Functional Threshold of Contrast Sensitivity Provides Benchmark for Visual Disability

A contrast sensitivity (CS) threshold that was linked to self-reported visual disability (SRVD) was identified in older adults aged 65 years and older using Medicare, according to a study published in JAMA Ophthalmology.¹ The identification of this threshold can help provide a benchmark in future research.

CS is the measurement used to assess how an individual differentiates shades and patterns, which can affect night driving and increase the risk of a fall if an individual cannot differentiate the colors of pavements.² CS can be a predictor of cognitive decline, self-reported visual acuity, and fall risk, but a consensus on the threshold for impaired CS that is clinically meaningful has not been reached. This study aimed to “generate a CS threshold at which impairment is no longer a simple deviation from the population mean but instead translates into tangible impairment noted by the individual.”¹

The researchers used the National Health and Aging Trends Study (NHATS) to collect data from adults aged 65 years and older who were using Medicare. Data from the NHATS spanned from 2011 through 2022 and included an oversample of Black and Hispanic participants to ensure representation of all groups. The current study used data from 2022 and 2023 from rounds 12 and 13 of the NHATS. Participants were included if they were community dwelling, were aged 65 years or older, and had a completed visual function test in both rounds of the study. Participants were excluded if they had missing information on any study variables.

SRVD was defined as participants who reported having difficulty or blindness to being able to see the TV across the room, reading newspaper print, or recognizing someone from across the street, with only 1 of these requirements needing to be met to classify the older adult as having SRVD. CS was measured in both eyes while the participant wore any corrective lenses. CS was measured with a tablet-based eBook and scored on logarithmic scales with a lower score indicating a worse CS.

There were 4475 participants who were included in this study who were primarily aged between 65 and 79 years (79.4%). Most of the participants were women (53.9%) and non-Hispanic White (80.1%).

A total of 6.7% of the participants reported SRVD in round 12 and round 13. The mean distance visual acuity (VA) was 0.10 logMAR, the mean near VA was 0.18 logMAR, and CS was 1.72 logCS in round 12. Round 13 saw a mean distance VA of 0.10 logMAR, near VA of 0.19 logMAR, and CS of 1.71 logCS.

Those with SRVD were more likely to be women (66.5% vs 53.0%) compared with those without SRVD. Those with SRVD were also more likely to have less than a high school education (22.7% vs 9.1%) and more likely to be in the lowest income quartile (38.0% vs 16.4%). Participants with SRVD had a higher burden of comorbidities (32.5% vs 18.1%) and frailty (37.9% vs 17.3%).

Between rounds 12 and 13, 3.8% of the study cohort reported persistent SRVD, 4.4% developed incident SRVD, and 4.1% reported SRVD in round 12 but not round 13. Those with persistent SRVD had a mean CS of 1.39 logCS. The mean CS of those with SRVD in round 12 but not 13 was 1.59 logCS and these participants gained 0.06 logCS in round 13. Those who had incident SRVD had a decline in CS from 1.62 logCS to 1.55 logCS. Patients without SRVD had a mean CS of 1.74 logCS. A multiple logistic regression model found that every 0.1 logCS loss at baseline was associated with higher odds of developing incident SRVD (OR, 1.12; 95% CI, 1.08-1.16).

There were some limitations to this study. There were only 2 time points included in this study where logCS was measured, which limits long-term analysis. SRVD is also subjective and can fluctuate day-to-day, which was not accounted for in this study. Uncorrected refractive error could have affected the measurements of CS and SRVD. Data on ocular health were not available for the participants. Generalizability is limited due to all participants being community-dwelling adults with preserved visual function. The threshold for specific ocular diseases was not assessed. Future studies will need to perform similar analyses across more racial, socioeconomic, and clinical populations to confirm the results.

The authors concluded that “a CS level of 1.60 logCS marked the point at which older adults tended to begin experiencing perceptible visual disability, providing a patient-centered benchmark for future research.” The authors hypothesized that incorporating this threshold into clinical practice can help to identify signals of visual decline and help with intervention timing for those who need it.

References

1. Xu S, Nguyen M, Zhou Y, Hu M, Ehrlich, De Lott LB. A clinically relevant threshold of impaired contrast sensitivity among older US adults. JAMA Ophthalmol. Published online May 28, 2026. doi:10.1001/jamaophthalmol.2026.1570
2. Kelley S. Contrast sensitivity testing. All About Vision. Updated November 19, 2025. Accessed May 29, 2026. https://www.allaboutvision.com/eye-care/eye-exam/types/contrast-sensitivity/