructed during a telephone call that is scheduled seven days later. During the call, participants are:

For phase 3 only, participants are also asked about the 15-item word list described above. During the call, participants are given one minute to write down as many words as they can remember and immediately seal the list in an envelope to be returned to the study team.

In phase 3, the Anoto AP-701 digital pen is incorporated for use in all pen and paper tasks: the Digital Symbol Substitution task (DSST), AMIPB complex figure, and the sentence and pentagon sections of the Mini Mental State Examination (MMSE). The Anoto AP-701 is commercially available and combines an ordinary ballpoint pen with a pressure sensor and a specialised camera that captures its coordinates on a page approximately 75 times per second, with a spatial resolution of ± 0.002 inches. These timestamped coordinates constitute a digital record of the drawing from which various metrics can be derived (e.g., number and order of strokes, “think time” vs. “ink time”), providing insights into how a task is performed [29].

Audio recordings are used in phases 2 (during the GNT) and 3 (during the GNT, Wechsler Memory Scale–Revised (WMS-R) Logical Memory Recall IIa immediate and delayed conditions, and serial subtraction). The main goals of the audio recordings are to maximise scoring accuracy (by allowing raters to listen back to responses) and to analyse other aspects of responses from which potentially sensitive cognitive measures can be derived, such as word choice (allowing classification of errors), response latencies, and speech-based metrics (e.g., fluency, tonal quality, and pitch) [30].

To measure physiological arousal during the cognitive assessments in phases 2 and 3, participants are asked to wear an Empatica E4 wristband on both wrists throughout the cognitive testing session (https://​www.​empatica.​com/​e4-wristband). This is a medical grade (CE-certified) wearable device that measures electrodermal activity (EDA), temperature, acceleration, and photoplethysmography (PPG), which derives heart rate and inter-beat interval (IBI) from the blood volume pulse (BVP). These real-time data may provide insights into how factors such as anxiety or stress can affect performance on cognitive testing. The wristband is the size of a small watch and does not impede movement of the hands.

The Clinical Dementia Rating (CDR) is a global scale designed to assess the severity of dementia [35] using a semi-structured interview with the participant and an informant who knows the participant well. The CDR is administered during phases 2 and 3 to rate the participant’s cognitive performance in six domains: memory, orientation, judgement and problem solving, community activities, home and hobbies, and personal care. A global CDR is determined by combining the scores from each domain in accordance with an established scoring algorithm. All assessors undertake a standardised CDR training program online.

A delirium questionnaire is administered during phases 2 and 3 and consists of a participant and an informant interview. The participant is asked to think about a time that they have been unwell, particularly if admitted to the hospital, and to state whether they have experienced any of the following symptoms over the course of a few hours or days: new or worsening confusion or disorientation, uncharacteristic drowsiness, agitation, aggression or violence, hallucinations, or thinking clearly but then more muddled. If the participant answers yes to any of the above, further questions probe which illness led to this, what was the approximate duration, and the year the symptoms occurred. A similar set of questions is used to obtain a history of delirium affecting the participant from the informant.

Lying and standing blood pressure are measured during each phase of Insight 46 using an Omron HEM-907 automated digital oscillometric sphygmomanometer and an appropriately sized cuff. Seated blood pressure, which was added to phase 2 and continued during phase 3, is assessed prior to lying and standing blood pressure.

The six-minute self-paced stepper test [41] is administered during phases 2 and 3 to assess the participant’s exercise capacity and the effects of exercise on physiological measures, including blood pressure (both phases 2 and 3) and heart rate (phase 3 only). The test is performed while the participant is wearing an accelerometer (LPMS-B Intertial Measurement Unit; Life Performance Research Inc.) on the lower back to assess movement. Participants are instructed to perform as many steps as they can at a pace that they can maintain for six minutes. Blood pressure is measured with a Tango M2 SunTech device at rest; at two, four, and six minutes during exercise; immediately post-exercise; and at a three-minute recovery. The number of steps completed by the participant, their duration of exercise, and their perceived level of exertion on a modified (0–10) Borg scale are recorded [72].

Four measures of physical capability (chair rising, standing balance, grip strength, and walking speed) are obtained in phase 3 using standardised protocols as described elsewhere [39]. In brief, chair rise time is measured as the time taken for the participant to rise from a chair and sit back down again ten times as quickly as possible. Standing balance is measured as the duration that the participant can stand on one leg, first with their eyes open and then with their eyes closed, up to a maximum of 30 s. Grip strength is determined using an electronic handgrip dynamometer (Jamar Plus), twice in each hand. Walking speed in metres per second is derived from the time taken for the participant to walk 2.44 m (8 feet) at their usual pace.

Lung function—measured as forced expiratory volume in one second (FEV1) and forced vital capacity (FVC)—is obtained in phase 3 using a spirometer (Easy on-PC sensor) [45, 46]. Participants are asked to stand and to take as deep a breath as possible to fill their lungs to capacity. They are then asked to seal their lips around the spirometer tube, place their tongue under the mouthpiece, and blow out as hard and fast as possible until no more air can come out of their lungs. Measurements are repeated until three similar readings are obtained, up to a maximum of five times.

Bioimpedance measurements are collected in phase 3 using a Tanita BC 418 body composition analyser (TANITA Cooperation, Japan). This provides measurements of body fat ratio, body fat mass, fat-free mass, estimated muscle mass, and base metabolic rate. Waist and hip circumference are also measured in phase 3 [11, 40].

To quantify sleep and circadian measures during phases 2 and 3, participants are asked to wear a wrist actigraphy device for seven days following their clinic visit (Philips Actiwatch Spectrum Pro). The device detects light and motion, allowing periods of rest and activity to be monitored, and it can be used to make inferences about sleep and wakefulness. To complement actigraphy data, participants are also asked to complete a sleep diary, documenting their self-reported sleep and waking times.

During phase 3, participants are asked to wear a FreeStyle Libre Pro IQ Sensor glucose monitor for seven days following their clinic visit [47]. The monitor provides a continuous measure of blood glucose levels recorded every 15 min.

As previously described, hearing and smell tests were administered during phase 1 [10, 49, 73]. During phase 2, a novel sensory test—a spoons taste test—was administered. Participants are asked to taste spoons that are made from different materials (stainless steel, silver, gold, copper, tin, and zinc), which have different metallic and bitter tastes [50]. The order in which the spoons are presented to each participant is randomized. The participants are then asked to place the spoons in order according to how metallic, bitter, salty, sweet, and unpleasant they found the tastes of each spoon.

The study participants are asked to complete a series of questionnaires in phases 2 and 3 that relate to the topics of sleep, head injuries, repetitive negative thinking, and general health. All questionnaires are completed during the research visit.

Three sleep questionnaires are administered: Epworth Sleepiness Scale (ESS) [51, 52], Pittsburgh Sleep Quality Index (PSQI) [53], and REM Sleep Behaviour Disorder Questionnaire (RSBDQ) [54].

The ESS provides a measure of the participant’s general level of daytime sleepiness [51]. The questionnaire lists eight situations, and the respondent rates how likely they are to doze off or fall asleep in each of those situations. The scores for each situation are summed, giving a total score between zero and 24. A total score greater than 16 indicates a high level of daytime sleepiness.

The PSQI is designed to measure sleep quality and assess sleep disturbances during the past month [53]. With permission from the author, the term “room mate” within the questionnaire was changed to “housemate” in keeping with UK norms. Participants are asked to provide responses based on experiences over the majority of days and nights during that one-month period. The questionnaire is divided into eight domains: sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbance, use of sleep medication, and daytime dysfunction. The scores from each domain are summed to create a global PSQI score. A global score less than or equal to five is associated with good sleep quality, while a score greater than five is associated with poor sleep quality.

The RSBDQ is a ten-item questionnaire designed to detect clinical features of REM sleep behaviour disorder, with one of the items having four parts [54]. Participants are asked to tick either “yes” or “no” in response to each statement in relation to their current sleep behaviours. Each “yes” answer is worth one point, resulting in a total score between zero and 13. A score of five or more is indicative of REM sleep behaviour disorder.

The study team created a questionnaire that is administered during phases 2 and 3 to gather information about head injury. The participants are asked:

One year into phase 2, the Perseverative Thinking Questionnaire (PTQ) [55] was introduced. The PTQ consists of 15 statements that aim to assess the core characteristics of repetitive negative thinking: repetitiveness, intrusiveness, and difficulties with disengagement. Participants are asked to use a scale from zero (never) to four (almost always) to rate the extent to which each statement applies when thinking about negative experiences or problems. The scores are summed to create a total score between zero and 60. Three subscores are also generated, representing core characteristics of repetitive negative thinking, unproductiveness of repetitive negative thinking, and repetitive negative thinking capturing mental capacity.

The final self-complete questionnaire is the General Health Questionnaire (GHQ)-28 [57, 58], which is used to gain a better understanding of the participant’s emotional health over recent weeks. The GHQ-28 was introduced in phase 3 and consists of 28 questions divided equally into four categories: measurement of somatic symptoms, anxiety and insomnia, social dysfunction, and depression. Participants are asked to respond to each question on a Likert scale of 0, 1, 2, or 3 [58]. The individual scores are then summed to create a total score between zero and 84. The GHQ-28 was administered during previous waves of data collection for the NSHD [74, 75], allowing longitudinal analyses.

In addition to continuing to collect blood (plasma, serum, and Guthrie cards) and urine samples for biomarker exploration as in phase 1 [10], cerebrospinal fluid (CSF) samples are also being obtained for this purpose from phase 2 onwards if the study participant is willing and eligible to undergo a lumbar puncture. Contraindications to lumbar puncture include neuroimaging evidence of a space-occupying lesion with mass effect, tonsillar herniation due to Chiari malformation, or increased intracranial pressure; known bleeding disorder or use of anticoagulation/antiplatelet medication (except aspirin 75 mg once daily); congenital spine abnormality; lignocaine allergy; or active rash over the proposed puncture site. During consent, participants are informed of possible side effects, including headache (1 in 10), backache, infection or bleeding, and nerve damage (very rare). With the participant in the lateral decubitus position or seated with lumbar forward flexion, an intervertebral space between L2/L3 and L5/S1 is first located by palpation. The skin is then prepared with chlorhexidine 2%/ethanol 70% solution, before local anaesthesia (lignocaine 2%, maximum 3 mg/kg) is administered. An atraumatic 22G spinal needle is then inserted into the anaesthetised space and CSF is collected by passive drip into two polypropylene tubes, aiming for a total of 20 ml. An aseptic technique (sterile gloves and field) is maintained throughout. The CSF samples are then transported on ice to the laboratory where they are centrifuged (1750 g/ 3000 rpm for 10 min at 4 °C) and aliquoted into up to 40 polypropylene cryovials, before being stored in a -80 °C freezer within 60 min of arrival. Planned biomarker analyses in plasma and CSF include neurofilament light chain, Aβ40 and Aβ42, glial fibrillary acidic protein (GFAP), and several phosphorylated tau moieties, as well as novel biomarkers for synaptic function (including SNAP-25 and β-synuclein).

During phases 2 and 3, blood for RNA analysis is collected into a 2.5 ml PAXgene tube and stored at -80 °C for later processing and RNA extraction. Blood samples are also being collected during phases 2 and 3 for the generation of cell lines, with the goal of generating induced pluripotent stem cells from individuals with low/high risk of disease. The blood is collected into an 8 ml cell preparation tube (CPT) and kept at room temperature, before being transported to the laboratory within 24 h of collection. Peripheral blood mononuclear cells (PBMCs) are extracted using the Ficoll-Paque protocol and cryopreserved for future reprogramming to induced pluripotent stem cells [60, 61].

Similar to phase 1, blood samples are being collected for clinical measures in phase 3. One 4 ml EDTA sample is collected to measure haemoglobin, platelet count and haemoglobin A1c (HbA1c). One 5 ml SST sample is collected to measure urea, creatinine, vitamin B12, folate, vitamin D (25-OH), random glucose, and thyroid-stimulating hormone (TSH).

In phase 2 only, participants are asked to provide a mid-stream early morning urine sample in a 25 ml urine collection pot for dried urine spot processing. The urine is pipetted onto four circles (each holding 25–80 µL) on Whatman 903 protein saver cards and allowed to dry at room temperature for 3–4 h before storage in a zip-lock plastic bag with desiccant at -80 °C. The remaining urine is aliquoted into three × 2 ml samples and stored at -80 °C. Metabolomic analyses are planned, with a view to validating biomarkers and studying the metabolism of drugs.

Phases 1 and 2 of the study followed the same brain imaging protocol design, as previously described [10], with the acquisition of imaging data on the Siemens Biograph mMR 3T PET/MRI scanner at the UCLH Macmillan Cancer Centre. For phase 3, the brain scanning takes place on both the PET/MRI at UCLH and the Siemens Prisma 3T MRI at Chenies Mews Imaging Centre, which enables the use of more advanced acquisition techniques due to better sensitivity and hardware performance. The advanced MR sequences acquired on the Prisma are as follows:

Table 3 provides the basic parameters for the MP2RAGE, PCASL, and IR-IVIM sequences acquired on the Siemens Prisma 3T MRI.

The PET/MRI imaging protocol has been revised for phase 3 to reflect the use of a different Aβ PET ligand—[¹⁸F]florbetaben (300 MBq) —compared to the one used in phases 1 and 2, as well as the use of the tau PET ligand [¹⁸F]MK-6240 (185 MBq). These two tracers require longer post-injection times for accurate quantification compared to [¹⁸F]florbetapir. A coffee-break protocol [79], see Fig. 5, is used to acquire PET data from 0–30 min and between 90–110 min post PET ligand injection. During the 0–30 min period, the following MRI sequences are acquired: volumetric T1, volumetric T2, volumetric FLAIR, and susceptibility-weighted imaging and quantitative susceptibility mapping. Participants are removed from the scanner during the 30–90 min period. During the 90–110 min period, multi-shell diffusion-weighted imaging and B0 field mapping are acquired. For phase 3, resting state fMRI data are not acquired, and the phase 1 and 2 arterial spin labelling (ASL) protocol is not run on the PET/MRI scanner because a more advanced multi-PLD ASL protocol has been implemented on the 3 T Prisma (see above). For all phases of Insight 46, PET data are acquired continuously during and following injection, apart from during the coffee-break, allowing the PET ligand dynamics to be assessed.

Dual-energy X-ray absorptiometry (DXA) scans were undertaken during the NSHD clinic visit at ages 60–64 [80], and DXA scans using a Hologic Bone Densitometer QDR Horizon system were added to phase 3 of Insight 46. The DXA scan assesses bone density, body composition (lean and fat mass), and aortic calcification, and it complements the other anthropometric and body composition measurements described above. The protocol includes a scan of the left hip (or right hip if the left is not possible), a scan of the lumbar spine, instant vertebral assessment (IVA), and a whole-body examination.

Optical coherence tomography (OCT)—a non-invasive eye scan that provides detailed retinal images, including images of small blood vessels—has also been added to phase 3. During the procedure, a trained technician first carries out an assessment of visual acuity in each eye, with appropriate spectacle and pinhole correction. The technician then completes the retinal imaging and OCT measurements for each eye using a Topcon DRI Triton device. The protocol includes fundus photography with fluorescence turned off, 3D OCT of the macula, 3D OCT of the optic nerve and disc, and angiography of the macula (4.5 × 4.5 mm).