In response to a request from the Social Security Administration (SSA), the National Academies of Sciences, Engineering, and Medicine (the National Academies) appointed a committee to conduct a study on identifying disabling medical conditions in adults that are likely to improve with treatment. Of particular interest to SSA are long-lasting conditions (12 months or more) in the categories of mental health disorders, cancers, and musculoskeletal disorders.
Specifically, SSA tasked the National Academies committee with the following:
- Identify and define the professionally accepted, standard measurements of outcomes improvement for medical conditions (for example, mortality and effectiveness of care);
- Identify specific, long-lasting (12-month duration or longer) medical conditions for adults in the categories of mental health disorders (such as depressive disorders, anxiety disorders, attention deficit/hyperactivity disorder), cancers (such as breast, skin, thyroid), and musculoskeletal disorders (such as disorders of the back, osteoarthritis, other arthropathies) that are disabling for a length of time, but typically (for most people with the condition) do not result in permanently disabling limitations, are responsive to treatment, and, after a specific length of time of treatment, improve to the point at which the conditions are no longer disabling; and
For the conditions identified in Objective 2 (above):
- Describe the professionally accepted diagnostic criteria, the average age of onset, and the gender distribution for each condition;
- Identify the types of medical professionals involved in the care of a person with the condition;
- Describe the treatments used to improve a person’s functioning, the settings in which the treatments are provided, and how people are identified for the treatments;
- Describe the length of time from start of treatment until the person’s functioning improves to the point of which the condition is no longer disabling and specific ages where improvement is more probable;
- Identify the laboratory or other findings used to assess improvement, and, if patient self-report is used, identify alternative methods that can be used to achieve the same assessment; and
- Explain whether pain is associated with the condition, and, if so, describe the types of treatment prescribed to alleviate the pain (including alternatives to opioid pain management such as non-pharmacological and multimodal therapies).
The findings of this National Academies study will assist SSA in the administration of its programs that provide disability benefits: the Social Security Disability Insurance (SSDI) program and the Supplemental Security Income (SSI) program. SSDI provides disability benefits to working-age Americans who are no longer able to work due to a disabling medical condition or terminal illness. SSI provides income assistance for disabled, blind, and aged people with limited income and resources regardless of their prior participation in the labor force. Both programs share a common disability determination process administered by SSA and a common definition of disability for adults, which is “the inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months.” SSDI and SSI beneficiaries must undergo periodic review to determine their continued eligibility. The timeframe of review varies by medical condition and is set based on SSA’s knowledge of the medical condition.
COMMITTEE’S APPROACH TO THE TASK
A 15-member committee was formed to address the task. Members with diverse backgrounds and expertise were appointed to focus on the different aspects of the task. Specifically, the members have expertise in various
branches of clinical medicine (mental health, oncology, and musculoskeletal disorders) and in biostatistics and epidemiology, health care policy, and health outcomes research.
The committee organized itself into three groups: cancer, mental health, and musculoskeletal. Each group had experts in the specific disease category that was being studied, in addition to a health outcomes researcher or a biostatistician and epidemiologist. Each group considered the specific disease categories listed in the Statement of Task and identified the diseases within those categories that they would study. Each disease outcome chapter provides criteria for the selection of the diseases chosen. The committee met five times in person.
In responding to the objectives in the Statement of Task, the committee examined systematic reviews, when available, for the medical conditions studied and also relied on published guidelines, particularly if they had been through an external review process. The committee instructed the National Academies staff to conduct targeted literature searches and to gather information from relevant texts, scientific journals and professional societies, and federal sources. The staff initially reviewed more than 1,157 titles and abstracts; those were narrowed to about 528 studies, which the committee members carefully reviewed for relevance to their task. Several disorders required additional searches to include randomized controlled trials, if there were a limited number of systematic reviews or meta-analyses.
COMMITTEE’S CONCEPTUALIZATION OF DISABILITY AND FUNCTION
The committee members considered the issue of general medical improvement versus functional improvement, as disability models have expanded beyond a one-dimensional conception of medical conditions as the sole determinants of disability. The models conceptualize disability as an entity that reflects an aggregate of individual, societal, and environmental factors. More specifically, an individual may be severely limited in one context, but able to maintain independence and gainful employment in another. The committee considered the entire range of function even though higher levels may not directly pertain to SSA’s definition of disability. That decision was driven by several factors. First, multiple impairments and medical morbidities typically combine to disable individuals with the target conditions. Second, these impairments and morbidities may affect different functional domains. Therefore, a measure of function that is specific to one domain may not capture deficits in other domains. Third, co-occurring impairments may interact synergistically rather than additively to disable patients. As a consequence, impairments with moderate impact on a discrete functional domain may exert profoundly disabling effects when combined with other
impairments. Those effects may become particularly potent when patients are also afflicted by symptoms and comorbidities.
DISABLING MEDICAL CONDITIONS AND LIKELIHOOD OF IMPROVEMENT WITH TREATMENT
The committee chose to examine disabling medical conditions within the categories of cancer, mental health, and musculoskeletal disorders to carry out the Statement of Task. The following sections summarize the committee’s conclusions by disease category.
Cancer is the second leading cause of death in the United States and a major cause of disability. In recent years, because of the development of new treatments such as immunotherapy and CAR T-cell therapy, there has been an increase in the overall survival of patients with cancers that would historically have had a poor prognosis. The committee notes the following cancers are likely to be disabling for a length of time (usually around the time of diagnosis) but might improve with treatment, particularly with recent developments in cancer therapy: breast cancer (excluding ductal carcinoma in situ), melanoma, renal cancer, head and neck cancers, advanced epithelial ovary cancer, non-small-cell lung cancer, and diffuse large B-cell lymphoma. The committee acknowledges that other cancers might also fit the criteria.
In addition to the effects of the medical condition itself, cancer treatments are well known to cause morbidity in cancer survivors. Though the treatments have generally improved to be both more effective and less debilitating, treatment-related impairments are still common and, in many instances, expected. Studies show that most types of cancers result in decreased work ability in patients, at least during active treatment or in the cancer’s terminal phase, and that decreased work ability is often associated not with the progression of the cancer itself, but rather with treatment, treatment-related side effects (also known as toxicities), and comorbidity with other health conditions. The adverse effects of some treatments can be profound, with serious implications for function and quality of life. At the core of cancer treatments are surgery, systemic therapy, and radiation therapy. Each of those modalities has evolved significantly in recent years. Systemic therapy, for instance, which historically centered on various combinations of cytotoxic chemotherapeutics, now includes hormonal and biologic (targeted, immune, and gene) therapies. The addition of these new agents has revolutionized the treatment of many types of cancer but also has introduced new types of morbidity. Treatment-related impairments
include pain, fatigue, cardiotoxicity, peripheral neuropathy, lymphedema, pulmonary dysfunction, and cognitive dysfunction. The residual effects of cancer treatments can present decades after treatment. Studies have shown that the majority of cancer patients will improve after treatment completion, although the time course is patient specific.
The most significant recent advance in our understanding of cancers that are likely to improve with treatment has been achieved through an influx of promising new pharmaceuticals. Improved prognoses for some cancers have been realized through the integration of novel, targeted immune checkpoint and PARP inhibitors (pharmacological inhibitors of the enzyme poly [ADP-ribose] polymerase), among others. The impact of those agents has, for some cancers, led to durable remissions in cancers previously considered to be imminently fatal. For example, their effects on metastatic melanoma have been particularly significant. The effective practice of precision medicine permitted by such agents will likely expand to include different types and stages of cancer as well as new agents. However, much uncertainty remains regarding their toxicities, and only patients whose tumors express targetable molecules are eligible for such therapies. Common, functionally morbid toxicities with the potential to affect all body systems have been attributed to those agents. Consequently, the body of evidence regarding their harms and benefits continues to evolve. Additional advances in cancer care that have improved treatment outcomes include enhanced imaging, earlier detection capabilities, and enhanced supportive care, among others.
Cancers are a heterogeneous class of medical conditions with impairments and recovery that are hard to generalize over the course of the disease. The committee developed three overall conclusions regarding its review of specific selected cancers. First, variation in the ability of a cancer to improve with treatment exists within cancers of a particular organ system—not only by stage, but also by cancer cell type and molecular and genomic characteristics. Prognosis and treatment decisions are likewise based on the cancer site, stage, cell type, and molecular and genomic characteristics. For example, triple-negative invasive breast cancer (breast cancer with tumors lacking estrogen, progesterone, and the HER2 gene) is much more aggressive and has a lower survival rate than many other invasive breast cancer cell types. Another example is that recent phase III trials show that targeted therapies demonstrate superior efficacy to chemotherapy in non-small-cell lung cancer patients with an activating EGFR (epidermal growth factor receptor) mutation and in patients with ALK (anaplastic lymphoma kinase) rearrangements. Patients’ ultimate survival varies dramatically based on the treatments available for the specific cancer sites, the stage of the disease, cell types, molecular and genomic markers, and the individual patient characteristics, including the presence of comorbid disease and the patient’s
functional status and social determinants of health. Additionally, a few studies suggest that for certain combinations of cancer site and treatment, the response varies by age, although the direction of the relationship varies among the studies.
Second, success in cancer treatment does not predict improved functional outcomes. Long-term cancer survivors often experience multiple comorbidities and impairments related to the toxic effects of the cancer therapies they underwent, including surgery, radiation, and systemic therapy (chemotherapy, biologic therapy). These impairments are a major cause of morbidity and have their own trajectories, treatments, and treatment response considerations. There are various types of side effects: acute side effects that develop during treatment but are transient, long-term side effects that develop during treatment but are chronic, late effects that develop after completion of the treatment, and secondary effects that result from acute and long-term side effects. The committee suggests that the following common cancer-related impairments can be disabling for a period of time but managed, though not necessarily cured, with treatment: pain, cancer-related fatigue, cardiotoxicity, chemotherapy-induced peripheral neuropathy, lymphedema, pulmonary dysfunction, and cognitive dysfunction. Additionally, the committee notes that improved functional outcomes do not predict return to work.
Finally, it is important to consider the recursive nature of cancer, cancer treatments, and impairments. Cancer is a dynamic process, and as cancer patients survive longer, they experience a higher probability of disease relapse that can reset an episode of treatment. Given that cancer treatments commonly result in functional impairment, and disease relapse is highly probable, the question of how long it takes from initiation of cancer treatment until functioning improves is a complex one. The committee suggests that the length of time from the start of cancer treatment until a person’s functioning improves to the point at which the condition is no longer disabling involves two timeframes: (1) the time to remission of the cancer, and (2) the time to recovery from toxicities, symptoms, and functional impairments. The committee notes that a cancer patient’s disease status (i.e., whether the cancer is in complete, partial, or no remission), more so than the cancer site and stage, is an appropriate indicator of whether the patient’s functional status should be assessed for improvement. If a patient’s cancer achieves complete remission, functional status improvement is probable, and it is reasonable to evaluate the patient’s functional status 12 months after achieving complete remission; if the cancer achieves stable partial remission, then functional status improvement is possible, and it is also reasonable to evaluate functional status 12 months after achieving stable partial remission; if the patient has no response to treatment or experiences progression of disease, then functional improvement is unlikely.
Mental Health Disorders
The committee selected eight mental health disorders for inclusion in the report: major depressive disorder, bipolar I disorder, bipolar II disorder, obsessive compulsive disorder (OCD), posttraumatic stress disorder (PTSD), panic disorder (PD), generalized anxiety disorder (GAD), and social anxiety disorder (SAD). Those mental health disorders are highly prevalent, are associated with significant functional impairment, and may respond to treatment. Professionally accepted diagnostic criteria for these conditions are detailed in the Diagnostic and Statistical Manual for Mental Disorders, Fifth Edition (DSM-5).
People diagnosed with a mental health disorder are directed to a specific treatment depending on clinical practice treatment guideline recommendations, their treatment history, their treatment preference, and treatment availability, among other factors. The committee expects that most patients who are disabled by psychiatric disorders are receiving psychiatric services by combinations of mental health professionals, including a prescriber (e.g., psychiatrist, advanced practice nurse), psychologist, licensed clinician social workers, and individuals with counseling or rehabilitation degrees. The conclusions here, however, should be interpreted with the caveat that for some populations (e.g., those in rural areas or small towns), care from qualified mental health professionals (e.g., specialized in evidence psychotherapy) might not be available. Importantly, the committee cautions that even under ideal treatment, full remission of mental health disorders, particularly when already determined as disabling, is seldom achieved.
Disorder-specific clinical practice guidelines detail evidence-based treatments for the eight disorders that the committee reviewed. Generally, those mental health disorders can be treated effectively with psychotherapy, pharmacotherapy or other biologic treatments, or a combination of both. There is no indication that improvement varies with age. However, some individuals do not improve after receiving evidence-based treatments, and among those who do improve, some will relapse. Furthermore, it is uncertain whether the rates of remission and response observed in the scientific literature can be generalized to those receiving SSDI or SSI on the basis of a mental health disorder.
For the most part, in the clinical trials of treatments for mental health disorders improvement is defined in terms of disorder-specific symptoms, not functioning. Work-related disability is rarely assessed as an outcome. Furthermore, because there are no evidence-based laboratory tests for mental health disorders, mental health outcomes are assessed using patient self-report measures or clinician assessments.
There is a dearth of data on the length of time from start of treatment until the person’s functioning improves to the point where the mental health
disorder is no longer disabling. Attempting to accurately describe time to functional improvement by drawing from the existing data has important limitations. First, as mentioned above, treatment efficacy in research trials is generally defined in terms of symptomatic improvement, not functional outcomes, and time to symptomatic improvement is restricted to the duration of the trials. Second, psychiatric disorders are often recurrent, so time until improvement cannot be adequately captured as a linear process. Thus, individuals may have periods of remission during which they no longer meet the criteria for disability and later have an exacerbation of illness and associated functional limitations during which they again meet the criteria for disability. Third, the relationship between changes in symptoms and functioning is complex, and symptomatic improvement may not correspond to contemporaneous improvements in functioning. Fourth, psychiatric disorders generally occur with other psychiatric disorders, chronic pain, and medical conditions, and time to improvement will depend on those and other factors. Any estimates of time to improvement needs to consider the fact that clinical trials generally exclude participants with comorbidities. Fifth, the mental health disorders discussed in the report are under-recognized and effective treatments, particularly evidence-based psychotherapies, are often unavailable. That is particularly true for OCD. However, based on the limited evidence, the committee made the following conclusions regarding time from the start of treatment to improvement in functioning.
With regard to major depressive disorder, functional improvement may lag behind or not occur even when a person is in symptomatic remission and may require rehabilitation that targets a return to work. Even then, recovery of occupational functioning, if it occurs, may take 1–2 years and may be limited by environmental contingencies. Early response to treatment might predict likelihood of improvement.
For bipolar I disorder, the acute phase of treatment lasts 6–12 weeks, while the maintenance phase of treatment, which focuses on functional recovery, lasts 6–24 months. Caveats include the fact that improvement in social and occupational functioning may be limited or delayed and require targeted rehabilitation efforts. High-quality research shows that even with the addition of vocational rehabilitation, the potential for return to gainful employment may be limited due in part to financial and other environmental impediments.
Time to improvement can range from 12 to 24 weeks in OCD. Individuals requiring higher doses of medication or more complex cases may take 1 year or more to receive the full treatment benefit. For PTSD, there is some evidence from clinical trials indicating that general functioning improves in response to psychotherapy modalities. The length of time to improvement in functioning varies across psychotherapy modalities and usually corresponds to clinical trial follow-up endpoints (e.g., 8 or 16 weeks). Evidence
regarding improvement in functioning from pharmacotherapy studies is less convincing, and the literature on improvements in work functioning specifically following PTSD treatment is scant.
For PD, GAD, and SAD, the time to improvement in symptoms in clinical trials is generally about 3 months. A longer treatment period may be required for partial responders or after relapse. Time to improvement in routine practice is likely considerably longer than in randomized controlled trials because of patient clinical complexity, treatment history, psychosocial factors, and variability in treatment delivery. Notably, the relationship between symptoms and functioning in individuals with anxiety disorders is weak. Even after treatment response or remission from an anxiety disorder, individuals may continue to have significant functional impairments, which in turn may predispose them to a relapse of the anxiety disorder.
The committee notes that all of those conditions may be associated with chronic pain, which may contribute to increased risk for mental health disorders, and mental health disorders may result in an increased risk of chronic pain. The types of chronic pain that commonly co-occur with mental health conditions include migraine headaches, neck and back pain, fibromyalgia, and abdominal pain.
Musculoskeletal disorders are a diverse set of conditions affecting bones, joints, muscles, and connective tissues. Those disorders may result in pain and a loss of function and are among the most disabling and costly conditions in the United States. Chronic pain and a loss of function are the primary mechanisms through which musculoskeletal disorders lead to disability and work loss.
SSA noted three categories of musculoskeletal disorders in its Statement of Task to the National Academies: disorders of the back, osteoarthritis, and other arthropathies. Based on the committee’s clinical expertise and knowledge of the medical and research literature on musculoskeletal disorders, the committee determined that those disorders encompass the most disabling musculoskeletal conditions and that although rheumatoid arthritis and psoriatic arthritis are classified by SSA as “immune disorders,” their most common—and, in many cases, most disabling—manifestation is inflammation of the joints leading to joint destruction and deformity. The committee thus decided that those conditions merited consideration as leading causes of musculoskeletal impairment.
Chronic low back pain is a primary musculoskeletal pain condition defined by pain for more than 3 months. It is highly prevalent in all adult age groups and is the top cause of years lived with disability. Chronic low back pain is sometimes associated with pain that radiates to the lower
extremity in a characteristic distribution (i.e., radicular pain, sometimes called “sciatica” or radiculopathy). The presence of radicular pain or radiculopathy is associated with worse chronic low back pain severity and functional outcomes. Other factors associated with worse functional outcomes are co-existing medical and psychiatric conditions and other chronic pain conditions. In addition, the overuse of biomedical approaches to treat chronic low back pain (e.g., opioids and spine surgery) has been identified as a potentially important contributor to disability. However, numerous treatments have been shown to be effective in improving function in chronic low back pain, including exercise therapies, behavioral/psychologic therapies, and manual therapies. Multidisciplinary approaches, including intensive chronic pain rehabilitation programs and less intensive primary-care-based collaborative care management interventions, also have demonstrated benefits for function. In general, medications are less beneficial for function than for pain in those with chronic low back pain, with most benefits demonstrated only in the short term. The committee did not identify evidence about the likelihood or duration of the treatment required to reach a point at which low back pain is no longer disabling. There is no evidence that the efficacy of chronic back pain treatments differs by age.
Osteoarthritis is a disease that destroys synovial joints over time. There is no known cure or method of reversing the process. Chronic pain and joint stiffness are hallmarks of this condition. Osteoarthritis can become disabling if it is severe enough to make work and daily tasks difficult. It is most common in older people, and gender differences vary by age. Before age 45 more men than women have osteoarthritis, but after age 45 it is more common in women. The prevalence of symptomatic knee osteoarthritis increases with each decade of life, with the annual incidence being highest between 55 and 64 years old. Although there are numerous treatments available, progressive osteoarthritis may result in reduced mobility and the resulting systemic complications of immobility and deconditioning. There is moderate to strong evidence suggesting that exercise therapy and psychosocial interventions are effective for relieving pain and improving function for many patients with osteoarthritis pain. Complications can result from the use of anti-inflammatory medications. Although joint arthroplasties and fusions can relieve pain and improve function, they can also cause infection, deep vein thrombosis, and even intraoperative mortality. For those reasons, joint replacements and fusions should generally be considered only when non-surgical approaches have not been effective in controlling pain and providing acceptable function.
Inflammatory arthropathies are conditions characterized by inflammation of the joints and often other tissues. They include rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, juvenile idiopathic arthritis,
and systemic lupus erythematosus, among others. Rheumatoid arthritis and psoriatic arthritis are among the most common inflammatory arthropathies and are important causes of disability in adults.
Rheumatoid arthritis and psoriatic arthritis are systemic inflammatory diseases whose most common and prominent clinical manifestations include inflammation and destruction of the joints. These conditions are an important cause of work-related functional impairment. Effective treatments exist for rheumatoid arthritis and psoriatic arthritis, and the number of treatment options has expanded significantly in recent years as newer biologic agents have been approved. Because physical functioning is commonly assessed as a secondary outcome in trials of rheumatoid arthritis and psoriatic arthritis therapies, there is more evidence available about the impacts of specific treatments on functional capacity than for many of the other disabling medical conditions considered here.
Many existing pharmacologic treatments for rheumatoid arthritis and psoriatic arthritis have been found to improve physical functioning as measured with the Health Assessment Questionnaire Disability Index (HAQ), including a number of biologic disease-modifying antirheumatic drugs (DMARDs), which are indicated for more severe disease. However, the extent to which those therapies can improve work-related functional capacity among individuals with such severe impairments as to qualify for SSDI remains uncertain, for several reasons. First, few clinical trials have tested therapies among individuals with such severe impairments, so the treatment outcomes in this population are not well understood. Second, because the likelihood of functional improvement declines as the duration of disease and the number of prior DMARDs trials increase, treatment response is likely to be more modest among those with refractory disease. Third, both rheumatoid arthritis and psoriatic arthritis can result in irreversible joint damage, which may limit how much functional improvement can be achieved through medical management alone in the absence of surgery. Early diagnosis and treatment to prevent joint destruction and deformity is therefore of critical importance for patients with rheumatoid arthritis and psoriatic arthritis. It is unclear how much improvement might be expected among patients who do not receive early DMARD therapy. Finally, evidence linking specific rheumatoid arthritis and psoriatic arthritis therapies directly to work outcomes is extremely limited, although HAQ scores are highly correlated with work disability.