The older people, omega-3, and cognitive health (EPOCH) trial design and methodology: A randomised, double-blind, controlled trial investigating the effect of long-chain omega-3 fatty acids on cognitive ageing and wellbeing in cognitively healthy older adults
- Vanessa Danthiir1Email author,
- Nicholas R Burns†2,
- Ted Nettelbeck†2,
- Carlene Wilson†1, 4 and
- Gary Wittert†3
© Danthiir et al; licensee BioMed Central Ltd. 2011
Received: 28 June 2011
Accepted: 20 October 2011
Published: 20 October 2011
Some studies have suggested an association between omega-3 long-chain polyunsaturated fatty acids (n-3 LC PUFAs) and better cognitive outcomes in older adults. To date, only two randomised, controlled trials have assessed the effect of n-3 LC PUFA supplementation on cognitive function in older cognitively healthy populations. Of these trials only one found a benefit, in the subgroup carrying the ApoE-ε4 allele. The benefits of n-3 LC PUFA supplementation on cognitive function in older normal populations thus still remain unclear. The main objective of the current study was to provide a comprehensive assessment of the potential of n-3 LC PUFAs to slow cognitive decline in normal elderly people, and included ApoE-ε4 allele carriage as a potential moderating factor. The detailed methodology of the trial is reported herein.
The study was a parallel, 18-month, randomised, double-blind, placebo-controlled intervention with assessment at baseline and repeated 6-monthly. Participants (N = 391, 53.7% female) aged 65-90 years, English-speaking and with normal cognitive function, were recruited from metropolitan Adelaide, South Australia. Participants in the intervention arm received capsules containing fish-oil at a daily dosage of 1720 mg of docosahexaenoic acid and 600 mg of eicosapentaenoic acid while the placebo arm received the equivalent amount of olive oil in their capsules. The primary outcome is rate of change in cognitive performance, as measured by latent variables for the cognitive constructs (encompassing Reasoning, Working Memory, Short-term Memory, Retrieval Fluency, Inhibition, Simple and Choice-Reaction Time, Perceptual Speed, Odd-man-out Reaction Time, Speed of Memory Scanning, and Psychomotor Speed) and assessed by latent growth curve modeling. Secondary outcomes are change in the Mini-mental State Examination, functional capacity and well-being (including health status, depression, mood, and self-report cognitive functioning), blood pressure, and biomarkers of n-3 LC PUFA status, glucose, lipid metabolism, inflammation, oxidative stress, and DNA damage.
Australia and New Zealand Clinical Trials Register (ANZCTR): ACTRN12607000278437
Ageing is accompanied by what is normally considered to be inevitable cognitive decline, although the extent to which this occurs is highly variable, and strongly affected by various disease processes. Cognitive function is a major determinant of quality of life in older age and decline in cognitive functioning is a primary contributing factor to increasing dependency in the elderly. Due to the ageing profile of the population, the financial, social, and other burdens that this dependency places upon society are of increasing concern. Thus, finding ways to prevent or ameliorate age-related cognitive decline is a public health imperative, with potential benefits not just in terms of lessening aged care costs but also in the enhancement of well-being in a growing segment of society.
Omega-3 polyunsaturated fatty acids (n-3 PUFAs) are one diet-related factor suggested to influence cognitive decline during ageing. The n-3 long-chain (LC) PUFAs eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are crucial to brain development and normal brain functioning . DHA is particularly important to brain functioning due to its influence on neural membrane properties, which modulate cell signalling . DHA concentration in the brain decreases with age in humans  and rats ; this has been postulated to be consequential to the age-related deterioration in central nervous system functions . Evidence from animal studies supports this; animals fed a low n-3 PUFA diet show cognitive deficits [4, 5] that are ameliorated by DHA supplementation [6, 7] and DHA supplementation improves memory performance in aged mice .
Cross-sectional and prospective studies
Oily fish are the major dietary source of the n-3 LC PUFAs, EPA and DHA. A consistent finding across studies is that higher intake of fish is related to less cognitive decline [9–12] and incidence of dementia [10, 13, 14] in prospective studies, and associated with better cognitive performance in non-clinical samples [11, 15–17], in cross-sectional analyses. However, variable associations have been found between dietary intake levels of n-3 LC PUFAs and cognitive outcomes; only a handful of the aforementioned studies that also examined relationships between cognitive outcomes and dietary intake levels of n-3 LC PUFAs have found significant positive relationships [9, 14, 15].
Stronger evidence of a positive relationship between cognitive outcomes in older age and levels of n-3 LC PUFAs has been found in studies assessing plasma or erythrocyte levels of n-3 PUFAs. In these studies, DHA and EPA levels, either individually or in combination, have been associated with better cognitive function in normal older adults in cross-sectional analyses [18, 19], and in prospective studies, with better cognitive outcomes over time  or reduced risk of cognitive decline , and lower risk of developing dementia ; in contrast, Laurin et al  reported a negative relationship between cognition and biomarkers of n-3 LC PUFAs.
Randomised controlled trials (RCTs)
There have been six published double-blind RCTs of n-3 LC PUFAs on cognitive functioning in the elderly within both clinical and non-clinical populations [23–28]. Only two studies have been conducted with cognitively healthy populations, screened for dementia at baseline, and no overall effect of supplementation was found. The "Older People and n-3 Long-chain polyunsaturated fatty acids" study  supplemented 867 cognitively healthy participants aged 70-79 years, with either 200 mg of EPA and 500 mg of DHA, or olive oil, daily for 24 months. After adjustment for confounders, there was no significant difference between the groups in performance on the California Verbal Learning Test, the main cognitive outcome measure, or any secondary cognitive outcomes (a global cognitive score, and domain scores on memory, processing speed, executive function, and delayed recall). In a 26-week RCT of fish oil on cognitive performance in 302 cognitively healthy adults aged 65+ years , no main effect of supplementation was seen for any of the cognitive domains (attention, sensorimotor speed, memory, and executive function), assessed by 5 tests. However, subgroup analysis revealed that carriers of the Apolipoprotein E (ApoE) ε4 allele, in both the low (226 mg EPA, 176 mg DHA) and high dose (1093 mg EPA, 847 mg DHA) fish oil groups, showed an improvement after 26 weeks compared with the placebo on the Digit-Span Forward task, their marker of attention.
In three of the studies with cognitively impaired (including Alzheimer's disease (AD)) samples, the interventions had a beneficial effect on cognitive outcomes but only in participants with mild cognitive dysfunction [23, 25, 28]. This finding was not replicated in a large RCT conducted over 18 months of 402 participants with mild to moderate AD. However, in this study in ApoE-ε4 non-carriers only, there was a positive effect on two cognitive outcome measures .
Apolipoprotein E-ε4 allele
The presence or absence of the Apolipoprotein E-ε4 allele appears to be a potentially important modifier of relationships between n-3 LC PUFAs and older-age cognitive outcomes but there are conflicting findings regarding the direction of this interaction. Epidemiological studies that have reported a differential relationship between cognitive outcomes and measures of n-3 LC PUFAs for carriers versus non-carriers of the ApoE-ε4 allele have found a beneficial relationship present only in non-carriers [19, 29]. However, the two RCTs that have included examination of the effect of ApoE-ε4 allele carriage in sub-analyses have reported a beneficial effect of supplementation dependent on ApoE-ε4 allele status, but differ on which group benefits from supplementation [26, 27]. Nonetheless, ApoE polymorphisms have been found to influence lipid responses to fish oil supplementation  and in a small group of males, those without ApoE-ε4 showed a greater increase in DHA and EPA after fish-oil supplementation . Further, possession of the ApoE-ε4 allele is associated with worse cognitive performance in old age  and greater risk of cognitive decline and dementia , hence, even in cognitively healthy samples screened for dementia, the genesis of any cognitive change may differ between carriers and non-carriers, providing another potential avenue for differential effects of supplementation on cognitive outcomes due to ApoE genotype.
These studies suggest that n-3 LC PUFA intake, in particular DHA, has a beneficial effect on cognition in older adults. However, the majority of evidence for the benefits of n-3 PUFA on cognitive functioning in adults is associational. Furthermore, work to date in this area has often not comprehensively assessed cognitive function or cognitive decline. Without a priori expectations of specific cognitive domains that will be affected, a comprehensive operationalisation of cognition is required in order to ascertain whether a null result is due to the absence of a relationship or effect, or the relevant cognitive domains were not assessed. Given the potential importance of ApoE-ε4 allele carriage as a determinant of n-3 LC PUFAs influence upon cognitive outcomes in older age, more randomised controlled trials are needed that incorporate consideration of its impact in order to determine whether n-3 LC PUFAs can influence age-related cognitive change in older adults.
The primary aim of the current trial was to comprehensively assess over 18-months the efficacy of a DHA-rich fish-oil supplement on slowing cognitive decline in non-demented older adults; the hypothesis was that, over 18 months, the n-3 LC PUFA supplementation group would show slower cognitive decline than the control group. The trial was set within the broader context of examining nutritional, health, and lifestyle factors associated with better cognitive and well-being outcomes in the elderly, both cross-sectionally and longitudinally.
Secondary aims were: 1) to explore relationships between measures of n-3 LC PUFA intake and measures of cognitive function at baseline; 2) to determine whether the effect of n-3 LC PUFA supplementation on cognitive decline is moderated by presence of the ApoE-ε4 allele, initial cognitive status, measures of dietary n-3 LC PUFA intake (including changes in n-3 LC PUFA status) or any nutritional, health-related, or psychological factors associated with cognitive outcomes in the elderly; 3) to examine the effect of the intervention on measures of functional capacity and well-being; and 4) examine relationships between measures of n-3 LC PUFA intake and functional capacity and well-being at baseline.
This paper details the methodology of the study, including the design, assessment battery, and research protocols used.
The trial was a parallel, 18-month, randomised, double-blind, placebo-controlled trial with repeated measures every 6 months, totalling 4 measurement points. Because the primary hypothesis related to change in a continual developmental process, multiple repeated measures enable determination of the rate of change on the measures, allowing assessment of an intervention effect on a trajectory, rather than effects at a specific time point. Decline of the cognitive abilities vulnerable to ageing effects begins by the early 30s so although participants were screened for cognitive impairment, all were expected to experience cognitive decline to some extent; for instance, meta-analyses of cross-sectional data suggest that processing speed declines, on average, by approximately 20% at age 40 and by 40-60% at age 80 . In terms of supplementation duration, studies have detected significant cognitive changes due to n-3 PUFA supplementation after 3-4 months [35, 36]. However, these studies focused on clinical populations; change in normal populations would not be expected to be as rapid, or of the same magnitude and the longer duration allowed for the multiple testing points required to best test the hypothesis and conferred greater power to detect an effect.
The trial was co-ordinated from CSIRO Food and Nutritional Sciences, Adelaide, Australia. The protocol was explained to participants and written informed consent was obtained prior to study commencement. All experimental procedures were approved by the Human Research Ethics Committee of CSIRO and in compliance with the Helsinki declaration. The study follows Good Clinical Research Practice. Participants were encouraged to inform the trial research officer of any adverse events; these were recorded by the officer. Adverse effects suspected as being related to supplement use were reported to the trial clinical doctor for follow-up. Participants attended the centre every 3 months; every 6 months, to undergo their assessments, and 3-months after each assessment to obtain additional capsules and be followed-up. At these visits, adverse events were recorded, along with any changes in medication, supplement use, or general health.
Sampling and criteria
Participants were recruited from metropolitan Adelaide, South Australia and surrounding areas via agencies and organisations for older citizens, public advertisements, and media releases. Information sheets explaining the study and asking for volunteers, were circulated in response to expressions of interest. To be eligible for inclusion, participants were required to be 65-90 years of age at screening, fluent in English, and agree to not commence their own n-3 fish-oil (or algal) supplementation throughout the duration of the study. Participants were excluded from study entry if they were taking n-3 fish-oil (or algal) supplements, had experienced any condition where cognitive impairment may be a consequence (e.g., head injury, other brain trauma, transient ischemic attacks, stroke, coronary artery bypass surgery, open heart surgery, degenerative neurological disease, history of alcohol/drug abuse), were taking doses of medications known to interfere with cognition, had a significant medical condition, were diagnosed with an intellectual disability, current major clinical depression, diabetes, dementia, or received a score less than 22/27 on a telephone-administered version  of the Mini-mental State Examination (MMSE) (equivalent to < 24/30 on the standard MMSE, the traditional cut-off for possible dementia).
Screening and information sessions
Individuals who expressed interest in participating were sent a questionnaire to obtain demographic information, history of health events, and specific questions to determine eligibility. Participants deemed eligible, based on the questionnaire responses, were contacted by a trained research officer and, if consenting, were administered a version of the MMSE via telephone - this method of administration has been validated for the MMSE . Participants scoring less than 22 correct out of a potential 27 points were excluded. Subsequently, approximately two to three months prior to baseline testing, participants attended an information session, during which the study protocol and requirements were explained to them, queries were answered, and practice trials and examples of the cognitive tasks were given. Participants were asked to maintain their usual diet over the course of the study. After written consent to participate was obtained, individual administration of the standard MMSE was undertaken, to corroborate the telephone screening and to serve as an outcome measure.
Randomisation, allocation, and blinding
Participants were given a unique study number on entry into the trial. An independent researcher prepared allocation to treatment. Age-stratified, permuted-block randomisation, with mixed block-sizes ranging from two to eight (size unknown to study investigators until unblinding), with 1:1 allocation was employed. Participants were placed into 5-year age strata, and randomisation for each stratum followed a computer generated randomisation schedule. The researchers, project staff, and participants remained blinded to treatment allocation until the trial was completed and the database locked. Capsule bottles were labelled with participants' unique study numbers by an external company. Active and placebo capsules (opaque brown, oblong, and softgel; Blackmores Pty Ltd, Australia) and containers were visually identical, the fish-oil was the low-odour type, and 1% fish-oil was added to the placebo oil to help maintain blinding in the event of, for example, accidental piercing of the capsules or an aftertaste. At the study-end, participants were asked to guess which study arm they were in, to assess adequacy of blinding.
Dietary intervention schedule
The intervention was administered for 18 months, starting immediately after baseline assessment, after which participants received their first 3-month supply of capsule bottles. For the study duration, participants returned to the CSIRO clinic every 3 months, received their next 3-month supply of capsules, and returned all unused capsules. The trial research officers recorded any reported side-effects of the capsules.
Compliance was monitored via a count of all unused supplements returned at three-monthly intervals, along with self-report calendars, mailed back on a monthly basis throughout the trial. If participants' compliance fell below 85% on the basis of their calendars, they were contacted by a research officer who noted the reasons. Compliance also will be assessed by examination of erythrocyte membrane n-3 LC PUFA status and changes in status, via a blood sample taken at every 6 month-assessment.
n-3 LC PUFA capsule
The supplement was chosen to be DHA-rich on the basis that evidence suggests DHA is more related to cognition in ageing, than is EPA. The dosage for the current trial followed that of the first reported n-3 LC PUFA placebo-controlled intervention study focusing on cognitive decline in the elderly (albeit in AD patients ) that found a positive effect; hence, the study dose was 1720 mg DHA and 600 mg EPA daily. Each capsule contained 430 mg DHA and 150 mg EPA (EPAX 1050 TG/N n-3 concentrate, EPAX, Norway); participants were instructed to have a total of four capsules a day, two in the morning and two in the evening. The total n-3 LC PUFAs in the daily supplement were equivalent to approximately 150 g of salmon. There are no known risks associated with n-3 LC PUFA consumption at this level. Evidence supports the effectiveness of n-3 fish-oil supplements as providers of n-3 LC PUFAs, with n-3 LC PUFAs from fish-oil supplements being incorporated into erythrocyte membranes to a similar extent as n-3 LC PUFAs from dietary fish intake .
The placebo arm received capsules, each containing 990 mg of olive oil (Oil Seed Products, New Zealand) and 10 mg of fish oil (equivalent to 1.8 mg EPA and 1.2 mg DHA; omega-3 18/12, Ocean Nutrition, Canada) and consumed a total of 4 capsules a day. Unlike other oils, olive oil is not rich in n-3 or n-6 fatty acids and should thus have minimal effect on the n-6:n-3 ratio. Olive oil is high in mono-unsaturated fatty acids (MUFAs), which have in some epidemiological studies been associated with lower cognitive decline , yet it is thought that it is the polyphenolic content of the oil that might bear this effect. The amount of MUFAs in the capsules was very minor compared to the dietary consumption of MUFAs in those studies that found a cognitively protective effect (i.e., mean daily MUFA intake = 42.1 g ). The 4 g of low-polyphenol olive oil per day was thus considered non-problematic. The minimal amount of fish oil was added to help mask any difference between the placebo and active capsules and was not processed under nitrogen, rendering it less potent than that used in the active formulation.
After the group information session, participants were scheduled one visit to the laboratory for assessment, every 6 months, for 18 months. Self-report questionnaires (see following section on study instruments) were mailed out to participants up to four weeks prior to each assessment visit, with instructions to complete and return them between one to two weeks prior to their assessment. Missing data or incorrectly completed questionnaires were thus identified beforehand and addressed during their assessment sessions. All physical measurements and cognitive testing were conducted at the same time in the morning at each session.
Summary of measures assessed over the study duration.
Outcome measures and covariates
Time points (months)
Physical activity, functional capacity, and well-being measures
Blood pressure, weight
Erythrocyte membrane n-3 and n-6 fatty acid status
Serum folate and Vitamin B12, and plasma carotenoid levels
Lipids, glucose, glycated haemoglobin, homocysteine
Medications, supplement use, adverse events and health changes
Inflammation (CRP), oxidative stress (MDA), and telomere length
Dietary data - general FFQ
Dietary data - n-3 consumption FFQ
Dietary data - historical seafood consumption FFQ
Primary outcome measures
Rate of change in cognitive performance, as measured by latent variables for the cognitive constructs detailed below.
Secondary outcome measures
Change in: MMSE; perceived health status, depressive symptoms, positive and negative affect, life satisfaction, self-reported cognitive functioning, and functional capacity, as assessed by the questionnaires below; blood pressure; biomarkers of glucose, glycated haemoglobin, triglycerides, total cholesterol, HDL, LDL, homocysteine, CRP, MDA, and telomere length.
Cognitive test battery
No previous definitive evidence suggests which cognitive domains may be amenable to change via n-3 LC PUFAs, thus the test battery used in this study comprehensively assessed all relevant domains that might benefit from a nutritional intervention. The focus of the battery was on constructs that are sensitive to age-related changes and that are most strongly associated with functional status, functional decline, and dementia.
Most of the cognitive domains were assessed using tasks in which speed of response was the main outcome measure. The psychometric properties of speed tasks make them more sensitive to subtle cognitive changes and obviate the need for parallel forms for repeated assessment. These properties, together with evidence that processing speed is integral to higher-order processes such as reasoning , may be a key indicator of changes in neural efficiency  and functional outcomes , and is an important mediator of age-related cognitive changes , make speed tasks ideal as outcome measures to test the effect of a nutritional intervention on cognitive performance in older-age.
With the exception of two individual tasks, at least two measures of each cognitive construct were used to indicate a latent construct for statistical analyses, thereby reducing task-specific variance and increasing construct validity and the reliability of assessment. The cognitive constructs assessed were Reasoning, Working Memory, Short-term Memory, Retrieval Fluency, Inhibition, Simple and Choice-Reaction Time, Perceptual Speed, Odd-man-out Reaction Time, Speed of Memory Scanning, Psychomotor Speed, Inspection Time, and Knowledge. The latter two were measured by only one task each and the knowledge measure was included as a 'control' task in that the construct it assesses does not decline with age nor would it be expected to be modifiable by intervention. Excepting the tasks measured by speed of response, parallel versions of each task were presented at each assessment visit unless otherwise specified. For a detailed description of the test battery see Appendix 1.
All measures were completed in small groups of up to 7 people, in CSIRO's Cognitive Laboratory, and administered by 2 trained research officers. The testing sessions lasted approximately four hours and consisted of four sections of around 50 min duration, with a break of approximately 10 min between each section. During this break participants were provided with standard refreshments, including decaffeinated tea/coffee. The speed tasks were interspersed between the accuracy based tasks and to also help minimise fatigue, paper and pencil tasks were interspersed with computerised tasks, which were administered in an adjacent room. The order of tasks was fixed across participants and testing sessions. Instructions and practice items preceded each test, and questions were clarified before commencement.
At study commencement, a questionnaire ascertained the following demographic and health information: sex, date of birth, and occupation (prior main occupation if retired); diagnosed medical conditions and regularly taken medications; supplement use and duration; and, for women, the use of hormone replacement therapy. A smoking questionnaire assessed current smoking status, the number of cigarettes/cigars/pipes smoked per day, and smoking history, from which pack years was derived (number of pack years = (number of cigarettes smoked per day × number of years smoked)/20) to measure lifetime direct smoking exposure. Changes to smoking habits were captured at study end.
Information was updated regarding medical and health details during the first and then subsequent visits to the CSIRO clinic.
During the assessment sessions, questionnaires were administered to gather other relevant demographic information: highest level of education achieved, country of birth and if relevant, the age of arrival in Australia and years of speaking English, parents' birth country and occupations, perceived ethnicity, cultural determinants of family diet in childhood, familial history of diagnosed dementia, current income level (as per the Australian census, 2006), and relationship status. Frequency of computer use, handedness, and possible hearing and vision impairments (including colour blindness) were also recorded. Participants' level of engagement in day-to-day mental or cognitive activities was assessed by self-reported minutes per week doing challenging puzzles, self-directed learning, managing finances, or teaching/attending classes. At study end participants stated which experimental group they thought they were in, to assess adequacy of blinding.
Physical activity, functional capacity, and well-being
To assess changes in functional capacity and well-being, and potentially control for any effect of changes in mood or physical activity on cognitive functioning, validated questionnaires assessing everyday functioning (The Late Life Function and Disability Instrument [47, 48]), depression (Centre for Epidemiology Studies Depression Scale ), self-reported cognitive functioning (Cognitive Failures Questionnaire , Prospective and Retrospective Memory Questionnaire ), subjective well-being (mood: Positive and Negative Affect Scale-Extended ; life satisfaction: Diener's Satisfaction with Life Scale , the Personal Wellbeing Index ), health-related quality of life (SF-36 , the sleep scale from the Nottingham Health Profile ), and physical activity (The Yale Physical Activity Survey ) were compiled as a general health and wellbeing questionnaire, which participants completed for each assessment. See Appendix 2 for descriptions of the questionnaires and derived scores.
The Victorian Cancer Council FFQ [58, 59] was used to measure average daily dietary intake, and a Polyunsaturated Fatty Acid FFQ  was used to assess usual dietary intake of n-3 and n-6 PUFAs. The PUFA FFQ was administered in paper format and scored using the electronic version. Lifetime seafood intake was assessed via the Lifetime Diet Questionnaire .
Anthropometry and blood pressure
Height was measured at baseline, using a wall-mounted stadiometer (SECA, Hamburg, Germany) and weight was measured at each assessment session using a calibrated digital precision scale (UC-321PBT; A&D Medical, Sydney, Australia). Body Mass Index was calculated as weight (kg)/height (m)2 to examine as a potential covariate. Blood pressure was measured in the left arm and after the subject had been seated supine for 5 minutes, using an automated sphygmomanometer (Sure Signs V3; Philips Medical Systems, Andover, MA), and the average of three consecutive readings was used as the measured value for each assessment.
Blood was collected into Vacutainer tubes containing EDTA, cooled to 4°C and separated into plasma (for malondialdehyde and carotenoids analyses) and erythrocytes (for analysis of fatty acids) by centrifugation. Erythrocytes were washed twice with isotonic saline and reconstituted to the original volume in water to lyse the cells. For glucose analyses, plasma was separated from blood collected into Vacutainer tubes containing EDTA + NaF. Serum was separated from clotted blood for lipid analyses. Aliquots of plasma, serum and lysed erythrocytes were stored at -80°C until analysis.
Serum vitamin B12 and folate, plasma homocysteine, and HbA1c were measured at an accredited pathology laboratory (IMVS, Adelaide, Australia). Plasma glucose and serum total cholesterol, triglycerides, HDL, and CRP were measured on a Hitachi 902 clinical analyser using diagnostic kits (Roche Diagnostics, Australia). LDL cholesterol was calculated using the Friedewald Equation . Plasma MDA was measured by a fluorescence high performance liquid chromatography (HPLC) method as detailed by Belobradjic et al .
Carotenoids were measured on a Shimadzu LC10 HPLC fitted with a refrigerated auto sampler and a SPD-MV10 Avp photodiode array detector with a class LC10 chromatography workstation. The chromatographic separations of the fat soluble vitamins and carotenoids were carried out on a Varian Microsorb-MV reverse phase C18 column, 4.6 mm ID × 250 mm length, 5 micron spherical particles .
The analysis of erythrocyte membrane fatty acids was based on a method reported by Ridges et al. , which uses a one-step extraction and transesterification procedure . Solid phase extraction on florisil (Sigma-Aldrich Pty Ltd, Castle Hill, NSW Australia) was used to clean up the toluene extract containing the fatty acid methyl esters (FAME) after the transesterification procedure. Briefly, the toluene was dried under nitrogen, residue redissolved in hexane and loaded onto florisil. The FAME were eluted from the florisil with 10% ether in hexane, dried under nitrogen and re-dissolved in isooctane. An aliquot was injected onto a bonded-phase vitreous silica BPX70 column 30 m × 0.53 mm × 0.5 micron (SGE, Australia) in an Agilent 6890N gas chromatograph equipped with a cool on-column injector. Peak identification was based on a comparison of retention times with Supelco 37 component FAME mix 47885-U (Sigma-Aldrich Pty Ltd, Castle Hill, NSW Australia) and peak areas were measured using ChemStation software.
ApoE genotyping was performed using Polymerase Chain Reaction (PCR) - Restriction Fragment Length Polymorphism method, as described by Hixson et al. . Absolute telomere length was measured by determining the number of TTAGGG hexamer repeats using quantitative real-time PCR as described by O'Callaghan et al .
Latent growth curve modelling (LGCM) will be the main analysis method used to assess the effect of the intervention. LGCM is an appropriate and rigorous methodology to employ for intervention studies that aim to alter a developmental trajectory and has greater power to detect an effect than models traditionally used in examining interventions, such as ANCOVA or repeated measures ANOVA .
Four equally-spaced testing time points were used in this study because, firstly, four time points are required to distinguish between different growth forms (e.g., linear, quadratic) and, secondly, four time points are estimated to give adequate power. The cognitive domains will be modelled as latent variables, providing the cognitive factors prove to be measuring the same construct across time, following the initial planned modelling of the tasks and cognitive factors on baseline data using confirmatory factor analysis (CFA). The main effect of the intervention on each cognitive domain will initially be assessed separately following Muthèn and Curran . The main intervention effect can be tested by modelling the shape of the curve, for the time-points of a construct, for treatment and control groups separately. Both groups are constrained to the same normative growth trajectory, and the treatment slope added to the treatment group. Statistical significance of the mean of the treatment slope indicates a significant main treatment effect. Within this framework, further tests will also examine interaction effects, to determine whether the magnitude of the treatment effect varies as a function of initial cognitive status, and whether other factors, such as carriage of the ApoE-e4 allele, sex, or initial n-3 status, moderate the treatment effect. Similar analyses will be undertaken for well-being.
Power was estimated based on a small effect size (Cohen's d) of .2, because there was no solid basis for an expected effect size, and followed power analysis based on LGCM of an artificial longitudinal treatment dataset . This effect size equated to a difference in slope of .23 SD of the control group slope (set at 20% of initial status). Power here refers to the power to detect a difference between the treatment and control groups in terms of mean rate of decline and/or in the variance of the decline rate at an alpha level of .05. A total sample-size of 350 was required for power of .84. To detect an interaction effect of .40 (moderate size) between initial status and treatment slope in the treatment group, with a main effect of .2, required a sample-size ≌ 275. We aimed for an initial sample of 420, allowing for an attrition rate of approx. 17%, resulting in N = 350 at time 4, giving sufficient power to detect the primary effects.
Data treatment of cognitive indicators
Treatment and analysis of reaction time data was conducted only on trials with correct responses. Unfeasibly long latencies on single trials and latencies less than 200 ms or greater than 3.5 intraindividual standard deviations above the individual's mean on a particular task were set to missing, and individuals' mean latencies then recomputed. Considering the above factors, if a participant had less than 60% of valid trial data on a particular task, their mean reaction time for the task was set to missing. Participants' scores were also set to missing on a particular speed task if their accuracy was less than 70% on that task. For the speed tasks and excepting the inhibition tasks' difference scores, the inverse of mean correct latencies were used, to normalise the distribution  and to express performance as a work rate. For all tasks (both speed and accuracy based), total or mean reaction time scores for data for tasks not completed, or not completed according to correct procedure, were considered missing.
Modifications to cognitive test battery after baseline assessment
Confirmatory factor analyses were planned on the baseline data (N = 391) to assess the relationships between the tasks and constructs and, in particular, to ensure no empirical redundancy between any of the constructs. The cognitive test battery was potentially to be abbreviated for future assessments, on the basis of the outcomes.
In total, on the basis of the above analyses, three tasks from the initial test battery were dropped from the battery for the subsequent assessments.
This methodology and protocol paper provides a detailed overview of the design and implementation of a double-blind, randomised, controlled trial assessing the efficacy of a DHA-rich fish-oil supplement on slowing cognitive decline, over 18-months, in cognitively-normal older adults. This is one of the few trials to date to examine the impact of n-3 LC PUFA supplementation on cognitive decline in cognitively healthy older adults. Advantages of the current study are the comprehensive assessment of cognitive functioning and the multiple measurement points, which allow for assessment of an intervention effect on a trajectory, rather than effects at a specific time point. The significance for society could be wide-ranging if such an available and inexpensive dietary treatment has the possibility to safely reduce cognitive decline in ageing, given the links between cognitive functioning and the societal and personal burden of dependency and decreased quality of life in older age.
Appendix 1: Cognitive test battery
Parallel versions of each task were presented at each assessment visit unless otherwise specified. The Flanker task and the delayed recall tasks for Face Memory and Word Memory were only presented at the first assessment because these tasks were found not to adequately mark the expected latent construct in confirmatory factor analyses of the baseline results, conducted between assessments 1 and 2.
Tests were administered via paper-and-pencil or computer. The order of presentation of tests is indicated in parentheses following the name of the task below. All computerised task scripts, excepting Inspection Time, were written and/or modified using the software Inquisit (v. 2©, Millisecond Software). The Memory Scanning and Inhibition tasks were newly devised.
Vocabulary (#17). 20 items from forms 1 and 2 of the standard, advanced, and extended range vocabulary tests from ETS Kit of Factor-Referenced Tests  comprise this multiple-choice task. A 4-minute time limit was imposed.
Raven's (Standard Plus) Progressive Matrices (#10) . Stimuli are a three by three array of symbols, with the bottom right hand symbol missing. Participants choose, from options given below the matrix, which symbol logically completes the matrix. Twenty items, selected to maintain progressive difficulty, were administered with a ten-minute time limit. The same version was administered at each assessment.
Letter Sets (#13). This task is from the ETS Kit of Factor-Referenced Tests . Stimuli consist of a series of five, four-letter items, which follow a predetermined alphabetical ordering. The task is to determine the rule or pattern linking four of the items and cross-out the item not fitting the rule. Fifteen items comprised this test, administered with a seven minute time limit. The same version was administered at each assessment.
Everyday Problems Test (#19) . This test is an objective measure of adults' ability to solve problems of daily living, presented as printed material, in seven domains considered critical for independent living. The seven domains, similar to those covered by the Instrumental Activities of Daily Living (IADL ), are health (medications), meal preparation/nutrition, phone usage, shopping, financial management, household management, and transportation. Stimulus material is in three formats - directions, charts, and forms - and number of questions is balanced across domains and format. For each item, a stimulus (e.g., prescription medication label, phone fee chart, mail-order form) is presented from which the participant must derive the answer through reasoning, to the accompanying item question. The test at each time point comprised 21 multiple-choice items (selected from the full-test 84-item pool), with a 15 min time limit imposed.
Word Endings (#18). This test is derived from the ETS Kit of Factor-Referenced Tests . The task is to write down as many words ending with a specific set of letters (e.g., 'in'). The score is the number of valid words written within 2 minutes.
Retrieval Fluency (#21). This test is derived from the Woodcock-Johnson III Tests of Cognitive Ability . Participants have to write as many words falling within a given category as possible, in two minutes. The score is the number of things written, bar repeats.
Face Memory - immediate recall (#1). This task, derived from Herzmann, Danthiir, Schacht, Sommer, and Wilhelm , is computerised. 20 photographic portraits are presented with 1.5 mins to study them. Each trial in the subsequent test period presents one of the studied faces next to a completely new face and the task is to indicate which face was the studied face.
First and Last Names (#11). In this paired associates test, derived from the ETS Kit of Factor-Referenced Tests , participants are given 2.5 mins to study a list of 15 first and last name pairs. They are then shown only the last names and allowed 1.5 mins to write down the associated first names.
Word Memory - immediate recall (#2). Word list A from the Rey Auditory Verbal Learning Test  was presented via computer at the first assessment and parallel word lists were presented for subsequent sessions. 15 words (nouns) were presented, one at a time, on screen for one sec duration, with an inter-trial interval of one sec. Immediately after presentation of the word-list, participants write down, in any order, as many of the words recalled.
Face Memory - delayed recall (#29). Approximately 3.5 hrs after presentation of the initial Face Memory task, participants were tested again on the faces studied in that task. Each trial presents one of the studied faces next to a completely new face. The task is to indicate the face that was studied in the initial study phase of the Face Memory task.
Word Memory - delayed recall (#6). Approximately 30 mins after presentation of the initial Word Memory task, participants write down as many words as they can recollect from the list, in any order.
Digit-Symbol Substitution (#12); from the revised version of the Wechsler Adult Intelligence Scale . A code with nine symbols, each uniquely associated with one of the digits "1" through to "9", is presented at the top of the page, with rows of only the digits below. The task is to fill in beneath each digit, in order of presentation, as many of the corresponding symbols as possible, within 90 secs. The score is the number of correct answers.
Finding As (#9); from the ETS Kit of Factor-Referenced Tests . Participants are required to examine columns of words; each column has five words containing the letter "A." The task is to place a line through words that contain "A" and ignore those that don't. The score is the number of words marked correctly within the one-minute time limit.
Number Comparison (#20); from the ETS Kit of Factor-Referenced Tests . This task consists of two columns of number-pairs, the numbers varying from 3 to 12 digits in length. Participants are required to place a cross between pairs of numbers if they are not identical. The score is the number marked correctly within the one and a half minute time limit.
The following tasks are all computerised tasks.
Each of the following tasks presents a processing task prior to the stimulus to be remembered. A reminder of the response keys for the processing task is present on the screen throughout. Each trial consists of two to five screens, depending on the difficulty level of the trial, prior to the cue to recall the stimuli. 12 items comprise the test, with three trials of set-size two, and two each of set-sizes three to five. The score is the mean of the proportion of numbers recalled in the correct serial position for each item.
Counting Span (#15) [79, 80]. Participants must remember digits while concurrently undertaking a counting task. For each screen, a number of blue circles and squares, and green circles are shown. The participant must count and remember the number of blue circles in each screen, and then press a key indicating whether the number was odd or even. After a response, the next set of figures is presented and after the last set, participants have to serially recall the number of blue circles in each of the preceding displays, responding using the keyboard and mouse. If a participant does not correctly recall the serial position of a minimum of one digit in a set-size, the task ends.
Operation Span (#4) [79, 80]. In this task, participants remember words (nouns) while concurrently solving a simple arithmetic problem. Each screen presents a maths equation followed by a word, of the form: "DOES (6 × 2) - 5 = 7? CLASS". Participants look at the word then press a key indicating whether the equation was correct or not. After the response, the next equation/word screen appears and after the final screen, participants write down the words recalled in the correct serial position. Parallel word sets were used in each session.
A windows based version of Inspection Time (#31)  was administered. Participants repeatedly view a target stimulus consisting of two vertical lines, joined at the top by a horizontal line, for varying durations. One vertical line is always shorter, with equiprobability. For each trial, the participant indicates by mouse-press whether the shorter line was on the left or right side. Using an adaptive staircase algorithm , a participant's inspection time is defined as the duration of stimulus exposure (in ms) which has an associated probability of 79% of making a correct response.
For the following reaction time tasks, unless otherwise specified, a practice session of 12 trials precedes the start of each test and each task consists of 30 trials. Feedback (accuracy) is presented after each practice trial. The stimuli in both the practice trials and the test proper are balanced such that an equal number of each potential response is required, and for tasks with more than one condition (e.g., Colour Stroop), task-relevant stimulus characteristics are also balanced across conditions.
Simple and Choice-reaction Time
A simple (#3) and two- (#5) and four-choice (#8) reaction time tasks are included, modified from Danthiir, Wilhelm, and Roberts (Further evidence for a multifaceted model of mental speed: Factor structure and validity of computerised tasks, submitted). The stimuli in these tasks are digits, each associated with arrows pointing in different directions. In all tasks, the stimulus-response coding is present at the top of the screen. Participants must press the appropriate arrow response key, depending on which number is displayed. The number of possible responses increases from one in the simple reaction time task to four in the 4-choice reaction time task.
Odd-man-out (OMO) Reaction Time
Two variants of the OMO task (Letter (#14) and Number (#23) OMO, modified from Danthiir et al. (Danthiir, Wilhelm, & Roberts: Further evidence for a multifaceted model of mental speed: Factor structure and validity of computerised tasks, submitted) are included. The OMO task requires an additional discrimination compared to typical CRT tasks. A string of eight equidistant stimuli are presented on screen. In each string, three stimuli are identical (ie., targets) and different from the other five. Two of the targets are always closer together in relation to the position of the third target. The task is to press a key to indicate if the odd-man-out (i.e., the target furthest from the two closer target stimuli) is to the left or right of the other targets.
Speed of Memory Scanning
Two variants of Sternberg's memory scanning paradigm  were devised, measuring speed of scanning items in short-term memory; one task using numbers (#7) as stimuli and the other, letters (#16). A string of two to five stimuli are presented simultaneously on the screen for 2200 ms. A probe stimulus is then presented and remains onscreen until participants press a key to indicate whether the probe was in the preceding string.
Inhibition (or Interference Control)
The following tasks that were devised tap inhibitory processes, also known as interference control or inhibition of prepotent responses. In all tasks the stimuli vary by a task-relevant and -irrelevant characteristic. The task-irrelevant characteristic is linked (e.g., spatially) via automatic prepotent response to the response dimension, providing two or three stimulus-response compatibility (SRC) conditions depending on the task; congruent, incongruent, and sometimes neutral. Trials are balanced in number across stimulus response congruency (SRC) condition and stimulus. During the test, if an incorrect response was given twice in a row for a particular trial-type (e.g., a green left oriented stimulus) a reminder of the correct response keys was presented for 5 sec.
Simon Task (#30). The typical Simon effect refers to the interference (manifesting both in delayed RT and increase in errors) experienced when the response required by a task is spatially opposite to the location of the stimulus, creating a stimulus-response conflict . Stimuli in this task were a red or green circle, presented on the left or right side of the screen. Participants press the right SHIFT key if the stimulus is green and the left SHIFT key if it is red. Two conditions are present in this task; stimulus-response congruent, where the location of the circle is on the same side as the required response (e.g., a green circle on the right side of the screen) and incongruent, where the location of the circle is on the opposite side of the required response (e.g., a green circle on the left side of the screen). The task comprises 60 trials. The dependent variable in analyses is the difference in RT between the congruent and incongruent conditions.
Colour Stroop Task (#22). This task is based on the well-known Stroop effect . Stimuli are the words blue, yellow, note, or engine, coloured yellow or blue, presented one at a time on the computer screen. The task is to press the appropriate key indicating the colour the word is displayed in. Three conditions are present in this task: stimulus-response congruent, where the word matches the colour it is presented in (e.g., the word "YELLOW, " coloured yellow); incongruent, where the word does not match the colour it is presented in (e.g., the word "YELLOW, " coloured blue); and neutral, where the word is not a colour word (e.g., the word "ENGINE, " coloured yellow). 90 trials are presented and the dependent variable in analyses is the difference in RT between the neutral and incongruent conditions.
Spatial Stroop Task (#25). This task is a measure of the so-called spatial Stroop effect (e.g., ). The word LEFT or RIGHT is presented on the left or right side of the screen. Participants press the Right SHIFT key for the word "RIGHT" and the Left SHIFT key for the word "LEFT." There are two conditions in this task; stimulus-response congruent, where the word matches the side it is presented on (e.g., the word LEFT presented on the left side of the screen) and incongruent, where the word does not match the side it is presented on (e.g., the word LEFT presented on the right side of the screen). 60 trials comprise this task and the dependent variable in analyses is the difference in RT between the congruent and incongruent conditions.
Flanker Task (#24). The flanker effect refers to the influence of flanking distracter stimuli on target identification, whereby identification is facilitated by response-congruent flankers and inhibited by response-incongruent flankers . The target in this case was either the letter "S" or the letter "H, " presented in the centre of the screen, flanked simultaneously by two letters displayed either side of the target letter. If the target letter is an "H" participants press the left SHIFT key and if the target letter is an "S" participants press the right SHIFT key. There are three different conditions: the congruent condition where the flanking letters are all the same as the target letter, the neutral condition where the flanking letters (the letter "P") are different from either of the target letters, and the incongruent condition where the flanking letters are the alternative target letter (i.e. if the target letter is "H" the flanking letters are "S"s and vice-versa). 90 trials are presented. The dependent variable in analyses is the difference in RT between the neutral and incongruent conditions.
For these tasks, participants use only the index finger of their preferred hand. For each task, ten practice trials and 20 test trials are given. The dependent variable is the time taken (in ms) between key presses. Following the prompt to begin, participants press either one key, repeatedly (Simple Movement Time task (#26)), or two keys (vertically adjacent - Up Movement Time task (#27); diagonally adjacent - Diagonal Movement Time task (#28)) as quickly as possible.
Appendix 2: Functional Capacity, physical activity, and well-being questionnaires
Unless stated otherwise, if the original questionnaire specified the time-period to which the questions were directed, this was specified as one month in the current study; the exception was questionnaires relating to mood, which were maintained as pertaining to the previous week to assess state, as opposed to trait, affective states.
The Australian SF-36 version 2  (SF36V2) health survey standard form was used as a measure of health-related quality of life. The SF36V2 comprises 36 items from which are derived eight norm-based scale scores: Physical Function, Role Physical, Bodily Pain, General Health, Vitality, Social Function, Role Emotion, and Mental Health. Two summary scale scores are derived from sums of the eight scale scores, weighted by factor score coefficients, representing physical and mental health (physical component score (PCS) and mental component score (MCS)). The eight scale scores were normed according to 2004 data from a large South Australian population-based health survey [88, 89] and the PCS and MCS were weighted using factor score coefficients computed in this population. On all scales, a higher score indicates better health.
The SF-36 was supplemented with the sleep scale from the Nottingham Health Profile , a generic validated health-related quality of life measure, to assess quality of sleep. The scale consists of five items weighted for severity, which have a maximum sum of 100. Participants answer yes or no to each item and the weights are added on items with a positive response; the higher the score the greater the health problem.
The Yale Physical Activity Survey is a validated [57, 90], interviewer-administered questionnaire specifically designed to be sensitive enough to capture the lower-intensity range of physical activities of older adults. It assesses the level of activity during a typical week in the last month. Minor changes to instructions were implemented to adapt the survey to a format suitable for self-completion. The questionnaire assesses work-, yardwork-, exercise-, caretaking-, and recreational-related physical activity, expressed as minutes per week, as well as current participation in several different types of activity categories believed to reflect physical activity dimensions. A total of eight indices are derived, three summary and five subscale indices, and a seasonal adjustment score can be calculated: 1) Total Time Summary Index, expressed as hours per week; 2) Energy Expenditure Summary Index, expressed as kcal/week-1; and 3) Activity Dimensions Summary Score (sum of five subscale indices, vigorous activity, leisurely walking, moving, standing, and sitting, each derived by multiplying frequency score by duration score for each activity type and then multiplied by a weighting factor (based on relative intensity of activity), expressed as a unit score), expressed as total units.
Depression was measured by the Centre for Epidemiology Studies Depression Scale (CES-D) , a 20-item tool measuring depressive symptoms in the general population; it has been extensively validated, including on older adults [91–93] and is widely used as a screening measure. Frequency of occurrence of symptoms is reported for the past week (as in original) and a higher score is indicative of more depressive symptoms, with a maximum score of 60; scores of 16 to 26 are considered indicative of mild depression and scores of 27 or more indicative of major depression. Because the CES-D uses a 4-point response scale it is more likely to be sensitive to change than other commonly used measures with a 2-choice response format.
The Positive and Negative Affect Schedule - extended version , is a widely validated 60-item scale that assesses two broad mood factors - positive and negative affect - as well as 11 specific affects. Participants indicated on a 5-point Likert scale the extent to which they had experienced the listed emotions over the last week, with higher scores indicating greater affect.
Two measures of life satisfaction were administered to participants; Diener's Satisfaction with Life Scale (SWLS)  and the Personal Wellbeing Index (PWI) . Both of these measures were included to assess cognitive rather than affective assessments of subjective wellbeing.
The SWLS assesses respondents' global evaluation of life satisfaction by stating their level of agreement with five statements regarding their life, using a seven point Likert scale. The total score represents the extent of global life satisfaction, with higher scores indicating greater satisfaction and a maximum score of 35; normative data are available for diverse populations including older adults. The scale has been shown to be sensitive to possible changes in levels of life-satisfaction over time due to the occurrence of positive and negative life-experiences .
The PWI scale measures subjective wellbeing as an aspect of quality of life. The scale consists of eight items assessing satisfaction with various aspects of life and participants rate their level of satisfaction on an 11-point Likert Scale. These eight domain scores are summed to yield an average score which represents subjective wellbeing, higher scores indicative of greater satisfaction, with a maximum of 100.
Self-reported cognitive functioning
The CFQ comprises 25 questions measuring self-reported failures in everyday perception, memory, and motor function. Participants rated on a 5-point Likert scale the frequency of such events over the last 4 weeks. A total score is typically derived, measuring proneness to everyday slips and errors, with higher scores indicating greater perceived frequency of mistakes with a maximum score of 100. However, an investigation of the latent structure of the CFQ, using a large representative adult sample, suggests its items assess three constructs - Forgetfulness, Distractibility and False Triggering - which are sensitive to age-effects .
The PRMQ is a validated [51, 96–98] self-report measure, comprising 16 questions, divided equally to assess the frequency of occurrence of prospective and retrospective memory errors in everyday life. Participants were asked to assess the frequency of these occurrences over the last 4 weeks, on a 5-point Likert scale. The PRMQ has been found to have a tripartite latent structure consisting of a general memory factor plus orthogonal specific factors of prospective and retrospective memory [98, 99], thus the PRMQ total scale can be used as a measure of general self-rated memory or specific memory scores can be derived. The maximum total score is 80, with higher scores equating greater perceived frequency of memory mistakes.
The Late Life Function and Disability Instrument [100, 101] (LLFDI) was designed to assess meaningful change in functioning and disability in older community dwelling adults. Concurrent and predictive validity of the LLPDI has been assessed .
The Physical Function component of the LLFDI  comprises 32 items measuring the level of difficulty experienced in performing specific activities as part of daily routines. The activities are divided into three domains: basic lower extremity functioning, advanced lower extremity functioning, and upper extremity functioning, with additional questions for users of assistive devices. Respondents rate the difficulty experienced on a five-point Likert scale; all items can be added to derive a total functioning score, with a maximum of 160 and higher scores indicative of better physical functioning.
The Disability component of the LLFDI  comprises 16 items measuring two dimensions: the frequency of performance of specific activities as part of daily routines, and the extent of limitation experienced in performing the activities. Respondents rate the frequency of performance and the extent of limitation experienced on a five-point Likert-type scale. Two sub-domains are identified within each dimension: Frequency dimension - 1) Social role (performance of various social and community tasks) and 2) Personal role (performance of various personal tasks); Limitation dimension - 1) Instrumental role (activities at home and in the community) and 2) Management role (tasks that involve minimal mobility or physical activity). Distinct summary scores can be calculated to represent the Limitation and the Frequency dimension, each with a maximum score of 80 and higher scores indicating less limitation and greater frequency, respectively.
a The current trial commenced prior to completion of either of those trials.
Akaike Information Criterion
Centre for Epidemiology Studies Depression Scale
confirmatory factor analysis
Comparative Fit Index
Cognitive Failures Questionnaire
fatty acid methyl esters
food frequency questionnaire
high performance liquid chromatography
latent growth curve modelling
The Late Life Function and Disability Instrument
Mini-mental State Examination
- n-3 LC PUFA:
omega-3 long-chain polyunsaturated fatty acid
Prospective and Retrospective Memory Questionnaire
the Personal Wellbeing Index
randomised controlled trial
Root Mean Square Error of Approximation
SF-36 version 2
Diener's Satisfaction with Life Scale.
We thank the participants for their continued effort in taking part in the trial. We gratefully acknowledge the assistance of the following organisations in recruiting participants: Council on the Ageing (SA), Active Ageing Australia (SA), University of the Third Age (Adelaide, Port Adelaide, Tea Tree Gully, Elizabeth, Campbelltown, and Noarlunga branches), Life Care Churches of Christ, and Lifestyles SA and their retirement villages. We thank EPAX for the generous donation of the Omega-3 concentrate used in the trial and Blackmores Pty Ltd for the generous donation of the placebo supplement and packaging of the Omega-3 concentrate. The trial was supported by the Brailsford Robertson Award 2007-2008 (University of Adelaide and CSIRO Food and Nutritional Sciences), and is funded by a National Health and Medical Research Project Grant (#578800). We would like to thank Dr. Peter Clifton for advice on the placebo supplement, guidance on participant eligibility, and assistance with application for funding and Dr. Manny Noakes for advice on the assessment of dietary intake and assistance with application for funding. We gratefully acknowledge the contributions of the research team to: trial co-ordination, Eva Calvaresi; cohort management, Donna Hughes and Chloe Mount; assistance with wellbeing assessment development, Eva Calvaresi and Chloe Mount; data collection, Emily Adcock, Eva Calvaresi, Kayla de la Haye, Laura Edney, Diane Hosking, Donna Hughes, Haidee Lease, Chloe Mount; nursing expertise, Debbie Davies and Lindy Lawson; medical monitoring for the trial, David Jesudason; performing the genomic assays, Nathan O'Callaghan; performing the biochemical assays, Mark Mano, Rob Muirhead, Vanessa Russell, Cathryn Pape, and Candita Dang; assistance with database management, Julie Syrette; assistance with programming of cognitive tasks and updating of the background information, Diane Hosking; technical support, Ben Robertson.
- Neuringer M, Anderson GJ, Connor WE: The essentiality of n-3 fatty acids for the development and function of the retina and brain. Annu Rev Nutr. 1988, 8: 517-541. 10.1146/annurev.nu.08.070188.002505.PubMedGoogle Scholar
- Uauy R, Dangour AD: Nutrition in brain development and aging: Role of essential fatty acids. Nutrition Reviews. 2006, 64 (5): S24-S33. 10.1301/nr.2006.may.S24-S33.PubMedGoogle Scholar
- Soderberg M, Edlund C, Kristensson K, Dallner G: Fatty-Acid Composition of Brain Phospholipids in Aging and in Alzheimers-Disease. Lipids. 1991, 26 (6): 421-425. 10.1007/BF02536067.PubMedGoogle Scholar
- Suzuki H, Park SJ, Tamura M, Ando S: Effect of the long-term feeding of dietary lipids on the learning ability, fatty acid composition of brain stem phospholipids and synaptic membrane fluidity in adult mice: a comparison of sardine oil diet with palm oil diet. Mech Ageing Dev. 1998, 101 (1-2): 119-128. 10.1016/S0047-6374(97)00169-3.PubMedGoogle Scholar
- Xiao YWL, Xu RJ, Chen ZY: DHA depletion in rat brain is associated with impairment on spatial learning and memory. Biomedical and Environmental Science. 2006, 19 (6): 474-480.Google Scholar
- Moriguchi T, Salem N: Recovery of brain docosahexaenoate leads to recovery of spatial task performance. J Neurochem. 2003, 87 (2): 297-309. 10.1046/j.1471-4159.2003.01966.x.PubMedGoogle Scholar
- Chung W-L, Chen J-J, Su H-M: Fish Oil Supplementation of Control and (n-3) Fatty Acid-Deficient Male Rats Enhances Reference and Working Memory Performance and Increases Brain Regional Docosahexaenoic Acid Levels. 2008, 138: 1165-1171.Google Scholar
- Sugimoto Y, Taga C, Nishiga M, Fujiwara M, Konishi F, Tanaka K, Kamei C: Effect of docosahexaenoic acid-fortified Chlorella vulgaris strain CK22 on the radial maze performance in aged mice. Biol Pharm Bull. 2002, 25 (8): 1090-1092. 10.1248/bpb.25.1090.PubMedGoogle Scholar
- van Gelder BM, Tijhuis M, Kalmijn S, Kromhout D: Fish consumption, n-3 fatty acids, and subsequent 5-y cognitive decline in elderly men: the Zutphen Elderly Study. 2007, 85: 1142-1147.Google Scholar
- Kalmijn S, Launer LJ, Ott A, Witteman JC, Hofman A, Breteler MM: Dietary fat intake and the risk of incident dementia in the Rotterdam Study. Ann Neurol. 1997, 42 (5): 776-782. 10.1002/ana.410420514.PubMedGoogle Scholar
- Morris MC, Evans DA, Tangney CC, Bienias JL, Wilson RS: Fish consumption and cognitive decline with age in a large community study. Archives of neurology. 2005, 62 (12): 1849-1853. 10.1001/archneur.62.12.noc50161.PubMedGoogle Scholar
- Kalmijn S, Feskens EJM, Launer LJ, Kromhout D: Polyunsaturated Fatty Acids, Antioxidants, and Cognitive Function in Very Old Men. Am J Epidemiol. 1997, 145 (1): 33-41.PubMedGoogle Scholar
- Barberger-Gateau P, Letenneur L, Deschamps V, Peres K, Dartigues J-F, Renaud S: Fish, meat, and risk of dementia: cohort study. BMJ. 2002, 325 (7370): 932-933. 10.1136/bmj.325.7370.932.PubMedPubMed CentralGoogle Scholar
- Morris MC, Evans DA, Bienias JL, Tangney CC, Bennett DA, Wilson RS, Aggarwal N, Schneider JA: Consumption of Fish and n-3 Fatty Acids and Risk of Incident Alzheimer Disease. Archives of neurology. 2003, 60 (7): 940-946. 10.1001/archneur.60.7.940.PubMedGoogle Scholar
- Kalmijn S, van Boxtel MPJ, Ocke M, Verschuren WMM, Kromhout D, Launer LJ: Dietary intake of fatty acids and fish in relation to cognitive performance at middle age. Neurology. 2004, 62 (2): 275-280.PubMedGoogle Scholar
- Nurk E, Drevon CA, Refsum H, Solvoll K, Vollset SE, Nygard O, Nygaard HA, Engedal K, Tell GS, Smith AD: Cognitive performance among the elderly and dietary fish intake: the Hordaland Health Study. Am J Clin Nutr. 2007, 86 (5): 1470-1478.PubMedGoogle Scholar
- Dangour A, Allen E, Elbourne D, Fletcher A, Richards M, Uauy R: Fish consumption and cognitive function among older people in the UK: Baseline data from the OPAL study. The Journal of Nutrition, Health and Aging. 2009, 13 (3): 198-202. 10.1007/s12603-009-0057-2.PubMedGoogle Scholar
- Whalley LJ, Fox HC, Wahle KW, Starr JM, Deary IJ: Cognitive aging, childhood intelligence, and the use of food supplements: possible involvement of n-3 fatty acids. American Journal of Clinical Nutrition. 2004, 80 (6): 1650-1657.PubMedGoogle Scholar
- Whalley LJ, Deary IJ, Starr JM, Wahle KW, Rance KA, Bourne VJ, Fox HC: n-3 Fatty acid erythrocyte membrane content, APOE varepsilon4, and cognitive variation: an observational follow-up study in late adulthood. Am J Clin Nutr. 2008, 87 (2): 449-454.PubMedGoogle Scholar
- Heude B, Ducimetiere P, Berr C: Cognitive decline and fatty acid composition of erythrocyte membranes--The EVA Study. Am J Clin Nutr. 2003, 77 (4): 803-808.PubMedGoogle Scholar
- Schaefer EJ, Bongard V, Beiser AS, Lamon-Fava S, Robins SJ, Au R, Tucker KL, Kyle DJ, Wilson PWF, Wolf PA: Plasma Phosphatidylcholine Docosahexaenoic Acid Content and Risk of Dementia and Alzheimer Disease: The Framingham Heart Study. Archives of neurology. 2006, 63 (11): 1545-1550. 10.1001/archneur.63.11.1545.PubMedGoogle Scholar
- Laurin D, Verreaulta R, Lindsay J, Dewailly E, Holube B: Omega-3 fatty acids and risk of cognitive impairment and dementia. Journal of Alzheimer's Disease. 2003, 5: 315-322.PubMedGoogle Scholar
- Chiu CC, Su KP, Cheng TC, Liu HC, Chang CJ, Dewey ME, Stewart R, Huang SY: The effects of omega-3 fatty acids monotherapy in Alzheimer's disease and mild cognitive impairment: A preliminary randomized double-blind placebo-controlled study. Progress in neuro-psychopharmacology & biological psychiatry. 2008, 32 (6): 1538-1544. 10.1016/j.pnpbp.2008.05.015.Google Scholar
- Dangour AD, Allen E, Elbourne D, Fasey N, Fletcher AE, Hardy P, Holder GE, Knight R, Letley L, Richards M, et al: Effect of 2-y n-3 long-chain polyunsaturated fatty acid supplementation on cognitive function in older people: a randomized, double-blind, controlled trial. American Journal of Clinical Nutrition. 2010, 91: 1725-1732. 10.3945/ajcn.2009.29121.PubMedGoogle Scholar
- Freund-Levi Y, Eriksdotter-Jonhagen M, Cederholm T, Basun H, Faxen-Irving G, Garlind A, Vedin I, Vessby B, Wahlund L-O, Palmblad J: Omega-3 Fatty Acid Treatment in 174 Patients With Mild to Moderate Alzheimer Disease: OmegAD Study: A Randomized Double-blind Trial. Archives of neurology. 2006, 63 (10): 1402-1408. 10.1001/archneur.63.10.1402.PubMedGoogle Scholar
- Quinn JF, Raman R, Thomas RG, Yurko-Mauro K, Nelson EB, Van Dyck C, Galvin JE, Emond J, Jack CR, Weiner M, et al: Docosahexaenoic Acid Supplementation and Cognitive Decline in Alzheimer Disease A Randomized Trial. Jama-J Am Med Assoc. 2010, 304 (17): 1903-1911. 10.1001/jama.2010.1510.Google Scholar
- van de Rest O, Geleijnse J, Kok F, van Staveren W, Dullemeijer C, OldeRikkert M, Beekman A, de Groot C: Effect of fish oil on cognitive performance in older subjects: A randomized, controlled trial. Neurology. 2008, 71: 430-438. 10.1212/01.wnl.0000324268.45138.86.PubMedGoogle Scholar
- Yurko-Mauro K, McCarthy D, Rom D, Nelson EB, Ryan AS, Blackwell A, Salem N, Stedman M: Beneficial effects of docosahexaenoic acid on cognition in age-related cognitive decline. Alzheimers Dement. 2010, 6 (6): 456-464. 10.1016/j.jalz.2010.01.013.PubMedGoogle Scholar
- Barberger-Gateau P, Raffaitin C, Letenneur L, Berr C, Tzourio C, Dartigues JF, Alperovitch A: Dietary patterns and risk of dementia: the three city cohort study. Neurology. 2007, 69 (20): 1921-1930. 10.1212/01.wnl.0000278116.37320.52.PubMedGoogle Scholar
- Minihane AM, Khan S, Leigh-Firbank EC, Talmud P, Wright JW, Murphy MC, Griffin BA, Williams CM: ApoE polymorphism and fish oil supplementation in subjects with an atherogenic lipoprotein phenotype. Arteriosclerosis, thrombosis, and vascular biology. 2000, 20 (8): 1990-1997. 10.1161/01.ATV.20.8.1990.PubMedGoogle Scholar
- Plourde M, Vohla M, Vandal M, Couture P, Lemieuxa S, Cunnanea S: Plasma n-3 fatty acid response to an n-3 fatty acid supplement is modulated by apoE 4 but not by the common PPAR-α L162V polymorphism in men. British Journal of Nutrition. 2009, 102: 1121-1124. 10.1017/S000711450938215X.PubMedGoogle Scholar
- Deary IJ, Whiteman MC, Pattie A, Starr JM, Hayward C, Wright AF, Carothers A, Whalley LJ: Ageing: Cognitive change and the APOE-ε4 allele. Nature. 2002, 418 (6901): 932-932. 10.1038/418932a.PubMedGoogle Scholar
- Panza F, D'Introno A, Colacicco AM, Capurso C, Pichichero G, Capurso SA, Capurso A, Solfrizzi V: Lipid metabolism in cognitive decline and dementia. Brain Research Reviews. 2006, 51 (2): 275-292. 10.1016/j.brainresrev.2005.11.007.PubMedGoogle Scholar
- Christensen H: What cognitive changes can be expected with normal ageing?. Australian and New Zealand Journal of Psychiatry. 2001, 35 (6): 768-775. 10.1046/j.1440-1614.2001.00966.x.PubMedGoogle Scholar
- Kotani S, Sakaguchi E, Warashina S, Matsukawa N, Ishikura Y, Kiso Y, Sakakibara M, Yoshimoto T, Guo J, Yamashima T: Dietary supplementation of arachidonic and docosahexaenoic acids improves cognitive dysfunction. Neuroscience Research. 2006, 56 (2): 159-164. 10.1016/j.neures.2006.06.010.PubMedGoogle Scholar
- Sinn N, Bryan J, Wilson C: Cognitive effects of polyunsaturated fatty acids in children with attention deficit hyperactivity disorder symptoms: a randomised controlled trial. Prostaglandins, leukotrienes, and essential fatty acids. 2008, 78 (4-5): 311-326. 10.1016/j.plefa.2008.04.004.PubMedGoogle Scholar
- Newkirk LA, Kim JM, Thompson JM, Tinklenberg JR, Yesavage JA, Taylor JL: Validation of a 26-point telephone version of the Mini-Mental State Examination. Journal of geriatric psychiatry and neurology. 2004, 17 (2): 81-87. 10.1177/0891988704264534.PubMedGoogle Scholar
- Brown AJ, Roberts DC, Pritchard JE, Truswell AS: A mixed Australian fish diet and fish-oil supplementation: impact on the plasma lipid profile of healthy men. Am J Clin Nutr. 1990, 52 (5): 825-833.PubMedGoogle Scholar
- Solfrizzi V, Colacicco AM, D'Introno A, Capurso C, Torres F, Rizzo C, Capurso A, Panza F: Dietary intake of unsaturated fatty acids and age-related cognitive decline: A 8.5-year follow-up of the Italian Longitudinal Study on Aging. Neurobiology of Aging. 2006, 27 (11): 1694-1704. 10.1016/j.neurobiolaging.2005.09.026.PubMedGoogle Scholar
- Foster TC: Bological markers of age-related memory deficits: Treatments of senescent physiology. CNS Drugs. 2006, 20: 153-166. 10.2165/00023210-200620020-00006.PubMedGoogle Scholar
- Dziedzic T: Systemic Inflammatory Markers and Risk of Dementia. American Journal of Alzheimer's Disease and Other Dementias. 2006, 21 (4): 258-262. 10.1177/1533317506289260.PubMedGoogle Scholar
- Katan MB, Deslypere JP, van Birgelen AP, Penders M, Zegwaard M: Kinetics of the incorporation of dietary fatty acids into serum cholesteryl esters, erythrocyte membranes, and adipose tissue: an 18-month controlled study. 1997, 38: 2012-2022.Google Scholar
- Danthiir V, Wilhelm O, Schulze R, Roberts RD: Factor structure and validity of paper-and-pencil measures of mental speed: Evidence for a higher-order structure?. Intelligence. 2005, 33: 491-514. 10.1016/j.intell.2005.03.003.Google Scholar
- Stankov L, Danthiir V, Williams LM, Pallier G, Roberts RD, Gordon E: Intelligence and the tuning-in of brain networks. Learning and Individual Differences. 2006, 16 (3): 217-233. 10.1016/j.lindif.2004.12.003.Google Scholar
- Gregory T, Nettelbeck T, Wilson C: Inspection time and everyday functioning: A longitudinal study. Personality and Individual Differences. 2009, 47 (8): 999-1002. 10.1016/j.paid.2009.07.009.Google Scholar
- Salthouse TA: The processing-speed theory of adult age differences in cognition. Psychological Review. 1996, 103 (3): 403-428.PubMedGoogle Scholar
- Jette AM, Haley SM, Coster WJ, Kooyoomjian JT, Levenson S, Heeren T, Ashba J: Late Life Function and Disability Instrument: I. Development and evaluation of the Disability component. Journal of Gerontology: Medical Sciences. 2002, 57A (4): M209-M216.Google Scholar
- Haley SM, Jette AM, Coster WJ, Kooyoomjian JT, Levenson S, Heeren T, Ashba J: Late Life Function and Disability Instrument: II. Development and Evaluation of the Function component. Journal of Gerontoly: Medical Sciences. 2002, 57A (4): M217-M222.Google Scholar
- Radloff LS: The CES-D scale: A self report depression scale for research in the general population. Applied Psychological Measurement. 1977, 1: 385-401. 10.1177/014662167700100306.Google Scholar
- Broadbent DE, Cooper PF, FitzGerald P, Parkes KR: The Cognitive Failures Questionnaire (CFQ) and its correlates. The British journal of clinical psychology/the British Psychological Society. 1982, 21 (Pt 1): 1-16.PubMedGoogle Scholar
- Smith G, Della Sala S, Logie RH, Maylor EA: Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory. 2000, 8 (5): 311-321. 10.1080/09658210050117735.PubMedGoogle Scholar
- Watson D, Clark LA: Manual for the Positive and Negative Affect Schedule - Expanded Form. 1994, The University of IowaGoogle Scholar
- Diener E, Emmons RA, Larsen RJ, Griffin S: The Satisfaction with Life Scale. Journal of Personality Assessment. 1985, 49 (1): 71-75. 10.1207/s15327752jpa4901_13.PubMedGoogle Scholar
- International Wellbeing Group: Personal Wellbeing Index. 2006, Melbourne: Australian Centre on Quality of Life, Deakin UniversityGoogle Scholar
- Ware JE, Kosinski MA, Dewey JE: How to score version 2 of the Sf-36 health survey. 2000, Lincoln: Quality Metric Inc.Google Scholar
- Hunt SM, Mcewen J, Mckenna SP: Measuring Health-Status - a New Tool for Clinicians and Epidemiologists. Journal of the Royal College of General Practitioners. 1985, 35 (273): 185-188.PubMedPubMed CentralGoogle Scholar
- Dipietro L, Caspersen CJ, Ostfeld AM, Nadel ER: A Survey for Assessing Physical-Activity among Older Adults. Medicine and Science in Sports and Exercise. 1993, 25 (5): 628-642.PubMedGoogle Scholar
- Giles GG, Ireland PD: Dietary Questionnaire for Epidemiological Studies (Version 2). 1996, Melbourne: Cancer Council VictoriaGoogle Scholar
- Hodge A, Patterson AJ, Brown WJ, Ireland P, Giles GG: The Anti Cancer Council of Victoria FFQ: relative validity of nutrient intakes compared with weighed food records in young to middle-aged women in a study of iron supplementation. Aust N Z J Public Health. 2000, 24 (6): 576-583. 10.1111/j.1467-842X.2000.tb00520.x.PubMedGoogle Scholar
- Swierk M, Williams PG, Wilcox J, Russell KG, Meyer BJ: Validation of an Australian electronic food frequency questionnaire to measure polyunsaturated fatty acid intake. Nutrition. 2011, 27 (6): 641-646. 10.1016/j.nut.2010.06.011.PubMedGoogle Scholar
- Hosking D, Danthiir V, Nettelbeck T, Wilson C: Assessing lifetime diet: reproducibility of a self-administered, non-quantitative FFQ. Public health nutrition. 2011, 14 (5): 801-808. 10.1017/S1368980010003174.PubMedGoogle Scholar
- Friedewa Wt, Fredrick Ds, Levy RI: Estimation of Concentration of Low-Density Lipoprotein Cholesterol in Plasma, without Use of Preparative Ultracentrifuge. Clinical chemistry. 1972, 18 (6): 499-Google Scholar
- Belobrajdic DP, Lam YY, Mano M, Wittert GA, Bird AR: Cereal based diets modulate some markers of oxidative stress and inflammation in lean and obese Zucker rats. Nutrition and Metabolism. 2011, 8: 27-10.1186/1743-7075-8-27.PubMedPubMed CentralGoogle Scholar
- Yang CS, Lee M-J: Methodology of Plasma Retinol, Tocopherol and Carotenoid Assays in Cancer Prevention Studies. Journal of Nutrition, Growth and Cancer. 1987, 4: 19-27.Google Scholar
- Ridges L, Sunderland R, Moerman K, Meyer B, Astheimer L, Howe P: Cholesterol lowering benefits of soy and linseed enriched foods. Asia Pac J Clin Nutr. 2001, 10 (3): 204-211. 10.1046/j.1440-6047.2001.00253.x.PubMedGoogle Scholar
- Lepage G, Roy CC: Direct Transesterification of All Classes of Lipids in a One-Step Reaction. J Lipid Res. 1986, 27 (1): 114-120.PubMedGoogle Scholar
- Hixson JE, Vernier DT: Restriction Isotyping of Human Apolipoprotein-E by Gene Amplification and Cleavage with Hhai. J Lipid Res. 1990, 31 (3): 545-548.PubMedGoogle Scholar
- O'Callaghan NJ, Dhillon VS, Thomas P, Fenech M: A quantitative real-time PCR method for absolute telomere length. Biotechniques. 2008, 44 (6): 807-809. 10.2144/000112761.PubMedGoogle Scholar
- Muthén BO, Curran PJ: General longitudinal modeling of individual differences in experimental designs: A latent variable framework for analysis and power estimation. Psychological Methods. 1997, 2: 371-402.Google Scholar
- Greaud VA, Green BF: Equivalence of conventional and computer presentation of speed tests. Applied Psychological Measurement. 1986, 10: 23-34. 10.1177/014662168601000102.Google Scholar
- Ekstrom R, French J, Harman H, D D: Kit of factor-referenced cognitive tests. 1976, Princeton, NJ: Educational Testing ServiceGoogle Scholar
- Raven J, Raven JC, Court JH: Section 3: The Standard Progressive Matrices. Manual for Raven's Progressive Matrices and Vocabulary Scales. 2000, San Antonio, TX: Harcourt AssessmentGoogle Scholar
- Willis SL, Marsiske M: Manual for the Everyday Problems Test. 1993, University Park: Pennsylvania State UniversityGoogle Scholar
- Lawton MP, Brody EM: Assessment of older people: self-maintaining and instrumental activities of daily living. The Gerontologist. 1969, 9 (3): 179-186. 10.1093/geront/9.3_Part_1.179.PubMedGoogle Scholar
- Woodcock RW, McGrew KS, Mather N: Woodcock-Johnson III Tests of Cognitive Ability. 2001, Itasca, IL: RiversideGoogle Scholar
- Herzmann G, Danthiir V, Schacht A, Sommer W, Wilhelm O: Towards a Comprehensive Test Battery for Face Processing: Assessment of the Tasks. Behavior Research Methods. 2008, 40 (3): 840-857. 10.3758/BRM.40.3.840.PubMedGoogle Scholar
- Rey A: L'examen clinique en psychologie. 1964, Paris: Presses Universitaires de FranceGoogle Scholar
- Wechsler D: Manual for the Wechsler Adult Intelligence Scale - Revised. 1981, New York: The Psychological CorporationGoogle Scholar
- Engle RW, Tuholski SW, Laughlin JE, Conway ARA: Working memory, short-term memory and general fluid intelligence: A latent variable approach. Journal of Experimental Psychology: General. 1999, 128: 309-331.Google Scholar
- Kane MJ, Hambrick DZ, Tuholski SW, Wilhelm O, Payne TW, Engle RW: The generality of working-memory capacity: A latent-variable approach to verbal and visuo-spatial memory span and reasoning. Journal of Experimental Psychology: General. 2004, 133: 189-217.Google Scholar
- Burns NR, Nettelbeck T: Inspection time and speed of processing: Sex differences on perceptual speed but not IT. Personality and Individual Differences. 2005, 39: 439-446. 10.1016/j.paid.2005.01.022.Google Scholar
- Wetherill GB, Levitt H: Sequential Estimation of Points on a Psychometric Function. British Journal of Mathematical & Statistical Psychology. 1965, 18 (1): 1-10. 10.1111/j.2044-8317.1965.tb00689.x.Google Scholar
- Sternberg S: High-Speed Scanning in Human Memory. Science. 1966, 153 (3736): 652-654. 10.1126/science.153.3736.652.PubMedGoogle Scholar
- Simon JR: Reactions toward Source of Stimulation. J Exp Psychol. 1969, 81 (1): 174-PubMedGoogle Scholar
- Stroop JR: Studies of interference in serial verbal reactions. J Exp Psychol. 1935, 18: 643-662.Google Scholar
- Macleod CM: Half a Century of Research on the Stroop Effect - an Integrative Review. Psychological Bulletin. 1991, 109 (2): 163-203.PubMedGoogle Scholar
- Eriksen B, Eriksen C: Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception and psychophysics. 1974, 16 (1): 143-149. 10.3758/BF03203267.Google Scholar
- Research HH, (ed.): Findings from Autumn 2004 Health Omnibus Survey. 2004, Adelaide: Harrison Health Research
- Hawthorne G, Osborne R, Taylor A, Sansoni J: The SF36 Version 2: critical analyses of population weights, scoring algorithms and population norms. Quality of Life Research. 2007, 16 (4): 661-673. 10.1007/s11136-006-9154-4.PubMedGoogle Scholar
- Harada ND, Chiu V, King AC, Stewart AL: An evaluation of three self-report physical activity instruments for older adults. Medicine and Science in Sports and Exercise. 2001, 33 (6): 962-970. 10.1097/00005768-200106000-00016.PubMedGoogle Scholar
- Zich JM, Attkisson CC, Greenfield TK: Screening for depression in primary care clinics: the CES-D and the BDI. International journal of psychiatry in medicine. 1990, 20 (3): 259-277. 10.2190/LYKR-7VHP-YJEM-MKM2.PubMedGoogle Scholar
- Hertzog C, Van Alsline J, Usala PD, Hultsch DP, Dixon R: Measurement properties of the Center for Epidemiological Studies Depression Scale (CES-D) in older populations. Psychological Assessment. 1990, 2: 64-72.Google Scholar
- Beekman ATF, Deeg DJH, Van Limbeek J, Braam AW, De Vries MZ, Van Tilburg W: Criterion validity of the Center for Epidemiologic Studies Depression scale (CES-D): results from a community-based sample of older subjects in the Netherlands. Psychological Medicine. 1997, 27 (01): 231-235. 10.1017/S0033291796003510.PubMedGoogle Scholar
- Pavot W, Diener E: Review of the Satisfaction With Life Scale. Psychological Assessment. 1993, 5 (2): 164-172.Google Scholar
- Rast P, Van Boxtel M: Factor structure and measurement invariance of the Cognitive Failures Questionnaire across the adult life span. Assessment. 2009, 16 (2): 145-158. 10.1177/1073191108324440.PubMedGoogle Scholar
- Kliegel M, Jager T: Can the Prospective and Retrospective Memory Questionnaire (PRMQ) predict actual prospective memory performance. Current Psychology. 2006, 25 (3): 182-191. 10.1007/s12144-006-1002-8.Google Scholar
- Mäntylä T: Remembering To Remember: Adult Age Differences in Prospective Memory. The Journal of Gerontology. 1994, 49 (6): 276-282.Google Scholar
- Ronnlund M, Mantyla T, Nilsson LG: The Prospective and Retrospective Memory Questionnaire (PRMQ): factorial structure, relations to global subjective memory ratings, and Swedish norms. Scandinavian journal of psychology. 2008, 49 (1): 11-18. 10.1111/j.1467-9450.2007.00600.x.PubMedGoogle Scholar
- Crawford JR, Smith G, Maylor EA, Della Sala S, Logie RH: The Prospective and Retrospective Memory Questionnaire (PRMQ): Normative data and latent structure in a large non-clinical sample. Memory. 2003, 11 (3): 261-275. 10.1080/09658210244000027.PubMedGoogle Scholar
- Jette AM, Haley SM, Coster WJ, Kooyoomjian JT, Levenson S, Heeren T, Ashba J: Late life function and disability instrument: I. Development and evaluation of the disability component. J Gerontol a-Biol. 2002, 57 (4): M209-M216. 10.1093/gerona/57.4.M209.Google Scholar
- Haley SM, Jette AM, Coster WJ, Kooyoomjian JT, Levenson S, Heeren T, Ashba J: Late life function and disability instrument: II. Development and evaluation of the function component. J Gerontol a-Biol. 2002, 57 (4): M217-M222. 10.1093/gerona/57.4.M217.Google Scholar
- Sayers SP, Jette AM, Haley SM, Heeren TC, Guralnik JM, Fielding RA: Validation of the Late-Life Function and Disability Instrument. Journal of the American Geriatrics Society. 2004, 52 (9): 1554-1559. 10.1111/j.1532-5415.2004.52422.x.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.