You can see the data accumulated by the Centers for Disease Control and Prevention, National Vital Statistics System in an article entitled, "Age-Adjusted Death Rates for Selected Causes, Death Registration States, 1900-32, and United States, 1933-98." Likewise, it would be a good idea for genealogists to understand the changing mortality rates. For example, see the following article.
Corsini, Carlo A. and Pier Paolo Viazzo ed., The Decline of Infant Mortality in Europe, 1800-1950, Four National Case Studies, United Nations Children Fund, International Child Development Center, “Hisper_decline_infantmortality.pdf.” 2016. Accessed March 31. http://www.unicef-irc.org/publications/pdf/hisper_decline_infantmortality.pdf.
Seriously, genealogists need to be aware of such things as the infant mortality rate in any given century and country and the overall life expectancy rate of a particular population and time. For example, the overall life expectancy of the average person living in England in the early modern era was about 30 years. If a person lived to the age of 21, their life expectancy increased to about 70 years. See Wikipedia: Life expectancy. It is not uncommon for genealogical researchers to accept as valid dates that are outside the normal range of mortality. This lack of awareness also extends to dates that are entirely unrealistic from the standpoint of reality. For example, mothers giving birth in early childhood and children born long after one or both parents have died. It may well be that some of these readily apparent errors are the result of transposition errors or similar typographical problems, but the persistence of these errors across generations in online family trees and the same errors copied from tree to tree, all indicate a lack of awareness on the part of the researchers.
Other factors that must be taken into consideration are the existence of epidemics and other natural disasters that can adversely affect mortality rates for relatively short periods of time. For example, the world-wide flu epidemic that is generally associated with the year 1918 resulted in the death of from 50 to 100 million people, approximately 3% to 5% of the world's population. All of us alive today had ancestors that lived through or died in this world-wide event. It is possible that unexplained deaths that occurred from 1918 to 1920 could have been caused by a flu infection.
One standard method of approaching the dates when children are born into a family when gaps of years are noted in the succession of children in a family is to assume that there are missing children who may have died in infancy. This is a rather safe assumption given the infant mortality rates across the world. If you are aware of the mortality rate for the time periods during the lives of your ancestors, you can plot all the ages on a graph and see if there are any days that are significantly outside of the norm. Any such outlying ages should then be re-evaluated and further supported by research. In short, genealogists should always maintain a sense of historical consistency and reality.