If the 25 worst-performing hospitals in Belgium were to perform as well as the median hospital in terms of mortality, over 4,000 deaths could be avoided annually, according to a large-scale study by the Leuven Institute for Healthcare Policy.
The study analysed data from 99 Belgian hospitals, and about 16 million patients, from 2008 to 2018. Doctoral researcher Astrid Van Wilder looked at the quality of hospitals, based on five parameters: hospital mortality rate, number of re-admissions, length of stay, complications, and patient satisfaction.
The researchers recorded a slightly positive evolution on the majority of the parameters. The average mortality fell from 3.4% to 3.1%, the total length of stay decreased by one day to an average of 6.5 days. Patient satisfaction also increased: in 2008, 56% of patients gave their hospital experience a score of 9 out of 10, but ten years later this rose to 61%.
Failure to rescue
However, the results showed major differences between hospitals, with the mortality rate varying between an average of 2.1% and almost 5%. If the 25 worst-performing hospitals were to perform as well as the median, more than 4,000 deaths per year could be avoided.
In terms of 'failure to rescue' (when a patient dies due to an easily treatable complication, such as pneumonia or gastrointestinal bleeding) some hospitals register up to four times more deaths than others. For example, after gastrointestinal bleeding, the risk of dying is five times higher in a poorly performing hospital, the study found.
Strikingly, the region, hospital size and university status could not explain the differences in variation. Patient characteristics were also included: groups with the same profile were compared, so that the explanation cannot lie with, for example, the patients' old age or underlying diseases.
Credit: Belga / James Arthur Gekiere
"It is a complex puzzle," Professor Kris Vanhaecht (KU Leuven), the promoter of the research, told Belga News Agency. "A combination of parameters may play a role. Socio-economic factors, the organisation and context of the hospital, and the atmosphere and cooperation within a team may also explain the variation. But this requires additional research."
For Vanhaecht, a solution to the major differences between hospitals lies in greater transparency and cooperation, which allows hospitals to look at each other critically and learn from each other's experiences.
The increased attention to quality assurance has now also resulted in the Flanders Quality Model consortium (FlaQuM): 25 hospitals are now affiliated and work together to provide sufficient quality care. Now, the researchers want to continue processing more recent data, from after 2018 and relating to the Covid-19 period.
