Trial

Vitamin D3 – Omega3 – Home Exercise – HeALTHy Ageing and Longevity Trial
Acronym: DO-HEALTH

Prof. Heike A. Bischoff-Ferrari, MD, DrPH
Centre on Aging and Mobility, Zurich, Switzerland

Main objectives:

• To improve healthy ageing in European seniors
• To reduce healthcare costs via the implementation of effective and broadly applicable disease prevention interventions

Specific Objectives:

• To establish whether vitamin D, omega-3 fatty acids, and a simple home exercise program will prevent disease at older age
• To assess comparative effectiveness and cost-benefit of the interventions

2152 seniors will be recruited; of these, at most 1291 (60%) will be apparently healthy (see inclusion criteria), and at least 861 (40%) will be pre-frail (low trauma fall event) community-dwelling seniors.

Study population justification: DO-HEALTH will enroll seniors age 70 years and older for their high risk of chronic disease, their high risk of vitamin D and omega-3 deficiency, and their high prevalence of physical inactivity. To represent the largest part of the senior population, DO-HEALTH will recruit community-dwelling seniors. However, to represent also the pre-frail population at risk of institutionalization, at least 40% of seniors will be enrolled based on a low trauma fall with or without a fracture in the year before DO-HEALTH enrollment

Under the assumption that vitamin D, omega-3 fatty acids, and the home exercise program have distinct mechanisms of action and therefore will have an additive effect on the study endpoints, we chose a 2×2×2 factorial study design as being the most efficient. All analyses will test the effect of each strategy, while controlling for the others

There will be 8 treatment arms.

The 3 primary treatment comparisons are:

1. dietary supplement of 2000 IU vitamin D per day compared to placebo (controlling for the other treatment strategies)
2. dietary supplement of 1 gram of omega-3 fatty acids (EPA+DHA, ratio 1:2, from marine algae) compared to placebo (controlling for the other treatment strategies)
3. Home exercise program (Strength) of 30 minutes 3 times a week compared to a control exercise program (Flexibility) 30 minutes 3 times a week

The DO-HEALTH trial will address 5 primary endpoints:

1. Bone: Incident non-vertebral fractures over 36 months (confirmed by X-ray or medical reports)
2. Muscle: Functional decline (assessed by Short Physical Performance Test Battery at baseline, 12, 24 and 36 months)
3. Cardiovascular: Systolic and diastolic blood pressure change (at baseline, 12, 24 and 36 months)
4. Brain: Cognitive decline (assessed by Montreal Cognitive Assessment at baseline, 12, 24 and 36 months) and in a subset of 802 seniors subject cognitive decline will be assessed at 36 months
5. Immunity: rate of any infection (at baseline, and every 3 months up to 36 months)

A Bonferroni adjustment for multiple comparisons will apply so that p<0.01 is required for significance.

The DO-HEALTH trial will assess additional secondary endpoints that support the primary endpoints and extend to other organ systems:

1. Bone: incidence of hip fractures (at 36 months), incidence of new vertebral fractures (vertebral morphometry in a subset of 1502 seniors with yearly DXA measurements; at 36 months), incidence of total fractures – (combined non-vertebral + new vertebral fractures in subset of 1502 seniors with yearly DXA measurements; at 36 months), risk of bone mineral density decrease in the spine and hip (in subset of 1502 seniors with yearly DXA measurements; at baseline, 12, 24, 36 months)
2. Muscle: rate of falling (rate of any low trauma fall, rate of injurious falls, number of persons who fell; assessed every 3 months over 36 months), reaction time and grip strength (at baseline, 12, 24, 36 months), incidence of muscle mass decrease at the upper and lower extremities (in a subset of 1502 seniors with yearly DXA measurements at baseline, 12, 24, 36 months), musculoskeletal pain (at baseline, 12, 24, 36 months), dual tasking 10 meter gait speed (at baseline, 12, 24, 36 months)
3. Cardiovascular: risk of incident hypertension (at 36 months)
4. Brain: mental health decline, incident depression (at baseline, 12, 24, 36 months), dual tasking gait variability in a subset of 250 participants (at baseline, 12, 24, 36 months)
5. Immunity: rate of any upper respiratory infection, incident flu-like illness, incident severe infections that lead to hospital admission (assessed every 3 months over 36 months)
6. Bone/Cartilage: Arthritis: primary outcome for osteoarthritis will be severity of knee pain (KOOS) in those with symptomatic knee osteoarthritis (based on modified clinical ACR criteria), secondary outcomes for osteoarthritis will be: rate of knee buckling, NSAID use because of knee pain; number of joints with pain (all cartilage endpoints are assessed at baseline, 12, 24, 36 months)
7. Dental:decline in oral health, tooth loss (at baseline, 12, 24, 36 months)
8. Gastro-intestinal: gastro-intestinal symptoms based on Rome III questionnaire (at baseline, 12, 24, 36 months)
9. Glucose-metabolic: fasting glucose and insulin levels, QUICKI andHOMA index, body composition and increase in body fat in the trunk and extremities by DXA (in subset of 1502 seniors with yearly DXA measurements; at baseline, 12, 24, 36 months)
10. Kidney: decline in kidney function – by blood creatinine levels and estimated glomerular filtration rate (at baseline, 12, 24, 36 months)
11. Global Health: quality of life (assessed every 6 months), incident frailty, incident disability regarding activities of daily living (at baseline, 12, 24, 36 months), incident nursing home admissions, rate of acute hospital admissions, mortality (assessed every 3 months)

The DO-HEALTH trial will include an organ-specific biomarker study to support primary and secondary endpoints at a mechanistic level(assessed in 2152 seniors at baseline, 12, 24, 36 months of follow-up):

1. Bone: calcium, phosphate, 25(OH)D, intact PTH, urinary calcium/creatinine ratio (second spot urine), Beta-Crosslaps serum, P1NP
2. Cardiovascular: Troponin T, NT-proBNP, homocysteine, CK, cholesterol, HDL-cholesterol, triglycerides
3. Inflammation: CRP, IL6
4. Gastrointestinal: AST, ALT, gGT, alkaline phosphatase, bilirubin
5. Glucose-metabolic:fasting glucose, insulin
6. Kidney: serum creatinine; calcium/creatinine ratio in second spot urine, serum urea, uric acid
7. Global Health: Ions: sodium, potassium, chloride, magnesium;
   Proteins: total protein, albumin, ferritin, transferrin;
   Hormones: TSH, fT4, fT3, cortisol;
   vitamins: folic acid, vitamin B12, 25(OH)D

Biomarker study for novel bone and muscle functionality: assessed at baseline, 12, 24, 36 months in 2152 seniors:

1. Bone: sclerostin;
2. Muscle: myostatin

Biomarker study for novel inflammation and immunity markers:assessed at baseline, 12, 24, 36 months in  the 802 seniors recruited by 2 centers (Zurich and Basel)

1. Inflammation: TNF-α, IL-10, IL-17, IL-22
2. Cellular immunity: CD3, CD4, CD25, CD127

The DO-HEALTH trial will assess exploratory endpoints that support the primary endpoints, but have limited statistical power (assessed over 36-month follow-up):

1. Bone: Incident repeat fractures (any repeat non-vertebral fractures in all participants, vertebral fractures and total fractures among subset of 1502 seniors with yearly DXA measurements)

Ancillary fracture healing study in all seniors with an incident major osteoporotic fractures at the arm (shoulder, humerus, forearm) and leg (hip, femur, ankle): (a) primary fracture healing endpoint: clinical fracture healing with 3 additional phone calls at 6, 12, 18 weeks after the fracture by PROMIS-HAQ questionnaire (b) secondary fracture healing endpoint: observed functional fracture healing measured with the Short Physical Performance Test Battery and grip strength at regular 12, 24, 36 month visits (c) exploratory fracture healing endpoint: radiological fracture healing with independent assessment of early (6-8 weeks) and late (12 to 14 weeks) consolidation assessment based on standard clinical care X-rays – participants will report a new fracture 7 days of the event (recruitment centre hotline)

2. Muscle: incident sarcopenia (among subset of 1502 seniors with yearly DXA measurements), incident frailty, decline in physical activity (at 36 months)
3. Cardiovascular: major cardiovascular events as a composite endpoint (any event: myocardial infarction, stroke, revascularization procedures of CABG and PCI, incident congestive heart disease, cardiovascular mortality); individual endpoints: myocardial infarction, stroke, incident congestive heart disease, and cardiovascular mortality (assessed every 3 months over 36 months)
4. Brain: incident dementia (at 36 months)
5. Immunity: incident cancer (any cancer, gastro-intestinal, breast cancer in women, prostate cancer in men); rate of implant infections after total hip or knee replacement (due to fracture or osteoarthritis); rate of gastro-intestinal infections (at 36 months)
6. Bone/Cartilage-Arthritis: incident symptomatic knee osteoarthritis; incident symptomatic hip osteoarthritis, incident symptomatic hand osteoarthritis; composite endpoint: incident symptomatic knee, hip or hand osteoarthritis; severity of hip pain in those with prevalent symptomatic hip osteoarthritis, severity of hand pain in those with prevalent symptomatic hand osteoarthritis (assessed every 12 months over 36 months)

Adherence laboratory (assessed at baseline, 12, 24, 36 months follow-up in 2152 seniors): serum 25(OH)D concentrations (measured both by an automated assay and HPLCMS/MS) and plasma PUFA concentrations (EPA, AA, DPA, DHA; measured by a sensitive and selective assay based on gas chromatography coupled to mass spectrometry detection (GC-MS)).

Safety laboratory (assessed at baseline, 12, 24, 36 months follow-up in 2152 seniors): serum calcium and serum creatinine

1. Age 70 years or older
2. Mini Mental State Examination Score of at least 24
3. Living in the community
4. Sufficiently mobile to come to the study centre
5. Able to walk 10 meters with or without a walking aid and able to get in and out of a chair without help
6. Able to swallow study capsules
7. Able and willing to participate, sign informed consent (including consent to analyze all samples until drop-out or withdrawal) and cooperate with study procedures

1. Consumption of more than 1000 IU vitamin D/day in the 6 months prior to enrollment, or unwillingness to limit vitamin D intake to the current standard of 800 IU/day of vitamin D during the course of the trial
   • Provision 1: an individual who consumed an average vitamin D dose between 1000 and 2000 IU vitamin D/day in the 3 months prior to enrollment, may be enrolled after a 3-month wash-out period where the maximum daily intake is limited to 800 IU vitamin D.
   • Provision 2: an individual who consumed an average vitamin D dose higher than 2000 IU/day in the 3 months prior to enrollment, may be enrolled after a 6-month wash-out period where the maximum daily intake is limited to 800 IU vitamin D.
2. Unwillingness to limit calcium supplement dose to 500 mg per day for the duration of the trial
3. Taking omega-3 fat supplements in the 3 months prior to recruitment and or unwilling to refrain from use of omega-3 supplements for the duration of the trial
4. Use of any active vitamin D metabolite (i.e. Rocaltrol, alphacalcidiol), PTH treatment (i.e. Teriparatide), or Calcitonin at baseline and unwillingness to forego these treatments during the course of the trial
5. Current or recent (previous 4 months) participation in another clinical trial, or plans of such participation in the next 3 years (corresponding to DO-HEALTH length)
6. Presence of the following diagnosed health conditions in the last 5 years: history of cancer (except non-melanoma skin cancer); myocardial infarction, stroke, transient ischemic attack, angina pectoris, or coronary artery intervention
7. Severe renal impairment (creatinine clearance ≤ 15 ml/min) or dialysis, hypercalcaemia (> 2.6 mmol/l)
8. Hemiplegia or other severe gait impairment
9. History of hypo- or primary hyperparathyroidism
10. Severe liver disease
11. History of granulomatous diseases (i.e. tubercolosis, sarcoidosis)
12. Major visual or hearing impairment or other serious illness that would preclude participation
13. Living with a partner who is enrolled in DO-HEALTH (i.e. only one person per household can be enrolled)
14. Living in assisted living situations or a nursing home
15. Temporary exclusion: acute fracture in the last 6 weeks
16. Epilepsy and/or use of anti-epileptic drugs
17. Individuals who fell more than 3 times in the last month
18. Osteodystrophia deformans (M. Paget, Paget’s disease)
19. For study center in Germany only: persons who are institutionalized / in prison by court order (§40, Abs. 1, Art. 4, “Gesetz über den Verkehr mit Arzneimitteln”)

2152 participants (269 in each of the 8 treatment groups). See study schema page 39 (Figure 3) for detailed allocation of treatment groups

The sample size of 2152 senior participants was chosen to have sufficient power for the most critical primary and secondary endpoints (any non-vertebral fracture and hip fracture) and, based on prior experience, the expectation that 68% will complete the entire 3-year study. However, the subjects who drop out early will provide partial data since our analyses will be based on the intention-to-treat principle. Under these assumptions and assuming an additive effect of the 3 interventions, as expected from our pilot study (7), we have > 90% power for all 5 primary endpoints (with Bonferroni adjustment for multiple comparisons).On the other side, if an interaction between interventions is assumed, the power still exceeds 80%. Contributing to the statistical power for the most critical fracture endpoints is the fact that at least 40% of seniors will be recruited based on a prior fall with or without a fracture, which confers an increased risk of sustaining a fracture endpoint during the course of the trial

Stratified Block Randomization with block sizes of 16 individuals (2 for each of the 8 treatment combinations) will be performed by the DO-HEALTH randomization software. Labeling of study intervention will be performed by a central randomization centre in Switzerland.

Stratification variables: recruitment centre (7 centers), low trauma fall during previous 12 months prior to the randomization day (yes/no), gender, and age (70 – 84 and 85+). The recruitment of at least 40% of participants who fell during the last year will be enforced at each of the 7 recruitment centers. Gender and age distribution will be monitored within each recruitment centre with the DO-HEALTH randomization software. If gross imbalance (less than 30% of participants in a stratum) is detected within a centre, recruitment strategies for the centre will be adapted to boost recruitment of participants of underrepresented category

Follow-up Period: 3 years

Four Clinical Visits: baseline, follow-up visits at 12, 24, and 36 months.
Ideally, all visits should be completed in one day, however, a center is allowed to split the visits over two days to streamline the workflow, or prevent the participants from being overwhelmed. Two conditions need to be respected:

• at baseline, the participant is randomized on the second day of the split visit
• the second part of any visit (baseline, 12, 24, 36 month) should take place no later than 1 week from the first part

Nine Telephone Follow-up Interviews: every 3 months between the clinical visits

Justification of in- person follow-up: Clinical visits are at baseline, 12, 24 and 36 months in all 2152 participants, with additional 3-monthly contact by phone calls in all participants. Based on prior experience, we designed DO-HEALTH to have an age appropriate follow-up in 3-month intervals. This will allow for a high-quality endpoint assessment. Seniors tend to forget health events, which will be avoided in the best possible manner in our follow-up structure. This will be also supported by an event diary. Furthermore, based on prior experience, seniors tend to be overwhelmed by questionnaires sent to them by mail. We believe that the in-person follow-up will address that concern.

Justification of 3-year treatment: Based on our pilot studies – pooled participant-level analysis of 11 RCTs (8) plus DO-HEALTH pilot trial (7), the 36 months of follow-up will provide enough participant-years to accumulate a sufficient number of events for the most critical primary (any non-vertebral fracture) and the most critical secondary endpoint (hip fracture), especially with respect to the 50% recruitment of pre-frail seniors with a prior fall/fracture event. Further, 3 years of follow-up will establish long-term efficacy and safety data for the 3 interventions.

Adherence to the interventions will be assessed at each clinical visit by:

1. Count of unused study capsules
2. Measurement of blood levels of 25(OH)D and PUFAs
3. 3-month in person adherence and comparative effectiveness assessed by a questionnaire (why and why not participants are adherent to the study interventions)
4. All participants will be given a personal diary to record their adherence to the study intervention and the exercise program. The diary will be collected at each clinical visit; it will not constitute a direct source of information for endpoint assessment, but might be used pro re nata for confirmation of participant reported events
5. Standardized motivation for the exercise program at 3-monthly in person contacts and reminder to take study capsules

DO-HEALTH will use specially designed electronic data collection system. The software will be able to detect and flag data entry errors in real time.

All data collected at the study centre will be securely transferred to the Central Data Bank at the end of each working day. Each study centre will have dedicated backup storage that will keep all data collected by the study centre. Central Data Bank will store the entire study database and will be equipped with a dedicated backup storage and additional secure off-site backup storage. Study database will be backed up daily.

DO-HEALTH will employ specially designed data collection procedures facilitating accurate data collection and double-checking of critical data points before a clinical visit or a phone call is complete. Data collection procedures will be augmented by error detection software routines programmed into the electronic data collection system.

Blood samples collected from the participants will be temporarily stored at each study centre at -80°C. Aliquots will be transferred to the central biobank on a 3-monthly basis in dry ice containers, so that the cold chain is not disrupted, and stored in the central DO-HEALTH biobank at Fisher Clinical Services GmbH (Allschwil, Switzerland) under the same temperature conditions until analyzed (for the DO-HEALTH biomarker study and later exploitation).

All statistical analyses of DO-HEALTH participant data will be performed centrally at the coordinating DO-HEALTH centre at the University of Zurich by the DO-HEALTH data management team, based on pre-defined analysis plans and under supervision of the DO-HEALTH head biostatistician (Prof. E. Orav). All analyses for the DO-HEALTH trial will be performed based on the intent-to-treat principle after all study data are collected. Further, a per-protocol analysis is foreseen among adherent individuals (≥80% adherence).

The statistical models will assess the effect of treatment, considering stratification variables, as well as pre-defined potential confounders. We will assess the effect of one treatment while adjusting for the other treatments, also testing for a potential interaction. Detailed statistical analyses plans for each endpoint will be developed prior to analysis as a part of the data control work package of DO-HEALTH. The plan will define statistical models for each endpoint and contain printouts of SAS procedures necessary to perform the analysis. All necessary SAS procedures will be pre-programmed and the code will be kept as electronic files to be used in the final analyses.

To test whether vitamin D3 and/or omega-3 fatty acids (EPA+DHA) supplements and/or the simple home exercise program reduce the risk of primary and secondary endpoints differentially by gender, age (70-84; 85+), body mass index, baseline physical activity, baseline serum 25(OH)D levels, baseline PUFA levels, previous fall (last year), previous fracture (last 10 years), FRAX – estimated absolute fracture risk, baseline symptomatic knee OA, and baseline calcium and protein intake (diet + supplements).

Each capsule of active study intervention contains either 1000 IU of Vitamin D₃ in the form of crystalline cholecalciferol in a medium chain triglycerides and/or 500 mg of eicosapentaenoic acid (EPA) plus docosahexanoic acid (DHA) in a ratio of 1:2. Placebo capsules contain high oleic sunflower oil replacing vitamin D₃ and/or omega-3 marine fatty acids.

Daily dose of study intervention: 2 capsules

Route of administration: orally – preferentially with dinner – or – when participants take all of their interventions (breakfast or lunch or even at bedtime)

Duration of administration: 3 years

All study intervention is provided by DSM and packaged in plastic bottles containing 62 capsules each (one-month supply). Each participant is supplied with 12 bottles once a year at the end of the clinical visit. Study capsules are to be stored at room temperature between 15 and 25 Celsius degrees. To avoid unpleasant after taste and thereby unblinding, capsules will be coated so that the capsule will dissolve only in the jejunum.

Dietary supplement of 2000 IU Vitamin D per day – In DO-HEALTH pilot trial, 2000 IU vitamin D compared to 800 IU vitamin D per day reduced hospital readmission by 39%, fall-related injuries by 60%, and severe infections by 90%. Further, 2000 IU vitamin D raised 25-hydroxyvitamin D levels to at least 75 nmol/l in over 90% of participants at 6 and 12 month follow-up, andwas safe as demonstrated by repeated serum and urinary calcium assessments. Moreover, in another pilot study to DO-HEALTH where we pooled the source data of 11 double-blind randomized trials, fracture reduction was only significant at the highest intake quartile of vitamin D (792-2000 IU/day).The safety of 2000 IU vitamin D per day is further supported by our benefit-risk analysiswhere a safe upper intake level of 10,000 IU of vitamin D per day was estimated. In addition, the newest recommendation by the Institute of Medicine increased the safe upper limit for vitamin D intake from 2000 to 4000 IU per day.

Dietary supplement of 1 g omega-3 fatty acids per day – Is recommended by the American Heart Association for cardio-protection and was beneficial with respect to secondary cardiovascular prevention in one trial. Because the optimal ratio of EPA to DHA is unknown we have selected a 1:2 ratio of EPA (eicosapentaenoic acid) to DHA (docosahexaenoic acid). Health authorities recommend 400 mg to 1 g/d for cardio protection. A total dose of 850 mg/d was used in the GISSI-Prevenzione Trial (EPA to DHA ratio, 1:2) and AREDS (EPA to DHA ratio, 2:1), and a dose of 1.8 g/d of EPA was used in JELIS. Given that the average intake of EPA+DHA is 100 to 350 mg/d in many parts of Europe, the proposed intervention of 1 g/d is expected to increase the average participant’s omega-3 intake by a factor of 3 to 10. Health risks associated with marine omega-3 fatty acids are believed to be minimal and doses of up to 3 g/d are generally recognized as safe. Although omega-3 fatty acids have potential antithrombotic effects, systematic reviews of data from small, short-term trials suggest that omega-3 fatty acid supplements do not increase the risk of clinically significant bleeding at doses of up to 4 g/d, even in combination with anticoagulant interventions such as aspirin or warfarin.

For the exercise intervention, we will test a simple exercise program that was validated in the DO-HEALTH pilot trial. The simple well-defined exercise home program was successful in reducing the rate of falling by 25% (significant) and the rate of fracture by 56% (95% CI: −82% to 9%; P =0.08; pilot study was not powered for fracture reduction) in the first year after hip fracture. Seniors who performed the intervention at least once a week also had a significant improvement in lower extremity function (timed up-and-go) and reaction time (repeated chair stands). The program will be instructed by a physiotherapist at the baseline visit. DO-HEALTH will develop a motivational and animated video to instruct the Strength exercise program as the main exercise intervention in DO-HEALTH. In addition, a paper format of the program will be developed. For the exercise control group, a similar animated video will be designed for a Flexibility exercise program, which will serve as a high quality control intervention. The Flexibility exercise program will focus on joint mobility rather than strength and balance. Participants are asked to do the exercise programs 3 times per week for 30 minutes. The study staff and participants will be informed that both interventions are beneficial and all in-person contacts (every 3 month) will be used as a standardized motivational opportunity. The physiotherapist who will instruct the exercise program during the baseline visit is not a part of the assessment team in DO-HEALTH. Seniors who want to step up the intensity of the program will be suggested to repeat the whole program, thus increasing the training time.

We have not included calcium supplementation as a component of the intervention as at the higher dose of vitamin D intake, additional high dose calcium supplementation (1000+ mg) was not beneficial for fracture reduction in a pilot study of DO-HEALTH pooling the source data of 11 double-blind RCTs; and calcium supplements given without vitamin D may increase the risk of hip fracture, the risk of myocardial infarction and the risk of kidney stones. In the participant information, we encourage to cover calcium needs sufficiently from dietary sources (1000 mg per day). A personal intake of calcium supplements will be allowed in DO-HEALTH up to a dose of 500 mg per day.

Feasibility of DO-HEALTH clinical visit burden and recruitment potential has been established in a pilot trial among the target population (Zurich Disability Prevention Trial: NCT01017354); and optimal dosing and safety of the interventions have been established in 2 pilot trials, one pooled analysis of 11 double-blind RCTs, a benefit risk assessment and 3 meta-analyses. DO-HEALTH has more than 80% power for the primary endpoints and all secondary endpoints. Further, DO-HEALTH will measure 25(OH)D and PUFA levels and extended biomarker studies in all participants at all 4 clinical visits. Also, a cost-benefit analysis will be included in DO-HEALTH. All 3 interventions have a high-quality control group (placebo for the two nutrients and Flexibility exercise programintervention for the exercise control group).

The trial will test only one dose of each agent rather than examining multiple doses. However, the dose for each agent was chosen on the basis of an extensive review of available evidence, including several pilot studies and meta-analyses. Because the trial population is older, the results may not be transferable to younger men and women

To validate self-reported study endpoints, the recruitment centres will send a medical release form to obtain confirmation from general practitioners and/or hospital/physician records for all endpoints reported during the last three months of participants’ follow-up time (participants authorize DO-HEALTH staff to do so with their written informed consent). The request to physicians will be accompanied by a copy of the consent form and a cover letter explaining that the participant is enrolled in the DO-HEALTH clinical trial expressing the scientific importance of record validation to the treating physician/ hospital. If there is no response within 1 month, a second request will be mailed, followed by a phone call by the respective DO-HEALTH recruitment centre (all efforts will be documented in the central data base).

After records are obtained, the central data management is notified and a de-personalized copy of the records only indicating the subject ID is faxed to the DO-HEALTH coordinating centre at the University of Zurich. There, an Endpoints Committee of Physicians, blinded to the randomized treatment assignment, will review the file and, and using a pre-defined protocol (diagnostic cards), will confirm or disconfirm the case.

DO-HEALTH ensures that GCP guideline compliance is a key quality target. Each recruitment centre will be monitored repeatedly and reports will be sent to the Data Safety and Monitoring Board.

We developed an electronic food frequency questionnaire (FFQ) as a tool that is consistent across countries in DO-HEALTH and targeted at the senior population. The development was led by Prof. Walter Willett (collaborator and advisor to DO-HEALTH) and Simonetta Salvini (diet expert and trainee of Prof. W. Willett at the Dept. of Nutrition at Harvard School of Public Health) in collaboration with the DO-HEALTH coordinating centre at the University of Zurich (Prof. Heike Bischoff-Ferrari).