Skip to main content

Coercive Control & Human Health

Coercive Control and Dementia Risk

What research reveals about narcissistic abuse and cognitive health

Sharon Mackenzie | Spirit Led Institute | November 2026

What happens to the brain of a person who spends twenty, thirty or forty years living under coercive control? We have become more familiar with the psychological consequences of narcissistic abuse: hypervigilance, anxiety, depression, the gradual erosion of self-trust and the loss of personal agency. But we must also ask what sustained exposure to an oppressive relational environment may do to cognitive health over the course of a lifetime.

Research published in Alzheimer’s & Dementia in 2025 gives us reason to take this question seriously. Emotional partner abuse and coercive control were associated with poorer cognitive functioning in a large cohort of women. Separate bodies of research link psychological trauma and chronic stress with a higher incidence of dementia. Together, these findings raise an important question for safeguarding and public health: could long-term exposure to coercive control contribute to vulnerability to dementia, including Alzheimer’s disease? The evidence warrants investigation, although a direct causal relationship has not yet been established.

Understanding the exposure

Narcissistic abuse is a term used to describe patterns of exploitation, manipulation, devaluation and control associated with pronounced narcissistic traits. It is not a standardised medical diagnosis. Coercive control, by contrast, describes a pattern of behaviour through which a person restricts another person’s freedom and autonomy. It may involve monitoring, intimidation, isolation, humiliation, financial restrictions or repeatedly undermining a person’s confidence in their own perceptions. The impact lies in the cumulative nature of the environment, rather than in any one incident.

Day and colleagues (2025), including researchers at the University of Wollongong, examined 135 people in close relationships with individuals reported to have pronounced narcissistic traits. The average relationship had lasted twenty years. The researchers found an association between pathological narcissism and coercive control, though not with abuse overall; broader impairments in personality functioning were associated with both. [1] This study helps connect the discourse around narcissistic abuse to measurable controlling behaviours. It did not examine dementia, and a diagnosis of narcissistic personality disorder is neither necessary nor sufficient to identify coercive control.

What the cognitive research found

Murchland and colleagues (2025) analysed data from 14,771 female participants in the US Nurses’ Health Study II. Experiences of intimate partner violence were reported in 2001 and 2008, and cognitive testing took place between 2014 and 2019. Emotional abuse showed the strongest association with lower cognitive scores among the abuse subtypes examined. Learning and working memory accounted for much of the difference. [2]

Of particular interest was a secondary analysis of 14,304 participants who answered questions about psychological abuse and coercive control in their current or most recent relationship. At the highest reported levels of control, the difference in average cognitive performance was comparable to the difference observed between women six to seven years apart in age. [2] This is an age-equivalent difference in test scores, not a finding that the brain had physically aged by that amount.

The researchers did not find an accelerated rate of cognitive decline associated with partner violence during the assessment period, and the study did not measure new dementia diagnoses. Cognitive functioning, the rate of cognitive decline and dementia are distinct outcomes. Nevertheless, the findings demonstrate why psychological violence deserves attention in research on brain health, including where there has been no physical assault.

Psychological trauma and later dementia

Coercive control is associated with trauma-related mental ill-health. A systematic review by Lohmann and colleagues (published online in 2023 and in print in 2024) identified 68 relevant studies and pooled data from 45. Coercive control was moderately associated with both post-traumatic stress disorder (PTSD) and depression. Evidence specifically concerning complex PTSD was too sparse to pool. [3]

In a 2022 study of 12,270 trauma-exposed middle-aged women, Roberts and colleagues found that those reporting high levels of PTSD symptoms experienced faster decline in learning and working memory, as well as speed and attention, than women without PTSD symptoms. [4] Unlike the 2025 partner-abuse study, this research detected a difference in the rate of cognitive change.

A 2020 systematic review led by Günak pooled eight studies covering nearly 1.7 million people. PTSD was associated with a 61% higher relative hazard of dementia (HR 1.61; 95% confidence interval 1.43–1.81). A 2026 meta-analysis of 18 studies reported a similar adjusted estimate of 56% (HR 1.56; 95% confidence interval 1.27–1.91), with substantial variation across studies. [5,6] These figures describe relative associations in populations, not an individual’s probability of developing dementia. They also do not establish that coercive control itself causes dementia.

Chronic stress and Alzheimer’s disease

Research into chronic psychological stress provides a further line of evidence. Johansson and colleagues followed 800 Swedish women from midlife into later life. Over approximately 37 years of observation, 153 developed dementia, including 104 with Alzheimer’s disease. Each additional psychosocial stressor recorded at baseline was associated with a 20% higher hazard of Alzheimer’s disease (HR 1.20; 95% confidence interval 1.07–1.35). The stressors included difficulties at work and within family life, rather than a specific measure of coercive control. [7]

A 2021 meta-analysis by Franks and colleagues also identified associations between perceived stress and later dementia (HR 1.44), and between exposure to at least two stressful life events and dementia (HR 1.72), compared with no events. [8] These results support the possibility that psychological stress experienced in adulthood has implications for cognitive ageing.

There are plausible biological pathways. Persistently dysregulated stress responses may affect the hippocampus, a region essential to learning and memory, and may interact with cardiovascular, metabolic and inflammatory processes. In a 2025 Framingham Heart Study analysis of 305 participants, Salardini and colleagues found an association between higher midlife cortisol and greater brain amyloid deposition fifteen years later in postmenopausal women. No significant association was observed with tau or in men. [9] Amyloid deposition is a biological marker of Alzheimer’s disease, but it is not equivalent to a clinical diagnosis. The study was not an investigation of abuse exposure.

The limits of the evidence

Research does not uniformly show a relationship between abuse and diagnosed dementia. Cations and colleagues (2022) followed 12,085 Australian women aged 70–75 at enrolment. Their analysis found no association between self-reported intimate partner violence and new dementia diagnoses (HR 1.02; 95% confidence interval 0.89–1.17), although historical violence was associated with earlier mortality. [10] This is an important counterpoint. A study that measures whether a person has experienced a violent relationship does not necessarily capture the duration or intensity of non-physical coercive control, but its findings cannot be dismissed.

Similarly, a 2025 UK Biobank study of 147,942 participants found an association between emotional neglect in adulthood and incident dementia (HR 1.35; 95% confidence interval 1.17–1.55), but no statistically significant association with the emotional abuse category alone (HR 1.01; 95% confidence interval 0.84–1.22). [11] Different definitions, populations and outcomes limit direct comparisons. The specific effect of sustained coercive control on diagnosed dementia remains an unanswered question.

Implications for safeguarding and care

At Spirit Led Institute, the concern is the environment in which harm occurs. A survivor who has endured years of control may describe forgetfulness, difficulty concentrating, problems making decisions or a loss of confidence in their own memory. These experiences deserve serious attention. They should not be treated as evidence of personal weakness or used to reinforce the controlling narrative. Nor should they automatically be interpreted as dementia: trauma-related cognitive difficulties and neurodegenerative disease are not the same. Persistent or progressive cognitive symptoms warrant appropriate medical assessment alongside an understanding of the person’s relational history.

The research also has implications for prevention. If a person’s long-term cognitive health is influenced by the conditions in which they live, safeguarding cannot stop at the identification of physical injury. Practitioners need to recognise patterns of coercion, preserve access to supportive relationships, restore autonomy and respond to psychological harm before years of exposure accumulate. These measures are justified by the established harms of coercive control even while its direct effects on dementia risk are being investigated.

A question deserving further research

We need longitudinal studies that measure coercive control itself: its severity, duration, forms, timing and persistence. Such studies should examine cognitive performance over time and clinically confirmed dementia subtypes, while accounting for age, genetics, childhood adversity, head injury, vascular disease and other potential influences. They should investigate whether ending the abuse and receiving effective support changes subsequent cognitive outcomes. They should include survivors of different genders and relational circumstances.

The emerging literature gives us a credible basis for asking whether prolonged coercive control is an under-recognised contributor to cognitive vulnerability and dementia risk. It does not yet give us a basis for claiming that narcissistic abuse causes Alzheimer’s disease. That distinction matters because survivors deserve evidence that accurately describes the harm they have experienced, and because research into the long-term consequences of abusive environments deserves to be taken seriously.

We should be asking what twenty years of psychological captivity may cost a person, not only during the relationship, but across the rest of their life.

Research references

[1] Day, N. J. S., et al. (2025). Coercive control and intimate partner violence: Relationship with personality disorder severity and pathological narcissism. Personality and Mental Health, 19(4), e70038. https://doi.org/10.1002/pmh.70038

[2] Murchland, A. R., et al. (2025). Intimate partner violence and cognitive functioning – toward quantifying dementia risk. Alzheimer’s & Dementia, 21(5), e70029. https://doi.org/10.1002/alz.70029

[3] Lohmann, S., Cowlishaw, S., Ney, L., O’Donnell, M., & Felmingham, K. L. (2024; online 2023). The trauma and mental health impacts of coercive control: A systematic review and meta-analysis. Trauma, Violence, & Abuse, 25(1). https://doi.org/10.1177/15248380231162972

[4] Roberts, A. L., et al. (2022). Association of posttraumatic stress disorder with accelerated cognitive decline in middle-aged women. JAMA Network Open. https://pubmed.ncbi.nlm.nih.gov/35771577/

[5] Günak, M. M., et al. (2020). Post-traumatic stress disorder as a risk factor for dementia: Systematic review and meta-analysis. British Journal of Psychiatry, 217(5), 600–608. https://doi.org/10.1192/bjp.2020.150

[6] Havermans, D. C. D., et al. (2026, online first). The role of posttraumatic stress disorder as a risk factor for dementia: A meta-analysis. Journal of Geriatric Psychiatry and Neurology. Published online 30 March 2026. https://doi.org/10.1177/08919887261435186

[7] Johansson, L., et al. (2013). Common psychosocial stressors in middle-aged women related to longstanding distress and increased risk of Alzheimer’s disease: A 38-year longitudinal population study. BMJ Open, 3, e003142. https://doi.org/10.1136/bmjopen-2013-003142

[8] Franks, K. H., Bransby, L., Saling, M. M., & Pase, M. P. (2021). Association of stress with risk of dementia and mild cognitive impairment: A systematic review and meta-analysis. Journal of Alzheimer’s Disease, 82(4), 1573–1590. https://doi.org/10.3233/JAD-210094

[9] Salardini, A., et al. (2025). Elevated serum cortisol associated with early-detected increase of brain amyloid deposition in Alzheimer’s disease imaging biomarkers among menopausal women: The Framingham Heart Study. Alzheimer’s & Dementia, 21(4), e70179. https://doi.org/10.1002/alz.70179

[10] Cations, M., Keage, H. A. D., Laver, K. E., Byles, J., & Loxton, D. (2022; online 2020). Intimate partner violence and risk for mortality and incident dementia in older women. Journal of Interpersonal Violence, 37(5–6). https://doi.org/10.1177/0886260520943712

[11] Li, J., et al. (2025). Adverse experiences in childhood and young adulthood, genetic susceptibility, and neurodegenerative disease incidence: A lifespan analysis. European Journal of Psychotraumatology, 16(1), 2599612. https://doi.org/10.1080/20008066.2025.2599612

Eucalyptus overlooking an Australian valley at sunsetGolden Australian wattle flowers at sunset