European comparisons - Background and data sources

General issues of comparison

Individual countries define and report workplace injuries differently. As a result, the statistics of injury in some countries are a by-product of reports made by employers (and others) to National Labour Inspectorates. Examples include the UK, Ireland and Denmark. In France, Germany, Spain, Italy, Portugal and Greece the main source of statistics are claims made through insurance or social security systems. In the Netherlands statistics are derived from a mix of employer reports and social security claims. A feature of insurance and social security systems is the relatively small level of under-claiming so that injury statistics are nearly or fully complete. In contrast, there is a considerable level of under-reporting of non-fatal injuries in systems based on employer reporting.

The published figures on workplace injury for most countries include accidents to people who are working, injured in road traffic conditions or by road vehicles. Statistics for Britain and Ireland exclude most such accidents which can account for between 20 and 40% of workplace fatalities in other member states.

Non-fatal injuries are reportable in most member states on the basis of absence from work being either more than three days or more than one day. These categories include any injuries specifically defined as "serious". For example, the rate of over 3 day injury in the tables for Great Britain includes major injuries as defined in the national reporting regulations.

Harmonisation project of Eurostat

Eurostat and member states are working on a project to produce and publish comparative European Statistics of injuries At Work (ESAW). The project shares some of the methodology of an HSE Study but improves the consistency of definitions of an accident at work, defined as "a discrete occurrence in the course of work which leads to physical or mental harm". This includes accidents in the course of work outside business premises and cases of acute poisoning. Occurrences having only a medical origin (such as a heart attack at work) or occupational illnesses are excluded. The data collected covers the nine branches of activity that are common to all member states; the nine branches are agriculture, manufacturing, utilities, construction, retail and wholesale distribution, hotels and restaurants, transport, financial services and real estate business activities. Since each member state may also collect information on other sectors, the individual figures for each member state quoted above may differ from those published at a national level.

The main proposals for ESAW are for member states to provide Eurostat with data on accidents which: exclude commuting accidents; are based on the over-3-day criterion for absence; identify road traffic accidents; cover the same industries; and cover all types of employment. 

Member states also provide estimates of the reporting level of injuries where notification systems are based on employer reporting or where underreporting is known to exist.  The British statistics of workplace injury, based on employer reports and the results of the Labour Force Survey, meet these key proposals.

Eurostat’s published rates of fatal and non-fatal injuries in Table 1 and Table 2 cover 9 branches covered by the injury notification systems in all member states. The sectors are: agriculture (with hunting and forestry), manufacturing, utilities, construction, retail and wholesale distribution, hotels and restaurants, transport, financial services, and real estate activities. These are categories A, D, E, F, G, H, J and K of the Standard Industrial Classification 1992.

Eurostat rates reproduced here are rounded for convenience. Eurostat has adjusted the rates for Sweden, the Netherlands, Ireland, Denmark and Great Britain to allow for the known under-reporting of injuries in those countries. The adjustments for Great Britain and Ireland are based on the results of the National Labour Force Surveys.

Different countries have different structures of employment in industry sectors. As a result, the rates of fatal and non-fatal injury for all industries combined in member states would not be comparable because some member states will have proportionally more workers in high risk industries. Eurostat has allowed for this by standardising the rates of fatal and non-fatal injury of each member state to a common basis of the EU industry mix of employment. Table 3 shows the rates of fatal and non-fatal injuries by the employed people covered within the schemes for individual countries.

Data from Eurostat also indicates the industry sectors of highest and lowest risk. Table 4 shows the EU average rate of fatal and of over 3 day injury in a number of industries, derived by Eurostat, together with the British rate of fatal and over 3 day injury for comparison. Eurostat included road traffic accidents to show the full risk of workplace injury. This should not affect the comparison where the EU average is substantially higher. Eurostat has not published comparative rates of fatal injury within published industries for individual member states.

Eurostat is the source for the EU average rates of fatal and over-3-day injury. The source of rates of fatal and over 3 day injury for Great Britain are the annual publications of ‘Health and Safety Statistics’ and ‘Statistics of Fatal Injuries’. The rates of fatal injury are based on reports from employers and others, and the rates of over 3 day injury are derived from the Labour Force Survey.

HSE Study - Great Britain, France, Germany, Spain and Italy

In 1991 HSE undertook its own analysis of injury statistics of large member states: France, Germany, Italy and Spain. The study adjusted the national injury statistics to bring them to a broadly comparable basis with Great Britain. For example, the study adjusted the national statistics to exclude road traffic or vehicle accidents, and to derive incidence rates for the five countries in the same way. The study also considered, mainly non-quantifiable, factors such as biases due to under or over claiming in national insurance systems. The study concluded that, despite the differences in statistical systems, rates of fatality and of injury, for all industries combined (and within most main industry sectors), are lower in Great Britain than in other large member states). Data for the most recent five-year period is in Table 5 and Table 6.