Cancer is a serious public health problem. Various hospitals based survey is conducted in various parts of the world. But the importance of community based survey is more important? To proceed survey, one step is the calculation of sample size?
Sample size depends on the type of study proposed.
For example at one extreme, total population incidence registers in place in many countries for general cancer and specific types do not need to consider sampling at all.
For studies based on sampling, standard normal,binomial or Poisson theory can be used to estimate sample size required for sufficient precision of prevalence/incidence/mortality estimates or power of statistical tests of comparisons made.
an excellent book which covers all types of cancer research in developed and underdeveloped countries is :
Cancer Epidemiolgy Principles and Methods by Isobel des Santos Silva.
IARC WHO 1999
Hello you can check this very helpful manual by World Health Organization which include very clear explanations of the requirements you need according to the study design: http://apps.who.int/iris/handle/10665/40062
For surveillance you need something close to a cohort study (cohort studies are always observational).
-If you just want to measure how frequent cancer is, then you need a descriptive cohort. In descriptive studies, you usually set the sample size based on the desired precision for the study.
-If what you want is to detect changes in the incidence, then the sample size is based on several parameters, like how much the incidence changes from one period to the next and how much difference is relevant to detect as significant.
When you set up a surveillance system, you normally first measure frequency, then you look for changes.
- Badges
- Science topic
- Similar topics
- Medicine
- Public Health