Prevention and treatment of cognitive impairments in the aging population have become priorities for stakeholders in health care research around the globe. In Canada, the Senate Report on the need for a Canadian Dementia Strategy was released in 2017 [1], and resulting recommendations started to be made public in 2019 [2]. For example, the importance of vision and hearing is mentioned in order to promote and enable early diagnosis, with the goals of increasing quality of life and improving social connectedness, belonging, and purpose. These initiatives specifically refer to the importance of vision and hearing health research. Internationally, recent publications in The Lancet [3,4] described dementia as the most challenging threat to population health in our century and pointed out that hearing loss may be the largest potentially modifiable risk factor for cognitive impairments. While low vision rehabilitation (eg, magnification strategies and reading rehabilitation) may serve as a potential prevention strategy for cognitive decline, this possibility is notably absent from this review. There is simply a lack of evidence from longitudinal or intervention studies regarding the links between visual loss and cognitive impairments [5]. Some researchers have indicated that improvement of vision through cataract surgery has resulted in improved scores on attention, orientation, memory, language, visual perceptual, and visuospatial skills as measured by the Addenbrooke’s Cognitive Examination or the Revised Hasegawa’s dementia scale [6,7]; however, improvements in global scores across these cognitive domains did not replicate when using the Telephone Interview of Cognitive Status [8,9]. This may in part be explained by the fact that some of these researchers chose to utilize cognitive measures [10] that contain items that require vision without any adaptations to confirm the visibility of these test items. Others utilized nonvisual measures of cognition [11,12], thereby avoiding the possible influence of improved vision masking as improved cognition on visual test items. Apart from medical interventions such as cataract surgery, the main technique to improve visual input for reading in the presence of low vision has been magnification; however, vision rehabilitation has never been systematically evaluated using cognitive outcome measures.

There is a growing body of evidence linking age-related eye diseases such as age-related macular degeneration (AMD) with changes in cognitive functioning and cognitive impairments due to Alzheimer’s disease (AD) [13-17]. The prevalences of AMD and AD increase with increasing age [18-21]; both conditions share many risk factors (eg, smoking, obesity, age, and unhealthy diet [22]), yet their comorbidity is higher than what would be expected if they were independent of each other [23-25]. The anatomical changes observed in both AMD (eg, drusen development in the retina) and AD (eg, formation of plaques in the brain) are possible symptoms of a common underlying disease mechanism within the central nervous system. Specifically, the buildup of beta amyloid found in plaques and drusen could indicate a common pathogenesis for both diseases [26-28]. Furthermore, declines in visual and cognitive functioning are correlated [29-31]; for example, higher cognitive function scores on the Mini-Mental State Examination (MMSE) were associated with better best-corrected visual acuity [29], when the visual items on this cognitive screening test are either included or excluded [31]. Declining scores on a modified and expanded version of the MMSE correlated with declines in visual acuity, contrast sensitivity, and stereo acuity impairments over 9 years in a population-based study of community-dwelling, highly physically functioning older adults [30]. It is less clear, however, whether these declines can be modified with vision interventions, given that the MMSE is a screening tool covering multiple aspects of cognition but none at great depth. There is some evidence that individuals who read frequently are at reduced risk of developing cognitive impairments [32]. In addition, it is not surprising that there is overlap between the behavioral aspects of AMD and AD; for example, social disengagement and functional impairments in activities of daily living may be present as a result of either or both conditions.

The Framework for Understanding Effortful Listening has been used to illustrate how listeners with a hearing impairment allocate more cognitive resources in challenging conditions (such as when an individual has hearing loss or there is noise) [33]. Similarly, reading becomes more effortful in the presence of central visual impairment or when visual input is suboptimal. For example, AMD generally causes a reduction in visual acuity, making it necessary for persons living with AMD to utilize magnification in order to read with peripheral retina that is still intact [34,35]. Given that fixation in the periphery is less stable [36], more concentration and effort are required to read magnified text with a peripheral retinal locus [37,38]. Effortful viewing or reading explains why the presence of central visual impairment (eg, the presence of central scotoma and resulting drop in visual acuity) has repeatedly been correlated with reduced reading speed as low as 20-40 words per minute [39], whereby poorer fixation stability has been associated with slower reading speeds [40]. Increased cognitive load leads to a decrease in the attentional visual window [41]. As individuals with vision loss find reading more effortful, thus experiencing a higher cognitive load, their attentional window should shrink. Processing of visual information in the retinal periphery (as is necessary in persons living with AMD) is slower and not as efficient as in the macular region [42], in both younger and older observers [43]. In addition, low vision affects eye movements during reading, thereby further shrinking the perceptual window where letters are processed, likely due to the increase in cognitive demand [44]. Importantly, when some cognitive resources are diverted to this reading effort, remaining cognitive resources may be insufficient for readers to rapidly or accurately process (comprehend or remember) the information that was seen or read.

Difficulty while reading is the most common functional complaint in persons with low vision, and improving the ability to read is often the main purpose of vision rehabilitation interventions [45]. The effectiveness of low vision rehabilitation has been demonstrated repeatedly [46,47], specifically for reading [48]. A systematic review confirmed that there is strong evidence that low vision reading rehabilitation services improve reading ability overall [49]. Both magnification devices (eg, handheld magnifiers, closed-circuit televisions, or zoom functions on an iPad [50,51]) and large print appear to be equally effective [52]. Increasing reading performance (eg, increasing reading speed at decreased print size or improving comprehension) is frequently the main target for improvement during rehabilitation [49,53]. It has been used as the primary outcome measure in recent clinical trials demonstrating the effectiveness of low vision treatments [49,53]. Individuals living with low vision can be trained to use either closed-circuit television video magnifiers or mechanical magnification devices (eg, handheld magnifiers, telescopes) to help improve reading performance by up to 200% [54]. One central underlying goal of low vision rehabilitation is to reduce the effort that is required to accomplish visual tasks in the presence of a visual impairment [55]. Measuring this effort (or its reduction) can be done subjectively by asking participants whether they perceive less effort during completion of the task. However, there are also a variety of observable variables that can be used as indicators of reduced reading effort. They include improved reading speed (eg, reading becomes faster as it becomes easier) and improved reading comprehension (eg, less effort liberates cognitive resources for processing and retention [33]). Interestingly, the findings for reading comprehension in the presence of low vision are mixed [56] insofar as reading speed (as an indicator of effort), scotoma size, and visual acuity all influence comprehension. However, most studies on low vision reading have small sample sizes and are underpowered. Therefore, it is hard to control for factors such as scotoma size or acuity impairment, a problem we hope to overcome in our study by recruiting a larger number of participants than is generally the case in studies on low vision and reading.

Reading is a complex process that involves bottom-up visual processing to enable grapheme (ie, letter) recognition that in turn enables grapheme-to-phoneme conversion, leading to word recognition and the identification of morphosemantic, syntactic, and pragmatic features of lexical items that are ultimately used in the comprehension of sentences and discourse such as stories [57]. Because reading is subserved by a number of cognitive processes, including attention, long-term memory, and working memory, there is a symbiotic relationship between reading and cognitive processing as has been documented extensively in the psycholinguistic/neurolinguistic and brain imaging literature [58,59]. Notably, a number of studies has shown that engaging in high-level cognitive activities, such as reading text, appears to preserve cognition in aging adults. The frequency of participation in activities that are mentally stimulating, such as reading, is associated with lower risk of incident AD [59-61]. Researchers have also linked reading and engaging in higher-level cognitive activities with increased cognitive reserve, that in turn is associated with more tolerance of AD pathology and stimulation of brain plasticity [62-65].

We previously found a positive correlation between reading speed and higher scores on the Montreal Cognitive Assessment (MoCA) in persons with AMD [66,67]. Given the information reviewed in the previous sections, the logical direction for our investigation into the functional connection between low vision and cognition is to examine the possible effects of reading rehabilitation on the cognitive abilities of individuals undergoing low vision rehabilitation. In order to disentangle the effects of visual and cognitive impairments in older adults, members of this research team have adapted cognitive tests so they can be administered to individuals with low vision [68,69]. In parallel, we are also in the process of developing a vision-screening test that can be administered to individuals with various levels of cognitive impairment [70,71]. This investigation will guide our future research efforts into the improvement of service provision in low vision rehabilitation for the purpose of increasing the independence and cognitive health of older adults living with low vision.

The overall goal of the study is to demonstrate that low vision rehabilitation will improve reading, which, in turn, will improve cognition.

The primary objective is to measure global changes in reading ability and cognitive functioning before and after 6 months and 12 months of reading rehabilitation. Hypothesis 1 is that improvements relative to pretreatment performance will be observed at 6 months and maintained at 12 months for participants with vision loss only and less so for those with dual sensory impairment (DSI). Control participants are expected to have stable performance and outperform both groups who receive reading rehabilitation. The global reading outcomes are reading speed, reading comprehension, and subjective perception of reading effort.

The global measure of cognition is the MoCA.

The secondary objective is to explore changes in more specific subdomains of reading ability and cognitive functioning. Hypothesis 2 is that improvements relative to pretreatment performance will be observed at 6 months and maintained at 12 months for participants with vision loss only and less so for those with DSI. Control participants are expected to have stable performance and outperform both groups who receive reading rehabilitation. The specific reading outcomes are reading acuity, critical print size, reading accuracy, and subjective reports of changes in reading habits.

The specific measures of cognition are episodic learning; memory encoding, storage, and retrieval; attention, speed, and mental flexibility; and semantic fluency.

Hypothesis 3 is that improved reading behavior will be correlated with improved cognitive functioning across all 3 groups, with the strongest relationship being in the AMD-only group. Such correlations will be observed across the global as well as the specific measures.

The tertiary objective is to explore factors that may influence the cognitive benefits of reading rehabilitation. Hypothesis 4 is that individual differences in the association between improved reading behavior and cognitive functioning may be related to participant characteristics such as demographic variables, hearing impairment severity, and mental health.

We present this protocol following the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines [72]. An overview of the trial registration data is provided in Table 1. We decided on a 3-arm, quasiexperimental, repeated-measures design (nonrandomized, pre-post intervention study) [73], whereby participants act as their own comparison across time points. Given the prevalence of hearing loss among older adults, we decided to include both patients with AMD who present with normal hearing [74,75] and those who experience both vision and hearing loss (DSI). We decided to add an age-matched control group with healthy vision and hearing in order to observe the size of possible practice effects and record general variability in the measures, given the expected advanced age of our participants. The rehabilitation center partners on this study generally provide reading rehabilitation within 3 months of the initial optometric assessment. Therefore, we chose 6 and 12 months as suitable follow-up time points. After 6 months, the initial interventions will be completed, and participants will have had 3 months to engage in reading postintervention. Overall, a design with 3 groups (AMD-only, DSI, comparison) x 3 test times (preintervention, 6 months, 12 months) is planned to allow us to examine whether cognitive performance will change over time and if the degree of change and the final performance on outcome measures will differ across groups.

The study will be conducted at 2 partnering vision rehabilitation centers in the Montreal area, the Centre de réadaptation Lethbridge-Layton-Mackay du Centre intégré universitaire de santé et de services sociaux du Centre-Ouest- de-l’Île-de-Montréal and the Institut Nazareth et Louis-Braille du Centre intégré de santé et de services sociaux de la Montérégie-Centre. Both sites are part of the Center for Interdisciplinary Rehabilitation Research of Greater Montreal and provide government-funded rehabilitation services, free of charge for eligible residents of Quebec, Canada.

Rehabilitation services at the partner centers will deliver the low vision evaluation and reading rehabilitation intervention, as regulated by the Quebec Ministry of Health. Typically, within 3 months of their initial low vision rehabilitation assessment, individuals will receive most of the recommended interventions. The clinical staff and rehabilitation professionals at either rehabilitation center decide which components of the full complement of services are suited for each participant. The 2 centers offer comparable services, including the provision of assistive devices and services, as regulated by the Quebec Health Insurance Program. These services are similar to those provided within the Blind Rehab Centers of the Veterans Affairs service in the United States [46,47]. They include, but are not limited to, a full optometric exam to determine functional vision, including refraction and the prescription of appropriate near and distance glasses and optical devices; an assessment by a low vision therapist or occupational therapist to establish the participant’s functional priorities and rehabilitation goals; and the provision of handheld optical magnification devices, electronic nonoptical magnification devices (eg, portable or tabletop closed-circuit TVs), or computer software for screen content magnification (eg, ZoomText). All devices are provided free of cost for the individual and with appropriate training and follow-up sessions at home within 3 months of the initial intervention, as required. The rehabilitation professionals may perform a systematic lighting assessment in the participants’ homes [81-83] and may make specific lighting recommendations that are intended to improve their reading ability. In addition, participants will have access to referral services such as orientation and mobility training (for independent travel) or registration with an adapted adult day center for individuals with sensory loss [80]. These centers provide access to psychosocial services, counsellors, social workers, or other mental health professionals. The provision of assistive devices for magnification and reading is generally linked to a follow-up visit in the individual’s home 3 weeks after the initial rehabilitation appointment. At this time, rehabilitation professionals observe the use of the devices and strategies in the environment where the participant lives. It is at this point that the professionals together with the participants decide whether the devices are useful and are assigned as a permanent loan. Should additional needs emerge, the participant is at liberty to contact the rehabilitation center at any time to initiate a new service episode. The rehabilitation professionals record all aspects of this intervention in the rehabilitation files, to which the research team will have access during and at the end of the 12-month study period.

Figure 1 illustrates the structure and timeline of the study. After intake, but before intervention begins, participants will complete the initial administration of all assessment measures. At 6 months and 12 months after the initial rehabilitation appointment, the research assistants will meet with each participant (at home for rehabilitation participants, in the lab with control participants) to repeat the administration of all measures. After the 12-month follow-up, data will be extracted from the rehabilitation charts detailing all rehabilitation interventions. A detailed overview of this process is presented in Figure 2.

The power analysis for this study is centered around the MoCA, the primary outcome measure of cognition. We are only aware of 1 study reporting MoCA scores (in its blind version) after a vision rehabilitation intervention, whereby 21 older adults participated in a program evaluation [80]. This study, however, contained reading rehabilitation in addition to various other interventions (eg, orientation and mobility, social engagement in an adapted day center); therefore, the reported effect size of d=0.794 (η² equivalent=0.136) after 6 months likely overestimates the outcomes we would expect in our study design. To be conservative, we halved this expected effect size for our power estimate. Please note that our control group is primarily used to examine possible practice effects and to observe variability on the measures in a cohort with potentially high age. We based our power analysis on a 3x3 (group x time point) within-between design for analysis of variance. Using gPower [113], a minimum sample size of n=81 will allow us to detect small effect sizes of .20 or greater, with an alpha of .05 and a desired power of .95. We also expect an attrition rate of around 10% at each time point based on our previous research with older, visually impaired adults [114,115]. We plan to include covariates in the analyses that will be identified from the analyses linked to the tertiary objective exploring factors that may influence the cognitive benefits of reading rehabilitation. The sample will need to be sufficiently large for these variables to be normally distributed in each group. Therefore, a sample goal of 225 (150 participants with sensory loss, 75 controls) should be sufficient for our protocol. This goal is feasible given an average of 3000 new referrals to rehabilitation in Montreal per year.

Our partner organizations will conduct recruitment for participants in the intervention groups (AMD-only and DSI) as part of their responsibility on the research team. We will recruit control participants through staff of the partner organizations, from family members of participants, as well as through open advertisements in public media and communications to senior groups. In addition, we have access to the Banque des participants of the Centre de recherche institut universitaire de gériatrie de Montreal (n>1000). Using this database, we can match our control participants to the intervention participants on variables such as age (by decade), presenting gender, education, and other potential variables of interest, such as MoCA score. Our pilot study [66,67] indicated that we will be able to recruit at least 3 AMD or DSI participants per week, making the recruitment target goal feasible to achieve within 18 months.

Data will be collected by 4 trained research assistants (2 teams of 2 people). The composition of the pairs who work together to collect the data will continuously rotate in order to facilitate harmonization among the teams and ensure consistency and accuracy in the administration of the measures. It will not be possible to conceal group membership from the research assistants because the administration of some of the measures will determine if participants are considered to have a hearing impairment. In addition, control participants will be identifiable because they are not visually impaired. However, the teams will be independent of the administration of the intervention, which will be provided by rehabilitation professionals outside of the research team. Administration of the measures will be audio-recorded throughout several of the tasks to ensure the highest level of precision and quality control for data entry and to improve testing efficiency and reduce testing time. We offer participants the choice of coming to the lab space at either rehabilitation center or receiving a home visit for data collection. Therefore, we developed our protocol to be easily portable such that the necessary measures and materials can be transported wherever needed by the research team using a wheeled carry-on suitcase.

In line with the data management requirements of the Institutional Review Board that approved this protocol and the recommendations by Michener for data management [116], all data will be anonymized and entered in a central password-protected Excel file that will be stored on the encrypted server of the Université de Montréal. Data integrity and quality control will be enforced through random spot-checking by a second member of the team. The data will be accessible for the period of the study approval, plus an additional 5 years thereafter, at which point all electronic and paper files will be destroyed.

The Comité d'éthique de la recherche of the Centre de recherché interdisciplinaire en readaptation du Montreal metropolitain (CRIR#1284-1217) has provided institutional review board approval. This committee is responsible for research protocols involving recruitment from and testing on the sites of the local clinical partners for this study. The principal investigator will obtain renewal of the ethics approval annually. We submitted all aspects of the study design and data analysis as a stage 1 pre-registered report (ClinicalTrials.gov ID: NCT04276610; February 19, 2020).

Any necessary changes to the protocol (eg, safety and security measures necessary due to COVID-19) are planned in collaboration with the partner sites in order to adhere to the requirements of the Quebec Ministry of Health. These updates and amendments are the submitted for approval to the Comité d'éthique de la recherche of the Centre de recherché interdisciplinaire en readaptation du Montreal metropolitain.

One of the 4 trained research assistants will obtain written informed consent at the first in-person meeting, either in the home or at the lab in the partner sites. Consent forms are available in large print in both English and French. Participants are free to abandon the study at any time, and this choice will not affect their care at the study partner sites.

All electronic study data will be stored in a password-protected Excel file on an encrypted server at the Université de Montréal and can only be accessed through a password-protected computer in a locked lab space. All records that contain personal identifiers, such as consent forms and questionnaires, will be stored separately from study records. Each participant file will be identified by code number. Study data will not be released to anyone outside the research team.

Data access is limited to the members of the research team and their trainees. In order to ensure confidentiality, the data that will be available to the research team members will only contain deidentified information.

All participants will be rehabilitation clients of 1 of the 2 partner sites. Therefore, they will have access to all available service and referral pathways that are part of the care offer. Should the research team suspect or a participant express a potential need for services or referral (eg, counseling), the research team will connect the participant with their respective rehabilitation file manager at the partner site for follow-up.

All members of the research team have the right to access and analyze all data, for the purpose of dissemination. The principal investigator will oversee dissemination activities in order to ensure that team members with the necessary topic expertise are involved in each aspect of dissemination. Study results will be made available in open-access format whenever possible and will be presented in formats that are accessible to researchers, clinicians, policy makers, members of the public, participants, and all other stakeholders. The study also distinguishes itself with its multifaceted approach to integrated knowledge translation and dissemination. The clinical partners are represented on the research team, directly influencing the study and maintaining the clinical relevance of the study goals. They are part of a local network of rehabilitation research sites, the Center for Interdisciplinary Rehabilitation Research of Greater Montreal [125], providing the research team a unique opportunity to disseminate their knowledge translation activities that are planned to occur after the study is completed. As an extension of this network, established collaborations with the Sense-Cog group in Europe [126], Envision University in the United States [127], and Canadian Consortium on Neurodegeneration in Aging knowledge translation team [128] provide access to sensory-cognitive-specific knowledge translation opportunities that will facilitate the distribution and implementation of our findings. Furthermore, Dr. Swenor, as a visually impaired scientist [129] and team member, brings a unique and important knowledge translation, equity, and inclusion perspective to the study. Finally, all collaborators and partners will participate in the development of a Canadian Institutes of Health Research Café scientifique [130] at study completion. This type of public, open science dissemination event is designed to disseminate the results at a clinical level and inform stakeholders as well as participants and their friends and family of the study outcomes