Sage Journals: Discover world-class research

Abstract

Background:

Lithium in drinking water may have significant mental health benefits. We investigated the evidence on the association between lithium concentrations in drinking water and their neuropsychiatric outcomes.

Methods:

We conducted a systematic review and meta-analysis and searched Pubmed, Embase, Web of Science, PsycINFO and CINAHL up to 19 January 2020, for peer-reviewed research examining the association between lithium concentrations in drinking water and neuropsychiatric outcomes. We used a pairwise analysis and a random effects model to meta-analyse suicide rates and psychiatric hospital admissions. We assessed for publication bias using Egger’s test and Duval and Tweedie’s Trim and Fill analysis.

Results:

Twenty-seven studies including 113 million subjects were included in this systematic review. Meta-analysis of 14 studies including 94 million people found higher lithium concentrations were associated with reduced suicide rates (r = −0.191, 95% confidence interval = [−0.287, −0.090], p < 0.001) and meta-analysis of two studies including 5 million people found higher lithium concentrations were associated with fewer hospital admissions (r = −0.413, 95% confidence interval = [−0.689, −0.031], p = 0.035). We found significant heterogeneity between studies (Q = 67.4, p < 0.001, I² = 80.7%) and the presence of publication bias (Egger’s test; t value = 2.90, p = 0.013). Other included studies did not provide sufficient data to analyse other neuropsychiatric outcomes quantitatively.

Conclusion:

Higher lithium concentrations in drinking water may be associated with reduced suicide rates and inpatient psychiatric admissions. The relationship with other neuropsychiatric outcomes and complications remains unclear. Further research is required before any public health recommendations can be made.

Trial registration number: The study was registered with PROSPERO, number CRD42018090145.

Keywords

Neuropsychiatric outcomes lithium drinking water suicide public health

Introduction

Even after 70 years since John Cade’s (2000) ground-breaking discovery, lithium remains the ‘gold standard’ for bipolar disorder (Malhi et al., 2015, 2017; Tondo et al., 2019). In the clinical setting, the recommended lithium serum level is 0.60–0.80 mmol/L (Malhi et al., 2018b, 2020; Nolen et al., 2019). This can be reduced to 0.40–0.60 mmol/L where there is good response but poor tolerance or, alternatively, increased to 0.80–1.00 mmol/L for insufficient response but good tolerance (Nolen et al., 2019). The mechanism of its action, however, is still not completely understood (Malhi et al., 2013; Malhi and Outhred, 2016). Active research is currently undergoing to elucidate the mechanism of action, as well as the expansion of lithium’s therapeutic horizons in light of its neuroprotective properties (Lauterbach and Mendez, 2011; Malhi, 2018) and its suicide-preventing qualities (Bauer and Gitlin, 2016; Tondo et al., 2019).

One developing area of interest is the presence of lithium in drinking water. Lithium is a natural trace element often present in drinking water and is a naturally occurring food ingredient (Dolara, 2014). Although the concentration in drinking water is markedly lower than that used in clinical practice (Pompili et al., 2015), a growing body of literature suggests that trace amounts of lithium in drinking water may be protective against various neuropsychiatric outcomes.

The potential benefit of lithium in drinking water was first suggested by Dawson et al. (1970) who found that higher concentrations of lithium in drinking water were associated with lower rates of hospital admissions. Since then, many large epidemiological studies have investigated the effects of lithium in drinking water on neuropsychiatric conditions and complications including suicide (Blüml et al., 2013; Giotakos et al., 2013; Ishii et al., 2015; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Schrauzer and Shrestha, 1990; Shiotsuki et al., 2016; Sugawara et al., 2013), dementia (Kessing et al., 2017a; Parker et al., 2018), bipolar disorder (Kessing et al., 2017b; Parker et al., 2018; Schullehner et al., 2019), schizophrenia (Schullehner et al., 2019) and depression (Parker et al., 2018). To our knowledge, there has been no study that has systematically reviewed and meta-analysed the potential therapeutic benefits of lithium in drinking water across different neuropsychiatric outcomes. In the current systematic review, we examined the effect of differing lithium concentrations in drinking water on a wide range of neuropsychiatric outcomes. We hypothesised that higher concentrations would be associated with lower rates of neuropsychiatric outcomes.

Method

Search strategy and selection criteria

We used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline (Moher et al., 2009). We searched Pubmed, Embase, Web of Science, PsycINFO and CINAHL until 19 January 2020, for studies investigating lithium in drinking water and neuropsychiatric outcomes. We also used SCOPUS to check the references and citations of included studies and relevant reviews. Our search terms focused on lithium, drinking water and the neuropsychiatric outcomes of interest (Supplementary Table 1). These included suicide, psychiatric hospital admissions, bipolar disorder, dementia, schizophrenia, major depressive disorder, psychotic experiences and depressive and anxiety symptoms. The inclusion criteria for our review included (1) an observational design (cohort, cross-sectional, case–control or longitudinal), (2) published in a peer-reviewed journal and represented original research and (3) an investigation of lithium concentrations in drinking water and its relationship with the chosen neuropsychiatric outcomes. We had no language restrictions. This study was prospectively registered with PROSPERO (registration number: CRD42018090145).

Two authors (E.M. and B.E-W.) independently screened titles and abstracts for eligible studies, followed by a full-text review for studies that met the inclusion criteria. Any disputes were settled through discussion with another author (S.S.). Two authors (E.M. and B.E-W.) independently extracted the following data from the included studies: authors, publication year, study region, study design, years investigated, sample size, region number, lithium concentration, neuropsychiatric outcomes, neuropsychiatric outcome assessment tools and all relevant results. Authors of eligible studies were contacted for additional information if further clarification was needed.

Data analysis

We found studies investigating the relationship between lithium concentration in drinking water and suicide, psychiatric hospital admissions, bipolar disorder, dementia, major depressive disorder, schizophrenia, schizophrenia spectrum disorders, psychotic experiences and depressive and anxiety symptoms. The correlations were reported using incidence rate ratios (IRR), relative risk (RR), rates of incidence, odds ratio and beta (β), T, r and r² values. Significance was reported as p values or confidence intervals (CI).

Regression coefficients extracted from each of studies were converted to the correlation coefficient ‘r’, using published formulas (Lipsey and Wilson, 2001; Peterson and Brown, 2005) in order to standardise the effect sizes between studies (Supplementary Table 2). Data from the most adjusted findings from each paper, including log-transformed lithium concentrations if the distribution of lithium was skewed, and adjustments accounting for confounders were used. Following this, Comprehensive Meta-Analysis 3.0 was used to pool the correlation coefficients from individual studies in a meta-analysis and conduct a pairwise analysis (Supplementary Table 3). Studies which only provided data by year (Pompili et al., 2015) or gender (Kapusta et al., 2011; Sugawara et al., 2013) were included as separate datapoints by year or gender, then combined to provide overall study data.

Subgroup analyses were applied to hospital admissions to examine differences by psychiatric diagnoses. Random-effects models were used throughout, in order to account for the expected heterogeneity between studies. A sensitivity analysis was also conducted excluding low-quality studies and Ohgami et al. (2009) as the standardised β coefficient was outside the −0.5 to 0.5 range that is recommended to transform β to r values (Peterson and Brown, 2005). We further excluded Liaugaudaite et al. (2017) in a separate sensitivity analysis, as it shared some data with a larger study conducted by Liaugaudaite et al. (2019). A sensitivity analysis of excluding studies that did not provide adjusted results was undertaken, as was a sensitivity analysis of removing each individual study to explore each study’s impact on overall effect.

As the lithium concentrations and its correlations with neuropsychiatric outcomes were measured across geographical regions rather than individuals, we used the number of municipalities investigated as the population in our analyses (Table 1). In addition, a meta-regression was conducted on studies investigating suicide rates where the mean lithium concentration was reported to assess whether a dose–response relationship existed.

We used a modified Newcastle-Ottawa scale to assess the risk of bias (Mata et al., 2015). The scale assesses the following: (1) the representativeness of the sample, (2) sample size, (3) non-respondents, (4) ascertainment of neuropsychiatric outcome and (5) quality of descriptive statistics reported. A score of 3 or more was classified as low risk, and a score of 2 or less was classified as high risk (Supplementary Table 4). Publication bias was investigated by Egger’s test (Egger et al., 1997), Duval and Tweedie’s Trim and Fill analysis (Duval and Tweedie, 2000) and examining the funnel plot.

Table 1.

Summary of included studies for the association between lithium in drinking water and neuropsychiatric outcomes: a systematic review and meta-analysis from across 2678 regions containing 113 million people.

Author (year)	Country	Regions studied	Year(s) investigated	Sample size	Lithium level range	Lithium levelMean (SD)	Neuropsychiatric outcomes (measurement used)	Neuropsychiatric outcome measurement tool	Main findings associated with lithium concentrations in rinking water
Ando et al. (2017)	Japan	24	2008–2009	3040	0.01–2.10 μg/L	0.48 μg/L (0.52)	Depressive Symptoms	Modified Generalised Health Questionnaire 12	Small, significant, inverse association with depressive symptoms
Ando et al. (2017)	Japan	24	2008–2009	3040	0.01–2.10 μg/L	0.48 μg/L (0.52)	Suicidal Ideation and Self-Harming behaviour	Self-reported Questionnaire	No association with Suicidal Ideation or Self-Harming
Blüml et al. (2013)	USA	254	1999–2007	NA	2.8–219 μg/L	NA	Completed Suicide (Suicide Mortality Rates per 100,000)	No formal diagnostic criteria	Significant, inverse association with suicide mortality rates
Dawson et al. (1970)	USA	27	1967–1971	NA	0–160 μg/L	NA	State Mental Health Hospital Admissions for Neurosis, Psychosis or Personality diagnoses	No formal diagnostic criteria	Significantly reduced mental health hospital admissions for all diagnoses
Dawson et al. (1972)	USA	24	1967–1969	6,000,000	0–139 μg/L	NA	State Mental Health Hospital Admissions for Neurosis, Psychosis or Personality diagnoses	Diagnostic and Statistical Manual of Mental Disorders (1968)	Significantly reduced mental health hospital admissions for all diagnoses.
Dawson et al. (1972)	USA	24	1967–1969	6,000,000	0–139 μg/L	NA	Completed Suicide per 10,000	No formal diagnostic criteria	No significant association with suicide
Giotakos et al. (2013)	Greece	34	1999–2010	10,086,000	0.1–121 μg/L	11.1 μg/L (21.16)	Completed Suicide (average standardised suicide rate per 100,000)	No formal diagnostic criteria	Significant, inverse association with suicide rates
Ishii et al. (2015)	Japan	274	2010–2013	14,646,121	0–130 μg/L	4.2 μg/L (9.3)	Completed Suicide (Suicide Mortality Rates per 100,000)	No formal diagnostic criteria	Significant, inverse association with suicide rates in some analyses and in men
Kabacs et al. (2011)	England	47	2006–2008	5,700,000	<1–21 μg/L	NA	Completed Suicide (Suicide Mortality Rates per 100,000)	ICD-10 codes: X60–X84	No association with suicide rates
Kapusta et al. (2011)	Austria	99	2005–2009	8,297,964	0–1300 μg/L	11.3 μg/L (27)	Completed Suicide (Suicide Mortality Rates per 100,000)	No formal diagnostic criteria	Significant, inverse association with suicide mortality rates
Kessing et al. (2017a)	Denmark	275	1970–2013	73,731	0.6–30.7 μg/L	NA	Diagnosis of Dementia	ICD-8 and ICD-10 Codes	Non-linear relationship that may indicate a significant inverse relationship with incidence of dementia
Kessing et al. (2017b)	Denmark	276	1995–2013	14,820	2–27 μg/L	NA	Diagnosis of Mania/Bipolar Disorder	ICD-8 and ICD-10 Codes	No significant association with incidence of bipolar disorder
Knudsen et al. (2017)	Denmark	275	1991–2012	3,740,113	0.6–30.7 μg/L	11.6 μg/L (6.8)	Suicide	ICD-8 and ICD-10 Codes	No protective effect of exposure to lithium on the incidence of suicide
Liaugaudaite et al. (2019)	Lithuania	54	2012–2016	2,273,508	1.0–39 μg/L	11.5 μg/L (9.9)	Suicide	ICD-10 codes	Significant non-linear relationship with suicide rates
Liaugaudaite et al. (2017)	Lithuania	9	2009–2013	1,190,261	0.48–35.53 μg/L	10.9 μg/L (9.1)	Completed Suicide (Suicide Mortality Rates per 100,000)	ICD-10 codes: X60–X84	Significantly associated with lower suicide rates overall and in men
Ohgami et al. (2009)	Japan	18	2002–2006	1,206,174	0.7–59 μg/L	NA	Completed Suicide (Suicide Standardised Mortality Rates per 100,000)	No Formal Diagnostic Criteria	Significantly associated with lower suicide mortality rates overall and in men
Oliveira et al. (2019)	Portugal	54	2011–2016	3,323,085	0.1–191 μg/L	10.88 μg/L (27.18)	Completed Suicide (Suicide Standardised Mortality Rates per 100,000)	ICD-10	No association was found with suicide
Oliver et at. (1976)	USA	1	1972–1973	384	0–32 μg/L	NA	Depressed mood and expressed unhappiness	CES-Depression scale, Lubin Depression Adjective Checklist, General Happiness Scores	Significantly associated with increased rates of worse mood symptoms in small subgroup of housewives with increasing lithium levels
Palmer et al. (2019)	USA	15	1999–2013	2,016,763	0.1–60.6 μg/L	NA	Completed Suicide rate per 100,000	No formal diagnostic criteria	Significantly decreased suicide rate overall and in men with higher lithium concentrations
Parker et al. (2018)	USA	174	1992–2012	4,227,556	0–>40 μg/L	27.4 μg/L (NA)	Diagnosis of Bipolar Disorder, Dementia or Major Depressive Disorder	No Formal Diagnostic Criteria	No significant association after adjusting for confounders for bipolar disorder, major depressive illness and dementia
Pokorny et al. (1972)	USA	254	1967–1969	NA	0–160 μg/L	NA	State Mental Health Hospital Admissions for Neurosis, Psychosis or Personality diagnoses	No Formal Diagnostic Criteria	No direct or causative relationship between drinking water lithium and mental hospital admissions
Pompili et al. (2015)	Italy	145	1980–2011	17,200,000	0.110–60.8 μg/L	5.28 μg/L (NA)	Completed Suicide (Standardised Mortality Rates per 100,000)	ICD-9 and ICD-10 Codes	Only significantly reduced suicide rates in women between 1980 and 1989 after adjusting for confounders
Schrauzer and Shrestha. (1990)	USA	27	1978–1987	10,086,000	0–160 μg/L	NA	Completed Suicide per 100,000	No Formal Diagnostic Criteria	Suicide significantly reduced in areas of higher lithium concentrations
Schullehner et al. (2019)	Denmark	NA	1985–2013	374,737	0.6–30.7 μg/L	12.0 μg/L (6.4)	Diagnosis of Schizophrenia spectrum disorder, Schizophrenia and Bipolar disorder	ICD-8 and ICD-10 Codes	Significantly increase the incidence of Schizophrenia spectrum disorders and Schizophrenia. No association with Bipolar incidence rates
Shimodera et al. (2018)	Japan	24	2008–2009	3040	0.01–2.1 μg/L	0.48 μg/L (0.52)	Psychotic experiences	Diagnostic Interview Schedule for Children	Significant inverse relationship with psychotic experiences
Shiotsuki et al. (2016)	Japan	153	2010–2011	16,981,717	0.1–43 μg/L	3.8 μg/L (5.3)	Completed Suicide (Suicide Standardised Mortality Rates per 100,000)	No Formal Diagnostic Criteria	Trace lithium is inversely associated with male but not female suicide rates with higher lithium concentrations
Shiotsuki et al. (2008)	Japan	1	2008	43	6.1 and 15.7 μg/L	NA	Anxiety and Depression	State-Trait Anxiety Inventory, Profiles of Mood States Test	Suggests that drinking mineral water containing very low lithium levels may increase serum lithium levels and improve mental state
Sugawara et al. (2013)	Japan	40	2011–2012	1,373,339	0–12.9 μg/L	NA	Completed Suicide (Suicide Standardised Mortality Rates per 100,000)	No Formal Diagnostic Criteria	Only significantly reduced suicide rates in females in higher lithium concentrations
Voors (1972)	USA	100	1965–1970	5,000,000	0–850 μg/L	NA	State Mental Health Hospital Admissions for Neurosis, Psychosis or Personality diagnoses	No Formal Diagnostic Criteria	Only significantly reduced neurosis hospital admission rates with higher lithium concentrations

SD: standard deviation; GHQ-12: General Health Questionnaire;; NA: not applicable or did not state; ICD: International Classification of Disease; IQR: interquartile range; CES: Centre for Epidemiological Studies.

No specific ethics approval was necessary for this study. Patient consent was not required for our study.

Results

Searches and sample of studies

We identified 2305 articles. After removing 1003 as duplicates, we screened the titles and abstracts of the remaining 1302 and identified 37 articles that met the inclusion criteria. After the full-text review, 10 papers were excluded. In total, 27 articles were included in our systematic review (Ando et al., 2017; Blüml et al., 2013; Dawson et al., 1970, 1972; Giotakos et al., 2013; Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Kessing et al., 2017a, 2017b; Knudsen et al., 2017; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Oliver et al., 1976; Palmer et al., 2019; Parker et al., 2018; Pokorny et al., 1972; Pompili et al., 2015; Schrauzer and Shrestha, 1990; Schullehner et al., 2019; Shimodera et al., 2018; Shiotsuki et al., 2008, 2016; Sugawara et al., 2013; Voors, 1972), with 16 used in our meta-analysis (Blüml et al., 2013; Dawson et al., 1970; Giotakos et al., 2013; Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Palmer et al., 2019; Pompili et al., 2015; Schrauzer and Shrestha, 1990; Shiotsuki et al., 2016; Sugawara et al., 2013; Voors, 1972). A PRISMA flow diagram outlines this process (Figure 1), and a table of excluded studies after full-text review is provided as Supplementary Table 5.

Figure 1.

PRISMA flowchart for the association between lithium in drinking water and neuropsychiatric outcomes: a systematic review and meta-analysis from across 2678 regions containing 113 million people.

The characteristics of the 27 included studies are shown in Table 1. In all, 20 studies were longitudinal (Blüml et al., 2013; Dawson et al., 1970, 1972; Giotakos et al., 2013; Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Palmer et al., 2019; Parker et al., 2018; Pokorny et al., 1972; Pompili et al., 2015; Schrauzer and Shrestha, 1990; Schullehner et al., 2019; Shiotsuki et al., 2016; Sugawara et al., 2013; Voors, 1972), 3 were case–control studies (Kessing et al., 2017a, 2017b; Knudsen et al., 2017) and 4 were cross-sectional studies (Ando et al., 2017; Oliver et al., 1976; Shimodera et al., 2018; Shiotsuki et al., 2008). Of the longitudinal studies, 14 investigated suicide rates (Blüml et al., 2013; Giotakos et al., 2013; Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Palmer et al., 2019; Pompili et al., 2015; Schrauzer and Shrestha, 1990; Shiotsuki et al., 2016; Sugawara et al., 2013), 3 examined hospital admissions (Dawson et al., 1970; Pokorny et al., 1972; Voors, 1972) and 1 studied both suicide and hospital admissions (Dawson et al., 1972). One study investigated bipolar disorder, dementia and major depression (Parker et al., 2018). Another study examined bipolar disorder, schizophrenia and schizophrenia spectrum disorder (Schullehner et al., 2019). The three case–control studies investigated the rates of suicide (Knudsen et al., 2017), dementia (Kessing et al., 2017a) and bipolar disorder (Kessing et al., 2017b), respectively. In terms of cross-sectional studies, one investigated depressive symptoms and suicidal ideation (Ando et al., 2017), another investigated depressed mood (Oliver et al., 1976), one examined depression and anxiety symptoms (Shiotsuki et al., 2008) and the fourth psychotic experiences (Shimodera et al., 2018). The mean lithium concentration in the studies ranged from 0.48 to 27.4 μg/L, and the sample size from 43 to 17.2 million. The regions included in each study ranged from 1 to 276, with a mean of 99. Eleven studies came from Europe (Giotakos et al., 2013; Kabacs et al., 2011; Kapusta et al., 2011; Kessing et al., 2017a, 2017b; Knudsen et al., 2017; Liaugaudaite et al., 2017, 2019; Oliveira et al., 2019; Pompili et al., 2015; Schullehner et al., 2019), nine from the United States (Blüml et al., 2013; Dawson et al., 1970, 1972; Oliver et al., 1976; Palmer et al., 2019; Parker et al., 2018; Pokorny et al., 1972; Schrauzer and Shrestha, 1990; Voors, 1972) and seven studies from Japan (Ando et al., 2017; Ishii et al., 2015; Ohgami et al., 2009; Shimodera et al., 2018; Shiotsuki et al., 2008, 2016; Sugawara et al., 2013). Studies commonly adjusted for age, gender and population size, although these were inconsistently applied and varied widely. Less common confounders included population density, urbanicity, annual mean temperature and availability of mental health services.

The quality of the included studies is shown in Supplementary Table 6. Out of the included 27 studies, 15 studies were classified as being at low risk of bias (Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Kessing et al., 2017a, 2017b; Knudsen et al., 2017; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Pompili et al., 2015; Schullehner et al., 2019; Shiotsuki et al., 2016; Sugawara et al., 2013; Voors, 1972). Studies that scored higher on our scale were generally from Japan or Europe.

Systematic review

We were unable to combine data from the following studies quantitatively.

Knudsen et al. (2017) conducted a nested case–control study in Denmark investigating the incidence of suicide. Each Danish citizen’s lithium exposure was determined by concentrations in their residential drinking water and was subsequently followed for 22 years assessing for suicide incidence rates. The findings indicated no association between suicide and lithium concentrations on an individual level.

Kessing et al. (2017b) found that lithium concentrations in drinking water were not associated with the incidence of bipolar disorder in a nested case–control study. Similar findings were demonstrated by Schullehner et al. (2019) which examined a subpopulation of Kessing et al. (2017b) Danish population. Parker et al. (2018) also found no association between high lithium concentrations and the prevalence of bipolar disorder in the United States (RR = 0·95, 95% CI = [0.75, 1.21]).

Parker et al. (2018) further assessed the RR of dementia and lithium concentrations to find no correlation between the two (RR = 1.02, 95% CI = [0.85, 1.23]). However, Kessing et al. (2017a) found a non-linear association between lithium concentrations and dementia rates. In the Danish sample consisting over 70,000 individuals, they found that compared to those with the lowest lithium levels in drinking water (2.0–5.0 μg/L), dementia was more common among those exposed to lithium concentrations between 5.1 and 10.0 μg/L (IRR = 1.22, 95% CI = [1.19, 1.25]). The study found no difference among those with the lithium concentration between 10.1 and 15.0 μg/L (IRR = 0.98, 95% CI = [0.96, 1.01]), and dementia was significantly less common (IRR = 0.83, 95% CI = [0.81, 0.85]) in the highest lithium concentration group (15.1–27.0 μg/L) (Kessing et al., 2017a).

Parker et al. (2018) found no association between lithium concentrations and the prevalence of major depressive disorder (RR = 1.12, 95% CI = [0.96, 1.29]).

Among 3040 junior high school pupils aged between 12 and 15 in Japan, Ando et al. (2017) found that the increased lithium concentration was significantly correlated with reduced depressive symptoms as measured using the General Health Questionnaire (β = −0.23, standard error [SE] = 0.09, p = 0.02). However, there was no association between lithium concentration and suicidal ideation (β = −0.06, SE = 0.28, p = 0.82). Oliver et al. (1976) used the Centre for Epidemiologic Studies Depression scale and found an association between increased lithium concentration and increased rates of depressed mood in housewives and retired women (high lithium concentrations: 42.6%, low lithium concentrations: 10.7%, p < 0.01). Finally, Shiotsuki et al. (2008) measured the anxiety and depression scores using the State-Trait of Anxiety Inventory and the Profiles of Mood States Test of 43 participants before and after they ingested spring water rich in lithium. They found a significant improvement in depression (51.1 ± 7.4 to 48.8 ± 8.0, p < 0.02) and tension and anxiety scores (47.6 ± 7.2 to 45.3 ± 9.2, p < 0.01).

An initial paper from Japan found an inverse association between lithium levels in tap water and psychotic experiences in adolescents as measured by the Diagnostic Interview Schedule for Children (β = −0.37, p = 0.021) (Shimodera et al., 2018). By contrast, a Danish study (Schullehner et al., 2019) found that those exposed to the highest concentrations of lithium in drinking water were at greater risk of developing a schizophrenia spectrum disorder (IRR = 1.21, 95% CI = [1.07, 1.35]) or schizophrenia (IRR = 1.23, 95% CI = [1.05, 1.44]), compared to those exposed to the lowest concentrations of lithium. However, unlike the Japanese study, outcomes were based on clinical diagnoses rather than derived from a structured psychiatric interview.

Pokorny et al. (1972) re-analysed the findings from Dawson et al. (1970) and found that geographic proximity to hospitals was a better explanation for hospital admission rates rather than lithium concentrations. Conversely, Dawson et al. (1972) re-analysed his data from his previous paper (Dawson et al., 1970) in 1972 and found that higher lithium concentrations were associated with reduced mental health hospital admissions (r = −0.5118, p < 0.01) but suicide rates were not affected (r = −0.2354, p > 0.05). Both studies were excluded from the meta-analysis, as they re-analysed the same population existing in Dawson et al. (1970), and thus provided no additional information.

Meta-analysis

We were able to meta-analyse the following outcomes: (1) suicide rates and (2) hospital admission rates for our chosen ecological studies.

Figure 2 shows the forest plot of the association between lithium concentration and suicide rate. Fourteen longitudinal studies with over 94 million people across 1223 geographical regions were included in this analysis (Blüml et al., 2013; Giotakos et al., 2013; Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Palmer et al., 2019; Pompili et al., 2015; Schrauzer and Shrestha, 1990; Shiotsuki et al., 2016; Sugawara et al., 2013). In the primary analysis, higher lithium levels were associated with reduced incidence of suicide across the geographical regions (r = −0.191, 95% CI = [−0.287, −0.090], p < 0.001), with significant heterogeneity between studies (Q = 67.4, p < 0.001, I² = 80.7%). These results remained significant when only high-quality studies were included (r = −0.227, 95% CI = [−0.362, −0.082], p = 0.002; Q = 54.5, p < 0.001, I² = 81.7%) (Ishii et al., 2015; Kabacs et al., 2011; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Ohgami et al., 2009; Oliveira et al., 2019; Pompili et al., 2015; Shiotsuki et al., 2016; Sugawara et al., 2013), when Ohgami et al. (2009) was excluded (r = −0.168, 95% CI = [−0.263, −0.070], p = 0.001), when excluding Liaugaudaite et al. (2017) (r = −0.190, 95% CI = [−0.288, −0.088], p < 0.001), and when studies that did not provide adjusted results (Giotakos et al., 2013; Schrauzer and Shrestha, 1990) were excluded (r = −0.135, 95% CI = [−0.220, −0.048], p = 0.002). In the sensitivity analysis of each study removed in turn, the results remained significant throughout. There was also a significant indication of publication bias influencing the findings (Egger’s test; t value = 2.90, p = 0.013). Duval and Tweedie’s Trim and Fill analysis found no missing studies to left of the mean, with five studies trimmed to the right of mean with adjusted values (r = −0.080, 95% CI = [−0.188, 0.029]).

Figure 2.

Forest plot of the meta-analysis between lithium in drinking water and suicide rates in the most adjusted models for the association between lithium in drinking water and neuropsychiatric outcomes: a systematic review and meta-analysis from across 2678 regions containing 113 million people.

The funnel plot for publication bias with imputed missing studies is provided as Supplementary Figure 1. A meta-regression analysis of eight studies reporting mean lithium concentrations ranging between 3.8 and 11.5 μg/L found no significant association of mean lithium levels and suicide rates (Giotakos et al., 2013; Ishii et al., 2015; Kapusta et al., 2011; Liaugaudaite et al., 2017, 2019; Oliveira et al., 2019; Pompili et al., 2015; Shiotsuki et al., 2016).

Two studies of over 5 million people were included in the meta-analysis of lithium levels and psychiatric hospitalisations (Dawson et al., 1970; Voors, 1972). We found that increased lithium concentration was associated with reduced hospital admissions (r = −0.413, 95% CI = [−0.689, −0.031], p = 0.035) (Figure 3). Sub-analyses showed only admissions for neurotic disorders were significantly reduced (r = −0.526, 95% CI = [−0.643, −0.385], p < 0.001) (Supplementary Figure 2), while admissions for personality (r = −0.358, 95% CI = [−0.754, 0.229], p = 0.227) (Supplementary Figure 3) and psychotic outcomes (r = −0.327, 95% CI = [−0.734, 0.253], p = 0.266) (Supplementary Figure 4) were both non-significant.

Figure 3.

Forest plot of the meta-analysis between lithium in drinking water and hospital admissions for the association between lithium in drinking water and neuropsychiatric outcomes: a systematic review and meta-analysis from across 2678 regions containing 113 million people.

Discussion

Findings

In our meta-analysis, we found that higher lithium concentrations were associated with reduced suicide rates and hospital admissions. This inverse relationship with suicide rates and hospital admissions was seen in seven studies and two studies, respectively. There was no observable dose–response relationship between lithium concentrations and suicide rates. We also found in our systematic review that while some studies suggested improvement in depression and anxiety symptoms scales, there was mixed evidence for dementia and psychosis rates and no association between lithium concentration and the rates of bipolar disorder or major depressive disorder.

Our main findings are mostly consistent with previously published studies that investigated sources of lithium in the environment other than drinking water. For example, König et al. (2017) demonstrated lower suicide rates in regions where higher concentrations of lithium were found in Chilean soil. In a randomised controlled trial, microdoses of lithium were shown to have positive mood effects when provided as an oral supplement (Schrauzer and de Vroey, 1994). Furthermore, higher lithium levels in drinking water significantly reduced the prevalence of depressive temperament in Japan (Ishii et al., 2018). A similar systematic review and meta-analyses found lithium in drinking water were dose-dependently associated with reduced suicide mortality (Barjasteh-Askari et al., 2020). However, this review was limited by the fact that the authors used unadjusted effect sizes and did not explore outcomes beyond suicide.

Our findings are also consistent with those relating lithium macro-doses with suicide. In a recent meta-analysis, Cipriani et al. (2013) found that using long-term lithium therapy in mood disorders significantly reduced the risk of suicide. These findings were supported by more recent studies (Lewitzka et al., 2015; Smith and Cipriani, 2017).

While its exact anti-suicidal mechanism of action remains uncertain, lithium’s effect on the brain appears to be multi-faceted. These include reducing impulsive and aggressive behaviours (Müller-Oerlinghausen and Lewitzka, 2016), increasing glutamine synthetase expression (Kalkman, 2011) and improved decision-making in those with mood disorders (Adida et al., 2015). Other effects include increasing the availability of vitamin B12 which may improve mood (Schrauzer et al., 1992) and reducing testosterone levels which may reduce suicidality in men (Sher, 2015).

Lithium’s neurobiology has also been a topic of intense research given its ability to affect multiple cellular pathways and expression of hundreds of genes (Bauer and Gitlin, 2016; Malhi et al., 2012; Rybakowski, 2020). Lithium-induced inhibition of glycogen synthase kinase 3β restores neurotransmission and corrects cellular dysfunction, which may underpin lithium’s mood stabilising, suicide-preventing and neurocognitive properties (Beurel and Jope, 2014; Malhi and Outhred, 2016; Malhi et al., 2012, 2018a). In addition, lithium’s effect on inositol monophosphatase influences neuroprotection by enhancing cell survival (Bauer and Gitlin, 2016). Lithium also stimulates neurotrophic factors, brain-derived neurotrophic factor and vascular endothelial growth factor (Forlenza et al., 2014) and has been found to influence the neuropathology of dementia (Bauer and Gitlin, 2016).

Lithium appears to modulate genes involved in neuroprotection, neuronal plasticity and other neuronal functions including neurotrophy (Malhi et al., 2012). This was further shown in a recent study where magnetic resonance imaging (MRI) scans showed increased grey matter, putamen, hippocampal and thalamic volumes in patients taking lithium as a consequence of lithium-induced gene expression (Anand et al., 2020). Further neuroprotection is evidenced as lithium prevents stress-induced dendritic damage (Wood et al., 2004) and shows anti-apoptotic activity (Bachmann et al., 2009; Chen and Chuang, 1999). Subtherapeutic lithium levels demonstrate anti-inflammatory properties by modulating the secretion of pro-inflammatory and anti-inflammatory cytokines in hippocampal and cortical neurons (De-Paula et al., 2016). The above-described neuroprotective properties may be apparent in those with and without mood disorders (Bauer and Gitlin, 2016), possibly at subtherapeutic levels (Vo et al., 2015).

Combining these multi-faceted mechanisms with a study showing animal models can accumulate lithium intracellularly (Foulks et al., 1952), it is plausible that long-term intake of subtherapeutic doses of lithium may reduce suicide and have additional neuropsychiatric benefits.

We further postulate that the anti-suicidal effect of lithium may not be as closely dose-dependent as its mood stabilising effects. For example, it has been previously discussed that lithium may demonstrate an anti-suicidal effect without having mood stabilising effects (Terao et al., 2009). A significant reduction in suicidal attempts has been noted among patients with affective disorders on lithium even if they do not have improvement in mood symptoms (Ahrens and Müller-Oerlinghausen, 2001).

Strengths and limitations

To our knowledge, this is the first systematic review and meta-analysis to examine the association between lithium concentrations in drinking water and rates of multiple neuropsychiatric outcomes. Our review included over 113 million subjects from 27 studies in nine different countries. That said, some important limitations need to be mentioned.

First, we emphasise the presence of publication bias within the meta-analysis and have tried to address this as follows. We investigated its presence using Egger’s test, Duval and Tweedie’s Trim and Fill analysis. We further examined the funnel plot for visual inspection.

In the absence of mandatory pre-registration, epidemiological studies such as those included here may be vulnerable to the ‘file-drawer problem’, thus increasing the possibility of publication bias. This, however, may occur both ways. A finding of no association between differences in drinking water lithium concentrations and mental health outcomes may not be considered sufficiently novel by researchers or journal editors to warrant publication. Conversely, journals may favour publishing papers with null findings, as some viewers may find lithium’s purported effects at such low concentrations to be implausible.

Future studies using pre-registered analyses are recommended to further explore the association between lithium and reduced suicide rates. Further analysis of publication bias present in this topic of literature would also be useful.

The formula for deriving r from the standardised coefficient was based on a previously published formula where validation was based on a range of β values between −0.5 and 0.5 where most of the data points were. Our results for values outside that range should be treated with caution. However, the results remained significant in a sensitivity analysis of the effects of only including β values between −0.5 and 0.5.

There was significant heterogeneity between studies, and as such these results should be treated with caution in both the size and the direction. We have attempted to address this using random effects models throughout, as well as undertaking sensitivity and subgroup analyses, as well as meta-regression.

Our results must also consider the ecological fallacy, being correlations found at a population-based level cannot be assumed to have the same effect on an individual-based level (Robinson, 1950). In the context of our findings, while the increased lithium concentration in drinking water is associated with the reduced risk of suicide or hospital admission at a population level, it is unknown whether any given particular individual’s susceptibility for suicide or hospital admission is improved.

In addition, many included studies also suffered from the lack of ascertainment of neuropsychiatric outcomes and low quality of descriptive statistics reported as evidenced in the quality appraisal. For example, the two studies that examined hospital admission rates were both from the 1970s and so may be less applicable to current practice or the availability of beds (Dawson et al., 1970; Voors, 1972). Even in the 1970s, hospital accessibility was suggested as a more plausible explanation than the effect of lithium concentration (Pokorny et al., 1972). Further contemporary research is required to better establish our hospital admissions results given only two older studies could be meta-analysed.

It should also be noted that suicide is the product of multiple factors including genetic, biological, psychosocial and environmental (Hawton and van Heeringen, 2009). While some of these potential confounders have been adjusted for in the included studies, they were not consistent between studies as was the methods of ascertaining lithium-level concentrations and the measurements used to report suicide and lithium levels. Despite this, we believe our methodology is robust enough to allow us to meta-analyse some of the findings.

Supplementation of drinking water with lithium may offer a cheap and effective opportunity to reduce the distressing symptoms and costly burden of neuropsychiatric outcomes. Although our findings suggest potential benefit, especially in suicide and hospital admissions, further research is needed to confirm these results. Safety of utilisation is of particular uncertainty, considering excessive lithium concentrations in drinking water may be associated with a number of physical complications such as compromised thyroid function (Broberg et al., 2011; Harari et al., 2015a), disruption of vitamin D and calcium homeostasis in pregnant women (Harari et al., 2016) and impaired birth size (Harari et al., 2015b). High concentrations in bottled water could also cause lithium toxicity in those already prescribed lithium (Dabeka et al., 2002). To better discern the net benefits of lithium in drinking water, future research should also focus on the precise concentration of lithium that provides the most significant neuropsychiatric benefit while demonstrating no substantial harms at the population and individual level.

Given the recent controversy around fluoridation (Green et al., 2019), there may also be broader clinical, societal and philosophical objections to adding a further element to the water supply. There may also be greater resistance to a measure to promote mental as opposed to dental health. Adding lithium to bottled water would allow greater individual choice but equally disadvantage those on low incomes. The economic and environmental effects should also be considered. For instance, there is laboratory evidence for the harmful effect of lithium on marine life (Shahzad et al., 2017), but data from the environment are more limited and generally confined to areas of high industrial pollution (Mathis and Cummings, 1973; Shahzad et al., 2017).

Conclusion

Our findings suggest that higher concentrations of lithium in drinking water may be associated with reduced suicide rates and psychiatric hospital admission. The effect on other neuropsychiatric outcomes is less clear. Further research is necessary on the risks and benefits of lithium in the water on neuropsychiatric and physical health outcomes before any public health recommendations could be made.

Supplemental Material

20200825_Lithium_Supplementary_tables_and_figures.docx – Supplemental material for The association between lithium in drinking water and neuropsychiatric outcomes: A systematic review and meta-analysis from across 2678 regions containing 113 million

Supplemental material, 20200825_Lithium_Supplementary_tables_and_figures.docx for The association between lithium in drinking water and neuropsychiatric outcomes: A systematic review and meta-analysis from across 2678 regions containing 113 million by Brenton Eyre-Watt, Eesharnan Mahendran, Shuichi Suetani, Joseph Firth, Steve Kisely and Dan Siskind in Australian & New Zealand Journal of Psychiatry

Footnotes

Author Contributions

B.E.-W., E.M., S.S., J.F., S.K. and D.S. designed the study. B.E.-W. and E.M. extracted all data. B.E.-W., D.S., S.K. and J.F. analysed the data. B.E.-W., E.M. and S.S. wrote the initial draft. All authors participated in the revision of the text and approved the final version of this paper. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship and/or publication of this article: All authors have completed the ICMJE uniform disclosure form at and declare no support from any organisation for the submitted work; J.F. is supported by a Blackmores Institute Fellowship; no other relationships or activities that could appear to have influenced the submitted work.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iDs

Brenton Eyre-Watt

Shuichi Suetani

Steve Kisely

Dan Siskind

Data Sharing Statement

Relevant study data can be found in the online supplementary files. Further statistical data are available on request to the corresponding author with publication, indefinitely for anyone that wishes to access the data, for any purpose.

Supplemental Material

Supplemental material for this article is available online.

References

Adida

Jollant

Clark

, et al. (2015) Lithium might be associated with better decision-making performance in euthymic bipolar patients. European Neuropsychopharmacology 25: 788–797.

Ahrens

Müller-Oerlinghausen

(2001) Does lithium exert an independent antisuicidal effect? Pharmacopsychiatry 34: 132–136.

Anand

Nakamura

Spielberg

, et al. (2020) Integrative analysis of lithium treatment associated effects on brain structure and peripheral gene expression reveals novel molecular insights into mechanism of action. Transl Psychiatry 10: 103.

Ando

Koike

Shimodera

, et al. (2017) Lithium levels in tap water and the mental health problems of adolescents: An individual-level cross-sectional survey. Journal of Clinical Psychiatry 78: E252–E256.

Bachmann

Wang

Yuan

, et al. (2009) Common effects of lithium and valproate on mitochondrial functions: Protection against methamphetamine-induced mitochondrial damage. International Journal of Neuropsychopharmacology 12: 805–822.

Barjasteh-Askari

Davoudi

Amini

, et al. (2020) Relationship between suicide mortality and lithium in drinking water: A systematic review and meta-analysis. Journal of Affective Disorders 264: 234–241.

Bauer

Gitlin

(2016) The Essential Guide to Lithium Treatment. Cham: Springer.

Beurel

Jope

(2014) Inflammation and lithium: Clues to mechanisms contributing to suicide-linked traits. Translational Psychiatry 4: e488.

Blüml

Regier

Hlavin

, et al. (2013) Lithium in the public water supply and suicide mortality in Texas. Journal of Psychiatric Research 47: 407–411.

10.

Broberg

Concha

Engström

, et al. (2011) Lithium in drinking water and thyroid function. Environ Health Perspect 119: 827–830.

11.

Cade

(2000) Lithium salts in the treatment of psychotic excitement: 1949. Medical Journal of Australia 78: 518–520.

12.

Chen

R-W

Chuang

D-M

(1999) Long term lithium treatment suppresses p53 and Bax expression but increases Bcl-2 expression A prominent role in neuroprotection against excitotoxicity. Journal of Biological Chemistry 274: 6039–6042.

13.

Cipriani

Hawton

Stockton

, et al. (2013) Lithium in the prevention of suicide in mood disorders: Updated systematic review and meta-analysis. BMJ 346: f3646.

14.

Dabeka

Conacher

HBS

Lawrence

, et al. (2002) Survey of bottled drinking waters sold in Canada for chlorate, bromide, bromate, lead, cadmium and other trace elements. Food Additives & Contaminants 19: 721–732.

15.

Dawson

Moore

McGanity

(1970) The mathematical relationship of drinking water lithium and rainfall to mental hospital admission. Diseases of the Nervous System 31: 811–820.

16.

Dawson

Moore

McGanity

(1972) Relationship of lithium metabolism to mental hospital admission and homicide. Diseases of the Nervous System 33: 546–556.

17.

De-Paula

Kerr

Scola

, et al. (2016) Lithium distinctly modulates the secretion of pro- and anti-inflammatory interleukins in co-cultures of neurons and glial cells at therapeutic and sub-therapeutic concentrations. Current Alzheimer Research 13: 848–852.

18.

Dolara

(2014) Occurrence, exposure, effects, recommended intake and possible dietary use of selected trace compounds (aluminium, bismuth, cobalt, gold, lithium, nickel, silver). International Journal of Food Sciences and Nutrition 65: 911–924.

19.

Duval

Tweedie

(2000) Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56: 455–463.

20.

Egger

Smith

Schneider

, et al. (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629.

21.

Forlenza

De-Paula

VdJR

Diniz

(2014) Neuroprotective effects of lithium: Implications for the treatment of Alzheimer’s disease and related neurodegenerative disorders. ACS Chemical Neuroscience 5: 443–450.

22.

Foulks

Mudge

Gilman

(1952) Renal excretion of cation in the dog during infusion of isotonic solutions of lithium chloride. American Journal of Physiology 168: 642–649.

23.

Giotakos

Nisianakis

Tsouvelas

, et al. (2013) Lithium in the public water supply and suicide mortality in Greece. Biological Trace Element Research 156: 376–379.

24.

Green

Lanphear

Hornung

, et al. (2019) Association between maternal fluoride exposure during pregnancy and IQ scores in offspring in Canada. JAMA Pediatrics 173: 940–948.

25.

Harari

Åkesson

Casimiro

, et al. (2016) Exposure to lithium through drinking water and calcium homeostasis during pregnancy: A longitudinal study. Environmental Research 147: 1–7.

26.

Harari

Bottai

Casimiro

, et al. (2015a) Exposure to lithium and cesium through drinking water and thyroid function during pregnancy: A prospective cohort study. Thyroid 25: 1199–1208.

27.

Harari

Langeen

Casimiro

, et al. (2015b) Environmental exposure to lithium during pregnancy and fetal size: A longitudinal study in the Argentinean Andes. Environment International 77: 48–54.

28.

Hawton

van Heeringen

(2009) Suicide. The Lancet 373: 1372–1381.

29.

Ishii

Terao

Araki

, et al. (2015) Low risk of male suicide and lithium in drinking water. Journal of Clinical Psychiatry 76: 319–326.

30.

Ishii

Terao

Matsuzaki

, et al. (2018) Lithium in drinking water may be negatively associated with depressive temperament in the nonclinical population. Clinical Neuropsychopharmacology and Therapeutics 8: 7–11.

31.

Kabacs

Memon

Obinwa

, et al. (2011) Lithium in drinking water and suicide rates across the east of England. British Journal of Psychiatry 198: 406–407.

32.

Kalkman

(2011) Circumstantial evidence for a role of glutamine-synthetase in suicide. Medical Hypotheses 76: 905–907.

33.

Kapusta

Mossaheb

Etzersdorfer

, et al. (2011) Lithium in drinking water and suicide mortality. British Journal of Psychiatry 198: 346–350.

34.

Kessing

Gerds

Knudsen

, et al. (2017a) Association of lithium in drinking water with the incidence of dementia. JAMA Psychiatry 74: 1005–1010.

35.

Kessing

Gerds

Knudsen

, et al. (2017b) Lithium in drinking water and the incidence of bipolar disorder: A nation-wide population-based study. Bipolar Disorder 19: 563–567.

36.

Knudsen

Schullehner

Hansen

, et al. (2017) Lithium in drinking water and incidence of suicide: A nationwide individual-level cohort study with 22 years of follow-up. International Journal of Environmental Research and Public Health 14: 627.

37.

König

Baumgartner

Blüml

, et al. (2017) Impact of natural lithium ressources on suicide mortality in Chile 2000–2009: A geographical analysis. Neuropsychiatrie 31: 70–76.

38.

Lauterbach

Mendez

(2011) Psychopharmacological neuroprotection in neurodegenerative diseases, part III: Criteria-based assessment: A report of the ANPA committee on research. The Journal of Neuropsychiatry and Clinical Neurosciences 23: 242–260.

39.

Lewitzka

Severus

Bauer

, et al. (2015) The suicide prevention effect of lithium: More than 20 years of evidence: A narrative review. International Journal of Bipolar Disorders 3: 15.

40.

Liaugaudaite

Mickuviene

Raskauskiene

, et al. (2017) Lithium levels in the public drinking water supply and risk of suicide: A pilot study. Journal of Trace Elements in Medicine and Biology 43: 197–201.

41.

Liaugaudaite

Naginiene

Raskauskiene

, et al. (2019) Relationship between lithium levels in drinking water and suicide rates: A nationwide study in Lithuania. Archives of Suicide Research. Epub ahead of print 9 October. DOI: 10.1080/13811118.2019.1674226.

42.

Lipsey

Wilson

(2001) Practical Meta-Analysis. Thousand Oaks, CA: SAGE.

43.

Malhi

(2018) Lithium: A global perspective. Pharmacopsychiatry 51: 220–221.

44.

Malhi

Outhred

(2016) Therapeutic mechanisms of lithium in bipolar disorder: Recent advances and current understanding. CNS Drugs 30: 931–949.

45.

Malhi

Bassett

Boyce

, et al. (2015) Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for mood disorders. Australian and New Zealand Journal of Psychiatry 49: 1087–1206.

46.

Malhi

Bell

Outhred

, et al. (2020) Lithium therapy and its interactions. Australian Prescriber 43: 91–93.

47.

Malhi

Das

Outhred

, et al. (2018a) Understanding suicide: Focusing on its mechanisms through a lithium lens. Journal of Affective Disorders 241: 338–347.

48.

Malhi

Gessler

Outhred

(2017) The use of lithium for the treatment of bipolar disorder: Recommendations from clinical practice guidelines. Journal of Affective Disorders 217: 266–280.

49.

Malhi

Outhred

Morris

, et al. (2018b) Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for mood disorders: Bipolar disorder summary. Medical Journal of Australia 208: 219–225.

50.

Malhi

Tanious

Das

, et al. (2012) The science and practice of lithium therapy. Australian and New Zealand Journal of Psychiatry 46: 192–211.

51.

Malhi

Tanious

Das

, et al. (2013) Potential mechanisms of action of lithium in bipolar disorder. CNS Drugs 27: 135–153.

52.

Mata

Ramos

Bansal

, et al. (2015) Prevalence of depression and depressive symptoms among resident physicians: A systematic review and meta-analysis. JAMA 314: 2373–2383.

53.

Mathis

Cummings

(1973) Selected metals in sediments, water, and biota in the Illinois River. Journal of the Water Pollution Control Federation 45: 1573–1583.

54.

Moher

Liberati

Tetzlaff

, et al. (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Journal of Clinical Epidemiology 62: 1006–1012.

55.

Müller-Oerlinghausen

Lewitzka

(2016) The contributions of lithium and clozapine for the prophylaxis and treatment of suicidal behavior. Biological Aspects of Suicidal Behavior 30: 145–160.

56.

Nolen

Licht

Young

, et al. (2019) What is the optimal serum level for lithium in the maintenance treatment of bipolar disorder? A systematic review and recommendations from the ISBD/IGSLI Task Force on treatment with lithium. Bipolar Disorder 21: 394–409.

57.

Ohgami

Terao

Shiotsuki

, et al. (2009) Lithium levels in drinking water and risk of suicide. British Journal of Psychiatry 194: 464–465.

58.

Oliveira

Zagalo

Madeira

, et al. (2019) Lithium in public drinking water and suicide mortality in Portugal: Initial approach. Acta Médica Portuguesa 32: 47–52.

59.

Oliver

Comstock

Helsing

(1976) Mood and lithium in drinking-water. Archives of Environmental & Occupational Health 31: 92–95.

60.

Palmer

Cates

Gorman

(2019) The association between lithium in drinking water and incidence of suicide across 15 Alabama counties. Crisis 40: 93–99.

61.

Parker

Gorges

Gao

, et al. (2018) Association between groundwater lithium and the diagnosis of bipolar disorder and dementia in the United States. JAMA Psychiatry 75: 751–754.

62.

Peterson

Brown

(2005) On the use of beta coefficients in meta-analysis. Journal of Applied Psychology 90: 175–181.

63.

Pokorny

Sheehan

Atkinson

(1972) Drinking-water lithium and mental-hospital admissions. Diseases of the Nervous System 33: 649–652.

64.

Pompili

Vichi

Dinelli

, et al. (2015) Relationships of local lithium concentrations in drinking water to regional suicide rates in Italy. World Journal of Biological Psychiatry 16: 567–574.

65.

Robinson

(1950) Ecological correlations and the behavior of individuals. American Sociological Review 15: 351–357.

66.

Rybakowski

(2020) Mood Stabilizers: Lithium. In: Riederer

Laux

Mulsant

Nagatsu

(eds) NeuroPsychopharmacotherapy. Cham: Springer. https://doi.org/10.1007/978-3-319-56015-1_45-1

67.

Schrauzer

de Vroey

(1994) Effects of nutritional lithium supplementation on mood: A placebo-controlled study with former drug users. Biological Trace Element Research 40: 89–101.

68.

Schrauzer

Shrestha

(1990) Lithium in drinking water and the incidences of crimes, suicides, and arrests related to drug addictions. Biological Trace Element Research 25: 105–113.

69.

Schrauzer

Shrestha

Floresarce

(1992) Lithium in scalp hair of adults, students, and violent criminals: Effects of supplementation and evidence for interactions of lithium with vitamin-B12 and with other trace-elements. Biological Trace Element Research 34: 161–176.

70.

Schullehner

Paksarian

Hansen

, et al. (2019) Lithium in drinking water associated with adverse mental health effects. Schizophrenia Research 210: 313–315.

71.

Shahzad

Mughal

Tanveer

, et al. (2017) Is lithium biologically an important or toxic element to living organisms? An overview. Environmental Science and Pollution Research 24: 103–115.

72.

Sher

(2015) Suicide in men. Journal of Clinical Psychiatry 76: e371–e372.

73.

Shimodera

Koike

Ando

, et al. (2018) Lithium levels in tap water and psychotic experiences in a general population of adolescents. Schizophrenia Research 201: 294–298.

74.

Shiotsuki

Terao

Ishii

, et al. (2016) Trace lithium is inversely associated with male suicide after adjustment of climatic factors. Journal of Affective Disorders 189: 282–286.

75.

Shiotsuki

Terao

Ogami

, et al. (2008) Drinking spring water and lithium absorption: A preliminary study. German Journal of Psychiatry 11: 103–106.

76.

Smith

Cipriani

(2017) Lithium and suicide in mood disorders: Updated meta-review of the scientific literature. Bipolar Disorder 19: 575–586.

77.

Sugawara

Yasui-Furukori

Ishii

, et al. (2013) Lithium in tap water and suicide mortality in Japan. International Journal of Environmental Research and Public Health 10: 6044–6048.

78.

Terao

Goto

Inagaki

, et al. (2009) Even very low but sustained lithium intake can prevent suicide in the general population? Medical Hypotheses 73: 811–812.

79.

Tondo

Alda

Bauer

, et al. (2019) Clinical use of lithium salts: Guide for users and prescribers. International Journal of Bipolar Disorders 7: 16.

80.

Perry

Ellerby

, et al. (2015) Is lithium a neuroprotective agent. Annals of Clinical Psychiatry 27: 49–54.

81.

Voors

(1972) Drinking-water lithium and mental hospital admission in North Carolina. North Carolina Medical Journal 33: 597–602.

82.

Wood

Young

Reagan

, et al. (2004) Stress-induced structural remodeling in hippocampus: Prevention by lithium treatment. Proceedings of the National Academy of Sciences of the United States of America 101: 3973–3978.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.47 MB

The association between lithium in drinking water and neuropsychiatric outcomes: A systematic review and meta-analysis from across 2678 regions containing 113 million people

Abstract

Background:

Methods:

Results:

Conclusion:

Keywords

Introduction

Method

Search strategy and selection criteria

Data analysis

Results

Searches and sample of studies

Systematic review

Meta-analysis

Discussion

Findings

Strengths and limitations

Conclusion

Supplemental Material

20200825_Lithium_Supplementary_tables_and_figures.docx – Supplemental material for The association between lithium in drinking water and neuropsychiatric outcomes: A systematic review and meta-analysis from across 2678 regions containing 113 million

Footnotes

Author Contributions

Declaration of Conflicting Interests

Funding

ORCID iDs

Data Sharing Statement

Supplemental Material

References

Supplementary Material