Quality of life is significantly associated with survival in women with advanced epithelial ovarian cancer: An ancillary data analysis of the NRG Oncology/Gynecologic Oncology Group (GOG-0218) study

OBJECTIVE

Evaluate association between baseline quality of life (QOL) and changes in QOL measured by FACT-O TOI with progression-free disease (PFS) and overall survival (OS) in advanced epithelial ovarian cancer (EOC).

METHODS

Patients enrolled in GOG-0218 with completed FACT-O TOI assessments at baseline and at least one follow-up assessment were eligible. Baseline FACT-O TOI scores were sorted by quartiles (Q1-4) and outcomes compared between Q1 and Q2-4 with log-rank statistic and multivariate Cox regression adjusting for age, stage, post-surgical residual disease size, and performance status (PS). Trends in FACT-O TOI scores from baseline to the latest follow-up assessment were evaluated for impact on intragroup (Q1 or Q2-4) outcome by log-rank analysis.

RESULTS

Of 1152 eligible patients, 283 formed Q1 and 869 formed Q2-4. Mean baseline FACT-O TOI scores were 47.5 for Q1 vs. 74.7 for Q2-4 (P<0.001). Q1 compared to Q2-4 had worse median OS (37.5 vs. 45.6months, P=0.001) and worse median PFS (12.5 vs. 13.1months, P=0.096). Q2-4 patients had decreased risks of disease progression (HR 0.974, 95% CI 0.953-0.995, P=0.018), and death (HR 0.963, 95% CI 0.939-0.987, P=0.003) for each five-point increase in baseline FACT-O TOI. Improving versus worsening trends in FACT-O TOI scores were associated with longer median PFS (Q1: 12.7 vs. 8.6months, P=0.001; Q2-4: 16.7 vs. 11.1months, P<0.001) and median OS (Q1: 40.8 vs. 16months, P<0.001; Q2-4: 54.4 vs. 33.6months, P<0.001).

CONCLUSIONS

Baseline FACT-O TOI scores were independently prognostic of PFS and OS while improving compared to worsening QOL was associated with significantly better PFS and OS in women with EOC.

Effect of nivolumab on health-related quality of life in patients with treatment-naïve advanced melanoma: results from the phase III CheckMate 066 study

In patients with advanced melanoma in the CheckMate 066 study, baseline health-related quality of life (HRQoL) with nivolumab was maintained over time, with statistically significant and clinically meaningful improvements in some exploratory analyses, and no HRQoL improvements with dacarbazine. Added to the survival benefit of nivolumab, the benefit-to-risk ratio favors nivolumab over dacarbazine.

Background

Nivolumab has shown significant survival benefit and a favorable safety profile compared with dacarbazine chemotherapy among treatment-naïve patients with metastatic melanoma in the CheckMate 066 phase III study. Results from the health-related quality of life (HRQoL) analyses from CheckMate 066 are presented.

Patients and methods

HRQoL was evaluated at baseline and every 6 weeks while on treatment using the European Organisation for Research and Treatment of Care (EORTC) Core Quality of Life Questionnaire (QLQ-C30) and the EuroQoL Five Dimensions Questionnaire (EQ-5D). Via a multi-step statistical plan, data were analyzed descriptively, cross-sectionally, and longitudinally, adjusting for baseline covariates, in patients having baseline plus ≥1 post-baseline assessment.

Results

Baseline-adjusted completion rates for all HRQoL questionnaires across treatment arms were 65% and 70% for dacarbazine and nivolumab, respectively, and remained similar throughout treatment. The mean baseline HRQoL scores were similar for patients treated with nivolumab and dacarbazine. Baseline HRQoL levels with nivolumab were maintained over time. This exploratory analysis showed a between-arm difference in favor of nivolumab on the EQ-5D utility index and clinically meaningful EQ-5D improvements from baseline at several time points for patients receiving nivolumab. Patients treated with nivolumab did not show increased symptom burden as assessed by the EORTC QLQ-C30. No HRQoL change was noted with dacarbazine patients up to week 43, although the high attrition rate after week 13 did not allow any meaningful analyses. Patients receiving nivolumab deteriorated significantly later than those receiving dacarbazine on several EORTC QLQ-C30 scales and the EQ-5D utility index.

Conclusions

In addition to prolonged survival, these exploratory HRQoL results show that nivolumab maintains baseline HRQoL levels to provide long-term quality of survival benefit, compared with dacarbazine in patients with advanced melanoma.

Patient experience in the treatment of metastatic castration-resistant prostate cancer: state of the science

Background:

Contemporary therapies for metastatic castration-resistant prostate cancer (mCRPC) have shown survival improvements, which do not account for patient experience and health-related quality of life (HRQoL).

Methods:

This literature review included a search of MEDLINE for randomized clinical trials enrolling ⩾50 patients with mCRPC and reporting on patient-reported outcomes (PROs) since 2010.

Results:

Nineteen of 25 publications describing seven treatment regimens (10 clinical trials and nine associated secondary analyses) met the inclusion criteria and were critically appraised. The most commonly used measures were the Functional Assessment of Cancer Therapy-Prostate (n=5 trials) and Brief Pain Inventory Short Form (n=4 trials) questionnaires. The published data indicated that HRQoL and pain status augmented the clinical efficacy data by providing a better understanding of treatment impact in mCRPC. Abiraterone acetate and prednisone, enzalutamide, radium-223 dichloride and sipuleucel-T offered varying levels of HRQoL benefit and/or pain mitigation versus their respective comparators, whereas three treatments (mitoxantrone, estramustine phosphate and docetaxel, and cabazitaxel) had no meaningful impact on HRQoL or pain. The main limitation of the data were that the PROs utilized were not developed for use in mCRPC patients and hence may not have comprehensively captured symptoms important to this population.

Conclusions:

Recently published randomized clinical trials of new agents for mCRPC have captured elements of the patient experience while on treatment. Further research is required to standardize methods for measuring, quantifying and reporting on HRQoL and pain in patients with mCRPC in the clinical practice setting.

Best supportive care in clinical trials: review of the inconsistency in control arm design

Background:

Best supportive care (BSC) as a control arm in clinical trials is poorly defined. We conducted a review to evaluate clinical trials' concordance with published, consensus-based framework for BSC delivery in trials.

Methods:

A consensus-based Delphi panel previously identified four key domains of BSC delivery in trials: multidisciplinary care; supportive care documentation; symptom assessment; and symptom management. We reviewed trials including BSC control arms from 2002 to 2014 to assess concordance to BSC standards and to selected items from the CONSORT 2010 guidelines.

Results:

Of 408 articles retrieved, we retained 18 after applying exclusion criteria. Overall, trials conformed to the CONSORT guidelines better than the BSC standards (28% vs 16%). One-third of articles offered a detailed description of BSC, 61% reported regular symptom assessment, and 44% reported using validated symptom assessment measures. One-third reported symptom assessment at identical intervals in both arms. None documented evidence-based symptom management. No studies reported educating patients about symptom management or goals of therapy. No studies reported offering access to palliative care specialists.

Conclusions:

Reporting of BSC in trials is incomplete, resulting in uncertain internal and external validity. Such studies risk systematically over-estimating the net clinical effect of the comparator arms.

Prescribing for comorbid disease in a palliative population: focus on the use of lipid-lowering medications

BACKGROUND

The balance of benefit versus burden of ongoing treatments for comorbid disease in palliative populations as death approaches needs careful consideration given their particular susceptibility to adverse drug effects.

AIM

To provide descriptive data regarding the medications being prescribed to patients who have a life-limiting illness at the time of referral to a palliative care service in regional Australia, with particular focus on lipid-lowering medications.

METHODS

A prospective case note review of 203 patients reporting the number of medications prescribed and, for lipid-lowering medications, the indication and level of prevention sought (primary, secondary, tertiary). Rates were compared by performance status, disease phase and comorbidity burden.

RESULTS

Mean number of regular medications prescribed was 7.2, with higher rates observed in those patients with a non-malignant primary diagnosis (rate ratio 1.28, confidence interval (CI) 1.11-1.50) or poorer performance status (rate ratio 1.37, CI 1.11-1.69) and lower rates for those in the terminal phase of disease (rate ratio 0.48, CI 0.30-0.76). Over one fifth of patients were prescribed a lipid-lowering medication, and two fifths of these prescriptions were for primary prevention of cardiovascular disease. Patients in the highest quartile of Charlson Comorbidity Index score were 4.6 (CI 2.06-10.09) times more likely to be prescribed a lipid-lowering medication than those in the lowest quartile.

CONCLUSIONS

Polypharmacy is prevalent for this group of patients, placing them at high risk of drug-drug and drug-host interactions. Prescribing may be driven by risk factors and disease guidelines rather than a rational, patient-centred approach.

Does palliative home oxygen improve dyspnoea? A consecutive cohort study

Palliative oxygen for refractory dyspnoea is frequently prescribed, even when the criteria for long-term home oxygen (based on survival, rather than the symptomatic relief of breathlessness) are not met. Little is known about how palliative home oxygen affects symptomatic breathlessness. A 4-year consecutive cohort from a regional community palliative care service in Western Australia was used to compare baseline breathlessness before oxygen therapy with dyspnoea sub-scales on the symptom assessment scores (SAS; 0-10) 1 and 2 weeks after the introduction of oxygen. Demographic and clinical characteristics of people who responded were included in a multi-variable logistic regression model. Of the study population (n = 5862), 21.1% (n = 1239) were prescribed oxygen of whom 413 had before and after data that could be included in this analysis. The mean breathlessness before home oxygen was 5.3 (SD 2.5; median 5; range 0-10). There were no significant differences overall at 1 or 2 weeks (P = 0.28) nor for any diagnostic sub-groups. One hundred and fifty people (of 413) had more than a 20% improvement in mean dyspnoea scores. In multi-factor analysis, neither the underlying diagnosis causing breathlessness nor the demographic factors predicted responders at 1 week. Oxygen prescribed on the basis of breathlessness alone across a large population predominantly with cancer does not improve breathlessness for the majority of people. Prospective randomised trials in people with cancer and non-cancer are needed to determine whether oxygen can reduce the progression of breathlessness compared to a control arm.

Impact of a psychosocial intervention on performance status and coping

9611

Background: Psychosocial distress is a critical cancer comorbidity; new interventions are needed. Pathfinders, a manualized psychosocial care program, provides patient navigation, counseling, coping skills training, mind/body techniques, and lifestyle advice. Methods: This prospective, single-arm, pilot study enrolled adult metastatic breast cancer patients with prognosis ≥6 months. Consenting participants met with a Pathfinder (trained social worker) at least monthly, with interim phone/email contact. Pathfinders worked with patients to identify inner strengths, teach coping skills, engage complementary/alternative providers, employ mind/body techniques, and support healthy lifestyle. At baseline, month 3 and month 6, patients completed surveys including Patient Care Monitor (PCM; a review of systems with 6 subscales and a global quality of life [QOL] score), and Functional Assessment of Chronic Illness Therapy - Fatigue subscale (FACIT-F). Results: Participants (n=50) were: mean age 51.2 years (SD 11.5); 24% non-white; 74% married; 50% did not complete college; the cohort had advanced cancer and short prognosis with 6-month attrition from death, 18%. Scores on the PCM Distress subscale improved from baseline to 3 months with a mean change of -3.42 (n=36; p=0.008) and from baseline to 6 months of -4.11 (n=28; p=0.002). PCM Despair subscale scores also improved: mean change of -4.53 (p=0.006) and -6.93 (p=0.016), respectively. PCM QOL and FACIT-F scores improved from baseline to 3 months; however, the change at 6 months, with smaller sample, was not statistically significant. Mean change in QOL from baseline to 3 and 6 months was 2.88 (n=30; p=0.006) and 2.66 (n=25; p=0.079), respectively. Mean change in FACIT-F from baseline to 3 and 6 months was 2.91 (n=39; p=0.020) and 1.29 (n=32; p=0.407), respectively. Conclusions: Pathfinders had significant positive effect on key psychosocial and QOL outcomes, notably distress and despair, for cancer patients despite advanced disease and worsening symptoms.

No significant financial relationships to disclose.

Survivorship care planning needs in diffuse large B-cell lymphoma (DLBCL)

e20703

Background: Cancer survivorship care plans inform and direct care in the survivorship setting. These care plans should be tailored to individual medical information, needs, and circumstances, as providing excess information can be overwhelming. According to survivors of DLBCL, what are important components of care plans? Methods: We developed a 22-question survey to define and rate important survivorship health and psychosocial concerns; items were developed based upon literature review and experience in survivorship clinics. Through the tumor registry, 178 patients were identified who had been treated with curative intent (including stem cell transplant) without evidence of recurrence since 1/2006 and who continue to receive care at Duke University Medical Center. Results: Sixty-five survivors consented and returned a completed IRB approved survey (response rate 37%). Responders: 58% female, 88% white, and 75% from North Carolina, with mean age at diagnosis of 59.7 years; 42% had stage four disease at diagnosis and 12% had had a transplant. The majority of survey participants (62%) indicated that they preferred their oncologist and primary care provider to jointly manage their survivorship care. On a 1–10 scale, the top scoring issue (mean 9.67) was “A plan to screen for possible return of your cancer.” Other top scoring issues (mean 8.81 - 9.48) related to cancer history (treatment, complications, stage or late effects) and non-cancer health monitoring. The lowest scoring needs related to social support, sexuality, financial/legal issues, alternative medicine, and mental health services (mean 5.45 - 7.12). There was greater agreement among responders on the importance ratings of the higher scoring issues than the lower scoring ones (standard deviation 1.01 - 2.34 vs. 3.18 - 3.56). Conclusions: DLBCL survivors prefer care plans focused on medical issues, and health care coordinated jointly by oncologists and primary care physicians. The lower importance of psychosocial issues and alternative medicine in this population differs from survivors of other cancers, underscoring the importance of tailoring care plans by cancer subgroup.

No significant financial relationships to disclose.

Electronic patient-reported data capture as the foundation of a learning health care system

6522

Background: In a “learning healthcare system” clinical decisions are supported by accurate information delivered at point of care; information gathered today iteratively informs future care and research. Methods: Customized software on wireless tablet personal computers presented a review of systems (ROS) instrument, validated research surveys (e.g., quality of life [QOL]), and a satisfaction survey, tailored by user. The system was piloted in the Duke Cancer Clinics and affiliated hospitals. We previously demonstrated equivalence of electronic and paper survey data. We conducted a series of studies using similar procedures to evaluate feasibility, acceptability, and utility. Results: First, we assessed the ability to collect ROS data at point of care to inform the clinic visit for participating breast (n = 65), gastrointestinal (n = 113), and lung (n = 97) cancer patients. Duke physicians reported that the system's clinical reports informed care and increased dictation efficiency. Second, we assessed patient satisfaction in the breast cancer cohort. Participants found the computers easy to read (94%), navigate (99%), and use (98%); the system helped 74% remember forgotten concerns to report to their clinician. Third, we assessed whether these data could contribute to current research. If the patient was on another clinical trial, PRO data (e.g., pain, QOL) were delivered to the investigator for research purposes in real time; data governance rules provided assurance to investigators. Fourth, we identified whether the PRO data could inform future research directions. Symptoms monitored longitudinally in aggregate uncovered unmet needs. Sexual distress was an underserved concern; intervention studies were initiated. Warehoused PRO data were integrated with clinical trials, genomic, biomarker, radiology, and administrative datasets for analyses. The approach has been scaled to 4 clinics and 3 hospitals. Conclusions: An integrated, real-time, electronic data capture system that interdigitates PROs with clinical and other data allows creation of a learning oncology environment that continuously improves care and research. Advantages include: patient-centeredness, description of the PRO phenotype, interoperability, and interface with caBIG infrastructure.

No significant financial relationships to disclose.

Changes in anticholinergic load from regular prescribed medications in palliative care as death approaches

Although there is an understandable emphasis on the side effects of individual medications, the cumulative effects of medications have received little attention in palliative care prescribing. Anticholinergic load reflects a cumulative effect of medications that may account for several symptoms and adverse health outcomes frequently encountered in palliative care. A secondary analysis of 304 participants in a randomised controlled trial had their cholinergic load calculated using the Clinician-Rated Anticholinergic Scale (modified version) longitudinally as death approached from medication data collected prospectively by study nurses on each visit. Mean time from referral to death was 107 days, and mean 4.8 visits were conducted in which data were collected. Relationships to key factors were explored. Data showed that anticholinergic load rose as death approached because of increasing use of medications for symptom control. Symptoms significantly associated with increasing anticholinergic load included dry mouth and difficulty concentrating (P < 0.05). There were also significant associations with increasing anticholinergic load and decreasing functional status (Australia-modified Karnofsky Performance Scale; and quality of life (P < 0.05). This study has documented in detail the longitudinal anticholinergic load associated with medications used in a palliative care population between referral and death, demonstrating the biggest contributor to anticholinergic load in a palliative care population is from symptom-specific medications, which increased as death approached.

Preference for place of care and place of death in palliative care: are these different questions?

Place of death is at times suggested as an outcome for palliative care services. This study aimed to describe longitudinal preferences for place of care and place of death over time for patients and their caregivers. Longitudinal paired data of patient/caregiver dyads from a prospective unblinded cluster randomised control trial were used. Patients and caregivers were separately asked by the palliative care nurse their preference at that time for place of care and place of death. Longitudinal changes over time for both questions were mapped; patterns of agreement (patient and caregiver; and preference for place of death when last asked and actual placed of death) were analysed with kappa statistics. Seventy-one patient/caregiver dyads were analysed. In longitudinal preferences, preferences for both the place of care (asked a mean of >6 times) and place of death (asked a mean of >4 times) changed for patients (28% and 30% respectively) and caregivers (31% and 30%, respectively). In agreement between patients and caregivers, agreement between preference of place of care and preferred place of death when asked contemporaneously for patients and caregivers was low [56% (kappa 0.33) and 36% (kappa 0.35) respectively]. In preference versus actual place of death, preferences were met for 37.5% of participants for home death; 62.5% for hospital; 76.9% for hospice and 63.6% for aged care facility. This study suggests that there are two conversations: preference for current place of care and preference for care at the time of death. Place of care is not a euphemism for place of death; and further research is needed to delineate these. Patient and caregiver preferences may not change simultaneously. Implications of any mismatch between actual events and preferences need to be explored.

Oxygen for relief of dyspnoea in mildly- or non-hypoxaemic patients with cancer: a systematic review and meta-analysis

The aim of this study was to determine the efficacy of palliative oxygen for relief of dyspnoea in cancer patients. MEDLINE and EMBASE were searched for randomised controlled trials, comparing oxygen and medical air in cancer patients not qualifying for home oxygen therapy. Abstracts were reviewed and studies were selected using Cochrane methodology. The included studies provided oxygen at rest or during a 6-min walk. The primary outcome was dyspnoea. Standardised mean differences (SMDs) were used to combine scores. Five studies were identified; one was excluded from meta-analysis due to data presentation. Individual patient data were obtained from the authors of the three of the four remaining studies (one each from England, Australia, and the United States). A total of 134 patients were included in the meta-analysis. Oxygen failed to improve dyspnoea in mildly- or non-hypoxaemic cancer patients (SMD=-0.09, 95% confidence interval -0.22 to 0.04; P=0.16). Results were stable to a sensitivity analysis, excluding studies requiring the use of imputed quantities. In this small meta-analysis, oxygen did not provide symptomatic benefit for cancer patients with refractory dyspnoea, who would not normally qualify for home oxygen therapy. Further study of the use of oxygen in this population is warranted given its widespread use.

Patient-centered validation of 11 symptom indices to evaluate response to chemotherapy for advanced cancer

6524

Background: Symptom burden in advanced disease has relevance both for clinical practice and in evaluating the efficacy of new chemotherapeutic agents. This study aimed to identify patients’ highest priority symptoms for 11 advanced cancers, compare priority ratings with those obtained from clinicians, and construct brief symptom indices based on their combined input. Methods: 534 patients with advanced bladder, brain, breast, colorectal, head/neck, hepatobiliary/pancreatic, kidney, lung, lymphoma, ovarian or prostate cancer from a subset of National Comprehensive Cancer Network (NCCN) member institutions and 4 Cancer Health Alliance of Metropolitan Chicago organizations completed a survey of priority symptoms and concerns and a disease-specific FACT QOL measure. 112 physicians at NCCN institutions completed a rating of whether symptoms and concerns were considered disease- or treatment-related. Symptoms endorsed more often than chance probability were retained. Expert clinician and patient ratings were equally weighted in item selection. Responses to symptom index items drawn from the QOL questionnaires allowed for validation analyses. Results: Items comprising 2 to 3 subscales (up to 20 items in length) were retained for each of the 11 disease-specific symptom indices. Content-determined subscales consisted of symptoms and concerns that were 1) exclusively or predominantly disease-related symptoms (DRS); 2) exclusively or predominantly treatment side effects (TSE); and 3) descriptive of general function or well-being (F/WB). For example, the NCCN/FACT Breast Cancer Symptom Index (FBSI) contains 17 items: 8 DRS, 4 TSE and 5 F/WB. Data on 14 of 17 FBSI items showed good internal consistency (a=.89). Correlations between FBSI and FACT-B scores were high for physical well-being, functional well-being and breast cancer subscales (r = 0.83, 0.77, and 0.61 respectively). Correlation with emotional well-being subscale was 0.55. FBSI scores differed across PSR groups in the appropriate direction (p<0.0001). Conclusions: NCCN/FACT disease-specific questionnaires have been transformed into brief, patient-centered symptom indices that can be used as stand-alone measures in oncology research and practice.

No significant financial relationships to disclose.

Comorbidity, age and stage at diagnosis in colorectal cancer (CRC)

6554

Background: Stage at diagnosis is a crucial predictor of outcome in CRC. The purpose of this study is to determine if comorbidity and age affect the stage at which CRC is diagnosed. Identifying variables that influence stage might improve outcomes in CRC. Due to frequent contact with the health care system, we hypothesize that patients with greater comorbidity and older age are more likely to be diagnosed with early-stage disease. Methods: We present data from two distinct patient populations: using the Cancer Care Outcomes Research and Surveillance Consortium (CanCORS) database, we identified CRC patients treated at 15 Veterans Administration (VA) hospitals from 2003-present. We also identified CRC patients treated from 2003-present at 10 non-VA, fee-for-service (FFS) practices in North and South Carolina. Data were abstracted by retrospective chart review. Comorbidity was calculated by the Charlson comorbidity index (CCI) with high comorbidity defined as CCI =3. Older age was defined as age =70 years. Data were analyzed using logistic regression where the odds of late stage at diagnosis were modeled as influenced by older age, high CCI, and race. The analysis included estimation of adjusted and unadjusted odds ratios. Results: 347 VA and 282 FFS patients were included. 98% VA vs 50% FFS were male; 43% VA vs 27% FFS were aged =70; 56% VA vs 70% FFS were white; 26% VA vs 44% FFS presented with metastatic CRC; and 21% VA vs 6% FFS had a CCI =3. In both patient populations, regression analysis showed that older age, high CCI and white race were not significant predictors of stage at diagnosis. VA 95% confidence intervals (CI's) were 0.52–1.41 (age =70), 0.50–1.75 (CCI =3), and 0.42–1.11 (white race). FFS 95% CI's were 0.52–1.53 (age =70), 0.36–2.78 (CCI =3), and 0.74–2.11 (white race). Broader 95% CI's in the FFS analysis were due to smaller sample size. Conclusions: In CRC patients, age and comorbidity were not related to stage at diagnosis. The findings are similar whether the patients were treated in a fee-for-service or VA health system. While older age and greater illness might provide more contact with the health care system, this exposure did not result in earlier diagnosis of CRC. Future studies will examine the impact of comorbidity on CRC treatment and survival.

No significant financial relationships to disclose.

Choice of adjuvant and first-line metastatic chemotherapy (CT) for colorectal cancer (CRC) treated in the Carolinas

17039

Background: CT choice is influenced by many factors including published evidence, guidelines, cost, reimbursement, patient considerations, key opinion leaders, and anecdote. Substantial locoregional variation in practice patterns can exist, and therefore studies of locoregional practice provide important information on local drivers of care. Methods: Using a population-based strategy, we identified CRC patients who developed metastatic disease since 6/1/03 from 9 Duke Oncology Network community practices and 1 academic practice in North and South Carolina. Demographic, comorbidity, diagnostic, stage, initial treatment, and metastatic treatment data were abstracted by retrospective chart review, double-entered and verified for accuracy. Results: Of the first 743 charts screened, 306 were eligible (mean age 61 (SD 13), 49% male; 65% white; 22% black; 77% colon cancer and 19% rectal; stages II 8%, III 16%, IV 64%). 26 earlier stage rectal cancer patients received neoadjuvant treatment, 50% infusional fluorouracil (5FU) and 42% capecitabine (Cap). 46 colon cancer patients received adjuvant CT, including 5FU/leucovorin (LVN; 54%), 5FU/LVN/oxaliplatin (21%), Cap (9%), and 5FU/LVN/irinotecan (7%). First-line CT for metastatic colon cancer (n=149) included FOLFOX+-bevacizumab (Bev; 42%), Cap/oxaliplatin +- Bev (23%), 5FU/LVN + Bev (9%), FOLFIRI +- Bev (7%), IFL +- Bev (7%), clinical trial (7%), Cap (3%), and unknown (1%). 54% of patients received Bev overall, reflecting 49% usage before 6/05 and 69% after 6/05. CT was not offered for 25 (8%) at initial diagnosis. Conclusion: Locoregional practice patterns in the Carolinas suggest that for adjuvant treatment of CRC, oxaliplatin has been used in 21% of adjuvant and 75% of first-line metastatic colon CT regimens, and that bevacizumab use has increased to 69% of first-line metastatic CRC patients.

No significant financial relationships to disclose.

E-tablets to collect research-quality patient-reported data

6609

Introduction: Programmed wireless notebook computers (e-tablets) can collect review of systems (ROS) data at point of care. Patients complete surveys in the clinic waiting area; a report is generated for the subsequent clinical visit. Are e-tablets a feasible, acceptable method for collecting data directly from patients in an academic cancer clinic? Can they reliably collect other survey data? Methods: We used PACE e-tablets (SOS, Inc.) to administer Functional Assessment of Cancer Therapy (FACT)-G, FACT-B, MD Anderson Symptom Inventory (MDASI), FACIT-Fatigue, and Self Efficacy instruments, in addition to the PACE Patient Care Monitor (PCM) ROS survey. Participants were 66 breast cancer patients in Duke Breast Cancer Clinic. At 4 visits in 6 months, participants completed all electronic and 1 paper survey. Subscales were compared using paired t-tests. Patients completed an electronic survey of satisfaction with PCM. Results: Mean age, 55 (SD 12); 77% Caucasian; 49% no college degree; 68% married; 61% metastatic cancer. Patients strongly supported e-tablets: easy to read (94%), easy to respond to questions (98%), weight of computer comfortable (87%). Satisfaction increased over time: helpful for reporting symptoms (75–88%), would recommend PCM to other patients (87–94%). 75% indicated PCM helped them remember symptoms to discuss with the clinician. Responses to paper and electronic surveys were nearly identical for 3 of 4 FACT-G, and all FACT-B, MDASI, and FACIT-Fatigue subscales (all p>0.31). Responses on FACT-G Social Wellbeing and 4 Self Efficacy subscales differed from paper to electronic (all p<0.006, all electronic scores poorer). Conclusions: Patients are satisfied with e-tablets. E- tablets furnish comparable data to those collected by paper on nearly all scales tested. PCM offers a valid, feasible method for collecting research-quality, clinically relevant data from patients in outpatient academic oncology.

No significant financial relationships to disclose.

Predictors of fatigue and quality of life in a prospective palliative care cohort

8571

Background: Palliative care seeks to minimize distress at the end of life. Fatigue significantly diminishes quality of life (QOL) in this population. Are there potentially modifiable factors that influence fatigue and QOL? Methods: This analysis focuses on a subset of 198 patients from a larger 2×2×2 factorial randomized trial of pain education and care coordination conducted in South Australia. Selected participants were adults referred to a community palliative care service with pain in the preceding 3 months and a hemoglobin assessment within 14 days of enrollment. Pain, other symptoms, and Australia-modified Karnofsky Performance Status (AKPS) were recorded at enrollment. Predictors considered were anxiety, depression, dyspnea, constipation, pain, AKPS, hemoglobin, age, and gender. Dependent variables were global QOL from the McGill QOL Questionnaire and fatigue. Using forward stepwise linear regression, multivariate models predicting fatigue and QOL were constructed from significant univariate variables. Results: Mean age was 69 (standard deviation (SD) 13); 97% had cancer. Most frequent diagnoses were lung (18%), hematological (15%), and colorectal (15%) malignancies. Mean hemoglobin was 11.4 gm/dL (SD 1.9); median AKPS 60%; mean worst pain 4.0 (SD 3.4; 0–10 scale). Distressing symptoms (3–4 on 0–4 scales) included dyspnea (22%), constipation (13%), anxiety (11%), and depression (6%). Mean QOL was 5.9 (SD 2.0) on a 0–10 scale; mean fatigue was 2.3 (SD 1.0) on a 0–4 scale. The final multivariate model predicting fatigue included AKPS (p<0.01), constipation (p=0.02), and dyspnea (p=0.06). Hemoglobin was not predictive of fatigue (univariate p=0.7069). QOL was significantly influenced by fatigue (p=0.03), anxiety (p< 0.01), and AKPS (p= 0.01). Conclusions: Fatigue was driven by performance status, constipation, and dyspnea. In contrast to an oncology population, hemoglobin was not a significant contributor to fatigue in this population, consistent with other palliative care cohorts. QOL was driven by fatigue, anxiety, and performance status. This analysis of a prospectively collected population suggests that performance status, constipation, dyspnea, and anxiety are potentially modifiable variables impacting fatigue and QOL in the palliative care setting.

No significant financial relationships to disclose.

Improving palliative care: A 2x2x2 factorial cluster randomized controlled trial of case conferencing and educational outreach visiting

8517

Background: Evidence-based palliative care service delivery models are needed. General practitioner (GP) and patient-centered case conferences may increase multidisciplinary interaction and enhance patient care. Educational outreach visiting in pain management may empower learners and improve care. Methods: Three interventions were tested against a routine care control in a 2×2×2 cluster factorial randomized controlled trial. Interventions were case conferencing, educational visiting for GPs, and educational visiting for patients. Subjects were adult patients referred to palliative care services in southern Adelaide, Australia, with any pain in the preceding 3 months. Participants were followed longitudinally until death. Main outcomes included performance status (Australian-modified Karnofsky Performance Status (AKPS)) and hospitalization rates. Longitudinal intention-to-treat analyses using cluster-specific methods were conducted. The sample goal was 460. Results: 461 participants were enrolled from 4/02–6/04. Mean age was 71 yrs, 50% were male, 91% had cancer. Mean and median survival was 146 and 87 days; median baseline AKPS was 60%. When participants had AKPS <70 (i.e. required a caregiver) at referral, those randomized to case conferencing or patient education had higher mean performance status than routine care (case conferencing: average daily AKPS, 54.9% vs 46.3%, p=0.0106; patient education: 54.7% vs 46.4%, p=0.0120). GP education did not improve performance status. Both case conferencing and patient education significantly decreased hospitalization rates compared to routine care (case conferencing: least-squares mean number of hospitalizations, 1.4 (standard error (SE) 0.1) vs 1.9 (SE 0.1), p=0.0002; patient education: 1.4 (SE 0.1) vs 1.8 (SE 0.1), p=0.0078). The addition of both interventions decreased hospitalizations to 1.2 (SE 0.1). Conclusion: At a time when declining function and higher healthcare utilization is the norm, palliative care that includes a case conference or patient educational visiting for pain leads to improved performance status and decreased hospitalizations. Patients with deteriorating performance status derive the greatest benefit.

No significant financial relationships to disclose.

Conflict associated with decisions to limit life-sustaining treatment in intensive care units

OBJECTIVE

To determine the incidence and nature of interpersonal conflicts that arise when patients in the intensive care unit are considered for limitation of life-sustaining treatment.

DESIGN

Qualitative analysis of prospectively gathered interviews.

SETTING

Six intensive care units at a university medical center.

PARTICIPANTS

Four hundred six physicians and nurses who were involved in the care of 102 patients for whom withdrawal or withholding of treatment was considered.

MEASUREMENTS

Semistructured interviews addressed disagreements during life-sustaining treatment decision making. Two raters coded transcripts of the audiotaped interviews.

MAIN RESULTS

At least 1 health care provider in 78% of the cases described a situation coded as conflict. Conflict occurred between the staff and family members in 48% of the cases, among staff members in 48%, and among family members in 24%. In 63% of the cases, conflict arose over the decision about life-sustaining treatment itself. In 45% of the cases, conflict occurred over other tasks such as communication and pain control. Social issues caused conflict in 19% of the cases.

CONCLUSIONS

Conflict is more prevalent in the setting of intensive care decision making than has previously been demonstrated. While conflict over the treatment decision itself is most common, conflict over other issues, including social issues, is also significant. By identifying conflict and by recognizing that the treatment decision may not be the only conflict present, or even the main one, clinicians may address conflict more constructively.

Families looking back: one year after discussion of withdrawal or withholding of life-sustaining support

OBJECTIVE

To identify critical psychosocial supports and areas of conflict for families of intensive care unit (ICU) patients during decisions to withdraw or withhold life-sustaining treatment.

DESIGN

Cross-sectional survey.

SETTING

Six intensive care units in a tertiary care academic medical center.

PARTICIPANTS

Forty-eight family members, one per case, of patients previously hospitalized in the ICU who had been considered for withdrawal or withholding of life-sustaining treatment.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Two raters coded transcripts of audiotaped interviews with family members about their experiences in the ICU and the decision-making process for withdrawing or withholding life-sustaining treatment. Codes identified sources of conflict and personal, institutional, and staff supports on which families relied during the decision-making process. Forty-six percent of respondents perceived conflict during their family member's ICU stay; the vast majority of conflicts were between themselves and the medical staff and involved communication or perceived unprofessional behavior (such as disregarding the primary caregiver in treatment discussions). Sixty-three percent of family members previously had spoken with the patient about his or her end-of-life treatment preferences, which helped to lessen the burden of the treatment decision. Forty-eight percent of family members reported the reassuring presence of clergy, and 27% commented on the need for improved physical space to have family discussion and conferences with physicians. Forty-eight percent of family members singled out their attending physician as the preferred source of information and reassurance.

CONCLUSIONS

Many families perceived conflict during end-of-life treatment discussions in the ICU. Conflicts centered on communication and behavior of staff. Families identified pastoral care and prior discussion of treatment preferences as sources of psychosocial support during these discussions. Families sought comfort in the identification and contact of a "doctor-in-charge." ICU policies such as family conference rooms and lenient visitation accommodate families during end-of-life decision-making.