Issue 25 – Epidemiology

Issue 25, April 2010 (scroll down for this month’s articles): Association, causation, and the usefulness of epidemiology :: Criticism of speed of EU chemical safety law :: New studies on mechanism of harm of BPA :: Plus 5&5, our revamped round-up of the best news and science from March.

Association, causation, and the usefulness of epidemiology

John Snow, said to be the father of epidemiology.

John Snow, said to be the father of epidemiology.

“Association is not causation.” Anyone who has read critiques of epidemiological studies purporting to show a human health problem is the result of exposure to an environmental chemical will have seen this statement.

It is, of course, true: no epidemiological study can truly show causation. In 1965, the observation of this basic challenge formed the backdrop to renowned epidemiologist Sir Austin Bradford Hill’s seminal paper “The Environment and Disease: Association or Causation?”.

Bradford Hill held that determining cause is relatively simple when effects are extreme. The problem is when effects are not immediately observable. Here, the researcher is dependent on noticing associations between changes in complex variables like occupation and environment and changes in health outcomes.

Epidemiology therefore treads a fine line, where although association is a necessary feature of inquiring into the effects of the environment on the health of a population, it can at the same time make it seem a weak science in the face of strong demand for certainty about causes.

Researchers such as Tamara Galloway, Professor of Ecotoxicology at the University of Exeter (UK), believe there is a misunderstanding of the role of epidemiology plays in building knowledge of how the environment affects health, which has led to a tendency to dismiss any study which doesn’t prove causation.

In fact, it appears the real issue has less to do with what epidemiology can prove in terms of cause, as much as what can be done with the results of an epidemiological study.

Researchers identify its value as a hypothesis generator. Even weaker associations can guide research, stimulating in vitro and in vivo laboratory tests to determine mechanisms of effect of an environmental chemical, to see if a hypothesis is realistic.

After that, epidemiology might be the only human data which is available – or is even possible. Environmental contaminants don’t have human models because it is not ethical to do randomised, controlled trials on people of chemicals other than medicines.

Galloway says: “Animal studies take you so far but humans respond differently to environmental toxics, which is why we need epidemiology to understand what chemicals are doing in a human population.”

This was recently illustrated by research from a University Exeter team, of which Galloway is a member, into stain-resistant PFOA chemicals and thyroid disruption.

The consensus from peer-reviewers was the epidemiology was unlikely to show any health effects, as primate research had already shown that PFOAs have no effect on thyroid disorders.

The study, however, found an association between PFOA exposure and effects on thyroid health in middle-aged women, the population most likely to suffer these problems. In contrast, the primate studies which gave PFOAs a clean bill of health had been on adolescent males.

The issue lurking under the surface, touched on by Bradford Hill, is about how epidemiology ought to influence public policy.

Epidemiologists already acknowledge their work has limits, while regulators demand high levels of certainty. The problem is, given the ethical constraints on research, epidemiological studies may on occasion be the best possible data available.

The temptation is always to wait for certainty, but in truth epidemiology seems to be a stark presentation of a question which has always troubled regulators: when it comes to protecting health, what should be done with imperfect evidence, if imperfect evidence is the best evidence possible?

Critics say EU chemicals safety law will “take decades” to implement

The European Commission

Critics say slow implementation is undermining the objectives of the EU's flagship chemical safety regulation, REACH.

Flagship EU regulation, introduced to protect the public from harm from chemical hazards in the environment, has been criticised both by environment groups and MEPs from the European Parliament’s Environment Committee.

Open letters to the Commission about the legislation, known as REACH, identified speed of progress as the main issue, with environment groups “deplor[ing]” the lack of progress in implementation.

Slow implementation means consumers will continue to be exposed to hazardous chemicals as they remain on the market longer than they would if the implementation process moved faster.

There is concern about speed because, since REACH passed into law in 2006, only 30 chemicals have made it as far as being classified as “substances of very high concern”, referred to as SVHCs.

When a given chemical is classified as an SVHC, consumers have the right to know if it is in products, and it should mean companies will prioritise it for elimination from consumer goods in favour of safer alternatives.

The SVHC list of 30 chemicals contrasts with around 350 on the alternative SIN list of priority chemicals for phase-out, compiled by public interest groups and NGOs, and the approximately 2,200 chemicals categorised as carcinogenic, mutagenic or reprotoxic.

Even the European Commission itself has estimated that 1500 chemicals fulfil the criteria for being SVHCs. Using that number, at its current speed REACH will take a century to implement.

In a press conference, the head of the European Commission’s Directorate-General for Environment, Janez Potočnik, acknowledged the concerns, describing REACH implementation as “bad” and saying “the route to substitution” needs to be faster.

Although lengthy internal discussions aimed at producing a deal acceptable to both business and environment Directorate Generals in the Commission are one reason for the delay, environment groups believe the main problem is one of resources.

The European Chemicals Agency is the EU body responsible for REACH implementation. However, it is not thought to be resourced to put together the necessary documentation which would allow SVHCs to be prioritised for being banned in Europe.

Prioritising chemicals for phase-out is a complex two-stage process which, critics argue, needs to be streamlined in order to speed up the process of removing hazardous chemicals from the European market.

On a more positive note, REACH is precipitating collection of data on chemicals on an unprecedented scale. By November of this year, information about roughly 10,000 chemicals is expected to be available.

Although incomplete, environment groups say it will still yield the most complete, publicly-accessible set of information about chemical safety ever made available.

Meanwhile, in contrast to REACH is the more rapid action being taken by companies, guided by ChemSec’s SIN list, already over 350 chemicals long and due to be further extended in 2010.

“Companies want to be ahead of the game when it comes to securing trust from consumers,” explains Jerker Ligthart, Project Coordinator for ChemSec, an NGO which works together with companies to strengthen chemicals regulation and developed the SIN list.

“They don’t want to have to explain to customers why their product has a toxic chemical in it – they want to be able to tell customers their products are safe from that perspective.”

The SIN list recently received a boost when the Nordic Council, the intergovernmental forum representing the Nordic countries, recommended all Nordic governments use the SIN list in public procurement.

Studies shed new light on how BPA may harm female reproductive health

Two new studies published in March looked at alternative mechanisms of BPA toxicity on the female reproductive system.

In one of the studies, researchers measured levels of proteins in the mammary tissue of rats which had been exposed in the womb to low levels of BPA.

Changes in the levels of proteins associated with cell proliferation were observed, suggesting that BPA may influence the likelihood that mammary gland cells will transform into cancer cells. [Betancourt et al. J Proteomics. 2010 (in press)]

In the second study, researchers were concerned with why female rats exposed to BPA have higher rates of breast cancer, uterine cancer and infertility.

The research is particularly interesting because it looks at the epigenetic effects of BPA, finding that the chemical strips methyl groups from the DNA, permanently altering some genes to make them more sensitive to oestrogen. [Bromer JG et al. FASEB J. 2010 (in press)]

The researchers said the effect is reminiscent to that of the drug DES, which was prescribed to women until 1971 to prevent miscarriage but was linked to increased incidence of rare forms of vaginal cancer. DES is also thought to have acted via an epigenetic mechanism.

[Article continues below.]

More information about DES exposure and epigenetic change

McLachlan, JA. (2006) “Commentary: Prenatal exposure to diethylstilbestrol (DES): a continuing story.” International Journal of Epidemiology 2006 35(4):868-870

  • “… early exposure to DES may cause persistent epigenetic changes in some genes and not others such that the fate of tissues or organs is altered”

Bromer, JG et al. “Hypermethylation of Homeobox A10 by in Utero Diethylstilbestrol Exposure: An Epigenetic Mechanism for Altered Developmental Programming.” Endocrinology Vol. 150, No. 7 3376-3382

  • “DES has a dual mechanism of action as an endocrine disruptor; DES functions as a classical estrogen and directly stimulates HOXA10 expression with short-term exposure, however, in utero exposure results in hypermethylation of the HOXA10 gene and long-term altered HOXA10 expression.”

For more detail on how epigenetic changes can affect health, you can also read our detailed post here, about what epigenetics might mean for cancer prevention.

In the past, much research has focused on the effect of BPA on hormone receptors, though recently attention has begun to move to other ways in which BPA can interfere with cell development and hormone signaling.

Concern about BPA is growing rapidly, with the first series of mainstream in-depth coverage of the issues appearing in the UK press, while Denmark became the first EU country to ban BPA as a precautionary measure and the French Senate backed a proposal to ban the use of the chemical in baby products.

5&5: News and science highlights from March



Institute to investigate cancer hotspot in Czech Republic: The Czech Republic already has very high rates of kidney and colorectal cancer, but the region of Plzen has rates 50% and 30% even higher, respectively. The EU is funding a new research centre to find out why.

Better living through green chemistryThe New Scientist reports on how principles of green chemistry are becoming more important as the pharmaceutical and chemical industries find themselves under increasing pressure to make lower carbon, less toxic products.

Canada proposes action to control risks associated with 22 substances: The Canadian Ministry of the Environment has given public notice of its intention to add eight more chemicals and step up restrictions on 14 others, as part of the national Chemical Management Plan.

EU bans triclosan from food packaging [Legal text]:  The European Commission has announced a ban on the antimicrobial triclosan from food packaging. The ban will not extend to other uses, such as in toothpaste and soaps, because these applications are covered under other legal directives.

Bad chemistry: The poison in the plastic that surrounds us: Part of the Independent‘s extensive coverage of BPA, this excellent piece of science journalism looks in depth at the evidence for harm from BPA and how the opinion of environmental health scientists and industry are divided as to its safety.


Formaldehyde Exposure and Asthma in Children: This review of seven studies finds a “significant positive association” between formaldehyde exposure and childhood asthma.

House Dust Concentrations of “Safer” Flame Retardants May Lower Semen Quality: Higher concentrations of organophosphate flame retardants, used as alternatives to PBDEs, are found in this study to be associated with lower semen quality and shows some interference with free thyroxine levels.

Low-Dose Exposure to Persistent Organic Pollutants and DNA Hypomethylation: Human study finding evidence that greater exposure to POPs results in increased DNA methylation, an epigenetic change likely to interfere with correct gene expression. Study limitations mean confounding factors cannot be ruled out.

Preconception Polychlorinated Biphenyl Concentrations and Infant Birth Weight: Interestingly, the authors of this study conclude the preconceptual period may be a sensitive time window for effects of PCBs on development. Given that low birth weight is a major predictor of adult health, this small (but prospective) study is significant.

Steroidogenesis and Trophoblast Differentiation in the Placenta in Relation to Prenatal Phthalate Exposure: The authors report that higher urinary concentrations of phthalate metabolites were associated with reduced expression of genes involved in trophoblast differentiation, which suggests phthalates could be interfering with cell differentiation.


RSS feed for comments on this post.

  1. […] Issue 25 – Epidemiology […]

  2. […] Issue 25 – Epidemiology […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at
Entries and comments feeds.

%d bloggers like this: