Lithium in drinking water and prescriptions effects on suicide mortality

Suicides are under the ten most common causes of death. An urban-rural analysis of the past three decades showed that suicide risk in urban regions has more rapidly decreased as compared to rural areas. However, the geographic variation of suicide rates in Austria is still not fully understood. Geographic analyses of the suicide mortality distribution in Austria employing a hierarchical Bayesian model that accounted for spatially correlated random effects showed that a significant proportion of the variance of regional suicide rates may predominantly be explained by population density - a marker of urbanity -, the mean per capita income of the region, and the religious composition of the population. However, given the existing evidence that lithium, a suicide preventive, naturally contained in usual drinking water from the tap, may have effects on suicide mortality, further analyses of the suicide distribution in Austria were done. Lithium in drinking water remained a significant contributor to the geographic pattern of suicide mortality in simple linear regression models as well as in a Geographically Weighted Regression (GWR) model. However, to challenge this result, other possible factors have to be considered. There are numerous reports pointing to the public health problem of pharmaceuticals in waste water. Pharmaceuticals like carbamazepine and ibuprofen were also found in finished drinking water in Canada and acetaminophen caffeine, carbamazepine, codeine, and other substances were reported in Californian drinking water. Also in the In the United States, the antidepressant Fluoxetine has been detected in 4.3% of all examined ground water sites. The effects of Fluoxetine and Carbamazepine concentrations in sediment and water systems are strong enough to affect health and behaviour in aquatic invertebrates. It is possible that the lithium distribution in drinking water is due to prescribed lithium rather than of natural origin only. Lithium is used in medicine since more than 60 years as a mood-stabilizer and suicide preventive and could have accumulated in sediments. If true, the geographic pattern of suicides would be not determined by natural lithium effects but by the geographical pattern of prescription practice, while levels of lithium in drinking water would suggest pollution with medication instead of natural levels. The main hypothesis of this project is that prescription rates of lithium are associated with lithium contents in drinking water.

The study examined the possible preventive effect of lithium drug prescriptions on the mortality by suicides in Austria. International studies have shown in recent years that lithium which occurs naturally in drinking water in small amounts is linked to suicide mortality. This means that the more lithium is contained in the drinking water in a region, the smaller is the suicide mortality in the corresponding region. Although the relationship remained significant even when economic factors of the region were taken into account (such as unemployment or for example, the density of physicians), an open question remained, namely whether this relationship is only by chance or can be explained by another factor. An important question was whether the prescription of lithium drugs which are used successfully in the treatment of recurrent depression or bipolar disorder for 60 years, could have an impact on regional suicide mortality. On one hand, these drugs are taken by persons with elevated suicide risk, which should have a suicide preventive effect in patients, on the other hand they are excreted and likely enter aquifers and by this way reach drinking water, where they could be taken up by the population - an undesirable but possible positive (cohort-)effect. Therefore, the geographic data of all Lithium prescriptions in Austria were assessed in the study. The results show, that lithium prescriptions are only slightly correlated with the lithium in drinking water and do not justify the hypothesis that the lithium content in drinking water is influenced by prescriptions. On the other hand, prescriptions show no correlation with suicide rates, so that the otherwise known suicide protective effect of lithium treatment cannot be found on an ecological level. Therefore the previously established hypothesis remains indicating that lithium deposits in drinking water stem from natural origins and are associated in a still not understood way with lower suicide mortality.