Measures of patient safety culture from the perspective of health workers can be used – along with patient-reported experiences of safety, traditional patient safety indicators (see indicator “Safe acute care – surgical complications and obstetric trauma”) and health outcome indicators (see, for example, indicator “Mortality following acute myocardial infarction”) – to give a holistic perspective of the state of safety in health systems.
A positive patient safety culture for health workers results in shared perceptions of the importance of safety, increased transparency and trust, and higher levels of shared responsibility, along with improved confidence in organisational and national safety initiatives. A growing body of research has found that positive patient safety culture is associated with a number of benefits, including better health outcomes and patient experiences, as well as improved organisational productivity and staff satisfaction (de Bienassis et al., 2020[36]). Improved models of patient safety governance and investment in improving the patient safety culture have a substantial and lasting impact on outcomes (G20 Health & Development Partnership, 2021[34]). Figure 6.25 illustrates two domains of the Hospital Survey on Patient Safety Culture, which asks hospital staff to provide information on aspects of their work environment and whether they are conducive to good patient safety. The safety of handoffs and transitions relates to staff perceptions of whether important patient care information is transferred across hospital units and during shift changes. Positive perceptions from staff on safety of handoffs and transitions range from 54% in Slovenia, to 32% in Belgium and Scotland (United Kingdom). On average across OECD countries, fewer than half of the hospital staff surveyed thought that handoffs and transitions were sufficient. Figure 6.25 also shows that just over half of health workers had positive overall perceptions of patient safety – meaning that staff think the procedures and systems at their workplace are good at preventing errors and that there is a lack of patient safety problems (OECD, forthcoming[37]).
Patient perspectives are also critical to make health systems more safe and people‑centred. Given this importance, the OECD developed a pilot survey instrument to measure patient-reported experience of safety (OECD, 2019[38]), and several OECD countries have tested this instrument. To strengthen health systems based on people’s voices, a few OECD countries have started utilising patient-reported safety indicators systematically. For example, Poland uses them as part of its provider accreditation mechanism, and Germany uses them as one of the inputs for an incident reporting project.
According to the Commonwealth Fund 2020 survey, the proportion of patients reporting experiences of medical mistakes in the past two years varied between 3% in Germany and 13% in Norway in 2020. Among hospitalised patients, the proportion was 5% in New Brunswick (Canada) and Estonia and 9% in Poland (Figure 6.26). New Brunswick (Canada) and Poland assessed the comparability of patient-reported incident rates and found that patient-reported data were comparable to the data collected in medical records. However, it should be noted that neither data source may capture certain types of harm. Patients may not report physical harms if they are not immediately recognisable (unlike pain and infection) and if they are not informed of their occurrence by a provider. Medical records may not include harms such as miscommunication, distress and worry, although responding to patients’ information and emotional needs is essential for delivering safe and people‑centred care.
Among different types of patient safety incident, medication-related incidents are most frequently reported across countries. The proportion of people who reported wrong medication or wrong dose given by a doctor, nurse, hospital or pharmacist in the past two years ranged from 3% in Australia to 7% in Norway (Figure 6.27). In Poland, 3% of hospitalised patients reported medication-related incidents. These data need to be interpreted with care: they may be underreported because patients may not know about all cases of medication error.
