Objective The aim of the study was to determine the relationships between functional outcomes, clinical symptoms, and health-related quality of life among patients with malignant brain tumors receiving interdisciplinary outpatient rehabilitation.
Design A prospective study of 49 adults with malignant brain tumors participating in outpatient therapies was performed. Outcome measures included the Functional Assessment of Cancer Therapy-Brain (FACT-Br) for health-related quality of life and the Patient-Reported Outcome Measures Instrument Survey (PROMIS) Depression and Pain Behavior scales measured at admission, discharge, 1 and 3 mos after discharge. Day Rehabilitation Outcome Scale (DayROS), a functional measure, was measured at admission and discharge.
Results The FACT-Br scores, PROMIS pain, and PROMIS depression scores did not significantly change. There were many negative associations seen between FACT-Br and PROMIS depression (all P < .0001) and less associations with PROMIS pain. There was a positive correlation between Day Rehabilitation Outcome Scale and FACT-Br (P = .0058) and a negative association with PROMIS pain (P = .028), but not with PROMIS depression. There were no correlations between Day Rehabilitation Outcome Scale gains and change in PROMIS depression, FACT-Br total, or PROMIS pain.
Conclusions Health-related quality of life, pain, and depression did not worsen. Patients who reported less depression and pain had better reported health-related quality of life. Level of function was also associated with HRQOL and pain, but not depression.
More than 138,000 people currently live with a malignant brain tumor in the United States, with more than 24,000 new cases in 2013.1 Tumors involving the brain can be a significant source of disability and reduced health-related quality of life (HRQOL), with the potential for causing a combination of physical, cognitive, and communication deficits, as well as cancer-related symptoms of pain and depression. This constellation of deficits and cancer-related symptoms may have devastating effects on HRQOL.2
Health-related quality of life, composed of physical, emotional, social, and role/functional aspects of well-being, has long been neglected in the brain tumor literature, as opposed to other traditional outcomes such as survival and neurologic functioning. This is likely due to the overall poor prognosis of brain tumors despite advances in standard therapy, difficulty obtaining these patient-reported outcome (PRO) measures, and perception that the disease may reduce neurocognitive function and ability to accurately report HRQOL.3 However, HRQOL is increasingly becoming an important secondary end point in clinical trials of new treatments for brain tumors because it provides the patient’s view on the burden of the disease, morbidity of the treatment, capacity to perform normal daily activities, and may be linked to overall survival.4,5
Health-related quality of life in patients with brain tumors has shown a consistent positive correlation with pain6 and depression.7 It has been reported that those who have less anxiety and depression have better HRQOL.8,9 Anxiety and depression have been found to be strong determinants of all of the components of HRQOL, except for physical functioning.8 In addition, pain has been reported to be independently related to HRQOL,7 and those with brain tumor recurrence have more pain and lower HRQOL than those without recurrence.10
There is also a correlation between functional status and HRQOL. A systematic review of 48 studies found a consistent relationship between physical function and HRQOL in patients with malignant brain tumors, along with other symptoms such as depression and fatigue.2 The negative effects on HRQOL and physical function in patients with malignant brain tumors are particularly evident in patients with high-grade primary and metastatic tumors.4,7,11–17
Unfortunately, despite advances in detection and treatment, the overall 5-yr survival rate of malignant brain tumors remains low.18 The disease is often rapidly progressive, and the cure rate is low, resulting in a limited life span and significant morbidity as the disease progresses. With advances in diagnosis and treatment, survival times are increasing but often at the expense of aggressive therapy that can also be a significant cause of disability and impaired HRQOL.4,19Interestingly, maintaining independent functional status has been associated with prolonged survival in newly diagnosed glioblastoma multiforme after surgical resection.20 Thus, there has been an increased focus on both HRQOL and disability in persons with brain tumors, as well as the role of rehabilitation in the management of functional impairments and disabilities. It also makes sense that those with less pain and depression would do better in rehabilitation outcomes in this patient population—this hypothesis still needs to be researched, however.
There are several studies that have looked at brain tumor rehabilitation outcomes, typically in an acute inpatient rehabilitation setting; however, few include HRQOL measures.21–30 One small study looking prospectively at 10 mixed brain tumor patients during inpatient rehabilitation found that patients made functional improvements with overall improvement in HRQOL 3 mos after discharge from acute inpatient rehabilitation.31
Although rehabilitation for patients with brain tumors often continues in an outpatient setting after discharge from acute inpatient rehabilitation, very few studies have looked at outcomes from brain tumor patients treated in an outpatient rehabilitation program. Cheville et al32 conducted a study of a multidisciplinary outpatient intervention on 103 adults undergoing radiation therapy for advanced cancer, including 12 brain tumor patients, and found an improvement in physical well-being, but not fatigue. Gehring et al33 studied an outpatient cognitive rehabilitation program in 140 adults with low-grade gliomas and found initial improvements in subjective reports of cognition with sustained benefits in attention, verbal memory, and mental fatigue. Sherer et al34 applied an outpatient traumatic brain injury rehabilitation program to 13 patients with primary brain tumors and found improvement in functional independence and work productivity. This study did not specifically address any HRQOL measures.34 Another study looked at the effect of a home neurorehabilitation program in a group of 121 patients with malignant brain tumors and found functional improvement in about a third of their patients.35 Furthermore, they found significant improvement in at least one domain of HRQOL after 3 mos of rehabilitation regardless of functional gain; thus, it did not explain whether the improvement in HRQOL was associated with the functional gain during rehabilitation.35
This study is one of the first studies to investigate the relationships among functional outcome, clinical symptoms, and HRQOL, in an interdisciplinary outpatient rehabilitation program. We hypothesized that patients with malignant brain tumors who received interdisciplinary outpatient rehabilitation services would have improvement in HRQOL, pain, and depression at the conclusion of rehabilitation and at 1 and 3 mos after discharge from the rehabilitation program. We also expected to find a correlation between subjective reports of HRQOL, pain, depression, and function in malignant brain tumor patients.
METHODS
Study Design
A nonrandomized, prospective, longitudinal study was conducted. The study staff approached all subjects with malignant brain tumors enrolled in day rehabilitation during the study period. McGaw Medical Center Northwestern University Institutional Review Board approval was obtained before study initiation (registry number STU00040479), and written informed consent was obtained from each subject. Baseline information was obtained from chart review at admission and included demographic information, oncologic history (time from diagnosis; tumor location and grade; metastatic vs. primary; treatment history; and active treatments), other significant medical and surgical history, and previous rehabilitation interventions received. The patient, with caregiver assistance if necessary, completed study measures upon enrollment and discharge from the program and were given two additional sets of forms with instructions to complete and mail back 1 and 3 mos after discharge from the program. The research staff maintained study contact by interval phone calls at each follow-up point.
Subjects
Forty-nine adults with malignant brain tumors were consecutively enrolled into an interdisciplinary outpatient rehabilitation program conducted at six affiliated day rehabilitation sites for 35 mos. Patients were not required to participate in this study to participate in the day rehabilitation program.
Because there is a difference in survival and outcomes among individuals with different brain tumor types, we chose individuals with higher-grade brain tumors to be the study group to keep consistency. Inclusion criteria were being 18 yrs or older with a diagnosis of a malignant brain tumor (anaplastic astrocytoma or grade 3 astrocytoma, glioblastoma multiforme or grade 4 astrocytoma, brainstem glioma, anaplastic oligodendroglioma, atypical/anaplastic meningioma, central nervous system lymphoma, and metastatic brain tumors) and participating in the day rehabilitation program.
Individuals with lower-grade brain tumors or other neurological diseases were excluded from the study. The exclusion criteria were a diagnosis of benign brain tumors (pilocytic astrocytoma or grade 1 astrocytoma, low-grade astrocytoma or grade 2 astrocytoma, low-grade oligodendroglioma, typical meningioma, acoustic neuroma, pituitary adenoma, dermoid tumor, epidermoid tumor, chordoma, colloid cyst, ganglioglioma, and pineal gland tumor), as well as tumors that are more common in the pediatric population (medulloblastoma, craniopharyngioma, and Rathke cleft cyst). Other exclusion criteria were a history of spinal cord tumors, traumatic brain injury, and stroke.
Day Rehabilitation Program
The protocol for this program is standard in our institution for all of the day rehabilitation sites. Other interdisciplinary outpatient rehabilitation programs are similar but may not be exactly the same as ours. A patient participated in at least two of the three disciplines of therapy (PT, OT, and SLP), generally from 2 half days per wk (3 hrs/d) to 5 full days per wk (6 hrs/d). Disciplines of therapy were determined by the patient’s needs after a full functional evaluation by a physiatrist. Other services offered included nursing and physician care, vocational rehabilitation, and psychology counseling. If psychological services were not covered by the patient’s insurance, funding was available via charitable funds as part of this study. Each patient was entered in a specific “track” based on his or her functional goals. These tracks included basic activities of daily living (ADLs), complex ADL, and return to work/school (Table 1). The length of treatment varied for each individual, depending on gains in therapy, goals, insurance limitations, and possible medical complications.
TABLE 1
Measures
The Functional Assessment of Cancer Treatment (FACT)-G questionnaire is a general HRQOL instrument that asks the patient to rate whether a statement is “not at all” to “very much” true for the past 7 days. The FACT-G domains include physical well-being, social/family well-being, emotional well-being, and functional well-being.36 The Functional Assessment of Cancer Therapy Additional Concerns for Brain Tumor patients (FACT-Br) includes the general FACT-G questions plus additional questions designed to assess concerns specific to patients with brain tumors including cognition, personality, functional independence, and neurologic symptoms that are written in the same style as the FACT-G questions.36 The FACT-Br questions were scored on a 0–4 Likert scale, with some reverse items, such that an overall domain score of 0 is the worst HRQOL and 200 is the best HRQOL. Patients were included in the analysis if they had more than 80% valid items in their responses.
The Patient-Reported Outcome Measures Instrument Survey (PROMIS) Depression and Pain Behavior scales (short forms) were used to measure depression and pain, respectively. The PROMIS instruments assess patient-reported health status for physical, mental, and social well-being to reliably and validly measure patient-reported outcomes via item banks in several domains.37,38 The eight PROMIS depression questions were scored on a 1–5 Likert scale, corresponding to patient ratings of “never” to “always,” with an overall PROMIS depression score of 40 being the most serious depression and 8 being limited depression. The seven PROMIS Pain Behavior questions were scored on a 1–6 Likert scale, corresponding to patient ratings of “had no pain” to “always,” with an overall PROMIS Pain Behavior score of 42 being the worst and 7 being the best pain behavior score.
Day Rehabilitation Outcome Scale (DayROS) is the Rehabilitation Institute of Chicago day rehabilitation outcome scale, similar to the Functional Independence Measure. The DayROS is a derivative of the Rehabilitation Institute of Chicago-FAS (the Rehabilitation Institute of Chicago Functional Assessment Scale), first presented by Cichowski, Heinemann, and Wu at the 64th session of the American Congress of Rehabilitation Medicine. The Rehabilitation Institute of Chicago-FAS was also used in an article published by Sliwa et al.39 The DayROS contains the following three scores: ADLs, mobility, and cognition. There is a possible 7 points for each item of these scores, with 7 being the most functional. The various tracts had different functional activities that made up the mobility, ADL, and communication scores (Table 1). An example of this would be mobility in the complex ADL tract that is only made up of the locomotion score, but mobility in the basic ADL tract includes transfers, bed mobility, wheelchair propulsion, and ambulation. A percent of the total possible score for each component of the mobility, ADL, and communication scores was derived. The total DayROS score for each tract was determined by adding the item scores and computing the percent of the total possible. The DayROS scores can range from 1 to 100. In addition, DayROS gain was determined as Discharge DayROS–Admission DayROS. The DayROS gain efficiency was computed by dividing the DayROS gain score by the total number of therapy days attended by the patient.
Statistical Methods
Distributions of each scale at the four evaluation time points were summarized using box plots. Correlation between pairs of scales at each evaluation time point was analyzed using univariate linear regression models. Adjusted R 2 from these models is reported as the summary of the strength of the association. The relationship between QOL changes from admission to discharge (outcome variable) and DayROS at admission or DayROS gain was analyzed similarly using linear regression models. Unadjusted P values from these analyses are reported.
A more comprehensive analysis looking at changes over time (admission, discharge, 1- and 3-mo follow-up) was conducted for all scales except DayROS, which was only measured at admission and discharge. First, scales were plotted over time, and a nonparametric lowess smoother was used as a graphical tool to visualize the pattern of change over time. Changes in each scale were then analyzed using separate generalized linear mixed models with repeated measures, which provide an appropriate mechanism for handing missing data and use all available observations. The scale score was the outcome variable, and the assessment point was a categorical fixed effect. Within-subject correlation between repeated assessments was modeled using AR (1) covariance structure. Repeated measures analysis of variance models were also used to assess the linear relationship between pairs of scales across all time points jointly, accounting for the within-patient correlation using AR (1) covariance structure. Analyses were performed using PROC MIXED40 in the SAS software, and model goodness-of-fit was assessed using standard diagnostic measures available in PROC MIXED. In addition, sensitivity analyses were performed to examine the robustness of the findings to patterns of data missingness by refitting these models using only complete cases.
RESULTS
Subject Characteristics
We enrolled 49 patients. Forty-six patients (94%) completed the day rehabilitation program. One patient signed the consent form but then did not return for rehabilitation. One patient died during the study, and one patient left the program prematurely to go to acute inpatient rehabilitation but died shortly after the inpatient stay. Thirty-five patients (76%) completed the admission and discharge study forms. Twenty-two patients (48%) were included in the 1-mo follow-up, and twelve patients (26%) completed the 3-mo follow-up; all study forms were completed in these 12 participants (Fig. 1). The average age of the participants was 45.7 yrs (range = 12–84 yrs). Other patient demographics are listed in Table 2.
FIGURE 1
TABLE 2
Day Rehabilitation Program
The average length of stay in day rehabilitation was 76.9 days (range = 38–227 days). The average number of missed days was 5.19 days (range = 0–22 days). The average number of therapy visits was 26.1 visits (range = 8–37). All patients received physical and occupational therapies; three patients did not receive speech language pathology. Unit of therapy was determined by taking the number of visits and multiplying by 4. The average total units of therapy were 108 units (range = 12–513 units) for PT, 119 units (range = 33–581) for OT, and 105 units (range = 0–484) for speech therapy. Twenty-seven patients (55%) received psychology services during rehabilitation, with the average number of psychology sessions being 4.8 (range = 0–28).
Health-Related Quality of Life, Pain, and Depression
To study HRQOL, FACT-Br was completed at admission, discharge, and at 1- and 3-mo postdischarge. No changes over time were found in the FACT-Br total score or any of the subscales (physical, social/family, emotional or functional well-being, or additional concerns) using generalized linear mixed models with repeated measures. In addition, no significant changes were found in PROMIS pain and depression scores. Sensitivity analyses in patients with complete data at all four times points confirmed these findings.
Generalized linear mixed models with repeated measures looking at the association between PROMIS scales and HRQOL scales at all time points jointly showed a significant negative association between PROMIS depression and FACT-G, FACT-Br, and each of their subscales (all P < 0.0001) except social and family well-being. The PROMIS pain was significantly associated with FACT-G (P < 0.0001), FACT-Br (P = 0.0003), physical (P < .0001) and functional (P = 0.003) well-being, and “additional concerns” (P < 0.0001), but not with emotional or social/family well-being (Table 3). The direction and strength of these associations were similar in the complete case sensitivity analysis, although some of the associations were no longer statistically significant (PROMIS depression and physical well-being, and PROMIS pain vs. FACT-G, functional and physical well-being) due to the reduced samples size (Fig. 2).
TABLE 3
FIGURE 2
Linear regression models at each evaluation time point demonstrated similar associations. At each time point, PROMIS depression was significantly negatively associated with FACT-G and FACT-Br total (all P< 0.0001), additional concerns (P < 0.001), emotional well-being (all P < 0.01), and with functional and physical well-being (all P < 0.05). PROMIS depression was not associated with social and family well-being except at discharge (P = 0.008) (Fig. 3). PROMIS pain scores were only mildly associated with Fact-G, Fact-Br, and additional concerns subscale at baseline and discharge (P < 0.01) and mostly uncorrelated with other subscales and at other time points. Supplemental Digital Content 1–4 (http://links.lww.com/PHM/A428, http://links.lww.com/PHM/A429, http://links.lww.com/PHM/A430, http://links.lww.com/PHM/A431) are available to see the relationships between PROMIS pain and FACT-Br and PROMIS depression and FACT-Br at all of the time points.
FIGURE 3
DayROS Correlations
Repeated measures models looking at the association between QOL scales (outcome) and DayROS (at admission and discharge jointly) showed a positive association between DayROS and FACT-Br total (slope = 0.505, P = 0.0058) and a negative association with PROMIS pain (slope = −0.132, P = 0.028). DayROS subscales (ADL, communication, and mobility) were also associated with FACT-BR total (all P < 0.05), but only DayROS-ADL was associated with PROMIS pain (P = 0.03). DayROS and its subscales were not associated with PROMIS depression. In addition, there were no significant association between DayROS gains (from admission to discharge) and change in PROMIS depression (P = .4467), FACT-Br total (P = .065) or PROMIS pain (P = .05658) over the same time (Fig. 4).
FIGURE 4
DISCUSSION
The main purpose of our study was to determine the relationship between HRQOL and cancer-related symptoms of pain and depression as well as functional gains in patients with malignant brain tumors while participating in an interdisciplinary outpatient setting and for 3 mos afterward.
We found that patients with malignant brain tumors who receive interdisciplinary outpatient rehabilitation services did not show a change in HRQOL, pain, or depression at the conclusion of rehabilitation and at 1- and 3-mo follow-up. There were more correlations seen between HRQOL and depression than between HRQOL and pain in the study patients.
Unlike previous studies, our study found that rehabilitation did not significantly improve overall HRQOL in patients with brain tumors in the outpatient setting. Huang et al31 followed 10 patients up to 3 mos after an acute inpatient rehabilitation program and found that HRQOL continued to improve at 3 mos after discharge. Pace et al35 looked at the effect of a home rehabilitation program and found that HRQOL improved in at least one domain after 3 mos. The HRQOL scores remained stable throughout our study. This could reflect a selection bias of subjects, because only patients with significant disability requiring at least two of three therapy disciplines were enrolled into our rehabilitation program. It could also reflect a difference in study design, because these other two studies were not examining directly the effects of an outpatient rehabilitation program. Cheville’s intervention was targeted at a mixed cancer population, so the direct effects on HRQOL in patients with brain tumors is not directly comparable.32 Gehring et al33 applied a cognitive rehabilitation program and did not target physical function, but they did find lasting improvements in subjective reports of mental fatigue. Sherer et al’s study of outpatient rehabilitation program for patients with brain tumors did not include measures of HRQOL, also making comparison with our findings difficult.34 However, it is important to note that although overall HRQOL did not improve during our outpatient rehabilitation program, it also did not decline. Patients with malignant brain tumors often have a gradual decline in function, cognition, and other symptoms with disease progression,2 and maintenance of HRQOL in this patient population is an important outcome to consider.
Many studies have shown a decline in HRQOL compared with general population normative values.3,4,13 A reduction in HRQOL could be explained by the different cancer-related treatments that they are experiencing during the rehabilitation period and the adjustment to their deteriorating function. Khan and Amatya6 and Cheng et al7 found that brain tumor patients reported between 60% and 80% overall decline in global HRQOL. Our patient population did not show a significant decline in HRQOL. This finding may be a result of the therapy these patients received in an interdisciplinary outpatient program. Our patients did not show an improvement in HRQOL, but the lack of decline in HRQOL is impressive in itself.
Other studies using the FACT-Br as a measure of HRQOL in patients with primary and metastatic brain tumors found a modest association with standard, widely used tools such as the 36 Item Short Form Survey, but some studies reported issues with reliably gathering HRQOL information in this population regardless of the tool used.8,41 Given the nature of our study design, each subjects’ baseline HRQOL at or before cancer diagnosis is unknown. Future studies may benefit from implementation of standard HRQOL measures along with psychosocial distress monitoring tools, which are currently mandated for all accredited cancer programs. Surveillance and ongoing monitoring of HRQOL, such as with the Commission on Cancer 2012 Program Standards, could also be linked to efforts to measure the effect of cancer on physical function.
A unique finding of our study was that HRQOL and depression were consistently related. Total FACT-Br, FACT-G, additional concerns, physical, emotional, and functional well-being were all negatively associated with PROMIS depression at all time points. Those who were less depressed had better reports of HRQOL. In contrast, Yavas et al16found that with disease progression, patients with high-grade gliomas reported a decline in HRQOL but no change in levels of anxiety or depression. Our study did not find a decline in HRQOL or depression for the duration of the study; however, we did not limit our study to subjects with disease progression. We did find a consistent relationship between better HRQOL and less depression at several time points, suggesting the importance of treating underlying mood disorders during any rehabilitation program to maximize the effects of rehabilitation on HRQOL. An alternate explanation is that perhaps treating functional limitations within this type of rehabilitation program is important in improving mood.
Unlike previous studies, our study found an inconsistent relationship between HRQOL and reports of pain. Khan and Amatya6 found that more than 50% of patients with brain tumors reported pain, which was associated with function and overall HRQOL. Cheng et al7 found that more than 70% of patients with gliomas reported pain at diagnosis, which was also associated with a decline in global HRQOL. We found that FACT-Br total, FACT-G, and additional concerns were associated with less pain at baseline and discharge, possibly alluding to the relationship between physical health and causes of pain. Our findings suggest that in patients with malignant brain tumors, an underlying mood disorder such as depression may be more related to HRQOL than pain.
Similar to the study by Khan and Amatya,6 our study also showed a positive association between pain and function. Those who were in less pain did better with the total DayROS and DayROS-ADL. There were not any correlations seen with DayROS and PROMIS depression. In this study, pain correlated more with function, whereas depression correlated more with HRQOL. One of the reasons for this finding may due to the nature of the FACT-Br having more questions that correlate more with depression than with pain. In addition, patients who are in more pain typically do not function as well.
We did not find any significant improvements in pain or depression during the course of the study. One would expect that pain or depression would improve with therapy. One reason why we did not find improvement in pain could be that many of these patients did not have a significant amount of pain at the start of the study. The PROMIS pain scale ranges from 7 to 42, with the average admission score of our patients being 15.3. There was improvement in pain at two of the three follow-up time points, but this difference was not significant. In general, the cognitive impairments seen in the brain tumor population are more severe than the degree of pain these patients have. Similarly, the depression scores were also low at the start, which could explain why there was not a significant improvement. The range of the PROMIS depression scale is 8 to 40, with our patients having an admission score of 15.2. The patients had worse depression scores at two of the follow-up time points, with one of these points being discharge. The findings are not significant, however. In addition, the cognitive decline that usually occurs with malignant brain tumors may have a role with the patients not realizing their level of depression or pain. In addition, one would expect over time that pain and depression worsen in this patient population. The fact that we did not see a significant decline is meaningful.
Our study has several limitations, including a small sample size and high dropout rate, which are inherent weaknesses of longitudinal studies in this type of relatively rare but fatal disease. Baseline data on HRQOL domains was not obtained, which would be helpful to track HRQOL throughout the cancer treatment spectrum. Attrition rate in our study increased over time, and only 12 patients completed all four evaluations. To address this, we used generalized linear mixed models with repeated measures, which provide a mechanism for handling missing values and use all available data.40 These models assume that data are missing at random, that is, that the probability of a missing HRQOL evaluation does not depend on HRQOL itself, conditional on other factors, for example, the underlying disease status at a given time. Although our statistical analysis approach does use all available data, later time estimates were still based only on the subset of patients with nonmissing data, and it is highly unlikely that these data were missing at random.
Another limitation was that we did not include a control or wait list group that did not receive rehabilitation services. There was already a small sample size, so having a control group would make the study group even smaller. Given the overall poor prognosis associated with malignant brain tumor patients, future studies would have to consider whether a wait list group would be an appropriate control group. In addition, there would need to be collaboration with other programs to have a control group, because our study was over several years and had a small sample size. In addition, the wide range of therapy sessions received by subjects is a limitation. Future studies with larger numbers of subjects would benefit from controlling for number of therapy sessions received to better understand the relationship of therapy to HRQOL, depression, and pain outcomes. Also, our interdisciplinary day rehabilitation program is a unique outpatient program, housing many rehabilitation specialists under one roof with ongoing, daily communication and regular team meetings, which may limit the generalizability of our findings. Potential future multisite studies could compare HRQOL of participants in our day rehabilitation program to those at institutions without a comprehensive rehabilitation program. This pilot study could serve as a model on which to base more comprehensive outpatient rehabilitation programs for brain tumor patients.
Further trials should be designed to address the question of how to best offer outpatient rehabilitation programs for patients with brain tumors, considering the costs and implementing the findings of this pilot trial by addressing the treatment of depression and pain. Future studies should consider a matched case control study design comparing the outcomes of malignant brain tumor patients to other neurologic diagnoses such as traumatic brain injury or stroke or compare malignant with benign brain tumors. In addition, future studies could follow subjects for an extended period to see whether a comprehensive rehabilitation intervention would affect not only HRQOL and cancer-related symptoms but also prognosis down the line. An interdisciplinary palliative care intervention was associated with improved HRQOL and mood, as well as prolonged overall survival in patients with metastatic lung cancer.7 Perhaps a similar approach incorporating rehabilitation would also benefit patients with malignant brain tumors.
CONCLUSIONS
In recent years, there has been an increased focus on HRQOL and disability in persons with brain tumors, as well as the role of rehabilitation in the management of functional impairments and disabilities.17,18 Patients with malignant brain tumors who receive interdisciplinary outpatient rehabilitation services showed overall stable HRQOL, pain, and depression at the conclusion of rehabilitation and 3 mos later. Although pain did not correlate consistently with HRQOL, there were significant correlations between HRQOL and depression in patients with malignant brain tumors. In addition, this study showed that patients who were in less pain had higher levels of function. Health-related quality of life, function, and cancer-related symptoms of pain and depression are important, and perhaps often neglected, outcomes to consider in developing rehabilitation and supportive care programs in this population. Further trials should be designed to address the question of how to best offer outpatient rehabilitation programs for patients with malignant brain tumors, considering costs and management of depression and pain.
ARYAN'S BLOG 