 |
|
|
| |
HIV+ Nonnative English Speakers Have Higher
Odds of Neurocognitive Impairment
|
| |
| |
IDWeek 2015, October 7-11, San Diego
Mark Mascolini
HIV-positive people who are not native English speakers had 4.5-fold higher odds of test-determined neurocognitive impairment than native English speakers in a small study of military beneficiaries [1]. HIV-related variables did not affect chances of neurocognitive impairment in this analysis, but years of education did.
Neurocognitive impairment, especially in mild and asymptomatic forms, persists in HIV populations responding well to antiretroviral therapy. Accurate determination of neurocognitive impairment, noted US military researchers who conducted this study, depends on neuropsychological testing. Previous research found that native English speakers outperformed nonnative English speakers on language-mediated tasks in these tests, and that difference affected several neuropsychological test measures [2]. The researchers added that ethnic and racial minorities with HIV in the United States tend to have higher rates of neurocognitive impairment and that reasons for this higher prevalence remain poorly understood.
To address these issues, the investigators conducted this cross-sectional study of 197 HIV-positive people who completed a standard battery of neuropsychological tests. The researchers normalized raw scores for age, education, sex, and race and figured a global deficit score to summarize these results. They defined neurocognitive impairment as a score at or above 0.5. The researchers used logistic regression analysis to identify independent predictors of neurocognitive impairment.
The analysis involved 19 nonnative English speakers and 178 native English speakers. Ten nonnative speakers were Hispanic, 6 Asian/Pacific Islander, 2 Caucasian, and 1 African American. The most frequent native languages after English were Spanish in 13 people and Filipino dialects in 4.
Most people in the nonnative and native groups were men (89.5% and 96.1%), and they had a similar median number of education years (15 and 14). Median monthly salary was the same in both groups ($4000), and about 20% in both groups were military officers. Nonnative and native speakers did not differ significantly in proportions with a lifetime history of alcohol use, drug abuse, or major depressive disorder.
Rates of HIV-related characteristics did not differ significantly between participants by native English speaking status, including years since HIV diagnosis, proportion taking antiretroviral therapy, proportion currently taking efavirenz, viral load (median about 50 copies in both groups), or nadir or current CD4 count.
The researchers recorded a global deficit score at or above 0.5, indicating neurocognitive impairment, in 42% of nonnative English speakers versus 17% of native speakers, a significant difference (P = 0.008). Higher proportions of nonnative speakers than native speakers had clinical impairment in 4 domains: verbal fluency (21% versus 8%, P = 0.079), attention/working memory (42% versus 21%, P = 0.035), learning (42% versus 19%, P = 0.020), and recall (37% versus 17%, P = 0.034).
Multivariable analysis adjusted for age, ethnicity, CD4 count, CD4 recovery, duration of HIV infection, and duration of antiretroviral therapy identified only 2 variables independently associated with neurocognitive impairment. Being a nonnative English speaker more than quadrupled the odds (adjusted odds ratio 4.5, 95% confidence interval 1.2 to 16.8, P = 0.025), and more years of education lowered the odds (adjusted odds ratio 0.81, 95% confidence interval 0.68 to 0.97, P = 0.024).
The researchers proposed that worse performance of nonnative English speakers in most neuropsychological testing domains suggests "a possible linguistic effect/bias of standard neuropsychological tests." But they noted that "unmeasured factors, including early lifetime experience and quality of education, may also have contributed to the decreased performance observed among nonnative speakers of English." The investigators called for further study of linguistic impact on neurocognitive impairment and refinement of neuropsychological testing.
References
1. Deiss R, Marquine M, Crum-Cianflone N, et al. Neurocognitive impairment among HIV-infected military beneficiaries: the impact of linguistic background. IDWeek 2015, October 7-11, San Diego. Abstract 1659. https://idsa.confex.com/idsa/2015/webprogram/Paper52720.html
2. Boone KB, Victor TL, Wen J, Razani J, Ponton M. The association between neuropsychological scores and ethnicity, language, and acculturation variables in a large patient population. Arch Clin Neuropsychol. 2007;22:355-365. http://www.sciencedirect.com/science/article/pii/S0887617707000170
|
| |
|
|
|
|
|
