The Case for Starting Lithium Early in Bipolar Disorder

For educational purposes only. This post is not medical advice and is not intended to guide treatment decisions. Please speak with your medical provider about your own care.

By the time most patients with bipolar disorder are offered lithium, they have already been unwell for a long time. A recent case-register analysis put hard numbers on the delay: a median of 659 days between first assessment and lithium initiation, with patients having passed through roughly 2.5 mood episodes and two antipsychotic trials on the way. [10] That is nearly two years of accumulated illness before a first-line maintenance agent enters the picture.

Whether that delay matters is one of the more interesting open questions in mood disorder treatment — and the evidence, while not definitive, points fairly consistently in one direction.

What the timing data actually show

The strongest direct evidence comes from a nationwide Danish register study of 4,714 patients. [1] Patients who started lithium early — at the point of first psychiatric contact, or after a single manic or mixed episode — had significantly lower rates of non-response than those who started later.

Two comparisons carried the finding:

• Initiation at first psychiatric contact versus later contact: HR 0.87 (95% CI 0.76–0.91)

• Initiation after a single episode versus after established bipolar disorder: HR 0.75 (95% CI 0.67–0.84)

The effect sizes are modest but the direction is consistent, and the sample is large and population-based rather than a selected trial cohort.

Why earlier might genuinely be better

Three separate lines of reasoning converge on the same conclusion, which is part of why the observational signal is taken seriously.

Episodes appear to degrade treatment response. A higher number of prior mood episodes predicts non-response — and not just to lithium, but across pharmacologic treatments generally. If each episode makes the next treatment attempt harder, there is a clear mechanistic argument for intervening before they accumulate. [2]

Lithium may be doing something structural. The drug has well-documented neuroprotective properties, raising the possibility that starting it before cognitive disturbances emerge preserves function that would otherwise be lost. [2] This shifts the framing from symptom suppression to something closer to prevention.

Bipolar disorder increasingly looks like a progressive condition. The field has moved toward conceptualizing it as chronic and potentially worsened by each episode — structurally similar to conditions where disease-modifying therapy is standard and started as early as possible. Lithium is notable for acting on many of the pathophysiologic mechanisms and surrogate markers implicated in that progression, and long-term outcomes are better with lithium-based regimens than without. If the disease-modifying frame is right, deploying the most disease-modifying agent late in the course is exactly the wrong sequencing. [3]

Supporting trial signals

No randomized trial has tested early versus late initiation directly. But two trials in early-course patients are worth knowing.

A 2017 Australian randomized trial in first-episode mania compared lithium monotherapy with quetiapine. Lithium came out ahead on secondary outcomes spanning depression, quality of life, functioning, and clinical global impression — a meaningful result in precisely the population where the early-initiation question is live. [2]

Separately, a randomized trial of early specialized treatment following first hospitalization for mania reduced re-hospitalization risk, improved medication adherence, and increased satisfaction with care compared with standard treatment. Effects were most pronounced in younger patients. [2] The 16-year follow-up of the Early Intervention in Affective Disorders trial found the intervention group maintained a lower absolute readmission risk across the entire follow-up period — though with a small sample, the separation did not reach statistical significance at the endpoint. A sustained but underpowered separation is not proof, but it is not nothing either. [4]

Where guidelines land

Lithium remains a first-line maintenance agent across international guidelines, with network meta-analysis showing a relapse risk ratio of 0.62 versus placebo and protection against both poles of the illness. [5][8][9] Multiple reviews go further and recommend that maintenance treatment begin shortly after illness onset, integrating pharmacologic, psychological, and lifestyle components from the first episode forward. [5][6][7]

The prescribing data tell a different story. Lithium use has been declining, and the 659-day median delay noted above suggests the gap between guideline position and practice is substantial. [8][10]

The caveats that matter

The timing evidence is observational. No double-blind or controlled trial has tested early versus late lithium on primary clinical outcomes in first-episode mania. [2] This is the central knowledge gap, and it is not a small one.

Confounding by presentation is plausible. Patients who come to attention early and are started on lithium early may differ systematically from those who present later — in illness severity, insight, social support, or the clarity of the initial diagnostic picture. Any of these could drive better outcomes independent of drug timing. The register finding is consistent with a causal effect of early initiation; it does not establish one.

Monitoring obligations are real. Long-term lithium carries risks of reduced renal function, hypothyroidism (13.9% versus 1.7%; OR 5.78), and hyperparathyroidism with hypercalcemia. Baseline and ongoing renal, thyroid, and calcium monitoring are required. Target maintenance serum levels typically sit at 0.6–0.8 mEq/L. [8][9][11] Starting earlier means monitoring longer, which is a genuine consideration in patients who may be years away from a stable care relationship.

Some presentations respond poorly. Lithium is less effective in dysphoric or mixed mania, rapid-cycling course, prominent psychosis, and comorbid alcohol use. [11] Early initiation is not a case for universal initiation.

The practical takeaway

The evidence for early lithium is good enough to change the default and not good enough to settle the question. What it argues against is the current pattern — cycling through multiple antipsychotics across multiple episodes before lithium is considered at all. That sequence is hard to justify on the data, and the register findings suggest it may cost something durable in terms of eventual response.

What would resolve this is a randomized trial of initiation timing in first-episode mania. Until one exists, the reasonable position is to treat lithium as a candidate at first presentation rather than a fallback after failure — while being honest with patients that the timing argument rests on observational evidence, and being clear-eyed about which presentations make lithium a poor fit.

Disclaimer: This article is provided for educational and informational purposes only. It is not intended as medical advice, diagnosis, or treatment, and should not be used as a substitute for consultation with a qualified healthcare professional. Lithium requires individualized dosing and ongoing laboratory monitoring. Do not start, stop, or change any medication without speaking with your medical provider.

References

1. Kessing LV, Vradi E, Andersen PK. Starting lithium prophylaxis early v. late in bipolar disorder. The British Journal of Psychiatry. 2014.

2. Bauer M, Andreassen OA, Geddes JR, et al. Areas of uncertainties and unmet needs in bipolar disorders: clinical and research perspectives. The Lancet Psychiatry. 2018.

3. Post RM, Li VW, Berk M, Yatham LN. Lithium as a disease-modifying drug for bipolar disorder. The Lancet Psychiatry. 2025.

4. Munkholm K, Kessing LV. Early specialised treatment for bipolar disorder: long-term follow-up from the Early Intervention in Affective Disorders (EIA) randomised controlled trial. Acta Psychiatrica Scandinavica. 2024.

5. Carvalho AF, Firth J, Vieta E. Bipolar disorder. The New England Journal of Medicine. 2020.

6. Grande I, Berk M, Birmaher B, Vieta E. Bipolar disorder. The Lancet. 2016.

7. Vieta E, Berk M, Schulze TG, et al. Bipolar disorders. Nature Reviews Disease Primers. 2018.

8. Nierenberg AA, Agustini B, Köhler-Forsberg O, et al. Diagnosis and treatment of bipolar disorder: a review. JAMA. 2023.

9. Malhi GS, Gessler D, Outhred T. The use of lithium for the treatment of bipolar disorder: recommendations from clinical practice guidelines. Journal of Affective Disorders. 2017.

10. Perez-Rodriguez V, Nguyen T, Beck K, et al. Time to lithium: a case register study of lithium initiation in bipolar disorder. Journal of Psychopharmacology. 2026.

11. Baldessarini RJ, Tondo L, Vázquez GH. Pharmacological treatment of adult bipolar disorder. Molecular Psychiatry. 2019.