Outcome measures for cognitive function study
Change in cognitive function at 24 months determined by the California Verbal Learning Test. This is a test of memory of 16 words.
Cognitive performance as measured by immediate and delayed recall of a short story, tests of prospective memory, timed letter search/cancellation task, verbal fluency, digit span forwards and backwards, symbol digit modalities test, simple and choice reaction time, and spatial memory.
Compliance determined by counting the number of capsules remaining every 3 months, and by measuring the change in n-3 LCP concentration in buccal epithelial cells over 24 months.
Change in Body Mass Index (a measure of body size).
Number of hospital admissions for cardiovascular events over 24 months.
The sample size required for this trial is based on a 0.3SD difference between trial arms in long-delay free recall of List A of the California Verbal Learning Test over the 24 months of intervention. To detect a 0.3SD difference between trial arms, with 90% power and 1% significance, 332 individuals are required per treatment group. Allowing for 20% drop-out over 24 months of intervention, the total sample size required for the study is 798 individuals. A difference of 0.3SD is clinically relevant and would be equivalent to individuals in the intervention arm being able to remember one more word out of the 16 in the California Verbal Learning Test than those in the placebo arm [18
In order to attain this sample size it is expected that a sample of approximately 4500 individuals, pre-screened for diabetes and dementia, will initially be drawn from the registers of 20 participating general practices (around 225 patients aged 70–79 years per practice). A conservative estimate of the proportion of eligible individuals who will agree to participate in the trial is 20%. It is estimated that it will take 6 months to enrol the full sample for this trial.
Baseline (pre-intervention) cognitive function data collection
At trial entry to enable randomisation
Date of birth
General practice number
Baseline data collection
1. Psychological health: the 30–item General Health Questionnaire (GHQ-30) will be used to assess the affective state of participants [19
]. The GHQ-30 is quick and easy to administer, and an assessment of mood is important as it may be associated with cognitive performance.
2. Educational achievement: the widely used Burnham Scale of educational achievement will be used [20
]. This is an important covariate as educational level has been shown to be strongly related to cognitive performance.
3. Blood pressure: this is necessary as hypertension is an important consequence of the metabolic syndrome, and individuals with high blood pressure may have a raised susceptibility to vascular and neural damage. Baseline blood pressure will therefore be used as a covariate in analysis. Blood pressure will be measured using the automated DYNAMAP device. Two measurements each of systolic and diastolic pressure will be taken in the sitting position (the mean of the two measurements will be used in analysis); participants will have rested for 5 minutes before their blood pressure is measured and there will be a minimum of 5 minutes between the two measurements. Normal general practice procedures will be followed for the management and treatment of high blood pressure.
4. Cardiovascular health: any history of hospital admission for myocardial infarction (MI) or stroke over the preceding 5 years will be abstracted from patient notes.
5. National Health Service number: So that the study organisers do not lose contact with patients should they move address and also to follow up on health status, participants are being asked to give their agreement for their names and NHS numbers to be sent to the NHS Central Register.
6. Basic anthropometric measures: height and weight will be measured using standardised procedures in order to enable the calculation of the Body Mass Index: a basic marker of body size.
7. Swab of cheek cells: nurses will collect a sample of buccal mucosal epithelial cells from each participant. This is a relatively non-invasive procedure in which a sterile wooden spatula is scraped briskly along the inside of the cheek (about 10 strokes on each cheek). The cheek cells will be sent to laboratories, in sterile containers, to be analysed for their n-3 LCP concentration.
8. Fish consumption: two simple questions regarding habitual fish consumption will be asked. Firstly, a question regarding the frequency of fish consumption, and second, participants will be asked to list the three species of fish they consume most frequently.
9. Cognitive ability: the participants will be asked to complete a validated series of cognitive tests. It is currently not known which, if any, cognitive domain will be most affected by n-3 LCP supplementation and therefore a series of tests have been selected which cover the major domains (memory, attention, psychomotor speed and executive function):
a. Brief assessment of subjective symptoms of cognitive impairment (memory, language/word-finding, concentration).
b. Immediate and delayed recall of a short story (from the Wechsler Memory Scale).
c. Immediate and delayed recall of a 16-item word list (California Verbal Learning Test).
d. Three tests of prospective memory i.e. remembering to carry out instructions without being reminded.
e. Timed letter search/cancellation task (attention/psychomotor speed/executive function).
f. Verbal fluency – naming animals (word-finding/executive function).
g. Digit span forwards and backwards (working memory/executive function) from the Wechsler Adult Intelligence Scale.
h. Symbol digit modalities test (attention/psychomotor speed/executive function).
i. Simple (psychomotor speed) and choice (decision speed) reaction time.
j. Spatial memory (memory/visual spatial function).
Following the completion of the series of tests detailed above, those participants not enrolled in the retinal function study will be introduced to the dietary intervention. The dietary intervention will be a daily dietary supplement in the form of capsules which will be identical in size, shape, colour and smell for both the intervention and placebo arms of the trial. Given the lack of trial evidence to support the use of any particular level of dietary supplementation, the dose selected in the OPAL study was based on the following considerations: the UK Food Standards Agency currently recommends that men and post-reproductive age women consume one to four, 140 g portions of oily fish a week [21
]; typical dietary recommendations, such as those of the World Health Organisation fall in the range 0.3–0.5 g n-3 LCPs daily [22
]; the dose of n-3 LCPs generally recognised as safe (GRAS) is 3 g per day.
Taking these considerations into account, a pragmatic decision was made on the dose for the OPAL study. The intervention arm will be asked to consume two 650 mg soft-gel capsules daily containing a total of 500 mg DHA and 200 mg EPA. This is equivalent to approximately 4.9 g of n-3 LCPs a week (0.7 g/day), a level that can be achieved via the consumption of approximately 300 g (slightly more than 2 portions) of fatty-fish a week. This level of supplementation falls well below the upper GRAS level and should thus be safe. The placebo arm will be asked to consume two identical 650 mg capsules containing olive oil daily, this oil is rich in n-9 fatty acids and will thus have a minimal effect on the 18:2 n-6/18:3 n-3 ratio. Changes in the n-6/n-3 ratio may affect prostanoid balance which affects vascular and inflammatory responses. The higher DHA to EPA ratio in the supplement is justified by the rationale of the study, which prioritises the neuroprotective effect of DHA over the vascular and anti-thrombotic effect of EPA.
The research nurses will explain the importance of consuming the capsules every day, suggesting that it should become part of their daily routine, for example by always consuming the capsules at breakfast-time. Participants will be given a 3-month supply of capsules at the baseline clinic visit, and asked to attend the clinic every 3-months throughout the 24-month course of the trial i.e. at 3, 6, 9, 12, 15, 18 and 21 months. Participants will be sent reminders to collect repeat supplies of capsules.
The participants enrolled in the retinal function study will be introduced to the dietary intervention after their baseline retinal function test (see below). These participants will be invited back to the general practice after retinal function testing to be given their dietary supplements.
Data collection at 3, 6, 9, 12, 15, 18, and 21 months
Participants will have appointments to meet the research nurse every 3 months throughout the study (8 repeat visits in total including 24 month assessment). During these appointments, participants will be reminded of the importance of the study and of the need to comply with the study protocol. The research nurses will also record any information the participants volunteer regarding any discomfort caused by the capsules. The recognised side-effects of n-3 LCP capsules include belching, flatulence, abdominal discomfort and loose stools. These discomforts are generally mild and decrease over time, and the research nurses will provide reassurance to any concerned participants.
Participants will be asked to bring their dietary capsule containers with them to the meeting so that any remaining capsules can be counted as a measure of compliance. Buccal cell swabs will also be taken during these repeat appointments from a random 20% sample every 6 months (i.e. at 6, 12 and 18 months) as a further measure of compliance. Participant records from the general practice will be consulted every 3 months in order to record any hospital admissions for MI or stroke over the intervening 3 month period.
Assessment at 24 months
A final evaluation of the cognitive function of all trial participants will be carried out after 24 months of intervention. Participants will be invited to attend their general practice and will be assessed for:
1. Psychological health: as at baseline.
2. Blood pressure: as at baseline.
3. Cardiovascular health: record of hospital admissions for MI or stroke between the 21 and 24 month appointments.
4. Basic anthropometric measures: as at baseline.
5. Buccal cell DHA concentration: as at baseline. Collected as a marker of the impact of the intervention on buccal cell DHA concentration, and compliance in the intervention arm of the trial.
6. Cognitive function: as at baseline.
Participants will be offered the use of a taxi, or their travel costs on public transport or mileage in private cars to attend the appointment up to a maximum of £10.
Data analysis for assessing change in cognitive function
Primary analysis will be carried out based upon the groups as randomised ("intention to treat"). Results will be presented as appropriate effects sizes with a measure of precision (95% confidence intervals). Covariates such age, gender, and baseline blood pressure and DHA concentration will be adjusted for in the analysis. Further exploratory analyses will be based on those patients who fully follow the treatment protocol.