Supportive Care Clinic Research Articles

Frequency and factors associated with falls in patients with advanced cancer presenting to an outpatient supportive care clinic

The aim of this study was to determine the frequency and factors associated with fall episodes in advanced cancer patients. We analyzed data that included demographic characteristics, utilization of assistive devices, cancer diagnosis, metastatic site, performance status, medications including hypnotics and opioids, Edmonton Symptom Assessment Scale (ESAS) score, and Memorial Delirium Assessment Scale (MDAS) score in 384 consecutive patients who were newly referred to the Supportive Care Clinic at the MD Anderson Cancer Center from January 1 to December 31, 2009. All patients completed standardized forms to report falls within the last month. Multivariate backward regression analyses were employed to identify factors predictive of falls in advanced cancer. The mean age of patients was 58 years, and 192 (50%) were male. Mean (SD)/median score for pain was 5 (2.8), 5; fatigue 5.6 (2.6), 6; sleep disturbance 5(2.7), 5; drowsiness 3.7(3), 3; and anorexia 5(3), 5. Some 31 patients (8%) reported fall episodes within the past month, 17 (55%) of whom reported the use of assistive devices. Using assist devices (OR = 5.5, 95% CI: 2.6-11.9, p < 0.0001) and taking zolpidem (OR = 3.39, 95% CI: 1.39-7.7, p = 0.008) were associated with an enhanced chance of falling. Higher MDAS score (4.00 vs. 1.42, p = 0.001) and MDAS positive screening for delirium (21 vs. 3.6%, p < 0.001) were also associated with falls. However, severity on the ESAS at the initial consult was not associated with falls. We conclude that 31 of 384 patients (8%) with advanced cancer receiving outpatient supportive care reported falls in the previous month. Patients with assistive devices, taking zolpidem, and with a higher MDAS score, and a positive delirium screening reported more frequent falls. Further studies are warranted.

Self-reported sleep disturbance in patients with advanced cancer: Frequency, intensity, and factors associated with response to outpatient supportive care consultation — A preliminary report

Self-reported sleep disturbance (SD) is a distressing symptom in patients with advanced cancer. There are limited data on the treatment of SD and predictors to response of SD to outpatient supportive care clinic (OPC) consultation. The aims of our study was to determine the frequency, intensity, and correlates of SD as assessed with the Edmonton Symptom Assessment System (ESAS) sleep item at the time of initial consultation and identify the predictors of improvement in SD at follow-up. We reviewed the records of consecutive patients with advanced cancer presenting to the OPC. ESAS scores were obtained at the initial and subsequent visits between January 2008 and February 2010. All patients underwent screening for SD (0-10 scale: 0 = best sleep, presence of SD defined as ≥ 3) and interdisciplinary assessment and treatment, including drug review, counseling, sleep hygiene review, and drug therapy. A response was defined as a 1-point improvement at the follow-up visit on the Edmonton Symptom Assessment Scale (ESAS) sleep item score. Baseline patient characteristics, medication use, and ESAS scores were analyzed to determine their association with response. The median age was 58 years, and 53% of patients were men. The most common cancer type was head and neck or lung (36%). Of the 442 patients, 330 had baseline SD (score ≥ 3/10, 75%). Median and mean (standard deviation) baseline SD scores were 5 and 5.1 (2.9). The multivariable regression model found the intensity of baseline ESAS sleep item scores to be associated with baseline sedative use, baseline ESAS pain scores, baseline ESAS fatigue scores, baseline ESAS feeling of well-being scores, and sedative use (R 2 = 0.22). Sleep disturbance response at first follow-up was seen in 196 of 330 patients (59%). Moderate to high SD score and anxiety at initial visit with odds ratios (OR) of 2.53 (p = 0.0007) and 1.59 (p = 0.048), respectively, were associated with a response. Both the frequency and severity of SD were high. Response to supportive care consultation was substantial. The severity of SD and anxiety at the initial visit predicted a response at first follow-up. Further research is needed.

The frequency of falls in patients with advanced cancer followed in an outpatient palliative care center.

225 Background: Falls are a concern in patients who are frail, have advance age or severe underlying medical condition as a significant cause of morbidity and reduced quality of life. It has been reported to occur in up to 50% of palliative care patients in different settings. Our aim was to determine the frequency of falls and identify its predictors and correlates in patients with advanced cancer. Methods: We reviewed 1,984 consecutive patients with advanced cancer seen in the outpatient supportive care center and determined the frequency of patient reported falls within the last month prior to the visit. Baseline patient characteristics, symptom severity scores in the Edmonton Symptom Assessment Scale (ESAS), medication use, functional status, and use of assistive devices were used to identify factors that are predictive of falls using backwards stepwise logistic regression. Results: 1,041 (52%) were female, 1,377 (69%) were non-Hispanic-white, 343 (17%) had peripheral neuropathy, 237 (12%) were on psychotropic medications, and 1,140 (58%) were on opioids. There were 211 (11%) patient reported falls. Presence of brain metastasis, use of assistive device, ECOG, ESAS depression, and weight prior to visit were significantly associated with patient reported falls in the multivariate model. Conclusions: One tenth of patients seen in the outpatient supportive care clinic reported falls. Our findings showing that falls to be associated with brain metastasis, assistive device, functional status, ESAS depression and weight are useful in stratifying high risk patients. They also show that certain medications deemed as risk factors in other populations were not associated with falls. [Table: see text]

Code status discussion: Just have one

code status discussion between a physician and a patient with cancer. The 2 videos were the same except that 1 video ended with the physician recommending a do-not-resuscitate (DNR) order for his patient, deemed by the investigators to be the ‘‘beneficence approach,’’ whereas the other video ended with the physician eliciting his patient’s choice of code status by asking a question, described as the ‘‘autonomy approach.’’ Participating patients were randomized to the sequence in which the videos were watched (recommendation video first or question video first). First, a note about methods. The recognized limitation of this study’s method (having all participants view both videos, thereby serving as their own control and lowering the necessary sample size), is that responses to the second intervention are biased by the first. The investigators overcome this built-in bias by basing their conclusions about what influenced DNR choice only on participants’ responses to the first video. However, in their prior work using the same methodology to compare patients’ reactions to sitting versus standing physicians delivering ‘‘bad news,’’ 2,3 patients rated the second physician as more compassionate despite their overall preference for the sitting physician. No such shift was observed in this study. The investigators suggest the high degree of compassion exhibited by the physician in the video may have masked patients’ tendency to shift their perception of compassion, and its impact on their DNR choice. Another likely explanation for patients’ consistency is the often-found phenomenon that individuals are strongly biased toward being consistent in their choices 4 to avoid cognitive dissonance. 5 The investigators’ hypothesis, that patients would more frequently choose DNR for the patient in the first video when the discussion ended with a recommendation, was not supported. The type of ending did not appear to impact DNR choice. This result is surprising. It is well established in other contexts that a physician’s recommendation is influential and sometimes determinative of patients’ choices. 6-8 Why this different result within this context?

Patient–physician communication about code status preferences

Code status discussions are important in cancer care, but the best modality for such discussions has not been established. The objective of this study was to determine the impact of a physician ending a code status discussion with a question (autonomy approach) versus a recommendation (beneficence approach) on patients' do-not-resuscitate (DNR) preference. Patients in a supportive care clinic watched 2 videos showing a physician-patient discussion regarding code status. Both videos were identical except for the ending: one ended with the physician asking for the patient's code status preference and the other with the physician recommending DNR. Patients were randomly assigned to watch the videos in different sequences. The main outcome was the proportion of patients choosing DNR for the video patient. A total of 78 patients completed the study, and 74% chose DNR after the question video, whereas 73% chose DNR after the recommendation video. Median physician compassion score was very high and not different for both videos. All 30 of 30 patients who had chosen DNR for themselves and 30 of 48 patients who had not chosen DNR for themselves chose DNR for the video patient (100% versus 62%). Age (odds ratio = 1.1/year) and white ethnicity (odds ratio = 9.43) predicted DNR choice for the video patient. Ending DNR discussions with a question or a recommendation did not impact DNR choice or perception of physician compassion. Therefore, both approaches are clinically appropriate. All patients who chose DNR for themselves and most patients who did not choose DNR for themselves chose DNR for the video patient. Age and race predicted DNR choice.

Consistency of symptom clusters among advanced cancer patients seen at an outpatient supportive care clinic in a tertiary cancer center.

Advanced cancer patients often develop severe physical and psychological symptom clusters (SCs), but limited data exist on their consistency or severity after an outpatient interdisciplinary team consultation led by palliative care specialists. The primary aim of the study was to determine the consistency and severity of SCs in advanced cancer patients in this setting. A total of 1373 patients with advanced cancer who were referred to The University of Texas MD Anderson Cancer Center's Outpatient Supportive Care Center between January 2003 and October 2008 with a complete Edmonton Symptom Assessment Scale (ESAS; 0-10 scale) occurred at initial and first follow-up visit were reviewed (median 14 days, range 1-4 weeks). We used a Wilcoxon signed-rank test to determine whether symptoms changed over time, and a principal components factor analysis with varimax rotation to determine SCs at baseline and at first follow-up. The number of factors calculated was determined based upon the number of eigenvalues. The patients' ratings of the following symptoms (mean, SD) at the initial and follow-up visits, respectively, were: fatigue 6.2 (2.3) and 5.7 (2.5, p < 0.0001), pain 5.4 (2.9) and 4.6 (3, p < 0.0001), nausea 2.2 (2.8) and 2.0 (2.6, p < 0.0001), depression 3.0 (2.9) and 2.5 (2.7, p < 0.0001), anxiety 3.4 (3.0) and 2.8 (2.8, p < 0.0001), drowsiness 4.8 (3.1) and 4.4 (3.1, p < 0.0001), dyspnea 3.0 (2.9) and 2.7 (2.8), p < 0.0001), loss of appetite 4.2 (2.7) and 3.9 (2.7, p < 0.0001), sleep disturbances 4.2 (2.6) and 3.8 (2.6, P < 0.0001), and well-being 4.3 (2.5) and 3.9 (2.3, p < 0.0001). Cluster composition differentiated into physical (fatigue, pain, nausea, drowsiness, dyspnea, and loss of appetite) and psychological (anxiety and depression) components at the initial visit, and these two SCs were consistent upon follow-up. We conclude that SCs remain constant between baseline and near-term follow-up but that the severity of those symptoms lessened during that interval. This knowledge may allow palliative care teams to provide more targeted and higher-quality care, but further studies are needed.

Distress management in a community oncology center.

107 Background: NCCN and the Institute of Medicine recommend distress screening for all cancer patients. Surveys have found that 20% to 47% of newly diagnosed and recurrent cancer patients experience significant levels of distress. Less than half of distressed patients with cancer are actually identified and referred for psychosocial help. The development of screening tools, processes, appropriate timing, and systems for follow up on concerns is an ongoing challenge for many cancer programs. Mountain States Tumor Institute p,iloted a distress screen process as part of our work with NCCCP. Methods: From October 24, 2011, to April 30, 2012, 133 patients completed distress screens. Most frequently reported concerns included fatigue (50%), sleep changes (32%), and anxiety/worry (31%). 51% of patients reported three or more concerns. 90% of all patients reporting anxiety also reported fatigue or sleep problems. All patients who reported concerns were offered social work and other supports. Screens were reviewed by social workers and discussed at weekly multidisciplinary psychosocial care rounds by social workers, psychiatrist, nurse practitioners, and chaplains. A variety of interventions were used to respond to distress including symptom management, education and referrals to on-site social workers, chaplains, supportive care clinic, integrative medicine and psychiatry. Results: A task was created in Mosaiq Electronic Medical Record to cue Health Information Specialists to schedule patients for distress screening as close as possible to 45 days from first chemotherapy treatment. Patients completed screens before seeing their medical provider. The majority of patients screened received one or more interventions in follow-up. Conclusions: EMRs can be effective tools to cue scheduled screening of patient distress. The use of multidisciplinary psychosocial care rounds can effectively help clinics plan interventions to alleviate patient distress. The use of rounds can ensure staff from different disciplines are not duplicating efforts and can help determine appropriate timing of various psychosocial interventions. Project funded with Federal funds from the NCI, Contract No HHSN261200800001E.

Conceptual models for integrating palliative care at cancer centers.

Palliative care programs are rapidly evolving in acute care facilities. Increased and earlier access has been advocated for patients with life-threatening illnesses. Existing programs would need major growth to accommodate the increased utilization. The objective of this review is to provide an update on the current structures, processes, and outcomes of the Supportive and Palliative Care Program at the University of Texas M.D. Anderson Cancer Center (UTMDACC), and to use the update as a platform to discuss the challenges and opportunities in integrating palliative and supportive services in a tertiary care cancer center. Our interprofessional program consists of a mobile consultation team, an acute palliative care unit, and an outpatient supportive care clinic. We will discuss various metrics including symptom outcomes, quality of end-of-life care, program growth, and financial issues. Despite the growing evidence to support early palliative care involvement, referral to palliative care remains heterogeneous and delayed. To address this issue, we will discuss various conceptual models and practical recommendations to optimize palliative care access.

Self-Reported Constipation in Patients With Advanced Cancer: A Preliminary Report

ContextConstipation is often inadequately assessed and underdiagnosed in patients with advanced cancer. Many studies use patient-reported constipation (PRC) as an outcome. ObjectivesThe aim was to compare the accuracy of PRC as compared with the modified Rome III (ROME) criteria and to determine the agreement between PRC, physician assessment of constipation, and objective assessment of constipation by modified ROME criteria among outpatients with advanced cancer. MethodsPatients with advanced cancer attending a supportive care clinic were screened. Constipation was assessed using the modified ROME criteria, patient report (yes or no and rated 0–10; 10=worst possible symptom), and physician assessments (yes or no and rated 0–10). ResultsOne hundred patients were enrolled, and 50 of 100 patients (50%) met the modified ROME criteria for constipation. Disagreement between ROME criteria and the patient report (yes/no) was found in 33 patients (33%) and between ROME criteria and the physician assessment (yes/no) in 39 patients (39%). The best combination of sensitivity (0.84) and specificity (0.62) was found with scores ≥3/10 for PRC. ConclusionWe found a high frequency of constipation. The limited agreement with modified ROME criteria suggests that a patient's self-report as yes or no is not useful for clinical practice. Patient self-rating on a 0 to 10 scale (score of three or greater) seems to be the best tool for constipation screening among this population. More research is needed to identify the best way to assess constipation in patients with advanced cancer.

Patient-physician communication about code status preferences: A randomized controlled trial.

9049 Background: Code status discussions are important in cancer care. The best modality for such discussions has not been established. Our objective was to determine the impact of a physician ending a code status discussion with a question (autonomy approach) versus a recommendation (beneficence approach) on patients’ do-not-resuscitate (DNR) preference. Methods: Patients in a supportive care clinic watched two videos showing a physician-patient discussion regarding code status. Both videos were identical except for the ending: one ended with the physician asking for the patient’s code status preference and the other with the physician recommending DNR. Patients were randomly assigned to watch the videos in different sequences. The main outcome was the proportion of patients choosing DNR for the video patient. Results: 78 patients completed the study. 74% chose DNR after the question video, 73% after the recommendation video (p=NS). Median physician compassion score was very high and not different for both videos (p=0.73). 30/30 patients who had chosen DNR for themselves and 30/48 patients who had not chosen DNR for themselves chose DNR for the video patient (100% v/s 62%, p&lt;0.001). Age (OR=1.1/year, p=0.01) and white ethnicity (OR=9.43, p=0.004) predicted DNR choice for the video patient. Conclusions: Ending DNR discussions with a question or a recommendation did not impact DNR choice or perception of physician compassion. Therefore, both approaches are clinically appropriate. All patients who chose DNR for themselves and most patients who did not choose DNR for themselves chose DNR for the video patient. Age and race predicted DNR choice.

Clinical needs of patients referred early to supportive and palliative care.

9065 Background: Palliative care is evolving from delivering care to patients at the end-of-life to those earlier in the disease trajectory. We evaluated the differences in clinical characteristics, symptoms burdens, and service utilization between traditional palliative care patients (late referrals, LR) and the new patients group who are in earlier in their course of disease (early referrals, ER). Methods: We reviewed consecutive cancer patients referred to the Supportive Care Clinic with follow up visit within 30 days between August 2008 and October 2010. Patients were divided into two groups: ER (defined as patients with expected survival ≥ 2 years or receiving treatment for curative intent) and LR (all others). We compared clinical characteristics, symptoms and service utilization between both groups using chi-square test and Wilcoxon rank sum test. Results: 58% (695/1208) patients had a 2nd visit within 30 days. Among them, 100 patients were classified as ER (14.4%) and 100/595 LR were randomly selected as the comparison group. ER patients were younger (median age 54 years vs 60 years, p=0.009), more likely to have head and neck cancer (67% vs 6%, p&lt;0.001) and to have a shorter interval between cancer diagnosis and palliative care consultation (3.8 m vs 16.2 m, &lt;0.001). ER patients were also more likely to be CAGE positive (15% vs 4%, p=.014), referred from radiation oncology (49% vs 3%, p&lt;0.001), referred for treatment related side effects (70% vs 9%, p&lt;0.001), and receiving active anti-cancer treatment at the time of consultation (74% vs 48%, p=0.0002). Baseline symptoms (Edmonton Symptom Assessment Scale) were similar between ER and LR except for insomnia (1.8 vs 2, p=0.004). LR patients experienced greater improvement in the symptom distress score (-5.5 vs -3, p=0.007). Overall median number of medical visits was 24 for ER vs 10.5 for LR (p&lt;0.001); however, median visit per month was 4.3 for LR and 2.1 for ER (p&lt;0.001). Conclusions: ER was associated with different patient characteristics; patients have similar distress but different needs and outcomes as compared to traditional LR patients.

Challenges of Managing Advanced Cancer Patients through Phone Triaging at an Outpatient Supportive Care Clinic: A Case Series of Palliative Care Patients

Palliative care strives to improve the quality of life for patients and their families by impeccable assessment and management using an interdisciplinary approach. However, patients with cancer-related pain and other symptoms tend to be undertreated because of limited follow-up visits due to late referrals and logistics. Thus, patients who present to the outpatient Supportive Care Center at The University of Texas MD Anderson Cancer Center often experience severe physical and psychological symptoms. The two case reports presented highlight the challenges of managing distressed patients with advanced cancer in the outpatient setting. These descriptions focus on addressing patient needs over the phone to enhance the care patients receive at the Supportive Care Center. Future prospective studies are needed to measure the effectiveness of using phone triaging in conjunction with standard outpatient palliative care.

Achievement of personalized pain goal in cancer patients referred to a supportive care clinic at a comprehensive cancer center.

Cancer pain initiatives recommend using the personalized pain goal to tailor pain management. This study was conducted to examine the feasibility and stability of personalized pain goal, and how it compares to the clinical pain response criteria. Records of 465 consecutive cancer patients seen in consultation at the Supportive Care Clinic were reviewed. Pain relief was assessed as clinical response (≥30% or ≥2 point pain reduction) and personalized pain goal response (pain ≤ personalized pain goal). One hundred fifty-two (34%), 95 (21%), and 163 (37%) patients presented with mild (1-4), moderate (5-6), and severe (7-10) pain, respectively. Median age (59 years), males (52%), and advanced cancer status (84%) did not differ by pain category. Median personalized pain goal at initial clinic consult was 3 (interquartile range, 2-3), was similar across pain groups, and remained unchanged (P = .57) at follow-up (median, 14 days). Clinical response was higher among patients with severe pain (60%) as compared with moderate (40%) and mild pain (33%, P < .001). Personalized pain goal response was higher among patients with mild pain (63%) as compared with moderate (44%) and severe pain (27%, P < .001). By using personalized pain goal response as the gold standard for pain relief, the sensitivity of clinical response was highest (98%) among patients with severe pain, but it had low specificity (54%). In patients with mild pain, clinical response was most specific for pain relief (98%), but had low sensitivity (52%). Personalized pain goal is a simple patient-reported outcome for pain goals. The majority of patients were capable of stating their desired level for pain relief. The median personalized pain goal was 3, and it was highly stable at follow-up assessment.

Preliminary Report: Vitamin D Deficiency in Advanced Cancer Patients with Symptoms of Fatigue or Anorexia

Vitamin D deficiency in noncancer patients is associated with symptoms of fatigue, muscle weakness, and depression. These symptoms are common among advanced cancer patients. We investigated the prevalence of low serum vitamin D levels in cancer patients with fatigue or poor appetite and their association with symptom burden and other correctable endocrine abnormalities. This was a retrospective review of 100 consecutive cancer patients with appetite or fatigue scores of ≥4 of 10 referred to a supportive care clinic. We investigated serum levels of 25(OH) vitamin D, cortisol, thyroid-stimulating hormone, and bioavailable testosterone. Symptoms were measured by the Edmonton Symptom Assessment Scale. Serum 25(OH) vitamin D <20 ng/mL was considered deficient; ≥20 ng/mL and <30 ng/mL were considered insufficient. Patients were predominantly male (68%) and white (66%), with a median age of 60 years (range, 27-91 years). Gastrointestinal (30%) and lung (22%) cancers were predominant. Forty-seven patients (47%) were vitamin D deficient and 70 (70%) were insufficient. Thirteen of 70 patients (19%) with vitamin D insufficiency were on supplementation. Vitamin D deficiency was more common among nonwhites (82% versus 36%) and females. No significant association was found between vitamin D and symptoms. Hypogonadic males had a significantly lower mean 25(OH) vitamin D level than eugonadic males. Low vitamin D levels were highly prevalent among advanced cancer patients with cachexia or fatigue. Vitamin D deficiency was more frequent among nonwhite and female patients. Vitamin D levels were also significantly lower in male patients with hypogonadism.

4164 ORAL A Supportive Nursing Care Clinic for Patients With Head and Neck Cancer – Effects on Nutritional Status, Nutritional Treatment and Admissions to Hospital

A supportive nursing care clinic for patients with head and neck cancer : effects on nutritional status, nutritional treatment and admissions to hospital

Frequency and predictors of patient deviation from prescribed opioids and barriers to opioid pain-management in patients (pts) with advanced cancer.

9081 Background: 75-80% of pts with advanced cancer report pain and almost all receive opioids. There is limited knowledge on deviations from prescribed opioids (MD-Rx) and barriers to pain control. Poor adherence to opioids has been reported in 49-70% of pts. Methods: In this prospective study, we examined the frequency and predictors of pt-reported opioid deviation, and barriers to opioid use in 198 outpatients followed at the Supportive Care Clinic (SCC). We collected pain score (0-10), MD-Rx, confidential patient-reported prescription (Pt-Rx) and opioid intake (Pt-In), barriers to pain management (Barriers Questionnaire II, BQ-II), and adherence to medications (modified Morisky scale, MMS). We defined opioid deviation as <70% or >130% from prescribed dose. Results: Median age was 55, 91 (46%) females, 187 (94%) had solid tumors. The median pain was 4 (IQR 3-7). The median morphine equivalent daily dose was 120mg [Q1-Q3 45-270] for MD-Rx, 100mg [40-270] for Pt-Rx, and 100mg [30-250] for Pt-In. There was a high correlation between MD-Rx and Pt-Rx for regular [r = 0.9, p<0.001], and regular + breakthrough opioids intake [r = 0.94, p<0.001]. 19/198 pts (9.6%) deviated: 11 (6%) used lower doses than prescribed, and 8 (4%) used higher doses. Deviation was more frequent in males [p=0.04] and non-whites [p=0.027]. Non-white race was also associated with higher barriers to opioid use (mean BQ-II 2.16 for other races and 1.76 for African-Americans vs. 1.6 for whites, p=0.01). Higher barrier by BQ-II was in turn associated with lower adherence according to MMS (mean BQ-II 2.0 vs. 1.6, p=0.007 in the motivation domain; mean BQ-II 2.13 vs. 1.57, p=0.0001 in the knowledge domain). Higher BQ-II scores were also associated with taking less/no opioids due to cost (p=0.0017). Conclusions: A small percentage of interdisciplinary supportive care pts reported confidential opioid deviation and most commonly towards less dose intake. These findings suggest SCCs are more successful than other settings in opioid management. Non-white race and male gender were associated with higher risk of opioid deviation. Non-white race was also associated with higher barrier to opioid use.

Achievement of personalized pain goal (PPG) in cancer patients referred to a supportive care clinic.

9060 Background: Cancer pain quality improvement initiatives recommend tailoring pain treatments to patient’s (pt’s) individual goals for comfort/functioning. The pt’s PPG for pain relief (0-10 pain-intensity that pt considers to be comfortable, in physical, functional, and psychosocial domains) is a simple measure of such goals. Pain and PPG are routinely assessed at our institution. Study was conducted to determine the proportion of cancer pts treated for pain, who achieve a pain response, as determined by their PPG, and identify factors associated with its non-achievement. Methods: Records of consecutive cancer pts seen in the Supportive-Care Clinic in consultation, who had follow-up visits within 1-5 weeks (n=348) were reviewed. All pts were treated by palliative-medicine specialists. Pain-response was defined as (1) Clinical-response (30% or ≥2 point pain reduction), and (2) PPG-response (if pain ≤ PPG). Logistic-regression model was used to estimate the effect of various factors on PPG-response. Results: Males 51%, median age 58 years, and survival 188 (IQR 92-412) days. At baseline, 239 pts (69%) had pain ≥4. Median pain was 7 (IQR 5-8) and PPG 3 (2-3). At follow-up (median 14-days), pain significantly decreased (7 vs 5, p < 0.001), while PPG remained stable (p=0.9). Initial PPG remained the same (51%) or changed by <2-points (31%) in 195 pts (82%). 122 pts (51%) achieved a clinical-response, but only 66 (28%) achieved their PPG (Table). Using PPG as gold-standard, sensitivity of clinical-response was 89%, and specificity 65%. Baseline factors associated with non-achievement of PPG were higher Memorial Delirium Assessment Scale scores (odds-ratio 1.43 per point, p= 0.002), positive alcoholism screening (CAGE ≥ 2) (2.79, 0.04) and insomnia (1.13, 0.04). Conclusions: PPG is a simple pt reported outcome for pain goals. Our preliminary findings suggest that PPG is stable over time, and that traditional definitions of clinical-response have a large percentage of false-positives. Predictors of lack of PPG-response include delirium, alcoholism history, and insomnia. Clinical-response by PPG-response. Achieved PPG Yes No Total Clinical Response Yes 59 60 119 No 7 111 118 Total 66 171 237 2 pts missing

Factors Associated with the Severity of Sleep Disturbance in Advanced Cancer Patients Presenting to Supportive Care Clinic (771)

natural death (and) directives. Sixty-one percent were uncomfortable discussing a child’s serious illness or impending death; 74%were uncomfortable discussing terminal/end-of-life care; 62% were uncomfortable discussing and directives; 74% were uncomfortable discussing death with families from a variety of cultural backgrounds. One hundred percent agreed that PC skills are an important competency for pediatric residents. Seventy-two percent believe PC skills would be best taught using standardized patient scenarios and real life clinical situations. Conclusion. Pediatric residents have minimal training in PC skills and report a large degree of discomfort in various PC situations. They identify PC skills as a vital part of their training. Implications for research, policy, or practice. PC services are vastly underused in the pediatric setting. Further research is needed on the most effective means to increase pediatric resident comfort level in PC situations.

Pharmacist-Led, Interdisciplinary Model for Delivery of Supportive Care in the Ambulatory Cancer Clinic Setting

To describe a pharmacist-led, interdisciplinary method of care delivery begun at the University of North Carolina. We describe the characteristics of the population seen and the role of the individual members of the interdisciplinary team, and provide an early analysis of the program's impact on symptom improvement. A supportive care consultation service was begun at the University of North Carolina Hospitals to serve adult outpatients with cancer undergoing treatment or follow-up. Patients data were entered into an institutional review board-approved database to permit detailed assessments over time. Patient demographics were analyzed using descriptive statistics, medications used and changes made were noted, and symptom scores from a previously described instrument were captured and compared over time. Patients were seen from all adult oncology services, including gynecologic, radiation, medical, and surgical. The characteristics of the population seen were similar to those of the hospital population as a whole. Most of the patients were seen for pain management, and many required a medication change. Symptom scores improved by the second visit, and the improvement was maintained. We are able to demonstrate that the use of a pharmacist-led, interdisciplinary team produced an improvement in symptom scores comparable to what has been seen in the inpatient palliative care service within our institution. Projected shortages of oncology providers may be mitigated by pharmacists working in collaborative practices, with prescriptive authority, in the ambulatory oncology setting.

Impact of a Palliative Care Consultation Team on Cancer-Related Symptoms in Advanced Cancer Patients Referred to an Outpatient Supportive Care Clinic

Patients with advanced cancer may develop severe physical and psychosocial symptoms. There are limited data on the impact of an outpatient palliative consultation (PC) team on cancer-related symptoms. To study the impact of the PC on symptoms in patients with advanced cancer receiving outpatient palliative care. Four hundred six consecutive patients referred to a supportive care outpatient center (OPC) from January 2006 to June 2007 with complete Edmonton Symptom Assessment Scale (0-10 scale) at the initial and follow-up visits were reviewed. Patient characteristics, change of symptoms at follow-up visit, and response rate were analyzed. Using logistic regression models, the predictors of improvement of pain and fatigue were assessed. Median age was 59 years; 53% were female. Median interval between visits was 15 days. Mean scores at baseline and follow-up visits were fatigue 6.8 and 5.3 (P<0.0001), pain 5.3 and 4.1 (P<0.0001), depression 3.2 and 2.5 (P<0.0001), anxiety 3.7 and 2.8 (P<0.0001), dyspnea 2.7 and 2.5 (P=0.05), sleep 5 and 4 (P<0.0001), and well-being 5.2 and 4.4 (P<0.0001). Dyspnea (odds ratio and P-value, 0.90, 0.03), nausea (0.92, 0.06), and depression (0.91, 0.04) were associated with improvement in fatigue; drowsiness (1.10, 0.04), and feeling of well-being (0.87, 0.02) were associated with improvement in pain. The initial consult by PC achieved significant symptom improvement in patients receiving treatment in the OPC. Further prospective studies are needed.