Introduction:

Vital statistics are essential to identify the health conditions of a population and design interventions. Vital statistics are essential to understand population health and design public health interventions. The accuracy and validity of the derived metrics and the relevance of public policy decisions depend on its quality.

Objetive:

To evaluate the quality of the vital statistics of children under five years of age in Colombia, between 2000 and 2018.

Methods:

A descriptive and exploratory study was carried out and the quality of records of live births and deaths published by DANE was analyzed using the criteria suggested by the United Nations: accuracy (precision), opportunity, and coverage (completeness).

Results:

Of the death records, 8.2% show codes useless for public health purposes as underlying cause of death. Of the live births, 97.8% were registered during the same year of occurrence. We estimated the underregistration of infant mortality at 28.9%, with differences between territories.

Conclusions:

The quality of vital statistics of children under five years of age has improved in Colombia in the 19 years analyzed, particularly in accuracy and opportunity. However, problems persist in assigning the underlying cause of death and in underregistration, with significant differences between territories. Improving the quality of the vital statistics system must be an ethical imperative, especially to understand the populations of geographic areas historically invisible in the country.