Bipolar disorder affects millions of people worldwide, with estimates suggesting that between one and three in every hundred individuals will experience the condition at some point in their lives.

In Malaysia, the prevalence of bipolar disorder is estimated to be between 0.4% and 0.6% of the population, which, at the lower end, translates to over 130,000 people affected.

Despite this relatively high prevalence, its treatment remains complex, demanding careful balancing of medications, monitoring of side effects, and long-term engagement with mental health services.

Characterised by intense shifts in mood, energy and behaviour, bipolar disorder can pull individuals between periods of elevated mood, high activity and impulsive decision-making, and phases of deep depression marked by fatigue, low motivation and a loss of interest in daily life.

For decades, one medication has stood at the centre of bipolar disorder treatment. Lithium, a naturally occurring element used in psychiatric care since the mid-twentieth century, is widely regarded as the most reliable mood stabiliser available. It reduces the frequency and severity of manic and depressive episodes, lowers the risk of self-harm and suicide, and helps many patients achieve long-term stability. Clinical guidelines across the globe continue to recommend lithium as a first-line therapy.

Yet clinical reality is less tidy than textbook guidance. A substantial proportion of people diagnosed with bipolar disorder either do not respond adequately to lithium or struggle with side effects that make long-term use difficult. These effects can include tremor, weight gain, gastrointestinal problems, thyroid dysfunction and kidney-related concerns. For some patients, even careful dose adjustments fail to resolve these issues. Others find that lithium alone does not fully protect them against relapse, especially during periods of stress or life change.

This gap between ideal treatment and real-world outcomes has long challenged psychiatrists. When lithium falls short, clinicians often turn to alternative medications or combine several drugs in the hope of achieving better control of symptoms.

However, high-quality evidence on which combinations work best over time has been limited. Randomised clinical trials rarely capture the complexity of long-term care, where patients move between treatments, stop medications, restart them, or receive add-on therapies as their condition evolves.

A new large-scale observational study published in the journal Nature Mental Health aims to address this uncertainty. Conducted by researchers from leading psychiatric institutions in Finland and Sweden, the study examined how different treatment strategies for bipolar disorder perform over time, both during lithium use and after it is discontinued. Drawing on health registry data from more than 160,000 individuals treated for bipolar disorder, the research offers one of the most comprehensive real-world assessments of medication effectiveness to date.

What sets this study apart is its methodological approach. Rather than comparing outcomes between different groups of patients, the researchers compared each patient with themselves across different periods of treatment. This means that an individual’s outcomes while taking lithium could be compared with their outcomes when using another medication or combination. By doing so, the study reduced the influence of fixed factors such as genetics, baseline illness severity, or long-standing lifestyle habits. These elements remain constant within a person, making it easier to isolate the impact of treatment changes.

The researchers focused on relapse as a key outcome, using psychiatric hospital admissions as a practical and objective measure. While hospitalisation does not capture every relapse, it reliably reflects periods when symptoms worsen to the point that specialised care becomes necessary. Fewer admissions generally indicate better long-term stability and more effective symptom control.

The findings provide both reassurance and new direction. Lithium, as expected, remained one of the most effective treatments overall. Patients taking lithium consistently showed lower rates of hospital admission compared with many other treatment options. This reinforces its role as a cornerstone of bipolar disorder management.

However, the study also highlighted several effective alternatives, particularly for patients who cannot continue lithium. Certain combinations of medications were associated with a reduced risk of relapse, suggesting that thoughtful pairing of drugs can compensate, at least in part, for lithium’s absence. Among the most notable findings was the apparent benefit of combining mood stabilisers with specific antipsychotic medications.

Antipsychotic drugs are often associated in the public mind with schizophrenia, yet they also play a crucial role in bipolar disorder, especially during manic episodes or when psychotic symptoms emerge. In this study, both oral and long-acting injectable antipsychotics were linked to improved outcomes. Long-acting formulations, administered by injection at regular intervals, stood out in particular. These treatments may help ensure consistent medication levels and reduce the risk of missed doses, a common challenge in long-term psychiatric care.

Certain drug pairings appeared especially promising. Combinations such as quetiapine with lamotrigine, or olanzapine with valproate, were associated with fewer hospital admissions among patients who had stopped lithium. Clozapine, an antipsychotic typically reserved for treatment-resistant cases, also showed a strong association with reduced relapse, although its use requires careful monitoring due to potential side effects.

Importantly, the study underscored that not all medication combinations are equally effective. Simply adding more drugs does not guarantee better outcomes. The choice of agents, their mechanisms of action, and how they interact all matter.

This finding carries a clear message for clinical practice. Personalised, evidence-informed decision-making remains essential, particularly when navigating treatment after lithium discontinuation. Always talk to your psychiatrist for the best available treatment suited for you.

Despite its scale and sophistication, the research does not claim to offer definitive answers. Observational studies, even very large ones, cannot establish cause and effect with the same certainty as randomised controlled trials. The authors stress that their findings should be interpreted with caution. Factors such as medication adherence, psychosocial support, and co-existing health conditions may still influence outcomes in ways that are difficult to fully account for.

Even so, the study fills a critical evidence gap. In everyday psychiatric practice, clinicians rarely have the luxury of waiting for perfect data. Decisions must be made in the face of incomplete information, guided by experience, patient preference and the best available research. Large observational studies like this one provide valuable insights into what tends to work in real-world settings, across diverse populations and healthcare systems.

The implications extend beyond medication choice alone. The findings also highlight the importance of continuity of care and regular monitoring. Treatments that reduce hospital admissions may also lessen the broader personal and social costs of bipolar disorder, including disruptions to work, education and relationships. Fewer relapses can translate into greater stability, improved quality of life and reduced strain on healthcare services.

Looking ahead, the researchers point to several avenues for further investigation. Hospitalisation is only one measure of treatment success. Future studies may examine functional outcomes, such as the ability to maintain employment, social connections and independent living. Quality of life, patient satisfaction and cognitive functioning are also crucial aspects of recovery that deserve greater attention.

There is also growing interest in more personalised approaches to bipolar disorder treatment. Advances in genetics, brain imaging and digital health tools raise the possibility of tailoring medication strategies to individual profiles. Machine learning techniques could one day help predict which treatment is most likely to work for a given person, reducing the trial-and-error process that currently characterises much of psychiatric care.

For now, the conclusion from this study is pragmatic and cautiously optimistic. Lithium remains a highly effective treatment for many people with bipolar disorder. When it works and is well tolerated, it continues to offer substantial benefits. When it does not, patients and clinicians are not without options. Several alternative strategies, particularly well-chosen combinations of mood stabilisers and antipsychotics, appear capable of reducing relapse risk and supporting long-term stability.

Always talk to your healthcare providers for medication advice.

As bipolar disorder continues to pose significant challenges for patients, families and healthcare systems, research that reflects real-world complexity is especially valuable.

By analysing treatment outcomes across hundreds of thousands of patient-years, this study brings clinicians closer to answering a question faced daily in mental health clinics: what to do when the gold standard is no longer enough. It offers hope and restarts meaningful conversation between doctors and patients.