Welcome ! Reducing saturated fat intake for cardiovascular disease: What's the evidence? You will be placed on hold until the webinar begins. The webinar will begin shortly, please remain on the line.
Welcome!Reducing saturated fat
intake for cardiovascular disease: What's the
evidence?
You will be placed on hold until the webinar begins. The webinar will begin shortly, please remain on the
line.
Poll Questions: Consent• Participation in the webinar poll questions is voluntary• Names are not recorded and persons will not be identified in any way• Participation in the anonymous polling questions is accepted as an
indication of your consent to participate
Benefits:• Results inform improvement of the current and future webinars• Enable engagement; stimulate discussion. This session is intended
for professional development. Some data may be used for program evaluation and research purposes (e.g., exploring opinion change)
• Results may also be used to inform the production of systematic reviews and overviews
Risks: None beyond day-to-day living
After Today• The PowerPoint presentation and audio
recording will be made available
• These resources are available at: – PowerPoint: http://
www.slideshare.net/HealthEvidence
– Audio Recording: https://www.youtube.com/user/healthevidence/videos
3
What’s the evidence? Hooper L., Martin N., Abdelhamid A., & Smith G.D. (2015). Reduction in saturated fat intake for cardiovascular disease. Cochrane Database of Systematic Reviews, Art. No.: CD011737.http://www.healthevidence.org/view-article.aspx?a=28821
Evidence Summary: http://www.healthevidence.org/documents/byid/28821/Hooper2015_EvidenceSummary_EN.pdf
Poll Question #1What sector are you from?A. Public Health PractitionerB. Health Practitioner (Other)C. EducationD. ResearchE. Provincial/Territorial/Government/Ministry/
MunicipalityF. Policy Analyst (NGO, etc.)G. Other
5
• Use Q&A or CHAT to post comments / questions during the webinar– ‘Send’ questions to
All Panelists (not privately to ‘Host’)
• Connection issues– Recommend using a wired
Internet connection (vs. wireless),
• WebEx 24/7 help line– 1-866-229-3239
Participant Side Panel in WebExHousekeeping
Housekeeping (cont’d)
• Audio – Listen through your speakers– Go to ‘Communicate > Audio Connection’• WebEx 24/7 help line– 1-866-229-3239
Poll Question #2
How many people are watching today’s session with you?
A. Just meB. 2-3C. 4-5D. 6-10E. >10
The Health Evidence™ Team
Maureen Dobbins Scientific Director
Heather HussonManager
Susannah WatsonProject Coordinator
Students:Emily Belita(PhD candidate)
Jennifer YostAssistant Professor
Olivia MarquezResearch Coordinator
Emily SullyResearch Assistant
Liz KamlerResearch Assistant
Zhi (Vivian) ChenResearch Assistant
Research Assistants:Lina SherazyClaire HowarthRawan Farran
What is www.healthevidence.org?
Evidence
Decision Making
inform
Why use www.healthevidence.org?
1. Saves you time2. Relevant & current evidence 3. Transparent process4. Supports for EIDM available 5. Easy to use
A Model for Evidence-Informed Decision
Making
National Collaborating Centre for Methods and Tools. (revised 2012). A Model for Evidence-Informed Decision-Making in Public Health (Fact Sheet). [http://www.nccmt.ca/pubs/FactSheet_EIDM_EN_WEB.pdf]
Stages in the process of Evidence-Informed Public Health
National Collaborating Centre for Methods and Tools. Evidence-Informed Public Health. [http://www.nccmt.ca/eiph/index-eng.html]
Poll Question #3
Have you heard of PICO(S) before?
A.YesB.No
Searchable Questions Think “PICOS”
1. Population (situation)
2. Intervention (exposure)
3. Comparison (other group)
4. Outcomes
5. Setting
How often do you use Systematic Reviews to inform a program/services?
A.AlwaysB.OftenC.SometimesD.NeverE.I don’t know what a systematic review is
Poll Question #4
Dr. Lee Hooper
Reader in Research Synthesis, Nutrition & Hydration in the Norwich Medical School at the University of East Anglia
Reducing saturated fat intake lowers the risk of cardiovascular eventsA.Strongly agreeB.AgreeC.NeutralD.DisagreeE.Strongly disagree
18
Poll Question #5
ReviewHooper L., Martin N., Abdelhamid A., & Smith G.D. (2015). Reduction in saturated fat intake for cardiovascular disease. Cochrane Database of Systematic Reviews, Art. No.: CD011737.
If you would like a full text copy of the review please visit the Cochrane Library or request a copy from Lee ([email protected]).
Springs from: Hooper L et al. (2012) Reduced or modified dietary fat for preventing cardiovascular disease. Cochrane Database of Systematic Reviews, Art No.: CD002137
Review authors• Lee Hooper, Norwich Medical School,
University of East Anglia, England• Nicole Martin, Managing Editor, Cochrane
Heart Group, London, England• Asmaa Abdelhamid, Royal College of
Paediatrics & Child Health, London• George Davey Smith, University of Bristol,
England
Rationale (a)Public health dietary advice on prevention of cardiovascular disease (CVD) has changed over time, with a focus on •fat modification during the 1960s and •fat reduction during the 1990s •In 2006 the American Heart Association (AHA) suggested “limit intake of saturated fat to 7% of energy” (Lichtenstein 2006). •In 2013 the AHA suggested “Aim for a dietary pattern that achieves 5% to 6% of calories from saturated fat” (strong evidence, Eckel 2013). US and European guidance are both based on dietary effects on lipids.
Rationale (b)If we reduce saturated fat in our diets we will replace the energy with other fats, carbohydrate, protein and/or alcohol. Which nutrients are used in place of saturated fat will affect our health. •Joint British Societies’ (JBS) guidance on preventing CVD recommends “Replace saturated fat with polyunsaturated fat” (JBS3 2014),•UK National Institute for Health and Care Excellence (NICE) guidance suggests that people at high risk of or with CVD eat so that “saturated fats are 7% or less of total energy intake… [and] replaced by mono- and poly-unsaturated fats” (NICE 2014).
Rationale (c)The World Health Organization (NUGAG subgroup) wanted to understand the following to enable them to set guidance for saturated fat intake: •the evidence of the effects on mortality and cardiovascular health of reducing saturated fat, and •how any effects differ depending on what type of energy is used to replace the saturated fat. •What cut-off of saturated fat to recommendWe chose to include only randomised controlled trials as dietary patterns are highly confounded by other lifestyle factors such as smoking, physical activity and socioeconomic status which themselves have a huge impact on our outcomes. This means that cohort studies provide less trustworthy answers than long term trials.
Review Focus:• Participants – adults, with or without CVD at
baseline
• Intervention – reduction in saturated fat by dietary advice, supplementation (of fats, oils or modified fat foods) or provision of a whole diet, over at least 2 years (24 months)
• Comparison – usual diet, placebo or control diet
• Outcomes – all-cause mortality, CVD mortality, CVD events (plus secondary outcomes)
Review Focus:Secondary Outcomes –•Myocardial infarction (MI)•Stroke including stroke incidence (type of stroke), stroke mortality, and stroke morbidity•CHD mortality (includes death from MI or sudden death•CHD events (includes any of: fatal or non-fatal myocardial infarction, angina or sudden death)•type II diabetes incidence•Blood measures (including serum blood lipids and measures of glucose tolerance)•Other outcomes & adverse effects reported by study authors (including cancer diagnoses & deaths, body weight, BMI, blood pressure (BP), quality of life
Methods - Searching• Searched to March 2014, on Cochrane
CENTRAL, Medline, EMBASE• Bibliographies & experts• assessed 23,471 titles & abstracts• 662 full text papers assessed• 15 RCTs
– planned an intervention of ≥24 months, AND – either stated an aim to reduce saturated fat OR
achieved statistically significant SFA reduction• These 15 RCTs were included in this review
Review flow diagram
Methods – review process
• Independently duplicated assessment of titles and abstracts, and of full text papers retrieved
• Duplicated data extraction and assessment of risk of bias
• We contacted authors to request missing outcome and risk of bias data
• Tabulated reasons for exclusion, characteristics of included studies, risk of bias of included studies
Risk of bias of included studiesWe assessed study risk of bias using the Cochrane Risk of Bias tool (see Cochrane Handbook, http://training.cochrane.org/handbook) and added other factors important to this review:•Free of systematic differences in care•Stated aim to reduce SFA•Achieved SFA reduction•Achieved serum cholesterol reduction
Methods – analysis (a)• Mantel-Haenszel random-effects meta-
analysis (RevMan 5) to assess risk ratios• I2 was used to assess heterogeneity
(considered important when I2 >50%) • Outcome data extracted for the latest time
point (always ≥24 months).• Effects of SFA reduction compared with usual
or standard diet on all (primary and secondary) outcomes and adverse effects.
• Funnel plots used to assess small study bias
Methods – analysis (b)Prespecified subgroups included:•energy substitution for SFA (MUFA, PUFA, carbohydrate, protein)•Baseline SFA intake •Sex (men, women and mixed populations)•Baseline CVD risk •Study durationWHO requested:•Degree of SFA reduction•Serum total cholesterol reduction achieved•Ethnic group
Methods – analysis (c)Sensitivity analyses excluded studies that:•Did not state an aim to reduce SFA•Did not report SFA intake during the trial, or find a significant reduction in SFA in the intervention compared to the control•Did not reduce total cholesterol (TC) •Were the largest study (WHI 2006)Analyses run with Mantel-Haenszel fixed-effect model and Peto fixed-effect model•GRADE assessment
15 Included RCTs:Baseline health status, people…•post-MI or with angina 6•with DM or glucose intolerance 4•with cancer risk or diagnosis 3•With no specific risks 2Geography •USA or Canada 6•Europe 7•Australia or NZ 2
What is the effect of saturated fat (SFA)
reduction on all-cause mortality?
SFA reduction on all-cause mortality
RR 0.97 (95% CI 0.90 to 1.05) I2 3%3276 deaths, >55000 people
SFA reduction on all-cause mortality - funnel
Replacement criteria• Replacement of SFA by PUFA, MUFA, CHO,
protein and trans were discerned from aims (if possible) or from dietary intake within the study (if necessary)
• categorised as any or all of PUFA, MUFA, CHO, protein
• AND • there was a statistically significant difference
(during the experimental diet) between intervention and control for PUFA, MUFA, CHO, or protein
SFA reduction on all-cause mortality – replacements
RR 0.96 (95% CI 0.82 to 1.13) I2 26%824 deaths, >4000 people
RR 3.00 (95% CI 0.33 to 26.99) 4 deaths, 52 people
RR 0.98 (95% CI 0.91 to 1.05) I2 0%2677 deaths, >53000 people
RR 0.98 (95% CI 0.91 to 1.06) I2 0%, 2673 deaths, >53000 people
protein
CHO
MUFA
PUFA
No effect of ↓ SFA on all-cause mortality• No sensitivity analysis (using 2 fixed
effects analyses, excluding largest RCT, excluding studies with non-fat dietary interventions, excluding studies with different intensity of interventions) altered the risk ratio (0.96 to 0.99) or altered the lack of statistical significance
• No subgrouping altered the verdict of no effect for all-cause mortality
What is the effect of saturated fat (SFA)
reduction on cardiovascular
mortality?
Effect of reduced SFA on CVD mortality
RR 0.95 (95% CI 0.80 to 1.12) I2 30%, 1096 CVD deaths, >53,000 participants
Effect of reduced SFA on CVD mortality
Reduced SFA on CVD mortality - replacements
No effect of ↓SFA on CVD mortality
• No sensitivity analysis altered the effect size (RR 0.92 to 1.00) or lack of statistical significance
• No subgrouping altered the verdict of no effect for CVD mortality except suggestion of effect with greater reduction in SFA – 1 study which reduced SFA by >8%E found a
30% reduction in CVD mortality, RR 0.70 (95% CI 0.51 to 0.96, Veterans Admin study 1969
What is the effect of saturated fat (SFA)
reduction on cardiovascular
events?Cardiovascular events included any of the following: cardiovascular deaths, cardiovascular morbidity (non-fatal myocardial infarction, angina, stroke, heart failure, peripheral vascular events, atrial fibrillation) and unplanned cardiovascular interventions (coronary artery bypass surgery or angioplasty)
Effect of reduced Saturated Fat on CV events
RR 0.83 (95% CI 0.72 to 0.96) I2 65%, 4377 events, >53000 participants
Reduced SFA on CV events – funnel plot
Effect of reduced Saturated Fat on CV events
RR 0.73 (95% CI 0.58 to 0.92) I2 69%, 884 events, >3000 participants
RR 1.00 (95% CI 0.53 to 1.89) 22 events, 52 participants
RR 0.93 (95% CI 0.79 to 1.08) I2 57%, 3785 events, >51000 participants
RR 0.98 (95% CI 0.90 to 1.06) I2 15%, 3757 events, >51000 participants
Reduced Saturated Fat on CV events - SAAnalysis RR (95% CI) of
CVD eventsI2 No. of
eventsNo. of
participants
Main 0.83 (0.72 to 0.96) 65% 4377 >53000
Sensitivity analyses
Stated aim to reduce SFA 0.84 (0.72 to 0.97) 69% 4354 >52000SFA significantly reduced 0.91 (0.79 to 1.04) 53% 4012 >52000
TC significantly reduced 0.81 (0.68 to 0.98) 77% 4092 >52000
Minus WHI 0.75 (0.61 to 0.91) 51% 932 >4000
Mantel-Haenszel Fixed effects
0.93 (0.88 to 0.98) 65% 4377 >53000
Peto Fixed effects 0.92 (0.86 to 0.98) 72% 4377 >53000
Reduced SFA on CV events - subgroupingAnalysis, RR (95% CI) of CVD
eventsI2 No. of
eventsNo. of
participants
Subgroup by replacementp=0.14
PUFA replacement 0.73 (0.58 to 0.92) 69% 884 >3000MUFA replacement 1.00 (0.53 to 1.89) NA 22 52CHO replacement 0.93 (0.79 to 1.08) 57% 3785 >51000Protein replacement 0.98 (0.90 to 1.06) 15% 3757 >51000
Subgroup by duration, p=0.15
Up to 24 months 0.96 (0.78 to 1.16) 0% 330 >2000>24 to 48 months 0.73 (0.56 to 0.95) 50% 383 >1000>48 months 0.93 (0.79 to 1.11) 75% 3599 >49000Unclear duration 0.43 (0.17 to 1.08) NA 65 >200
Subgroup by baseline SFA, p=0.13
Up to 12%E SFA NA >12 to 15%E SFA 0.98 (0.91 to 1.05) 6% 3765 >51000>15 to 18%E SFA 0.41 (0.22 to 0.78) NA 28 55>18%E SFA 0.79 (0.63 to 1.00) NA 219 846
Subgroup by SFA change, p=0.005
Up to 4%E SFA difference
0.98 (0.91 to 1.05) 6% 3763 >51000
>4 to 8%E SFA difference
0.40 (0.22 to 0.74) 0% 30 >100
>8%E SFA difference 0.79 (0.63 to 1.00) NA 219 >800
Reduced Saturated Fat on CV events - subgrouping
Analysis, RR (95% CI) of CVD events
I2 No. of events
No. of participan
tsSubgroup by sex, p=0.05
Men 0.80 (0.69 to 0.93) 24% 859 >3000Women 1.00 (0.88 to 1.14) 60% 3445 >48000Mixed, men & women 0.59 (0.23 to 1.49) 71% 73 >500
Subgroup by CVD risk, p=0.67
Low CVD risk 0.89 (0.75 to 1.06) 40% 3130 >47000Moderate CVD risk 0.59 (0.23 to 1.49) 71% 73 >500Existing CVD 0.86 (0.71 to 1.05) 63% 1174 >5000
Subgroup by serum TC reduction, p=0.03
TC ↓ by ≥0.2mmol/L 0.74 (0.59 to 0.92) 63% 887 >4000TC ↓ by <0.2mmol/L 0.99 (0.90 to 1.08) 15% 3488 >49000Unclear TC change 0.20 (0.01 to 4.15) NA 2 >100
Reduction of CV events with SFA reduction
• Sensitivity analyses –Consistent reduction in CV events with
reduced SFA for almost all sensitivity anal• Subgrouping explained some
heterogeneity - greater reduction in CV events with – (SFA replaced by PUFA)–Greater SFA reduction–Greater serum cholesterol reduction
Meta-regression – effect of individual factors on degree of reduction of CVD events
• greater reduction in serum total cholesterol was associated with greater improvement in CVD events with SFA reduction (p=0.04, accounting for 99% of between study variation)
• greater reductions in SFA intake and greater baseline SFA intake were loosely associated with reduced CVD events
• Gender, study duration and baseline cardiovascular risk did not appear to influence effect size
Effects of SFA reduction on serum chol.
Pooled effect on serum total cholesterol was a fall of 0.24mmol/L (95% CI -0.36 to -0.13), I2 60%, >7000 participants
0.24 mmol/L total cholesterol = 9.3 mg/dl
Secondary outcomesThere were no statistically significant effects of reducing saturated fats on •MI: RR 0.90 (95% CI 0.80 to 1.01, p=0.09) I2 10%, 1714 MI•Stroke: RR 1.00 (95% CI 0.89 to 1.12) I2 0%, 1125 events•Cancer deaths: RR 1.00 (95% CI 0.61 to 1.64) I2 49%, 2472 events•Cancer diagnoses: RR 0.94 (95% CI 0.83 to 1.07) I2 33%, 5476 events•Diabetes diagnoses: RR 0.96 (95% CI 0.90 to 1.02) I2 NA, 3342 events•CHD mortality: RR 0.98 (95% CI 0.84 to 1.15), I2 21%, 886 deaths •CHD events: RR 0.87 (95% CI 0.74 to 1.03, p=0.12), I2 66%, 3307 events
Cut-offsTesting cut-offs for saturated fat intake
• While the review suggests that reducing saturated fat reduces cardiovascular events there are no clear data suggesting what cut-offs may be appropriate
• This is one way of exploring what cut-offs may be appropriate
• I used the forest plot of the effects of saturated fat reduction on CV events:
Effect of reduced Saturated Fat on CV events
RR 0.83 (95% CI 0.72 to 0.96) I2 65%, 4377 events, >53000 participants
Testing cut-offs for saturated fat intake
• I tested cut-offs from 7% of energy from SFA to 15% of energy from SFA
• For each cut-off I chose the studies that had an intervention group intake less than the cut-off, and the control group greater than the cut-off
Testing cut-offs for saturated fat intake
• Example 1: the only study with an intervention group achieving <7% E from SFA and control >7%E from SFA was Black 1994, so this was the only study in the 7% analysis.
• Example 2: Ley 2004 obtained 10%E from SFA in the intervention group, and 13.4%E from SFA in the control. This study appears in the cut-offs for 11%, 12% and 13%.
Testing cut-offs for saturated fat intake
Graph of RR of a CVD event vs. cut off points (as % energy from saturated fat) tested.
Testing cut-offs for saturated fat intake
Graph of RR of a various outcomes vs. cut off points (as % energy from saturated fat) tested.
WHO Specific questions (a)• In adults what is the effect in the population
of reduced percentage of energy (%E) intake from saturated fatty acids (SFA) relative to higher intake for reduction in risk of non-communicable diseases (NCDs)?
• We see clear reductions in cardiovascular events
• Marginally statistically significant reductions in myocardial infarction
• No clear effects (over these time scales) on all-cause mortality or cardiovascular mortality, stroke, CHD mortality or CHD events
WHO Specific questions (b)
• What is the effect on coronary heart disease mortality and coronary heart disease events?
• There are no clear effects of SFA reduction on CHD mortality or CHD events BUT evidence here is limited
WHO Specific questions (c)• What is the effect in the population of
replacing SFA with PUFAs, MUFAs, CHO (refined vs. unrefined), protein or trans fatty acids (TFAs) relative to no replacement for reduction in risk of NCDs?
• SFA replacement with PUFA is – associated with reductions in CVD events – Marginal significance for reduced MI
• Replacement with CHO, protein – Associated with no clear effects on outcomes
• No trans fat data available• Very limited MUFA data
WHO Specific questions (d)
• What is the effect in the population of consuming <10%E as SFA relative to >10%E as SFA for reduction in risk of NCDs?
• Limited RCT evidence• What evidence there is supports better
health a <10%E from SFA
Reduction in saturated fat intake compared to usual saturated fat intake for adults
Outcomes No of Participants (studies) Follow up
Quality of the evidence (GRADE)
Relative effect (95% CI)
Anticipated absolute effects Time frame is at least 2 years Risk with Usual saturated fat intake
Risk difference with Reduction in saturated fat intake (95% CI)
All-cause mortality 55858 (11 studies) 56 months1
⊕ ⊕ ⊕⊕ HIGH2,3,4,5,6
RR 0.97 (0.9 to 1.05)
Study population 57 mortality per 1000
2 fewer mortality per 1000 (from 6 fewer to 3 more)
Moderate
Cardiovascular mortality
53421 (10 studies) 53 months1
⊕ ⊕ ⊕⊕ HIGH2,3,4,6,7
RR 0.95 (0.8 to 1.12)
Study population
19 CV mortality per 1000
1 fewer CV mortality per 1000 (from 4 fewer to 2 more)
Moderate
-
Cardiovascular events
53300 (11 studies) 52 months1
⊕ ⊕ ⊕⊝ MODERATE2,4,6,8,9,10 due to inconsistency
RR 0.83 (0.72 to 0.96)
Study population 83 CV events per 1000
14 fewer CV events per 1000 (from 3 fewer to 23 fewer)
Moderate
Questions?
Reducing saturated fat intake lowers the risk of cardiovascular eventsA.Strongly agreeB.AgreeC.NeutralD.DisagreeE.Strongly disagree
68
Poll Question #6
Poll Question #7Do you agree with the findings of this review?A.Strongly agreeB.AgreeC.NeutralD.DisagreeE.Strongly disagree
A Model for Evidence-Informed Decision
Making
National Collaborating Centre for Methods and Tools. (revised 2012). A Model for Evidence-Informed Decision-Making in Public Health (Fact Sheet). [http://www.nccmt.ca/pubs/FactSheet_EIDM_EN_WEB.pdf]
Poll Question #8
The information presented today was helpful
A.Strongly agreeB.AgreeC.NeutralD.DisagreeE.Strongly disagree
What can I do now? Visit the website; a repository of over 4,600 quality-rated systematic reviews related to the effectiveness of public health interventions. Health Evidence™ is FREE to use.Register to receive monthly tailored registry updates AND monthly newsletter to keep you up to date on upcoming events and public health news.Tell your colleagues about Health Evidence™: helping you use best evidence to inform public health practice, program planning, and policy decisions! Follow us @Health Evidence on Twitter and receive daily public health review-related Tweets, receive information about our monthly webinars, as well as announcements and events relevant to public health. Encourage your organization to use Health Evidence™ to search for and apply quality-rated review level evidence to inform program planning and policy decisions. Contact us to suggest topics or provide feedback.
Poll Question #9What are your next steps?
A.Access the full text systematic reviewB.Access the quality assessment for the review
on www.healthevidence.org C.Consider using the evidence D.Tell a colleague about the evidence
Thank you!Contact us:
For a copy of the presentation please visit:http://www.healthevidence.org/webinars.aspx