Volume 39, Issue 4 , Pages 691-701, April 2010
Outcomes of a Cancer-Related Fatigue Clinic in a Comprehensive Cancer Center
Article Outline
Abstract
Cancer-related fatigue (CRF) is a significant issue for cancer patients and frequently precipitates increased stress and anxiety for patients and caregivers alike. CRF may present well after the initial phase of cancer diagnosis and treatment, regardless of whether the cancer is in remission, widely metastatic, or somewhere in between. Determining whether the etiology of fatigue is potentially reversible and whether it is an effect of treatment or another unrelated cause is often perplexing. Because of the significant impact of CRF on patients at our institution, we organized a CRF clinic and began evaluating patients for fatigue in 1998. Our goal has been to initiate a more focused and, at the same time, more comprehensive effort in educating, evaluating, and treating CRF. The purpose of this report was to present a retrospective review of patients treated in our CRF clinic between 1998 and 2005, to examine the outcomes of our patients, and to briefly describe some of the challenges encountered in treating these patients. This information may help reassess and improve approaches in addressing CRF and subsequently improve fatigue in these patients.
Key Words: Cancer-related fatigue, fatigue clinic, symptom management, intervention
Introduction
Because of advances in treating many types of cancers and the resultant increases in the rates and lengths of survival, cancer-related fatigue (CRF) is now a frequent complaint among those who have had cancer.1, 2 However, this complaint is seldom dealt with to the satisfaction of the patient, in part, because of the difficulty in treating it. Moreover, patients themselves may be reluctant to voice concerns about this symptom because they may believe it to be a normal part of the disease and its treatment. Patients also may fear that fatigue is evidence of cancer recurrence or progression and, thus, may dread raising CRF as an issue. Sadly, the current reality is that health care providers, patients, and caregivers may feel inadequate in dealing with this symptom.
The purpose of this report was to briefly describe the methods of evaluation, the interventions provided, outcomes, and other key information, by retrospectively analyzing the data for patients evaluated in our CRF clinic between 1998 and 2005 and to offer insights into the challenges frequently encountered in patients experiencing CRF. This information may help reassess and improve the approaches used in the CRF clinic and elsewhere and support advances in better managing CRF.
Methods
Cancer-Related Fatigue Clinic Evaluation
Physicians directly refer most of the patients who come to our CRF clinic. Most of these patients are already receiving treatment at our institution; a few have been referred from external institutions; and some have had their primary cancer treatments at other centers. The patients themselves often initiate these referrals, motivated by their desire to address the CRF they are experiencing. Because we believe it important for health care providers to work as a team to treat patients, we contact the self-referred patients' primary oncologists to solicit their referrals. The treating physician usually meets this request with great enthusiasm. Our CRF clinic evaluation closely aligns with the National Comprehensive Cancer Network (NCCN) Cancer-Related Fatigue Guidelines.3 These guidelines outline the best level of care for the evaluation and management of CRF.
Before a patient's first CRF clinic appointment, our advanced practice nurse (APN) contacts the patient for scheduling, reviews and arranges for the necessary preliminary diagnostic assessments, and either mails or arranges for the patient to pick up a CRF Assessment Packet. (Table 1 details the specific contents of the CRF Assessment Packet.) The initial diagnostic workup includes values for thyroid-stimulating hormone (TSH) and electrolytes, a chemistry panel, and a complete blood count (CBC). The TSH test must be performed within the month before the appointment and other tests within the previous two weeks. For patients receiving active cancer treatment, a CBC must be obtained within one week before the evaluation.
Table 1. Cancer-Related Fatigue Clinic Assessment Packet
| Instrument | For Assessment of | No. of Items | Response Scale | Score Interpretation |
|---|---|---|---|---|
| BFI20 | Fatigue | 9 | 10 point | Mild: <4 |
| Moderate: 4–6 | ||||
| Severe: ≥7 | ||||
| Brief Pain Inventory21 | Pain | 15 | 10 point: 0 (no pain) to 10 (worst pain) | Mild: <4 |
| Moderate: 4–6 | ||||
| Severe: ≥7 | ||||
| Beck Depression Inventory-II22 | Depression | 21 | 4 point | Minimal: 0–13 |
| Mild: 14–19 | ||||
| Moderate: 20–28 | ||||
| Severe: ≥29 | ||||
| Patient-Generated Subjective Global Assessment of Nutritiona | Nutrition | 4 | Fill in, checklist, and others | |
| Brief Sleep Disturbance Scaleb | Sleep | 5 | Yes/no; 11 point | Mild: 22–29 |
| Moderate: 30–34 | ||||
| Severe: ≥35 | ||||
| M.D. Anderson Cancer-Related Symptom Inventory23 | Multiple cancer-related symptoms | 32 | 11 point (0–10) | |
| Functional Status Index24, 25 | Physical function, mobility | 18 | 5 point (assistance, difficulty); 4 point (pain) | Normal: 54 |
| Significant decrease in function: >97 | ||||
| Short Form-12 Health Survey: Standard Scoring26 | Patient self-report on health status | 12 | Varies | |
| Beck Anxiety Inventory27c | Anxiety | 21 | 4 point | Minimal: 0–7 |
| Mild: 8–15 | ||||
| Moderate: 16–25 | ||||
| Severe: 26–63 |
aDeveloped for the cancer-related fatigue clinic as an easy checklist for patients. |
bDeveloped by the Symptom Research Group at The University of Texas M.D. Anderson Cancer Center. |
cRecently added to assessment packet (i.e., in February 2003). |
At a patient's initial clinic appointment, an APN scores all individual assessment instruments from the CRF Assessment Packet and reviews all test results. Next, the physician reviews the assessment scores from all measurement instruments. The physician then obtains a complete patient history and performs a thorough examination, followed by a review of the patient's laboratory data. Only after such evaluations do the physician, patient, and caregiver (if present) discuss the findings. An individualized treatment plan, which may often include an exercise regimen and treatment of any reversible symptoms identified through the evaluation process (e.g., anemia, depression, uncontrolled pain, hypothyroidism, or other comorbidities), is then developed for the patient. Further consultations with other specialists may be recommended. Critical components of the patient's initial clinic appointment are the education and counseling provided by both the APN and the physician, which include take-home literature pertaining to issues surrounding CRF (such as sleep hygiene and energy conservation, which is extremely important in treating CRF) and other relevant information. For most of the cases, approximately 1.5 hours are required to complete the evaluations by the APN and physician, to formulate a treatment plan, and to conduct the necessary educational and counseling activities. (More details on CRF clinic operations and logistics have been described previously.3, 4) Most patients schedule a follow-up appointment six to eight weeks after their initial assessment. Because of the nature of CRF, most of the treatment plans require at least this much time to determine whether there has been any improvement after the implementation of the treatment plan. For example, an exercise program often requires several weeks of routine implementation before it can affect CRF.
Patients evaluated in our clinic often present with complex problems and have high expectations. It is crucial to address expectations at the beginning of the initial appointment and set achievable goals and a general timeline for measuring improvement. Managing expectations is vital, as physicians and staff must not mislead patients into believing that evaluation and treatment in a CRF clinic will completely cure them of their fatigue.
Outcome Measures
We identified all patients who attended the CRF clinic between July 20, 1998, and April 18, 2005; their data were collected and then validated in our study's database by means of cross-checks with original patient data (e.g., CRF Assessment Packet documentation, institutional electronic medical records). Results from patient responses to the Brief Fatigue Inventory (BFI) at their initial (i.e., baseline) CRF clinic visits and all subsequent visits served as our primary outcome measure. BFI scores range from 0 to 10, with higher scores indicating greater fatigue; BFI fatigue scores are typically dichotomized into severe (BFI score: 7–10) and nonsevere (BFI score: <7). Nonsevere fatigue can be further dichotomized into moderate (BFI score: 4–6.9) and mild (BFI score: 0–3.9) subcategories. We defined treatment success as a reduction in patient baseline fatigue to a lower fatigue category or subcategory (i.e., baseline severe fatigue reduced to nonsevere fatigue on follow-up or baseline moderate fatigue reduced to mild fatigue on follow-up). Using reported BFI scores, patient baseline and follow-up categories and/or subcategories were determined, and category and subcategory baseline vs. follow-up differences were calculated. For patients achieving treatment success (i.e., reduction to a lower fatigue category or subcategory), the follow-up visit number (e.g., first or second follow-up visit) at which the patient reported a fatigue reduction was recorded. The number of the follow-up visit at which fatigue reduction was observed then served as our secondary outcome measure.
Analyses
Descriptive statistics (e.g., count, percent, median, minimum, maximum) were computed for patient demographic, clinical, and self-report assessment measures. We compared patients by follow-up status (attended at least one follow-up visit vs. did not attend a follow-up), fatigue severity status (severe fatigue vs. nonsevere), and fatigue treatment success status (baseline fatigue category/subcategory reduction vs. no category/subcategory reduction); associations between patient status groups and categorical patient characteristics were investigated by means of Chi-square or Fisher's exact testing, whereas differences in continuous characteristics were examined using analysis of variance or a nonparametric equivalent (e.g., Mann-Whitney rank-sum test) for non-normally distributed dependent variables. The level of significance for all statistical tests conducted was set at P
<
0.05.
Results
We collected retrospective data for 262 patients who had been referred to and evaluated by our CRF clinic between July 20, 1998, and April 18, 2005. Two of these patients did not have complete baseline fatigue data and were excluded from our analyses. Our sample was the remaining 260 CRF clinic patients for whom complete baseline fatigue data were available. This cohort of patients had been referred to our specialized clinic either directly by their physicians or through patient- or family-initiated referrals. Thus, these patients are not necessarily representative of all the cancer patients at our institution who experience underlying fatigue. Before collecting our data, we obtained institutional review board approval.
Of the 260 CRF patients in our sample, most were female (66%, n
=
172). This is not surprising given that our institution's Breast Cancer Center frequently refers patients to the CRF clinic. The fact that the most common cancer diagnosis of our CRF patients was breast cancer (34%, n
=
87) supports this association with referral patterns. The predominance of nonminority patients (82%, n
=
212) reflects the general cancer population of the institution. CRF patients averaged 56 years of age (range: 24–86 years); 65% (n
=
168) were married. Overall, nearly two-thirds of patients (63%, n
=
146) no longer had any evidence of malignancy. In contrast, among the patients with solid tumors visiting our clinic, 18% (n
=
46) had metastatic disease. Slightly more than two-thirds of patients (72%, n
=
173) had good performance status (i.e., Zubrod performance status scores <2) at the time of their initial fatigue evaluation.
We observed several statistically significant (P
<
0.05) associations between patient follow-up status (attended at least one follow-up visit vs. did not attend) and fatigue severity status (severe vs. nonsevere fatigue). Follow-up status also was found to be associated with no evidence of disease (NED) and with the distribution of patient cases across the CRF clinic's top three cancer diagnoses. Seventy-six percent (n
=
91) of the follow-up patients had NED compared with 50% (n
=
55) of the patients with no follow-up. In terms of the type of cancer, more breast cancer patients (65% vs. 52%) than gynecologic (ovarian/cervical/endometrial) cancer patients (5% vs. 23%) returned for follow-up than those with no follow-up. Patient fatigue severity status was found to be associated with solid tumor patients' treatment response and with Zubrod performance score. Forty-two percent of patients with solid tumor cancer were uncontrolled, in terms of their treatment response, compared with only 16% of nonsevere fatigue patients with solid tumor cancer. Forty percent of patients with severe fatigue had poor Zubrod performance scores, but only 19% of patients with nonsevere fatigue had poor performance scores. No associations were found between patient demographic and clinical characteristics and patient fatigue treatment success status.
Nearly 30% (n
=
75) of patients were receiving cancer treatment (i.e., chemotherapy, hormonal, or radiation therapy) at the time of their initial evaluation for fatigue. The most common form of ongoing treatment was chemotherapy (16%, n
=
42), followed by hormonal therapy (11%, n
=
29). Tamoxifen was the most common form of hormonal treatment being received (72%, n
=
21). Only six patients (2%) were undergoing radiation therapy at the time of presentation to our clinic.
The large proportion of breast cancer patients who attended our CRF clinic and the underlying patient referral patterns mentioned previously likely explain why hormonal therapy was the second most common treatment reported by our patients and why a combination of fluorouracil, doxorubicin, and cyclophosphamide was the most common chemotherapy regimen noted.
Predictably, almost all patients (98%, n
=
254) had previously had some form of cancer treatment, most commonly chemotherapy, which had been received by 70% (n
=
183) of patients. A similarly high percentage of patients (68%, n
=
177) had undergone surgical treatment, whereas slightly more than half of the patients (55%, n
=
142) had received radiation therapy. Seventy-eight percent (n
=
203) of patients had received more than one treatment modality, with 47% (n
=
122) receiving two modalities and 26% (n
=
69) receiving three. We found no associations between the current and/or prior patient treatment methods, and patient follow-up status, fatigue severity status, or fatigue treatment success.
Most of the patients (74%, n
=
193) had underlying comorbidities in addition to their cancer diagnosis. The most frequently observed comorbidity was hypertension (36%, n
=
93), followed by depression (28%, n
=
72), hypothyroidism (22%, n
=
57), and diabetes mellitus (13%, n
=
34). Of the 193 patients with comorbidities, most had either one or two comorbidities; only 6% of patients (n
=
15) had three or more comorbidities. In most cases, patient comorbidities were fairly well controlled, and the CRF clinic physician evaluator did not believe them to be a direct cause of patient fatigue.
The most common symptoms reported during initial patient CRF evaluations were weakness (50%, n
=
140) and depression (39%, n
=
101). The medications that patients were most frequently taking on presentation were antidepressants (42%, n
=
109), opioids (35%, n
=
92), and antihypertensives (34%, n
=
87). Twenty-five patients (10%) were on a stimulant at referral. The medians of patients' initial laboratory values were within normal ranges according to our diagnostic tests. Most of the patients were not anemic (median hemoglobin: 12.7
g/dL; range: 7.1–17.2
g/dL) and had normal albumin (median: 3.9
g/dL; range: 1.0–5.7
g/dL) and TSH levels (median: 1.8
mcu/mL; range: 0–54.2
mcu/mL).
Patient scores from the CRF Assessment Packet are presented in Table 2. Using the nonparametric Mann-Whitney rank-sum test, we observed no statistically significant differences (P
<
0.05) between patient follow-up status or fatigue treatment success status and any of the packet's measures. However, when basing patient groups on their fatigue severity status (severe fatigue vs. nonsevere), significant differences were observed across all CRF packet measures, excluding nutrition. Patients with severe fatigue reported significantly worse symptom status than patients with nonsevere fatigue, that is, pain (8.0 vs. 5.0), sleep disturbance (24.0 vs. 20.0), depression (22.0 vs. 14.0), functional status (84.0 vs. 70.0), and anxiety (18.5 vs. 8.5).
Table 2. Cancer-Related Fatigue Assessment Tool Scores at Initial Evaluation
| Assessment Tool | Scores | Fatigue Status: Severe vs. Nonsevere | P-valuea | |
|---|---|---|---|---|
| Median | Range | |||
| Brief Fatigue Inventory | 6.8 | 0–10 | ||
| Brief Pain Inventory | 6.0 | 0–10 | 8.0 vs. 5.0 | 0.000b |
| Brief Sleep Disturbance Scale | 21.0 | 1–52 | 24.0 vs. 20.0 | 0.001 |
| Beck Depression Inventory-II | 17.0 | 0–52 | 22.0 vs. 14.0 | 0.000b |
| Functional Status Index | 76.0 | 8–181 | 84.0 vs. 70.0 | 0.000b |
| Beck Anxiety Inventory | 12.0 | 0–44 | 18.5 vs. 8.5 | 0.000b |
| Patient-Generated Subjective Global Assessment of Nutrition | 3.0 | 1–3 | 3.0 vs. 3.0 | 0.414 |
aMann-Whitney rank-sum test. |
bP |
Prescribed treatment interventions for CRF are listed in Table 3. The treatment interventions most frequently discussed and recommended at initial patient visits were energy conservation (98%, n
=
256), sleep hygiene (97%, n
=
253), and exercise plans (95%, n
=
247) (Table 3). Using two-sided Fisher's exact testing, with a significance level of P
<
0.05, prescribing practices did not differ among patients by either follow-up visit status or treatment success status. However, for two of the nine possible interventions, prescribing practices were found to be associated with patient fatigue severity status. Patients with severe fatigue were prescribed two interventions more frequently than were patients with nonsevere fatigue: analgesics (31% vs. 19%) and antidepressants (33% vs. 21%). In terms of the total number of prescribed interventions, prescribing practices were also found to differ by patient fatigue severity status (Table 4).
Table 3. Prescribed Patient Interventions for CRF
| Intervention | n (%) |
|---|---|
| Energy conservation | 256 (98.5) |
| Sleep hygiene | 253 (97.3) |
| Exercise | 247 (95.0) |
| Relaxation | 71 (27.3) |
| Antidepressant | 70 (26.9) |
| Analgesic | 64 (24.6) |
| Stimulant | 56 (21.5) |
| Anxiolytic | 43 (16.5) |
| Nutritional counseling | 26 (10.0) |
| Number of prescribed interventions | |
| 169 (65.0) | |
| 91 (35.0) | |
| Total | 260 |
Table 4. Interventions for Cancer-Related Fatigue by Fatigue Severity Status
| Intervention | Prescribed | Nonsevere Fatigue, n (%) | Severe Fatigue, n (%) | P-valuea |
|---|---|---|---|---|
| Analgesica | No | 111 (81) | 85 (69) | 0.031 |
| Yes | 26 (19) | 38 (31) | ||
| Antidepressanta | No | 108 (79) | 82 (67) | 0.035 |
| Yes | 29 (21) | 41 (33) | ||
| No. of prescribed interventionsa | 0–4 | 98 (72) | 71 (58) | 0.027 |
| 5–9 | 39 (28) | 52 (42) | ||
aFisher's exact test (two-sided). |
One hundred and fifty-eight referrals for consultation (mean of 1.4 referrals per patient) were submitted on 117 (45%) patients at the initial CRF evaluation. The most common referral was to social services for self-hypnosis and relaxation techniques (n
=
41, 26%). Other commonly referred services included physical medicine and rehabilitation (includes rehabilitation, physical therapy, and occupational therapy; n
=
33, 21%); psychiatry (includes psychiatry, neuropsychology testing, and psychology; n
=
28, 18%); and pain management (n
=
9, 6%). Some patients were already following with some of these services, and, therefore, initiation of a referral was unnecessary. Patients were educated regarding energy conservation and sleep hygiene by the physician and APN at the initial evaluation. Also, during the period of this study, our institution did not have a Sleep Clinic. Presently, a Sleep Clinic is available within our center and referral of our patients for further evaluation of sleep dysfunction is frequent.
Patient follow-up intervals were usually six to eight weeks after initial evaluations, although this varied to some extent. Some of this follow-up variation was dependent on the distance a patient lived from the institution. For the most part, follow-up visits for patients living at some distance from M.D. Anderson were coordinated with other return visits.
Treatment Success
Of the 260 patients, 122 (47%) reported severe fatigue and 138 (53%) reported some form of nonsevere fatigue—110 (42%) reported moderate fatigue and 27 (10%) mild fatigue (Fig. 1, Table 5). Fifty-three percent (n
=
138) of patients attended at least one follow-up visit. Among these patients, 69 (50%) had had severe fatigue on initial presentation. For patients with severe fatigue, 46 of 69 (67%) were able to reduce their fatigue from a severe to a nonsevere level. For those patients with nonsevere moderate fatigue at initial evaluation, 28 of 57 (49%) were able to reduce their fatigue from a moderate to a mild level. Thus, overall, 59% of patients (74 of 126) successfully achieved their targeted fatigue reduction level. There was no statistically significant association (P
<
0.05) observed between distribution of fatigue-level categories and patient follow-up status (i.e., patients with no follow-up visits vs. those with at least one follow-up) for either the severe/nonsevere distribution (two-sided Pearson Chi-square
=
0.86, P
=
0.36) or the severe/moderate/mild distribution (two-sided Pearson Chi-square
=
1.33, P
=
0.52).
Table 5. Severity of Fatigue and Patient Follow-Up Status (n
=
260)
| Fatigue-Level Distribution | Baseline Visit Only, n (%) | Baseline | Total |
|---|---|---|---|
| Severe | 54 (44.3) | 69 (50.0) | 123 (47.3) |
| All nonsevere | 68 (55.7) | 69 (50.0) | 137 (52.7) |
| Nonsevere moderate | 53 (43.4) | 57 (41.3) | 110 (42.3) |
| Nonsevere mild | 15 (12.3) | 12 (8.7) | 27 (10.4) |
| Total | 122 (46.9) | 138 (53.1) | 260 (100.0) |
Treatment Length
As mentioned previously, a follow-up visit is typically scheduled six to eight weeks after a patient's initial consultation. Of the 74 successfully treated patients, 62% (n
=
46) reported treatment success by their first follow-up visit; an additional 22% (n
=
16) reported success by their second follow-up visit. Thus, 84% (62 of 74) of patients were able to achieve their targeted fatigue reduction within two follow-up visits.
Treatment Maintenance
Of the 74 patients who successfully achieved targeted fatigue reduction levels, 60% (n
=
44) returned to the CRF clinic for additional, posttreatment success follow-up. We defined treatment maintenance as a patient reporting average BFI scores within his or her originally targeted fatigue reduction category at all subsequent follow-ups. Of the 44 patients with post-treatment-success follow-ups, 77% (n
=
34) maintained their targeted fatigue reductions.
Twenty-nine (66%) of the 44 patients attending follow-ups after successful treatment had had severe fatigue at baseline; 26 of these 29 patients (90%) successfully maintained a nonsevere fatigue status across their subsequent clinic visits. Fifteen (34%) of the 44 patients attending follow-ups after successful treatment had had nonsevere moderate fatigue at baseline; 8 of these 15 patients (53%) successfully maintained a nonsevere mild fatigue status at all their subsequent visits.
Discussion
Fatigue is a frequent and challenging symptom for cancer patients and requires a multidisciplinary approach for successful treatment. Because fatigue may be attributed to discrete disease entities and also correlated with other symptoms our patients often experience, it requires a comprehensive evaluation, including both a detailed history of cancer treatment and attention to comorbidities, medication use, and key aspects related to the presentation of fatigue (pattern, intensity, type, factors improving or worsening feelings of fatigue, and other symptoms experienced).
Based on our findings, we support the establishment of a CRF clinic at our institution. Most of the patients returning after their baseline visit had improved energy levels. In addition, these patients attained this improvement within a few visits to our clinic. Most maintained their improvement over time.
It is interesting to note that those patients with severe fatigue also had greater levels of pain, sleep disturbance, depression, anxiety, and poorer functional status than those with nonsevere levels. Whether the severe fatigue was a result or a cause of these other symptoms is unknown. Patients with higher total symptom burdens would be expected to frequently express higher levels of fatigue. The concept of symptom clusters has previously been described in the literature and supports our findings.5, 6, 7, 8, 9
On analysis of patients who did not return for follow-up visits vs. those who did, including whether they had severe or nonsevere fatigue, we noted no statistical differences. Therefore, we can only speculate on why approximately half of our patients did not return after their initial visits. There may be various reasons from logistical concerns (because many of our patients are not local) to disinterest and apathy. Occasionally, patients only come for the evaluation at the urging of a family member. It is often evident during a visit that the patient has no interest in pursuing treatment interventions, commonly including exercise, or in returning for subsequent visits. It is also possible that some patients follow treatment recommendations and feel their fatigue has improved to such an extent that they do not need additional follow-up. We plan to query those not returning after their initial evaluations so that we may answer this question and plan interventions to increase follow-up compliance, if needed.
We were unable to identify causative factors among our patients to explain why some had fatigue reductions and some did not. Although we compared demographic variables, treatment types, diagnostic testing results, and symptoms, none were predictive of treatment success.
We may be able to identify predictive factors in a subpopulation of our patients by enriching our follow-up data. Presently, we have a great deal of data at the initial visit but collect less standardized data at follow-ups. By serially collecting similar data on all patients at follow-up, we may be able to better recognize factors related to success.
Study Limitations
A limitation of this study is that this group of patients may be more motivated than others because they took advantage of this specialized clinic. Our data are clinic based; therefore, the sample is not homogeneous in the sense of controlling for factors that may affect fatigue. However, caring for patients is infrequently as uniform as the criteria set forth for studies.
Moreover, our cancer patients may not be representative of patients at other sites or from other areas of the country, because our institution is a comprehensive cancer center and draws patients internationally. Therefore, the organization, operations, and outcomes of our experience may not be generalizable to other practices. In addition, we do not have a large minority population seen at our institution, limiting the results of our experiences. Also, information regarding treatment adherence in this sample was limited, and plans to enhance this data collection are underway.
It should be noted that our approach to conduct multiple comparisons across patient demographic and clinical characteristics and patient follow-up, fatigue severity, and treatment success status used unadjusted P-values. Thus, there may be an increased risk of Type I error in our hypothesis testing, that is, error in rejecting a null hypothesis when it should be retained. However, our research endeavors to identify potentially important associations with CRF and predictors of CRF are in an early, exploratory stage; therefore, a more liberal approach to develop theory and, subsequently, testable hypotheses may be more appropriate. It is worth noting that our most robust findings, involving associations between patients' fatigue severity statuses and their symptom statuses across a set of multiple symptoms, remain statistically significant when a Bonferroni adjusted P-value (here, P
<
0.008) is used.
Challenges Encountered
Our clinic commonly faces a challenge in convincing patients that exercise can improve their fatigue. It is not unusual for patients to express astonishment at the suggestion of exercise, given that they have just finished describing their lack of energy for routine daily activities (such as food preparation, housekeeping, and occasionally, even self-care activities like bathing). We must then explain the importance of integrating an exercise program into the patient's daily routine, often as a first-line treatment for fatigue. There are numerous studies published and two recent meta-analyses demonstrating the importance of exercise in CRF.10, 11
Depending on the presence of particular comorbidities, some patients may require further assessments, including cardiac and pulmonary evaluations, before commencing an exercise routine. For example, a patient with a history of hypertension, diabetes, and questionable angina, who has not had a recent cardiac workup and, at presentation, does not exercise, should undergo cardiac assessment before beginning an exercise regimen.12
Many patients require physical therapy before initiating an exercise regimen to build stamina and physical conditioning to acceptable levels. For example, patients with neutropenia or bony metastasis should be cautioned regarding the type and duration of exercise that they can safely undertake.
Another challenge frequently encountered is helping patients understand that they will need help from family and friends while dealing with CRF and that they must be willing to ask for it. It is unrealistic for many patients to continue performing all the tasks they had performed before their cancer treatment. Family and friends are, in most cases, available to support the patient; it is usually the patients themselves who resist asking for help.
During the patient's initial evaluation, it is important to establish a sense of what a “usual day” in the patient's life is like to determine whether patients are overexerting themselves and setting unrealistic goals. Suggesting energy-conserving measures, such as using a stool with rollers when cooking and a shower chair during bathing, may be helpful.13 It may also help to suggest that patients delegate some household tasks to family or hire help. Despite patients' reluctance to seek help, families very often prove enthusiastic about participating in treatment plan recommendations. For patients working outside the home, a different sort of help may be required. These patients may need to decrease their work schedule to part time and then transition back to full time as their symptoms improve.
It is all too common for cancer survivors to have such unremitting fatigue that they are no longer able to sustain regular employment.14, 15, 16, 17, 18, 19 This can be devastating to both the patient and the patient's family, particularly, if the patient's income provided the sole means of family support. Unfortunately, it is also all too common for private disability insurers and state and federal agencies to deny benefits to patients applying for assistance based on debilitating fatigue, especially if the patient's malignancy has been cured. Patients regularly report being told by their insurers that, because they no longer have cancer, they should be able to work. In other words, insurers often believe that patients with no underlying disease are complainers and have no cause prohibiting their successful employment. This scenario frequently becomes a battle in which health care providers must advocate for patients. Physicians must educate insurers about the realities of persons living with CRF. Because of enormous successes in curing or stabilizing numerous forms of cancer, we will have to explain increasingly over the next decade that CRF is a debilitating yet potentially treatable condition for which patients should receive normal insurance and benefit coverage.
Unlike patients with, for example, severe coronary artery disease, who can have coronary angiograms or echocardiograms performed to demonstrate poor cardiac functioning, CRF patients have minimal objective evidence documenting their condition. It has only been with the development of validated instruments, such as the BFI20 and others, that health care providers have been able to quantify the level of fatigue a patient feels. Despite the lack of diagnostic tests to validate the existence of CRF, appropriately identified patients experiencing severe and persistent fatigue should receive disability benefits.
The challenges we have experienced have been described in the literature and are not unique to our patients. The NCCN CRF Fatigue Guidelines12 address these also.
Future Plans
As we continue to make headway in our approach to CRF, we plan to focus on identifying factors that affect or might predict responders vs. nonresponders and explain how patients who maintain their fatigue reductions differ from those who do not. Do patients with single visits differ from those paying more visits? What factors affect fatigue reduction in those responders? Is it treatment adherence or other factors? To advance the treatment of CRF, we also need well-designed clinical trials evaluating pharmacological agents. There is a paucity of studies presently; therefore, any further data would be most helpful.
Further research to identify physiological causes of CRF is necessary so that targeted interventions may be developed. This is the most exciting area to research that, if successful, could advance treatment options by leaps and bounds rather than simply treating symptoms.
In summary, CRF is a common and frequently debilitating symptom for patients that is routinely challenging to providers. It is best approached in a multidisciplinary fashion. Our CRF clinic has demonstrated improvements in functioning for many patients. This improvement can be accomplished within a few visits, and most of them maintain this improvement over time. The experiences of this clinic may be helpful to others caring for such patients. In addition, our outcomes are supported by published literature, and they validate both findings in our cancer population and those previously reported. As we continue to explore and refine CRF, it is important to continue to document our findings so that we continually expand the literature of this important symptom.
References
- Assessment and management of cancer-related fatigue in adults. Lancet. 2003;362:640–650
- . Fatigue and cancer: causes, prevalence and treatment approaches. Br J Cancer. 2004;91(5):822–828
- A fatigue clinic in a comprehensive cancer center: design and experiences. Cancer. 2001;92:1708–1713
- . A cancer-related fatigue clinic: opportunities and challenges. J Natl Compr Canc Netw. 2003;1:333–343
- Symptom clusters: concept analysis and clinical implications for cancer nursing. Cancer Nurs. 2005;28:27–82
- A cytokine-based neuroimmunologic mechanism of cancer-related symptoms. Neuroimmunomodulation. 2004;11:279–292
- . Occurrence of symptom clusters. J Natl Cancer Inst Monogr. 2004;32:76–78
- . Symptom clusters: the new frontier in symptom management research. J Natl Cancer Inst Monogr. 2004;32:17–21
- . Cancer-related fatigue, depressive symptoms, and functional status: a mediation model. Nurs Res. 2006;55:366–372
- . Cancer-related fatigue: a systematic and meta-analytic review of non-pharmacological therapies for cancer patients. Psychol Bull. 2008;134:700–741
- . Exercise for the management of cancer-related fatigue in adults. CD006145 Cochrane Database Syst Rev. 2008;(Issue 2):
- Cancer-related fatigue. Clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2007;5:1054–1078
- A pilot study examining energy conservation for cancer treatment-related fatigue. Cancer Nurs. 2002;25:333–341
- . Recovered from cancer but still ill: strategies used to legitimize extreme persistent fatigue in disease-free cancer patients. Eur J Cancer Care (Engl). 2009;18:28–36
- . Is fatigue a long-term side effect of breast cancer treatment?. Cancer Control. 1999;6:256–263
- . Ensuring quality care for cancer survivors: implementing the survivorship care plan. Semin Oncol Nurs. 2008;24:208–217
- . Psychosocial factors in lung cancer: quality of life, economic impact, and survivorship implications. J Psychosoc Oncol. 2008;26:69–80
- Survivorship issues following treatment completion—results from focus groups with Australian cancer survivors and health professionals. J Cancer Surviv. 2008;2:20–32
- . Survivorship, late effects and cancer of the cervix. Clin Oncol (R Coll Radiol). 2008;20:479–487
- The rapid assessment of fatigue in cancer patients. Cancer. 1999;35:1186–1196
- . Pain assessment: global use of the brief pain inventory. Ann Acad Med Singapore. 1994;23:129–138
- Self-reported depressive symptoms that differentiate recurrent-episode major depression from dysthymic disorders. J Clin Psychol. 1987;43:246–250
- Assessing symptom distress in cancer patients. The M.D. Anderson Symptom Inventory. Cancer. 2000;89:1634–1646
- . Functional performance assessment: a program approach. Arch Phys Med Rehabil. 1981;62:456–460
- . Patient profiles: utilization in functional performance assessment. Arch Phys Med Rehabil. 1983;64:268–271
- . The MOS short-form general health survey. Reliability and validity in a patient population. Med Care. 1988;26:724–735
- An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol. 1988;56:893–897
PII: S0885-3924(10)00083-7
doi:10.1016/j.jpainsymman.2009.09.010
© 2010 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Volume 39, Issue 4 , Pages 691-701, April 2010

