PURE Rubbish

Let me preface this post by saying I'm a bit time constrained, and computer issues have made my life miserable these past few days.   However a new article from the PURE study was recently published in Lancet, and I wish to add some limited commentary to the mix.  

First things first, the links:

• Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study

• The first study:  The Prospective Urban Rural Epidemiology (PURE) study: Examining the impact of societal influences on chronic noncommunicable diseases in low-, middle-, and high-income countries

Secondly, I've tweeted these links out.  Commentary by others:

• Kevin Klatt:  On PURE
• Nutrition Wonk:   Research Review The 5 Continent (PURE) Study: A Prospective Cohort Study

And now for my favorite headline!

I'm going to limit my commentary to three topics:

1. Choice and variety of countries
2. Bias of Investigators
3. Exclusion of participants with existing cardiovascular disease in a study including those aged 35 to 70 at baseline.

I bolded the last one because it is ultimately the most important.  

The Eighteen Countries Study

From the 2009 announcement of the PURE study, I bring you an annotated map of the countries included.  

"The choice and number of countries selected in PURE reflects a balance between involving a large number of communities in countries at different economic levels, with substantial heterogeneity in social and economic circumstances and policies, and the feasibility of centers to successfully achieve long-term followup. Thus, PURE includes sites in which investigators are committed to collecting good-quality data for a low-budget study over the planned 10-year follow-up period and did not aim for a strict proportionate sampling of the entire world. The countries involved in the PURE study include 4 LIC: Zimbabwe, Bangladesh, India, and Pakistan; 10 MIC**: South Africa, Brazil, Argentina, Colombia, Chile, Poland, China, Malaysia, Iran, and Turkey; and 3 HIC: Canada, Sweden, and UAE"

** For whatever reason, the occupied Palestinian territory (the 11th MIC) is not listed in the section on the selection of counties.  This country in the current study is not mentioned in the original paper.

The distribution of these countries alone is enough to dismiss this study. Surely they could have engaged such committed low budget investigators in more stable countries in Africa than Zimbabwe?  In the Middle East the occupied Palestinian territory seems perfect, eh?  Did Dr. Shai turn them down in Israel, or is her team too expensive?  

We get little in the way of itemized un-"massaged" data on a per country basis.  What we get is rather itemized as:  China, S. Asia, Europe/North America, South America, Middle East, SE Asia and Africa.   Come on!!!

• All of Africa is represented by the southern-most tip, and one notably unstable country with a ravaging HIV/AIDS epidemic and famine.
• The Middle East suddenly includes the occupied Palestinian territory
• North America is lumped with Europe and represented by Canada, Sweden and Poland
• China and India pretty much have that region covered, but let's throw in Bangladesh
• South America is nicely covered, but what about Australia?  
• Not a single former USSR country?  Why not?  

Bottom Line:  There is no attempt to right any "wrongs" of past epidemiological studies here in terms of the selection of countries.

Researcher Bias

The "tell" here is in the aftermath.  It's ALL about these supposedly wrong and deadly dietary guidelines that were initiated by the United States.   A country not included in the study, because I guess it's too expensive to find motivated idealoguenvestigators willing to work on the cheap.  Forgive me if I'm not going to change a thing based on these lackluster results.  

We've got Salim Yusuf of fame for this speech 

I think you'll find this slide interesting in terms of ramifications of PURE

Another co-author will be at Low Carb Breckenridge 2018.  Make no mistake about it, these conferences are revenue generators pure and simple.  Not serious medical conferences.

The Missing Seven(+) Thousand

"For the current analysis, we included all outcome events known to us until March 31, 2017. 148723 participants completed the FFQ, of which 143934 participants had plausible energy intake (500–5000 kcal per day)  and had no missing values on age and sex [4789 had non-plausible energy intake or missing values]. We excluded 1230 participants (0·8% of the cohort) because follow-up information was not available and 7369 participants with a history of cardiovascular disease (5·0% of the cohort). The remaining 135335 individuals were included in this analysis (appendix p 19)."

So.  Here we have a study on mortality, and CVD mortality specifically.  Can we agree that mortality generally correlates with age, and that nutrition-related mortality due to chronic disease does so especially?  The participants ranged in age from 35 to 70 at baseline.  I can understand if this was a study following 35-40 year olds into their 50s or beyond.  But this was not such a study.  

Why in an OBSERVATIONAL study of dietary factors on mortality due to heart disease (or even mortality in general) would you exclude people with existing heart disease?  

Keep in mind:  they didn't exclude anyone with any other existing disease.  How about HIV status in Zimbabwe?  Was that assessed/documented?  We're talking almost 15% adult infection rate ....

In the remaining ~135,000 participants,  they documented:

• 5796 deaths 
• 4784  major cardiovascular disease events

How might THOSE numbers have changed if they had NOT excluded 7369 people with existing heart disease?   Perhaps *THIS* is why no associations with CVD mortality were observed for any macronutrient?  

Yes.  This study is PURE rubbish.