Intimate Partner Violence Tied to Working Memory Decline Into Menopause

CHICAGO — Women who had experienced physical intimate partner violence experienced a decline in working memory as they transitioned into menopause, a subgroup analysis of the longitudinal Study of Women’s Health Across the Nation (SWAN) found.

Reports of physical intimate partner violence in the year before SWAN baseline — when participants were pre- or perimenopausal — were associated with declines in working memory performance on the Digit Span Backwards test of -0.17 points per year after age 61 (95% CI -0.28 to -0.06, P=0.002) compared to the performance of women who had not experienced such violence in the year prior to baseline, Jillian Baker, MPH, of the University of Michigan in Ann Arbor, reported in a poster presentation at the annual meeting of the Menopause Society.

“Over the course of 8 to 10 years, that seems like it could be a pretty meaningful decline in working memory, and working memory allows us to hold and manipulate information, and it’s critical for things like remembering directions or remembering a phone number, for example,” Baker told MedPage Today. “And so that could have pretty serious implications for quality of life.”

Around a quarter of U.S. women have experienced severe physical violence by a partner in their lifetime, which, along with rape and stalking by an intimate partner, is associated with a higher prevalence of a range of mental and physical health conditions compared with those without a history of intimate partner violence. Previous studies have shown associations between intimate partner violence and neuropsychiatric function, including working memory.

Notably, intimate partner violence in the year prior to baseline was not associated with significant declines in other cognitive domains.

“That’s one of the questions that we have moving forward, is why we saw such a strong signal in working memory, but not necessarily processing speed or verbal memory?” Baker said. “So that is a question that we can’t answer right now.”

There were also no statistically significant interactions between depressive and anxiety symptoms and intimate partner violence, according to the presentation.

Baker said it’s also likely that there were women who experienced intimate partner violence prior to or after baseline, “but if they didn’t experience it then, or they chose not to report it, then they’re not captured in this, so I would expect that the relationship is even more stark than we’re seeing,” Baker said.

Previous SWAN research had established 61 as the best-fitting age for their model of working memory performance, Baker said. Age 61 was an “inflection point” where a decline in working memory was observed in the study population.

Still, “Age 61 isn’t prescriptive and women with a history of [intimate partner violence] shouldn’t expect to have immediate declines in working memory after age 61, nor should they necessarily expect no change before age 61,” Baker wrote in an email. “More practically, around that age … is where the data indicate that working memory performance may reflect a greater change.”

Potential mechanisms for this disparity, the researchers theorized, could be traumatic brain injury related to the physical violence or mental health conditions.

SWAN is a multisite, longitudinal cohort study that began in 1996 and enrolled 3,302 participants to define and understand the menopause transition in an ethnically and racially diverse sample of women. This analysis included 1,713 women, including 53 women (3.1%) who had reported physical intimate partner violence in the year prior to SWAN baseline in 1996-1997. They underwent baseline cognitive testing from 2002-2003, and at least one of five other visits between 2003-2016.

In this subgroup, mean age at baseline cognitive testing was 54, and 50.7% of the participants were white women.

Cognitive tests included the Digit Span Backwards test, Symbol Digit Modality Test, and East Boston Memory Test-Delayed, which assessed, respectively, working memory, processing speed, and verbal memory. The Digit Span Backwards test required participants to recite backwards numbers recited by the examiner, with one point given for each number sequence recalled correctly. To minimize practice effects, cognitive testing earlier than the 2002-2003 visit was not included in the analysis.

Researchers controlled for education, depressive and anxiety symptoms, trouble sleeping, bodily pain, and race. Limitations included self-reported definitions of intimate partner violence, loss to follow up over time, and the possibility of unmeasured confounders.