Saturday, 17 October 2009

Scary Figures

Suicide rate 'one of the highest'

Jersey has one of the highest suicide rates in the world after a rise in the number of people taking their own life, the medical officer of health has said. Last year the number of people who killed themselves in Jersey increased by 40% compared to the previous year. The medical officer of health said the island's suicide rate, 14.9 per 100,000 people, puts Jersey in the globally high category for suicide. Dr Rosemary Geller said work was being done to reduce the number of suicides. She said she was "very concerned" after a "particularly high" number of people killed themselves in 2008. The health authorities commissioned a study by the University of Southampton into why this might have been. Dr Geller said the study had highlighted a number of issues, which the department was addressing. (Source: JEP (1))

Usual scary figures by the JEP. Actually the figure from 2000 to 2008 was 113, with "16 last year compared to a previous average of about 10".

It is not stated where that "previous average comes from". Certainly not the 113, because that gives (113-16)/8= 12.125% over 7 years from 2000 to 2007 by my reckoning.

They also present a figure of 12.8 deaths per 100,000 population.

Now 16 suicides, while a figure of concern is not so large in a population of, say, 90,000 approx.

1600/90000 = 0.02% (approximately)

Moreover, with such small numbers, any increase is going to be significant if expressed as a percentage rise. This is the same game often played by any headlines which cite percentages for "fastest growing". As a few examples:

Neopaganism, for example, is apparently the "fastest growing religion". This is mainly because it is the smallest. Growth is expressed as a percentage and seems larger than, for example, Catholicism. But if you count increase in total Catholics and total Neopagans, then increases in numbers of Catholics each year is considerably greater.

The Guardian has an article which says "Wind power becomes Europe's fastest growing energy sourceEurope installs 20 wind turbines a day and 10 EU states reach wind power capacity of more than 1GW"(2)

Again this sounds impressive until you reach the figures further down, which say: "The new wind power capacity, costing €11bn (£9.9bn), should, in a normal year, produce 142 TWH (terawatt hours) of electricity or about 4.2% of EU demand and abate 100m tonnes of CO2 a year – equal to taking more than 50m cars off Europe's roads."

That's still good for the greenhouse effect, but 4.2% of demand still means the increase is not a huge proportion of energy needs!

So returning to the suicides. With such small numbers, any change is going to be hugely significant if expressed as a percentage. Other factors may come into play - an increase in population size, for example, may lead to increase in the numbers of suicides.

There may be timing differences involved, the kind of statistical variation that occurs because some suicides just happen to fall in one year because the cut off point is a calendar year - this needs the statistical method of "moving averages" to smooth out differences over years. This is extremely important. Scotland's official statistics, for example, note that:

The numbers of suicides can fluctuate markedly from year to year, particularly for the smaller Health Board and Council areas. Therefore, some of the tables include 5-year moving annual averages, as these should provide a better indication of the overall long-term trend than the figures for the individual years.  As well as the figures for Scotland as a whole and the 5-year moving average, the Chart also shows the likely range of values around the moving average. This likely range of statistical variability in the figures is estimated by assuming that the numbers represent the outcome of a Poisson process, with the underlying rate of occurrence in each year being the same as the value of the 5-year moving average which is centred on that year.(4)

I'm not saying that the matter of suicides should not be a cause for concern. Of course it should. But scaremongering headlines like those in the JEP take the worst possible comparisons. The UK has a much vaster population. Scotland, for example, has 843 suicides listed in 2008. (or to be more precise, as they are "Deaths for which the underlying cause was classified as 'intentional self-harm' or 'event of undetermined intent'  registered in Scotland".

The rest of the UK is probably higher, although The Samaritans website reveals (4) that the rest of the UK figures will not be released until December 2009, which begs the question of where the JEP was getting its comparison figures from.

A headline such as 60% is an instant grabber. A headline such as "Jersey suicide rates rise massively to nearly 0.02% of the population" would not be nearly as attention seeking. I do think that journalists should consider their responsibility when highlighting figures in this way. While there is a need to draw attention to suicide as a social concern, there should be a more responsible way of doing so.


1 comment:

JTG said...

I think the simplest approach to the question is a desire for funding.