Mortality: rates and measures

From the MSF book 'Refugee Health; An Approach to Emergency Situations'
When working in emergency situations, there are certain measures of the target population that can assist in determining the severity of a situation. The Crude Mortality rate (CMR) is the most useful indicator in the emergency phase as it alone measures the gravity of the situation and follows its evolution. Calculating mortality rates per 10,000-population/per day enables each situation to be compared against reference values. The expected CMR in a developing country is in the order of 25 deaths/per 1,000 population/per year, i.e. 0.68/10.000/day; but in emergencies it can rise above 1/10.000/day. Sources of mortality data When cemeteries or burial sites are clearly identified, they represent the best source of information on the number of deaths occurring in the community; this system requires the permanent presence of previously trained staff ('grave watchers'), assigned to provide 24-hour coverage and report on the number of daily burials; they should be closely supervised. Medical records in health facilities provide reliable information (especially on cause of death) but this source of information is not sufficiently representative, given that many deaths occur within the home setting. Analysis and transmission of data MORTALITY The crude mortality rate and the under-five mortality rate are the two basic indices for data on mortality (expressed per day); these should be computed and followed over time. There are a number of different indices used for describing mortality and morbidity in public health surveillance. Crude mortality rates
  • total numbers of deaths reported over a given time
  • estimated mid-period population Under-five specific mortality rate
  • number of under-five deaths reported over a given time
  • estimated mid-period under-five population Cause-specific mortality rate
  • number of deaths attributed to a specific cause over a period of time
  • estimated mid-period population Cause-specific mortality proportion
  • number of deaths attributed to a specific cause over a period of time
  • total number of deaths from all causes reported over the same period The comparison of epidemiological indicators, calculated over a given period (week or month), with normal or reference values, constitutes the first phase of data analysis. With regard to mortality and morbidity, it must be taken into account that the reference values were obtained from large populations for use at a national level. When applied to small refugee populations (populations denominator less than 10,000), these references might lead to either an under-estimation or an over-estimation of the situation. This is why surveillance really begins once the trends of these indicators can be followed in time.