The U.S. Department of Veterans Affairs (VA) provides compensation to military veterans with service-connected injuries or conditions, defined as “disabilities that are the result of a disease or injury incurred or aggravated during active military service” (
1). Although President Lincoln’s famous charge “to care for him who shall have borne the battle” was issued in 1865, a formalized disability payment system for injured veterans did not appear until 1917 in an amendment to the Bureau of War Risk Insurance program (the predecessor of the VA) that authorized the creation of a schedule “of ratings of reductions in earning capacity from specific injuries or combinations of injuries of a permanent nature” (
2).
Today, veterans can receive cost-free VA health care for any service-connected condition and monthly tax-free disability payments. More than one-quarter of all U.S. military veterans—more than 5 million in total—receive these benefits (
3,
4). For fiscal year 2022, $139 billion of the total $270 billion VA budget was allocated to the disability benefits program managed by the Veterans Benefits Administration (VBA). A substantial portion of this budget is for mental health conditions, including posttraumatic stress disorder (PTSD). Congressional testimony by VA officials in 2017 indicated that the number of PTSD claims had nearly tripled since 2007 (
5), far outstripping the corresponding increase in non–mental health claims. In 2016, >1 million veterans were receiving service-connected disability benefits for PTSD, nearly 100,000 were receiving them for anxiety, and another 250,000 were receiving them for major depression. Service-connected veterans had, on average, five service-connected conditions, and 13% of those with a disability were rated at 100% (
6,
7).
Service connection and the percentage of disability associated with each condition are determined with the VA Schedule for Rating Disorders (VASRD), which describes what symptoms or impairments qualify for specific amounts of compensation for each possible condition under federal regulation. Ratings derived from the VASRD can range from 0% to 100% disability (
8). Under the current VASRD, the percentage of disability awarded is based on the presence of 31 psychiatric symptoms and the purported level of impairment associated with each. Veterans usually receive the disability rating associated with the symptom that has the highest assigned impairment percentage.
The current schedule—or rubric—for rating PTSD and other mental disorders has not been updated in >25 years. On February 15, 2022, proposed changes to the rating schedule for mental disorders were published in the
Federal Register (
8). The changes are substantial and reflect a reconceptualization and significant overhaul of the current rating rubric. Rather than basing the percentage of service-connected disability on the presence of particular symptoms, the new rubric will base the rating percentage for a given condition on the extent of functional impairment in five different domains.
Here, we review the current symptom-based rating rubric as it pertains to mental disorders. We then discuss the proposed changes and compare the new rubric with other rating systems used for mental disorders. In examining the implications of the proposed changes for the VA disability system writ large, we draw on research regarding psychiatric symptoms and their relationship to functional impairment and on best-practice assessment of psychiatric disorders.
The Symptom-Based Rating Schedule
The current VASRD relies on a list of 31 general psychiatric symptoms that are coded as present or absent on a standardized form completed by a mental health examiner, who conducts what is referred to as a compensation and pension (C&P) examination. This method of disability determination has several problems. First, the list of symptoms is somewhat arbitrary and excludes core PTSD diagnostic symptoms even though PTSD is the most prevalent service-connected mental health condition. Second, the rating for each symptom is dichotomous (present or absent), although in clinical reality, symptoms exist on a continuum of severity that has direct implications for related impairment. In addition, despite many years of reliance on this list of symptoms for disability calculation, there are no published guidelines for when the severity of a veteran’s symptom meets the threshold for endorsement. For example, the symptom “disturbance of motivation and mood” could be endorsed by some examiners for relatively minor disturbances but endorsed by others only for major problems that cause substantial impairment (see Meisler and Gianoli (
9) for a more detailed discussion of problems inherent in the current symptom list).
A third and more fundamental problem with symptom-based ratings is that significant discordance often exists between symptoms and function. It has long been recognized that symptom severity is only modestly correlated with functioning among individuals with serious mental illness (
10–
13). For example, in a sample of older patients with schizophrenia, Bowie and colleagues (
10) found that a performance-based measure of function was more predictive of functioning (as rated by caretakers) than were symptoms. Directly pertinent to the preponderance of PTSD claims within the VA, a study of 73 women with PTSD found that neither clinician-rated PTSD symptom severity nor self-reported PTSD symptoms were associated with measures of functional capacity (
14). These problems are the basis for the VBA’s proposal to shift away from a symptom-based disability rating schedule.
Proposed Changes to the Rating Schedule
The proposed changes to the rating formulas are based on recommendations solicited from the VBA, Veterans Health Administration (VHA), Board of Veterans’ Appeals, U.S. Department of Defense, and veterans service organizations. The original impetus for the changes dates back 15 years to when the Institute of Medicine (now the National Academy of Medicine) published the report
The Future of Disability in America (
15), urging a greater emphasis on function than on symptoms in assigning disability ratings. This emphasis on function had previously been articulated by the framework of the
International Classification of Functioning, Disability and Health (
ICF). The
ICF defines
disability as “an umbrella term for impairments, activity limitations and participation restrictions” that arise from dynamic interactions between health conditions and contextual factors (including both environmental and personal factors) (
16). Its framework of disability does not center around symptoms because function does not necessarily align with symptoms. As the plan for VA’s proposed changes notes (
8), “While symptoms determine the diagnosis, they do not necessarily translate directly to functional impairment.”
One of the main purposes of the VA compensation program is to provide benefits to “make up for the potential loss of civilian wages to civilian working time” (
17). VBA’s proposed shift to a focus on function aligns with this emphasis on vocational impairment. The emphasis on function allows for a rating rubric that is not specific to mental health conditions and might be applied to other diseases or disorders that involve impairments, activity limitations, participation restrictions, or all of these. The proposed rubric emphasizes function in the context of specific environments (e.g., work, social, and family relationships) in a way that symptoms do not. The extent of the disability is estimated by considering the difference between the person and a population norm for people without the condition (
15). Thus, the new schedule still requires assignment of the role the condition plays in an individual’s restricted participation in activities. The new VASRD also aligns with the
DSM-5 in its reliance on the structure of the World Health Organization (WHO) Disability Assessment Schedule 2.0 (WHODAS 2.0) (
18) and the
ICD-10 in considering the extent of functional impairment among people with diagnosed health conditions.
The new rating schedule shifts the focus from symptoms to severity of functional impairments resulting from a mental health condition across different life domains. The six domains of the WHODAS 2.0 were consolidated into five by combining “getting along with people” and “participation in society” to yield “interpersonal interactions and relationships.” The four other areas of functioning evaluated are cognition, task completion and life activities, navigating environments, and self-care. Each domain is to be evaluated on the basis of functioning during the past month and rated with a five-level scale. Guidelines are based on a combination of intensity and duration of impairment, a structure similar to that used in the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) (
19).
Table 1 describes how the new rubric combines severity of impairment and frequency.
The new general rating formula will assign 10%, 30%, 50%, 70%, and 100% disability ratings on the basis of the severity of impairments experienced in the five domains according to the rubric outlined in
Table 2. All veterans with a mental disorder diagnosis are given a minimum rating of 10%, which entitles them to cost-free treatment for the condition. The minimum rating is assigned because
DSM-5 diagnoses reflect mental disorders that involve “clinically significant disturbance” and are “usually associated with significant distress or disability in social, occupational, or other important activities” (
20).
The new VASRD clarifies that ratings should “consider any ameliorating effects of medications prescribed for a mental disorder” (
8). That is, examiners should consider symptoms and related functional impairment that remain while the veteran is taking medications rather than speculate about how severe the veteran’s symptoms or impairments would be if they were not taking medications. The new VASRD specifically states that “the ratings should be based on the actual elements of disability present.”
The
Federal Register announcement for the new rating schedule indicated that a proof-of-concept study had been done to “derive the appropriate level to assign to each domain (e.g., 0 through 4)” (
8). The announcement provided little information about how the study was conducted, noting only that it involved 100 veterans who were already connected to services for a mental disorder. The results, as reported, indicated that the proposed rubric, on average, increased the level of disability award, with fewer veterans rated at ≤50% disability, a similar proportion rated at 70%, and more receiving a 100% disability rating. Thus, overall, the average disability rating was higher with the proposed rubric. However, given the modest correlation between symptoms and function, it is possible that for a small percentage of veterans who are symptomatic but functioning reasonably well, service-connected disability ratings could be lower under the new rubric.
The New VASRD and Other Disability Rating Systems
The proposed VASRD for mental disorders is similar to the rubric used by the Social Security Administration (SSA) to determine disability because the SSA also considers abilities and impairments rather than symptoms per se. The SSA guidelines for determining disability are operationalized in its Program Operations Manual System. The section on mental limitations and the ability to work begins with a list of the “basic mental demands of competitive, remunerative, unskilled work” (
21). The four demands listed—following instructions; making simple work-related decisions; responding appropriately to supervision, coworkers, and work situations; and dealing with changes in a routine work setting—overlap with the domains of impairment the VBA is now considering. In the proposed VASRD, the ability to meet these worksite demands would be compromised by social and cognitive impairments. It is noteworthy that under the existing VASRD guidelines, veterans may have had symptoms but may not have had any difficulty with work-specific SSA criteria.
A critical difference between the VBA and SSA systems, however, is that the SSA system is designed to yield yes-or-no decisions about individuals’ ability to work for pay, whereas the VBA rating determines the percentage of impairment attributable to conditions related to military service. In other words, the VBA system recognizes military service–connected impairments that cause reductions in vocational functioning without causing total disability or completely precluding gainful employment. A second major difference is that the VBA system provides compensation for conditions that impair individuals’ vocational and social functioning. The VBA’s consideration of social impairment reflects its commitment to compensating veterans for service-connected harm as a whole, not only for harm that causes occupational impairment. Although the SSA system is designed to provide financial protection only for permanent impairments that preclude work for pay, the VBA compensates veterans for losses related to their military service; these losses and associated compensation can range from minimal to total.
The proposed changes also bring VA’s disability rating rubric for mental disorders in closer alignment with the American Medical Association’s
AMA Guides to the Evaluation of Permanent Impairment (6th ed.) (
22). As with the SSA’s guidelines, the
Guides was developed to mitigate the risk for being unable to work. The
Guides, which is also based on the WHO’s model of disability and impairment, is “the most widely used basis for determining impairment and is used in state workers’ compensation systems, federal systems, automobile casualty, and personal injury, as well as by the majority of state workers’ compensation jurisdictions” in the United States (
23). As described in a thorough review by Warren (
24), the
Guides was first introduced in 1952
to develop and incorporate more empirically supported methodology into the disability determination process so that there is a more standardized means to evaluate an individual who may have physical or psychological issues that impede functioning in everyday life, in particular, regarding the individual’s ability to function within the workplace.
Unlike the SSA rubric but akin to the VA system, the Guides is used to calculate a percentage of impairment for each condition rather than rendering a yes-or-no opinion on vocational capacity.
The rubric outlined in the most recent edition of the
Guides requires examiners to base impairment ratings for mental and behavioral disorders on three separate measures: a 24-item list of symptoms and mental status observations, each rated on a 7-point severity scale; the Global Assessment of Functioning, which combines symptom severity and level of impairment; and the Psychiatric Impairment Rating Scale (PIRS) (
25). The PIRS is an observer-rated scale on which the examiner is asked to provide an impairment rating ranging from 0 to 5, with higher scores indicating higher levels of impairment, for six areas of function: self-care and personal hygiene, social and recreational activities, travel, social functioning (relationships), concentration, and employability or adaptation. The scores are then aggregated across the domains to provide a score for percentage of impairment. Thus, although the
Guides does retain the use of symptom presence and severity, it also incorporates the PIRS as a measure of functional capacity in the rubric for determination of impairment. The proposed changes to the VASRD will make the VA’s rating rubric more consistent with one of the most widely used impairment determination systems available.
Implications for Disability Examinations
Given the new rubric’s focus on functional impairments, C&P examinations will need to go beyond an assessment of symptoms. Although examiners may now assess functioning, the current VBA rubric is not designed to capture that information for rating purposes. Although information about symptoms and their severity will still be needed to determine whether a veteran meets the diagnostic criteria for a disorder (which, per DSM-5, are largely symptom based), examinations will need to delineate how the psychiatric symptoms correlate with specific function and impairment.
Self-Report Assessment Instruments
The WHODAS 2.0 (
18), a self-reported measure of functional impairment, is alluded to in the announcement of the proposed schedule. The WHODAS 2.0 was developed on the basis of the concept that assessment of impairment does not have to be diagnosis specific and should provide information about an individual’s global disability status, reflecting disability caused by any general medical, psychiatric, or comorbid condition (
18,
26). Because the WHODAS 2.0 does not differentiate impairments related to a service-connected condition from those resulting from other causes, use of this measure could inflate disability ratings by including impairments that are unrelated to military service. Regardless of the assessment measures used, fair and accurate adjudication of claims will still require the challenging task of disentangling impairments caused by a service-connected mental health condition from those that preceded military service or are otherwise unrelated to service.
A second risk of using the WHODAS 2.0 and other measures of functional disability is that—much like measures of symptom severity—they are largely reliant on self-report. Instructions for the WHODAS 2.0 state that “if the clinician determines that the score on an item should be different based on the clinical interview and other information available, he or she may indicate a corrected score” (
18). In common use, however, results are based largely on self-report, and scores are therefore susceptible to biases such as exaggeration. For example, in a study using a simulation design to examine the potential for malingering on the WHODAS 2.0, college students who were instructed to malinger to obtain stimulant medications or extra-time accommodations for examinations reported higher levels of disability on all domains of the WHODAS 2.0 compared with students in the control (nonmalingering) group and even compared with students with diagnosed attention-deficit hyperactivity disorder (
27). The authors concluded that the WHODAS 2.0 should not be “relied upon solely as the measure of disability.” Similar conclusions were drawn in a study of those being evaluated for the impact of mild traumatic brain injury. In a sample of Canadian disability claimants who had been referred to a clinic for multidisciplinary independent medical evaluation, those whose neuropsychiatric battery indicated that they were likely malingering scored significantly higher on the WHODAS 2.0 than those whose results did not suggest malingering (
28).
Given that approximately half of veterans who are connected to services for treatment of mental disorders receive compensation for PTSD, consideration of measures developed to specifically assess PTSD-related functional impairments is advised. This specific consideration for PTSD is of particular importance because few measures exist to assess functional impairment associated with PTSD, and unstructured assessments of function by C&P examiners have been found to be remarkably poor to identify PTSD-associated impairment (
29). One measure that has been developed to specifically assess PTSD-related functional impairments is the Inventory of Psychosocial Functioning (IPF) (
30). The IPF is an 80-item self-report questionnaire that asks respondents about PTSD-related difficulties they have experienced in the past 30 days. It inquires about individuals’ functioning or difficulties in seven major life domains: romantic relationships, family, work (including home-based work), friendships and socializing, parenting, education (including distance learning), and self-care. The IPF has high content validity, and psychometric testing has found that both the IPF and its abbreviated form—the seven-item Brief Inventory of Psychosocial Functioning (B-IPF)—have strong test-retest reliability as well as high construct- and criterion-related validity (
31).
Although the IPF and the B-IPF are psychometrically sound and were designed to specifically measure PTSD-related functional deficits, they, like other self-report measures, are susceptible to demand characteristics. This limitation is of particular concern in disability evaluations because some individuals may overreport functional impairments to obtain benefits. It has also been suggested that individuals with some psychiatric conditions, including PTSD, may have difficulty communicating their symptoms and functioning clearly or accurately; thus, information obtained from self-report measures may be limited or deficient (
32). Finally, despite having been developed to specifically measure PTSD-related functional difficulties, the IPF’s ability to discriminate among different sources of impairment—similar to the abilities of the WHODAS 2.0 and other measures—remains unknown.
Performance-Based Measures
The problems inherent in reliance on self-report measures of functional impairment in a disability examination setting suggest the possible utility of measures of actual function and performance. A 2019 report from the National Academies of Sciences, Engineering, and Medicine (
33) provides a detailed review of several performance-based assessment techniques that could be considered. A review of all such measures is beyond the scope of this article. However, one example of such a measure is the University of California, San Diego, Performance-Based Skills Assessment (UPSA) (
34), which takes the examinee through a series of role-plays of various tasks (e.g., making a telephone call or reading a utility bill) required for independent everyday functioning. The UPSA, along with its variants—including abbreviated, computerized, and mobile formats—has been found to accurately predict independent living among middle-age and older adults with schizophrenia (
35,
36) and to be a stronger predictor of employment status than either symptoms or cognitive performance (
33). The UPSA parallels objective-based protocols to evaluate disability in cases involving physical or musculoskeletal impairments such as the functional capacity evaluation (
37).
Although performance-based measures such as the UPSA provide a more direct assessment of functional capacity, the UPSA, in particular, was developed to be used with cognitively disabled populations, and therefore its utility—and the utility of similar measures—with less disabled populations may be limited (
38). Findings regarding the effect of age on performance on the UPSA have also been mixed. One study found that among a sample of healthy elderly participants, UPSA scores were unaffected by age (
39), whereas another found that performance on the UPSA was age dependent, with younger participants performing better than middle-age participants (
40).
Clinical Interviews
Although some of the measures mentioned here may be helpful, clinical interviews are likely to remain the central method of assessing function. It will be important for examiners to learn how to assess functioning in detail with initial questions, follow-up probes, and assessments of over- or underreporting. More study is needed to understand the most informative and effective way to inquire about functioning. Because domains such as social functioning are broad, open-ended questions are likely to be the most informative, at least initially. Broad questions about veterans’ relationships, work history, and hobbies or recreational activities can provide useful information about their social, cognitive, and emotional abilities and difficulties. Examiners can then use this information to elicit specific examples that illustrate veterans’ level of functioning in both social and vocational settings. Examiners should be encouraged to gather corroborating information when available.
Challenges in Implementing the New Functional Rating Schedule
In a critical review of the challenges and complexities associated with the assessment of psychiatric disability and determination of benefits, Schultz (
41) observed that “the clinical definitions of impairment and disability, unlike legal and administrative definitions, must adhere to methodological standards for measurability, reliability, and validity of the evaluative approach used.” The author noted, however, that research on assessment and its application to disability determination rubrics (e.g., the AMA
Guides), “particularly in the area of complex conditions such as pain or psychiatric disability, has not kept pace with needs.”
For the VA, bringing examination procedures in line with the new function-based rating schedule will require significant changes to standard current practice for both examiners and VBA raters. The new rubric will make it necessary to more specifically define, anchor, and operationalize the domains of function in the rating schedule to improve reliability and validity. For example, per the proposed Schedule for Rating Disabilities: Mental Disorders (
8), a moderate impairment is defined as one that poses “clinically significant difficulties,” but the term
clinically significant is not defined. To improve reliability, standardized assessment methods, specific anchor points with illustrative examples of difficulties, or both are needed. Standardized and ongoing training for all examiners—both VA and contractors—will be needed to facilitate the reliable translation of examination findings into appropriate VBA ratings.
Studies of symptom-based rating systems have suggested the importance of precise, standardized measurement in rating decisions. Extensive literature findings have shown the limitations of unstructured psychological diagnostic evaluations (
29,
42–
44) and, in VA disability examinations, their increased proneness to negative examiner bias (
45). In a cluster-randomized trial in which C&P examiners were randomly assigned to training using the CAPS-5 and the WHODAS 2.0 or to no training, the examinations conducted by the trained examiners were far more detailed with less between-examiner variance than were those conducted by the untrained examiners (
29). Given what is known about the limitations of unstructured evaluations, this may well be the perfect time for the VBA to promote or even require that examiners adopt the use of standardized measures in their examination process and implement a standardized quality assurance monitoring system for both examinations and ratings.
Conclusions and Recommendations
The VA has proposed fundamental changes to its rating schedule for mental disorders that have the potential to substantially affect rating decisions and compensation for millions of veterans with military service–connected conditions. The new system reflects an overdue shift away from a symptom-based formula and toward one based on real-world functioning, bringing the VA rubric more in line with other disability systems, including those used by the SSA and the AMA and the system endorsed by the WHO.
Data from a small pilot study reported in the
Federal Register (
8) suggest that the new system would, on average, increase ratings for veterans determined to have service-connected psychiatric disabilities, with a greater number of veterans rated at ≥50% disabilities. The reasons for this outcome are unknown, although possible explanations include the nonspecific nature of functional ratings that relate to all conditions without discriminating or apportioning impairments arising from different causes. The real-world impacts of the new schedule have yet to be determined. Awarding some compensation—at least a 10% rating—to all veterans with a service-connected psychiatric condition reflects a change from the current system that allows for 0% awards, which may make a veteran eligible for at least some health care benefits but not for financial compensation. The change to a 10% rating would be expected to increase the overall proportion of veterans who are compensated monetarily. Veterans who have functional impairment despite fewer or less severe symptoms would be expected to see increases in their compensation, whereas those with lesser functional impairment than would be suggested by their symptoms should see a decrease.
The effects of the change in the VASRD will also be affected by the complex systems that surround and have an impact on service-connection determinations. Defining and refining the VASRD guidelines is only the first step of the process. Proper training and supervision of examiners, VBA raters, and others involved in the process, as well as reliable implementation of and adherence to the new guidelines, will be critical to their utility. Justifiable service-connection awards depend on the quality of the evidence that is gathered. It appears likely that the new VASRD will demand more from examiners—more information elicited during an interview, greater judgment about functioning as extrapolated from interview and other claimant information, and perhaps more careful attention to the accuracy of self-reported information. A good-quality mental disorder disability evaluation requires gathering comprehensive information not only from the diagnostic interview with the veteran but also from detailed chart reviews (both medical and military) and, if available, collateral sources. To ensure that adequate information is collected during the C&P examination and reported afterward, it will be crucial to have reporting forms that capture functional information in an objective, detailed, and reviewable way.
VBA’s adaptation of the function-based rating schedule for determining service-connection awards will bring opportunities for quality improvement projects and research to study the implications of the new rating rubric for veterans. Important questions that have been raised about racial (
45) and gender (
46) disparities in service-connection determinations with respect to the symptom-based assessments should be reexamined with the new function-based schedule. The change also provides an opportunity to improve the mental health C&P examinations currently being conducted, ideally leading to a more standardized process. Such standardization would likely reduce the wide variability in methods and quality of examinations that currently exists (
47,
48).
We recommend testing the validity of standardized assessments of functioning in the disability examination setting. Should the assessments be valid for service-connection rating, we recommend that the VBA encourage or require the use of validated tools to complete C&P examinations. Consideration should be given to measures that focus specifically on occupational functioning and that are performance based as opposed to relying only on self-report.
Although the final impact of the proposed changes to the VASRD on individual veterans remains unknown, the new system may very well have ripple effects on other parts of veterans’ lives affected by service-connection awards. Changes in the VASRD will affect who is eligible for free VHA health care and, in turn, who uses this health care. Changes in the rating rubric will also have a direct financial impact on at least some veterans, possibly causing changes in their ability to meet the basic costs of living—improving it for some and reducing it for others. Much as the current system has been subjected to extensive research and scrutiny, once the new rating rubric is implemented, it and its impacts will also be open to examination and critique.
The proposed changes to the VASRD are being made in a program whose policies, procedures, and implementation are designed to compensate veterans for conditions caused or worsened by their military service and to give veterans’ claims the benefit of the doubt (
49). These policies, as currently implemented, have received considerable criticism for allowing for malingering and reinforcing illness behavior (
50–
52). The rating rubric is only one part of the overall compensation system. Therefore, in addition to basing disability determination on function rather than symptoms, it will be important to address previously noted concerns that examinations are not rigorous and recovery is disincentivized. The rollout of the new rubric, and the procedural changes it will require, offers an ideal opportunity to implement strategies to address these concerns. The new rubric aligns with other disability determination programs and holds promise for more accurately assessing veterans’ conditions, related impairments, and accompanying disabilities, both socially and vocationally. In doing so, the proposal offers the opportunity to improve compensation determinations for veterans with mental disorders related to their military service.