• phase 1;
  • experimental therapeutics;
  • palliative care;
  • supportive care;
  • symptom burden


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References


Phase 1 oncology trial participants often are excluded from hospice. However, it is not known whether they would benefit from hospice services. The objectives of the current study were to define the palliative care needs of these patients and to determine whether their needs are greater than those of other cancer patients.


Two hundred ninety-seven patients who were undergoing cancer therapy and 69 patients who were enrolled in phase 1 trials at 7 oncology clinics in an urban cancer network were recruited and consented to participate in interviews. Interviewers assessed the prevalence and severity of 10 symptoms using the Global Distress Index of the Memorial Symptom Assessment Scale and patients' perceived need for 4 services typically provided through hospice: a chaplain, counselor, home health aide, and visiting nurse.


Patients in the 2 groups had a similar symptom burden. However, after adjusting for Eastern Cooperative Oncology Group performance status scores, phase 1 patients were more likely to have 5 of the 10 symptoms and reported greater severity for 6 of the 10 symptoms. Compared with other patients, phase 1 patients were less likely to say they needed a home health aide (4 of 69 patients [6%] vs 198 of 297 patients [67%]), a chaplain (7 of 69 patients [10%] vs 134 of 297 patients [45%]), or a counselor (11 of 69 patients [16%] vs 160 of 297 patients [54%]; chi-square test: P < .001 for all). They were equally likely to say they needed a visiting nurse (30 of 69 patients [44%] vs 142 of 297 patients [48%]; chi-square test: P = .516).


Compared with other patients who had cancer, patients who were participating in phase 1 trials were less likely to want several home care services, although they experienced a greater symptom burden. Further research will be needed to define the palliative care needs of this population. Cancer 2009. © 2009 American Cancer Society.

Phase 1 oncology trial participants enroll in trials that, by definition, are not designed with curative intent or for the direct benefit of the enrolled patient. Instead, their primary goal is to establish the maximum tolerated drug doses and toxicity profiles of cancer treatments in human participants.1 Only a small portion of all cancer patients enroll in phase 1 oncology studies. However, >1500 phase 1 or phase 1/2 clinical trials currently are open or are preparing to recruit adult patients with cancer.2, 3

Hospices have become significant providers of end-of-life care in the US over the past 25 years. The 1982 Medicare Hospice Benefit was established to provide palliative care services to patients with a prognosis of ≤6 months who are willing to forgo life-prolonging or disease-modifying treatment.4 Currently, >500,000 patients with cancer enroll in hospice every year,5 and growing evidence indicates that hospice is able to provide high-quality care with excellent patient and family satisfaction.6–10 Unfortunately, because phase 1 agents are considered disease-directed therapy, trial participants are unable to enroll in hospice.

The benefits of early or simultaneous palliative or hospice care are recognized increasingly for all patients with advanced cancer, and advocates of palliative care have argued that hospice should be available to patients who have enrolled in phase 1 trials.11–13 Phase 1 participants often have advanced cancers for which curative treatments are not available, and phase 1 agents are unlikely to prolong life or modify the course of disease. Furthermore, the life expectancy of phase 1 patients averages between 5 months and 6.5 months.14, 15 Hospice providers and phase 1 trialists agree that patients on phase 1 protocols should be able to enroll in hospice concurrently.16 However, even if dual enrollment is possible technically under Center for Medicare and Medicaid Services (CMS) regulations, the perception of complex payment structures and excessive trial-associated costs limit hospice enrollment of this unique group of patients.12, 16 Current regulations do not routinely allow for flexible, comprehensive palliative care and hospice planning for patients who are unwilling to give up cancer therapy.

However, it is not known whether patients who enroll in phase 1 trials have palliative care needs that hospice could meet or whether they want the services that hospice could provide. If phase 1 patients have substantial needs for care, then it may be necessary to reconsider hospice eligibility criteria. Conversely, if they do not have substantial needs for palliative care, then concerns regarding exclusion from hospice may be misdirected. Thus, the objective of this study was to define the palliative care needs of patients who enroll in phase 1 oncology trials compared with those of cancer patients who receive routine chemotherapy with similar life expectancies.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Patients with cancer were recruited over a period of 2 years from 7 clinics within the University of Pennsylvania Comprehensive Cancer Center Network, including 1 clinic that is devoted to early-phase trials. All patients aged >18 years with clinical evidence of cancer were eligible for participation if they were able to speak and read English and could give informed consent. Phase 1 patients were eligible at the time of phase 1 trial enrollment regardless of their prognosis, while other patients were eligible if their oncologists estimated that the patients would have a prognosis of ≤6 months if their current therapy was discontinued. These broad inclusion criteria were designed to capture all cancer patients who most likely would have been hospice-eligible if they were not receiving active cancer treatment. Nursing staff at each clinic assisted a rotating team of 3 interviewers to identify and approach eligible patients. This study was approved by the institutional review boards of participating sites.

After they provided their informed consent, patients provided basic demographic information, including self-described ethnicity, sex, marital status, and education. Patients completed a 3-part interviewer-assisted questionnaire. First, they completed the Global Distress Index of the Memorial Symptom Assessment Scale (MSAS-GDI).17 The GDI provides an assessment of the presence and impact of 10 symptoms that are common in cancer. The score is the average of the single dimension scores of 4 psychologic symptoms (feeling sad, worrying, feeling irritable, and feeling nervous) and 6 physical symptoms (lack of appetite, lack of energy, pain, feeling drowsy, constipation, and dry mouth). The 4 psychologic symptoms are scored from 0 to 4, indicating whether or not the symptom was experienced and its frequency from 1 (‘rarely’) to 4 (‘frequently’). The 6 physical symptoms also are scored from 0 to 4, so that 0 indicates the absence of the symptom, 0.8 indicates ‘not at all’, 1.6 indicates ‘a little bit’, 2.4 indicates ‘somewhat’, 3.2 indicates ‘quite a bit’, and 4 indicates ‘very much’ in the severity of the distress caused by the symptom.17

Next, patients were asked to identify home care services that would be helpful to them (nurse, home health aide, emotional/psychologic counseling, and spiritual support). Participants responded ‘yes’ or ‘no’ to questions regarding their desire for each described service. These services were selected because previous research demonstrated that they are among the most valued by patients and families and because they are easily understood.18–20 Interviewers described each service but did not identify them as hospice services.

Finally, functional status was assessed using standardized questions to assign an Eastern Cooperative Oncology Group (ECOG) performance status score. An ECOG performance scale is scored from 0 to 4, as follows: 0, fully functional; 1, able to perform normal activities of daily living and some work, but strenuous activity restricted; 2, unable to work, up and active >50% of the time; 3, limited self-care and in bed or chair >50% of the time; 4, completely disabled and confined to bed or chair.21

Nonparametric tests (eg, the Wilcoxon rank-sum test) and the chi-square test were used for bivariate analyses. Logistic or ordinal logistic regression, as appropriate, was used for multivariable models. The software package Stata for Windows version 10 (Stata Corporation, College Station, Tex) was used for all statistical analyses. Sample size calculations were based on a fixed sample of general cancer patients (n = 297). We calculated that a sample of ≥65 phase 1 patients would provide adequate power (1-β = 0.80) to detect a 20% difference (0.10 vs 0.30) in the proportions of patients in each group who said they would want each of the 4 hospice services (α = .0125; corrected for multiple comparisons).


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Patients who agreed to participate (366 of 418 patients; 88%) were similar to those who refused to participate with respect to age, sex, ECOG score, ethnicity, and education. Patient characteristics are described in Table 1. None of the patients were receiving hospice or home care program services. Nonphase 1 trial patients were similar to phase 1 patients with regard to mean age (58 years vs 57 years; P = .330), sex (men, 147 of 297 patients [49%] vs 37 of 69 patients [54%]; P = .537), marital status, and education level (Table 1). However, nonwhite patients were under-represented in the phase 1 population (2 of 69 patients [3%] vs 81 of 297 patients [33%]; chi-square test; P < .001), which is typical of phase 1 trials.22

Table 1. Patient Characteristics
 No. of Patients (%) 
CharacteristicConventional Treatment (N = 297)Phase 1 (N = 69)P
  1. ECOG indicates Eastern Cooperative Oncology Group.

Mean age [range], y58 [20–89]57 [22–82].330
Sex, men147 (49)37 (54).537
 White199 (67)67 (97)<.001
 Other98 (33)2 (3) 
Marital status
 Single45 (15)7 (10) 
 Married/living  with partner186 (62)51 (74).288
 Divorced/separated45 (15)6 (9) 
 Widowed21 (7)5 (7) 
Educational level
 Did not graduate from  high school24 (8)5 (7) 
 High school84 (28)25 (36) 
 Some college or  technical school76 (26)16 (23).673
 College55 (19)9 (13) 
 Graduate school58 (20)14 (20) 
Cancer type
 Breast84 (28)2 (3) 
 Lung35 (12)2 (3) 
 Sarcoma5 (2)2 (3) 
 Hematologic37 (13)0 (0) 
 Gastrointestinal51 (17)8 (12) 
 Glioblastoma1 (>1)0 (0)<.001
 Prostate35 (12)1 (2) 
 Genitourinary18 (6)18 (27) 
 Melanoma2 (1)29 (44) 
 Oropharynx28 (9)1 (2) 
 Other0 (0)3 (5) 
ECOG score
 089 (30)49 (71) 
 132 (11)17 (25) 
 2124 (42)1 (1)<.001
 344 (15)2 (3) 
 48 (3)0 (0) 

Melanoma (29 of 69 patients; 44%), genitourinary cancers (18 of 69 patients; 27%), and gastrointestinal cancers (8 of 69 patients; 12%) were the most common diagnoses among phase 1 trial participants. The most common diagnoses of nonphase 1 participants included breast cancers (84 of 297 patients; 28%), gastrointestinal cancers (51 of 297 patients; 17%), hematologic cancers (37 of 297 patients; 13%), lung cancers (35 of 297 patients; 12%) and prostate cancers (35 of 297 patients; 12%). The ECOG performance status scores for phase 1 trial participants were significantly better than for nontrial participants (median, 2 vs 0; rank-sum test; P < .001).


Both phase 1 participants and other patients reported a median of 5 symptoms (mean, 5.0 symptoms vs 4.7 symptoms; rank-sum test; P = .416). Physical symptoms were equally common in the 2 groups (median, 3; rank-sum test; P = .538) as were psychologic symptoms (median, 2; rank-sum test; P = .486). Ninety-six percent (66 of 69 patients) of phase 1 patients reported ≥1 symptoms compared with 94% (279 of 297 patients) among nontrial participants. After adjusting for ECOG scores in an ordinal logistic regression model, phase 1 patients had a greater number of symptoms than other patients (β = 1.42; 95% confidence interval [95% CI], 0.70–2.14; P < .001).

When each symptom was evaluated individually, phase 1 patients were no more likely than other patients to have any of the 10 symptoms. However, after adjusting for ECOG scores in a logistic regression model, phase 1 patients were more likely to report a lack of appetite (odds ratio [OR], 2.41; 95% CI, 1.31–4.43 [P = .005]), lack of energy (OR, 3.90; 95% CI, 2.01–7.56 [P < .001]), pain (OR, 2.46; 95% CI, 1.35–4.48 [P = .003]), drowsiness (OR, 2.39; 95% CI, 1.31–4.35 [P = .004]), and nervousness (OR, 1.89; 95% CI, 1.05–3.37 [P = .032]) (Table 2).

Table 2. The Presence and Severity of Global Distress Index Symptoms
 Unadjusted Data  
 No. of Patients (%)  Adjusted OR*
Symptom PresenceConventional Treatment (N = 297)Phase 1 (N = 69)OR [95% CI]POR [95% CI]P
  • OR indicates odds ratio; 95% CI, 95% confidence interval.

  • *

    Adjusted for Eastern Cooperative Oncology Group performance status score.

Lack of appetite110 (37)30 (43)1.31 (0.77–2.22).3222.41 (1.31–4.43).005
Lack of energy202 (68)51 (74)1.33 (0.73–2.40).3403.90 (2.01–7.56)<.001
Pain162 (55)40 (58)1.15 (0.68–1.95).6062.46 (1.35–4.48).003
Drowsiness169 (57)43 (62)1.25 (0.73–2.15).4122.39 (1.31–4.35).004
Constipation93 (31)17 (25)0.72 (0.39–1.31).2771.29 (0.66–2.53).454
Dry mouth135 (45)30 (43)0.92 (0.54–1.56).7661.54 (0.86–2.78).147
Sadness115 (39)27 (39)1.02 (0.59–1.74).9501.54 (0.85–2.80).155
Worrying169 (57)46 (67)1.51 (0.87–2.63).1391.78 (0.99–3.23).056
Irritability121 (41)25 (36)0.83 (0.48–1.42).4911.07 (0.59–1.92).835
Nervousness115 (39)33 (48)1.45 (0.86–2.46).1661.89 (1.05–3.37).032
 Severity (Range)  Adjusted OR*
Symptom severityConventional Treatment (N = 297)Phase 1 (N = 69)OR [95% CI]POR [95% CI]P
Lack of appetite2.23 (0.8–4)2.18 (0.8–4)1.29 [0.78–2.13].3272.53 [1.42–4.51].002
Lack of energy2.87 (0.8–4)2.68 (0.8–4)1.02 [0.64–1.61].9423.74 [2.17–6.45]<.001
Pain2.78 (0.8–4)2.67 (1.6–4)1.04 [0.64–1.67].8862.82 [1.63–4.90]<.001
Drowsiness2.42 (0.8–4)2.42 (0.8–4)1.17 [0.73–1.87].5092.60 [1.53–4.42]<.001
Constipation2.77 (0.8–4)2.54 (0.8–4)0.67 [0.36–1.22].1901.24 [0.63–2.42].532
Dry mouth1.95 (0.8–3.2)2.17 (0.8–4)1.02 [0.61–1.70].9411.76 [1.00–2.09].050
Sadness2.37 (1–4)1.85 (1–3)0.82 [0.49–1.37].4531.40 [0.79–2.50].243
Worrying2.35 (1–4)2.37 (1–4)1.42 [0.88–2.28].1511.89 [1.12–3.19].016
Irritability2.26 (1–4)2.33 (1–4)0.83 [0.48–1.41].4891.13 [0.63–2.04].681
Nervousness2.45 (1–4)2.27 (1–4)1.29 [0.79–2.12].3110.916 [0.40–2.09].834

The severity of each of the 10 GDI symptoms in the 2 groups was equivalent for all physical and psychologic symptoms. However, after adjusting for ECOG scores in an ordinal logistic regression model, phase 1 participants reported more severe lack of appetite (OR, 2.53; 95% CI, 1.42–4.51 [P = .002]), lack of energy (OR, 3.74; 95% CI, 2.17–6.45 [P < .001]), pain (OR, 2.82; 95% CI, 1.63–4.90 [P < .001]), drowsiness (OR, 2.60; 95% CI, 1.53–4.42 [P < .001]), dry mouth (OR, 1.76; 95% CI, 1.00–2.09 [P = .050]), and worrying (OR, 1.89; 95% CI, 1.12–3.19 [P = .016]) (Table 2).

The mean GDI score (which combines symptom prevalence and severity) was identical in the 2 groups (mean, 1.15 for both groups; rank-sum test; P = .706). The mean scores for the physical component (1.24 vs 1.27; P = .775) and the psychologic component (1.05 vs 1.03; P = .805) of the GDI also were equivalent. However, after adjusting for ECOG scores in a linear regression model, phase 1 participants had higher (worse) GDI scores (β = 0.372; 95% CI, 0.16–0.583 [P = .001]). Adjusted physical subscale scores were higher for phase 1 participants (β = 0.496; 95% CI, 0.271–0.721 [P < .001]), but adjusted scores for the psychologic subscale were not (β = 0.247; 95% CI, −0.020–0.515 [P = .070]).


Phase 1 participants said they needed fewer services than other patients (mean, 0.75 vs 2.14 services; median, 0 vs 2 services). Forty-nine percent (34 of 69 patients) of phase 1 participants and 79% (235 of 297 patients) of other patients said they would like to receive at least 1 of the 4 services. This difference persisted after adjusting for ECOG scores in an ordinal logistic regression model (OR, 0.28; 95% CI, 0.17–0.48 [P < .001]).

Both groups were equally likely to want home nursing services (30 of 69 patients [44%] vs 142 of 297 [48%]; P = .516) in bivariate analysis and after adjusting for ECOG score in a logistic regression model (OR, 1.16; 95% CI, 0.65–2.06 [P = .623]) (Table 3). Phase 1 patients were significantly less likely to want home health aide services (4 of 69 patients [6%] vs 198 of 297 patients [67%]; P < .001), counselors (11 of 69 patients [16%[ vs 160 of 297 patients [54%]; P < .001), and a chaplain (7 of 69 patients [10%] vs 134 of 297 patients [45%]; P < .001). These differences persisted after adjusting for ECOG score (Table 3).

Table 3. Self-Reported Needs for Services
 Unadjusted Data  
 No. of Patients (%)  Adjusted OR*
Service DesiredConventional Treatment (N = 297)Phase 1 (N = 69)OR [95% CI]POR [95% CI]P
  • OR indicates odds ratio; 95% CI, 95% confidence interval.

  • *

    Adjusted for Eastern Cooperative Oncology Group performance status score.

Nurse142 (48)30 (44)0.84 [0.48–1.47].5161.16 [0.65–2.06].623
Home health aide198 (67)4 (6)0.03 [0.01–0.09]<.0010.05 [0.02–0.14]<.001
Counselor160 (54)11 (16)0.16 [0.07–0.33]<.0010.23 [0.11–0.47]<.001
Chaplain134 (45)7 (10)0.14 [0.05–0.32]<.0010.18 [0.08–0.42]<.001


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

This study helps define the symptoms of phase 1 trial participants and generates questions regarding the best model for providing palliative care to this unique population. Our data suggest that phase 1 participants have a greater symptom burden at the time of trial enrollment after adjusting for ECOG performance status. However, these patients are less likely to want some typical hospice services, such as home health aide assistance, counseling, or chaplain support.

One explanation for this discrepancy may be that phase 1 patients experience their symptoms differently. For instance, in a previous study, there was a correlation between higher quality-of-life scores and patient belief in benefit from clinical trial participation23; perhaps expectations influence a patient's experience of disease. Indeed, many phase 1 participants maintain hope for therapeutic benefit despite signing informed consent explicitly acknowledging the dose-finding and toxicity identification goals of phase 1 studies.24

This study also raised questions regarding which services would best meet the symptom management needs of phase 1 participants. These patients had fewer needs for hospice home care services, such as home health aides and chaplain or social work visits. However, it is possible that other services could be more useful. A recent study that used the same general cancer population described in our current report indicated that those patients placed a higher value on alternative services, such as vouchers for assistance in the home, transportation, peer support, and meal delivery.25 Further research also is needed to define the optimal set of services to meet the needs of patients in phase 1 trials.

One intriguing palliative care model is that of simultaneous care programs, which have been studied in advanced cancer populations. These programs combine cancer care and palliative care or hospice services and appear to offer many benefits, including increased rates of hospice enrollment and earlier hospice use,26–29 decreased emergency room visits and hospitalization,30 increased retention in studies,26 improved quality of life,29, 31 and possible cost savings.27, 32 Care coordination provided by a trained palliative care nurse is a common feature of many of these programs. This may be a feasible palliative care service model for advanced cancer patients, including phase 1 participants, who are well enough to receive anticancer treatment. Care coordination could use nursing resources that are readily available in existing experimental therapeutic programs, making it a practical option for palliative care in this environment. However, any program should be flexible enough to tailor services to meet the needs of diverse patients. In 1 simultaneous care study of phase 1 and 2 patients, a wide variability in symptoms was observed over time, suggesting a need for services that could be adapted to changing needs.27

Funding mechanisms for integrated palliative care and cancer care are as challenging as service design. CMS does not explicitly restrict phase 1 patients' access to hospice, but it is unclear whether funding for concurrent care is possible under current per diem payment structures.12 In open-access or simultaneous care models, the cost for disease-modifying treatment and associated imaging for many cancer patients would exceed hospice-level reimbursement, which, in 2008, is estimated at $135 and $601 for home and inpatient care, respectively.33 The financial complexity of the current reimbursement system was demonstrated during the University of Michigan-Hospice of Michigan partnership, which suffered financial loss in the hospice because of increased medication costs per patient but overall cost savings based on estimated Medicare billing for the partnership.30 New payment structures will need to be created if comprehensive end-of-life cancer care, including palliative and hospice care, becomes the standard of care.

The current study had 4 main limitations. First, palliative care needs among patients near the end of life change over time.27, 34 Our cross-sectional study design did not allow a longitudinal assessment of symptoms or correlation with disease burden, functional decline, service needs, and end-of-life choices. Second, patients with advanced cancer typically have multiple symptoms,35–37 and accurate symptom assessment is challenging.38 We relied on the MSAS-GDI for this study, but we recognize that alternative instruments or methods, such as palliative care consultations, surveys of caregiver burden or distress, or after-death surveys, may have provided additional information.

Third, there may be inherent differences in the psychologic or psychosocial characteristics of patients who participate in phase 1 oncology trials compared with other patients. Some of these differences, and particularly those related to patients' attitudes or beliefs regarding medical care, were not assessed in this study. Although the current study relied on ECOG performance status to adjust for differences between these patient groups, it is likely that other unobserved differences exist as well. Furthermore, data regarding disease-specific characteristics, such as stage and disease burden, which were unavailable in our study, may be interesting characteristics to adjust for in future studies. More generally, because performance status assesses function, and an increased performance status has been associated with increased symptom burden,39 it offers a valuable way to measure and adjust for limitations in activity that may be associated with needs for services.

Finally, we did not use paired comparisons of patients in each of these groups adjusting for duration of disease, treatment regimen, and psychologic factors. Doing so well may have produced different conclusions. However, our objective in this study was not to draw patient-level conclusions. Instead, we wanted to compare the symptoms and service needs of phase 1 trial participants as a group with those of other seriously ill patients with cancer and with selected patients who are typical of each of these groups. Thus, these findings should be of considerable interest in assessing the symptoms and service needs of these 2 populations. Nevertheless, further research is needed to explain the differences and similarities reported here, and such research should include paired comparisons with adequate adjustment for clinical, situational, and psychologic variables.

This study suggests that phase 1 trial participants have symptoms that should be assessed systematically and that targeted interventions should be designed to meet their unique needs. Oncology care is evolving, and treatment options for patients with advanced disease may include molecularly targeted cancer treatment, experimental therapies, palliative chemotherapy, and supportive care. Access to clinical trials, psychosocial support, and palliative care services all are critical components of high-quality cancer care.40–43

Although treatments for cancer have changed dramatically over the last 20 years, the Medicare Hospice benefit has not. Consequently, it risks becoming antiquated in the face of current oncology care standards. Further research is needed to better define the palliative care needs of advanced cancer patients, including phase 1 participants, and to explore cost-effective service delivery models for hospice and palliative care at different points along the cancer care continuum.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Supported by grant R01CA109540, by a Veterans Administration Advanced Research Career Development Award, and by a Presidential Early Career Award for Scientists and Engineers (D.J.C.).


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References
  • 1
    Roberts TGJr,Goulart BH,Squitieri L, et al. Trends in the risks and benefits to patients with cancer participating in phase 1 clinical trials. JAMA. 2004; 292: 21302140.
  • 2
    National Institutes of Health. Database of federally and privately sponsored clinical trials. Available at: Accessed on May 1, 2008.
  • 3
    National Cancer Institute. Clinical trials home page, vol 2008. Available at: Accessed on May 1, 2008.
  • 4
    Public Health Code. 42 Code of Federal Regulations, Part 418, Section 418.22(b): Medicare Hospice Regulations. Washington, DC: US Government Printing Office; 1996.
  • 5
    National Hospice and Palliative Care Organization. National Trend Summary 2004. Washington DC: National Hospice and Palliative Care Organization; 2005.
  • 6
    Kane RL,Klein SJ,Bernstein L,Rothenberg R,Wales J. Hospice role in alleviating the emotional stress of terminal patients and their families. Med Care. 1985; 23: 189197.
  • 7
    Teno J,Clarridge B,Casey V, et al. Family perspectives on end-of-life care at the last place of care. JAMA. 2004; 291: 8893.
  • 8
    Casarett D,Karlawish J,Crowley R,Mirsch T,Morales K,Asch DA. Improving use of hospice care in the nursing home: a randomized controlled trial. JAMA. 2005; 294: 211217.
  • 9
    Miller SC,Mor V,Teno J. Hospice enrollment and pain assessment and management in nursing homes. J Pain Symptom Manage. 2003; 26: 791799.
  • 10
    Miller S,Mor V,Wu N,Gozalo P,Lapane K. Does receipt of hospice care in nursing homes improve the management of pain at the end of life? J Am Geriatr Soc. 2002; 50: 507515.
  • 11
    Byock I,Miles SH. Hospice benefits and phase I cancer trials. Ann Internal Med. 2003; 138: 335337.
  • 12
    Byock I. Palliative care and the ethics of research: Medicare, hospice, and phase I trials. J Support Oncol. 2003; 1: 139141.
  • 13
    Paprstein H. Hospice patients in clinical cancer treatment trials. Clin J Oncol Nurs. 2005; 9: 729732.
  • 14
    Janisch L,Mick R,Schilsky RL, et al. Prognostic factors for survival in patients treated in phase I clinical trials. Cancer. 1994; 74: 19651973.
  • 15
    Bachelot T,Ray-Coquard I,Catimel G, et al. Multivariable analysis of prognostic factors for toxicity and survival for patients enrolled in phase I clinical trials. Ann Oncol. 2000; 11: 151156.
  • 16
    Casarett D,Karlawish J,Henry M,Hirschman K. Must patients with advanced cancer choose between a phase I trial and hospice? Cancer. 2002; 95: 16011604.
  • 17
    Portenoy RK,Thaler HT,Kornblith AB, et al. The Memorial Symptom Assessment Scale: an instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer. 1994; 30A: 13261336.
  • 18
    Casarett D,Crowley R,Hirschman K. How should clinicians describe hospice to patients and their families? J Am Geriatr Soc. 2004; 52: 19231928.
  • 19
    Casarett D,Crowley R,Stevenson C,Xie S,Teno J. Making difficult decisions about hospice enrollment: what do patients and families want to know? J Am Geriatr Soc. 2005; 53: 249254.
  • 20
    Rickerson E,Harrold J,Carroll J,Kapo J,Casarett D. Timing of hospice referral and families' perceptions of services: are earlier hospice referrals better? J Am Geriatr Soc. 2005; 53: 819823.
  • 21
    Oken MM,Creech RH,Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982; 5: 649655.
  • 22
    Seidenfeld J,Horstmann E,Emanuel EJ,Grady C. Participants in phase 1 oncology research trials: are they vulnerable? Arch Intern Med. 2008; 168: 1620.
  • 23
    Cheng JD,Hitt J,Koczwara B, et al. Impact of quality of life on patient expectations regarding phase I clinical trials. J Clin Oncol. 2000; 18: 421428.
  • 24
    Agrawal M,Emanuel EJ. Ethics of phase 1 oncology studies: reexamining the arguments and data. JAMA. 2003; 290: 10751082.
  • 25
    Casarett D,Fishman J,O'Dwyer PJ,Barg FK,Naylor M,Asch DA. How should we design supportive cancer care? The patient's perspective. J Clin Oncol. 2008; 26: 12961301.
  • 26
    Meyers FJ,Linder J. Simultaneous care: disease treatment and palliative care throughout illness. J Clin Oncol. 2003; 21: 14121415.
  • 27
    Meyers FJ,Linder J,Beckett L,Christensen S,Blais J,Gandara DR. Simultaneous care: a model approach to the perceived conflict between investigational therapy and palliative care. J Pain Symptom Manage. 2004; 28: 548556.
  • 28
    Bakitas M,Stevens M,Ahles T, et al. Project ENABLE: a palliative care demonstration project for advanced cancer patients in 3 settings. J Palliat Med. 2004; 7: 363372.
  • 29
    Ford Pitorak E,Beckham Armour M,Sivec HD. Project safe conduct integrates palliative goals into comprehensive cancer care. J Palliat Med. 2003; 6: 645655.
  • 30
    Beresford L,Byock I,Sheils Towhig J.Financial Implications of Promoting Excellence in End-of-Life Care. Vol 2008. New Brunswick, NJ: Robert Wood Johnson Foundation; 2002.
  • 31
    Finn JW,Pienta KJ,Parzuchowski J,Worden F. Palliative care project: bridging active treatment and hospice for terminal cancer [abstract]. Proc Am Soc Clin Oncol. 2002; 21. Abstract 1452.
  • 32
    Byock I,Twohig JS,Merriman M,Collins K. Promoting excellence in end-of-life care: a report on innovative models of palliative care. J Palliat Med. 2006; 9: 137151.
  • 33
    MEDPAC. Payment Basics: Hospice Services Payment System, revised 2008. Washington, DC: Medicare Payment Advisory Commission; 2008.
  • 34
    Casarett D,Coffey J,Hirschman K,Pierre L. Does a palliative care clinic have a role in improving end-of-life care? J Palliat Med. 2002; 5: 387396.
  • 35
    Walsh D,Donnelly S,Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1,000 patients. Support Care Cancer. 2000; 8: 175179.
  • 36
    Homsi J,Walsh D,Rivera N, et al. Symptom evaluation in palliative medicine: patient report vs systematic assessment. Support Care Cancer. 2006; 14: 444453.
  • 37
    Cleeland CS. Symptom burden: multiple symptoms and their impact as patient-reported outcomes. J Natl Cancer Inst Monogr. 2007; 37: 1621.
  • 38
    Paice JA. Assessment of symptom clusters in people with cancer. J Natl Cancer Inst Monogr. 2004; 32: 98102.
  • 39
    Cleeland CS,Mendoza TR,Wang XS, et al. Assessing symptom distress in cancer patients: the M. D. Anderson Symptom Inventory. Cancer. 2000; 89: 16341646.
  • 40
    [No authors listed] ASCO-ESMO consensus statement on quality cancer care. J Clin Oncol. 2006; 24: 34983499.
  • 41
    Levy MH,Back A,Bazargan S, et al. Palliative care. Clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2006; 4: 776818.
  • 42
    Holland JC,Andersen B,Breitbart WS, et al. Distress management. J Natl Compr Canc Netw. 2007; 5: 6698.
  • 43
    Adler NE,Page AEK. Cancer Care for the Whole Patient: Meeting Psychosocial Health Needs. Committee on Psychosocial Services to Cancer Patients/Families in a Community Setting. Board of Health Care Services, US Institute of Medicine. Washington, DC: National Academies Press; 2008.