Low‐level lithium in drinking water and subsequent risk of dementia: Cohort study

• Published in: International journal of geriatric psychiatry Vol. 38; no. 3; pp. e5890 - n/a
• Main Authors: Duthie, Ashleigh C., Hannah, Jean, Batty, G. David, Deary, Ian J., Starr, John M., Smith, Daniel J., Russ, Tom C.
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.03.2023
• Subjects: Animal models; Cohort analysis; Cohort Studies; Dementia; Dementia disorders; Drinking behavior; Drinking water; Drinking Water - adverse effects; Drinking Water - analysis; Electronic medical records; epidemiology; Female; Geriatric psychiatry; Humans; Lithium; Lithium - adverse effects; Male; Neuroprotection; Population studies; Sampling; Surveys and Questionnaires; United Kingdom > UK; Scotland; epidemiology; lithium; dementia
• ISSN: 0885-6230; 1099-1166
• DOI: 10.1002/gps.5890

Background Lithium, a mood stabilizer, is known to exhibit neuroprotective effects in animal models and may have anti‐dementia effects. Aims We used data from Scottish Mental Survey 1932, a population‐based cohort study, to investigate the association between lithium in drinking water and dementia rate in humans. Method Lithium levels in drinking water from 285 sampling sites across Scotland dating from 2014 were obtained from the sole public water provider (Scottish Water). Dementia and non dementia cases were identified from cohort data by electronic health records until 2012, and linked to postcode. Results The mean lithium level at all sampling sites was 1.45 μg/L (SD 1.83, range 0.5–18.2) and was 1.26 (SD 0.63, range 0.55–9.19) for sites matched to participant data. Of 37,597 study members, 3605 developed dementia until June 2012. Lithium levels were positively associated with the risk of dementia in women (highest in second quartile, HR 1.17, 95%CI 1.04–1.32), but there was no relationship in men (highest in second quartile, HR 0.95, 95% CI 0.81–1.12). The pattern of association was explored further by decile, and in females there was an association between lithium level and increased dementia risk compared to the lowest decile (0.55–0.68 μg/L) in all deciles except the highest, corresponding with lithium levels 0.68–2.1 μg/L. Conclusions Lithium levels in drinking water are very low across Scotland which limited detection of potential effect. Our results do not support an association between extremely low levels of lithium and later dementia risk. We found a trend to increased risk in females at lithium levels below but not above 2.1 μg/L. Key points Lithium levels in drinking water are very low across Scotland which limited detection of any potential effect. Our results do not support an association between extremely low levels of lithium in drinking water and later dementia risk. At very low lithium levels the trend for dementia risk in females was not linear.