Assessment of the quality of cause-of-death data in Greenland, 2006-2015

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review


  • Kim Moesgaard Iburg
  • Lene Mikkelsen, University of Melbourne, Australien
  • Nicola Richards, University of Melbourne, Australien

Background: While the system of registration of mortality and cause of death (COD) in Greenland was established several decades ago, reporting procedures follow a complicated administrative process. Timely and reliable reporting on mortality and COD is of high importance for the usability of the collected data for research, health planning and decision making. Methods: COD data collected by the Chief Medical Office in Greenland from 2006 to 2015 (4490 registered deaths) were analysed with the software Analysis of National Causes of Death for Action (ANACONDA) v4.0. Unusable or insufficiently specified ICD codes are identified. The Vital Statistics Performance Index for Quality (VSPI(Q)) is estimated for the overall quality conclusions of the register's usability. Results: Sixty-eight per cent of the input data for Greenland was coded with a usable underlying COD, 24% with an unusable cause and 8% of deaths with an insufficiently specified cause. Almost 700 deaths were coded to an unusable code of 'very high impact'. The most prevalent unusable underlying causes were other ill-defined and unspecified causes, including no death certificate available, followed by senility, heart failure, sepsis and shock and cardiac arrest. The VSPI(Q) score was 66%, representing medium quality. Conclusions: In the 10 years' worth of data analysed, the true underlying COD in many cases was unknown. Several likely explanations for this include lack of systematic COD training for physicians, logistic and capacity challenges in Greenland that potentially could reduce the quality of the collected data and its usability in providing essential information about the true pattern of mortality in Greenland.

BogserieScandinavian Journal of Public Health
Sider (fra-til)801-808
Antal sider8
StatusUdgivet - dec. 2020

Se relationer på Aarhus Universitet Citationsformater

ID: 175726597