Testosterone linked to cognitive decline but behavioral improvements in dementia patients

A new study from Taiwan suggests that higher levels of testosterone in the blood may simultaneously worsen cognitive performance while easing behavioral and psychological symptoms in older adults with dementia. These findings, published in the journal Psychiatry Research, point to a complex and potentially conflicting role for testosterone in patients experiencing the behavioral and emotional challenges that often accompany cognitive decline.

Testosterone is often recognized for its roles in physical development, sexual function, and mood regulation, but growing evidence indicates that it also influences brain health. In younger adults, testosterone tends to support neuroplasticity, learning, and memory. However, its effects in aging individuals are less consistent and may depend on a variety of biological conditions.

Prior studies have reported mixed results regarding testosterone and cognition in older populations. Some findings suggest protective effects, while others report no association or even potential harm. In particular, elevated testosterone levels may interact with the brain’s oxidative stress systems in ways that either help or harm cognitive function depending on the context. This dual potential led the Taiwanese researchers to consider whether testosterone might affect not only cognitive performance but also the behavioral symptoms frequently seen in dementia.

Behavioral and psychological symptoms of dementia, often referred to as BPSD, affect the vast majority of people with moderate to severe cognitive decline. These symptoms include hallucinations, aggression, anxiety, depression, and disruptions to sleep or eating patterns. They often impose a heavy burden on caregivers and are a leading reason for institutionalization.

Antipsychotic medications are commonly prescribed, but they come with significant side effects, especially in older adults. If testosterone could be shown to influence these symptoms, it might offer new therapeutic insights or highlight biological pathways worth targeting.

“Behavioral and psychological symptoms of dementia are very common, affecting up to 90% of people with dementia, and can worsen the quality of life, lead to increased caregiver burden, and result in higher healthcare costs,”

explained study authors Chieh-Hsin Lin, a professor at Chang Gung University, and Hsien-Yuan Lane, a professor and director of the Center for Addiction and Mental Health & Department of Psychiatry at China Medical University Hospital. “Prior research showed diverse effects of testosterone on behaviors and cognition in the elderly. Therefore, this study aimed to explore how testosterone influences these two domains in adults with dementia.”

The study included 94 Taiwanese patients diagnosed with Alzheimer’s disease who also showed clinically significant behavioral and psychological symptoms. Participants were between 50 and 100 years old and were recruited from two major medical centers in Taiwan. To qualify, they needed to meet several criteria, including at least six years of education or work experience, a confirmed Alzheimer’s diagnosis, and evidence of behavioral disturbances. Patients with other serious neurological or psychiatric conditions were excluded.

Cognitive function was assessed using the Alzheimer’s Disease Assessment Scale – Cognitive Subscale, or ADAS-Cog, which evaluates memory, language, and other key cognitive domains. Behavioral and psychological symptoms were measured using the Neuropsychiatric Inventory, or NPI, a widely used tool that captures a range of symptoms such as depression, delusions, and nighttime disruptions.

Blood samples were drawn from each participant, and plasma testosterone levels were measured using enzyme-linked immunosorbent assay (ELISA), a standard laboratory technique. The researchers then used statistical models to explore how testosterone levels were associated with scores on the cognitive and behavioral assessments, while also accounting for factors like age, education, disease severity, and use of medications.

The results indicated that testosterone played contrasting roles. Higher testosterone levels were associated with worse cognitive outcomes but better behavioral profiles.

In terms of cognitive performance, the analysis showed that higher testosterone concentrations predicted significantly worse scores on the ADAS-Cog. This suggests a detrimental influence of testosterone on memory and other cognitive skills in this group. The effect appeared somewhat stronger in male patients, although it did not reach statistical significance in the gender-specific analysis.

At the same time, higher testosterone levels were linked to better behavioral and psychological outcomes, as measured by the NPI. Patients with more testosterone tended to show fewer signs of mood disturbances, agitation, or abnormal behaviors. This relationship remained significant even after controlling for age, education, and use of antipsychotic medications.

When researchers broke down the NPI into symptom clusters, they found that testosterone levels were most strongly related to improvements in symptoms like depression and nighttime behavior disturbances. In women, testosterone was also linked to a reduction in a cluster that included hallucinations, delusions, and irritability, once other variables were taken into account.

“We were surprised to learn from this study that testosterone may be a double-edged sword: while it can improve behavioral performance, it may also impair cognitive ability,” the researchers told PsyPost.

These findings provide support for the idea that testosterone interacts with brain chemistry in ways that are context-dependent. In earlier stages of life, or under healthy conditions, testosterone may support brain function. But in older individuals with dementia, particularly in environments of high oxidative stress, testosterone may have a more complicated impact. It could amplify damage in cognitive systems while simultaneously helping to regulate neurotransmitters like dopamine that contribute to behavioral disturbances.

This interpretation is consistent with animal research, which has shown that testosterone can worsen cognitive outcomes in models of neurodegeneration. Other studies have suggested that testosterone’s effects depend on its interaction with other hormones, such as estrogen, and with proteins like sex hormone-binding globulin, which regulates its availability in the body.

But the study, like all research, has some limitations. “First, its findings were based upon a cross-sectional design, which was unable to clarify if higher testosterone directly negatively impacted cognition or if it was a side effect of the disease’s progression,” Lin and Lane noted. “Future longitudinal studies are needed. Second, exclusively Taiwanese people were recruited. Whether the findings can be applied to other populations remains unclear. Third, the sample size was modest. Further studies with larger sample sizes are warranted.”

“Hopefully, testosterone at an optimal level range can improve behavioral/psychological symptoms with minimal, if any, damage to cognitive function, at least in a subpopulation of the people with dementia.”

“In the current sample, a testosterone concentration range between 1.5-2.5 ng/mL could be recommended for balancing both behaviors and cognition: this range yielded a substantial improvement in behavioral/psychological symptoms with a trivial decrease in cognitive function,” the researchers added. “Nonetheless, as aforementioned, more studies are necessary to determine or individualize the optimal level.”

The study, “The double-edged roles of testosterone in cognitive function and behavioral/psychological symptoms of patients with dementia,” was authored by Chieh-Hsin Lin, Shi-Heng Wang, and Hsien-Yuan Lane.