Group One Carcinogens, revisted

Cancer is scary, so the usual lot are again pumping the alcohol-cancer link.

The Cancer Society of New Zealand says Kiwis are only now sobering up to the link between alcohol and cancer, just as we did more than 30 years ago with smoking and lung cancer.

Strong links between drinking more than two or three units a day have been established to deadly digestive tract cancers including mouth, throat, larynx and oesophageal cancers.

There are also strong links between alcohol and bowel, breast and prostate cancers.

The warning comes on as some Kiwis sign up for Dry July, a month without drinking alcohol. Those taking part are encouraged to sign up sponsors in support of cancer services.

 ...

Dr Jan Pearson, of the Cancer Society, said it was time Kiwis started talking about the risks of excessive drinking.

"We are probably at the stage now that we were at 30 years ago with tobacco," she said.

...

Otago University public health professor Tony Blakely said alcohol "absolutely" contributed to cancer rates.

"Alcohol has more obvious impacts on injuries, deaths, social bedlam, unwanted pregnancies and suicides," he said.

...

Christchurch Professor Doug Sellman, of the National Addiction Centre, said that 25 per cent of alcohol-related deaths "are actually cancer deaths.

"The ethanol in alcohol is a group one carcinogen, like asbestos," he said.

First, they're right: alcohol is a known Group 1 carcinogen, along with diesel exhaust, ciclosporin, estrogen-based oral contraceptives and menopausal therapy, risky sex (Hep B & C, HPV, HIV), the sun, mineral oils, salted fish, wood dust, painting, boot & shoe manufacture and repair, plutonium, and rather a few other things.

Alcohol has a bundle of positive and negative health effects. At low doses, for most people, the positive dominates. That's why we get a J-curve with about a 16% reduction in the risk of all-source mortality around one standard drink per day. After about 3-4 standard drinks per day, you start  exceeding non-drinkers' mortality risk. Unless you have a strong family history of particular diseases that are positively or negatively affected by alcohol, it makes far more sense to watch the aggregate J curve than to follow the disease-by-disease effects.

Update: Here's Thomas Lumley over at StatChat:

The phrase “group one carcinogen” is only relevant in an argument over whether the risk is zero or non-zero. Its use in other contexts suggests that someone doesn’t know what it means, or perhaps hopes that you don’t.

Dave Guerin at Ed.co.nz summarises:

1. Dodgy Advocacy Otago Uni’s Profs Doug Sellman and Tony Blakely commented on alcohol and cancer, with Sellman saying that the “alcohol in ethanol is a group one carcinogen”. This is a great example of Sellman’s dodgy advocacy work. Canterbury’s Eric Crampton points out a few other group 1 carcinogens (the sun, salted fish, wood dust), while Auckland’s Prof Thomas Lumley points out that being in group 1 simply means that a cancer risk has been established, high or low. Lumley went on to say that using the group 1 term in most contexts “suggests that someone doesn’t know what it means, or perhaps hopes that you don’t”. Quite.

Previously:

• A few things Alcohol Action NZ won't tell you about alcohol
• Group One Carcinogens
• The J-Curve
• The J-Curve: Science vs Politics
• Moderate drinking and health
• Moderate drinking and health - more evidence
• NZ's Ministry of Health and the J-Curve
• Differential standards of evidence
• Nanny's drinking problem