Thesis:

Factors associated with all-cause in-patient mortality between 2018 and 2019 at Kisumu county level four hospital, Kenya

A lack of demand for the analyzed mortality data has contributed to dwindling data management on inpatient mortality. Due to insufficient information on patient mortality, healthcare planners rarely use local data in resource allocation and hospital management. This results in missed opportunities to build hospital capacity to address common causes of death, and also results in a poor hospital reputation, fewer patients seeking hospital care, increased medical errors, and increased inpatient mortality. Deaths due to medical error are unmeasured and prevention discussions take place in limited and confidential forums such as hospital death conferences or hospital mortality meetings. The main aim of this study was to determine factors associated with all-cause in-patient mortality between 2018 and 2019 at Level Four Kisumu County Hospital, Kenya. Specifically, the study sought to establish patient related factors, trends, institutional and patient care factors associated with hospital mortality. The study was a cross sectional prospective study in which files of patients who died between January 2018 and December 2019 were randomly picked and examined across medical, pediatrics, surgery and reproductive health department. Sample size was determined using Yamane Taro formula which yielded 203 as sample size. The findings of this study suggest that 3% of deaths that occurred at Kisumu County level four hospital between 2018 and 2019 were avoidable and were contributed to by unavailability of supplies, patient care and existing medical condition of patient. According to the current study, medical ward had the highest 2-year in-hospital mortality at 13.86% while obstetrics and gynecology (reproductive health) had the least mortality at 0.47%. The current study found that infections caused 42% of deaths in patients aged below 35 years while noncommunicable diseases caused 41% of hospital deaths in patients aged >60 years. The study found that there was a significant difference in the odds of a patient dying when a nurse and a doctor were all present (OR=0.697) meaning that presence of doctor and nurse protected the patient from death. Comorbidity was a significant factor associated with death among the patients who died in 2018 and 2019 (p=<0.05). Patient related factors such as age, education level and gender were not significantly associated with hospital deaths (p>0.05). This research clearly illustrates that hospital deaths among aged people is due to noncommunicable disease and deaths among young populations is due to infectious diseases but also raises the question of the need to enhance nurse-doctor relationship to reduce avoidable deaths among patients admitted in hospitals. This study was only conducted in one hospital and hence further research is needed to cover more hospitals before generalized conclusions can be drawn.