Skip to main content

Table 1 Meta-analytic findings of studies of lithium versus risk of suicides or attempts: summary

From: Prevention of suicidal behavior with lithium treatment in patients with recurrent mood disorders

Conditions

Studies

(n)

Risk Ratio (RR) [95% CI]

p-value

All studies

35

3.77 [2.72–5.24]

 < 0.0001

Outcomes

 Suicides

15

4.03 [2.37–6.85]

 < 0.0001

 Attempts

8

3.73 [1.50–9.33]

 < 0.005

 Both

12

3.66 [2.28–5.87]

 < 0.0001

Diagnoses

 Bipolar

14

3.84 [2.34–6.30]

 < 0.0001

 Major affective

21

3.72 [2.40–5.77]

 < 0.0001

 Major depression

8

4.69 [1.66–13.3]

 < 0.0001

Trial types

 Open

29

3.95 [2.79–5.58]

 < 0.0001

 Randomized, placebo-controlled

6

1.91 [0.89–4.08]

0.096

Comparator type

 vs. placebo or no lithium

27

4.22 [2.91–6.10]

 < 0.0001

 vs. anticonvulsants

8

2.42 [1.15–5.10]

0.02

  1. Analyses based on random-effects modeling, with stratification. Data from Baldessarini et al. (2006), Guzzetta et al. (2007), Collins and McFarland (2008), Lauterbach et al. (2008), Smith et al. (2009), and Oquendo et al. (2011). RR is for differences between lithium vs. comparators (values > 1 favor lithium)