A national quality improvement effort: cancer registry data

Need of informatics in designing interoperable clinical registries

Clinical registries are designed to collect information relating to a particular condition for research or quality improvement. Intuitively, informatics in the area of data management and extraction plays a central role in clinical registries. Due to various reasons such as lack of informatics awareness or expertise, there may be little informatics involvement in designing clinical registries. In this paper, we studied a clinical registry from two critical perspectives, data quality and interoperability, where informatics can play a role. We evaluated these two aspects of an existing registry, Gynecology Surgery Registry, by mapping data elements and value sets, used in the registry, to a standardized terminology, SNOMED-CT. The results showed that majority of the values are ad-hoc and only 6 of 91 procedures in the registry could be mapped to the SNOMED-CT. To tackle this issue, we assessed the feasibility of automated data abstraction process, by training machine learning classifiers, based on existing manually extracted data. These classifiers achieved a reasonable average F-measure of 0.94. We concluded that more informatics engagement is needed to improve the interoperability, reusability, and quality of the registry.

Adverse drug event reporting systems: a systematic review

AIM

Adverse drug events (ADEs) are harmful and unintended consequences of medications. Their reporting is essential for drug safety monitoring and research, but it has not been standardized internationally. Our aim was to synthesize information about the type and variety of data collected within ADE reporting systems.

METHODS

We developed a systematic search strategy, applied it to four electronic databases, and completed an electronic grey literature search. Two authors reviewed titles and abstracts, and all eligible full-texts. We extracted data using a standardized form, and discussed disagreements until reaching consensus. We synthesized data by collapsing data elements, eliminating duplicate fields and identifying relationships between reporting concepts and data fields using visual analysis software.

RESULTS

We identified 108 ADE reporting systems containing 1782 unique data fields. We mapped them to 33 reporting concepts describing patient information, the ADE, concomitant and suspect drugs, and the reporter. While reporting concepts were fairly consistent, we found variability in data fields and corresponding response options. Few systems clarified the terminology used, and many used multiple drug and disease dictionaries such as the Medical Dictionary for Regulatory Activities (MedDRA).

CONCLUSION

We found substantial variability in the data fields used to report ADEs, limiting the comparability of ADE data collected using different reporting systems, and undermining efforts to aggregate data across cohorts. The development of a common standardized data set that can be evaluated with regard to data quality, comparability and reporting rates is likely to optimize ADE data and drug safety surveillance.

Laryngeal adenocarcinoma not otherwise specified: A population-based perspective

OBJECTIVES/HYPOTHESIS

Laryngeal adenocarcinoma not otherwise specified (LAdC NOS) is a category to which variants of minor salivary gland tumors of the larynx that do not fit other well-characterized histological subtypes are assigned. Its rare nature and inconsistency in available reports has hindered the investigation and further understanding of this malignancy. In this study, a national population-based resource was used to evaluate the epidemiology and survival of this rare entity.

STUDY DESIGN

Retrospective population-based analysis.

METHODS

The Surveillance, Epidemiology, and End Results database was searched for patients diagnosed with LAdC NOS between 1973 and 2012. Patient demographics, tumor location, TNM stage, grade, incidence, and survival trends were collected and analyzed.

RESULTS

One hundred eleven patients met criteria for diagnosis of LAdC NOS, of which the majority were male (80.2%), white (84.7%), with a mean age of 65 years. The supraglottis was the most common site at presentation (38.7%). The majority presented with grade II tumor (45.7%). TNM staging revealed T2 (36.8%), N0 (72.2%), and M0 (88.9%) to be the most common classification. The overall incidence between the years of 2000 and 2012 was 0.008/100,000 individuals. The overall 5-year disease-specific survival (DSS) was 60.1%, compared to 85.7% in patients treated with combination surgery and radiotherapy.

CONCLUSIONS

LAdC NOS is an uncommon malignancy. It most commonly affects men in their mid-60s, indiscriminate of race. Lesions most commonly present in the supraglottis and are more often low grade histologically. DSS is highest in patient treated with combination surgery and radiotherapy.

LEVEL OF EVIDENCE

4 Laryngoscope, 2016 127:424-429, 2017.

Strategies for improving the data quality in national hospital discharge data system: a delphi study

Introduction:

National hospital discharge data system can play a critical role in community health assessment, disease surveillance, strategic planning, policymaking, service quality control, and research. Moreover, the quality of hospital discharge data affects the usefulness of the data and is one of the prerequisites for effective utilization of the data. Thus, the present study aimed to identify the necessary actions for improving the data quality in the national hospital discharge data system and present a model for Iran based on the experiences of England, Canada, and New Zealand.

Methods:

In doing so, the measures performed in these countries were investigated. The related data were organized in six categories of standards and procedures, training and coordination with the users, assurance from the capability of the system’s software, data modification, data quality control, and documentation and reporting the data quality. According to the gathered data, the primary model was designed. Then, the model was assessed using a two-round Delphi technique by 33 and 31 experts, respectively.

Conclusion:

According to the findings, a model was presented in order to improve the data quality of Iran’s national hospital discharge data system.

Clinical presentation, histology, and treatment in 430 patients with primary tumors of the spinal cord, spinal meninges, or cauda equina

OBJECT Patients having a primary tumor of the spinal cord, spinal meninges or cauda equina, are relatively rare. Neurosurgeons encounter and treat such patients, and need to be aware of their clinical presentation, tumor types, treatment options, and potential complications. The purpose of this paper is to report results from a series of 430 patients with primary intraspinal tumors, taken from a larger cohort of 9661 patients with primary tumors of the CNS. METHODS Extensive information on individuals diagnosed (in the year 2000) as having a primary CNS neoplasm was prospectively collected in a Patient Care Evaluation Study conducted by the Commission on Cancer of the American College of Surgeons. Data from US hospital cancer registries were submitted directly to the National Cancer Database. Intraspinal tumor cases were identified based on ICD-O-2 topography codes C70.1, C72.0, and C72.1. Analyses were performed using SPSS. RESULTS Patients with primary intraspinal tumors represented 4.5% of the CNS tumor group, and had a mean age of 49.3 years. Pain was the most common presenting symptom, while the most common tumor types were meningioma (24.4%), ependymoma (23.7%), and schwannoma (21.2%). Resection, surgical biopsy, or both were performed in 89.3% of cases. Complications were low, but included neurological worsening (2.2%) and infection (1.6%). Radiation therapy and chemotherapy were administered to 20.3% and 5.6% of patients, respectively. CONCLUSIONS Data from this study are suitable for benchmarking, describing prevailing patterns of care, and generating additional hypotheses for future studies.

The Lombardia Stroke Unit Registry: 1-year experience of a web-based hospital stroke registry

This paper presents methodological aspects of the Lombardia Stroke Registry. At the registry start-up, 36 recruiting centres were identified according to a regional survey. The registry recruits consecutive patients with acute stroke or transient ischaemic attacks (TIAs). A 3-month follow-up was planned to correlate acute care with outcomes. On 31st December 2007, data concerning 6,181 patients discharged alive were available. The registry aims at measuring performance parameters, identifying guidelines non-compliance and analysing care processes. In this first phase, 30% of the Lombardia acute stroke and 10% of TIA patients have been enrolled, thus the sample can be considered informative for the disease care in the region. The proportion of completed data items is very high with very small differences among items. The following critical points were highlighted: (1) lack of data input staff for 30% of centres, and (2) difficulty of obtaining the informed consent for post-discharge follow-up.

Laryngeal Cancer in the United States: Changes in Demographics, Patterns of Care, and Survival

BACKGROUND

Survival has decreased among patients with laryngeal cancer during the past 2 decades in the United States. During this same period, there has been an increase in the nonsurgical treatment of laryngeal cancer.

OBJECTIVE

The objectives of this study were to identify trends in the demographics, management, and outcome of laryngeal cancer in the United States and to analyze factors contributing to the decreased survival.

STUDY DESIGN

The authors conducted a retrospective, longitudinal study of laryngeal cancer cases.

METHODS

Review of the National Cancer Data Base (NCDB) revealed 158,426 cases of laryngeal squamous cell carcinoma (excluding verrucous carcinoma) diagnosed between the years 1985 and 2001. Analysis of these case records addressed demographics, management, and survival for cases grouped according to stage, site, and specific TNM classifications.

RESULTS

This review of data from the NCDB analysis confirms the previously identified trend toward decreasing survival among patients with laryngeal cancer from the mid-1980s to mid-1990s. Patterns of initial management across this same period indicated an increase in the use of chemoradiation with a decrease in the use of surgery despite an increase in the use of endoscopic resection. The most notable decline in the 5-year relative survival between the 1985 to 1990 period and the 1994 to 1996 period occurred among advanced-stage glottic cancer, early-stage supraglottic cancers, and supraglottic cancers classified as T3N0M0. Initial treatment of T3N0M0 laryngeal cancer (all sites) in the 1994 to 1996 period resulted in poor 5-year relative survival for those receiving either chemoradiation (59.2%) or irradiation alone (42.7%) when compared with that of patients after surgery with irradiation (65.2%) and surgery alone (63.3%). In contrast, identical 5-year relative survival (65.6%) rates were observed during this same period for the subset of T3N0M0 glottic cancers initially treated with either chemoradiation or surgery with irradiation.

CONCLUSIONS

The decreased survival recorded for patients with laryngeal cancer in the mid-1990s may be related to changes in patterns of management. Future studies are warranted to further evaluate these associations.

Quantifying data quality for clinical trials using electronic data capture

Background

Historically, only partial assessments of data quality have been performed in clinical trials, for which the most common method of measuring database error rates has been to compare the case report form (CRF) to database entries and count discrepancies. Importantly, errors arising from medical record abstraction and transcription are rarely evaluated as part of such quality assessments. Electronic Data Capture (EDC) technology has had a further impact, as paper CRFs typically leveraged for quality measurement are not used in EDC processes.

Methods and Principal Findings

The National Institute on Drug Abuse Treatment Clinical Trials Network has developed, implemented, and evaluated methodology for holistically assessing data quality on EDC trials. We characterize the average source-to-database error rate (14.3 errors per 10,000 fields) for the first year of use of the new evaluation method. This error rate was significantly lower than the average of published error rates for source-to-database audits, and was similar to CRF-to-database error rates reported in the published literature. We attribute this largely to an absence of medical record abstraction on the trials we examined, and to an outpatient setting characterized by less acute patient conditions.

Conclusions

Historically, medical record abstraction is the most significant source of error by an order of magnitude, and should be measured and managed during the course of clinical trials. Source-to-database error rates are highly dependent on the amount of structured data collection in the clinical setting and on the complexity of the medical record, dependencies that should be considered when developing data quality benchmarks.

Inter-rater reliability of data elements from a prototype of the Paul Coverdell National Acute Stroke Registry

BACKGROUND

The Paul Coverdell National Acute Stroke Registry (PCNASR) is a U.S. based national registry designed to monitor and improve the quality of acute stroke care delivered by hospitals. The registry monitors care through specific performance measures, the accuracy of which depends in part on the reliability of the individual data elements used to construct them. This study describes the inter-rater reliability of data elements collected in Michigan's state-based prototype of the PCNASR.

METHODS

Over a 6-month period, 15 hospitals participating in the Michigan PCNASR prototype submitted data on 2566 acute stroke admissions. Trained hospital staff prospectively identified acute stroke admissions, abstracted chart information, and submitted data to the registry. At each hospital 8 randomly selected cases were re-abstracted by an experienced research nurse. Inter-rater reliability was estimated by the kappa statistic for nominal variables, and intraclass correlation coefficient (ICC) for ordinal and continuous variables. Factors that can negatively impact the kappa statistic (i.e., trait prevalence and rater bias) were also evaluated.

RESULTS

A total of 104 charts were available for re-abstraction. Excellent reliability (kappa or ICC > 0.75) was observed for many registry variables including age, gender, black race, hemorrhagic stroke, discharge medications, and modified Rankin Score. Agreement was at least moderate (i.e., 0.75 > kappa >/=; 0.40) for ischemic stroke, TIA, white race, non-ambulance arrival, hospital transfer and direct admit. However, several variables had poor reliability (kappa < 0.40) including stroke onset time, stroke team consultation, time of initial brain imaging, and discharge destination. There were marked systematic differences between hospital abstractors and the audit abstractor (i.e., rater bias) for many of the data elements recorded in the emergency department.

CONCLUSION

The excellent reliability of many of the data elements supports the use of the PCNASR to monitor and improve care. However, the poor reliability for several variables, particularly time-related events in the emergency department, indicates the need for concerted efforts to improve the quality of data collection. Specific recommendations include improvements to data definitions, abstractor training, and the development of ED-based real-time data collection systems.

Analysis of data-collection methods for an acute stroke care registry

This study aims to assess and compare the completeness and reliability of data collected by prospective and retrospective methods for the Paul Coverdell National Acute Stroke Registry. The prototypes consisted of eight states that used the same data elements but differed in their collection approach. Three prototypes employed retrospective case ascertainment (n=1218), and five prototypes used prospective or a combination of prospective and retrospective case ascertainment (n=1602). RTI International performed an audit analysis of the eight prototypes. Completeness, exact match, and discrepancy analyses were performed with data elements grouped into 12 categories for this analysis. A sample of 2820 (37.6%) from a total of 7494 records from 91 hospitals was studied. The "in-hospital complications" section had the highest percentage of completeness (99.6%), followed by "demographic data" (97.7%), and "in-hospital diagnostic procedures" (93.4%). The section with the lowest percentage of completeness was "thrombolytic treatment" (53.5%), followed by "reasons for nontreatment with thrombolytics" (57.1%), and "signs and symptoms onset" (63.5%). Across all prototype elements, exact matches with audit data ranged from 62.8% to 95.9%. Documentation of the date/time of stroke onset and of arrival in the emergency department had a high number of discrepancies with audit data, with exact match percentages of 69.7% and 64.5%, respectively. No significant difference was found between retrospective and prospective case ascertainment in completeness or matching with audit data. Combined retrospective and prospective data-collection approaches for different types of data elements may be best in terms of both completeness and accuracy.

Validity of a population-based cancer register in Sweden--an assessment of data reproducibility in the South-East Region Prostate Cancer Register

BACKGROUND

With a population-based setting, high coverage and accurately recorded data, the validity of a register is guaranteed. The South-East Region Prostate Cancer relies on the National Cancer Register as a basic source of data, thereby ensuring a high coverage of the corresponding geographic area. To assess the reproducibility of the data recorded a random sample of the cases were reviewed a second time and compared to the original recording.

MATERIAL AND METHODS

The South-East Region Prostate Cancer Register was started in 1987. In addition to the basic data acquired from the Swedish National Register, it also includes tumour stage, grade, treatment and, since 1992, PSA. In the first stage of quality assessment 10 cases for each of the years 1987-1996 from Linköping University Hospital were randomly selected for two independent recodings according to the same protocol as the original registration. In the second step 10 cases each for the same years from the remaining 8 hospitals in the region were selected for a single recoding.

RESULTS

No systematic deviations were seen between the two independent recodings from Linköping, a single recoding was therefore considered sufficient for assessing the reproducibility of the data from the remaining hospitals in the region. The Kappa values for agreement between the original registration and the single recoding ranged from 0.589 to 0.869.

CONCLUSION

The population-based setting and high coverage guarantees the external validity of the register. The internal validity is ensured by the high reproducibility shown in the present study.

Defining and improving data quality in medical registries: a literature review, case study, and generic framework

Over the past years the number of medical registries has increased sharply. Their value strongly depends on the quality of the data contained in the registry. To optimize data quality, special procedures have to be followed. A literature review and a case study of data quality formed the basis for the development of a framework of procedures for data quality assurance in medical registries. Procedures in the framework have been divided into procedures for the co-ordinating center of the registry (central) and procedures for the centers where the data are collected (local). These central and local procedures are further subdivided into (a) the prevention of insufficient data quality, (b) the detection of imperfect data and their causes, and (c) actions to be taken / corrections. The framework can be used to set up a new registry or to identify procedures in existing registries that need adjustment to improve data quality.

Towards case-based performance measures: uncovering deficiencies in applied medical care

Measures are designed to evaluate the processes and outcomes of care associated with the delivery of clinical (and non-clinical) services. They allow for intra- and interorganizational comparison to be used continuously to improve patient health outcomes. The use of performance measures always means to abstract the complex reality (medical scenarios and procedures) in order to provide an understandable and comparable output. Measures can focus on global performance. The more detailed data are available the more specific judgements with respect to the appropriateness of clinical decision-making and implementation of evidence are feasible. Externally reported measures are intended both to inform and lead to action. By providing this information, deficiencies in patient care and unnecessary variations in the care process can be uncovered. Such variations have contributed to disparities in morbidity and mortality. The developments in information technology, especially world-wide interconnectivity, standards for electronic data exchange and facilities to store and manage large amounts of data, offer the opportunity to analyse health-relevant information in order to make the delivery of healthcare services more transparent for consumers and providers. Global performance measures, such as the overall life expectancy (mortality) in a country, can give a rough orientation of how well health systems perform but they do not offer general solutions nor specific insights into care processes that have to be improved. In contrast to population-based measures, case-based performance measures use a defined group of patients depending on specific patient characteristics and features of disease. By means of these measures we are able to compare the number of patients that receive a necessary medical procedure against those patients who do not. The use of case-based measures is a bottom-up approach to improve the overall performance in the long run. They are not only a tool for global orientation but can offer a straightforward link to the areas of deficient care and the underlying procedures. Performance measures are relevant to providers as well as consumers, from their own individual perspective. Cased-based measures focus on the management of individual patient. This approach to performance measurement can inform physicians in a meaningful and constructive way by monitoring their individual performance and by pointing out possible areas of improvement.

National survey of head and neck verrucous carcinoma: patterns of presentation, care, and outcome

BACKGROUND

Verrucous carcinoma is distinguished by controversy regarding appropriate diagnosis and treatment. This study provides a contemporary survey of demographics, patterns of care, and outcome for this disease in the United States.

METHODS

The National Cancer Data Base had 2350 cases of verrucous carcinoma of the head and neck diagnosed between 1985 and 1996. Statistical procedures included chi-square analyses, Student t tests, and relative survival.

RESULTS

Tumors originated most frequently in the oral cavity (55.9%) and larynx (35.2%). Although most patients were male (60.0%), oral cavity tumors were more common among older females. The most prevalent treatment was surgery alone (69.7%), followed by surgery combined with irradiation (11.0%) and irradiation alone (10.3%). For oral cavity tumors, surgery alone was more common among early (85.8%) than advanced cases (56.9%); a larger proportion of advanced disease received radiation alone or surgery and irradiation combined. Most laryngeal tumors were treated with surgery (60.3% for early and 55.6% for advanced disease), but a higher proportion received radiation alone or surgery combined with radiation compared with oral cavity cases. Five-year relative survival was 77.9%. For localized disease, survival after surgery was 88.9% compared with 57.6% after irradiation.

CONCLUSIONS

Demographic differences implicate different mechanisms of carcinogenesis for verrucous carcinoma arising in the oral cavity and the larynx. Although selection bias may account for the differences observed, patients receiving initial treatment with surgery had better survival than those treated with irradiation, especially for cases originating in the oral cavity.

Evaluating the implications of clinical practice guidelines for patient care

Patient care evaluation studies have been developed by the Commission on Cancer of the American College of Surgeons. The studies were primarily designed to monitor trends in diagnosis, therapy, and outcome of specific oncologic diseases in hospitals and cancer centers. As they reflect the current standards of patient care, patient care evaluation studies have become valid tools of quality management in medicine. In an international pilot project that began in 1996, this approach was redefined to evaluate the impact of current clinical practice guidelines in oncology. Close cooperation between medical societies in the United States and Germany under the coordination of the Commission on Cancer and the Institute of Medical Informatics at the Justus-Liebig-University of Giessen was established. This infrastructure for data collection, data management, analysis, and interpretation of results allows for the recognition of international differences in patient care. Our results indicate discrepancies between current state-of-the-art patient care represented by clinical practice guidelines and the diagnostic and therapeutic procedures in the clinical routine. Patient care evaluation studies are designed as exploratory, not confirmatory, trials. In contrast with confirmatory trials, their aims may not always lead to predefined hypotheses. They reflect routine practice and are not the basis of the formal proof of efficacy, although they may contribute to the total body of relevant evidence. Without this comprehensive approach to evaluation, the potential of clinical practice guidelines to improve patient care remains unknown.

Patterns of care for patients with primary differentiated carcinoma of the thyroid gland treated in Germany during 1996. U.S. and German Thyroid Cancer Group

BACKGROUND

To determine current patterns of care and disease characteristics for patients with thyroid carcinoma, a Patient Care Evaluation Study was initiated in 1996 in the U.S. and Germany. This project addresses ongoing concerns with respect to the diagnostic evaluation and treatment of patients diagnosed with thyroid carcinoma and raises questions concerning how physicians are interpreting current standards and acting on the basis of these recommendations.

METHODS

Patients with primary thyroid carcinoma were entered into a prospective multicenter observational study with free choice of treatment (no control group) between January 1, 1996 and December 31, 1996 in Germany. This resulted in a total of 2537 cases under observation and analysis; 1685 patients had papillary carcinoma (66.4%), 691 had follicular carcinoma (27.2%), 70 had medullary carcinoma (2.8%), and 91 had anaplastic carcinoma (3.6%). The 2376 patients with carcinoma of either papillary or follicular histology were included in the current analysis.

RESULTS

The major symptoms reported for patients with papillary and follicular thyroid carcinoma was neck mass (reported in 76% and 79%, respectively) followed by dysphagia (reported in 25% and 27%, respectively), stridor (reported in 9% and 14%, respectively), and neck pain (reported in 7% and 8%, respectively). Greater than 50% of the patients with papillary thyroid carcinoma were reported to have American Joint Committee on Cancer/International Union Against Cancer Stage I disease. Between 37-39% of the follicular carcinoma patients had Stage I and Stage II disease. Only slight differences in the diagnostic approach to patients with papillary or follicular carcinoma were noted. The majority of patients underwent an ultrasound of the thyroid region (78.1%), which was suggestive of carcinoma in only 39% of the cases. A thyroid scan was performed on 76.6% of patients, and the results were suggestive of carcinoma in 44.8% of the individuals. In contrast, fine-needle aspiration biopsy of the thyroid is highly recommended in the current Clinical Practice Guidelines (CPG) but results were obtained in only 27.4% of the patients. Total thyroidectomy without lymph node dissection was the most commonly used surgical procedure in the treatment of patients with papillary and follicular thyroid carcinoma. Only approximately 2% of patients at low risk in the group with Stage I disease were treated with a lobectomy. In 80% of the patients with Stage I papillary thyroid carcinoma and approximately 90% of those patients diagnosed with Stage II, III, and IV disease treating physicians chose to utilize radioiodine as adjuvant treatment after disease-directed surgery. External beam radiation was added to the treatment regimen for many patients diagnosed with Stage III and IV disease (30% in patients with papillary thyroid carcinoma and 33% in patients with follicular thyroid carcinoma).

CONCLUSIONS

To the authors' knowledge no single effective diagnostic test for thyroid carcinoma currently is available and in the majority of cases a combination of ultrasound, thyroid scan, or fine-needle aspiration biopsy together with the clinical findings (e.g., thyroid mass) led to a diagnosis of carcinoma. The authors suspect that the high prevalence of concomitant pathologic findings such as goiter, even in the healthy population in Germany, reduces the accuracy of all diagnostic test methods and may account for the frequent use of imaging techniques. The majority of patients underwent a total or near-total thyroidectomy. Total thyroidectomy with radical lymph node dissection was used very frequently in those patients with papillary thyroid carcinoma (22%). German physicians tend to surgically treat early stage thyroid carcinoma somewhat more radically than recommended in the CPG. With respect to other treatment options employed as part of the first course of treatment, radioiodine appears to play the most important role. [See commentary o

Brain tumor survival: results from the National Cancer Data Base

Hospital-based data reported to the National Cancer Data Base (NCDB) were available for over 60,000 patients with a primary brain tumor diagnosed from 1985-1988 and 1990-1992. The most common histologies were glioblastomas, astrocytomas and meningiomas. Five-year survival rates for these tumors were 2%, 30% and 70% respectively. Histology, age at diagnosis, behavior, and location were important variables in estimating survival. Comparisons with population-based registry data suggest that the malignant tumors are well represented in NCDB, but the benign histologies are under-reported. Survival estimates for the malignant tumors are comparable to previously reported studies. The NCDB provides recent information on brain tumor distribution and survival patterns not available in other large databases.

The National Cancer Data Base 10-year survey of breast carcinoma treatment at hospitals in the United States

BACKGROUND

The National Cancer Data Base (NCDB), a joint project of the American College of Surgeons Commission on Cancer and the American Cancer Society, is a cancer management and outcome data base for health care organizations. It provides a comparative summary of patient care that is used by participating hospitals and communities for self-assessment. The most current (1995) data are described herein.

METHODS

Since 1989, seven calls for data have been issued, yielding reports on a total of 240,031 breast carcinoma patients for the years included in this analysis. A total of 1849 hospital cancer registries responded to at least 1 of the calls for data.

RESULTS

A continuous improvement in care was reported. By 1995, 45.8% (nearly one-half) of breast carcinoma patients were diagnosed early as Stage 0 or I, and early stage patients (Stage 0 or I) were most often treated with partial mastectomy (in 58% of cases). Favorable 10-year relative survival rates for Stage 0 (95%) and Stage I (88%) breast carcinoma patients were reported. Patients who were presumed to be Stage I and were not selected for axillary dissection had poorer survival. Survival differences were reported for different treatment groups within individual stage strata. Over the 10-year observation period, fewer patients from lower-income neighborhoods were diagnosed with early stage breast carcinoma. In general, the annual relative survival rate remained constant over the 10-year observation period (with no plateau after 5 years) within each stage and for all stages combined.

CONCLUSIONS

Improvements in diagnosis and treatment during the period 1985-1995 were demonstrated by these data. The NCDB breast carcinoma data are appropriate norms for formal quality assurance purposes, such as those specified by the Standards of the Commission on Cancer published by the American College of Surgeons Commission on Cancer. Cancer committees and other clinicians working within the hospital setting should assess and compare stage distribution, stage specific treatment patterns, and the correlations between the outcomes of patients and both disease stage and treatment.

The growth and maturation of the National Cancer Data Base

BACKGROUND

The National Cancer Data Base (NCDB), a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society, is a cancer management and outcomes data base for health care organizations. It provides a comparative summary of patient care that is used by communities and participating hospitals for self-assessment. The most current (1994) data are described here.

METHODS

Six calls for data have yielded a total of 4,580,000 cases for the years 1985-1994. A total of 1735 hospital cancer registries have each participated in at least one of the calls for data.

RESULTS

Summing the last year's report from each of the 1227 hospitals that participated in 1994, the cases represent the equivalent of 57% of the estimated 1994 U.S. cancer cases. These data were received from all six regions of the country, including all 50 states. Ninety-seven percent of patients received all or part of their treatment at the reporting hospital. The four most common cancers are carcinomas of the breast (15.7%), lung (14.3%), prostate (13.1%), and colon (7.7%), and collectively they comprise a majority of new cases.

CONCLUSIONS

The NCDB is a cancer management and outcomes data base for health care organizations that currently provides data on 57% of the estimated new cases in the U.S. Past data have been used extensively to assess patterns of care and outcomes.

The National Cancer Data Base report on patterns of childhood cancers in the United States

BACKGROUND

Patterns of and progress against childhood cancer have been reported on multi-institution, regional, national, and international bases by several sources in the past. These sources have included clinical cooperative group trials and population-based registries. In general, the population-based surveys have excluded brain tumors of either benign or uncertain behavior. The authors of this article investigated the patterns of data reported for the period 1985-1993, motivated by their interest in assessing the potential of National Cancer Data Base (NCDB) data to 1) facilitate individual institution review and 2) cover institutions that are not members of the Pediatric Oncology Group or the Children's Cancer Group, which are both national clinical cooperative groups.

METHODS

Six annual calls for data, starting with a call for 1985 and 1988 cases, were issued to approximately 2100 hospitals with cancer programs (1340 programs approved by the Commission on Cancer of the American College of Surgeons and 760 other programs). The baseline data items of the NCDB included patient demography, tumor characteristics, initial treatment, and follow-up. The data for each patient were coded in the traditional manner by trained cancer registrars before being transmitted to the NCDB in standard format.

RESULTS

In the most recent year for which data were reported, the NCDB included 42% of all estimated U.S. childhood cancers. The cases were reported by institutions that were members of the Pediatric Oncology Group and the Children's Cancer Group as well as nonmember institutions. The distribution of diagnostic groups reported to the NCDB was generally similar to that reported to SEER, except for lymphomas and brain cancer (the NCDB series included benign as well as malignant brain tumors). The distribution of diagnostic groups reported to the NCDB did not change over the 9-year reporting period (1985-1993). With regard to ethnicity, the most varied distribution of diagnostic groups was found among African American patients. For many types of cancer, the survival of those patients reported to the NCDB was similar to that of patients included in the SEER population-based series. These cancers included Wilms' tumor (NCDB 89% vs. SEER 88%), non-Hodgkin's lymphoma (NCDB 74% vs. SEER 70%), soft tissue sarcomas (NCDB rhabdomyosarcomas 70% and sarcomas 79% vs. SEER soft tissue sarcomas 71%), and neuroblastoma (NCDB 58% vs. SEER 57%).

CONCLUSIONS

The authors concluded that the number of brain tumors of benign and uncertain behavior being diagnosed were significant enough in number that they should be included in regional and national cancer registries that report data for clinical purposes. They further concluded that for reasons of data inclusion and institutional coverage, the NCDB will be an important data base for pediatric cancers that will warrant increased use by pediatric investigators.

Changes in patterns of prostate cancer care in the United States: results of American College of Surgeons Commission on Cancer studies, 1974-1993

BACKGROUND

Advances in medical and public health practice have led to many changes in patterns of prostate cancer care. Data from several studies of prostate cancer by the Commission on Cancer of the American College of Surgeons provide information on the directions, magnitudes, and consequences of these changes.

METHODS

The Commission on Cancer conducts patient care evaluation (PCE) studies based on the voluntary participation of hospital cancer programs and their tumor registries. PCE studies have been conducted repeatedly for prostate cancer covering patients diagnosed as early as 1974 and as recently as 1990. In addition, the National Cancer Data Base of the Commission on Cancer collects data for all forms of cancer from throughout the country. The Commission on Cancer, the American Cancer Society, and the American Urologic Association also has conducted a focused survey of radical prostatectomy outcomes. In aggregate, these multiple studies have accrued 179,366 reports on treatment of prostate cancer patients.

RESULTS

Predominant among practice changes are new techniques of prostate cancer detection and initial evaluation which have led to shifts in disease stage at the time of initial therapy. The proportion of prostate cancer that is localized at the time of detection has increased. Use of radiation therapy and radical prostatectomy has increased as the selection of hormone treatment and no cancer-directed treatment have decreased. Five-year prostate cancer survival has improved for every stage of disease.

CONCLUSIONS

The multiple studies by the Commission on Cancer provide data that are not available from other sources. Continued monitoring of prostate cancer patterns of care may be useful in measuring progress in control of this common disease.

Pancreatic cancer and comparison of a hospital-based tumor registry with a National Cancer Data Base

BACKGROUND

The Commission on Cancer of the American College of Surgeons has called upon institutions providing cancer care to compare practice patterns and outcomes with the National Cancer Data Base (NCDB). Using data from the Virginia Mason Tumor Registry (VMTR), we sought to compare our pancreatic cancer care patterns with those reported nationally, while critically evaluating the accuracy and usefulness of our registry.

METHODS

A review of the 906 computerized patient files in the VMTR from 1973 to 1995 was performed, with more detailed data on patients from the last 5 years retrieved from 224 manual abstracts. These data were compared with the 1991 NCDB for pancreatic cancer.

RESULTS

The percent of cases according to AJCC stage in the NCDB (n = 9,715) versus the VMTR (n = 149), respectively, with cases of unknown stage excluded, were stage I 22% versus 22%, stage II 9% versus 12%, stage III 17% versus 28% (P <0.05) stage IV 52% versus 38% (P <0.05). One-third of the cases in the VMTR 1991 to 1995 were of unknown stage; number of cases with unknown stage for NCDB was 26.6%. The percent of surgical procedures for the NCDB (n = 7,802) versus the VMTR (n = 224), respectively, was pancreatectomy 14% versus 11%, local excision 1% versus 0%, no cancer-directed surgery 83% versus 89% (P <0.05), unknown 2% versus 0% (P <0.05). The actuarial relative survival rates for the 1991 NCDB versus 1987 to 1995 VMTR was 3-year 18% versus 38%, and 5-year 14% versus 35%.

CONCLUSIONS

In comparison with the NCDB, VMTR may have fewer stage IV pancreatic cancers, but improvement is needed in decreasing the number of patients for whom the stage is unknown, as many of these likely represent late stage disease. We have a similar resection rate and a higher survival compared with the NCDB, but a mechanism is not in place to statistically compare our survival data with those of the NCDB. Even though all accredited hospitals are required to have a tumor registry, our data were difficult to compare with those of the NCDB because of coding and reporting deficiencies and inability to statistically compare survival data. Before our practice patterns and outcomes can be compared with national standards, both the VMTR and the NCDB must have standardized data collection and better access to the data.

The National Cancer Data Base report on patterns of hospital reporting

BACKGROUND

Previous Commission on Cancer data from the National Cancer Data Base (NCDB) have examined time trends in stage of disease, treatment patterns, and survival for selected cancers. The most current (1993) data are described here.

METHODS

Five calls for data have yielded a total of 3,700,000 cases for the years 1985 through 1993, including 477,679 cases for 1988, and 608,593 cases for 1993, from hospital cancer registries across the U.S.

RESULTS

The most recent call for data for 1993 comprised 52% of the estimated new cases of cancer in the U.S. The country was comprised of 6 regions, with the Mountain and Southeast regions having the highest regional reporting of new cases of cancer (69% and 55%, respectively) and the Northeast and Pacific regions having the lowest (47% each). Approximately 96% of patients received their treatment at the reporting hospital. The 4 most common carcinomas were breast (15.7%), lung (14.6%), prostate (14.2%), and colon (7.5%) and comprised the majority of new cases. Trends in patterns of care for breast carcinoma were analyzed for possible bias in the 1988 and 1993 periods. When hospitals reporting only in 1988 or in 1993 were compared with hospitals reporting at both time points, the only differences were small differences in ethnic participation. These differences were less than 1.5% in the proportion of African Americans reported in the different time periods. There were no significant differences in the downstaging of breast carcinoma, or the role of conservative surgery or adjuvant radiation therapy.

CONCLUSIONS

The NCDB is a cancer management and outcomes data base for health care organizations that presently comprises 52% of the estimated new cases in the U.S. This will increase to 80% as the approved hospitals of the Commission on Cancer are required to report to the NCDB. Comparison of breast carcinoma findings at two time periods appeared similar regardless of hospital reporting set (i.e., set of hospitals reporting for one period versus both periods).