DISCUSSION
We examined potential racial disparities in QoL two or more years post-HSCT in a regional cohort of pediatric patients. Our results demonstrate no significant racial disparities in QoL across a variety of functioning scales during this post-transplant window. Socioeconomic factors did not appear to have any significant impact on this relationship between race and QoL in our patient cohort. However, transplant -related factors did impact certain QoL metrics, even after accounting for certain socioeconomic factors. These findings can provide critical information to providers and patient families concerning potential avenues for growth in managing long-term patient wellbeing following HSCT.
Given multiple reports of racial disparities affecting clinical outcomes related to HSCT.19-22, the current findings of no effect of patient race in long-term QoL scores were surprising. Prior published studies that draw upon much larger databases or registries and include participants from varied socioeconomic backgrounds have found racial and socioeconomic disparities in access to fully-matched unrelated donors for HSCT and in overall completion of HSCT for malignancy.18,27,28 There are numerous explanations for the lack of racial disparities in QoL scores post-HSCT within the current study. As a single institution evaluation in a moderately-sized pediatric hospital, the absence of racial disparities in readjustment rates at least two years post-HSCT may not necessarily indicate a lack of racial disparities in transplant outcomes within the United States. For example, it was noted in this study that a significant majority of patients (74.1%) came from estimated annual income bracket of above $50,000; this indicates a patient population with a higher socioeconomic status than that of the general population, which had a median income of $55,000 in 2015.26 Such higher income levels could allow patients and families to better handle post-HSCT life and afford resources to improve their physical and emotional health.29-31
These findings are potentially attributable to a very organized supportive care program present throughout the transplant process within this institution. During both the immediate and later post-transplant periods, we utilize dedicated mental health specialists, hospital-based schoolteachers, social worker services, inpatient and outpatient pain services, multidisciplinary survivor clinics, and post-HSCT community engagement programs to optimize patient wellbeing; such programs may identify and resolve early declines in physical, emotional, social, and school/work functioning following treatment.32-35Findings from our institution may be able to provide insight to other transplant centers regarding global changes that can be made to ensure patients from marginalized groups experience optimal outcomes. This includes educating transplant center leadership and staff about racial disparities that may be present in transplant referral, ensuring racial and ethnic minorities have adequate psychosocial support from community and hospital services, and implementing multidisciplinary survivor clinics into other transplant centers to monitor all patients years after transplant.36,37
The results from this study can be contrasted with other work evaluating QoL in healthy controls, cancer patients, and transplant patients via the PedsQL TM 4.0 questionnaire (Table 5). While the QoL scores of patients in this study were still slightly below those of healthy controls,23 their post-transplant QoL scores were greater than those of pediatric cancer patients across all the categories of readjustment,38 and were higher than cited transplant QoL scores for emotional and work/school functioning.39 The report comprised of healthy controls evaluated a group of patients from California with similar racial and ethnic demographics to the transplant population at our institution, making this a reasonable comparison.23,24The study sample’s QoL scores were also comparable to that of pediatric liver transplant recipients 10 years post-transplant39as well as pediatric HSCT recipients in other published studies .40,41 These non-statistical comparisons indicate that patients undergoing HSCT at our institution are adaptive, on average exhibiting nearly healthy functioning two years post-transplant. Further research should examine the role of institution-specific influences, such as effective long-term monitoring of possible complications including heart failure and endocrinopathies and an emphasis on psychosocial readjustment and transition to adult care.
When pediatric patients were divided based on their diagnosis of malignant or non-malignant conditions, regardless of race, patients with malignant conditions reported significantly lower physical and work/school functioning in preliminary analyses. Patients with malignant conditions likely face more physical limitations due to their cancer diagnosis; the physical side effects of cancer therapies and myeloablative conditioning regimens also likely persist for years following transplant, inhibiting long-term improvements in QoL. Those with malignant conditions necessitating HSCT also likely spend a greater amount of time away from school or work due to hospitalizations or treatments related to their diagnosis, impacting their work/school functioning.42 The significant relationship between type of diagnosis and physical functioning seen in multivariate regression analysis persisted when estimated household income was factored into the analysis, suggesting that even those with higher income experience lasting physical challenges. Conversely, relationships of diagnosis and school functioning were not present in multivariate analyses. These differences in outcomes highlight a need for provider intervention and follow-up to optimize wellbeing in patients with malignancies. Outside of racial and ethnic groups, particular care and focus should be given to patients with malignant conditions within institutions and transplant centers to improve physical and academic outcomes.
Limitations of this study include retrospective data collection, which precludes inferring causal relationships between covariates. The relationships we describe are based upon the inclusion of self-reported QoL from pediatric patients and estimates of socioeconomic status based upon medical records; it is possible that the nature of these metrics biased subsequent analyses. While there is no gold standard for measuring pediatric QoL, the PedsQL not only has strong psychometric properties,23,24 but is also appropriate for use in our population of interest.38 Nonetheless, there is the potential for differences in self-reporting across race and ethnicity that impact metrics like the PedsQL.43 We have no means of quantifying potential reporting bias, but rely on previous research suggesting the strength of parent-reported health status of children.23,44 Lastly, the results of our study should be interpreted in the context of a small sample size (n = 86) as a single center experience. While we had a diverse sample, subsequent work would benefit from exploring the relationships described herein using larger samples sizes and multi-center approaches.