Sunday, 16 June 2013

Public health realpolitik

I've written two posts recently about the role of the industry in alcohol and tobacco policy.  The principle behind these was that, following a particular model of how policymaking could or should be done, there's a role for both industries - so long as it's at the right stage of the process: defining what's a problem about tobacco or alcohol, and what sort of government actions are legitimate.  There's less of a role for the industry (if any) in assessing the validity of research evidence.

I started to think about this issue again recently when the Daily Express, amongst others, ran a story noting that the risk of cancer increased with the consumption of just two drinks in a year.  Some responded to this idea by suggesting that the public health lobby was in danger of becoming a 'lunatic fringe', and comparing it to the temperance movement.

This critique is based on the idea that such a small level of risk is largely irrelevant to people who don't live their lives as risk minimisers.  There's also the additional concern around 'crying wolf' that I've noted before with reference to 'binge' drinking social marketing: if you tell people a couple of drinks a year (or even a month) is seriously bad for you, then they might not listen when you're telling them that more than 21 units a week is genuinely risky (if that's a more important message to get across).

There are some fundamental arguments here about the nature of alcohol and the role of public health professionals and researchers.  One of the key distinctions between smoking and drinking in terms of public health ambitions and tactics, apart from the issue of passive smoking, is that there is generally considered to be a 'safe' limit of alcohol consumption.  Indeed, sometimes certain amounts of alcohol are understood to be beneficialThis impression is only strengthened by 'responsible' or 'sensible' consumption levels, which then offer a concept on which to hang a narrative that constructs your own drinking as unproblematic.*

The presentation of this sort of finding that two beers a year might harm your health could be seen as attempt to change this impression.  If just two beers a year can increase risk levels, then the message seems to be that there's no safe level of consumption.  This is certainly what Stephen McGowan thought on Twitter, and to some extent it seems to be the motivation of Ian Gilmore in commenting on the findings that drinking even within the government guidelines can be risky and has links with cancer.

There are genuine arguments here about how best to communicate public health messages.  However, I'm not sure that this is actually based on calculations about how to best to engage people or change their approach to alcohol.  It seems more sensible to view it in the context of the arguments about the appropriate role of the alcohol industry in policymaking and the assessment of evidence.

Taking this perspective, the whole approach of both 'sides' in the debate (public health and the industry) is somewhat dispiriting.  On the one hand the industry steps in to rubbish research findings, when what it's really saying is that regulation of the industry doesn't fit the principles of maintaining the free market and personal responsibility that the government is often keen on in other contexts.  On the other, public health lobbyists are driven to stress that any form of alcohol consumption is problematic, in order to compete with the attempts of the industry to downplay the link between consumption and health harm.

Neither of these positions is helpful for an open, clear debate.  I can get particularly frustrated with a public health position because I think there's real opportunities for great work in this area.  With the move to local authorities there could be steps to improve wellbeing by integrating work with transport networks to encourage walking and cycling, with schools to encourage healthy eating, and with adult and community services to look at fostering the sort of social capital that we know improves health in the long term.  There's the opportunity to take a broader view of people's health, and really consider the wellbeing element as well as health as the absence of sickness.

However, the sort of risk minimisation approach symbolised by the 'two beers' story and all the talk of 'avoidable deaths' is too narrow.  Risk is unavoidable, and all lives end in death.  This isn't just about a critique of the Longer Lives project; I want to suggest that a grown-up debate about wellbeing would accept that sometimes drinking at a 'risky' level could still be beneficial for someone's wellbeing - particularly later in life.  (And this is in addition to the point I've made before about how rationalism needn't be a universal aspiration.)

The reason the public health lobby can't bring itself to do this, I'd suggest, is that it sees itself as locked into some kind of dialectical confrontation with the alcohol industry.  It's taking the view of a hardened haggler in a market: start with a ridiculously low offer, and you'll end up with something reasonable and acceptable, because the seller will start with a price far higher than the item's worth.

I'd suggest that this kind of adversarial approach does nobody any favours.  Public health gets branded as a 'lunatic fringe', and the industry gets labelled as misrepresenting, distorting and undermining research evidence.  Drinkers probably simply get on with their lives none the wiser either way.

My soft spot for corporatism wants me to suggest that this would be the solution, getting everyone together to discuss the issues.  But actually this is a little naive.  I forget that this approach brought down both Tory and Labour governments in the 1970s.  To some extent you could see corporatism, with its views of opposing factions, as cementing this adversarial approach.

But, in that case, how else can we approach policymaking?  How can we ensure that stakeholders take a grown-up approach?  That fundamentally requires a certain level of trust from both sides that the other will be sensible too.  It requires the different players to understand their roles, and stick to them.  Public health won't want to stop (excessively) highlighting the health harms alcohol can cause while the industry keeps its current tack, as that would mean the overall balance of the message to the public on alcohol would be (in their eyes) too positive.

The only way to encourage this that I can see is to have a strong government that is clear about the different roles of the various stakeholders in the policy-making process.  I can't see that being the government we have now.

*This is particularly the case when 'binge' drinking is largely defined in terms of an attitude towards to alcohol (drinking to get drunk), and so those who deny this motivation are able to construct an idea of 'responsibility' that isn't exactly what the public health lobby, at least, would like to see.


  1. I'm reasonably convinced by the DiCastelnuovo et al metastudy on all-source mortality and alcohol consumption showing a J-curve, with minimum risk somewhere around 1-2 drinks per day and risk equivalent to non-drinking somewhere around 4 drinks per day.

    Where alcohol has beneficial effects that more than offset harmful effects, relative to non-drinking, I get a bit nervous about public health folks who push the "even two drinks increases cancer risk" line. Sure, it's relevant for folks with lots of family history of cancer and no family history of heart disease. But for everyone else, isn't all-source mortality the more relevant measure?

  2. Cheers Eric.

    I can see that there might be some benefits to low alcohol consumption (particularly for specific conditions), and this is one of the key reasons that alcohol can't be treated in the same way as tobacco for Public Health.

    However, this then has implications for Public Health communication, as it opens up the door to people explaining that their own drinking isn't problematic - it's some of that 'responsible' drinking, when 'responsible' is a very malleable concept.

    As far as I can see, as I'm going to post soon, the sorts of alcohol consumption that are deemed problematic are those which are associated with those with less symbolic power in society. This is why the idea of 'unproblematic' alcohol consumption is difficult - because who defines what's problematic?

    (This isn't any kind of answer, but it's the thought the constantly drifts around my head whenever anyone condemns or defends a form of drinking.)

  3. Truth, as I read the lit, is that best advice is to drink about a glass of wine a day (bit less than one for women, bit more than one for men) and not to drink more than a couple of pints of beer at a session.

    So we then have a couple of options.

    1) Tell people not to drink more than a couple pints per day and that while there are some health benefits from drinking about a glass of wine or a glass of beer per day, those disappear if you have a couple pints and that you're probably hurting your health if you make a habit of having more than 2 pints a day - and that the health costs rise substantially with consumption from that point.

    2) Focus on all the risks so as not to encourage people to justify their own harmful drinking habits. I have two worries about this approach:
    i) some will heed the advice - especially those who would otherwise not have that hard a time sticking with moderate drinking. They'll forego some consumption benefits.
    ii) others will listen, then look around at their moderate-drinking friends and neighbours, and start discounting everything the public health folks have to contribute because the one bit they hear a lot about conflicts so drastically with what they see in the real world.

    Suppose it's an empirical question, but it would take a lot of very substantial evidence to convince me it's a good idea to lie to people.