Wednesday 27 July 2011

Medication data show no link between wind farm and "syndrome"

A survey of four towns in Victoria and South Australia has shown that the symptoms of “Wind Turbine Syndrome” -- a collection of adverse health effects that have been associated with low frequency noise from wind turbines -- are experienced no more frequently near wind farms than in comparable areas. The research, lead by Head of the University of Adelaide Discipline of Medicine Professor Gary Wittert, is yet to be published, but was covered by ABC News here, and featured in Monday’s Four Corners program, “Against the Wind.”

Wittert used Pharmaceutical Benefits Scheme (PBS) data to map the prescriptions of medications for sleeplessness, blood pressure and other cardiovascular conditions, and found that these conditions are no more prevalent within 10 km of wind farms at Waubra, Yambuk, Snowtown and Hallett Hill than in control areas with similar demographics. This defies the expectations of those who argue that prolonged exposure to high levels of low frequency noise and infra-sound can cause adverse health effects up to 10 km from a wind farm.

Wind Turbine Syndrome, which encapsulates a range of symptoms such as sleeplessness, high blood pressure, migraines, depression, tinnitus and nausea, was defined by American paediatrician Dr Nina Pierpont in her 2009 book of the same name, based on a voluntary study of 10 families (38 subjects) in 5 countries. In Australia, the campaign has been led by Dr Sarah Laurie, an unregistered GP from South Australia and the Medical Director of the Waubra Foundation. Dr Laurie’s research so far (summarised here) is based on interviews with 60 subjects, mostly residents living near wind farms.

The worrying findings of Pierpont and Laurie have so far been based on small samples of voluntary respondents, and sceptics are quick to point out that no residents with wind turbines on their property have made health complaints. This year’s federal Senate inquiry into The Social and Economic Impact of Rural Wind Farms cited literature reviews by National Health and Medical Research Council (NHMRC) in 2010, and by the American Wind Energy Association (AWEA) and Canadian Wind Energy Association (CWEA) in 2009, which both found no suggestion in the existing research for a positive link between wind farm noise and adverse health effects. Neverthless, the Inquiry and NHMRC did recommend further research.

Based on Laurie’s findings, the Waubra Foundation calls for health studies in the form of residents’ personal health journals and ongoing noise monitoring. These studies may yield valuable data, but will take many years to deliver viable scientific conclusions. But if vibration and low frequency noise from wind farms cause adverse health effects, any number of historical data records should reflect the systematic increase of these symptoms in the area of wind farms. As Wittert points out, we have up to 25 years of experience of living and farming around wind farms in Australia, and over 30 years internationally:

[...] if you think about the density of turbines in some European countries, such as Germany and Scandinavia, one would expect that if there was that much infrasound causing people and animals to vibrate that you would notice some significant effects. There is no data coming out of Europe that there are any changes in lambing, in milk production, in the quality of beef or in any other form of agricultural products that involve livestock. Nor are there any reports of significant health effects coming out of any of those countries.

If exposure to wind farm noise was the cause of these severe health effects in humans and livestock, historical records should be sufficient to show the link. While acting as an expert witness for wind farm developer Acciona, Wittert conceived of an elegant method for quantifying the geographical and historical prevalence of these health effects. Before undertaking the study, he explained his planned methodology to the Senate inquiry:

Senator ADAMS—You were talking earlier about trying to get independent people. How would you go about that? You were talking about a 10-kilometre area for doing your PBS investigation. How would you find a cohort that was large enough within that particular area to do these other observations and have independent witnesses?

Prof. Wittert—That is why GIS methodology is probably the optimal way of doing it if you can get enough pool data. Let me give you an example. Let us say there were 2,000 dwellings within 10 kilometres of a set of turbines and I knew that the turbine was erected in 2007 or 2008. PBS could give me data over 4½ years. So let us say I can get data from 2006-07 and I know that there are a certain number of prescriptions being issued for blood pressure medication. Over the period of the next few years I could then determine whether there were any change in the provision of blood pressure medication. If I could do the same thing across two, three or four areas where there were turbines erected, I could compare those to other areas of similar demographics where there were no turbines erected. I would now be able to compare across time within an area where there is a turbine. I could compare across time in an area where there is no turbine. I could then compare between the two areas. I could do that across a range of conditions—assuming, of course, that there is a prescription issued for that condition and it is diagnosed. What you can then use is geographic information systems to overlay things like distance to the nearest doctor, a whole bunch of socioeconomic data, the age profile of the population et cetera.

In contrast with the dozens of voluntary respondents featured in Pierpont’s and Laurie’s studies, the PBS data account for more than 10,000 subjects near four different wind farms. This kind of geostatistical study removes questions of selection bias, and allows for quantitative comparison within each community before and after the installation of the wind farm, and against comparable communities without wind farm developments.

Given the health complaints of dozens of residents near wind farms, the NHMRC and Senate inquiry, alongside the Waubra Foundation, are right to call for more research. But the aims of that research, its methodology and the tone of its discussion in the community, deserve some scrutiny. Wittert’s study, based on historical data collected for an independent purpose, shows no link between proximity to wind farms and prevalence of the expected symptoms of Wind Turbine Syndrome. The methodology is straightforward, and the academic report is unlikely to be inflammatory. By contrast, the kind of small scale studies carried out by Pierpont and Laurie -- based on interviews and personal health diaries -- privilege the most dramatic accounts of aggrieved residents, and provide powerful anecdotes for emotive media coverage.

This focus on the subjective and emotional often pushes the actual science to the end of the article, without advancing knowledge about the issue. At the height of the Senate inquiry Professor Simon Chapman of the University of Sydney vented some frustration with a cutting analysis at the ABC's The Drum. In this context, Wittert's study is a welcome contribution to the scientific discourse. As he said to the Senate inquiry, “there is also an awful lot of anxiety being generated, and it would be nice to have some objectivity and rationality around it.”