Veterans with TBI-related Ocular Injury and Vision Dysfunction: Recommendations for Rehabilitation

Abstract

Significance.

We know the prevalence of traumatic brain injury (TBI) related vision impairment and ocular injury symptoms. Lacking is an understanding of healthcare utilization to treat these symptoms. Utilization knowledge is important to structuring access to treatment, identifying clinical training needs, and providing evidence of the effectiveness of treatment.

Purpose.

This manuscript reports rehabilitation, glasses/contacts, and imaging/photography/ video recommendations made by optometrists and ophthalmologists as part of the VA mandated Performance of Traumatic Brain Injury Specific Ocular Health and Visual Functioning Examination administered to Veterans with TBI at VA polytrauma specialty facilities.

Methods.

Using a retrospective design, natural language processing, descriptive and regression statistics, data were analysed for 2,458 Operation Enduring Freedom/Operation Iraqi Freedom veterans who were administered the mandated exam between 2008 and 2017.

Results.

Of the 2,458 veterans, vision rehabilitation was recommended for 24%, glasses/contacts were recommended for 57% and further imaging/photography/video testing were recommended for 58%. Using key words in the referral, we determined that 37% of veterans were referred to Blind Rehabilitation, 16% to Occupational Therapy, and 3% to Low Vision Clinics. More than 50% of the referrals could have been treated by Blind Rehabilitation, Occupational Therapy, or Low Vision clinics. Rehabilitation referrals were significantly associated with younger age, floaters, photosensitivity, double vision, visual field and balance deficits, dizziness, and difficulty reading. In comparison, prescriptions for glasses and contacts were associated with older age, photosensitivity, blurred vision, decreased visual field and night vision, difficulty reading, and dry eye. Imaging/photography/video testing were associated with floaters, photosensitivity and headache.

Conclusions.

Findings delineate service delivery models available to veterans with TBI-related vision impairment. The challenge these data address is the lack of clear paths from diagnosis of TBI to identification of vision dysfunction deficits to specialized vision rehabilitation, and finally to community reintegration and community based-vision rehabilitation.

Increasingly, vision function-related symptoms are recognized as possible long-term sequelae of traumatic brain injury (TBI), which may not be surprising considering that 70% of the sensory processing capacity of the brain is related to vision.^1–3 Visual sequelae associated with combat and TBI include closed eye and optic nerve injury (blindness, visual field loss, low vision), orbital fracture or eye muscle injury (diplopia, restricted eye movement, strabismus), eyelid injury (disfigurement, exposure, dry eye), brain focal contusion/hemorrhage (central or peripheral visual field loss), and diffuse brain injury (convergence insufficiency, reduced accommodation, fixation instability, impaired saccades and pursuits, nystagmus).⁴ Ocular injuries as a percentage of total injuries in combat have increased from 2% in World Wars I and II to 7% in the Vietnam War and 13% in the Gulf War.^5–6 Between 2001 and 2017, the average annual incidence of eye injury in the military was 15,681, with 304 hospitalized and 298 at high risk of blindness. During that same period there were 4,394 annual TBI cases without injury to the eye but with visual dysfunction. The total cost of TBI-related visual dysfunction is $1.9 billion annually in 2017 U.S. dollars.⁵ Veterans with blast related TBIs report significantly decreased visual quality of life and score the same or worse than other study groups with cataract, thyroid eye disease, or optic neuropathy.⁶

More than 65% of patients with TBI treated in the VA Polytrauma Network System of Care report vision problems.^{1, 7} Recognizing the need for specialized services related to returning veterans with polytrauma, the Veterans Health Administration (VHA) established the Polytrauma/TBI System of Care in 2004 to ensure that returning war veterans with loss of limb and other severe and lasting injuries had access to the best of both modern medicine and integrative holistic therapies. VHA Directive 2008–065, “Performance of Traumatic Brain Injury Specific Ocular Health and Visual Functioning Examinations for Polytrauma Rehabilitation Center Patients,” mandated that all patients with a diagnosis of TBI admitted to one of the five inpatient Level 1 Polytrauma Rehabilitation Centers must have a TBI-specific ocular health and visual functioning examination performed by an optometrist or ophthalmologist.⁸ The mandate provided a template that specified a history of symptoms, a complete clinical examination and patient recommendations and treatment plan.

Since the 2008 Directive, several studies^2,5,6,8–10 have identified the prevalence and symptomology of veterans with TBI-related vision and ocular dysfunction. A 2014 VA Quality Enhancement Research Initiative (QUERI) Evidence-based Synthesis Project¹ investigated TBI incidence and prevalence (13 studies) and the types of visual dysfunction (four studies) among individuals with TBI (veterans, active-duty military, and civilians). Prevalence varied by symptom and study: the prevalence of visual field deficit ranged from 0% to 39%, and convergence dysfunction ranged from 11% to 63%. Variation was attributed to different levels of severity of TBI, setting (inpatient versus outpatient), and protocols that treated only patients with current symptoms. A systematic review and meta-analysis¹¹ of 22 publications used random-effects models to calculate combined prevalence estimates for the outcomes associated with TBI-related vision dysfunction. A systematic review¹⁰ found that accommodative dysfunction was the most frequent (42.8%), followed by convergence insufficiency (36.3%), and visual field loss (18.2%).

The purpose of this study was to understand how the VA’s current system of care emerged and to use lessons learned to create a standardized and successful system of vision care that could be implemented within both the VA and community care. We accomplished this by analyzing data generated by the directive to identify patterns of service delivery, including access to and utilization of health care services by veterans with TBI-related ocular injuries and visual dysfunction. The rationale for the study was trifold. First, knowledge of utilization patterns should provide opportunities to structure services so that care can be available outside of the five VA Polytrauma Rehabilitation Centers, thereby increasing access to care in the communities where many of these veterans now reside. Second, the clinical knowledge base created by clinicians during the post-9/11 era, if not adequately captured, would be lost not only to the next generation of clinicians treating service members in future conflicts, but also to non-specialized clinicians currently treating Operation Enduring Freedom/Operation Iraqi Freedom veterans in community versus specialty center settings. Finally, systematic examination of vision function-related rehabilitation outcomes will provide evidence to amend and update practice guidelines for the management of TBI. The research questions asked (1) What specific recommendations were made by the optometrists and ophthalmologists performing the exams and (2) What were the relationships between the recommendations and the demographic variables, comorbidity, severity of TBI injury, specific visual and ocular symptoms, and geographical access to vision-related services.

METHODS

The study was approved by the Institutional Review Boards of the University of South Florida and the University of Florida and the Research and Development Committees of the James A Haley VA Hospital and the North Florida South Georgia Veterans Health System. Using a retrospective design, 2,458 veterans who had served in the Operation Enduring Freedom or Operation Iraqi Freedom conflicts, had received inpatient or outpatient care at one of the five VA Polytrauma Rehabilitation Centers between January 1, 2008, and December 31, 2017, and had received the mandated evaluation were identified.⁸ All analyses were performed within the secure VA’s Informatics and Computing Infrastructure environment.

The first step in understanding how the mandated evaluation affected the health care utilization of these 2,458 veterans was to identify the recommendations made by the optometrists and ophthalmologists performing the exams. The researchers used natural language processing algorithms to extract recommendations from the text of the evaluation clinical notes. For reliability and validity, two annotators independently identified recommendation key words in the patients’ notes. Disagreements between the annotators were adjudicated by an Optometry expert, e.g., whether prism was an optometry or rehabilitation recommendation or both/either. Agreement statistics were calculated. Annotation and adjudication continued until a .7 correlation agreement between annotators was achieved. A rule-based algorithm was constructed that sorted the keywords (e.g., orientation, o and m, O&M, blind rehab, tint, advanced blind rehab, blind rehabilitation, etc.) into three categories: rehabilitation, glasses/contacts, or imaging/photography/tests. The frequencies of recommendations by category were stratified by the five Polytrauma Rehabilitation Centers to show variation across evaluating facilities.

The second step in understanding how the mandated exam affected utilization was to determine how the recommendation categories were influenced by demographic variables, visual symptoms, comorbidity, severity of TBI injury, geographic access to care, and the evaluating Polytrauma Rehabilitation Center. The visual symptoms were specific to the history section of the mandated evaluation. Comorbidities were calculated based on work conducted by Pugh et al.¹² with the veteran and military TBI population. Glasgow Coma Scale values at the time of injury (abstracted from the Defense and Veterans Eye Injury and Vision Registry, was used as a proxy for severity of TBI¹³. For patients with missing Glasgow Coma Scale scores, severity of TBI was extrapolated from the VA Functional Status Outcomes Database Functional Independence Measure score. For patients with missing Glasgow Coma Scale and Functional Independence Measure scores, the research team conferment severity of TBI in veterans’ individual electronic medical records. Previous studies indicated that geographic distance barriers negatively impact utilization of inpatient and outpatient services, leading to a seven-fold decrease in use of VA services by Operation Enduring Freedom/Operation Iraqi Freedom veterans.¹⁴ Thus, access measured as distance and travel time to the treating facility were included as covariates and used to map the relationship of the veterans’ home zip code at the time the evaluation was performed to the treating facility. Initially, student t-tests and chi-square analyses were conducted to identify demographic and clinical characteristic factors univariately associated with each clinic category. Three logistic regression models (one for each category) using forward stepwise selection were fit to identify factors independently associated (at P <.05).

We further examined the rehabilitation recommendation category to learn more about which VA vision rehabilitation clinics were veterans were referred to. Possible clinics included: Blind Outpatient/Blind Rehabilitation Outpatient Specialist (BROS), Blind Inpatient/ Blind Rehabilitation Center, Vision Impairment Services Team (VIST) Coordinator, Vision Impairment Services Outpatient Rehabilitation (VISOR) & Advanced Blind Rehabilitation, Orthoptics, Advanced Low Vision, Intermediate Low Vision/Intermediate Low Vision Care, Low Vision Care, Occupational Therapy, and Physical Therapy (for vestibular treatment). Keywords from the rehabilitation recommendations (with the guidance of the expert panel) were mapped to the VA clinics. The frequencies of rehabilitation recommendations by clinic were stratified by evaluating Polytrauma Rehabilitation Center.

RESULTS

Description of Cohort

We identified 20,268 unique patients (with and with no TBI) who served in the Operation Enduring Freedom/Operation Iraqi Freedom conflicts. Approximately 12% of the 20,268 (2,458 unique patients) were treated at a VA Polytrauma Rehabilitation Center between January 1, 2008, and December 31, 2017 and administered the mandated exam. Study participants had an average of nine comorbidities, with a mean age of 37 years. They were 95% male, 10% Hispanic, 78% White, and 11% African American. Figure 1 shows the residences of patients on the date the mandated exam was administered. Unique patients are color-coded to indicate which Polytrauma Rehabilitation Center performed the mandated evaluation. Figure 2 reports dates of the mandated exams by year and treating Polytrauma Rehabilitation Center.

Figure 1.

Location of veterans’ home residence (by ZIP code centroid) on date of mandated exam, color-coded by evaluating Polytrauma Rehabilitation Center.

Figure 2.

Mandated evaluations performed by year and by site.

Recommendations by Category

Table 1 displays the number of veterans who received recommendations by category (rehabilitation, glasses/contacts, and imaging/photography/tests) and by treating facility. The total number of referral recommendations (3,387) exceeded the total number of veterans in the study cohort (2,458), because some veterans received referrals for more than one category. As an example, of the cohort of 2,458 patients with mandated exams, rehabilitation was recommended for 581 (23.6%) during the 10-year study period. More than 50% of the cohort was referred for further testing and/or glasses, not mutually exclusive. Less than 25% were referred for vision rehabilitation. Variation across sites was found with Tampa recommending further testing and glasses/contacts for the highest percentage of veterans while San Antonio recommended vision rehabilitation for the highest percentage of veterans. Adjusted analyses showed that photosensitivity was the only symptom that was a significant predictor of recommendation to all three categories of recommendation.

Table 1.

Number and column percentages show the percent of veterans who received a recommendation per site.

Recommendation Category	Total n=2,458	Minneapolis n=321	Palo Alto n=275	Richmond n=388	San Antonio n=134	Tampa n=1340
Imaging, Photo	1426 (58.0)	143 (44.6)	125 (45.5)	51 (13.1)	16 (11.9)	1091 (81.4)
Glasses/Contacts	1380 (56.1)	154 (48.8)	67 (24.4)	120 (30.9)	55 (41.0)	984 (73.4)
Rehabilitation	581 (23.6)	68 (21.2)	99 (36.0)	121 (31.2)	80 (59.7)	213 (15.9)

Percent is calculated as 100 times number of Veterans with recommendation in the category/total number of Veterans in the center who received a referral. Column percentages do not add up to 100 because some Veterans had multiple recommendations from one or more categories.

Figure 3 uses a circular bar chart to illustrate Table 1 data. Bars are shaded by recommendation category and grouped by evaluating facility. A longer bar indicates more veterans received a recommendation. Figure 4 shows the overlap in categories of recommendations for the total cohort (Table 1, column 2). Table 2 and Figures 5A–5C show demographic and clinical variables significantly associated with referral for rehabilitation (Figure 5a\A), glasses/contacts (Figure 5b), and imaging/photos/videos/tests (Figure 5C).

Figure 3.

Circular bar charts of the percentage of veterans with recommendations by category by polytrauma rehabilitation centers. A=Minnesota, B=Palo Alto, C=Richmond, D=San Antonio, and E=Tampa Polytrauma Rehabilitation Centers

Figure 4.

Venn diagram veterans who received recommendations for one, two, or all three categories of recommendations.

Table 2.

Demographic and clinical variables significantly associated with recommendations per category of recommendation.

	Rehabilitation		Glasses		Imaging, Photo, VideoTests
c-statistic	.689		.792		.761
Variable	OR*	P-value	OR*	P-value	OR*	P-value
Age (per 10 years)	.84	<.001	1.21	<.001
Floaters	.74	.024			1.54	.001
Photosensitivity	1.32	.032	1.67	<.001	1.63	.002
Blurred vision			2.10	<.001
Double vision	1.29	.022
Decreased visual field	1.34	.024	0.75	.030
Decreased night vision			1.48	.010
Balance problems/dizziness	1.36	.022
Difficulty reading	1.63	.001	1.52	.001
Dry eye			1.44	.003
Headache					1.36	.015
Ocular pain					1.34	.050
Comorbidity count	1.02	.147	1.02	.09	0.97	.013
Non Tampa site	3.94	<.01	0.46	<.001	0.21	<.001
Access in minutes from facility (per 30 min.)	1.00	.941	0.89	.073	1.03	.777
Access in miles from facility (per 20 miles)	1.01	.905	1.07	.084	0.98	.792

^*OR=odds ratio

Figure 5.

(A) Factors significantly associated with a rehabilitation recommendation (vestibular, low vision, blind rehab, and low vision + blind rehab). A=Minnesota, B=Palo Alto, C=Richmond, D=San Antonio, and E=Tampa Polytrauma Rehabilitation Centers. (B) Factors significantly associated with a recommendation for glasses or contacts. (C) Factors significantly associated with imaging, photography, or test recommendation.

Rehabilitation Recommendations.

Table 3 shows the rehabilitation recommendations by type of rehabilitation clinic by evaluating center. Many key words from the evaluation recommendations mapped to more than one clinic, therefore five groups were created: (1) referrals based on key words definitively mapped to Blind Rehabilitation (inpatient and outpatient) or to (2) Low Vision; (3) keywords mapped to both Blind Rehabilitation and Low Vision; (4) keywords mapped to vestibular treatment typically treated by Physical Therapy and sometimes Occupational Therapy; and (5) keywords mapped to Occupational Therapy. The fifth group, Occupational Therapy, may have included non-vision related referrals. Of the 2,458 veterans who received the mandated exam, 581 (23.6%) received some type of vision rehabilitation referral. Figure 6 graphs the Table 3 data showing recommendations per rehabilitation clinic for the first four groups defined above, within and across evaluating Polytrauma Rehabilitation Centers.

Table 3.

Veterans with rehabilitation recommendations by rehabilitation clinic and evaluating facility.

	Total Cohort n = 2,458	Minneapolis n=321	Palo Alto n=275	Richmond n=388	San Antonio n=134	Tampa n=1340
Clinic(s)	581 (23.6)	68 (21.2)	99 (36.0)	121 (31.2)	80 (59.7)	213 (15.9)
Blind Rehab	253 (37.2)	23 (33.8)	68 (68.7)	90 (74.4)	70 (87.5)	2 (.9)
Low Vision	23 (3.4)	0 (.0)	14 (14.1)	2 (1.7)	2 (2.5)	5 (2.4)
Blind Rehab/Low Vision	382 (56.2)	46 (67.6)	65 (65.7)	69 (57.0)	25 (31.3)	177 (83.1)
Vestibular (OT/PT)	22 (3.2)	1 (1.5)	1 (1.0)	0 (.0)	1 (1.3)	19 (8.9)
OT**	399 (16.2)	49 (72.1)	67 (24.4)	70 (18.0)	26 (19.4)	187 (14.0)

^**May include OT referrals not related to vision therapy

OT = occupational therapy, PT = physical therapy

Figure 6.

Circular bar charts of the percentage of veterans with recommendations to specific rehabilitation clinics.

DISCUSSION

Anecdotally, vision problems tend to be the most sensitive sign of TBI and typically the last to recover in patients. Patients report that “until my vision was corrected, I really didn’t move on in therapy,” and “I couldn’t see what the therapist wanted me to do – it was hazy and blurred.” Other Veterans did not realize they had vision dysfunction but complained of not being able to follow the football during a game on TV. There was no existing system of care to accommodate the post-9/11 service members returning home to the U.S. for medical treatment for TBI and other blast-related injuries. This manuscript reports on VA service delivery and utilization for veterans with TBI-related ocular injury and vision dysfunction that emerged following an unpredicted need that was met by dedicated clinicians using resources available at the time.

Variation in referrals for rehabilitation, glasses, and further testing was found. This is not surprising as each of the Polytrauma Rehabilitation Centers evolved independently of the others. The variation in delivery of vision rehabilitation was at least partly attributed to how patient care was uniquely delivered by the VA Polytrauma Rehabilitation Centers and the VA Blind Rehabilitation Service at each Polytrauma Rehabilitation Center site. The VA Polytrauma System of Care, created in 2004, provides comprehensive specialized services related to veterans with TBI and polytrauma, i.e., those with a significant associated secondary injury. The VA Blind Rehabilitation Service Continuum of Care, established in 2006, consists of inpatient Blind Rehabilitation Centers, Vision Impairment Services in Outpatient Rehabilitation programs, Advanced Low Vision clinics, Intermediate Low Vision Clinics, Vision Impairment Services Team Coordinators, and Blind Rehabilitation Outpatient Service specialists. The variation, albeit unavoidable at the time, now, two decades later, poses a challenge to the standardization of VA care and precludes a clear pathway to treatment.

Overall, the data suggest a need for more standardized accounting of key symptoms and diagnostic testing protocols. In addition, the current service-delivery model may be outdated in that it does not adequately address both the TBI and the related ocular injuries and visual dysfunctions incurred by military personnel serving during the 9/11 and post-9/11 conflicts. Additionally, the current delivery model demonstrates a lack of understanding about how the visual system works. There are two visual systems to consider for vision rehabilitation. The afferent system transmits sensory information from the eyes to the brain. Damage to the afferent system causes ‘visual impairment’ that cannot be repaired. Instead, patients are trained to compensate for their remaining visual system. Patients with moderate to severe TBI typically have visual impairment. The efferent system transmits sensory information from the brain to the eyes. Damage to the efferent system causes visual ‘dysfunction.’ Typically, patients with mild to moderate TBI have damage to the efferent system and can be retrained or rehabilitated to at least age-appropriate norms. These differentiations are important for the move towards clinical guidelines for community-based care. See Table 4 for a summary of the VA service delivery models in actual practice, contributed by the optometrists and ophthalmologists performing the mandated examinations at each of the five Polytrauma Rehabilitation Centers. These variations in service delivery models explain at least some of the variance in referrals.

Table 4.

Summary of actual service delivery across polytrauma facilities.

	Rehabilitation Centers						Network Sites
Rehabilitation Services	Minneapolis	Palo Alto	Richmond	San Antonio	Tampa	Lexington		Tucson
Moderate to severe TBI
Visual field/neglect	OT	OD	BRCT (LV, O&M, Stroke, TBI, binocular)	BR, OD	OTVS	LV		BR
Accommodation
Orientation & Mobility	VISOR	BR		BR
Magnification		OD, BR, LV		BR, OD		OD, LV
Glarewear		OD, BR				OD		BR, OD
Prism visual field loss	OD	OD	OD		OD			OD
Legally/totally blind	VISOR	VIST	VISOR		BR, OTVS	BR		BR
Mild to moderate TBI
Oculomotor	OT	OD	BRCT	BR, OD	OTVS	OD, office, telehealth		BR
Vergence
Accommodation
Glarewear	VISOR	OD, BR			OTVS, LV			BR, OD
Prism ocular alignment*	OD	OD	OD		OD			OD

OD=optometry, OT=occupational therapy, OTVS=OT vision specialist, BR=blind rehabilitation specialist, BRCT=BR cross trained, LV=low vision specialist, O&M=orientation and mobility, TBI=traumatic brain injury; VISOR=Vision Impairment Services Outpatient Rehabilitation, VIST=Vision Impairment Services Team

^*Fresnel and ground prism

One way the VA has responded to these challenges was to establish advanced fellowships in optometry to provide post-residency training in diagnosis, treatment, and rehabilitation of post-traumatic visual disturbances. The VA National Enterprise System for Telerehabilitation^{15, 16} may be another option to integrate the treatment of TBI with blind rehabilitation. A question posed by this study asks how should the delivery of vision and TBI rehabilitation services be paired at the community level? The task at hand is to take the best practices from the Polytrauma and Blind Rehabilitation Services to develop a standard of care that can be applied across the VA System of Care, to include community-based care.

This is the first known study to examine service delivery of vision rehabilitation for veterans with TBI related vision dysfunction and ocular injury. Included in our analyses were calculations of prevalence of symptoms. Our findings were similar to symptoms identified by Capo-Aponte et al.² in a study that examined the medical reports of 500 U.S. military personnel diagnosed with deployment-related mTBI who had received eye care at the Landstuhl Regional Medical Center in Germany. Common visual symptoms at the acute stage were blurred vision (66%), reading problems (62%), and light sensitivity (40%), compared to the symptoms of the cohort group at the time of mandated testing at 53%, 46%, and 58% respectively. In the Capo-Aponte et al.² cohort, 89% reported dizziness, and 93% reported headache, compared to patients in the current study’s cohort, reporting 41% and 39% respectively. This study built on prevalence of symptoms by identifying significant relationships between symptoms and other clinic and demographic variables and healthcare utilization.

There were limitations. First, data were captured solely from large administrative databases, thus, the variables selected for analysis were limited to those available in administrative data sets. For example, while we identified veterans with difficulty reading, our data did not enable us to learn more specifics such as the cause of the difficulty reading. Second, this was the first time that the mandated evaluation data (described above in the Introduction) have been evaluated. The authors were able to achieve all of the study’s data analysis goals, except they were unable to follow optometry follow-up referrals. The inconsistent wording used by the optometrists jeopardized the validity of the data during natural language processing. As a result, ‘Optometry’ was eliminated as a fourth category of recommendations. In the next phase of analyses, the authors plan to take a more in-depth look at optometry assessments and treatment by using natural language processing to examine treatment notes, versus analysis of treatment recommendations in the mandated evaluation used for the analyses reported in this study. A third limitation is that some participants may have received rehabilitation as active-duty service members prior to being admitted to a VA Polytrauma Rehabilitation Center, and this study does not include data from the Department of Defense rehabilitation services.

CONCLUSIONS

The VA Polytrauma System of Care and mandated comprehensive vision evaluation were important first steps in identifying and documenting the many visual and ocular signs and symptoms associated with TBI. Prior to this system, vision rehabilitation meant only blind rehabilitation. This work represents another body of evidence that documents significant prevalence of visual dysfunction following TBI, dysfunction that can impact daily life, therapies, and improvement across many systems. Additionally, this work goes on to objectively identify a need for next steps in TBI-related visual dysfunction, prioritizing clear pathways for a broad spectrum of treatment options within the VA and community reintegration outside the VA.