Hospital performance

Health spending on average represents 8.8% of GDP in OECD countries, ranging from 4.2% in Turkey to 17.1% in the United States. Hospitals account for nearly 40% of health spending. In most OECD countries, more than 80% of this expenditure is covered by government and compulsory schemes. Hospitals are thus important targets in national efforts to improve health system performance.

The OECD currently collects a number of acute care measures of hospital performance on a national level. The mortality rate within 30 days of patients being admitted to hospital after an acute myocardial infarction (AMI) is one key hospital performance measure. The measure reflects processes of care such as timely transport of patients and effective medical interventions. This indicator is also influenced by differences in hospital transfers, average length of stay, and AMI severity. Several effective strategies to lower AMI mortality rates in OECD countries have been identified.

‌‌ Hospital-performance

HOSPITAL PERFORMANCE PROJECT

The OECD launched the Hospital Performance Project in 2015 to understand hospital performance across countries and to strengthen international comparisons. The project built on international initiatives such as the ECHO and euroHOPE Projects and expanded these approaches to all OECD countries.

Following two years of methodological work, risk adjusted, hospital-level 30-day AMI mortality rates were collected for over 3,000 hospitals across 12 OECD member countries for the years 2013 to 2015. Additional information on this methodology including risk adjustment and indicator definitions are available in documents included in the Further reading section below.

The scope and scale of this data collection are groundbreaking, being the first time hospital-level performance data have been collated and compared across OECD countries in America, Europe and Asia. Results are published in the OECD’s flagship publication Health at a Glance 2021: OECD Indicators (see Chapter 6 on Quality and Outcomes of Care).

The OECD is now expanding the hospital performance work beyond AMI mortality to include a focus on stroke mortality.

EXPLORE THE DATA

Unlinked data

National and hospital level data on 30-day AMI mortality using unlinked data along with hospital characteristic data are available in the data visualisation and data link below. Hospital-level rates are based on aggregated data from 2013-2015 and are risk-standardised for age, sex, and co-morbidity. Rates based on unlinked data focus on single hospital admissions and use these admissions as the denominator to calculate rates. These calculations only consider deaths occurring in the hospital where the admission occurred. These rates may be lower than those using linked data (below) because multiple admission may be counted for the same patient. Data for download include national and hospital mortality rates as well as hospital characteristics.

Linked data

National and hospital level data on 30-day AMI mortality using linked data along with hospital characteristic data are available in the data visualisation below. Hospital-level rates are based on aggregated data from 2013-2015 and are risk-standardised for age, sex, and co-morbidity, and previous AMI. Rates based on linked data focus on each AMI patient and use these as the denominator (regardless of the number of admissions or readmissions) to calculate rates. These calculations consider deaths occurring anywhere including inside or outside the admission hospital. These rates may be higher than those using unlinked data as a single patient is only counted once in the denominator. Data for download include national and hospital mortality rates as well as hospital characteristics.

A‌ccess the data visualisation / Download the underlying dataset

While the linked data based method is considered more robust, it requires a unique patient identifier to link the data across the relevant datasets which is not available in all countries.

Health policies and data

Hospital performance