In this study, the actigraphy revealed important findings about SWD in HSCT recipients. TST and SL remained relatively stable over time, and medians were close to or within normal ranges seen in healthy adults. Experts recommend seven to eight hours of sleep, and normative values for SL in healthy adults is usually < 30 minutes [33, 34]; TST is likely to decrease and SL to increase with age [34]. SE showed a small improvement for five participants at 180 days. Of these five participants, only one was prescribed a sleep aid prior to transplant, suggesting participants were beginning to increase actual sleep even without sleep-aid use. Our finding that the median and mean SE was suboptimal at each time point is consistent with Nelson et al. (2018), who reported an average SE of 78% in autologous HCT recipients at 6–18 months after transplant [35]. SE ≥ 85% is considered an indicator of good sleep quality; SE below 84% is poor, and ratings below 75% can be indicative of sleeping disorders like insomnia [34, 36]. SE decreases with age, which could partially explain poor levels in this sample [34]. SE also reflects the duration of WASO, which remained twice the desired ≤ 30 minutes per night in this study, in line with WASO estimates in the survivorship period following HSCT [35]. Overall, the findings are consistent with the literature describing sleep-maintenance problems among HSCT recipients [3, 7, 35]. Such SWD may be improved by decreasing fragmented sleep stemming from wake time in bed, perhaps with behavioral interventions to reduce the number and duration of nighttime arousals.
The plot illustrating individual trajectories of TST and SE suggests there are distinct phenotypes for TST: a) consistently adequate TST, b) adequate TST that worsens, and c) inadequate TST that improves. Likewise, SE also demonstrates individual phenotypes: a) consistently good SE, b) moderate SE with some improvement or worsening at T3, and c) poor SE with variable improvement. These phenotypes are comparable to classifications based on sleep disruption in HSCT recipients developed by Jim and colleagues [37]. Overall, our findings suggest that SWD can be very individual and treatment factors may influence sleep adequacy, efficiency, and quality in this population [37].
Median ESS levels < 10 indicate that participants experienced minimal sleepiness; however, over time there was a trend toward heightened sleepiness, which did not reach significance. Median ISI scores indicate most participants experienced minimal or subthreshold (transient) insomnia symptoms, which is generally consistent with estimates in the survivorship period following HSCT [3, 35, 38, 39]. Investigations of SWD in HSCT recipients, however, often include autologous transplant recipients with different rates of SWD than allogeneic recipients, who receive steroids that can interfere with sleep [35, 40]. In one study, for example, investigators found SWD was more pronounced in allogeneic transplant patients than in autologous [41]. One would expect that the ISI score incrementally improves after transplant; however, ISI scores remained stable across time similar to TST and SL, suggesting that SWD was relatively unchanged between day 100–180 post-HSCT. Lack of improvement may reflect a mild degree of insomnia with limited room to improve, recall bias, or small sample size.
FCRI scores were at clinically significant levels at T1 and T2, then decreased to a subclinical level at T3. Consistent with our findings, Nelson and colleagues reported moderate FCRI scores (M = 15.67, SD = 8.04) in 84 autologous HSCT recipients at 6–18 months post-transplant [35]. In contrast, other investigators used the short version of the FoP-Q-SF [42] to assess FCR and reported high levels of FCR at 100 days and one year or more after allogeneic HSCT [43, 44]. A systematic review of 130 studies of mixed cancer survivors corroborated low–moderate levels of FCR that remained stable over the survivorship trajectory [45].
Median anxiety scores indicate mild–moderate levels with slight increase at T2 and then a decrease at T3. Evidence suggests anxiety is highest in anticipation of the transplant but reaches stability post-HSCT [46]. As our findings suggest, a subset of HSCT patients continue to experience significant levels of anxiety after HSCT, which may impact recovery, function, and health outcomes [47].
For depression, median scores indicate mild fluctuating levels of distress during the early post-HSCT period. Other researchers have reported that depression is prevalent (35%) during all stages of HSCT treatment up to five years after the transplant [46, 48]. In a prospective study of autologous and allogeneic transplant recipients, El-Jawahri et al. found that depressed patients had a three-fold greater risk of dying than non-depressed patients between 6 and 23 months after HSCT [48]. This highlights the need for timely assessment and treatment of depression using diagnostic tools developed specifically for HSCT patients [47].
Median fatigue levels remained in the mild range. Cancer-related fatigue may be due to SWD, cytotoxic therapy-induced cellular damage and repair efforts, cytokine activation, anemia, poor nutritional status, and/or the demands of coping with treatment [49, 50]. The increase in fatigue from T1–T3 in the present study raises the question of whether participants become more physically and mentally active at home as the distance from transplant increases, resulting in higher perceived fatigue.
There were significant associations between ISI and anxiety, FCRI and anxiety, and FCRI and fatigue. In previous studies of women with breast cancer, distress and FCR have been associated with SWD [51, 52]. Other studies have reported an association between FCR and anxiety and depression in cancer survivors [45, 53]. Few studies, however, have examined the relationship between psychosocial factors and FCR or SWD in HSCT survivors [35]. One recent study of patients with hematological cancer reported a strong relationship between FCR and anxiety, but not depression [54].
Regarding the association between FCR and fatigue, a possible explanation is that ongoing stress from the threat of treatment failure may contribute to tiredness and fatigue. SWD, fatigue, and psychological distress may also share common mechanisms and have synergistic effects. SWD and resulting fatigue may arise from alterations in sleep-regulating hormones, effects of adjuvant medications, and/or psychological disturbances due to worries about health, family, and financial impact of illness [55].
The enrollment and retention of willing participants indicates study procedures were acceptable and associated measures were not overly burdensome. Other factors contributing to the studies success included support from nursing leadership and staff, a dedicated study coordinator, ample time to explain study processes, monetary incentives, and willingness to enroll. Barriers included scheduling participant appointments, varying levels of enthusiasm for repeated measures, and questions about how many days to wear the Actiwatch. Participant responses demonstrate that retaining HSCT participants in a study examining sleep and psychosocial factors is possible on a larger scale.
The current study has several strengths including a clinically meaningful question, prospective longitudinal design, and subjective and objective measures of sleep. There are several limitations that should also be noted. The purpose of this study was exploratory; findings from such a small sample and recruitment at a single center limit generalizability. Moreover, patients’ sleep patterns before transplant were not measured, which would have been helpful to compare with SWD post-transplant. Finally, the ESS, ISI, and PROMIS® anxiety, depression, and fatigue instruments ask participants to evaluate their status retrospectively during the past one to two weeks. Although this approach is valuable, it requires participants not only to remember their subjective sleep, anxiety, depression, and fatigue but also to average and evaluate these perceptions.
Despite study limitations, our findings call attention to SWD issues HSCT recipients experience, underscoring the limited understanding of underlying factors and consequences of SWD, psychological distress, and fatigue that accompany HSCT. Larger sample size and extending the study beyond 180 days may help determine at what point SWD consistently improves. Although pharmacological and behavioral interventions have been successful in managing fatigue and psychosocial challenges after HSCT, additional research will determine the most optimal assessment tools, intervention strategies, and long-term care to address psychiatric comorbidity in this understudied patient population [47]. Our results highlight the need for more research to prevent or mitigate long-term morbidity after HSCT, since late effects are associated with poor health, diminished functional status, and higher healthcare costs. Possible sleep/fatigue and psychosocial interventions could be tested at the time of transplantation or during early post-transplantation, including comprehensive supportive care, patient education, and closer monitoring in high-risk patients as a part of long-term survivorship. Close coordination with the transplantation center and encouraging self-management support for patients and families has the potential to improve survival and quality of life.