OECD Interactive Tool: International Comparisons, Sources and Methodology

Printer-friendly version

Back to OECD Tool

Sources

National-level data was obtained from the Organisation for Economic Co-operation and Development (OECD) and The Commonwealth Fund International Health Policy Surveys. Provincial data was obtained from CIHI and Statistics Canada databases using the OECD’s methodology.

The most recent available data was used, spanning the period 2010 to 2014.

Explore details regarding the specific data sources and methodological considerations for each indicator.

Methodology

OECD indicators
34 countries were involved in the OECD Health at a Glance 2015. National-level data was downloaded from the OECD, and provincial-level data was purchased or downloaded from Statistics Canada or calculated by CIHI using the OECD’s indicator methodologies.

Commonwealth Fund Indicators
11 countries, including Canada, participated in The Commonwealth Fund’s 2013 and 2014 International Health Policy surveys. Both the national-level and provincial-level indicators were calculated according to the specifications in The Commonwealth Fund surveys. Data was weighted by age, gender, education, knowledge of official languages and geographic distribution according to census data.

For comparison purposes, Canada’s peer countries were identified. Based on a scan of 26 international studies, peer countries were selected according to size of economy, level of resources to devote to health and data collection systems.

OECD Countries

Australia∗†

Czech Republic

Germany∗†

Israel

Mexico

Portugal

Switzerland†

Austria

Denmark

Greece

Italy

Netherlands∗†

Slovak Republic

Turkey

Belgium

Estonia

Hungary

Japan

New Zealand∗†

Slovenia

United Kingdom∗†

Canada

Finland

Iceland

Korea

Norway†

Spain

United States∗†

Chile

France∗†

Ireland

Luxembourg

Poland

Sweden∗†

 

∗Peer country.
†Country that participated in The Commonwealth Fund surveys.

Methodological Considerations

OECD Indicators
To compare Canadian provinces with OECD countries, CIHI applied the OECD’s indicator methodologies to provincial data. Thus results for provincial indicators in the eTool may not match those for similar indicators published by CIHI in other reports (e.g., Health Indicators).

Here are examples of differences in indicator methodologies that can affect provincial results:

  • CIHI adjusts some indicator results on the basis of risk, including comorbidities, whereas OECD indicators are not risk-adjusted.
  • For some indicators, CIHI uses multiple years of data to reduce year-to-year variability, whereas the OECD uses 1 year of data.
  • CIHI uses data from the Canadian Census of Population to standardize its indicators according to age and sex, whereas the OECD’s indicators are standardized using the population of all OECD countries combined.

These methodological differences limit the comparability of these OECD indicator results with those of other CIHI reports.

Canadian methods that have been specifically designed for regional comparisons in Canada should be used when making interprovincial comparisons.

The Commonwealth Fund indicators
Each year, the target population of interest changes, rotating on a 3-year cycle. For 2013, the target population was the general adult population age 18 and older; for 2014, it was older adults age 55 and older. The general adult population is used most frequently due to larger sample sizes.

Suppression Rules

To maintain privacy and for public reporting purposes, provincial indicator results were suppressed when the underlying count was small or the rates were unstable. The following suppression rules were applied to the indicator results:

  • The indicator rates are suppressed for public reporting if the number of cases is less than 5 (i.e., 1 to 4).
  • The indicator rates are classified as unstable and not reported if adding 1 case to the numerator changes the rate for a province by more than 10% of the national rate.
  • For the cancer mortality rates, some results are suppressed by Statistics Canada to meet the confidentiality requirements of the Statistics Act.
  • For the cancer 5-year relative survival rates, age-standardized estimates are suppressed if
    • 1 or more of the age-specific strata used in their calculation are empty;
    • 2 or more of the age-specific estimates used in their calculation are based on fewer than 5 cases; or
    • 1 of the age-specific estimates is based on fewer than 5 cases and the total number of cases for all age groups combined is less than 100.

The Canadian results include data from all provinces and territories, even if the province-specific result was suppressed. When data was not reportable for 2 or more provinces, 2 years of data were combined, where possible, for all provinces to allow for reporting across most provinces. As a result, 2 years of data were combined for the indicator OECD: Rate of Foreign Body Left in During Procedure.

For additional methodological information, please contact us at hsr@cihi.ca