Related Experiment Videos
Has there been progress in cancer care in Croatia? Assessing outcomes in a partially complete mortality follow-up
A Znaor1, H Brenner, B Holleczek
1Croatian National Cancer Registry, Croatian National Institute of Public Health, Rockefellerova 7, 10000 Zagreb, Croatia. ariana.znaor@hzjz.hr
Background:
We examine the possibility of assessing progress in cancer care with partially complete mortality follow-up information, and report outcomes from Croatia.
Methods:
Follow-up based on death certificates indicating cancer as the cause of death was available from the Croatian National Cancer Registry. The effect of partially complete follow-up was first examined with data from the Saarland Cancer Registry by comparing absolute, relative, and cancer death certificate based survival estimates. Survival changes between 2000 and 2006 are reported for 21 common cancers amongst patients aged 15-49 and 50-59 in Croatia.
Results:
Survival estimates based on cancer death specific follow-up could well approximate absolute and relative survival for patients aged 15-49, and relative survival for patients aged 50-59: overestimation by more than one standard error occurred 1 and 2 and 5 times, respectively, amongst 21 cancers. In Croatia, significant survival increases occurred for patients aged 15-59 with colorectal and breast cancers, patients aged 15-49 with thyroid cancer and patients aged 50-59 with malignant melanoma and prostate cancer.
Conclusions:
Outcome evaluation is limited with partially complete follow-up information. Internationally comparable cancer information continues to lack from South-Eastern Europe, and the provision thereof remains a highly important public health task.
Related Concept Videos
Cancer Survival Analysis
Kaplan-Meier Approach
Comparing the Survival Analysis of Two or More Groups
Longitudinal Research
Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
Assumptions of Survival Analysis