Causes of Death at the University of Ghana Hospital in Accra - A 37 Year Review (1979-2015)

Abstract

The aim of the study was to conduct an analysis of the causes of death at the University of Ghana Hospital, Legon (UGHL) from 1979 to 2015. The objectives were to conduct cross-sectional and temporal analysis of the causes of death and also evaluate the quality of the data. Secondary data from medical cause of death certificates was extracted and analysed using SPSS. Diseases were grouped into three main categories and analysed by age, gender and time in years. Of 3263 deaths recorded, 45.2% were adults 60 years and above and 37.7% were adults aged 19-59 years. Male deaths (58.06%) were higher than female deaths (41.94%). Findings revealed a common pattern of yearly deaths decreasing till 1991-1995 after which they increased. It revealed deaths in adults older than 19 years are increasing whilst those in younger age groups are declining. Almost 60% of all deaths were caused by non-communicable diseases. HIV/AIDS deaths rose sharply from 1996.Within the last decade, deaths by Malaria, Tuberculosis and Diarrheal diseases in individuals’ ≥ 18 years has declined. In addition, deaths from malaria and diarrheal diseases in individuals < 18 years of age declined. Deaths by Tuberculosis in individuals < 18 years of age have been zero since 1985. Measles ceased to be a cause of death after 1990.Tetanus was not a cause of death after 1985 until it re-emerged in 2011-2015. A shift from communicable to non-communicable diseases was observed within this period, depicting a dual burden of disease. Amongst the top causes of non-communicable deaths were cardiovascular disease and cancer. The UGHL cause of death database was assessed to be 95.95 - 97.3% complete. Non- documentation of time interval between disease onset and death, recording “mode of dying” as the immediate cause of death, the use of non-conventional abbreviations, ill-defined causes of death and improperly written cause based sequence were some of the data quality issues identified. In conclusion, generally t h e d e s c r i p t i v e findings at the UGHL are quite similar to those of similar research studies conducted in sub-Saharan Africa and the world. It supports well established findings and projections made on the Global burden of Disease and the epidemiological transition. However, the U GHL study setting had exceptional findings relating to the overall disease and age distribution of death. This suggests that targeted public health interventions need to be put in place to deal with this issue. Findings emphasized the need to upgrade physician knowledge on death certification practices to ensure good quality cause of death data. The study has yielded results that may be applicable to the national populace as it is the longest retrospective study of its kind in Ghana.