In 1990 we reported that the mean suicide rates in 27 Texas counties over a 10-year period were consistently lower in those with ‘high’ natural lithium content in the drinking water (70–160 μg/l) than in counties with ‘medium’ (12–60 μg/l) or ‘low’ (0–10 μg/l) water lithium levels. Reference Schrauzer and Shrestha1 Ohgami et al Reference Ohgami, Terao, Shiotsuki, Ishii and Iwata2 have since argued, without proof, that these associations may have been spurious owing to what they considered an arbitrary division of the data. It is necessary, therefore, to emphasise that the data were partitioned in accord with accepted methods of statistical trend analysis and not in an arbitrary fashion, and that tests were conducted to assure that the partitioning of the data did not produce spurious associations.
Within the same study, Reference Schrauzer and Shrestha1 we found the rates of homicide, rape, robbery, burglary and theft to be also lower in the high-lithium counties. In addition, a statistically significant reciprocal relationship between the water lithium levels and the arrest rates for possession of opium, cocaine and their derivatives was observed, while the arrest rates for lesser crimes such as possession of marijuana, drunkenness and driving under the influence showed no consistent dependence on the water lithium levels. The studies were later extended to include arrest rates of juveniles, yielding statistically significant results for possession of narcotic drugs and, interestingly, ‘runaway from home’. Reference Schrauzer, Shrestha, Schrauzer and Klippel3
In the interest of historical accuracy it needs to be pointed out that in 1972 Dawson et al Reference Dawson, Moore and McGanity4 reported mental hospital admissions and homicide rates to be lower in high-lithium Texas counties. They also found the suicide rates to be lower in these counties, but the differences did not reach statistical significance, as incidence data for only a 2-year period (1968–1969) were compared.
Thus, the evidence in favour of beneficial effects of low levels of lithium on human behaviour is already strong, and since lithium is close to be officially recognised as a nutritionally essential trace element, Reference Schrauzer5 emphasis should be placed on assuring adequate lithium intakes in populations at risk of developing lithium deficiency.
eLetters
No eLetters have been published for this article.