FormalPara Key Summary Points

Why carry out this study?

Chronic graft-versus-host disease (GVHD) following hematopoietic stem cell transplantation (HSCT) is associated with poor health-related quality of life (HRQoL) and functional status.

Data are lacking on GVHD-related disability, particularly regarding activity limitations from the patient perspective.

The objective of this analysis is to assess physical, cognitive, and work disability and describe factors predictive of disability in patients with chronic GVHD in the potentially employable general workforce using patient-reported data.

What was learned from the study?

Severe cognitive, severe physical, and work disability were associated with chronic GVHD, and symptom burden, sex, race, and transplant history were predictive of disability.

These findings help to define and better understand chronic GVHD-associated disability.

Introduction

Allogeneic hematopoietic stem cell transplantation (HSCT) represents the only potentially curative treatment modality for many high-risk hematologic malignancies and bone marrow disorders [1,2,3]. Despite routine use of immunosuppressive prophylaxis, chronic graft-versus-host disease (GVHD) develops in approximately 30% to 50% of HSCT recipients [4,5,6,7,8] and is a leading cause of late morbidity and mortality post-transplant [7, 9,10,11]. Chronic GVHD is characterized by multiorgan pathology and burdensome symptoms that negatively impact health-related quality of life (HRQoL) and physical function [12, 13]. Studies of long-term HSCT survivors have demonstrated that active chronic GVHD is associated with significantly impaired physical function compared with patients without chronic GVHD or those with resolved chronic GVHD [14, 15]. Further, patients with chronic GVHD may be unable to return to work [16, 17], which further underlines the potential impact of GVHD-related disability.

Despite the growing evidence of physical and functional impairment associated with chronic GVHD, there are limited data on disability as an outcome measure of disease burden or response to treatment in clinical trials [18]. The World Health Organization defines disability as any physical or mental impairment that limits people’s ability to do certain activities or interact with their environment and encompasses impairment in physical or mental functioning, activity limitation, and participation restrictions [18, 19]. Few studies have evaluated disability specifically in the context of chronic GVHD, and attempts to describe disability have focused largely on physician-assessed clinical manifestations [e.g., Flowers criteria (bronchiolitis obliterans, keratoconjunctivitis sicca, scleroderma, joint/fascia involvement, and esophageal stricture)] [16, 18]. A previous study from the Chronic GVHD Consortium explored definitions of disability among patients with chronic GVHD based on Flowers criteria as well as general health status (Karnofsky performance status) and patient-reported measures (Human Activity Profile) [18]. Disability as defined by Flowers was associated with worsening Human Activity Profile and Karnofsky performance status scores over an 18-month follow-up period. Further, progressive disability correlated with low response to therapies [18].

Chronic GVHD–related disability in the context of specific activity limitations and restrictions from a patient perspective still needs to be better characterized. Further, identification of factors associated with chronic GVHD–related disability could aid in (1) recognition of patients most at risk for limitations and (2) optimization of disease management strategies. Previously reported findings from the online Living With Chronic GVHD Patient Survey have demonstrated that chronic GVHD symptoms severely interfere with patients’ ability to perform activities of daily living [20, 21]. Many respondents reported adverse employment and economic outcomes due to their chronic GVHD [20, 21]. Notably, in a separate study, inability to work was significantly associated with mortality among patients with chronic GVHD based on a multivariable analysis of sociodemographic factors [17]. Thus, our objective was to assess physical, cognitive, and work disability and identify factors contributing to disability in patients with chronic GVHD in the potentially employable general workforce using direct patient input from the Living With Chronic GVHD Patient Survey.

Methods

Study Design and Respondents

The Living With Chronic GVHD Patient Survey was an online, noninterventional, cross-sectional survey administered in May–August 2020 to adults living in the USA. Participants were recruited through consumer panels (survey providers) and patient advocacy groups via email blasts, newsletters, and Facebook posts [20]. Further details regarding advocacy groups have been described [20]. Survey eligibility criteria included respondent self-reports of receipt of prior allogeneic HSCT, chronic GVHD diagnosis from a healthcare provider, and active chronic GVHD within the past 5 years. Full survey details have been previously published [20]. In brief, the survey had 74 questions, and respondents were asked to provide information on demographics, disease diagnosis, chronic GVHD symptoms per the Lee Symptom Scale (LSS), impact of the respondent’s most severe chronic GVHD symptoms on activities of daily living, and work status. The LSS assesses 30 individual symptoms grouped into seven subdomains (eyes, energy, lung, mouth, nutrition, psychological, and skin); total LSS score ranges from 0 to 100, with higher scores indicating greater symptom burden [22]. Activities of daily living were graded on a scale of 0 (“chronic GVHD had no effect on my daily activities”) to 10 (“chronic GVHD completely prevented me from doing my daily activities”).

Definitions of Cognitive, Physical, and Work Disability

Descriptive and correlational analyses were used to create three composite definitions of disability: (1) severe cognitive disability, (2) severe physical disability, and (3) work disability. Assessment of cognitive disability included evaluation of the impact of chronic GVHD on specific activities of daily living, including managing personal finances, using a computer, and interacting socially with friends and family. Analysis of physical disability included effects on personal hygiene, dressing, eating, ability to use the restroom, ability to move around the house, and ability to get around outside the house as well as ability to prepare meals, shop, and do housework. A score of at least 7 (0–10 scale) on any item related to cognitive or physical disability was considered a severe limitation in the respective domain. Work disability was assessed based on two questions related to work status (“ever taken disability leave because of chronic GVHD” and “ever left a job because of chronic GVHD”). Respondents who answered “yes” to at least 1 of these questions were considered to have experienced work disability. Composite disability in each of the three subdomains was calculated as the percentage of respondents meeting the criteria for severe limitation on any individual item in the cognitive or physical domains or any disability in the work domain when symptoms were at their worst.

Ethical Considerations

The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. The survey was reviewed and approved by the New England Institutional Review Board before administration. All respondents provided informed consent via an online consent form before initiation of the survey. All study participants from the initial study consented to their data being used for further research purposes.

Statistical Analyses

Only respondents who were employed (reporting full- or part-time employment at time of transplant) or considered potentially employable in the general workforce (responses of unemployed and seeking employment, unemployed and not seeking employment, disabled or unable to work, or other) at the time of transplant were included in this analysis. Respondents identifying as self-employed, retired, or homemaker were excluded. Exclusion of self-employed respondents was based on the potential lack of required work hours, and therefore, the impact on employment may be different between self-employed versus standard employed participants. Respondent demographics and clinical characteristics were summarized using descriptive statistics.

To identify factors associated with disability, a two-step regression analysis was undertaken. First, univariable regression models were used to estimate the unconditional (marginal) association between each disability outcome and each predictor variable, with likelihood-ratio tests against the null model (excluding the variable) used to estimate the association. At the second step, predictor variables meeting threshold (P < 0.2) were considered for inclusion in a multivariable logistic regression model to estimate conditional associations. A backward step selection approach was adopted for the final model, with application of a 0.05 significance level.

Results

Respondents

As published previously [20], of 580 participants who entered the survey screener, 165 met all eligibility criteria and completed the survey (full analysis set). Briefly, primary reasons for study exclusion among initial survey respondents included declined or missing consent (n = 186), not having received prior HSCT (n = 111 including unknown/missing response), and not having a diagnosis of chronic GVHD (n = 48 including unknown). The majority of included respondents were recruited from BMT InfoNet (n = 110). Twenty-eight respondents in the full analysis set reported being self-employed, retired, or a homemaker at the time of their most recent transplant and were excluded from this analysis; the remaining 137 respondents were considered part of the potentially employable general workforce (i.e., employed or potentially interested in employment at the time of most recent transplant) and were included in this analysis. Demographics and clinical characteristics of the potentially employable respondent population were similar to those previously reported for the full analysis set (Table 1) [20]. Among the 137 respondents in the potentially employable analysis set, median (range) age was 54.0 (18–76) years, 88 (64.2%) were women, and 110 (80.3%) were white. More than half of respondents [n = 81 (59.1%)] had a bachelor’s degree [n = 44 (32.1%)] or advanced degree [n = 37 (27.0%)]. Among respondents considered potentially employable at the time of transplant, 59 (43.1%) reported being disabled or unable to work, and 28 (20.4%) and 12 (8.8%) reported being employed full- and part-time, respectively, at the time they completed the survey. Median (range) chronic GVHD duration was 4.5 (0.1–36.7) years. Respondents reported seeing a median (range) of 4.0 (0–10) specialists for their chronic GVHD, which most frequently included hematologists/oncologists (n = 128 [93.4%]) and ophthalmologists [n = 88 (64.2%)].

Table 1 Characteristics of living with chronic GVHD patient survey respondents (potentially employable general workforce subset)
Full size table

Disability

Nearly half of evaluable respondents [63/134 (47.0%)] met the study criteria for severe cognitive limitations based on their ability to perform specific tasks, and approximately two-thirds each met the criteria for severe physical disability [91/135 (67.4%)] and work disability [86/137 (62.8%)] associated with chronic GVHD (Fig. 1). With respect to cognitive disability, the greatest percentage of survey respondents met study criteria for severe limitations in their ability to interact socially with friends or family members [56/137 (40.9%); Fig. 2]. The highest rates of study-defined physical disability occurred for subdomains of shopping [60/137 (43.8%)], housework [57/137 (41.6%)], and eating [51/137 (37.2%); Fig. 3]. More than half of survey respondents [78/137 (56.9%)] reported taking at least one disability leave owing to chronic GVHD (Fig. 4).

Fig. 1
figure 1

Respondents reporting severe cognitive, severe physical, or work disability. The analysis was conducted in the potentially employable general workforce subset (N = 137). Respondents were asked to consider impact of chronic GVHD at the time symptoms were at their worst. Missing observations [i.e., respondents reporting that any of the activity limitation items were not applicable with no report of severe disability (≥ 7 on a 0–10 scale in any of the other items)] were not included in the percentage calculations; data labels above bars indicate n/N evaluable. Severe cognitive disability was defined as severe limitation (≥ 7 on a 0–10 scale) in ≥ 1 of the specified items (managing personal finances, social interaction, and computer use). Severe physical disability was defined as severe limitation (≥ 7 on a 0–10 scale) in ≥ 1 of the specified items (personal hygiene, dressing, eating, using the restroom, moving around the house, preparing meals, shopping, doing housework, and getting around outside the house). Work disability was defined as response of “yes” on ≥ 1 of the specified items (took disability leave or left a job). GVHD graft-versus-host disease

Full size image
Fig. 2
figure 2

Respondents reporting limitations in specific cognitive activities. The analysis was conducted in the potentially employable general workforce subset (N = 137). Respondents were asked to consider impact of chronic GVHD at the time symptoms were at their worst. Severe limitation was defined as a score of ≥ 7 on a 0–10 scale. GVHD graft-versus-host disease

Full size image
Fig. 3
figure 3

Respondents reporting limitations in specific physical activities. The analysis was conducted in the potentially employable general workforce subset (N = 137). Respondents were asked to consider impact of chronic GVHD at the time symptoms were at their worst. Severe limitation was defined as a score of ≥ 7 on a 0–10 scale. GVHD graft-versus-host disease

Full size image
Fig. 4
figure 4

Respondents reporting limitations in specific work ability. The analysis was conducted in the potentially employable general workforce subset (N = 137). Respondents were asked to consider impact of chronic GVHD at the time symptoms were at their worst. Severe limitation was defined as a score of ≥ 7 on a 0–10 scale. GVHD graft-versus-host disease

Full size image

Factors Associated with Disability

Results of a univariable analysis demonstrated that chronic GVHD severity/duration, LSS symptoms (eye, mouth, lung, skin, nutrition, energy, and psychological), and number of transplant specialists consulted were associated with all types of disability (Table 2). Furthermore, number of treatments was associated with both severe cognitive and physical disability. Nonwhite race was predictive of work disability. Age, sex, education, employment, and annual income were not associated with any type of disability in the univariable analysis, although female sex was associated with the physical disability attributes of preparing meals, shopping, and housework (chi-square test, 8.0; P = 0.005).

Table 2 Univariable analysis of factors associated with disability (potentially employable general workforce subset; N = 137)
Full size table

Findings from a multivariable analysis of factors associated with disability demonstrated that LSS skin and psychological symptoms as well as number of transplant specialists consulted remained significantly associated with severe cognitive disability (Table 3). Female sex and LSS mouth, nutrition, and energy symptoms were associated with physical disability. Nonwhite race and LSS energy symptoms were associated with work disability.

Table 3 Multivariable analysis of factors associated with disability (potentially employable general workforce subset; N = 137)
Full size table

Discussion

Among HSCT survivors who completed the Living With Chronic GVHD Patient Survey and were considered part of the potentially employable general workforce at the time of their most recent transplant, nearly half of respondents (47.0%) met the study criteria for severe cognitive limitations based on their ability to perform specific tasks, and approximately two-thirds met the criteria for severe physical disability and work disability (67.4% and 62.8%, respectively) associated with their chronic GVHD. In particular, more than one-third of respondents were categorized in this analysis as having severe limitations in the individual domains of interacting socially with friends/family members, shopping, housework, eating, and ability to get around outside of the house. More than half (57%) took disability leave from work, with 28% leaving a job due to chronic GVHD. These findings highlight the immense burden of chronic GVHD on activities of daily living that may be underappreciated with respect to its impact on HRQoL. Our findings are largely consistent with previous studies that demonstrated the negative overall impact of chronic GVHD on HRQoL, including the impact on psychological state, mental functioning, social interactions, and physical functioning and functional status [12,13,14,15, 23,24,25,26]. Furthermore, two longitudinal studies of patients who underwent HSCT demonstrated neurocognitive and motor dexterity deficits 80 days after transplant that often improved by 1 year, although deficits persisted for at least 40% of survivors at 5 years [27, 28]. Although neither study focused on patients with chronic GVHD, those receiving chronic GVHD treatment experienced impaired motor dexterity but not cognitive ability 1 year following HSCT [28].

Data that address the association between disability and chronic GVHD are limited. Previous analyses have begun to explore definitions of disability in chronic GVHD, demonstrating associations with chronic GVHD organ impairment, functional status, activity, and employment, as well as response to therapy. Analyses from the Living With Chronic GVHD Patient Survey have demonstrated that patients frequently experience protracted and severe symptoms that interfere with both basic and instrumental activities of daily living [20]. Moreover, many survey respondents reported adverse employment outcomes including missed time as well as lost career opportunities and income as a result of their chronic GVHD [21]. Findings from the present analysis build upon these previous reports to further delineate the effects of physical, cognitive, and work disability among patients with chronic GVHD. Exploratory analyses revealed several factors potentially associated with composite disability, including specific symptoms as assessed by the LSS and the number of transplant specialists consulted, which were generally predictive of all types of disability in the univariable analysis. Interestingly, on multivariable analysis, the number of transplant specialists consulted remained significantly associated with severe cognitive disability only, female sex with severe physical disability, and nonwhite race with work disability. It remains unclear what factors contributed to the disproportionate effects of chronic GVHD on physical disability in women versus men as the survey did not collect information on the physical demands of respondents’ lives. The effects of nonwhite race on work disability may reflect the diminished capacity to leverage support networks and provisions allowing access to work (e.g., remote home-based work) that minority communities face in general [29,30,31]. However, these explanations are speculative. Additional studies are needed to confirm and further explore the relationships.

This study was potentially limited by several factors. Recruitment was restricted to patient advocacy groups and consumer panels. Additionally, self-employed respondents were excluded based on the assumption that set working hour requirements differed from those of standard employed respondents. Therefore, the potential for selection bias exists regarding who completed the survey. Further, representation of racial and ethnic minorities was relatively low. A greater percentage of the survey respondents held a bachelor’s or advanced degree than in the general US population [32]. As a result, respondent demographics may not be reflective of the transplant and chronic GVHD patient population at large. These limitations may impact the findings as demographics including race, ethnicity, and level of education are known to be associated with disability [33, 34]. For example, disability rates are higher among individuals with lower levels of education and non-Hispanic Black adults than those with higher levels of education and white or Hispanic adults [33, 34]. Additionally, associations between disability and race identified in the current analysis were based on small population sizes. Chronic GVHD diagnoses and clinical characteristics were self-reported by survey respondents and were not confirmed by a physician. Finally, the survey may have been subject to recall bias, as respondents were asked about the time when symptoms were at their worst over a potentially long recall period. Additional investigations, including longitudinal studies of patients at the same stage of their clinical course, analyzing the impact of work type, sex and racial disparities, and specific chronic GVHD symptoms are needed to further understand this relationship and improve the overall HRQoL of long-term HSCT survivors.

Conclusions

In the potentially employable subset of survey respondents with chronic GVHD, nearly half met the study criteria for severe cognitive disability based on their ability to perform specific tasks, and two-thirds each met the study criteria for severe physical disability and work-related disability associated with chronic GVHD. Furthermore, our findings suggest that factors, including presence of specific symptoms and the number of transplant specialists consulted, may be associated with severe disability. These observations reveal a need for improved social support for patients with chronic GVHD, highlighting potential indicators for those most in need.