The study, conducted at six tertiary hospitals in China from October 2024 through March 2025, included 370 people with multiple sclerosis (MS) and 88 with neuromyelitis optica spectrum disorder (NMOSD). The NEI VFQ-25 measures how visual problems affect general vision, near and distance activities, peripheral and color vision, social functioning, mental health, role difficulties, dependency and ocular pain. Scores range from 0 to 100, with higher values indicating better vision-related quality of life. Depressive symptom burden was assessed with the seven-item depression subscale of the Hospital Anxiety and Depression Scale (HADS-D), which produces scores from 0 to 21. A score of at least 11 was used as the operational threshold for screen positivity, while a score of at least 15 was used in a sensitivity analysis. These thresholds identify questionnaire-based symptom burden rather than psychiatric diagnoses.

The association was visible across the distribution of vision-related quality-of-life scores. In the lowest NEI VFQ-25 quartile, 90.1% of participants screened positive for depressive symptoms, compared with 72.1% in the second quartile, 62.2% in the third and 38.3% in the highest quartile; the trend was statistically significant at p<0.001. Median HADS-D scores also declined from 16.0 in the lowest quartile to 9.0 in the highest, with a statistically significant difference across quartiles. These results describe a graded cross-sectional relationship: patients reporting more severe vision-related difficulties were more likely to report greater depressive symptom burden.

Patients with NMOSD reported poorer vision-related quality of life overall than those with MS. Their median composite score was 78.9, compared with 88.7 in MS, and they had a higher prevalence of WHO-defined blindness, 27.3% versus 6.8%. Lower scores were reported in most questionnaire subscales, including near activities, distance activities, social functioning, mental health, role difficulties, dependency, color vision and peripheral vision. Ocular pain was the exception, with no statistically significant difference between groups at a nominal p-value of 0.146. The difference in HADS-D scores was smaller: the median was 13.0 in NMOSD and 12.0 in MS, while screen-positive prevalence was 70.5% and 64.9%, respectively. Neither comparison reached conventional statistical significance.

The central result did not depend on which of the two diseases a patient had. In analyses that excluded objective blindness and variables describing ocular history, the association between lower NEI VFQ-25 scores and depression-screen positivity remained significant in MS, with an odds ratio of 2.14 per 10-point decrease, and in NMOSD, with an odds ratio of 1.86. A formal interaction analysis found no evidence that the association differed by disease subtype, with p=0.537. This does not mean that the two groups were otherwise clinically identical: the NMOSD cohort was older, more frequently female, less educated and more likely to meet the study’s definition of blindness. The analysis instead indicates that the association between reported visual-function burden and depressive symptoms was detectable within both disease groups.

Sequential regression models also suggested that the patient-reported questionnaire captured information not fully represented by the binary blindness classification. When WHO-defined blindness was examined without the NEI VFQ-25 composite score, it was associated with depression-screen positivity, with an adjusted odds ratio of 2.12 and a 95% confidence interval of 1.01–4.44. When the questionnaire score was included without blindness, each 10-point decrease was associated with higher odds, with an adjusted odds ratio of 1.56 and a 95% confidence interval of 1.36–1.79. In the model containing both exposures, the blindness estimate weakened to an odds ratio of 1.21, with a 95% confidence interval of 0.53–2.75, and was no longer statistically significant, while the NEI VFQ-25 association remained nearly unchanged, with an odds ratio of 1.55 and a 95% confidence interval of 1.35–1.78. The researchers emphasized that these sequential models were not formal mediation analyses; the change in estimates cannot establish that blindness causes the association or that reported quality of life mediates it.

Domain-specific analyses pointed to psychosocial questionnaire content as particularly informative. The study grouped four subscales into a Core Visual Function score—near activities, distance activities, peripheral vision and color vision—and four into a Psychosocial score—mental health, role difficulties, dependency and social functioning. When Core Visual Function was analyzed alone, each 10-point decrease was associated with higher odds of depression-screen positivity, with an odds ratio of 1.69 and a 95% confidence interval of 1.38–2.08. After the two scores were entered together, however, the Core Visual Function estimate was attenuated to an odds ratio of 1.06, with a 95% confidence interval of 0.81–1.38, and was no longer statistically significant. The Psychosocial score remained associated, with an odds ratio of 1.62 and a 95% confidence interval of 1.30–2.02. A sensitivity analysis that removed the Mental Health subscale from the Psychosocial score produced a similar pattern. These were conceptually defined questionnaire summaries, not factors derived through a formal analysis in this cohort, so the results should not be interpreted as proof of a causal pathway.

The findings persisted when the researchers used the higher HADS-D threshold of 15 or more. In that sensitivity analysis, each 10-point decrease in the NEI VFQ-25 composite score was associated with an odds ratio of 1.53 for the higher symptom-burden outcome, with a 95% confidence interval of 1.35–1.74 and p<0.001. The authors also noted that the questionnaire includes dimensions such as independence, social participation, role performance and emotional well-being. Lower scores may therefore represent the wider consequences of living with visual impairment, not simply the amount of impairment measured by acuity or visual-field testing. The data support treating a patient-reported measure as complementary to objective examination, rather than as a replacement for it.

Several features limit the conclusions. The cross-sectional design cannot establish whether visual difficulties preceded depressive symptoms, whether the two evolved together or whether other factors influenced both. Both the NEI VFQ-25 and HADS-D are self-reported, leaving the possibility of shared reporting tendencies or common-method bias. HADS-D is a screening instrument rather than a structured psychiatric interview, and recruitment from tertiary centers may have influenced symptom prevalence. Eleven otherwise eligible participants were excluded because they were taking antidepressant or anxiolytic medication, which may have limited generalizability to treated patients. The analysis also did not account for clustering by recruiting center and lacked standardized Expanded Disability Status Scale scores, leaving possible confounding from disability, fatigue, pain, cognition, disease activity and treatment status.

Within those boundaries, the study offers a detailed picture of how visual health and depressive symptoms co-occur in MS and NMOSD. The striking result is not that blindness alone failed to matter, but that a binary measure of severe visual impairment captured less of the association with depressive symptoms than a questionnaire asking patients about the practical and psychosocial effects of seeing. Across the full sample, lowest vision-related quality of life was common alongside high symptom burden; at the other end, better reported vision-related functioning was common alongside fewer positive depression screens. The authors conclude that validated patient-reported outcomes may add clinically useful information about disease burden that is not fully visible in standard visual classifications, while longitudinal studies with standardized neuro-ophthalmic, neurological, treatment and psychiatric assessments are needed to determine how these relationships unfold over time.

Subject of Research: Association of vision-related quality of life with depressive symptom burden in adults with MS and NMOSD

Article Title: Association between psychosocial consequences of visual dysfunction and depressive symptom burden in multiple sclerosis and neuromyelitis optica spectrum disorder

Article References: Zhang, G., Wang, H., Sun, X., Liu, Y., Li, C., Yang, X., Guo, J., Meng, H., Ye, X., Zhong, D., Wang, H., He, Y., Sun, D., Zhu, J., Ge, L., Zhou, Q., Chiu, K., & Zhao, R. (2026). Association between psychosocial consequences of visual dysfunction and depressive symptom burden in multiple sclerosis and neuromyelitis optica spectrum disorder. Frontiers in Neurology, 17, Article 1975600. https://doi.org/10.3389/fneur.2026.1975600

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1975600

Keywords: multiple sclerosis, neuromyelitis optica spectrum disorder, depressive symptoms, vision-related quality of life, visual impairment, patient-reported outcomes, NEI VFQ-25, HADS-D, psychosocial functioning, Association, between, psychosocial