Related Experiment Video
Updated: Jun 3, 2026

Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Using maps and funnel plots to explore variation in place of death from cancer within London, 2002-2007
P Madden1, Vh Coupland, H Møller
1King's College London, Thames Cancer Registry, London, UK.
Abstract:
London has a high proportion of hospital deaths, which health policy seeks to reduce. We explore variation and trends in place of death from cancer within London between 2002 and 2007. Mortality data based on death certificates were used to define deaths from cancer at home, hospice, hospital and nursing home and examine trends over time for London. Proportions of deaths in each place were presented in maps for 31 London primary care trusts (PCTs). Funnel plots were used to identify consistent performance outside the control limits of three standard deviations. There was little overall change in place of death for London, but consistent variation between PCTs. Outer London PCTs had higher proportions of home deaths and inner London PCTs higher proportions of hospice deaths. Funnel plots identified consistent high outlying performance for home, hospice and hospital deaths. No PCT showed a change of 10% or more in home deaths, but five showed decreasing hospital deaths and three increasing hospice deaths. Maps and funnel plots appear useful for identifying areas with differing performance for home, hospital, nursing home and hospice deaths. These methods may help further investigation of how local services may successfully support deaths outside hospital.
More Related Videos
Related Concept Videos
Cancer Survival Analysis
Comparing the Survival Analysis of Two or More Groups
Statistical Methods for Analyzing Epidemiological Data
Kaplan-Meier Approach
Life Tables
Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...

