Low-dose lithium sits at the center of this dementia and brain health question.

A pilot study from the University of Pittsburgh offers tentative hope for slowing cognitive decline in older adults with mild memory impairment. Researchers found that low-dose lithium treatment slowed the rate of verbal memory decline—the ability to remember and recall words and sentences—by roughly half compared to placebo, particularly in participants with evidence of Alzheimer’s-related brain changes.

This isn’t a cure or a reversal of damage, but it represents meaningful progress in understanding how a well-known psychiatric medication might play a role in preventing Alzheimer’s disease progression. The trial, called LATTICE (Lithium as a Treatment to prevent Impairment of Cognition in Elders), ran from 2017 to 2024 and its results, published in March 2026, are now generating interest among researchers worldwide. This article explores what the study found, why the results matter despite falling short of some initial goals, and what comes next for lithium research in Alzheimer’s prevention.

How Did Researchers Test Low-Dose Lithium in Older Adults?
What Happened to Memory in the Lithium Group Versus Placebo?
Why Does Slowing Verbal Memory Decline Matter for Alzheimer’s Prevention?
How Safe Was Low-Dose Lithium in This Population of Older Adults?
Why Didn’t the Study Achieve Its Original Primary Goals?
How Will Researchers Use These Findings to Design Larger, More Definitive Trials?
What Larger Prevention Trials Are Underway or Planned?
Conclusion

How Did Researchers Test Low-Dose Lithium in Older Adults?

The LATTICE study began in September 2017 at the University of Pittsburgh Department of Psychiatry under the direction of Dr. Ariel Gildengers. The research team enrolled 80 community-dwelling adults aged 60 and older who had already been diagnosed with mild cognitive impairment (MCI)—the stage where memory problems are noticeable but haven’t yet progressed to dementia. The average participant was 72 years old; 45 were women and 35 were men. This wasn’t a study of healthy people hoping to prevent decline, but rather people who were already experiencing early cognitive slippage and needed intervention. The study used a randomized, double-blind, placebo-controlled design, the gold standard in clinical research.

Half the participants received low-dose lithium targeting blood levels of 0.6 to 0.8 mmol/L—roughly one-quarter to one-half the dose typically used for bipolar disorder treatment. The other half received an identical-looking placebo. Importantly, neither the participants nor the researchers administering the treatment knew who was receiving the real medication until the study ended. This design prevents bias and ensures any differences observed are genuine treatment effects, not wishful thinking or expectation. The study ran for two full years, long enough to track meaningful changes in cognition while minimizing the risk of people dropping out. At the end of the two-year period, researchers measured multiple aspects of cognitive function and brain health. They compared how quickly the lithium and placebo groups experienced decline across different mental abilities.

How Did Researchers Test Low-Dose Lithium in Older Adults?

What Happened to Memory in the Lithium Group Versus Placebo?

The most striking finding involved verbal memory measured with the California Verbal Learning Test-II (CVLT-II). This test assesses how well people can learn, retain, and recall a list of words—a standard way to measure memory decline in aging and dementia research. In the placebo group, verbal memory scores declined at a rate of 1.42 points per year. In the lithium group, the decline was significantly slower: 0.73 points per year. That’s roughly a 50% reduction in the rate of decline. For someone in the lithium group, this difference means that in one year, their memory decline would be about half as severe as someone in the placebo group. Over five years, that difference compounds considerably. However, this benefit didn’t apply equally to everyone in the lithium group.

Researchers found that participants with evidence of amyloid beta—a hallmark Alzheimer’s biomarker that accumulates in the brains of people at risk for or developing Alzheimer’s disease—showed the most pronounced benefit. This is a critical finding because it suggests that lithium isn’t a universal cognitive booster. Rather, it appears to work specifically in people whose cognitive decline is being driven by Alzheimer’s-related pathology. Someone experiencing memory problems from another cause, such as small blood vessel disease or Lewy body pathology, might not experience the same benefit. It’s important to note what this finding does not mean. A 50% reduction in decline rate is encouraging, but it’s not a reversal or stabilization of function. Even in the lithium group, memory was still declining. The benefit was a slowing of loss, not a recovery of what was already lost.

Why Does Slowing Verbal Memory Decline Matter for Alzheimer’s Prevention?

Verbal memory is one of the first cognitive abilities to deteriorate in early Alzheimer’s disease. When someone struggles to remember a conversation from yesterday or recall the names of new acquaintances, verbal memory decline is often the culprit. This makes it a reliable marker of disease progression. Slowing its decline, even by a modest amount, could delay the transition from mild cognitive impairment to dementia—a progression that typically occurs in about 10 to 15 percent of people with MCI each year. The significance of the LATTICE findings lies not in curing Alzheimer’s but in the possibility of buying time. If lithium can slow cognitive decline in a subset of older adults, that extra time becomes clinically meaningful.

A person whose decline is slowed by 50% might remain independent and functional for significantly longer than they otherwise would. They might retain the ability to manage finances, live alone, or participate fully in social activities for years longer than expected. This has profound implications for quality of life, family burden, and healthcare resource utilization. Moreover, verbal memory decline is measurable and trackable. It’s easier to monitor than abstract concepts like “slowing disease” or “preventing cognitive loss.” This measurability makes lithium a practical tool for future research. Researchers can track changes precisely and use the measurements to identify whether participants are responding to treatment.

Why Does Slowing Verbal Memory Decline Matter for Alzheimer's Prevention?

How Safe Was Low-Dose Lithium in This Population of Older Adults?

Safety is paramount in research with older adults, who often take multiple medications and have reduced kidney function. The good news from LATTICE is clear: low-dose lithium was safe and well tolerated overall. No serious adverse events requiring hospitalization or discontinuation of the medication occurred as a result of the lithium treatment. This matters because it means the dose used in the study—0.6 to 0.8 mmol/L—sits in a sweet spot. It’s high enough to potentially confer cognitive benefits but low enough to avoid the more serious side effects associated with standard-dose lithium use in psychiatric patients. That said, some participants in the lithium group did experience side effects more frequently than those on placebo.

Diarrhea, fatigue, and tremor (fine shaking of the hands) were more common in the lithium-treated participants. These are not dangerous complications, but they are noticeable. A person starting low-dose lithium might experience loose stools or feel more tired than usual, or notice their hands shake slightly when holding a coffee cup. Most people tolerated these effects without stopping treatment, but some individuals chose to discontinue. This highlights an important real-world consideration: tolerability matters. A cognitive benefit is only meaningful if the person taking the medication can stick with it long enough to realize the advantage.

Why Didn’t the Study Achieve Its Original Primary Goals?

The LATTICE trial was designed to test three primary endpoints: whether lithium could improve overall cognition (not just verbal memory), slow brain atrophy (shrinkage of brain tissue over time), and increase blood levels of BDNF (brain-derived neurotrophic factor, a protein thought to support brain health). The study failed to achieve any of these three primary endpoints. Lithium did not improve overall cognitive function across the board. It did not prevent or reduce brain shrinkage. And it did not increase blood BDNF levels. On the surface, this might sound like the study failed. Many news readers skip past these details and assume “no primary endpoints” means “no benefit.” In reality, this is a common scenario in clinical research. The primary endpoints are the gold-standard outcomes a study is powered to detect. Failing to achieve them is sobering but not unusual, especially in early-stage pilot studies.

However, the secondary finding—the significant slowing of verbal memory decline—is valuable in its own right. It’s a real effect that invites further investigation. Think of it this way: you’re testing a new cancer drug hoping it will shrink tumors overall. It doesn’t. But you notice that in a subset of patients with a specific genetic mutation, tumors do shrink. That subset finding becomes the basis for the next, larger trial specifically targeting that population. The failure to show improvements in the primary endpoints also tempers expectations. Lithium is not a broad cognitive enhancer. It’s not going to reverse brain shrinkage or dramatically overhaul brain chemistry in the ways researchers initially hoped. Instead, it appears to have a more specific, more modest role in protecting verbal memory in people with Alzheimer’s pathology.

Why Didn't the Study Achieve Its Original Primary Goals?

How Will Researchers Use These Findings to Design Larger, More Definitive Trials?

The path forward is clear, and researchers at Pittsburgh and elsewhere are already pursuing it. The next phase will involve larger, confirmatory trials powered to detect the verbal memory benefit that the LATTICE study identified. But there’s a key strategic difference in how these trials will be designed. Rather than enrolling everyone with mild cognitive impairment, future trials will use blood-based biomarkers to identify individuals most likely to benefit from lithium treatment—namely, those with evidence of amyloid beta and related Alzheimer’s pathology. This is a precision medicine approach.

Instead of the shotgun approach of giving lithium to all older adults with memory problems and hoping some benefit, researchers will identify who has Alzheimer’s-related pathology first, and then offer lithium to that group. Blood tests that measure amyloid, tau, and phosphorylated tau are now widely available and affordable. This makes pre-screening participants feasible and practical. By focusing on the people most likely to respond, researchers increase the chance of detecting a true treatment effect in a smaller, less costly trial. They also make the treatment more ethical—patients who don’t have Alzheimer’s pathology won’t be unnecessarily exposed to lithium’s side effects.

What Larger Prevention Trials Are Underway or Planned?

Beyond the confirmatory trials being planned at Pittsburgh and other institutions, there’s already a Phase 4 lithium prevention trial running in Australia and Europe. This study, registered on ClinicalTrials.gov, is testing lithium’s effects in people at risk for cognitive decline due to mood disorders—a population that may also be at elevated risk for Alzheimer’s disease. This trial is expected to conclude in June 2026, providing additional real-world evidence about how lithium performs in a larger, geographically diverse population.

The timeline is important. With results from LATTICE published in March 2026 and the Australian-European trial concluding in June 2026, the scientific community will have multiple independent data points about low-dose lithium’s effects within the same year. This convergence of evidence will inform whether lithium deserves a place in the arsenal of Alzheimer’s prevention strategies. Researchers are actively seeking funding for the larger confirmatory trials, so expansion of these studies is contingent on securing grants and resources.

Conclusion

The LATTICE trial doesn’t yet prove that low-dose lithium prevents Alzheimer’s disease or stops cognitive decline in older adults. But it does provide the first solid evidence from a well-designed study that lithium slows a key marker of decline—verbal memory—in people with mild cognitive impairment and Alzheimer’s pathology. This finding shifts lithium from being an interesting hypothesis to being a biologically plausible candidate worth investigating further. For people currently experiencing memory loss, it suggests a concrete possibility worth discussing with their neurologist or memory care specialist, particularly if they have biomarker evidence of Alzheimer’s disease.

For the broader field of dementia prevention, LATTICE opens a door. It demonstrates that even a medication with a 70-year history, typically used in psychiatry, can teach us something new about protecting cognition in aging. The next few years will be critical. Larger trials targeting people with Alzheimer’s pathology, combined with results from ongoing international studies, will determine whether low-dose lithium becomes part of standard cognitive care for older adults at risk. Until then, the message is cautiously optimistic: there may be a way to slow memory decline, at least for some people, and researchers are working hard to confirm and expand that promise.