Machine Learning to Allocate Palliative Care Consultations During Cancer Treatment.

PURPOSE: For patients with advanced cancer, early consultations with palliative care (PC) specialists reduce costs, improve quality of life, and prolong survival. However, capacity limitations prevent all patients from receiving PC shortly after diagnosis. We evaluated whether a prognostic machine learning system could promote early PC, given existing capacity. METHODS: Using population-level administrative data in Ontario, Canada, we assembled a cohort of patients with incurable cancer who received palliative-intent systemic therapy between July 1, 2014, and December 30, 2019. We developed a machine learning system that predicted death within 1 year of each treatment using demographics, cancer characteristics, treatments, symptoms, laboratory values, and history of acute care admissions. We trained the system in patients who started treatment before July 1, 2017, and evaluated the potential impact of the system on PC in subsequent patients. RESULTS: Among 560,210 treatments received by 54,628 patients, death occurred within 1 year of 45.2% of treatments. The machine learning system recommended the same number of PC consultations observed with usual care at the 60.0% 1-year risk of death, with a first-alarm positive predictive value of 69.7% and an outcome-level sensitivity of 74.9%. Compared with usual care, system-guided care could increase early PC by 8.5% overall (95% CI, 7.5 to 9.5; P < .001) and by 15.3% (95% CI, 13.9 to 16.6; P < .001) among patients who live 6 months beyond their first treatment, without requiring more PC consultations in total or substantially increasing PC among patients with a prognosis exceeding 2 years. CONCLUSION: Prognostic machine learning systems could increase early PC despite existing resource constraints. These results demonstrate an urgent need to deploy and evaluate prognostic systems in real-time clinical practice to increase access to early PC.

Medical oncologists' and palliative care physicians' opinions towards thromboprophylaxis for inpatients with advanced cancer: a cross-sectional study.

BACKGROUND: Patients with advanced cancer are increasingly discharged from inpatient settings following focused symptom management admissions. Thromboprophylaxis (TP) is recommended for patients with cancer admitted to acute care settings; less is known about TP use in palliative care (PC) settings. This study explored the opinions of Canadian medical oncologists (MO) and PC physicians regarding the use of TP for inpatients with advanced cancer. METHODS: A fractional factorial survey designed to evaluate the impact of patient factors (age, clinical setting, reason for admission, pre-admission performance status (Eastern Cooperative Oncology Group; ECOG), and risk of bleeding on anticoagulation) and physician demographics on recommending TP was administered by email to Canadian MO and PC physicians. Each respondent received eight vignettes randomly selected from a set of 32. Hierarchical regression was used to evaluate the odds of prescribing TP adjusted for patient factors. RESULTS: 606 MO and 491 PC physicians were surveyed; response rates were 11.1% and 15.0%, respectively. MO were predominantly male (59.7%); PC female (60.3%); most worked in academic environments (90.3% MO; 73.9% PC). Multivariable hierarchical logistic regression demonstrated that all patient factors except age were associated with prescribing TP (ORs range: from 1.34 (95% CI 1.01 to 1.77) for good ECOG, to 2.53 (95% CI 1.9 to 3.37), for reversible reason for admission). Controlling for these factors, medical specialty was independently associated with recommending TP (OR for MO 2.09 (95% CI 1.56 to 2.8)). CONCLUSIONS: MO have higher odds of recommending TP for inpatients with advanced cancer than PC physicians. Further research exploring the drivers of these differing practices is warranted.

Evaluation of Adolescents' and Young Adults' Attitudes Toward Participation in Cancer Clinical Trials.

PURPOSE: Participation in cancer clinical trials (CCTs) for adolescents and young adults (AYAs) remains the lowest of any patient group with cancer. Little is known about the personal barriers to AYA accrual. The aim of this study was to explore AYA attitudes that influence CCT participation. METHODS: A mixed-methods approach was used. AYAs and non-AYAs (≥ 40 years) completed the Cancer Treatment subscale of the Attitudes Toward Cancer Trials Scales and 9 supplementary questions formed from interview analysis. Differences between AYA and non-AYA cohorts were analyzed using the Mann-Whitney U test, and logistic regression models were constructed to evaluate the effect of demographics on perceptions of CCTs. RESULTS: Surveys were distributed to 61 AYAs (median age, 29 years; range, 17-39 years) and 74 non-AYAs (median age, 55 years; range, 40-88 years). Compared with non-AYAs, AYAs perceived CCTs to be unsafe/more difficult (Personal Barrier/Safety domain; P = .01). There were no differences based on age in other domains. AYAs were also more concerned with CCT interference in their long-term goals (P = .04). Multivariable ordered logistic regression identified increased personal barriers in the Personal Barrier/Safety domain for AYAs (P = .01), in patients with English as a second language (ESL; P < .01), and in patients previously not offered a clinical trial (P = .03). Long-term goals were identified as a barrier in particular tumor types (P = .01) and in patients with ESL (P < .01), with a trend identified in AYAs (P = .12). CONCLUSION: Age-related differences in attitudes toward CCTs suggest that tailored approaches to CCT accrual are warranted. Patient-centered delivery of information regarding CCTs, particularly in patients with ESL and who are trial naïve, may improve accrual.

Setting Quality Improvement Priorities for Women Receiving Systemic Therapy for Early-Stage Breast Cancer by Using Population-Level Administrative Data.

Purpose Routine evaluation of quality measures (QMs) can drive improvement in cancer systems by highlighting gaps in care. Targeting quality improvement at QMs that demonstrate substantial variation has the potential to make the largest impact at the population level. We developed an approach that uses both variation in performance and number of patients affected by the QM to set priorities for improving the quality of systemic therapy for women with early-stage breast cancer (EBC). Patients and Methods Patients with EBC diagnosed from 2006 to 2010 in Ontario, Canada, were identified in the Ontario Cancer Registry and linked deterministically to multiple health care databases. Individual QMs within a panel of 15 QMs previously developed to assess the quality of systemic therapy across four domains (access, treatment delivery, toxicity, and safety) were ranked on interinstitutional variation in performance (using interquartile range) and the number of patients who were affected; then the two rankings were averaged for a summative priority ranking. Results We identified 28,427 patients with EBC who were treated at 84 institutions. The use of computerized physician electronic order entry for chemotherapy, emergency room visits or hospitalizations during chemotherapy, and timely receipt of chemotherapy were identified as the QMs that had the largest potential to improve quality of care at a system level within this cohort. Conclusion A simple ranking system based on interinstitutional variation in performance and patient volume can be used to identify high-priority areas for quality improvement from a population perspective. This approach is generalizable to other health care systems that use QMs to drive improvement.

Establishing achievable benchmarks for quality improvement in systemic therapy for early-stage breast cancer.

BACKGROUND: Setting realistic targets for performance is a consistent challenge in quality improvement. In the current study, the authors used administrative data to define achievable targets for a panel of 15 previously developed quality indicators (QIs) focusing on systemic therapy in patients with early-stage breast cancer. METHODS: Deterministically linked administrative databases were used to identify patients with TNM stage I to stage III breast cancer who were diagnosed between 2006 and 2010 in Ontario, Canada. For each individual indicator, data-driven empirical benchmarks were calculated using the pared-mean benchmark approach. Variation in institution-level performance for each indicator was examined through the construction of funnel plots. RESULTS: A total of 28,303 patients with early-stage breast cancer were identified, 43% of whom received adjuvant chemotherapy. For the 9 QIs for which receiving the service or outcome was desirable (ie, consultation with a medical oncologist), the benchmark varied from 40.9% to 100%. For the 6 indicators for which not receiving the service or outcome was desirable (ie, incidence of febrile neutropenia), the benchmark varied from 0% to 49.0%. There was substantial variation noted with regard to the number of institutions meeting the target and the amount of interinstitution variation between the QIs. Top performing institutions varied by indicator, with no individual institution meeting the benchmark for all indicators. For the majority of indicators, institution size was not found to be correlated with performance. CONCLUSIONS: Data-derived benchmarking can be used to facilitate quality improvement by identifying areas of both good as well as suboptimal performance while defining an achievable target for which to strive. Cancer 2017;123:3772-3780. © 2017 American Cancer Society.

Time to Viremia for Patients Taking their First Antiretroviral Regimen and the Subsequent Resistance Profiles.

BACKGROUND: The resistance profiles for patients on first-line antiretroviral therapy (ART) regimens after viremia have not been well studied in community clinic settings in the modern treatment era. OBJECTIVE: To determine time to viremia and the ART resistance profiles of viremic patients. METHODS: HIV-positive patients aged ≥16 years initiating a three-drug regimen were retrospectively identified from 01/01/06 to 12/31/12. The regimens were a backbone of two nucleoside reverse transcriptase inhibitors (NRTIs) and a third agent: a protease inhibitor (PI), non-nucleoside reverse transcriptase inhibitor (NNRTI), or an integrase inhibitor (II). Time to viremia was compared using a proportional hazards model, adjusting for demographic and clinical factors. Resistance profiles were described in those with baseline and follow-up genotypes. RESULTS: For 653 patients, distribution of third-agent use and viremia was: 244 (37%) on PIs with 80 viremia, 364 (56%) on NNRTIs with 84 viremia, and 45 (7%) on II with 11 viremia. Only for NNRTIs, time to viremia was longer than PIs (p = 0.04) for patients with a CD4 count ≥200 cells/mm(3). Of the 175 with viremia, 143 (82%) had baseline and 37 (21%) had follow-up genotype. Upon viremia, emerging ART resistance was rare. One new NNRTI (Y181C) mutation was identified and three patients taking PI-based regimens developed NRTI mutations (M184 V, M184I, and T215Y). CONCLUSIONS: Time to viremia for NNRTIs was longer than PIs. With viremia, ART resistance rarely developed without PI or II mutations, but with a few NRTI mutations in those taking PI-based regimens, and NNRTI mutations in those taking NNRTI-based regimens.

Use and impact of intermittent versus continuous systemic treatment strategies in metastatic colorectal cancer in routine practice.

BACKGROUND: Randomized trials have shown that intermittent treatment may reduce toxicity without compromising survival in patients with metastatic colorectal cancer (mCRC). A population-based study examined patterns of use of chemotherapy-free intervals (CFIs) in routine practice in Ontario and their impact on survival and toxicity. METHODS: Patients treated with first-line intravenous chemotherapy for mCRC in Ontario between 2007 and 2009 were identified from administrative data. A CFI was defined as more than 56 days between 2 chemotherapy doses. A propensity score analysis was used to compare the survival of patients with CFIs and patients without CFIs, stratified by the type of first-line treatment: irinotecan (IRI), irinotecan plus bevacizumab (IRI-B), and oxaliplatin (OX). Toxicity was estimated on the basis of the rate of emergency room visits and hospitalizations. RESULTS: There were 1989 patients who started first-line chemotherapy for mCRC in Ontario between 2007 and 2009, and 489 (25%) had at least 1 CFI. The median time to the first CFI was 155 days (interquartile range, 82-217 days). There was no difference in survival for the propensity score-matched patients with or without CFIs in the IRI (hazard ratio [HR], 0.93; P = .70) and OX groups (HR, 0.73; P = .06). Survival was worse in the CFI group for patients treated with IRI-B (HR, 1.28; P = .03). Toxicity was lower for patients with at least 1 CFI (0.17 vs 0.25 acute visits per person-month of treatment, P = .007), although the magnitude varied with the treatment type. CONCLUSIONS: Intermittent treatment strategies are being used in routine practice for patients with mCRC. The impact on survival and toxicity varies with the type of first-line chemotherapy.

Palliative care and the aggressiveness of end-of-life care in patients with advanced pancreatic cancer.

BACKGROUND: We examined the impact of palliative care (PC) on aggressiveness of end-of-life care for patients with advanced pancreatic cancer. Measures of aggressive care included chemotherapy within 14 days of death; and at least one intensive care unit (ICU) admission, more than one emergency department (ED) visit, and more than one hospitalization, all within 30 days of death. METHODS: A retrospective population-based cohort study using administrative data was conducted in patients with advanced pancreatic cancer from 2005 to 2010 in Ontario, Canada. Multivariable logistic regression was performed with the above measures of aggressive care as the outcomes of interest and PC as the main exposure, adjusting for covariables. Secondary analyses examined intensity of PC as the main exposure defined in two ways: 1) absolute number of PC visits before the outcome of interest (0, 1, 2, 3+ visits) and 2) monthly rate of PC visits. RESULTS: The cohort included 5381 patients (median survival 75 days); 2816 (52.3%) had received a PC consultation. PC consultation was associated with decreased use of chemotherapy near death (odds ratio [OR] = 0.34, 95% confidence interval [CI] = 0.25 to 0.46); lower risk of ICU admission: OR = 0.12, 95% CI = 0.08 to 0.18; multiple ED visits: OR = 0.19, 95% CI = 0.16 to 0.23; multiple hospitalizations near death: OR = 0.24, 95% CI = 0.19 to 0.31). A per-unit increase in the monthly rate of PC visits was associated with lower odds of aggressive care for all four outcomes. CONCLUSION: PC consultation and a higher intensity of PC were associated with less aggressive care near death in patients with advanced pancreatic cancer.

Invasive procedures in the elderly after stage IV cancer diagnosis.

BACKGROUND: Invasive procedures are resource intense and may be associated with substantial morbidity. These harms must be carefully balanced with the benefits gained in life expectancy and quality of life. Prior research has demonstrated an increasing aggressiveness of care in cancer patients at the end-of-life. To better characterize surgical care in this setting, we sought to examine trends in the use of invasive procedures in patients diagnosed with metastatic cancer on presentation. MATERIALS AND METHODS: Using Surveillance Epidemiology and End Results -Medicare data, we identified invasive procedure claims from 1994-2009 for patients diagnosed with incident stage IV breast, colorectal, lung, and prostate cancer patients in 1995-2006. We grouped procedures into surgically relevant categories, using an adaptation of the Clinical Classifications Software, and measured utilization and relative changes over time. RESULTS: Of stage IV patients diagnosed in 2002-2006, 96% underwent a procedure during the course of their cancer care including 63% after the diagnostic period, and 25% in the last month of life. Between 1996 and 2006, minimal change was observed in utilization during the diagnostic period (+1.5%). However, there were significant increases during continuing care (+20.7%) and the last month of life (+21.5%). Procedures consistent with primary tumor resection decreased, whereas those with probable palliative intent and those unrelated to cancer increased. CONCLUSIONS: Nearly all patients who present with metastatic cancer undergo invasive procedures. Although overall utilization is increasing, the specific procedure types indicate that it may be appropriate, enhancing the quality of life in this vulnerable population.

Time to virologic failure for patients taking their first antiretroviral regimen and the subsequent resistance profiles.

INTRODUCTION: The resistance profiles of first-line antiretroviral therapy (ART) regimens after virologic failure have yet to be studied in a clinic setting in the modern treatment era. Time to virologic failure among three standard first-line regimens and the resistance profiles of these failures were compared. MATERIALS AND METHODS: All HIV-positive persons aged 16 and over starting a three-drug first-line ART regimen were retrospectively identified at a Toronto community clinic (1 January 2006-1 January 2013). The regimens included a backbone of two NRTIs and a third agent; a PI, an NNRTI, or an II. Patients must have been on treatment for at least 14 days and have at least one VL test within 6 months after starting treatment. The primary outcome was virologic failure defined as either: no suppression by 6 months, or after suppression, two consecutive, detectable VL200 copies/mL at least 14 days apart or one VL>200 copies/mL. Time to failure was compared using a proportional hazards model adjusting for demographic and clinical factors. Resistance profiles of NRTIs and third agents are described in patients with virologic failure who had both baseline and virologic failure genotypes. RESULTS: Six hundred sixty patients (93% male) were included with a mean age of 38.9 and a median follow-up period of 35.3 (32.2-39.3) months. Distribution of third agent use was: PI 37.3% (n=246), NNRTI 55.9% (n=369) and II 6.8% (n=45). Virologic failures occurred in 81/246 (33%) with PI, 87/369 (24%) with NNRTI and 11/45 (24%) with II. Compare to PIs, time to failure was longer with NNRTIs (p=0.0013) and similar for IIs (p=0.1562). No evidence that failure with NNRTIs was different from IIs (p=0.9139). Of the 660 patients, 567 (86%) had a baseline genotype. Of the 567 patients, 179 had virological failure. Of the 179, 145(81%) had a baseline genotype and only 37 (21%) had both a baseline and follow-up genotype. Upon failure, emerging ART resistance was rare. No new PI or II mutations were identified and one new NNRTI (Y181C) mutation was identified. Three patients taking PI-based regimens developed NRTI mutations (M184V, M184I, T215Y). CONCLUSIONS: Time to virologic failure was significantly greater in the NNRTI group compared to the PI group. If failure did occur, ART resistance rarely developed with no PI mutations but a few NRTI mutations in those taking PI-based regimens, and NNRTI mutations in those taking NNRTI-based regimen.

Interinstitutional variation in management decisions for treatment of 4 common types of cancer: A multi-institutional cohort study.

BACKGROUND: When clinical practice is governed by evidence-based guidelines and there is consensus about their validity, practice variation should be minimal. For areas in which evidence gaps exist, greater variation is expected. OBJECTIVE: To systematically assess interinstitutional variation in management decisions for 4 common types of cancer. DESIGN: Multi-institutional, observational cohort study of patients with cancer diagnosed between July 2006 through May 2011 and observed through 31 December 2011. SETTING: 18 cancer centers participating in the formulation of treatment guidelines and systematic outcomes assessment through the National Comprehensive Cancer Network. PATIENTS: 25 589 patients with incident breast cancer, colorectal cancer, lung cancer, or non-Hodgkin lymphoma. MEASUREMENTS: Interinstitutional variation for 171 binary management decisions with varying levels of supporting evidence. For each decision, variation was characterized by the median absolute deviation of the center-specific proportions. RESULTS: Interinstitutional variation was high (median absolute deviation >10%) for 35 of 171 (20%) oncology management decisions, including 9 of 22 (41%) decisions for non-Hodgkin lymphoma, 16 of 76 (21%) for breast cancer, 7 of 47 (15%) for lung cancer, and 3 of 26 (12%) for colorectal cancer. Forty-six percent of high-variance decisions involved imaging or diagnostic procedures and 37% involved choice of chemotherapy regimen. The evidence grade underpinning the 35 high-variance decisions was category 1 for 0%, 2A for 49%, and 2B/other for 51%. LIMITATION: Physician identifiers were unavailable, and results may not generalize outside of major cancer centers. CONCLUSION: The substantial variation in institutional practice manifest among cancer centers reveals a lack of consensus about optimal management for common clinical scenarios. For clinicians, awareness of management decisions with high variation should prompt attention to patient preferences. For health systems, high variation can be used to prioritize comparative effectiveness research, patient-provider education, or pathway development. PRIMARY FUNDING SOURCE: National Cancer Institute and National Comprehensive Cancer Network.

Validating billing/encounter codes as indicators of lung, colorectal, breast, and prostate cancer recurrence using 2 large contemporary cohorts.

BACKGROUND: A substantial proportion of cancer-related mortality is attributable to recurrent, not de novo metastatic disease, yet we know relatively little about these patients. To fill this gap, investigators often use administrative codes for secondary malignant neoplasm or chemotherapy to identify recurrent cases in population-based datasets. However, these algorithms have not been validated in large, contemporary, routine care cohorts. OBJECTIVE: To evaluate the validity of secondary malignant neoplasm and chemotherapy codes as indicators of recurrence after definitive local therapy for stage I-III lung, colorectal, breast, and prostate cancer. RESEARCH DESIGN, SUBJECTS, AND MEASURES: We assessed the sensitivity, specificity, and positive predictive value (PPV) of these codes 14 and 60 months after diagnosis using 2 administrative datasets linked with gold-standard recurrence status information: CanCORS/Medicare (diagnoses 2003-2005) and HMO/Cancer Research Network (diagnoses 2000-2005). RESULTS: We identified 929 CanCORS/Medicare patients and 5298 HMO/CRN patients. Sensitivity, specificity, and PPV ranged widely depending on which codes were included and the type of cancer. For patients with lung, colorectal, and breast cancer, the combination of secondary malignant neoplasm and chemotherapy codes was the most sensitive (75%-85%); no code-set was highly sensitive and highly specific. For prostate cancer, no code-set offered even moderate sensitivity (≤ 19%). CONCLUSIONS: Secondary malignant neoplasm and chemotherapy codes could not identify recurrent cancer without some risk of misclassification. Findings based on existing algorithms should be interpreted with caution. More work is needed to develop a valid algorithm that can be used to characterize outcomes and define patient cohorts for comparative effectiveness research studies.

Impact of oncology-related direct-to-consumer advertising: association with appropriate and inappropriate prescriptions.

BACKGROUND: Little is known about the impact of direct-to-consumer advertising (DTCA) on appropriate versus inappropriate prescribing. Aromatase inhibitor (AI) therapy for breast cancer provides an ideal paradigm for studying this issue, because AIs have been the focus of substantial DTCA, and because they should only be used in postmenopausal women, age can serve as a simple surrogate marker of appropriateness. METHODS: Data regarding national DTCA spending for the AIs were obtained from TNS Multimedia; hormonal therapy prescription data were obtained from IMS Health. Time series analyses were performed to characterize the association between monthly changes in DTCA spending for the AIs and monthly changes in the proportion of all new hormonal therapy prescriptions represented by the AIs from October 2005 to September 2007. Analyses were stratified by age, considering prescriptions for women ≤ 40 (likely premenopausal) to be inappropriate and those for women > 60 (likely postmenopausal) to be appropriate. RESULTS: Monthly dollars spent on AI-associated DTCA varied considerably ($118,600 to $22,019,660). Time series analysis revealed that for every million dollars spent on DTCA for the AIs, there was an associated increase 3 months later in the new AI prescription proportion of 0.15% for all ages (P < .0001) and 0.18% for those > 60 years (P < .0001), but no significant change for those ≤ 40 at any time from 0 to 6 months. CONCLUSIONS: DTCA for the AIs was associated with increases in appropriate prescriptions with no significant effect on inappropriate prescriptions, suggesting that DTCA may not foster inappropriate medication use for certain drug classes.

Use of stereotactic radiosurgery for brain metastases from non-small cell lung cancer in the United States.

PURPOSE: The indications for treatment of brain metastases from non-small cell lung cancer (NSCLC) with stereotactic radiosurgery (SRS) remain controversial. We studied patterns, predictors, and cost of SRS use in elderly patients with NSCLC. METHODS AND MATERIALS: Using the Surveillance, Epidemiology, and End Results-Medicare (SEER-Medicare) database, we identified patients with NSCLC who were diagnosed with brain metastases between 2000 and 2007. Our cohort included patients treated with radiation therapy and not surgical resection as initial treatment for brain metastases. RESULTS: We identified 7684 patients treated with radiation therapy within 2 months after brain metastases diagnosis, of whom 469 (6.1%) cases had billing codes for SRS. Annual SRS use increased from 3.0% in 2000 to 8.2% in 2005 and varied from 3.4% to 12.5% by specific SEER registry site. After controlling for clinical and sociodemographic characteristics, we found SRS use was significantly associated with increasing year of diagnosis, specific SEER registry, higher socioeconomic status, admission to a teaching hospital, no history of participation in low-income state buy-in programs (a proxy for Medicaid eligibility), no extracranial metastases, and longer intervals from NSCLC diagnosis. The average cost per patient associated with radiation therapy was 2.19 times greater for those who received SRS than for those who did not. CONCLUSIONS: The use of SRS in patients with metastatic NSCLC increased almost 3-fold from 2000 to 2005. In addition, we found significant variations in SRS use across SEER registries and socioeconomic quartiles. National practice patterns in this study suggested both a lack of consensus and an overall limited use of the approach among elderly patients before 2008.

Associations between end-of-life discussion characteristics and care received near death: a prospective cohort study.

PURPOSE: National guidelines recommend that discussions about end-of-life (EOL) care planning happen early for patients with incurable cancer. We do not know whether earlier EOL discussions lead to less aggressive care near death. We sought to evaluate the extent to which EOL discussion characteristics, such as timing, involved providers, and location, are associated with the aggressiveness of care received near death. PATIENTS AND METHODS: We studied 1,231 patients with stage IV lung or colorectal cancer in the Cancer Care Outcomes Research and Surveillance Consortium, a population- and health system-based prospective cohort study, who died during the 15-month study period but survived at least 1 month. Our main outcome measure was the aggressiveness of EOL care received. RESULTS: Nearly half of patients received at least one marker of aggressive EOL care, including chemotherapy in the last 14 days of life (16%), intensive care unit care in the last 30 days of life (9%), and acute hospital-based care in the last 30 days of life (40%). Patients who had EOL discussions with their physicians before the last 30 days of life were less likely to receive aggressive measures at EOL, including chemotherapy (P = .003), acute care (P < .001), or any aggressive care (P < .001). Such patients were also more likely to receive hospice care (P < .001) and to have hospice initiated earlier (P < .001). CONCLUSION: Early EOL discussions are prospectively associated with less aggressive care and greater use of hospice at EOL.

High-cost imaging in elderly patients with stage IV cancer.

BACKGROUND: Medicare expenditures for high-cost diagnostic imaging have risen faster than those for total cancer care and have been targeted for potential cost reduction. We sought to determine recent and long-term patterns in high-cost diagnostic imaging use among elderly (aged ≥65 years) patients with stage IV cancer. METHODS: We identified claims within the Surveillance, Epidemiology, and End Results (SEER)-Medicare database with computed tomography, magnetic resonance imaging, positron emission tomography, and nuclear medicine scans between January 1994 and December 2009 for patients diagnosed with stage IV breast, colorectal, lung, or prostate cancer between January 1995 and December 2006 (N = 100,594 patients). The proportion of these patients imaged and rate of imaging per-patient per-month of survival were calculated for each phase of care in patients diagnosed between January 2002 and December 2006 (N = 55,253 patients). Logistic regression was used to estimate trends in imaging use in stage IV patients diagnosed between January 1995 and December 2006, which were compared with trends in imaging use in early-stage (stages I and II) patients with the same tumor types during the same period (N = 192,429 patients). RESULTS: Among the stage IV patients diagnosed between January 2002 and December 2006, 95.9% underwent a high-cost diagnostic imaging procedure, with a mean number of 9.79 (SD = 9.77) scans per patient and 1.38 (SD = 1.24) scans per-patient per-month of survival. After the diagnostic phase, 75.3% were scanned again; 34.3% of patients were scanned in the last month of life. Between January 1995 and December 2006, the proportion of stage IV cancer patients imaged increased (relative increase = 4.6%, 95% confidence interval [CI] = 3.7% to 5.6%), and the proportion of early-stage cancer patients imaged decreased (relative decrease = -2.5%, 95% CI = -3.2% to -1.9%). CONCLUSIONS: Diagnostic imaging is used frequently in patients with stage IV disease, and its use increased more rapidly over the decade of study than that in patients with early-stage disease.

End-of-life care discussions among patients with advanced cancer: a cohort study.

BACKGROUND: National guidelines recommend that physicians discuss end-of-life (EOL) care planning with patients with cancer whose life expectancy is less than 1 year. OBJECTIVE: To evaluate the incidence of EOL care discussions for patients with stage IV lung or colorectal cancer and where, when, and with whom these discussions take place. DESIGN: Prospective cohort study of patients diagnosed with lung or colorectal cancer from 2003 to 2005. SETTING: Participants lived in Northern California, Los Angeles County, North Carolina, Iowa, or Alabama or received care in 1 of 5 large HMOs or 1 of 15 Veterans Health Administration sites. PATIENTS: 2155 patients with stage IV lung or colorectal cancer. MEASUREMENTS: End-of-life care discussions reported in patient and surrogate interviews or documented in medical records through 15 months after diagnosis. RESULTS: 73% of patients had EOL care discussions identified by at least 1 source. Among the 1470 patients who died during follow-up, 87% had EOL care discussions, compared with 41% of the 685 patients who were alive at the end of follow-up. Of the 1081 first EOL care discussions documented in records, 55% occurred in the hospital. Oncologists documented EOL care discussions with only 27% of their patients. Among 959 patients with documented EOL care discussions who died during follow-up, discussions took place a median of 33 days before death. LIMITATIONS: The depth and quality of EOL care discussions was not evaluated. Much of the information about discussions came from surrogates of patients who died before baseline interviews could be obtained. CONCLUSION: Although most patients with stage IV lung or colorectal cancer discuss EOL care planning with physicians before death, many discussions occur during acute hospital care, with providers other than oncologists, and late in the course of illness. PRIMARY FUNDING SOURCE: National Cancer Institute and Department of Veterans Affairs.

Pathological predictors for site of local recurrence after radiotherapy for prostate cancer.

PURPOSE: Rational design of targeted radiotherapy (RT) in prostate cancer (Pca) hinges on a better understanding of spatial patterns of recurrence. We sought to identify pathological factors predictive for site of local recurrence (LR) after external beam RT. METHODS AND MATERIALS: Prospective databases were reviewed to identify men with LR after RT from 1997 through 2009. Patients with biochemical failure and biopsy-confirmed Pca more than 2 years after RT were evaluated. Prediction for site of recurrence based on the following pretreatment factors was determined on independent and cluster-sextant basis: presence of malignancy, dominant vs. nondominant percentage core length (PCL) involvement, PCL ≥ or <40%, and Gleason score. Sites of dominant PCL were defined as sextants with peak PCL involvement minus 10%, and >5% for each patient. RESULTS: Forty-one patients with low-intermediate risk Pca constituted the study cohort. Median time to biopsy after RT was 51 months (range, 24-145). Of 246 sextants, 74 were involved with tumor at baseline. When sextants are treated as independent observations the presence of malignancy (77% vs. 22%, p = 0.0001), dominant PCL (90% vs. 46%, p = 0.0001), and PCL ≥40% (89% vs. 68 %, p = 0.04) were found to be significant predictors for LR, although PCL ≥40% did not retain statistical significance if sextants were considered correlated. The vast majority of patients (95%) recurred at the original site of dominant PCL or PCL ≥40%, and 44% also recurred in regions of nondominant PCL <40% (n = 8) and/or benign sampling (n = 14) at baseline. CONCLUSIONS: LR after RT predominantly occurs in regions bearing higher histological tumor burden but are not isolated to these sites. Our data highlights the value of spatially resolved baseline pathological sampling and may assist in the design of clinical trials tailoring RT dose prescriptions to subregions of the prostate gland.

Accelerated partial breast irradiation using brachytherapy for breast cancer: patterns in utilization and guideline concordance.

BACKGROUND: Accelerated partial breast irradiation using brachytherapy (APBIb) is an alternative to whole-breast irradiation (WBI) after breast-conserving surgery. We evaluated patterns of APBIb use with respect to 2009 American Society for Radiation Oncology consensus guidelines (ASTRO-G) in a population-based cohort. METHODS: From Surveillance, Epidemiology, and End Results data, we identified 138 815 American women with breast cancer diagnosed between January 1, 2000, and December 31, 2007, who underwent WBI or APBIb after breast-conserving surgery and classified them as suitable, cautionary, or unsuitable for APBIb according to ASTRO-G criteria. Logistic regression was applied to study APBIb use overall and within each guideline category. All P values are from two-sided tests. RESULTS: Overall, 2.6% of patients received APBIb, and 65.8% of them were classified as cautionary or unsuitable. APBIb was used by 5% of suitable, 3.4% of cautionary, and 1.6% of unsuitable patients by ASTRO-G criteria (P < .001). APBIb use increased from 0.4% in 2000 to 6.6% in 2007 and varied widely (0%-7%) between localities. Variables associated with APBIb use among suitable patients included other vs white race (odds ratio [OR] = 0.51, P < .001), region (OR = 2.60-8.62, P < .001), and more recent year (OR = 20.3, P < .001). Among cautionary patients, variables associated with APBIb use included black vs white race (OR = 0.76, P = .027), other vs white race (OR = 0.57, P < .001), Hispanic ethnicity (OR = 0.75, P = .036), region (OR = 3.10-10.2, P < .001), nonmetropolitan or rural location (OR = 0.53, P = .012), and more recent year (OR = 17.6, P < .001). Among unsuitable patients, black vs white race (OR = 0.77, P = .008), other vs white race (OR = 0.46, P < .001), region (OR = 3.33-21.6, P < .001), and more recent year (OR = 12.7, P < .001) were associated with APBIb use. CONCLUSIONS: APBIb after breast-conserving surgery has been rapidly adopted in the United States. Use varied by race, ethnicity, and widely by region, especially among patients who may not be suitable for this radiation technique.

Effect of bodychecking on rate of injuries among minor hockey players.

BACKGROUND: Bodychecking is a leading cause of injury among minor hockey players. Its value has been the subject of heated debate since Hockey Canada introduced bodychecking for competitive players as young as 9 years in the 1998/1999 season. Our goal was to determine whether lowering the legal age of bodychecking from 11 to 9 years affected the numbers of all hockey-related injuries and of those specifically related to bodychecking among minor hockey players in Ontario. METHODS: In this retrospective study, we evaluated data collected through the Canadian Hospitals Injury Reporting and Prevention Program. The study's participants were male hockey league players aged 6-17 years who visited the emergency departments of 5 hospitals in Ontario for hockey-related injuries during 10 hockey seasons (September 1994 to May 2004). Injuries were classified as bodychecking-related or non-bodychecking-related. Injuries that occurred after the rule change took effect were compared with those that occurred before the rule's introduction. RESULTS: During the study period, a total of 8552 hockey-related injuries were reported, 4460 (52.2%) of which were attributable to bodychecking. The odds ratio (OR) of a visit to the emergency department because of a bodychecking-related injury increased after the rule change (OR 1.26, 95% confidence interval [CI] 1.16-1.38), the head and neck (OR 1.52, 95% CI 1.26-1.84) and the shoulder and arm (OR 1.18, 95% CI 1.04-1.35) being the body parts with the most substantial increases in injury rate. The OR of an emergency visit because of concussion increased significantly in the Atom division after the rule change, which allowed bodychecking in the Atom division. After the rule change, the odds of a bodychecking-related injury was significantly higher in the Atom division (OR 2.20, 95% CI 1.70-2.84). INTERPRETATION: In this study, the odds of injury increased with decreasing age of exposure to bodychecking. These findings add to the growing evidence that bodychecking holds greater risk than benefit for youth and support widespread calls to ban this practice.