The Efficacy of Individualized Dietary Advice in Improving Diet Quality and Cardiovascular Health

Queen's University Belfast

Status: Completed

Conditions: Cardiovascular Diseases

Treatments: Behavioral: Personalised dietary advice and educational material; Behavioral: Educational material

Study type: Interventional

Funder types: Other

Identifiers

NCT03854461

DK120870-01

Details and patient eligibility

About

This study is a a randomized controlled parallel group dietary intervention conducted over six months in participants at high risk of cardiovascular disease living in Ireland (North and South) to evaluate the efficacy of individualized dietary advice incorporating biomarker profiles in improving diet quality and cardiometabolic outcomes.

Full description

Study design: a randomized controlled parallel group dietary intervention will be conducted over six months in participants at high risk of CVD living in Ireland (North and South) to evaluate the efficacy of individualized dietary advice incorporating biomarker profiles in improving diet quality and cardiometabolic outcomes.

Population and recruitment

Population characteristics: Participants will be at high risk (>20% over 10 years) of developing CVD. Established recruitment mechanisms, used in several previous and ongoing dietary intervention studies, from general practices, hospital outpatient clinics, and from the general population (via press release, radio interviews, poster placement) will be utilized.Recruitment will take place over an 18 month period.

Inclusion criteria: Participants will be considered eligible to enroll in the study if they are overweight (BMI >27 and ≤ 45 kg/m2), have low Prime Diet Quality Score (PDQS <15)27 and a combination of risk factors which places them at high total risk (estimated multifactorial CVD risk ≥ 20% over ten years) of developing atherosclerotic CVD for the first time.

Exclusion criteria: Participants will be ineligible to enroll in the study if they have established diabetes mellitus, CVD or a medical condition or dietary restriction(s) that would substantially limit their ability to complete the study requirements, have an excessive alcohol consumption (>28 Units/week for men or >21 Units/week for women), have a low predicted likelihood to change dietary habits or are unable to provide informed consent.

Randomization: A personal health and medical history questionnaire, together with Joint British Societies CVD risk prediction charts and an assessment of participant's baseline PDQS using a 42-item questionnaire will be used to assess eligibility. This questionnaire is validated in the US, but will be validated in the Irish population. Those eligible and willing to take part will give informed written consent, and will be randomized (randomization scheme generated using www.randomization.com with random block sizes and research staff administering the randomization or conducting endpoint assessment blinded) to either an education only intervention or individually tailored food marker-based dietary advice.

The intervention will take place over a six month period. Participants will be interviewed towards the end of the intervention (for those in individually tailored support arm), or at drop-out (for all participants who drop out), to elicit their views on the intervention.

Dietary intervention: Participants will be randomized to one of two groups:

1. Control - participants will be given written educational material to encourage dietary change towards a better quality diet. Educational material will be adapted from resources used in previous dietary intervention studies.
2. Personalized dietary advice - participants will also be given the educational material, but will be encouraged to make dietary change using individualized recommendations incorporating food markers of specific components of a high quality diet. Food markers of diet quality will be used to develop an algorithm to deliver personalized dietary advice. Based on biomarker data, a system of categorization of biomarker status will be developed, alongside dietary advice related to this biomarker categorization, and decision trees created, as previously described in the Food4Me study, to ensure standardized delivery of advice within this intervention arm.

The dietary advice will be delivered via email and online, with monthly contact and the messages being reinforced during each direct contact with the study team for outcome assessment. Therefore frequency and duration of contact will be similar in both intervention groups.

Outcome measures: Outcomes will be assessed during a clinic visit at 0, 3 and 6 months, and will include dietary intake, anthropometry, other lifestyle behaviors, blood pressure, biological sample collection for nutritional status biomarker and health biomarker analysis, as well as fecal samples for metagenomic and transcriptomic profiling. Outcome data will be collected using the following methodology:

• Demographic/lifestyle information (e.g., alcohol consumption, medication use).
• Dietary intake using an online version of the PDQS. This will also be assessed six months after the intervention completes, to determine whether the intervention influences longer term adherence to a better quality diet.
• Physical activity using an objective physical activity monitor.
• Anthropometric data (weight and height measured using calibrated scales and stadiometer respectively, waist and hip circumference); body composition by bioelectrical impedence (Bodystat 1500).
• Blood pressure will be measured twice from the right arm, using an automated Omron sphygmomanometer, with the participant sitting quietly for at least five minutes.
• Fasting blood samples will be drawn from the antecubital vein and immediately separated into plasma/serum for the proposed assays detailed below. It will be stored at -70°C until analysis.
• Fasting morning urine sample stored at -70ºC until analysis.
• In a subset of participants (n=70, 35 in each arm), stool samples will be collected at baseline and the end of intervention for the analysis of changes in gut microbiota composition and functionality, as previously described.

Laboratory methodology: All methods are in routine use in participating laboratories and are performed with careful attention to quality control and with participation in external quality control schemes where available. All laboratory analysis will be conducted blind, i.e. the analyst will not know to which intervention the participant has been assigned.

Biomarkers of cardiometabolic risk: Serum lipids (total cholesterol, HDL-C, and TG) and whole blood HbA1c will be assessed by enzymatic assays (Randox Ltd, Crumlin, NI and Glen Bio, Antrim, NI) on an I-Lab 600 auto-analyzer. LDL-C will be calculated.

Biomarkers of diet quality: the following biomarkers will be used to assess nutritional status response to the dietary intervention: red meat - methylhistidines and carnosine; white meat - TMAO, guanidoacetate, anserine; fruit and vegetables -- proline betaine, isothiocyanates, S-methyl-L-cysteine sulfoxide, and tartaric acid; wholegrains - alkylresorcinols; coffee - phenolic acids, N-methylpyridinium, trigonelline, 2-furoylglycine; SSBs -- formate, citrulline, taurine, and isocitrate)red meat- methylhistidines; oily fish- omega-3 index; fruit and vegetables- total carotenoids and proline betaine; wholegrain- alkylresorcinols.

Gut microbiota composition and functionality: Sample handling and nucleic acid purifications will be performed following an established protocol.166 Illumina DNA-seq libraries with fragment length of ~300 bp will be generated, with Pippen prep size selection. For RNA-seq, 5ug of RNA will be enriched for mRNA and subjected to strand-specific cDNA synthesis using established protocols. Paired-end sequencing (e.g. 150x150 nt) will be performed on the Illumina HiSeq platform.

Primary Outcome: Change in Prime Diet Quality Score.

Secondary Outcomes:

1. Change in markers of cardiometabolic risk, including blood pressure, lipid profile, HbA1c, body weight.
2. Change in biomarkers of diet quality
3. Gut microbiota composition and functionality (to be conducted in a subset).

Power calculation: Sample size calculations will be confirmed during the validation of the PDQS within an Irish population. However, based on published US data, with a standard deviation (SD) of PDQS of 2.3, and an assumption of a difference in increase in PDQS of 5 between the personalized dietary advice and control groups then, if there are 60 participants in each group, the study would have over 90% power to detect this difference in mean PDQS as statistically significant, at the α=0.05 level. A total of 134 participants will be recruited to allow for approximately 10% drop-out. For the health outcomes, and, based on what was observed in a dietary intervention based on improving diet quality in similar high CVD risk participants, if there are 60 participants in each group, and it is assumed that (i) the SD of the change in diastolic blood pressure (DBP) is 9 mmHg at the end of the study; (ii) a difference in DBP of 4.5 mmHg between the individualized dietary advice and control groups; then the study would have 80% power to detect, as statistically significant at the 5% level, these differences in mean change between the two groups. The investigators will have similar levels of power (~80%) to detect effect sizes of similar magnitude in total cholesterol, HbA1c and body weight. The analysis of gut microbiota composition and functionality will be exploratory in nature.

Data analytical strategy: The repeated end-point measures generated by the study design will be analyzed with techniques appropriate for longitudinal data using the xt panel study procedures available in Stata release 11 (College Station, TX). Initial examination of the correlation structure of the repeated measures will help guide model fitting. The influence of the dietary intervention under study will be assessed by comparing the means of changes in measurements from baseline to 6 months between the biomarker-based individually tailored advice and education only groups. This difference in means will provide an estimate of the effect of intervention and 95% confidence limits will be calculated to indicate its precision. If appropriate, measurements will be logarithmically transformed prior to analysis and interpretation will be made in ratio terms on the original scale.

Enrollment: 150 patients

Sex: All

Ages: 18 to 75 years old

Volunteers: Accepts Healthy Volunteers

Inclusion and exclusion criteria

Participants will be at high risk (>20% over 10 years) of developing CVD.

Inclusion Criteria:

Participants will be considered eligible to enroll in the study if they are:

1. overweight (BMI >27 and ≤ 45 kg/m2),
2. have low Prime Diet Quality Score (PDQS <15) and
3. have a combination of risk factors which places them at high total risk (estimated multifactorial CVD risk ≥ 20% over ten years) of developing atherosclerotic CVD for the first time.

Exclusion Criteria:

Participants will be ineligible to enroll in the study if they have:

1. established diabetes mellitus, CVD or a medical condition or dietary restriction(s) that would substantially limit their ability to complete the study requirements,
2. have an excessive alcohol consumption (>28 Units/week for men or >21 Units/week for women),
3. have a low predicted likelihood to change dietary habits
4. are unable to provide informed consent.

Trial design

Primary purpose: Prevention

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: Single Blind

---

150 participants in 2 patient groups

Educational material

Active Comparator group

Description:

Participants will be given written educational material to encourage dietary change towards a better quality diet. Educational material will be adapted from resources used in previous dietary intervention studies.

Treatment:

Behavioral: Educational material

Personalised dietary advice and educational material

Experimental group

Description:

Participants will also be given the educational material, but will be encouraged to make dietary change using individualized recommendations incorporating food markers of specific components of a high quality diet. Food markers of diet quality will be used to develop an algorithm to deliver personalized dietary advice. Based on biomarker data, a system of categorization of biomarker status will be developed, alongside dietary advice related to this biomarker categorization, and decision trees created, as previously described in the Food4Me study, to ensure standardized delivery of advice within this intervention arm.

Treatment:

Behavioral: Personalised dietary advice and educational material