© Marcel Crozet / ILO

Occupational Safety and Health Statistics (OSH database)

Table of Contents

The Occupational Safety and Health Statistics (OSH) database includes indicators on fatal and non-fatal injuries at work and labour inspection. 

Occupational injuries


Occupational safety and health at work are vital components of decent work. Statistics on occupational injuries are essential to assess the extent to which workers are protected from work-related hazards and risks. In this regard, indicators on occupational injuries are complementary to those on labour inspection, also included in ILOSTAT.
Given the usefulness of occupational injuries statistics in conveying information on the conditions of workers, this indicator was included as one of the indicators to measure progress towards the achievement of the Sustainable Development Goals (SDG), under Goal 8 (Promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work for all).1SDG indicator 8.8.1 refers to the frequency rates of fatal and non-fatal occupational injuries, by sex and migrant status. For the official list of SDG indicators, see here

ILOSTAT presents statistics on both fatal occupational injuries and non-fatal occupational injuries. In order to assess the work-related risks to which workers are exposed, it is preferable not to look at the gross number of occupational injuries, but rather to present them as a ratio. This is why ILOSTAT features  statistics on the number of cases of fatal and non-fatal occupational injuries per 100’000 workers, disaggregated by sex and migrant status and also by economic activity, compiled from national sources. For users interested in more detailed statistics, ILOSTAT also contains statistics on days lost due to occupational injuries by sex and migrant status and by economic activity.

Concepts and definitions

An occupational injury is defined as any personal injury, disease or death resulting from an occupational accident. An occupational injury is therefore distinct from an occupational disease, which is a disease contracted as a result of an exposure over a period of time to risk factors arising from work activity.

An occupational accident is an unexpected and unplanned occurrence, including acts of violence, arising out of or in connection with work, which results in one or more workers incurring a personal injury, disease or death. A case of occupational injury is the case of one worker incurring an occupational injury as a result of one occupational accident. An occupational injury could be fatal (as a result of occupational accidents and where death occurred within one year of the day of the accident) or non-fatal, with lost work time.

Incapacity for work is the inability of the victim, due to an occupational injury, to perform the normal duties of work in the job or post occupied at the time of the occupational accident. Incapacity can be permanent or temporary. Cases of permanent incapacity for work are cases of occupational injury where the persons injured were never again able to perform the normal duties of work in the job or post occupied at the time of the occupational accident. Cases of temporary incapacity are cases of occupational injury where the workers injured were unable to work from the day after the day of the accident, but were later able to perform again the normal duties of work in the job or post occupied at the time of the occupational accident (within a period of one year from the day of the accident).

The workers in the particular group under consideration and covered by the source of the statistics of occupational injuries are known as the workers in the reference group. In the case of a notification system, it is the number of workers in, for example, the establishments or selected economic activities covered by the system as set out in the relevant legislation or regulations.

Days lost due to temporary incapacity refers to the total number of calendar days during which those persons temporarily incapacitated were unable to work, excluding the day of the accident, up to a maximum of one year. Temporary absences from work of less than one day for medical treatment are not included.

For more detailed information, refer to the Resolution concerning statistics of occupational injuries (resulting from occupational accidents), adopted by the 16th International Conference of Labour Statisticians (October 1998).

Method of computation

Given that relative measures are easier to interpret and favour comparability between countries, activities and over time, ILOSTAT presents statistics on the occupational injuries incidence rate, calculated as follows:

Fatal occupational injuries incidence rate = Number of new cases of fatal occupational injuries during the reference period / Number of workers in the reference group  x 100’000

Non-fatal occupational injuries incidence rate = Number of new cases of non-fatal occupational injuries during the reference period  /   Number of workers in the reference group  x 100’000

Data sources

Statistics on occupational injuries could come from a variety of sources, including various types of administrative records (insurance records, labour inspection records, records kept by the labour ministry or the relevant social security institution), establishment surveys and household surveys.

The recommended data sources for occupational injuries statistics are national systems for the notification of occupational injuries (such as labour inspection records and annual reports, insurance and compensation records, and death registers), supplemented by household surveys (especially in order to cover informal sector enterprises and the self-employed) and/or establishment surveys.

It is worth noting that fatal and non-fatal occupational injuries tend to be notified to and compensated by different agencies, so when using statistics from administrative records, statistics on fatal and non-fatal occupational injuries would very likely come from different records. This means that the sources may have different coverage, and thus fatal and non-fatal occupational injuries, even though very complementary, may not be strictly comparable.

Interpretation and uses

Data on occupational injuries are essential for planning preventive measures. The role of the indicators described in this document is to identify important areas to which attention should be paid. In order to be able to design more targeted prevention mechanisms and related policies, it is recommended to disaggregate and analyse this indicator by sex, occupation, economic activity, or any combination of these. For instance, workers in occupations and activities of highest risk can be targeted more effectively for inspection visits, development of regulations and procedures, and also for safety campaigns.

There may be problems of under-reporting of fatal occupational injuries, and proper systems should be put in place to ensure the production of high quality data. Under-reporting is thought to be present in countries at all levels of development, but may be particularly problematic in developing countries. 

Because data quality issues may be present, it may be more relevant to analyse indicator trends rather than levels. When measured over a period of time, the data can reveal progress or deterioration in occupational safety and health, and thus point to the effectiveness of prevention measures. Indicators on occupational injuries are volatile and large annual fluctuations may occur due to unexpected but significant accidents or national calamities. The underlying trend should therefore be analysed.


The variety of possible sources of data on occupational injuries, which includes administrative records, establishment surveys and household surveys (such as labour force surveys), hinders the comparability of the data across countries, since each type of source provides information on different specific concepts.

Even data coming from administrative records are not strictly comparable, since there are numerous types of records that follow different rules and are maintained by different agencies. Two main sources of data are records of notifications by employers to the competent authority and insurance records of the authority compensating the victims. These two would clearly yield different results, since it is possible that not all injuries that are compensated are reported by the employer and vice versa. It is also possible that these records have a different a geographical coverage or that they cover different economic activities. Ideally, all records pertaining to the same topic kept by different agencies should be linked and/or consolidated (using unique unit identifiers, for example) so that the statistics are truly comprehensive and representative of the country as a whole.

When statistics come from an establishment survey, the results would be closer to those from records of notifications made by employers, since it is also the employer who provides the information. However, establishment surveys tend not to cover the informal sector, nor establishments of a very small size. When statistics come from a household survey (such as a labour force survey), their reliability depends on the accuracy of the respondents. However, if enough questions are used about accidents and injuries to ensure the accuracy of the information, household surveys can be an effective means of obtaining data cross-tabulated by various disaggregations.2For more information on deriving occupational injuries statistics from household and establishment surveys, refer to ILO: “Occupational injuries statistics from household surveys and establishment surveys” (ILO, Geneva, 2012)

It is important to note that there might be a difference in units from source to source: insurance records and notifications records will most likely give the number of cases of injuries (if one worker had suffered from several injuries throughout the year, he/she would appear as many times as the number of injuries suffered), whereas the information from household surveys would refer to the number of persons having suffered from at least one injury (unless the survey reliably collects information on how many injuries each person suffered and the results are summed up).

Labour inspection


Proper application of labour legislation depends on an effective labour inspectorate. Labour inspectors examine whether national labour standards are applied in the workplace and advise employers and workers on how to improve the application of national law in such matters as working time, wages, occupational safety and health, and child labour. In addition, labour inspectors inform national authorities regarding loopholes and defects in national law. They play an important role in ensuring that labour law is applied equally to all employers and workers. Because the international community recognizes the importance of labour inspection, the ILO has made the promotion of the ratification of two labour inspection conventions (ILO Conventions Nos. 81 and 129) a priority.3For more information on International Labour Standards on Labour inspection, see here

Labour inspection statistics play an important role in assisting governments, their ministries of labour and labour inspectorates, in the development of national policies, systems, programmes and strategies for labour inspection. Labour inspection statistics allow governments to observe trends in labour markets and to better analyse compliance issues.4ILO Labour Administration, Labour Inspection and Occupational Safety and Health Branch : “Guide on the Harmonization of Labour Inspection Statistics” (ILO, Geneva, 2016)

ILOSTAT presents information on various indicators pertaining to labour inspection, obtained from national sources. It features statistics on labour inspection visits, the number of inspectors per 10’000 persons employed and the average number of labour inspection visits conducted per inspector. For users interested in more detailed statistics, data on the number of registered workplaces that could be selected for labour inspection, and the number of labour inspectors by sex are also available.

Concepts and definitions

Labour inspectors are public officials or other authorities who are responsible for three key labour inspection activities: a) securing the enforcement of the legal provisions relating to conditions of work and the protection of workers while engaged in their work, such as provisions relating to hours, wages, safety, health and welfare, the employment of children and young persons, and other connected matters, in so far as such provisions are enforceable by labour inspectors; b) supplying technical information and advice to employers and workers concerning the most effective means of complying with the legal provisions; c) bringing to the notice of the competent authority defects or abuses not specifically covered by existing legal provisions. Labour inspectors have the authority to initiate processes that may lead to legal action.

Labour inspection visits refer to a physical presence of a labour inspector in a workplace for the purpose of carrying out a labour inspection and which is duly documented as required by national legislation.

A workplace can be defined as any physical space, whether a physical construction (such as a building or set of buildings) or not, in which at least one employed person carries out their work activities. Only those workplaces that are registered and could potentially be selected for labour inspection should be included in the total number.

Employment comprises all persons of working age who during a specified brief period, such as one week or one day, were in the following categories: a) paid employment (whether at work or with a job but not at work); or b) self-employment (whether at work or with an enterprise but not at work).5Resolution concerning statistics of work, employment and labour underutilization, adopted by the 19th International Conference of Labour Statisticians, Geneva, 2013

Also refer to the guide on the harmonization of labour inspection statistics:

Method of computation

Given that relative measures are easier to interpret and favour international comparability by removing the effect of the country size, ILOSTAT presents statistics on the following two calculated indicators:

Inspectors per 10’000 persons employed = Number of labour inspectors  / Number of employed persons liable to labour inspection  x 10’000

Ideally, the denominator should have the same coverage as the numerator (geographical coverage, population covered, economic activities covered, etc.). When the denominator is not available from the same source as the numerator, however, statistics on the employed population from other sources (such as a labour force survey) can be used as a proxy.

Labour inspection visits per inspector =  Number of inspection visits conducted during the reference year / Number of labour inspectors

Data sources

Labour inspection records are the prominent source of statistics on the labour inspectorate’s staff and activities. Other types of administrative records can also be used, where appropriate.

Interpretation and uses

Statistics on labour inspection can provide a powerful resource for labour inspectorates to properly understand the environment in which they act, the resources at their disposal in relation with the population they have to cover and the needs of workers and employers. Based on accurate and updated information, labour inspectorates can design scenarios to anticipate change, prepare approaches to observed trends, define strategic options and address challenges with the best methodological models. In addition, patterns in economic sectors, regions and enterprises can be identified, providing helpful background information to map risks.

Countries where labour inspection systems are underfunded and understaffed face difficulties to ensure the proper implementation of labour legislation. Labour inspection can help prevent problems such as occupational accidents and illnesses, absenteeism, abuse of workers and labour conflicts, thus contributing to enhancing productivity and economic development. Having statistics on labour inspection enables the assessment of the ability of inspection services to conduct their important role in an efficient manner.


Given that administrative records are the main source of data on labour inspection, there are a number of factors inherent to this type of source which could affect the reliability of the resulting statistics. In order to ensure the consistency and validity of the statistics, administrative records must be complete and comprehensive (without missing observations), kept up to date, and based on concepts, definitions and classifications clearly established for statistical purposes. Dates of occurrences must also be recorded with precision, and reference periods must be respected to ensure the reliability of statistics referred to for each period (and so, for instance, statistics on the number of inspection visits conducted and the number of labour inspectors would refer to the same reference period if the two indicators are being analysed together).

It is also worth noting that statistics on the number of visits conducted and the number of labour inspectors do not convey any information on the quality of the inspection services, the duration or type of the visits (first visit, specialized visit, preventive or reactive visit, etc.), or the results obtained through the visits conducted.


Note: Many publications are available only in English. If available in other languages, a new page will open displaying the options on the right. 

Decent Work and the Sustainable Development Goals: A Guidebook on SDG Labour Market Indicators

This Guidebook provides a detailed overview of the labour market indicators included in the Sustainable Development Goals Global Indicator Framework. It is intended to serve as a manual of best practices for calculating and interpreting the SDG labour market indicators, with a view to monitoring progress made at the national and international levels towards the achievement of the SDGs.

Guide on the Harmonization of Labour Inspection Statistics

This publication provides a methodology for the use of common terms and definitions and common procedures for the collection and compilation of labour inspection data. It also incorporates several additional recommendations that further the objective of harmonizing labour inspection statistics.

Resolution concerning statistics of occupational injuries (resulting from occupational accidents)

Adopted by the 16th ICLS (1998), this resolution sets standards of good practice for the collection and presentation of statistics of occupational injuries as guidance for countries wishing to revise their existing statistical systems in this field, or establish new ones. Its provisions should not undermine any existing national systems, nor should they lead to duplication of effort.

Scroll to Top
Skip to content