Tuesday 18 October 2011

Vitamins aren't helping you (unless you are genuinely ill)... in fact, they may be killing you

Taking vitamin pills are a great idea if you suffer a genuine vitamin deficiency. Whether that's due to some chronic illness or special diet (vegans, I'm looking at you), there are those who need them and absolutely should be taking them.

The overwhelming majority have no need of them, however. If you are already reasonably healthy (and that's a surprisingly low barrier), vitamins are going to do nothing for you. Mankind evolved to deal with scarce food resources, and the body is awfully good at scavenging necessary vitamins and minerals from our everyday diets. "More is better" is never the way biology works - too little is bad, but enough is fine, and more won't help. Antioxidant vitamins do nothing.

At some point, too much will be bad for you, no matter how benign the substance appears. Everything becomes toxic at a high enough dose. Even vitamin overdoses occur, though they are very rarely fatal.

More dangerous and insidious (and the real point of this post) is the possibility that by assuming you are doing something healthy, you will then justify other unhealthy activity- smoking, exercising less, eating less well, etc. This is "risk compensation". Overall, you take vitamins (which don't help) and then go on to do something bad for you - a net loss for your health. This is bad in itself - and serves as an example of how things that defenders pass off as "harmless even if they don't work" can be of real harm.

Street Lights Are Not Making You Safer

Long time since last post, got some other things brewing, but while we're waiting, here's a quickie...

Most people would assume that street lighting is a good thing for many reasons, but principal among them would be safety. I would intuitively have thought the same, and given the choice between a dark alley and brightly-lit street, there's no doubt in my mind which I would take. But maybe I'm wrong about that, and many others along with me.

This post is precipitated by a story that solar streetlighting in Haitian camps has coincided with a large drop in reports of crime. So far, so great, good news everybody. But I use "coincided" for a reason. Security patrols improved over the same period, and the population of the camp dropped. While I'm sure the lights had many other advantages, there is too much happening here to assume that correlation is causation.

And wider research seems to show that, in fact, street lighting does not (in general) decrease crime. Of course, there are many other advantages - replacing wood-burning for light, increasing hours of labour, making pedestrian areas more pleasant for the cafe culture - but crime reduction may not be among them. Security legend Bruce Schneier drew attention to this counterintuitive finding back in 2007. Lighting may increase safety where it increases likelihood that crimes will be noticed and reported, but do not necessarily prevent crime in the first place. And a bit of thought produces hypotheses why this would be; for example, a burglar using a flashlight is more obvious than an brightly-lit figure fiddling with a lock. Bright lighting can give potential victims night-blindness while their potential muggers wait in the shadows. Overall, those lights are not making you safer - check out the evidence.

As I said before, I can still see lots of reasons to use public lighting in specific situations, where the light itself is the point of the exercise. But it comes with downsides - cost, pavement space, prevention of astronomy, and (as a country boy in the city) blocking the simple beauty of the night sky. If it isn't making people more secure (particularly when it is used in secluded areas, where the other benefits are minimal), a reappraisal is in order.