National Cancer Institute Designated Cancer Center Research Articles

Differences in Patients, Surgical Complexity, and Outcomes After Cancer Surgery at National Cancer Institute-designated Cancer Centers Compared to Other Hospitals

Interest in comparing hospital surgical quality continues to increase, particularly with respect to examining certain hospital designations such as National Cancer Institute-designated Cancer Centers (NCI-CC). Our objectives were to compare patients, surgical complexity, and risk-adjusted 30-day outcomes following major cancer surgery at NCI-CC versus non-NCI centers. From the American College of Surgeons National Surgical Quality Improvement Program, patients were identified who underwent colorectal, pancreatic, or esophagogastric resection for cancer (2007-2011). Regression methods were used to evaluate characteristics associated with undergoing treatment at NCI-CCs and surgical-complexity-adjusted 30-day morbidity, mortality, and prolonged length-of-stay at NCI-CC versus non-NCI centers. NCI-CCs performed 20.2% of colorectal (10,555/52,265), 53.5% of pancreatic (6335/11,838), and 49.8% of esophagogastric (1596/3208) operations for cancer. NCI-CCs were more likely to treat patients who were younger, white, and with fewer comorbidities, but were more likely to perform more complex procedures including synchronous liver resection (eg, colorectal), adjacent organ resections (rectal cancer), and vascular reconstructions (eg, pancreas) (all P<0.05). NCI-CCs had a lower mortality rate for colorectal surgery only (1.2% vs. 1.9%) and increased rates of superficial surgical site infection (SSI) for colorectal (9.8% vs. 7.1%) and pancreatic (10.7% vs. 8.8%) surgery. No differences existed for the remaining complications by NCI-CC designation status. NCI-CCs were distributed throughout hospital quality rankings for all procedures and complications assessed. NCI-CCs treated younger, healthier patients, but performed more complex procedures. Patients treated at NCI-CCs had a lower risk of mortality for colorectal resection, but morbidity was similar to non-NCI centers. Comparison of cancer surgery hospital quality is feasible and should adjust for differences in patient demographics, comorbidities, and surgical complexity.

Can we measure the quality of complex cancer surgery? Comparison of patients and outcomes at National Cancer Institute Designated Cancer Centers (NCI-CC) vs. Non-NCI centers

Can we measure the quality of complex cancer surgery? Comparison of patients and outcomes at National Cancer Institute Designated Cancer Centers (NCI-CC) vs. Non-NCI centers

Abstract B79: Disparities in access to the EGFR assay

Abstract Introduction: This study examines regional and site of care differences in access to lung tumor genotyping. We test the hypothesis that utilization rate by community hospitals of the epidermal growth factor receptor (EGFR) assay is a function of presence and number of National Cancer Institute (NCI) designated cancer centers within the state. Additional hypotheses to be tested are: Community hospital participation in NCI clinical research groups increases the likelihood and rate of utilization of the EGFR assay. Counties with higher percentage of Blacks, lower education and income levels are likely to have lower rates of EGFR assay utilization. Description of the research: Genzyme Genetics holds the license to distribute and market the EGFR assay. It provided the authors a dataset which contained hospital name, city, state, zip code, patient gender, and payor for EGFR assays processed in fiscal year 2010. Genzyme has not enforced its license with cancer centers that have CLIA approved labs which independently process EGFR assays. Therefore, this data set represents community hospital utilization of the EGFR assay. The Genzyme data was merged with publicly available information detailing county/state level lung cancer incidence, location of cancer centers (obtained from the Center for Disease Control (CDC) and NCI), population demographic, educational and income statistics published by the Census bureau, and longitude and latitude data for each institution ordering the EGFR assay and for NCI designated cancer centers. Multiple regression analysis was conducted at the state level to analyze the relationship between the community hospital utilization rate of the EGFR assays (defined as the number of EGFR assays ordered by community hospitals divided by the lung cancer incidence rate in that state) the number of NCI cancer centers, race, education, and income levels within that state. A similar analysis was conducted with number of cancer centers replaced by presence of cancer center. Preliminary results aggregated at the state level were obtained for this abstract. Analysis is ongoing and results presented at the conference will be detailed at the zip code level with location and distance to NCI cancer center calculated and mapped. Further, NCI has provided an unencrypted version of its provider of service (POS) file which includes variables about community hospital participation in NCI sponsored clinical research groups. Relationship between EGFR utilization rate and community hospital participation in clinical research will also be analyzed. Summary of results: The presence and number of NCI cancer centers were both highly significant with p values of .023 and .000. When the cancer center variables were included in the models, other variables were insignificant. However, when cancer center variables were removed from the models, income over $75,000 trended toward significance with a p value of .08. Conclusion: Preliminary analysis illustrates that important regional and site of care differences exist in utilization of the EGFR assay at community hospitals. As molecular diagnostic technologies become an increasingly important component of lung cancer care, oncology biomarker and drug development, there is growing urgency to elucidate structural barriers impeding access to advanced cancer technologies and limiting diversity of tumor tissue included in translational research. Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):B79.

Wide variation in content of inpatient do-not-resuscitate order forms used at National Cancer Institute-designated cancer centers in the United States

Determining resuscitation status (RS) for inpatients with advanced cancer is emotionally charged and often conflictual. Available data suggest that clinicians have inconsistent practices when establishing and documenting do-not-resuscitate (DNR) orders. Lack of standardization may contribute to ineffective and unclear discussions regarding RS. To inform revisions of DNR order forms used at one comprehensive cancer center, we surveyed National Cancer Institute-designated cancer centers (NCICCs) to determine if a standardized approach to documenting inpatient DNR orders exists. Over a 12-week period in 2002-2003, the 50 NCICCs providing inpatient care were contacted regarding participation in this Institutional Review Board-approved study. Using a 69-item database, inpatient DNR order forms were analyzed for content and elements of process used to establish and document RS. Each form was evaluated by two raters to assess inter-rater reliability. Analysis was descriptive; inter-rater agreement was summarized using the kappa statistic. Sixty percent (30 out of 50) of NCICCs participated. Eighty percent of responding sites confined the order process exclusively to cardiopulmonary resuscitation and did not include other interventions for possible limitation. Two thirds of responding sites used preformatted order forms specific for establishing inpatient RS; forms varied widely in content and elements of process used to establish and document DNR orders. NCICCs do not have a standardized approach to establishing and documenting DNR orders. Lack of standardization may reflect the absence of a common understanding of these difficult issues which may contribute to unclear and ineffective communication when addressing RS.

A comparative analysis of inpatient Do Not Resuscitate (DNR) forms used at National Cancer Institute-designated cancer centers (NCICC)

8222 Background: The decision-making process establishing inpatient DNR status is difficult and may lead to conflict among the patient, family and clinician. In this study, we examined DNR forms from NCICC to determine content and process. Our goal was to summarize the range of currently accepted practice and frequency of specific components of inpatient DNR orders at NCICC in the U.S. Methods: Letters were mailed to NCICCs inviting participation in this IRB approved study, with repeat mailing at 4 weeks and follow up phone calls at 8 weeks. Stamped envelopes were provided for return of inpatient DNR order forms. 2 physicians (MD) evaluated each form for content and components of the process used to establish DNR. Descriptive analysis was performed using SAS. Results: 34/50 (68%) eligible sites responded and 30 (60%) participated. 67% of the participating sites used specific inpatient DNR forms; the remainder established DNR status exclusively by order sheets and without forms. 70% of form titles were negative in connotation, 30% neutral and none were favorable. 30% provided a definition of cardiopulmonary resuscitation or of DNR. 57% were confined exclusively to resuscitation efforts, with the remainder offering additional levels of treatment limitation. The majority of forms were indefinite in duration, with 20% time-limited. 25% specified exceptions for the operating room and 10% contained references to minors. Requirement for MD signature was universal, with 25% each also requiring patient and/or surrogate signatures. Additional requirements included documentation of persons involved in the decision (20%) and content of DNR discussion (67%). Conclusions: There is a wide range of practice in establishing DNR status at NCICC. Institutions were almost equally divided between addressing resuscitation exclusively versus resuscitation with limitation of other supportive therapies. Documentation of patient participation in the process is uncommon. A more consistent approach to addressing appropriate care options at the end of life could potentially facilitate decision making around this sensitive issue. No significant financial relationships to disclose.

A comparative analysis of inpatient Do Not Resuscitate (DNR) forms used at National Cancer Institute-designated cancer centers (NCICC)

8222 Background: The decision-making process establishing inpatient DNR status is difficult and may lead to conflict among the patient, family and clinician. In this study, we examined DNR forms from NCICC to determine content and process. Our goal was to summarize the range of currently accepted practice and frequency of specific components of inpatient DNR orders at NCICC in the U.S. Methods: Letters were mailed to NCICCs inviting participation in this IRB approved study, with repeat mailing at 4 weeks and follow up phone calls at 8 weeks. Stamped envelopes were provided for return of inpatient DNR order forms. 2 physicians (MD) evaluated each form for content and components of the process used to establish DNR. Descriptive analysis was performed using SAS. Results: 34/50 (68%) eligible sites responded and 30 (60%) participated. 67% of the participating sites used specific inpatient DNR forms; the remainder established DNR status exclusively by order sheets and without forms. 70% of form titles were negative in connotation, 30% neutral and none were favorable. 30% provided a definition of cardiopulmonary resuscitation or of DNR. 57% were confined exclusively to resuscitation efforts, with the remainder offering additional levels of treatment limitation. The majority of forms were indefinite in duration, with 20% time-limited. 25% specified exceptions for the operating room and 10% contained references to minors. Requirement for MD signature was universal, with 25% each also requiring patient and/or surrogate signatures. Additional requirements included documentation of persons involved in the decision (20%) and content of DNR discussion (67%). Conclusions: There is a wide range of practice in establishing DNR status at NCICC. Institutions were almost equally divided between addressing resuscitation exclusively versus resuscitation with limitation of other supportive therapies. Documentation of patient participation in the process is uncommon. A more consistent approach to addressing appropriate care options at the end of life could potentially facilitate decision making around this sensitive issue. No significant financial relationships to disclose.