Simulation Modeling to Set Priorities for the Development of Cancer Quality Measures
This project, a component of the Cancer Quality of
Care Measures Project, uses simulation modeling to set priorities in the development
of quality of care measures for colorectal cancer diagnosis and treatment. Dr. Karen Kuntz
(Associate Professor of Decision Science, Harvard
School of Public Health) will be primary modeler for this project. Dr. Kuntz has
previously collaborated with NCI in developing a CISNET colorectal cancer simulation model. This model
mapped key clinical processes that contribute to improved survival among stage I-III
colorectal cancer patients:
- Patients initially diagnosed with stage I-III cancer face a monthly risk of
developing metastatic recurrence, which is a function of age at diagnosis, stage, and
whether or not the patient received adjuvant chemotherapy.
- Patients who develop metastatic recurrence have a chance that their metastasis is
rescectable, which is a function of whether or not they were receiving post-treatment
- Patients with resectable metastases who are successfully resected and thus have no
evidence of disease (NED) face a monthly risk of progressing to nonresectable metatatic
disease. This risk is a function of whether or not the patients received chemotherapy
- Patients with non-resectable metastatic disease face a cancer-specific hazard rate,
which is a function of age at nonresectable metastases and whether or not the patient
received chemotherapy following the diagnosis of nonresectable metastases.
In this project, Outcomes Research staff worked with Dr. Kuntz to focus primarily on
mapping clinical processes around effectiveness of care. Dr. Kuntz modified the existing
simulation model to provide estimates of the contribution of four processes, or nodes, to
improved cancer outcomes. She identified those nodes that have a potential for net
positive impact on survival or mortality and rank ordered them by degree of impact and
improvement potential (i.e., the difference between current practice and optimal
practice). Current practice was based on analyses of SEER and SEER-Medicare linked data.
Optimal practice was based on reasonable estimates of maximal diffusion of care. The
simulation modeling found that process improvement in the receipt of adjuvant chemotherapy
after surgery and the receipt of chemotherapy following the diagnosis of nonresectable
metastatic disease provided the greatest potential for improving survival for this
population. Post-treatment surveillance contributed little to survival within the
In a related project, in May 2006 the Outcomes Research Branch (ORB) and the Health Communication and Informatics Research Branch (HCIRB) organized a symposium on Patient-centered Communication (PCC) in cancer care. The symposium addressed several key issues related to PCC in cancer care, including:
- pathways by which the functions of PCC are likely to affect desired patient health outcomes across the cancer continuum;
- key variables that are likely to facilitate or mitigate the relationship between the functions of PCC and patient health outcomes;
- challenges in measuring and monitoring the functions of PCC; and
- interventions needed to ensure the functions of PCC are accomplished as part of routine cancer care delivery.
Symposium participants were asked to review a conceptual framework developed by Ronald M. Epstein, MD, and Richard L. Street, Jr., Ph.D., as part of their work on the NCI monograph Patient-Centered Communication in Cancer Care: Promoting Healing and Reducing Suffering. Two patient case studies were introduced to illustrate the application of the PCC conceptual framework and generate discussion.