Lithium in the public water supply and suicide mortality in Texas

https://doi.org/10.1016/j.jpsychires.2012.12.002Get rights and content

Abstract

There is increasing evidence from ecological studies that lithium levels in drinking water are inversely associated with suicide mortality. Previous studies of this association were criticized for using inadequate statistical methods and neglecting socioeconomic confounders. This study evaluated the association between lithium levels in the public water supply and county-based suicide rates in Texas. A state-wide sample of 3123 lithium measurements in the public water supply was examined relative to suicide rates in 226 Texas counties. Linear and Poisson regression models were adjusted for socioeconomic factors in estimating the association. Lithium levels in the public water supply were negatively associated with suicide rates in most statistical analyses. The findings provide confirmatory evidence that higher lithium levels in the public drinking water are associated with lower suicide rates. This association needs clarification through examination of possible neurobiological effects of low natural lithium doses.

Introduction

The positive effects of lithium on mood and mental health were first described in the late 1940s (Cade, 1949). Lithium has since been widely used as a first-line treatment for bipolar disorder, and as an augmentation treatment for unipolar depressive disorder (Bauer et al., 2006). The suicide-protective properties of lithium in therapeutic dosages for treating bipolar (Baldessarini et al., 2006; Goodwin et al., 2003) and major depressive disorder (Baldessarini et al., 2003; Guzzetta et al., 2007) are well established.

There is growing evidence from ecological studies that lithium traces in ground and drinking water may protect against suicide. A negative association between lithium concentrations in the municipal water supply and suicide rates was first reported for 27 Texas counties (Schrauzer and Shrestha, 1990). This finding was affirmed in the Oita prefecture in Japan (Ohgami et al., 2009). Several methodological concerns were raised about these results regarding the used statistical methods (Huthwaite and Stanley, 2010) and the failure of the investigators to adjust for possible socioeconomic factors that influence suicide rates (Chandra and Babu, 2009).

A study conducted in the East of England, using separate measurements of lithium for 47 subdivisions, found no correlation between lithium levels in tap water and suicide rates (Kabacs et al., 2011). A nationwide Austrian study, however, which used 6460 lithium measurements from 99 districts, found an inverse correlation between lithium concentrations in drinking water and suicide rates (Kapusta et al., 2011). Well-known socioeconomic determinants of suicide were adjusted for in the analysis. In a subsequent commentary on that study, use of more appropriate statistical methods was recommended for validation purposes for future research (Yang, 2011).

This current study incorporates salient socioeconomic determinants of suicide and employs pertinent statistical procedures in revisiting the research question on a nexus between lithium levels in drinking water and suicide rates. Our geographic focus was the area where the association was first documented, namely Texas (Schrauzer and Shrestha 1990). In performing both linear and Poisson regression analyses, using more recent data for a very large set of Texas counties, we adjusted for various potential socioeconomic confounders and utilized a large number of water samples.

Section snippets

Data acquisition

Applying public health informatics techniques, we linked data abstracted from three sources: the Texas Department of State Health, the US Census Bureau, and the Texas Water Development Board Groundwater Database. We retrieved the most recent (1999–2007) crude and age-adjusted suicide mortality rates per 100,000 for 254 Texas counties from the Texas Department of Health (http://soupfin.tdh.state.tx.us/deathdoc.htm), which are based on a subset of variables drawn from the Texas Certificate of

Results

The geographical distributions of SRadj. and lithium levels for our selected set of Texas counties are illustrated in Figs. 1 and 2.

We considered linear regression, weighted linear regression for adjusted rates, Poisson rate regression using the crude rate as a surrogate for the age-adjusted rate, and quasi-Poisson regression on the age-adjusted rates. In assessing model adherence to the underlying assumptions for each modelling strategy, we determined that the optimal approach was Poisson rate

Discussion

Our study provides further evidence that lithium levels in the public water supply in Texas are negatively associated with suicide mortality rates at the county level. Utilizing more appropriate statistical procedures and a more recent time period, this research agrees with the findings of a study conducted in Texas during the 1970–1980s (Schrauzer and Shrestha, 1990), and also those from studies conducted in Austria (Kapusta et al., 2011) and the Oita prefecture in Japan (Ohgami et al., 2009).

Role of the funding source

None.

Contributors

Nestor Kapusta, Ian Rockett, and Victor Blüml designed the study and wrote the protocol. Benjamin Vyssoki, Victor Blüml, and Tom Bschor managed the literature searches and analyses. Michael D. Regier, Gerald Hlavin, and Franz König undertook the statistical analysis. Nestor Kapusta and Victor Blüml wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.

Conflict of interest

We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.

We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us.

We confirm that we

Acknowledgements

We thank Gerald Hobbs, PhD, for statistical support.

References (19)

  • R.J. Baldessarini et al.

    Decreased risk of suicides and attempts during long-term lithium treatment: a meta-analytic review

    Bipolar Disorders

    (2006)
  • R.J. Baldessarini et al.

    Lithium treatment and suicide risk in major affective disorders: update and new findings

    Journal of Clinical Psychiatry

    (2003)
  • J.F. Cade

    Lithium salts in the treatment of psychotic excitement

    The Medical Journal of Australia

    (1949)
  • R.J. Carroll et al.

    Measurement error in nonlinear models: a modern perspective

    (2006)
  • P.S. Chandra et al.

    Lithium in drinking water and food, and risk of suicide

    British Journal of Psychiatry

    (2009)
  • G. Desai et al.

    Lithium in drinking water and food, and risk of suicide

    British Journal of Psychiatry

    (2009)
  • F.K. Goodwin et al.

    Suicide risk in bipolar disorder during treatment with lithium and divalproex

    Journal of the American Medical Association

    (2003)
  • P. Gustafson

    Measurement error and misclassification in statistics and epidemiology: impacts and Bayesian adjustments

    (2004)
There are more references available in the full text version of this article.

Cited by (85)

View all citing articles on Scopus
View full text