The study unearthed a 20% diagnostic error rate with non-sex-specific parameters—10% false negatives in women and 10% false positives in men—which the authors found may be eliminated with gender-specific norms and cut scores.
Erin E. Sundermann, PhD
In amnestic mild cognitive impairment (MCI), a lack of sex-specific parameters is associated with a 20% diagnostic error rate in verbal memory testing, according to the results of a new study.1
Conducted by Erin E. Sundermann, PhD, assistant project scientist, psychiatry, University of California­—San Diego, and colleagues, the results suggest that applying specific cut scores for men and women when defining verbal memory impairment can improve diagnostic accuracy. This, in turn, may result in earlier detection of MCI in women while preventing false diagnoses in men.2
"If these results are confirmed, they have vital implications,” Sundermann said in a statement. Sundermann and colleagues noted that longitudinal analysis is warranted based on these data, and their replication may have an impact both in clinical and research practices.
The investigators acknowledged that participants were mostly well-educated compared to the general population, so the results may not apply to all groups, marking it as a limitation of the study.
“If women are inaccurately identified as having no problems with memory and thinking skills when they actually have mild cognitive impairment, then treatments are not being started and they and their families are not planning ahead for their care or their financial or legal situations,” Sundermann explained. “And for men who are inaccurately diagnosed with mild cognitive impairment, they can be exposed to unneeded medications along with undue stress for them and their families."
In total, the study included 985 individuals, 453 women and 532 men, from the Alzheimer’s Disease Neuroimaging Initiative, all of whom took the Rey Auditory Verbal Learning Test (RAVLT) in the Mayo Clinic Study of Aging­—from which a separate group of older adults were used to calculate typical score sets.
Women significantly outperformed men in RAVLT scores with a mean of 42.3 (SD, 11.5) compared to 35.6 (SD, 11.1) as well as Delayed Recall with a mean score of 6.2 (SD, 4.5) compared to 4.5 (SD, 3.9) for men (P <.001). The difference in the scores for men and women was an average of 6 points within a score range of 0 to 75, and an average of 2 points on the test taken after a delay within a score range of 0 to 15.
Women in the study were younger (71.9 years; standard deviation [SD], 6.8) than men (73.6 years [SD, 7.0]), and also had significantly higher mean Mini-Mental State Examination (MMSE) scores (mean, 28 [SD, 1.6]) and Geriatric Depression Scale scores (mean, 1.5 [SD, 1.5]) compared to men (MMSE: 28.1 [SD, 1.7]; Geriatric: 1.3 [SD, 1.4]; P <.05).
When using typical cut scores, the frequency of MCI diagnosis was significantly higher in men than women (P <.001), though when utilizing sex-specific cut scores this difference disappeared (P = .62). In total, 35% (n = 184) of men and 26% (n = 120) of women were deemed to have true positive diagnoses, while true negatives were determined for 55% (n = 293) of men and 64% (n = 288) of women. False negatives occurred in 10% (n = 45) of women, and false positives occurred in 10% (n = 55) of men. No women were categorized as false positive and no men were categorized as false negative.
All told, the sex-specific diagnosis criteria resulted in 65% (n = 348) of men and 64% (n = 288) of women being characterized as cognitively normal, compared to rates of 55% (n = 293) and 73% (n = 73), respectively. Similarly, rates of MCI diagnosis increased from 26% (n = 120) to 36% (n = 165) of women and decreased from 45% (n = 239) to 35% (n = 184) of men once using sex-specific cut scores.
"When the typical average cut-off scores are used for diagnosis, women might respond less to treatments in a clinical trial than men because they are at a more advanced stage of the disease, while men might not respond because some of them do not actually have MCI," Sundermann said. "These combined factors would result in research that reduces the estimate of how well treatments work for both men and women."
1. Sunderman EE, Maki P, Biegon A, et al. Sex-specific norms for verbal memory test improve diagnostic accuracy of amnestic MCI. Neurology. First published October 9, 2019. doi: 10.1212/WNL.0000000000008467.
2. Should scores on mild cognitive impairment tests be adjusted for sex? [press release]. Published October 9, 2019. aan.com/PressRoom/Home/PressRelease/2751. Accessed October 10, 2019.