Ethical decision making about end-of-life care issues by pediatric oncologists in economically diverse settings.

PURPOSE: Pediatric cancer represents 1% to 4% of all cancers worldwide, with the majority of diagnoses in developing countries where mortality remains much higher than that in high-income countries. We sought to describe differences in ethical decision-making at the end of life among an international sample of pediatric oncologists practicing in countries with a variety of income levels and resource settings. METHODS: Pediatric oncologists subscribing to an educational international oncology Web site were invited to complete a 38-item web-based survey investigating ethical domains related to end-of-life care: level of care, fiduciary responsibility, decision making, and justice. RESULTS: Responses were received from 401 physicians in 83 countries, with most respondents practicing in middle-income or high-income countries. Significant differences in attitudes toward ethical issues existed across the national developmental indices. CONCLUSIONS: Further education on ethical principles is warranted in pediatric oncology, particularly among oncologists practicing in low-income or middle-income countries.

Improving the histopathologic diagnosis of pediatric malignancies in a low-resource setting by combining focused training and telepathology strategies.

BACKGROUND: Accurate diagnosis is critical for optimal management of pediatric cancer. Pathologists with experience in pediatric oncology are in short supply in the developing world. Telepathology is increasingly used for consultations but its overall contribution to diagnostic accuracy is unknown. PROCEDURE: We developed a strategy to provide a focused training in pediatric cancer and telepathology support to pathologists in the developing world. After the training period, we compared trainee's diagnoses with those of an experienced pathologist. We next compared the effectiveness of static versus dynamic telepathology review in 127 cases. Results were compared by Fisher's exact test. RESULTS: The diagnoses of the trainee and the expert pathologist differed in only 6.5% of cases (95% CI, 1.2-20.0%). The overall concordance between the telepathology and original diagnoses was 90.6% (115/127; 95% CI, 84.1-94.6%). CONCLUSIONS: Brief, focused training in pediatric cancer histopathology can improve diagnostic accuracy. Dynamic and static telepathology analyses are equally effective for diagnostic review.

Implementation of a data management program in a pediatric cancer unit in a low income country.

INTRODUCTION: Pediatric cancer units in low-income countries lack data on which to base quality improvement initiatives. We implemented a data management program in the oncology unit of the children's hospital of Tegucigalpa, Honduras, and then we assessed training and supervision of data managers, data accuracy, and completeness as well as obstacles encountered. METHODS: Training included 2 days of off-site hands-on instruction in the use of an online database, daily on-site supervision by physicians, periodic online meetings for education and problem-solving, and continuous e-mail support. RESULTS: Of the 652 patients diagnosed with acute leukemia between July 1995 and June 2005, 150 (23%) had not yet been registered in the database at the time of audit and 65 (10%) had missing medical records. The remaining 437 charts (67%) were reviewed by an external auditor and compared to the data entered previously by the two trained data managers. Protocol information was incomplete in 30% of cases, and the cause of death was inaccurate in 18%. All other data were 99% accurate and 93%-100% complete. Obstacles included a limited medical records system, poor organization of the charts, missing records, inconsistently documented protocol information, data managers who lack a medical background, and slow or unreliable internet connections. CONCLUSION: Data managers can be trained to effectively collect basic pediatric oncology data in a low-income country. Addressing inadequacies in the medical record system while providing specific training in protocol-based care and determination of cause of death for both physicians and data managers will improve data quality.