Saturday, October 31, 2015

Addressing Some Questions from my Presentation: Part 2

The Importance of Total Mortality

To save time I only presented results for major coronary events, but not coronary mortality and total mortality.  But both coronary mortality and total mortality are both important to determine whether an intervention is effective.  Consider the following scenarios


RR for
CHD Events
RR for
CHD Mortality
RR for
Total Mortality
Scenario 1
0.90
0.90
1.00
Scenario 2
0.90
1.00
1.00

Which treatment would you prefer?  I would prefer scenario 2

In scenario 1 there are less CHD events and mortality in experimental group.  But because total mortality is the same, there is more non-CHD mortality in the experimental group, and therefore non-CHD events is also likely to be higher

In scenario 2 there are less non-fatal CHD events in experimental group, but no difference in CHD mortality, non-CHD mortality and total mortality.  Therefore it’s unlikely for there to be a difference in non-fatal non-CHD events

To summarise:


CHD
Events
CHD
Mortality
Non-CHD
Events
Non-CHD
Mortality
Total
Mortality
Scenario 1
Down
Down
Up???
Up
Equal
Scenario 2
Down
Equal
Equal???
Equal
Equal

Therefore:


Morbidity
Mortality
Scenario 1
Likely Equal
Equal
Scenario 2
Likely Down
Equal

When pooling the results from all the saturated fat vs. polyunsaturated fat trials together the results are like scenario 1.  People who promote the replacement of SFA with PUFA like to cite the Mozaffarian meta-analysis.  This is probably because of its large and significant risk reductions for coronary events and coronary mortality.  However, that meta-analysis found no benefit for total mortality.  And so with this information such an intervention wouldn’t be worth our time because total mortality is the same and the benefit of quality of life from reduction of CHD events is likely to be offset by the loss of quality of life from the increase in non-CHD events

Saturated Fat and Inflammation

I was asked about whether saturated fat increases inflammation.  This is certainly an interesting and relevant question.  My understanding is that most of evidence here comes from: (1) cell culture studies where very high levels of saturated free fatty acids are added to induce inflammation; (2) and animal studies (see below); and (3) studies measuring postprandial inflammation, which probably has more to do with the presence or absence of beneficial substances (blog)

The animal studies used as evidence that saturated fat is inflammation usually involves putting mice who are genetically susceptible to obesity from a high fat diet (C57BL/6 mice) on a high fat diet.  It doesn’t seem to matter whether the fat comes from SFA, MUFA or n-6 PUFA, all are inflammatory for these mice (MCTs and n-3 PUFA tend not to be).  On such high fat diets these mice dramatically overeat and develop impaired glucose tolerance, insulin resistance and hypothalamic inflammation in a few days to a week, before obesity has time to develop.  Chronic systemic inflammation usually develops later

Despite these high fat rodent diets being high in SFA and MUFA, due to the regular use of lard, it’s often SFA that get the blame.  However, these animals likely become inflamed and develop those health problems simply because a fat intake of at least 40% of total calories is inappropriate for their physiology.  The proportion of energy from fat in rodent high fat diets is often 40% and sometimes 60%.  Notably 40% fat is about the proportion of energy from fat in the Mediterranean diet.  Clearly the Mediterranean diet does not have these effects 

The purpose of animal studies is not to provide specific dietary advice for humans, but rather to test mechanisms in a controlled way

Friday, October 30, 2015

Addressing Some Questions from my Presentation: Part 1

Monounsaturated Fat and Coronary Heart Disease

My presentation (and the meta-analysis) is about saturated fat (SFA) and polyunsaturated fat (PUFA), but what about monounsaturated fat (MUFA)?  Despite the promotion of MUFA as being heart healthy (because it reduces LDL-C and the total-C:HDL-C ratio) it actually received very little attention until much later on with the promotion of the Mediterranean diet.  As a result, my understanding is that there is only one clinical trial that has manipulated MUFA intake without also altering several other dietary variables, which is what occurs in the Mediterranean diet trials and as a result the Mediterranean diet trials are inappropriate to test the effect of MUFA on the incidence of coronary heart disease (CHD)

The only clinical trial is the ‘Rose Corn Oil Trial’ (RCOT; hat tip to Zahc for reminding me).  RCOT had a control/saturated fat group, an olive oil group and a corn oil group.  The control group performed slightly better than both oil groups, but the differences between the groups was hardly significant as the number of participants and events in the groups was very small.  So I wouldn’t want make much of a conclusion on MUFA and CHD to for a few reasons: (1) there seems to be only one relevant trial; (2) the trial was very small; and (3) the fact that both oil groups did slightly worse may indicate an unfavourable health effect of the ‘oil advice’ per se, rather than being related to MUFA and PUFA


Saturated Fat
Group
(N = 26)
Olive Oil
Group
(N = 26)
Corn Oil
Group
(N = 28)
Major Coronary Events
6
9
12
Total Coronary Events
11
11
15
Coronary and Total Mortality
1
3
5
Participants who Completed the Trial
24
21
25

Regarding observational studies, 3 out of 4 meta-analyses of observational studies have found no association between MUFA and CHD, including Jakobsen [1], Skeaff [2] and Chowdhury [3].  The Mente meta-analysis was the only one to find a significant association between MUFA and CHD, with MUFA being inversely associated with CHD [4]

The reason why MUFA were not well researched until recently is likely because the original diet heart hypothesis related to total cholesterol, rather than LDL-C or the total-C to HDL-C ratio, likely due to limitations in measurement.  MUFA doesn’t affect total-C as it increases HDL-C to the same extent as it reduces LDL-C [5], so it was considered neutral and largely ignored

Is Nutrition Science Bad?

In order to accurately convey the evidence regarding saturated fat and coronary heart disease I had to point out some of the problems with the saturated fat vs. polyunsaturated fat trials as well as in relying on mechanisms/risk factors and observational studies.  I can understand how someone could come away from my presentation with a negative attitude towards nutrition science, but I don’t think it’s as bad that

One of the problems is that the early to mid-20th century was like the dark ages of epidemiology.  Diet and disease relationships were only beginning to be explored in observational studies and tested in clinical trials and some of the limitations of certain study designs didn’t appear to be well understood at the time

Regarding observational studies, prospective cohort studies are the most reliable*, however such cohorts had not yet been set up.  Instead a lot of observational epidemiology at the time was based on ecological studies – comparing the dietary patterns and incidence of coronary heart disease between different countries.  Ecological studies are some of the most susceptible to confounding variables and it seems the early ecological studies didn’t adjust to adjust for confounding variables.  This is understandable considering the lack of evidence at the time to identify confounders and perhaps the lack of software to execute such adjustments

There were some issues with the early clinical trials.  Some of them weren’t randomised like the Anti-Coronary Club and the Finnish Mental Hospital Study.  It was not uncommon to put participants on a diet, and then after a year or so allocate the good adherers to the ‘experimental group’ and the poor adherers to the ‘control group’.  Not to mention that the health effects of trans fats were not widely understood at the time and the investigators preferred to test a ‘cholesterol lowering diet’ rather than rigorously testing the effects of replacing SFA with PUFA

The design and execution of observational studies and clinical trials appears to be much better these days.  I think we need to recognise the issues in certain earlier studies and not disregard modern and better designed epidemiological evidence based on the faults of their predecessors

With the diet heart story the main problem is that so few people have actually read the papers and have relied on reviews and meta-analyses, which in turn seemed to assume everything is ok with the trials, and many simply take some participant characteristics and the results then leave it at that

* Prospective cohorts are the most reliable because they avoid the issue of reverse ‘causality’.  For example in a cross-sectional design it could be possible to have a positive association between calcium intake and fracture risk.  This is because the people with the highest intake of calcium may be those who are taking calcium supplements because they are at high risk of osteoporosis or have been diagnosed with osteoporosis or osteopenia 

** Although adherence may be a bit worse and using patients in mental hospitals to achieve very high adherence and a double-blind design is probably unethical in most countries these days

Thursday, October 29, 2015

AHSNZ's International Symposium: 'Looking Back, Moving Forward'

The Ancestral Health Society of New Zealand (AHSNZ) held their first international symposium “Looking Back, Moving Forward” last weekend in Queenstown

There was a diverse group of speakers coming from a variety of different backgrounds, many of whom are big names in academia and the ancestral health community.  This diversity was really quite a strength of the symposium.  It allowed for the sharing of many ideas across a wide range of topics such as behaviour change, exercise, sustainability, active transport, nutrition and mental health.  Secondly getting people with a broad range of backgrounds and inviting people outside the ancestral health community helps to introduce new ideas and prevent groupthink

On that note, the presentations I found the most interesting tended to be those I wasn’t that familiar with, such as Brad Norris on personal values and behaviour, Professor Simon Kingham and Julie Anne Genter (MP) on active transport and Professor David Raubenheimer on nutritional ecology and protein leverage

Even with the diversity built in to the symposium, there were several common themes throughout the presentations, for example:

  • Health promotion is not as simple as telling people to be healthy.  Personal responsibility is still important, but you actually need to engage people with where they are at and change the environment to one that supports health promoting behaviours
  • There are several cost effective solutions to transport and sustainability issues.  The problem is inertia and not being prepared to give things a go
  • The high intake of low quality, highly processed foods (which tends to have a lot of acellular carbohydrates and added fats, but little protein, micronutrients and other beneficial substances) is the main dietary issue (for physical and mental health) that needs to be addressed and is unfortunately the type of product that is being aggressively marketed 

There were plenty of breaks for discussion and long lunch breaks to explore Queenstown and engage in movement sessions (Primal Play with Darryl Edwards and floor based movement with Phillip Beach).  I didn’t get much chance to participate but the Primal Play that Darryl introduced us to in his presentation was quite fun

I gave a presentation on saturated fat and coronary heart disease, and using that as a way to discuss some issues in nutrition science, such as the limitations of mechanisms/risk factors, observational studies and nutrient based approach to diet.  It may have surprised some people that the presentation wasn’t related to my PhD project (some of which I’ve discussed here), but my PhD project is quite technical and more theoretical than practical (not to mention I just started and have very few results).  So it would be inappropriate for an audience with a large number of clinicians (which again was a good design for the dissemination of knowledge and enabling an efficient discussion around some semi-technical concepts and research) and interested lay people.  On the other hand, Melyssa Roy gave a great talk on her PhD project which is very interesting and related to how different diet and exercise advice is put into practice 

Most if not all the talks were filmed, so I imagine they will be uploaded at some point.  You can also see some of the content of the presentations via the AHSNZ and AHSNZ15 hashtags.  I’ve seen/heard nothing but good feedback about the event, so if you can make the trip I would recommend going to the next one

Sunday, October 11, 2015

Some Further Thoughts on Observational Studies

Confounding Variables in Observational Studies

A while ago I had a look at the observational studies regarding saturated fat (SFA) and coronary heart disease (CHD).  I haven’t blogged about it because my experience was pretty disappointing as very few of the papers referenced by meta-analyses included much detail besides risk ratios (RR) for SFA and CHD, and some didn’t even do that.  The paper Jakobsen et al cited for the Adventist Health Study didn’t report the RRs for nutrient intake (SFA, etc), but reported the RR for nuts and other foods.  Not to mention that this is a pretty inappropriate cohort, as Seventh Day Adventists are supposed to vegetarian and so vegetarian food patterns would be associated with even greater adherence to lifestyle factors that promote health.  Notably this cohort was an outlier for SFA>PUFA and SFA>CHO RRs (see figure 1).  This is similar to Jim Mann’s health conscious cohort that included a large number of vegetarians and semi-vegetarians (see figure 2 from Siri-Tarino et al)

Anyway, only a few observational studies reported baseline characteristics which can be sources of confounding variables.  The following tables come from a new paper published on the Nurse’s Health Study (NHS) and the Health Professionals Follow-up Study (HPFS)*.  The first shows a consistent trend whereby people who eat more SFA tend to engage in less health conscious behaviours.  The trends for each factor alone aren’t very impressive, but the results suggest they add up quite a bit

  
The trend between age-adjusted SFA intake and CHD is highly significant in both studies and the RR in the highest quintile is 1.54 for NHS and 1.32 for HPFS.  However, in the multivariate analysis, which adjusts for many of the baseline characteristics (and therefore attempts to control for confounding variables), there is no significant association between SFA and CHD and the RR in the highest quintile is 1.01 for NHS and 0.87 for HPFS**


This adjustment doesn’t work all the time as evidenced by examples such as hormone replacement therapy, but it does bring us much closer to the real relationship between SFA and CHD (I don’t know what it is but it’s probably close to neutral)

* NHS and HPFS have typically being the studies driving the inverse association between PUFA and CHD in a couple of the meta-analyses of observational studies.  This paper doesn’t change the overall evidence base much as it’s hardly a new cohort of people

** The results of this paper suggest the RR for 5% SFA is 0.94 and RR for 5% MUFA is 1.00, so I don’t know how they came up with MUFA being better than SFA in a substitution analysis

What Do We Mean By Saturated Fat?

One of the problems with the focus on nutrients is that they don’t necessarily tell you much about food quality.  We can get some information on this from the 2010 USDA dietary guidelines, which included some data from NHANES on the sources of saturated fat in the average US diet (below)


When these sources of saturated fat are roughly grouped into ‘higher’ and ‘lower’ food quality (based roughly on added sugars, refined starches and added fats) it seems that probably about half the saturated fat in the US diet comes from lower quality foods (higher quality = 37.4%, lower quality = 37.9%, unknown – 24.5%) (see below).  This is really not that surprising since the average US diet contains a lot of calories from ‘lower quality foods’, but it brings up two points

1.      Any association between saturated fat and disease could easily be due lower food quality rather than saturated fat
2.      Since consumption of refined sugars, starches and fats, etc is high, dietary advice should focus on reducing those rather than the current focus on things with a really debatable effect on health (such as SFA and salt).  It’s insane that dietary guidelines allow up to 3 serves of refined grains, but you better not eat fatty meat and full fat dairy, because SFA

This applies to other nutrients as well.  This discussion has occurred with carbohydrates.  It’s time for it to happen more widely with protein, fats, etc as well.


Higher Quality Foods (%)
Lower Quality Foods (%)
Regular cheese
8.5

Pizza

5.9
Grain-based desserts

5.8
Dairy desserts

5.6
Chicken and chicken mixed dishes
5.5

Sausage, franks, bacon, ribs
4.9

Burgers

4.4
Tortillas, burritos, tacos

4.1
Beef and beef mixed dishes
4.1

Reduced-fat milk
3.7

Pasta and pasta dishes

3.7
Whole milk
3.4

Eggs and egg mixed dishes
3.2

Candy

3.1
Butter

2.9
Potato/corn/other chips

2.4
Nuts and seeds, and nut and seed mixed dishes
2.1

Fried white potatoes
2.0

All other food categories (24.5%)
?
?
TOTAL
37.4
37.9