Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario, Canada; Department of Psychiatry and Department of Medicine (Division of Neurology), University of Toronto, Toronto, Ontario, Canada; Department of Clinical Epidemiology, College of Medicine, University of the Philippines Manila, Manila, Philippines; Concussion Canadian Center, University Health Network, University of Toronto, Toronto, ON, Canada.
Department of Clinical Neurological Sciences, London Health Sciences Centre, Western University, London, ON, Canada; Concussion Canadian Center, University Health Network, University of Toronto, Toronto, ON, Canada.
Mult Scler Relat Disord. 2024 Jul;87:105645. doi: 10.1016/j.msard.2024.105645. Epub 2024 May 5.
BACKGROUND: Cognitive impairment (CI) is common in people with MS (PwMS). Evidence is lacking for the self-reported CI's mediation effect on employment status and objective cognitive performance. Self-reported CI was found to be unreliable and seemed to be more associated with depression rather than formal cognitive performance. We hypothesized that the link between subjective and objective assessments of cognitive functions, mood, and employment status may be more complex in PwMS than previously reported. OBJECTIVE: The aims of this study are the following: (Romero-Pinel et al., 2022) to determine whether employment status could affect performance in cognitive function testing and (Rao et al., 1991) whether their relationship may be mediated by self-reported CI; and (Deluca et al., 2013) to determine whether self-reported depression interacts with self-reported CI in influencing performance in various cognitive domains in PwMS. METHODOLOGY: A retrospective study was performed involving PwMS who completed the self-report Multiple Sclerosis Neuropsychological Questionnaire (MSNQ), Hospital Anxiety and Depression Scale-depression scale (HADS-D), Minimal Assessment of Cognitive Function in MS (MACFIMS) and had data regarding employment status. Included PwMS were classified as employed or unemployed. A structural equation modeling (SEM) approach was taken due to the advantage of examining multiple cognitive outcomes simultaneously while accounting for shared associations. First, a latent factor of memory and executive functioning modeled the error-free associations between both factors and a processing speed task (SDMT). Next, the model tested for the indirect effect of self-reported cognition (MSNQ) on employment status differences in each outcome (memory, speed, and executive functioning). Finally, we tested interactions between MSNQ and HADS-D on each of the outcomes. RESULTS: We included 590 PwMS: 72.5% female, mean age 44.2 years (SD = 10.5), mean disease duration 8.6 years (SD 9.0). The majority (n = 455, 77.1%) had relapsing MS; 357 (60.5%) were employed. About half (n = 301, 51%) did not report CI on the MSNQ; of those, 213 (70.8%) were employed. The mean MSNQ for employed PwMS was 24.5 (SD = 10.7) and 29.8 (SD = 11.2) for unemployed PwMS. Employed PwMS had significantly better memory (β = .16, p < .05), executive functioning (β = .25, p < .05), and processing speed (β = .22, p < .05). MSNQ partially indirectly mediated the effect of employment status on memory (Δβ = .03, p < .05) and executive functioning (Δβ = .03, p < .05) and processing speed (Δβ = .04, p < .05), indicating that self-report CI partially explains the influence of employment status on these cognitive domains. The association between MSNQ with both memory and executive functioning was moderated by depression, meaning that in PwMS with high HADS-D scores, MSNQ was more strongly related to worse memory and executive functioning. The final model was an acceptable fit to the data (χ = 465.07, p < .05; CFI = .90, RMSEA = .08, 90% CI [.06, .09], SRMR = .05) explaining 41.20%, 38.50% and 33.40% of the variability in memory, executive functioning, and processing speed, respectively. CONCLUSION: Self-reported CI partially explains the associations between employment status and objective cognitive assessment in PwMS. Depression may moderate the relationship between self-reported cognitive assessment and objective cognitive performance. Thus, employment status and mood may guide the interpretation of self-reported CI.
背景：认知障碍（CI）在多发性硬化症患者（PwMS）中很常见。缺乏证据表明自我报告的 CI 对就业状况和客观认知表现有中介作用。自我报告的 CI 被发现不可靠，似乎与抑郁的关系更密切，而不是与正式的认知表现。我们假设，在 PwMS 中，主观和客观认知功能、情绪和就业状况之间的联系可能比以前报道的更复杂。
目的：本研究的目的如下：（Romero-Pinel 等人，2022）确定就业状况是否会影响认知功能测试的表现，（Rao 等人，1991）他们之间的关系是否可能由自我报告的 CI 介导；以及（Deluca 等人，2013）确定自我报告的抑郁是否会影响各种认知领域的自我报告 CI 在 PwMS 中的影响。
方法：进行了一项回顾性研究，纳入了完成自我报告多发性硬化症神经心理学问卷（MSNQ）、医院焦虑和抑郁量表-抑郁量表（HADS-D）、最小多发性硬化症认知功能评估（MACFIMS）且有就业状况数据的 PwMS。纳入的 PwMS 被分为在职或失业。由于同时考虑多个认知结果并解释共享关联，因此采用结构方程建模（SEM）方法。首先，记忆和执行功能的潜在因素模拟了两个因素与处理速度任务（SDMT）之间的无误差关联。接下来，该模型测试了自我报告认知（MSNQ）对每个结果（记忆、速度和执行功能）的就业状况差异的间接影响。最后，我们测试了 MSNQ 和 HADS-D 之间在每个结果上的相互作用。
结果：我们纳入了 590 名 PwMS：女性占 72.5%，平均年龄 44.2 岁（标准差=10.5），平均疾病持续时间 8.6 年（标准差=9.0）。大多数（n=455，77.1%）为复发型 MS；357 人（60.5%）在职。约一半（n=301，51%）在 MSNQ 上未报告 CI；其中，213 人（70.8%）在职。在职 PwMS 的平均 MSNQ 为 24.5（标准差=10.7），失业 PwMS 为 29.8（标准差=11.2）。在职 PwMS 的记忆（β=.16，p<.05）、执行功能（β=.25，p<.05）和处理速度（β=.22，p<.05）明显更好。MSNQ 部分间接介导了就业状况对记忆（Δβ=.03，p<.05）和执行功能（Δβ=.03，p<.05）和处理速度（Δβ=.04，p<.05）的影响，表明自我报告的 CI 部分解释了就业状况对这些认知领域的影响。MSNQ 与记忆和执行功能的关联受抑郁的调节，这意味着在 HADS-D 得分较高的 PwMS 中，MSNQ 与记忆和执行功能的相关性更差。最终模型对数据的拟合良好（χ=465.07，p<.05；CFI=.90，RMSEA=.08，90%CI[.06，.09]，SRMR=.05），分别解释了记忆、执行功能和处理速度的 41.20%、38.50%和 33.40%的变异性。
结论：自我报告的 CI 部分解释了 PwMS 中就业状况和客观认知评估之间的关联。抑郁可能调节自我报告认知评估与客观认知表现之间的关系。因此，就业状况和情绪可能指导自我报告 CI 的解释。
Mult Scler Relat Disord. 2019-9-26
IBRO Neurosci Rep. 2022-11-2
Arch Clin Neuropsychol. 2023-2-18
Zotero 插件