Safety of care refers to the ability of the system to provide necessary care without causing harm to the patient, like e.g. complications or infections caused by a treatment or other procedure, harmful drug interactions, etc. In this section, we have assessed this dimension through indicators relating to :
- Healthcare-related infections (QS-1, QS-2, QS-7, QS-8)
- Healthcare providers’ perception of patient safety (QS-9, QS-10)
The indicator on polypharmacy (taking at least 5 medications at the same time, which comes with an increased risk of interactions) previously included in this section has been moved to the section on care for older people (OLD-13). In this same section, the indicators on the incidence of bed sores (OLD-8) and falls (OLD-7, which can be linked, among other causes, to medication intake) in residential care facilities are also closely related to safety of care, as are the indicators on inappropriate prescription of antipsychotic and anticholinergic medications (OLD-10 to OLD-12).
Other indicators relating to safety can be found in the section on adequacy (indicators on antibiotics use (QA-3 to QA-5), use of medical imaging (QA-6) and caesarean section (QA-8)), accessibility and sustainability (e.g. indicators on nurse staffing and education levels (A-11, S-9).
Healthcare-associated infections (QS-1, QS-2, QS-7, QS-8)
People normally go to hospital to have their health issues diagnosed, treated and hopefully resolved. However, matters do not always go so smoothly and some patients might also acquire a new infection during their stay. In a world where an increasing number of germs are resistant to existing (antibiotic) treatments, these healthcare-related infections can pose a serious threat to patient safety, sometimes causing severe complications or even death. This also makes them a significant financial burden for the community.
In Belgium, infections caused by methicillin-resistant Staphylococcus aureus and multi-resistant Escherichia coli (a bacteria found in the gut) are subject to compulsory surveillance by hospitals since 2006 and 2015 respectively. This surveillance data is centralised by a dedicated unit within Sciensano.
Monitoring these two germs is important for a variety of reasons. Aside from the fact that they are among the most frequent causes of healthcare-related infections, they are also considered to be primary indicators of antibiotics resistance in humans. Surveillance of MRSA is also a good indicator of the effectiveness of measures aiming to prevent or control the spread of infections like hand hygiene, use of protective equipment, etc.
Results
-
Prevalence of healthcare-associated infections in hospitalised patients (QS-1)
Note: Caution is warranted for two reasons when comparing the new figures for this indicator (based on the 2022 data) to the previous ones (2017). First, the 2022 figures also include healthcare-associated infections imported from long-term care facilities (like e.g. nursing homes) to acute care hospitals, while the 2017 figures included only healthcare-associated infections that were acquired in hospitals. Second, the 2022 figures also take into account the healthcare-associated COVID-19 infections.- In 2022, on any given day, an estimated 9.2% of patients in Belgian acute care hospitals had one or more healthcare-associated infections. When excluding healthcare-associated infections acquired in long-term care facilities, this figure was a little lower (8.5%), but still higher than in 2017 (7.3%) and 2011 (7.1%).
- The three most frequent healthcare-associated infections in acute care hospitals were pneumonia (approximately 22%), urinary tract infections (18%), and surgical site infections (14%). COVID-19 came 6th with a little under 9% of cases. About one in five (21%) of these infections was linked to a device introduced in the patient’s body (invasive device).
- The most common healthcare-associated infections acquired in long-term care facilities were pneumonia (36%), urinary tract infections (19%) and blood stream infections (12%).
- In total, a little under 900 different germs were reported in patients with a healthcare-associated infection in the participating facilities. The most frequent were E. coli (17%), S. aureus (11%; see QS-2 below for more information) and K. pneumoniae (10%).
- Figures on healthcare-associated infections are difficult to compare between countries, because they can be influenced by a lot of factors (e.g. the number of facilities participating in the survey, differences in their patient populations, differences in the way infections are diagnosed and reported, etc.). However, Belgian figures remain above the European average of 7.1%.
- Belgian figures are higher than might have been expected and show a deteriorating trend, which warrants further investigation and prevention efforts.
Link to the technical sheet and detailed results
-
Incidence of healthcare-associated MRSA infections in hospitals (QS-2)
Staphylococcus aureus is a common bacteria which frequently causes skin, mucosa and (postoperative) wound infections but can also affect a wide range of other organs. Over time, a number of strains have become resistant to common antibiotic treatments; they are known as methicillin-resistant Staphylococcus aureus or MRSA. This indicator focuses on the incidence of MRSA infections per 1000 hospital admissions.- The incidence of healthcare-associated MRSA infections has been decreasing fairly consistently in acute care hospitals in all Belgian regions since 2005, falling to 0,3/1000 admissions in 2021. This improvement is probably linked to a better uptake of the recommendations to control the infection and a more rational use of antibiotics. However, it should be mentioned that screening practices vary among hospitals, which might influence the results.
Note: the obligation to collect surveillance data was temporarily lifted during the COVID-19 pandemic. Along with the disruption of hospital activities, this may have influenced results for 2020 and 2021.
Link to the technical sheet and detailed results
-
Percentage of infections caused by MRSA (among all S. aureus infections) (QS-7) and by 3GC-resistant E. coli (among all E. coli infections) in acute care hospitals (QS-8)
These indicators focus on the percentage of S. aureus and E. coli infections which are caused by treatment-resistant strains; those common germs are considered to be primary indicators of antibiotics resistance in humans in general.- In 2021, 9.1% of S. aureus infections and 8.2% of E. coli infections in Belgian acute care hospitals were caused by resistant strains.
- After an overall steady decrease since 2004, the proportion of resistant S. aureus (MRSA) in Belgian acute care hospitals slightly increased between 2015 and 2016 and, more recently, between 2020 and 2021.
- Of those infections, 23% were considered to be healthcare-associated. Overall, the percentage of MRSA infections that are specifically associated with care has been decreasing since 1994, falling from over 90% to less than 25% for Belgium. This trend was observed in all regions.
- In 2021, the proportion of S. aureus and E. coli infections caused by resistant strains in Belgium was relatively low compared to other European countries.
Link to the technical sheet and detailed results
Note: the obligation to collect surveillance data was temporarily lifted during the COVID-19 pandemic. Along with the disruption of hospital activities, this may have influenced results for 2020 and 2021.
Perception of patient safety among healthcare workers (QS-9, QS-10)
Under otherwise similar conditions, a nurse tending to 15 patients or working long hours is obviously more likely to drop the ball than one who only has 8 people under his or her care and works shorter shifts, and the impact of staffing levels on patient outcomes is as unsurprising as it is well-demonstrated. However, the way healthcare professionals perceive overall patient safety and staffing levels at their hospital is also indicative of the attention the facility devotes to patient safety, i.e. its patient safety culture (PSC).
This PSC has recently been identified as an essential component in creating and maintaining safe healthcare systems and is associated with positive outcomes for both patients and staff, like increased satisfaction, better health outcomes and improved productivity. In times of crisis, like the recent COVID-19 pandemic, a better PSC had been shown to be linked to better adaptability and resilience.
Assessing the safety culture of hospitals is also a crucial step towards improving patient safety. The Belgian government has run multiple programmes on this topic since 2007, and incentives were introduced in 2018 through a pay-for-performance mechanism.
Results
-
Average percentage of health workers who think that staffing levels in hospitals (acute, psychiatric, and rehabilitation) are sufficient to handle the workload and work hours are appropriate to provide the best care for patients (QS-9)
This indicator is based on the “staffing” dimension of the Hospital Survey on Patient Safety culture, which is assessed through 4 items/questions: “we have enough staff to handle the workload”, “staff in this unit work longer hours than is best for patient care”, “we use more agency/temporary staff than is best for patient care” and “we work in “crisis mode”, trying to do too much, too quickly”.- Overall, between 2010 and 2022, a little less than half of surveyed healthcare workers in Belgium believed that the staffing levels at their workplace were appropriate for ensuring patient safety.
- The perception of staffing levels in hospitals tends to vary considerably depending on the category of health worker but also from year to year for a specific professional group. This is probably due to differences in sample sizes, which went from a few dozen responders in some years to tens of thousands in others. Due to this variability, overall trends across years and professionals probably sketch a more reliable image.
- For 2019, Belgian results for this indicator (43% based on fairly large sample of 27 hospitals and 9050 surveyed professionals) were very slightly above the OECD-13 average.
Link to the technical sheet and detailed results
Data source: Belgian Hospital Survey on Patient Safety Culture (v 1.0), Hasselt University
Note: The size and composition sample of hospitals and health workers varied from year to years.
Data source: OECD Patient Safety Culture Pilot Data (2020-2021)
Note: The size and composition of the sample of hospitals vary between countries.
-
Average percentage of health workers who have positive overall perceptions of patient safety in hospitals (acute, psychiatric and rehabilitation) (QS-10)
This indicator is based on the “overall perceptions of patients safety” dimension of the Hospital Survey on Patient Safety culture, which is assessed through 4 items/questions: “patient safety is never sacrificed to get more work done”, “our procedures and systems are good at preventing errors from happening”, “it is just by chance that more serious mistakes don't happen around here”, “we have patient safety problems in this unit.”
-
- In 2022, 58% of surveyed healthcare workers in Belgium had a positive perception of patient safety in their hospital. This percentage has remained relatively stable around 50% on average between 2011 and 2021; overall, variation between professional groups was limited.
- For 2019, the average percentage of Belgian healthcare workers who had a positive perception of patient safety (47%) was lower than the OECD-13 average (53%).
Link to the technical sheet and detailed results
Data source: Belgian Hospital Survey on Patient Safety Culture (v 1.0), Hasselt University
Note: The size and composition of the sample of hospitals and health workers varied from year to year.
Data source: OECD Patient Safety Culture Pilot Data (2020-2021)
Note: The size and composition of the sample of hospitals vary between countries.
Patient-to-nurse ratio in general nursing units (secondary indicator of QS-9 and QS-10)
In 2019, the overall patient-to-nurse ratio in the 84 Belgian hospitals that responded to the survey was 9.4. This is a slight improvement compared to 2009 (10.7), but still remains well above the level that is considered safe (8.0).