Over 12.8 million adults in the United States report vision difficulty (ie, blind/low vision [B/LV]) [1]. This number is projected to increase as the population ages, in addition to an increase in chronic conditions that lead to being B/LV. For example, recent estimates indicate that the number of people who are legally B/LV over the age of 40 years in the United States is expected to double to more than 8 million by 2050, and another 16.4 million are expected to have difficulty seeing with correctable vision loss [2]. People who are B/LV are more likely than those who are not to have chronic conditions including diabetes, heart disease, hypertension, injuries, depression, and premature death [3-8]. The etiologies of B/LV are numerous and vary widely ranging from genetic, congenital, or acquired conditions, and may be stable or progressive. Patients who are B/LV require a comprehensive care coordination approach, which includes accessible health care facilities and health care providers with B/LV expertise.

Disability civil rights law in the United States mandates accessible health care environments for people with disabilities, including people who are B/LV. The primary legislation addressing accessibility in health care settings is the Americans With Disabilities Act (ADA) of 1990 [9]. The ADA prohibits discrimination against individuals with disabilities in all areas of public life, including employment, transportation, public accommodations, and access to state and local government services. Regarding health care specifically, the ADA requires health care providers, facilities, and programs to ensure equal access for individuals with disabilities. This includes making reasonable modifications to policies, practices, and procedures to accommodate the needs of people with disabilities, providing effective communication, and removing architectural and other barriers that may prevent access. In addition to the ADA, other laws and regulations may also impact accessibility in health care settings, such as Section 504 of the Rehabilitation Act of 1973 and the Patient Protection and Affordable Care Act (ACA) of 2010. These laws further emphasize the importance of ensuring equal access to health care for individuals with disabilities. More recently, the ACA’s Section 5307 mandates the U.S. Department of Health and Human Services to collaborate with external organizations to develop, evaluate, and disseminate curricula to address “cultural competency, prevention, public health proficiency, reducing health disparities, and aptitude for working with individuals with disabilities” [10].

Despite these mandates, patients who are B/LV experience disability-related accessibility barriers in health care settings [11-13]. Among these include inaccessible information, lack of communication support, navigational/wayfinding challenges, inaccessible medical equipment, digital accessibility issues, and attitudinal barriers. Print materials such as forms, brochures, and instructions are often not available in accessible formats, such as braille, large print, or electronic text, making it challenging for B/LV patients to access important health care information independently. Moreover, B/LV patients may encounter challenges in communicating with health care providers if there are no provisions for alternative communication methods such as screen readers for electronic health records, tactile sign language interpreters (for people who are DeafBlind), large print materials or audio descriptions of visual information and instructions [12]. In addition, patients who are B/LV may also encounter attitudinal barriers from health care providers or staff who may have misconceptions or lack awareness (due to lack of training) about the capabilities and needs of B/LV individuals, leading to stigma, discrimination, or inadequate care [12,13].

B/LV patients often encounter various architectural barriers in health care settings that can hinder their ability to navigate independently and access health care services effectively [12,14]. Architectural features such as narrow doorways, high thresholds, stairs without handrails or tactile indicators, and uneven flooring can pose significant challenges for B/LV individuals to navigate safely. Health care facilities with complex layouts, multiple corridors, and interconnected buildings can be particularly challenging for individuals who are B/LV to navigate independently without clear signage and wayfinding cues. This can make it challenging for B/LV individuals to find their way to reception areas, exam rooms, or other health care services [14]. Addressing these navigational barriers is necessary to ensure accessibility (and compliance with disability civil rights law) and requires addressing the environment through which a B/LV patient moves.

The overall aim of this study was to develop a program to improve navigational access for patients who are B/LV by developing and evaluating a human guide training for staff and volunteers at a large academic medical center in southeast Michigan. Specifically, our aims were to (1) develop and implement an in-person awareness and skill-building training for hospital staff and volunteers, teaching how to provide Human Guide technique; (2) adapt the in-person Human Guide training for electronic delivery via an asynchronous, web-based learning module; and (3) conduct a process evaluation of the in-person training and web-based (asynchronous video) learning module.

As described in the forthcoming sections, the in-person training was not implemented due to infection prevention precautions (related to COVID-19). Information on the design of the in-person training is provided to facilitate implementation by other health care organizations.

Michigan Medicine is an academic medical center operated by the University of Michigan Medical School. Located in southeast Michigan, Michigan Medicine operates a level 1 trauma center emergency department for adults and children, and operates an adult and children’s hospital, and specialty centers in ophthalmology, oncology, and cardiovascular medicine.

All activities involving human participants reported in this study were reviewed by the Medical Institutional Review Board at the University of Michigan and deemed quality improvement activities (HUM00208772). Therefore, the study was exempt from ethics review.

A needs assessment was conducted through two methods: (1) cohort discovery and (2) feedback from subject matter experts. These methods were chosen based on recommendations from the Centers for Medicare and Medicaid Services [19,20] to identify the number of people who are B/LV and identify barriers at points of contact throughout the health system.

This study was designed as a process evaluation using methods common in health promotion program evaluation [27]. Process evaluations are a type of formative evaluation used to assess the quality and delivery of program activities and components. Although process evaluations can include a variety of measures, most process evaluations include measures of reach (ie, how many people received the program) and early indicators if the program is working as intended (ie, early impact indicators) [27].

In our process evaluation, we included four measures: (1) reach, (2) knowledge, (3) behavioral capability, and (4) satisfaction. Reach was measured based on the number of people who accessed and completed the training. Knowledge was operationalized through a 5-item program team-developed knowledge questionnaire focused on aspects included in the training (see Table 2). Participants were asked to respond to these knowledge questions as being true, false, or don’t know. Incorrect and “don’t know” responses were recoded as incorrect. Behavioral capability, or self-reported understanding of having the skill needed to complete a task, was asked through three items focused on (1) providing clear directions, (2) providing a human guide technique, and (3) providing a human guide technique with a service animal. Satisfaction was operationalized through three questions: (1) relevance to the organization, (2) overall satisfaction, and (3) recommending to colleagues. Behavioral capability and satisfaction questions were responded to on a 5-point Likert-type scale (ie, strongly disagree, disagree, neutral, agree, and strongly agree), with a score of 5 being “strongly agree.” Individual questions are provided in the Results section.

Data were analyzed in the R statistical environment using JASP (Jeffrey’s Amazing Statistics Program). We summarize the reach and knowledge questions using frequencies and percentages. We calculated means to describe behavioral capability and satisfaction. Pre- and posttest scores for behavioral capability were assessed for change using paired samples t tests with Cohen d effect sizes.

There are inadequate published definitions of diagnostic codes related to capturing patients who are B/LV. We used ICD-9-CM and ICD-10-CM diagnostic codes used by Ratakonda and colleagues [28] when studying potentially preventable hospitalizations among people with sensory disabilities. We identified that there were 55,159 B/LV patients who had an encounter in Michigan Medicine during the 2 years prior to the program (Table 3).

Subject matter experts agreed that navigating the health care system’s clinics and hospitals was taxing on patients who are B/LV. The timeliness of addressing this issue was underscored by experts as the Office of Patient Experience recently reported there had been patients arriving late to appointments, despite having been on the premises before the appointment start-time, due to challenges visually navigating the environment. In these cases, not all patients were able to be worked back into the clinic’s schedule leading to a delay in receiving care. In one instance, a complaint was escalated and social workers became involved to better support the patient.

In total, 87 people completed the web-based learning module and were included in the process evaluation dataset. These participants took the training between February 21, 2023, and December 31, 2023. The majority of these trainees were volunteers at Michigan Medicine (45/87, 52%). Over one-fourth (25/87, 29%) of trainees were unaffiliated with Michigan Medicine or the University of Michigan (eg, some trainees were from other states). Less than one-third of participants (24/87, 28%) had previously completed a human guide training. The most common advertising sources for learning about the training were from the Office of Patient Experience (24/87, 28%) and recommendations from friends or colleagues (24/87, 28%).

Completing the human guide training required scoring a 100% on the 5-item knowledge quiz; therefore, we did not calculate pre- and posttest score changes. At pretest, however, the most difficult question was “If a person identifies as blind, it is appropriate to call the person vision impaired” with almost 65% (55/85) of participants scoring incorrectly (see Table 4). This contrasted to knowledge related to not making noises to distract service dogs (correct: 79/84, 94%). There was a significant increase in self-reported behavioral capability of (1) providing clear directions, (2) providing a human guide (without a service dog), and (3) providing a human guide with a service dog. Effect sizes of the mean differences were high (ds>0.9) but expected given the immediate posttest nature of the survey. Rainfall plots of pre- and posttraining behavioral capability show that while there was large variability in pretest responses, the responses at posttest clustered around 4 (“agree”) and 5 (“strongly agree”; see Figure 2).

Trainees reported being satisfied with the training (mean 4.5, SD 0.57) and would recommend it to their colleagues.

Participants were asked to provide feedback on how to improve the web-based learning module; 16 participants provided free-text comments. Although some of these comments were positive, most responses provided critical feedback about trainees’ expectations and needs. These needs were well aligned with the goals of the in-person training, including more demonstration, enactive mastery experiences, and simulation in clinical settings.

Trainees also indicated concern with navigating through doorways.

Other comments indicated that the training was “informative, but also rather general.” This trainee further indicated that additional practice would be needed for proficiency. Other trainees commented that they wanted to be asked additional knowledge questions related to human guide to ensure they understood the content. Lastly, three comments focused on platform or technology issues, including the training platform not allowing full screen for the videos and also a captioning error which was fixed.

Patients who are B/LV face obstacles in physically navigating health care environments, particularly complex and maze-like hospitals. These obstacles serve as barriers to effectively and efficiently engaging in health care and may implicate health care systems due to regulatory requirements with the ADA and ACA. Therefore, it is necessary to improve accessibility to effectively navigate health care facilities. The aim of this study was to develop a program to improve navigational access for patients who are B/LV by developing and evaluating a Human Guide training for staff and volunteers at a large academic medical center. Specifically, we aimed to (1) develop an in-person training, (2) adapt the in-person training for electronic delivery as a web-based learning module, and (3) conduct a process evaluation of both the in-person training and web-based learning module. As described earlier, we only accomplished Aim 2 and Aim 3. This web-based learning module focused on teaching human guide techniques to improve wayfinding assistance for people who are B/LV. This training was aligned with priorities set by the U.S. Centers for Medicare and Medicaid Services and needs in the local setting—particularly given the priorities to address poor patient experiences due to navigational barriers.

This early process evaluation suggests that the training is accurately targeting the intended skills and knowledge related to human guide technique. This training has also been adopted by the HOPE Ambassador program, improving the skillset of these wayfinding volunteers to meet the needs of people who are B/LV. The program could be improved, however. Results from the evaluation, and the early theory of change from the program, suggest the importance of providing in-person training allowing learners to practice human guide technique and receive just-in-time feedback. This would improve learners’ experience and also aid in cementing knowledge and behavior change.

Improving wayfinding assistance through human guide does not address the root cause of navigational inaccessibility in health care settings. Addressing architectural barriers requires a proactive approach to accessibility in health care facility design and renovation, as well as ongoing efforts to improve wayfinding and navigation assistance for people who are B/LV. This may involve incorporating universal design principles, installing tactile signage and markings, ensuring clear pathways and accessible entrances, providing adequate lighting, training staff to assist patients and visitors who are B/LV, and providing web-based audio descriptions or tactile maps [29,30]. Some additional recommendations are provided in Textbox 1. By removing architectural barriers, health care facilities can create environments that are more inclusive and accessible for people with disabilities, including those who are B/LV.

There are several limitations to our program and program evaluation. The lack of full program implementation due to infection control policies led to potentially missed opportunities to provide just-in-time feedback to trainees on how to improve their human guide technique. This would have been responsive to trainee’s needs, as indicated by the open-ended feedback. Despite this, the web-based learning module was liked by participants and had an associated increase in perceived behavioral capability. Still, we lack medium- and long-term outcome measurement. Collecting medium-term outcomes was too burdensome for this resource-limited training, particularly given the low incidence of patients and visitors needing human guide. In addition, the immediate pretest and posttest design is known to inflate effect size measures. To address these limitations, we recommend that future human guide trainings consider hosting an in-person training, in addition to web-based videos, and also consider longer posttest observations. The evaluation of this training could also be improved by reducing reliance on self-reported knowledge and behavioral capability, and instead use an observation checklist to evaluate trainees’ skills. Orientation and mobility specialists may be particularly well-suited to evaluating trainee skills. This could serve a dual purpose as a data source and a feedback mechanism.

Patients who are B/LV experience barriers to physically navigating health care environments. Human guide is a technique that can be used to assist blind patients navigating the environment. Participants who completed our very brief, web-based video learning module reported improvements in behavioral capability with providing human guide technique to patients who are blind. We recommend the implementation of disability competency trainings in health provider preparation programs, and as required trainings for staff and volunteers working in clinical settings.