Provider Based Costing Sampling

Asked: 31 Jul 2018 | 991

I am conducting research in health centers. Two types of costing providers, center specific and non center specific. What criteria do I have to choose for one health center to be representative of others or what number of health center should I use to make it representative?

Costing

Expert Replies:

Marufa Sultana

Research Investigator | Bangladesh | Replied: 08 Aug 2018 at 07:30

Hi Yigezu,
I agree with the others as the selection of the center is solely depends on the objective of the research. As such, if it is a multicenter study, you can select center(s) based on the geographical region, or hierarchical level (according to the health system of the respective country), public-private or any other according to the study objective.

Dr Abdur Razzaque Sarker

Health Economist | Bangladesh | Replied: 03 Aug 2018 at 21:26

I, agree with professor Wanrudee as the sampling criteria and even sample size depends on the objective of this research. In addition to the geographical region, types, and size of health center we should think about the available health care services provided by the health centers and even the catchment area., i.e., healthcare coverage.

Wanrudee Isaranuwatchai

Assistant Professor | Canada | Replied: 02 Aug 2018 at 14:28

This is a great and important question. Thank you for the question.

To learn more about your scenario, is collecting data at all health centres an option? If yes, then you could consider that option in order to comprehensively provide costing estimates. Or if your project objective is to provide a comprehensive review of provider-based costing at all health centers, then you could consider the plan to collect data from all sites.

If your objective is to provide an average/sample costing estimate or if you have to select certain number of health centers, then there are a few approaches you could consider in choosing health centers to ensure that they are representative to your objective. For example, you could select health center(s) by geographical regions, by rurality (urban vs rural areas), by type of health center (e.g., provincial level vs district level), by size of health center, or by other criteria/category you are interested in.

Deciding the number of health center to collect data from would depend on your objective and feasibility (e.g., resources available) in term of data collection.

Hope this helps. Please let us know if you have additional questions.